Synchronous motor drive system and synchronous motor drive method

Abstract

The present disclosure is constructed on the prior art inverter architecture, a pulse code width modulation (PCWM). This is an open loop motor control system without sensing its rotor position. The present disclosure employs a closed loop method to track the optimum efficiency motor operating point directly. A bench load test is conducted to gather information for an AI type control, which includes both load angle vs. voltage command charts and power factor vs. voltage command charts, with load levels as parameters for certain frequency command ranges. This way, the optimum efficiency motor operating points are generated a priori. The AI type control is mechanized to track the optimum efficiency motor operating points.

Claims

1. A synchronous motor drive system comprising: a synchronous motor; a load angle sensor for measuring a load angle of the synchronous motor; and a controller for generating a drive signal based on an input frequency command and the measured load angle, and supplying the drive signal to the synchronous motor, wherein the synchronous motor is a permanent magnet motor, the permanent magnet motor comprises: a rotor comprising a permanent magnet; and a stator comprising an armature, the synchronous motor drive system further comprises a permanent magnet magnetic flux sensor for detecting a permanent magnet magnetic flux, and the load angle sensor measures a phase difference between an armature magnetic flux and the permanent magnet magnetic flux to measure the load angle.

2. The synchronous motor drive system as claimed in claim 1, wherein the controller sends an armature magnetic flux phase signal representing a phase of the armature magnetic flux to the load angle sensor, the permanent magnet magnetic flux sensor sends a permanent magnet magnetic flux phase signal representing a phase of the permanent magnet magnetic flux to the load angle sensor, and the load angle sensor measures the phase difference between the armature magnetic flux and the permanent magnet magnetic flux based on the armature magnetic flux phase signal and the permanent magnet magnetic flux phase signal.

3. The synchronous motor drive system as claimed in claim 2, wherein the controller sends, as the armature magnetic flux phase signal, an on/off signal representing a magnitude of the armature magnetic flux, and the permanent magnet magnetic flux sensor sends, as the permanent magnet magnetic flux phase signal, an on/off signal representing a magnitude of the permanent magnet magnetic flux.

4. The synchronous motor drive system as claimed in claim 1, wherein the controller applies a sine wave voltage to the permanent magnet motor, expresses a phase of the voltage inn ways (n is an integer equal to or greater than 2), and sends n pulses to the load angle sensor during one period of the voltage, and the load angle sensor measures the load angle by measuring the number of pulses which corresponds to phase difference between the armature magnetic flux and the permanent magnet magnetic flux.

5. The synchronous motor drive system as claimed in claim 1, wherein the permanent magnet magnetic flux sensor is a Hall sensor.

6. The synchronous motor drive system as claimed in claim 1, wherein the controller comprises: a load angle control block for generating a voltage command based on the frequency command and the measured load angle to control the load angle; a PWM signal generator for generating a PWM signal based on the frequency command and the voltage command; and an inverter for generating the drive signal based on the PWM signal.

7. The synchronous motor drive system as claimed in claim 6, wherein the load angle control block comprises: a voltage command generator for generating the voltage command; a target load angle table storing a target load angle to be targeted for a frequency and a voltage applied to the synchronous motor; a target load angle determination block for determining the target load angle based on the frequency command and the voltage command by referring to the target load angle table; and a load angle error calculator for calculating a load angle error between the target load angle and the measured load angle; and the voltage command generator adjusts the voltage command to be generated based on the load angle error.

8. A synchronous motor drive method for driving a synchronous motor, comprising: a step of receiving input of a frequency command; a load angle measurement step of measuring a load angle of the synchronous motor; and a step of generating a drive signal based on the frequency command and the measured load angle, and supplying the drive signal to the synchronous motor, wherein the synchronous motor is a permanent magnet motor, the permanent magnet motor comprises: a rotor comprising a permanent magnet; and a stator comprising an armature, the synchronous motor drive method further comprises a step of detecting a permanent magnet magnetic flux; and the load angle measurement step measures a phase difference between an armature magnetic flux and the permanent magnet magnetic flux to measure the load angle.

9. The synchronous motor drive method as claimed in claim 8, wherein the load angle measurement step defines an armature magnetic flux axis at a position which is delayed from an armature voltage axis by π/2 in electric angle, and measures a phase difference between the armature magnetic flux and the permanent magnet magnetic flux relative to the armature magnetic flux axis.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a block diagram showing an embodiment of a motor drive system according to one embodiment of the prior patent invention.

(2) FIG. 2 is a block diagram showing the detailed inside embodiment of an ASIC and a gate drive and power transistor circuitry according to one embodiment of the prior patent invention.

(3) FIG. 3 is a unit sine function table referred to by the PCWM signal encoder when obtaining a unit sine function numeric by inputting a unit sine function table position number according to one embodiment of the prior patent invention.

(4) FIG. 4 is an illustration showing a relation between the instantaneous amplitude value of the fractional sine function and the pulse width numeric in the unit PWM pulse interval according to one embodiment of the prior patent invention.

(5) FIG. 5 is a functional relation graph between the pulse width numeric in the unit PWM pulse interval and the fractional sine function numeric according to one embodiment of the prior patent invention.

(6) FIG. 6 is a flow chart of a subprogram showing how a machine frequency and a machine voltage are updated every CK5 interrupt period (approximately 3.6864 ms) according to one embodiment of the prior patent invention.

(7) FIG. 7 is a unit sine function table scanning speed updating subprogram working with the subprogram in FIG. 6 according to one embodiment of the prior patent invention.

(8) FIG. 8 is a flow chart of a subprogram showing how a unit sine function table position number is updated every CK3 interrupt period (51.2 μs) according to one embodiment of the prior patent invention.

(9) FIG. 9 is a PCWM signal encoding subprogram working with the subprogram in FIG. 8 in which a center modulation front-half off period numeric is updated according to one embodiment of the prior patent invention.

(10) FIG. 10 is a time chart illustrating how PCWM numerics generated in the PCWM signal encoder are converted into a pulse width numeric in the unit PWM pulse interval in the PCWM signal encoder according to one embodiment of the prior patent invention.

(11) FIG. 11 is a comparison chart for the PCWM signal decoder output waveforms illustrating how the side modulation and the center modulation upper arm signals look in the unit PWM pulse interval according to one embodiment of the prior patent invention.

(12) FIG. 12 is a block diagram showing the detailed inside of the PCWM signal decoder in the ASIC according to one embodiment of the prior patent invention.

(13) FIG. 13 illustrates an example a unit PWM pulse polarity characteristic for the upper a arms' signal when the unit sine function is positive as a function of the normalized time in the unit PWM signal interval according to one embodiment of the prior patent invention.

(14) FIG. 14 is a comparison chart showing that center modulation PWM signal can create the power transistors' inter-conduction twice per each unit PWM pulse interval by one on/off operation of each power transistor in the compared to once for a side modulation PWM signal according to one embodiment of the prior patent invention.

(15) FIG. 15 is a block diagram showing a motor drive feedback control system using a load angle as the primary controlled variable according to one embodiment of the present disclosure.

(16) FIG. 16 is a comparison chart between vector control and torque control according to one embodiment of the present disclosure.

(17) FIG. 17 is a chart of target load angle vs. voltage command with load levels as the parameter for certain frequency ranges obtained by a load test according to one embodiment of the present disclosure.

(18) FIG. 18 is an external rotor type permanent magnet motor simplified drawing with the extended view according to one embodiment of the present disclosure.

(19) FIG. 19 is a vector diagram showing relations among motor variables according to one embodiment of the present disclosure.

(20) FIG. 20 is a time chart of showing relations among terminal voltage, armature magnetic flux, and permanent magnetic flux, for a U-phase upper arm as the example according to one embodiment of the present disclosure.

(21) FIG. 21 is a vector diagram defining a load angle as a function of motor variables according to one embodiment of the present disclosure.

(22) FIG. 22 is a vector diagram showing the motor torque is defined by the product of armature magnetic flux, permanent magnetic flux, and load angle for a small load angle according to one embodiment of the present disclosure.

(23) FIG. 23 is a control diagram and the associated time chart showing how the control is mechanized using the load angle as a controlled variable according to one embodiment of the present disclosure.

(24) FIG. 24 is a detailed block diagram showing how a control system is mechanized using a load angle as the primary controlled variable according to one embodiment of the present disclosure.

(25) FIG. 25 is a chart showing relation of base voltage vs. frequency command as shown in FIG. 24 according to one embodiment of the present disclosure.

(26) FIG. 26 is a block diagram showing an alternate motor drive feedback control system using a power factor angle as the primary controlled variable according to another embodiment of the present disclosure.

(27) FIG. 27 is a chart of power factor angle vs. voltage command with load levels as the parameter for certain frequency ranges obtained by a load test according to another embodiment of the present disclosure.

(28) FIG. 28 is a control diagram when a voltage phase is advanced from a current phase and the associated time chart showing how the control is mechanized using the power factor as a controlled variable according to another embodiment of the present disclosure.

(29) FIG. 29 is a control diagram when a voltage phase is delayed from a current phase and the associated time chart showing how the control is mechanized using the power factor as a controlled variable according to another embodiment of the present disclosure.

(30) FIG. 30 is a detailed block diagram showing how a control system is mechanized using a power factor angle as the primary controlled variable according to another embodiment of the present disclosure.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Example Problem to be Solved

(31) Globally, motors consume nearly 60% of the whole electric power usage. In order to alleviate the global warming effect, realization of low carbon society is preached. There is enough room to reduce CO2 by enhancing the efficiencies of motors.

(32) An object of the present disclosure is to present a synchronous motor drive system and a synchronous motor drive method with high efficiency.

Example Means for Solving the Problem

(33) According to the first aspect of the present disclosure, a synchronous motor drive system comprises: a synchronous motor; a load angle sensor for measuring a load angle of the synchronous motor; and a controller for generating a drive signal based on an input frequency command and the measured load angle, and supplying the drive signal to the synchronous motor.

(34) Here, the synchronous motor may be a permanent magnet motor, the permanent magnet motor may comprise: a rotor comprising a permanent magnet; and a stator comprising an armature, the synchronous motor drive system may further comprise a permanent magnet magnetic flux sensor for detecting a permanent magnet magnetic flux, and the load angle sensor may measure a phase difference between an armature magnetic flux and the permanent magnet magnetic flux to measure the load angle.

(35) Here, the controller may send an armature magnetic flux phase signal representing a phase of the armature magnetic flux to the load angle sensor, the permanent magnet magnetic flux sensor may send a permanent magnet magnetic flux phase signal representing a phase of the permanent magnet magnetic flux to the load angle sensor, and the load angle sensor may measure the phase difference between the armature magnetic flux and the permanent magnet magnetic flux based on the armature magnetic flux phase signal and the permanent magnet magnetic flux phase signal.

(36) Here, the controller may send an on/off signal representing a magnitude of the armature magnetic flux, as the armature magnetic flux phase signal, and the permanent magnet magnetic flux sensor may send an on/off signal representing a magnitude of the permanent magnet magnetic flux, as the permanent magnet magnetic flux phase signal.

(37) Here, the controller may apply a sine wave voltage to the permanent magnet motor, express a phase of the voltage in n ways (n is an integer equal to or greater than 2), and send n pulses to the load angle sensor during one period of the voltage, and the load angle sensor may measure the load angle by measuring the number of pulses which corresponds to the phase difference between the armature magnetic flux and the permanent magnet magnetic flux corresponds to.

(38) Here, the permanent magnet magnetic flux sensor may be a Hall sensor.

(39) Here, the controller may comprise: a load angle control block for generating a voltage command based on the frequency command and the measured load angle to control the load angle; a PWM signal generator for generating a PWM signal based on the frequency command and the voltage command; and an inverter for generating the drive signal based on the PWM signal.

(40) Here, the load angle control block may comprise: a voltage command generator for generating the voltage command; a target load angle table storing a target load angle to be targeted for a frequency and a voltage applied to the synchronous motor; a target load angle determination block for determining the target load angle based on the frequency command and the voltage command by referring to the target load angle table; and a load angle error calculator for calculating a load angle error between the target load angle and the measured load angle, and the voltage command generator may adjust the voltage command to be generated based on the load angle error.

(41) According to the second aspect of the present disclosure, a synchronous motor drive method for driving a synchronous motor, comprises: a step of receiving input of a frequency command; a load angle measurement step of measuring a load angle of the synchronous motor; and a step of generating a drive signal based on the frequency command and the measured load angle, and supplying the drive signal to the synchronous motor.

(42) Here, the synchronous motor may be a permanent magnet motor, the permanent magnet motor may comprise: a rotor comprising a permanent magnet; and a stator comprising an armature, the synchronous motor drive method may further comprise a step of detecting a permanent magnet magnetic flux, and the load angle measurement step may measure a phase difference between an armature magnetic flux and the permanent magnet magnetic flux to measure the load angle.

(43) Here, the load angle measurement step may define an armature magnetic flux axis at a position which is delayed from an armature voltage axis by π/2 in electric angle, and measure a phase difference between the armature magnetic flux and the permanent magnet magnetic flux relative to the armature magnetic flux axis.

(44) According to the third aspect of the present disclosure, a synchronous motor drive system comprises: a synchronous motor; a power factor angle sensor for measuring a power factor angle of the synchronous motor; and a controller for generating a drive signal based on an input frequency command and the measured power factor angle, and supplying the drive signal to the synchronous motor.

(45) Here, the synchronous motor drive system may further comprise a terminal current detection sensor for detecting a terminal current of the synchronous motor, and the power factor angle sensor may measure a phase difference between a terminal voltage of the synchronous motor and the terminal current to measure the power factor angle.

(46) Here, the controller may send a voltage phase signal representing a phase of the terminal voltage to the power factor angle sensor, the terminal current detection sensor may send a current phase signal representing a phase of the terminal current to the power factor angle sensor, and the power factor angle sensor may measure the phase difference between the terminal voltage and the terminal current based on the voltage phase signal and the current phase signal.

(47) Here, the controller may send an on/off signal representing a magnitude of the terminal voltage, as the voltage phase signal, and the terminal current detection sensor may send an on/off signal representing a magnitude of the terminal current, as the current phase signal.

(48) Here, the controller may apply a sine wave voltage to the synchronous motor, express a phase of the voltage in n ways (n is an integer equal to or greater than 2), and send n pulses to the power factor angle sensor during one period of the voltage, and the power factor angle sensor may measure the power factor angle by measuring the number of pulses which corresponds to the phase difference between the terminal voltage and the terminal current corresponds to.

(49) Here, the controller may comprise: a power factor angle control block for generating a voltage command based on the frequency command and the measured power factor angle to control the power factor angle; a PWM signal generator for generating a PWM signal based on the frequency command and the voltage command; and an inverter for generating the drive signal based on the PWM signal.

(50) Here, the power factor angle control block may comprise: a voltage command generator for generating the voltage command; a target power factor angle table storing a target power factor angle to be targeted for a frequency and a voltage applied to the synchronous motor; a target power factor angle determination block for determining the target power factor angle based on the frequency command and the voltage command by referring to the target power factor angle table; and a power factor angle error calculator for calculating a power factor angle error between the target power factor angle and the measured power factor angle, and the voltage command generator may adjust the voltage command to be generated based on the power factor angle error.

(51) According to the fourth aspect of the present disclosure, a synchronous motor drive method for driving a synchronous motor, comprises: a step of receiving input of a frequency command; a power factor angle measurement step of measuring a power factor angle of the synchronous motor; and a step of generating a drive signal based on the frequency command and the measured power factor angle, and supplying the drive signal to the synchronous motor.

(52) Here, the synchronous motor drive method may further comprise a step of detecting a terminal current of the synchronous motor, and the power factor angle measurement step may measure a phase difference between a terminal voltage of the synchronous motor and the terminal current to measure the power factor angle.

Example Effect of Certain Embodiments

(53) According to the present disclosure, a synchronous motor drive system and a synchronous motor drive method with high efficiency can be presented.

(54) In the following, by referring to the figures, embodiments of the present disclosure will be described in detail.

First Embodiment

(55) In the first embodiment of the present disclosure, a permanent magnet motor (more specifically, a three phase permanent magnet motor) is used as a synchronous motor.

(56) The first embodiment of the present disclosure relates to an inexpensive inverter system which can yield equal or higher efficiency compared to a vector control inverter in a simpler method for a three phase permanent magnet motor, which is characterized as a higher efficiency motor.

(57) The present embodiment presents an inverter system for a permanent magnet motor which directly tracks the optimal efficiency by torque control using a load angle, despite its inexpensive structure.

(58) Problems to be resolved by the present embodiment are as follows.

(59) (1) Defining an attracting action between an electromagnet (armature) of a motor stator and a permanent magnet of a motor rotor by using a numerical expression model of electromagnetic induction theory.

(60) (2) Realizing a control system model in a static coordinate system which can drive a permanent magnet motor by sinusoidal wave to obtain the optimal efficiency, contrary to a vector control counterpart.

(61) (3) Defining a motor load angle as an index of efficiency control.

(62) (4) Forming a table consisting of a set of values of load angles at which the motor efficiency is optimal for the number of rotations of the drive motor in accordance with the magnitude of the load, by conducting a load test on the motor in advance, and adjusting the applied voltage on the motor to make the counted load angle value coincide with the ideal value stored in the table.

(63) (5) Conducting measurement of the motor load angle by converting the sine wave of each of the armature magnetic flux and the permanent magnet magnetic flux to an on/off signal (e.g. signal which becomes on when the value is zero or more, and off when the value is less than zero), and detecting the phase difference between the signals after the conversion.

(64) (6) Enabling to count digitally what percentage of the wavelength at the current drive frequency the magnitude of the load angle becomes.

(65) (7) Making the control circuit simpler and smaller.

(66) The PCWM inverter, on which the present embodiment is based, is one which drives the PMAC motor by sinusoidal wave signal in an open loop mode without sensing the rotor position.

(67) The present embodiment enables the inverter system to operate constantly at the optimal efficiency point in closed loop control by installing only one sensor on the motor stator to detect the phase of the rotor.

(68) The first feature of the present embodiment is to investigate an attracting action between the electromagnet (armature) of the motor stator excited by the inverter and the permanent magnet installed on the rotor under the condition of the fixed rotation axis. Specifically, it is to figure out what rotating motion the rotor does by using a numerical expression model based on the electromagnetic induction theory, when the magnetic flux by the rotating magnetic field of the electromagnet (armature magnetic flux) is defined as the driving side and the magnet flux by the permanent magnet (permanent magnet magnetic flux) of the rotor is defined as the tracking side.

(69) The second feature of the present embodiment is (a) to define the armature magnetic flux axis which is delayed from the armature voltage axis by ¼ wavelength as the base armature magnetic flux axis, and (b) to conduct control using a fixed coordinate system (stationary coordinate system). The present embodiment employs a fixed coordinate system (stationary coordinate system) contrary to vector control which uses a rotating coordinate system. In the PCWM method of the present embodiment, the number of data in the sine wave 360° function table is stored as digital information of a multiple of 6. Such method is impossible for an existing conventional inverter system.

(70) The third feature of the present embodiment is (a) to define a load angle as an index for realizing the optimal efficiency, and obtain the value of the load angle at which the efficiency is optimal in accordance with the magnitude of the load by conducting a load test on the inverter and motor in advance, (b) to obtain the relationship between the command voltage and the target load angle which realizes the optimal efficiency in the form of a table in advance, and (c) to build a model follower adaptive control system which adjusts the command voltage so that the value of the counted load angle coincides with the ideal value stored in the table in a real machine.

(71) The fourth feature of the present embodiment is to conduct measurement of the motor load angle by converting the sinusoidal wave of each of the armature magnetic flux and the permanent magnet magnetic flux to a 50% duty on/off signal, and detecting the phase difference between the signals after the conversion.

(72) The fifth feature of the present embodiment is that in contrast with the first carrier used in the PCWM method which has a constant frequency, the period of the sine wave of the voltage applied to the motor changes in accordance with the drive frequency, and therefore by using the second carrier which is synchronized to the sine wave frequency, the present embodiment makes it possible to count digitally what percentage of the wavelength at the current drive frequency the magnitude of the load angle becomes. Especially, using the second carrier for measuring the load angle is impossible for an existing conventional inverter system. In the PCWM method of the present embodiment, carrier frequency control is operated as digital information.

(73) The sixth feature of the present embodiment is characterized as a full-digital construction by not using ND converters and many sensor circuits and consisting of fewer and less expensive parts in order to make the control circuit simpler and smaller.

(74) FIG. 15 illustrates a permanent magnet motor drive system (synchronous motor drive system) in the first embodiment of the present disclosure, and represents a circuit block diagram for an efficiency control of either an air conditioner or a fan. The permanent magnet motor drive system 60 of the present embodiment comprises a controller 62, a permanent magnet motor 64, and a load angle sensor 66. In the present embodiment, as a load 70, for example, an air conditioner or a fan is assumed. When a user inputs a room temperature command to a host CPU 50, the host CPU 50 inputs a frequency command which depends on the room temperature command, to the controller 62 inside the permanent magnet motor drive system 60. The controller 62 generates a drive signal which depends on the frequency command, and supplies it to the permanent magnet motor 64. The permanent magnet motor 64 operates in accordance with the drive signal, and supplies a speed and a torque to the load 70. The load (air conditioner or fan) 70 operates in accordance with the speed and the torque, and the room temperature changes.

(75) In the present embodiment, a load angle is controlled by regulating a voltage in an inner loop. That is, the load angle sensor 66 measures the load angle of the permanent magnet motor 64, and supplies it to the controller 62. The controller 62 generates the drive signal based on the frequency command and the measured load angle. Here, for the number of rotations of the permanent magnet motor 64, in accordance with the magnitude of the load, there exists the value of the load angle (target load angle) at which the efficiency of the permanent magnet motor 64 becomes optimal. The controller 62 generates the drive signal by controlling (adjusting) the applied voltage independently of the applied frequency so that the supplied measured load angle approaches the target load angle, and thereby the optimal efficiency can be achieved.

(76) Further, in the present embodiment, frequency control is conducted in an outer loop. That is, a room temperature sensor 80 measures the room temperature, and supplies it to the host CPU 50. The host CPU 50 controls (adjusts) the frequency command in accordance with the supplied room temperature.

(77) Further, it is noted that a compressor, etc. may be employed as the load 70.

(78) FIG. 16 is a comparison chart of vector control with torque control (inverter) of the present embodiment.

(79) FIG. 17 is a diagram showing one example of an optimal efficiency data table obtained as a result of conducting a load test on a motor. We will explain this further later.

(80) The motor rotor equation of motion described here is defined using stationary cylindrical coordinate system, which is employed throughout the present analysis. The stator winding located inside the motor is magnetized by a digitalized sinusoidal wave from the driving inverter. Ferrite magnets (permanent magnets) are attached to the inside of the motor rotor forming the motor outer shell, and are magnetized by sinusoidal waveform as well.

(81) In this analysis, we initially assume a two-pole/six slot motor and expand it onto the 2-D plane. For simplicity, we assume a motor configuration having no salient poles. Further, the analysis is performed for the U phase as the representative axis. FIG. 18 shows a synchronous motor model and its 2-D expansion view used in the present embodiment.

(82) In FIG. 18, assuming the moment when the armature current in the U phase is maximum, the rotating magnetic flux Φa generated by the armature current (armature reaction magnetic flux) which includes magnetic fluxes generated by the V phase and W phase currents is on the winding axis of the U phase. By noting that the rotating magnetic flux is rotating at the synchronous speed toward the CCW (counterclockwise) direction in the direction of the phase order, the electromotive force Ea=−jXaI generated by this magnetic flux is delayed from the current I by π/2. When this is treated as a voltage drop, the sign changes and it becomes advanced from I by π/2.

(83) The voltage drop jxlI by the armature leakage reactance is also advanced from I by π/2, and the sum of the two voltage drops j(Xa+xl) I=jXsI=the synchronous reactance drop is also advanced from I by π/2.

(84) If the center of the magnetic pole at the moment when the current of the U phase is maximum is at the position which is delayed from the winding axis of the U phase by ζ as shown in FIG. 18, the electromotive force E (We assume that the direction opposite to I is positive.) generated in the U phase by the rotation of the magnetic flux Φm by the magnetic pole at the synchronous speed toward the CCW direction in the direction of the phase order is delayed from jXsI in phase by ζ.

(85) Further, when the armature winding and inverter lead resistance r is considered, the winding resistance equivalent magnetic flux Φr by this is in phase with I, and is delayed from Φa by π/2.

(86) The vector summation of E, rI and jXsI must be the power supply voltage V applied to the U phase, and therefore the vector diagram of the U phase is obtained as shown in FIG. 19.

(87) By referring to FIG. 20, we start the analysis using the following equation of electromagnetic induction.
V=dΦe(θ)/dt=pΦe(θ)  (1)

(88) where

(89) V=V1 sin (θ);

(90) V1: Maximum voltage applied from the inverter at the U phase terminal;

(91) θ: Rotational angle from the U phase terminal voltage axis Ue, CCW positive;

(92) Φe=−Φe1 cos (θ): Armature magnetic flux at a rotational angle θ

(93) Φe1: Positive maximum value of the armature magnetic flux induced by the U phase terminal voltage; and

(94) p: Differentiation operator.

(95) The equation (1) indicates that the phase of the armature magnetic flux Φe is delayed from the phase of the terminal voltage V by π/2.

(96) In other words, the armature magnetic flux Φe at a phase angle of θ from the U phase terminal voltage axis Ue is
Φe=−Φe1 cos(θ)  (2)

(97) In order to deal with the interaction with the permanent magnet magnetic flux, it is convenient to define the armature magnetic flux axis Um by rotating the base axis of the armature magnetic flux Φe in the clockwise direction by π/2 in the x-y plane from the U phase terminal voltage axis Ue. By using this redefinition, the armature magnetic flux Φe can be written as.
Φe=Φe1 sin(θ)  (3)

(98) where

(99) θ: Rotational angle from the U phase armature magnetic flux axis Um, CCW positive.

(100) A further manipulation of the equation (1) results in

(101) $\begin{array}{cc}V=p\Phi e=j\omega \Phi e& \left(4\right)\end{array}$

(102) Rewriting equation (4) for Φe results in

(103) $\begin{array}{cc}\Phi e=V/j\omega =-jV/\omega & \left(5\right)\end{array}$

(104) Similarly to the equation (1), the following equation of electromagnetic induction can be defined.
E=dΦm(θ)/dt=pΦm(θ)  (6)

(105) where

(106) E=E1 sin (θ);

(107) E1: Maximum voltage induced by the permanent magnet;

(108) θ: Rotational angle from the U phase permanent magnet voltage axis q, CCW positive;

(109) Φm=−Φm1 cos (θ): Permanent magnet magnetic flux at a rotational angle θ

(110) Φm1: Positive maximum value of the permanent magnet magnetic flux induced by the permanent magnet;

(111) p: Differentiation operator;

(112) The equation (6) indicates that the phase of Om is delayed from the phase of the permanent magnet voltage E by π/2.

(113) When the permanent magnet magnetic flux axis d is defined by rotating the base axis of the permanent magnet magnetic flux Φm from the permanent magnet voltage axis q by π/2 in the clockwise direction, the permanent magnet magnetic flux Φm can be written as
Φm=Φm1 sin(θ)  (7)

(114) where

(115) θ: Rotational angle from the U phase permanent magnet magnetic flux axis d, CCW positive.

(116) Similarly to the equation (5),
Φm=−jE/ω  (8)

(117) When considering the winding resistance r of the armature winding including the inverter lead, the winding resistance equivalent magnetic flux Φr is defined as follows.
Φr=−jrI/ω  (9)

(118) By referring to FIG. 19, we will analyze the following conventional motor circuit equation.
V=E+rI+jXsI  (10)

(119) where

(120) r: Winding resistance;

(121) Xs: Synchronous Reactance=ωLs;

(122) Ls: Synchronous Inductance.

(123) Multiplying both sides of equation (10) by −j/ω, it becomes

(124) $\begin{array}{cc}-\mathrm{jV}/\omega =-\mathrm{jE}/\omega -\mathrm{jrl}/\omega ⫬+\mathrm{Xsl}/\omega =-\mathrm{jE}/\omega -\mathrm{jrl}/\omega +\mathrm{Lsl}& \left(11\right)\end{array}$

(125) Substituting equations (5), (8) and (9) into equation (11) yields
Φe=Φm++LsI  (12)

(126) Newly defining the synchronous inductance magnetic flux Φa as follows.
=LsI  (13)

(127) The equation (12) further turns to the equation (14) as a vector relationship of magnetic fluxes.
Φe=Φm+Φr+Φa  (14)

(128) The relationship of the equation (14) is illustrated as a vector diagram in FIG. 21.

(129) Traditionally, we have the following motor torque equation,

(130) $\begin{array}{cc}T=k\left\{.Math.V.Math..Math.E.Math./\left(\omega Xs\right)\right\}\sin \delta =k\left\{V1E1/\left(\omega \mathrm{Xs}\right)\right\}\sin \delta & \left(15\right)\end{array}$

(131) where

(132) k=3P/2: Constant number;

(133) P: Number of motor poles;

(134) V1: Maximum voltage applied from the inverter at the U phase terminal;

(135) E1: Maximum voltage induced by the permanent magnet;

(136) ω: Motor rotation angular speed.

(137) Substituting the equation of the synchronous reactance Xs=ωLs which is defined in equation (10) in equation (15) results in

(138) $\begin{array}{cc}T=k\left\{V1E1/\left(\omega 2\mathrm{Ls}\right)\right\}\sin \delta =k\left\{\left(V1/\omega \right)*\left(E1/\omega \right)/\mathrm{Ls}\right)\}\sin \delta & \left(16\right)\end{array}$

(139) where
|Φe|=Φe1=V1/ω  (17)
|Φm|=Φm1=E1/ω  (18)

(140) Substituting equations (17) and (18) in equation (16) and arranging it result in
T=k|Φe∥Φm|sin δ/Ls  (19)

(141) where

(142) δ: Load angle=Included angle between the permanent magnet magnetic flux axis and the armature magnetic flux axis

(143) Equation (19) indicates the motor torque is proportional to the area surrounded by the oblique sides Φe and Om and their included angle δ. This motor torque equation is illustrated in FIG. 22.

(144) From this, it is noted that the motor torque is approximately proportional to the load angle δ when the load angle δ is small.

(145) The maximum value of the permanent magnet magnetic flux induced by the permanent magnet is given and unchangeable. However, the present inverter can precisely control the voltage value of the applied voltage (i.e. the terminal voltage of an armature winding) independently from the applied frequency. Therefore, it can realize the optimum efficiency by changing the magnitude of the armature magnetic flux in accordance with the speed and the magnitude of the load of the motor.

(146) The load angle measurement and control of the present embodiment are not always executed during the entire motor drive. They are operational within certain motor speed ranges in which the motor has entered into a steady operation. For the motor drive in a transient state of increasing or decreasing the speed of the motor, an open loop control is executed by fully utilizing the character of a PCWM scheme employed by the present embodiment. The frequency of the load angle measurement and control of the present embodiment may be control of an extremely long interval such as on a “minute” basis. However, at the time of detecting the load angle, real time processing of a short interval by a counting signal PCK which is output from the PCWM signal encoder, described later, is preferable.

(147) A 24-poles/18-slots external rotor type motor is preferable for practical use. A method for measuring the motor load angle for such motor will be explained by using FIG. 23 which shows one embodiment.

(148) The rotational angle of the rotor (electric angle) 8 during one cycle of the driving sinusoidal wave of the motor is
θ=2π/(24/2)=π/6  (20)

(149) As the above-described equation (1) indicates that the armature magnetic flux axis Um is delayed from the armature voltage axis Ue by ¼ wavelength, when the equation (20) is multiplied by this value, the included angle between the two axes is given as follows:
¼θ=π/(6*4)=π/24=7.5°  (21)

(150) In the present embodiment, the motor load angle is measured by defining the armature magnetic flux axis Um at the position which is delayed from the armature voltage axis Ue by π/2 (¼ wavelength) in electric angle, and measuring the phase difference between the armature magnetic flux and the permanent magnet magnetic flux relative to the armature magnetic flux axis Um.

(151) The measurement of the motor load angle is made by converting the sinusoidal wave of each of the armature magnetic flux and the permanent magnet magnetic flux to 50% duty on/off signal, and measuring the phase difference between the signals after the conversion. Thereby, an ND converter for measuring an amplitude becomes unnecessary, and a signal processing circuit which is tolerant to external noises can be realized. Further, in contrast with the first carrier (CK3 described later) used in the PCWM scheme which has a constant frequency, the period of the sine wave of the voltage applied to the motor changes in accordance with the drive frequency, and therefore by using the second carrier (the counting signal PCK described later) which is synchronized to the sine wave frequency, what percentage of the wavelength at the current drive frequency the magnitude of the load angle becomes, is digitally counted.

(152) In FIG. 23, the sinusoidal wave of the armature magnetic flux is converted to an on/off signal δD representing the magnitude of the armature magnetic flux. The conversion is conducted inside the PCWM signal encoder 116, which is described later. That is, the sinusoidal wave is converted to a signal which becomes on when the magnitude of the armature magnetic flux is zero or more, and off when the magnitude of the armature magnetic flux is less than zero. Here, the ON signal indicates the N pole. Further, the sinusoidal wave of the permanent magnet magnetic flux is detected at the Hall sensor installed on the Hall sensor PCB (Print Circuit Board), and it is converted to an on/off signal δH representing the magnitude of the permanent magnet magnetic flux. That is, the sinusoidal wave is converted to a signal which becomes on when the magnitude of the permanent magnet magnetic flux is zero or more, and off when the magnitude of the permanent magnet magnetic flux is less than zero. Here, the ON signal indicates the S pole. In the present embodiment, both OD and OH are made to be 50% duty on/off signals.

(153) In FIG. 23, the sinusoidal wave table position means the table position of FIG. 3. Thus, in one wavelength (one period) of the drive frequency, the PCWM signal encoder outputs 720 pulses as the counting signal. For example, as a result of digital measurement, if the phase difference (measured load angle δL) between the armature magnetic flux and the permanent magnet magnetic flux is 36 of the above-described pulses, this is 5% (=36/720) of the wavelength, and therefore the measured load angle δL is 18° (=360°×0.05).

(154) When the number of rotations of the drive motor is given, there exists the load angle value which gives optimum efficiency for the varying load magnitude. A model follower adaptive control method is employed by adjusting the motor applied voltage to have the value of the counted load angle become the ideal value stored in the table. For this purpose, a load test is conducted on the motor in advance, to get the table shown in FIG. 17. The relationship between the voltage command VC and the target load angle δT shown in FIG. 17 is prepared for each of the certain frequency ranges, and stored in the target load angle table 104. Thus, the target load angle table 104 stores the target load angle (OT) to be targeted for the frequency (FC) and the voltage (VC) applied to the permanent magnet motor 64.

(155) If the present load angle measurement system is likened to measurement of the passing time of a train passing at a railroad crossing, it resembles conducting fixed point observation of the time difference from the closing of the crossing gate to the arrival of the train. That is, measurement of the delay time of the permanent magnet magnetic flux axis from the armature magnetic flux axis is understood as the phase difference between the respective pulse trains of the armature magnetic flux and the permanent magnet magnetic flux which have been converted to 50% duty.

(156) By referring to FIG. 24 which shows one embodiment of the load angle control, further detailed explanation will be given. FIG. 24 specifically shows the constitution of the permanent magnet motor drive system 60 of the present embodiment shown in FIG. 15.

(157) In the present embodiment, the load angle sensor 66 and the load angle control block 101 perform the load angle measurement and control. However, as described previously, the load angle measurement and control are not always done during the entire motor drive. They are performed within certain motor speed ranges in which the motor enters into steady operation.

(158) The load angle control block 101 generates the voltage command VC based on the frequency command FC and the measured load angle δL to control the load angle.

(159) The frequency command FC supplied from the outside of FIG. 24, and the voltage command VC which is calculated on real time inside the figure and described later, are input to a target load angle determination block 102 inside the load angle control block 101. The target load angle determination block 102 determines the target load angle δT based on the frequency command FC and the voltage command VC by referring to the target load angle table 104. The target load angle table 104 is a set of values of load angles at which the efficiency becomes optimal, and which are obtained by conducting load test on the motor in advance. The target load angle table 104 is given in the format shown in FIG. 17 as described previously.

(160) The load angle sensor 66 in FIG. 24 subtracts the Hall sensor phase signal (signal representing the phase of the Hall sensor, i.e., phase of the permanent magnet magnetic flux) OH output from the Hall sensor 136, from the armature magnetic flux phase signal (signal representing the phase of the armature magnetic flux) OD output from the PCWM signal encoder 116, which is described later, to obtain the measured load angle δL as the output.

(161) The load angle sensor 66 outputs the measured load angle δL of which measurement is made by the method described in FIG. 23. The load angle sensor 66 may output the measured load angle δL after averaging it. The measured load angle δL output from the load angle sensor 66 is then input to a load angle error calculator 106.

(162) The load angle error calculator 106 subtracts the measured load angle δL from the target load angle δT to obtain a load angle error δE. The load angle error δE is input to a voltage command accumulator 112 inside a voltage command generator 107.

(163) The voltage command generator 107 generates a voltage command VC. The voltage command generator 107 comprises a base voltage determination block 108, a V/F base voltage table 110, and the voltage command accumulator 112.

(164) The frequency command FC is input to the base voltage determination block 108. The base voltage determination block 108 determines a base voltage VB by referring to the V/F base voltage table 110. The V/F base voltage table 110 is a one obtained by conducting a load test on the motor in advance, and given in the format shown in FIG. 25. The base voltage VB output from the base voltage determination block 108, is input to the voltage command accumulator 112 having an addition/subtraction/storage function.

(165) At the time of entering into the load angle measurement and control loop, the voltage command accumulator 112 outputs the base voltage VB as the initial value of the voltage command VC, to the target load angle determination block 102 and the PCWM encoder 116. Thereafter, the voltage command accumulator 112 receives the load angle error δE from the load angle error calculator 106, and adjusts the voltage command VC based on the load angle error δE. Specifically, when the load angle error δE is plus, it means the measured load angle is less than the target load angle. Therefore, it works so that the voltage command VC is decreased. Contrary, when the load angle error δE is minus, it means the measured load angle is more than the target load angle. Therefore, it works so that the voltage command VC is increased.

(166) On the other hand, at the time of exiting from the load angle measurement and control loop, the voltage command accumulator 112 continues to renew the value held by itself toward the value of the base voltage VB so that the held value matches the base voltage VB in the end.

(167) A PWM signal generator 114 comprises the PCWM signal encoder 116, a PCWM signal decoder 128, and a sine wave 360° function table 120. For example, the PWM signal generator 114 can be realized as an LSI or an ASIC. A logic part DC voltage 138 is supplied to the PWM signal generator 114. The PWM signal generator 114 generates a PWM signal based on the frequency command FC and the voltage command VC.

(168) Here, the PWM signal generator 114 can be configured similarly to the ASIC 06 shown in FIGS. 1 and 2. However, in the present embodiment, the clocks CK1, CK3, CK4 and CK5 shown in FIG. 2 have been made eight times greater. That is, CK1, CK3, CK4 and CK5 have been made to be 1.6 μs, 409.6 μs, 409.6 μs and 29.4912 ms, respectively. Further, until the machine frequency FM reaches the frequency command (command frequency) FC, the process shown in FIG. 6 is conducted, and load angle control is not conducted. After the machine frequency FM reaches the frequency command (command frequency) FC, the voltage command VC is input as a machine voltage VM to the PCWM signal encoder 116, and load angle control is conducted.

(169) The PCWM signal encoder 116 receives the frequency command FC and the voltage command VC as inputs, and receives data stored in the sine wave 360° function table 120 shown in FIG. 3 as a write signal 122 by a read signal 118 of a prescribed period. The PCWM signal encoder 116 outputs an encoded PCWM signal 126 by real time processing the write signal 122 and the information of the frequency command FC and the voltage command VC.

(170) The format of the sine wave 360° function table 120 is the same as the unit sine function table in FIG. 3. The table consists of 720 8-bit binary signals. 720 is selected as a multiple of 6. In the present embodiment, a three phase permanent magnet type motor is used, and therefore it is preferable to select a multiple of 6.

(171) A figure showing how the information of the sine wave 360° function table 120 is processed in the PCWM signal encoder 116, is the fractional sine function numeric nf representing the instantaneous amplitude value and the pulse width numeric pw of FIG. 4. It shows that after the inverter starts at the origin of the circle, the scanning speed increases with the operation point moving toward the outside as it is accelerated, and when it reaches the maximum speed, the operation point circulates on the outer periphery of the circle.

(172) A figure showing how the fractional sine function numeric nf in FIG. 4 is converted to the pulse width numeric pw, is the relationship between the pulse width numeric pw and the fractional sine function numeric nf of FIG. 5. However, as described above, in the present embodiment, CK1 and CK3 are made as 1.6 μs and 409.6 μs, respectively.

(173) As shown in the left side of FIG. 5, based on the calculation formula shown in FIG. 5, the encoded pulse width numeric pw in the unit PWM pulse interval is obtained, and this is output as the PCWM signal 126.

(174) The PCWM signal decoder 128 decodes the PCWM signal 126 input from the PCWM signal encoder 116 on real time as a PWM signal 130, and outputs it to the inverter (gate drive) 132 of the next stage. The decoding method of the PCWM signal decoder 128 is described previously with reference to FIGS. 10-14.

(175) The inverter 132 generates a motor drive signal 134 based on the PWM signal 130. The inverter 132 can be configured similarly to the gate drive and power transistor circuitry 08 shown in FIGS. 1 and 2. The motor drive signal 134 output from the inverter 132 drives the three phase permanent magnet motor 64 which is directly coupled to the load 70. A main circuit DC voltage 140 is supplied to the inverter 132.

(176) The controller 62 applies the voltage to the permanent magnet motor 64, and therefore it knows the state of the armature magnetic flux at each time. The PCWM signal encoder 116 of the controller 62 outputs, among its outputs, the armature magnetic flux phase signal δD which is an on/off signal of 50% duty representing the magnitude (zero or more, or less than zero) of the sine wave signal of the armature magnetic flux.

(177) In the case of FIG. 24, the Hall sensor 136 is installed on the stator side of the three phase permanent magnet motor 64 (see also FIG. 23). The Hall sensor 136 detects the sine wave signal generated by the rotation of the three phase permanent magnet motor 64, converts the signal to the Hall sensor phase signal δH which is an on/off signal of 50% duty representing the magnitude (zero or more, or less than zero) of the signal by a comparator in the Hall sensor 136, and outputs it.

(178) The PCWM signal encoder 116 outputs the counting signal (read signal) PCK which becomes on/off at every occurrence of the write signal 122. The counting signal PCK outputs as many pulses as the number of data stored in the sine wave 360° function table 120 during one period of the drive frequency signal regardless of the magnitude of the drive frequency. This can be called the second carrier which is synchronized to the period of the drive frequency. When the phase difference between the armature magnetic flux phase signal δD and the Hall sensor phase signal δH is digitally measured, the phase difference becomes an effective means as an index showing the ratio to the wavelength of the drive frequency signal.

(179) The load angle sensor 66 receives the armature magnetic flux phase signal OD and the Hall sensor phase signal δH as inputs, counts the phase difference between the two by the counting signal PCK, and outputs the resulting number of counts as the measured load angle M.

(180) In this way, it is possible to count digitally what percentage of the wavelength at the current drive frequency the magnitude of the load angle becomes. Namely, the inverter 132 of the controller 62 applies the sine wave voltage to the permanent magnet motor 64. Further, the phase of the voltage is expressed in n=720 ways (see FIGS. 3 and 4). Further, the PCWM signal encoder 116 of the controller 62 sends n=720 counting signals PCK (pulses) to the load angle sensor 66 during one period (one wavelength) of the voltage. Then, the load angle sensor 66 measures the load angle by measuring the number of pulses which corresponds to the phase difference between the armature magnetic flux and the permanent magnet magnetic flux corresponds to. Thus, the load angle sensor 66 can count digitally what percentage of the wavelength at the current drive frequency the magnitude of the load angle becomes.

(181) In the present embodiment, the phase of the voltage is expressed in n=720 ways. However, another value (integer equal to or greater than 2) can be employed as n. Here, as the value of n, 6 or more is preferable. Specifically, a multiple of 6 which is 6 or more is preferable.

(182) Further, in the present embodiment, a Hall sensor is used as a permanent magnet magnetic flux sensor detecting the permanent magnet magnetic flux, but another permanent magnet magnetic flux sensor may be used.

Second Embodiment

(183) In the above-described first embodiment, control is conducted based on load angle, but in the second embodiment of the present disclosure, control is conducted based on power factor angle.

(184) FIG. 26 illustrates a permanent magnet motor drive system (synchronous motor drive system) in the second embodiment of the present disclosure, and represents a circuit block diagram for an efficiency control of either an air conditioner or a fan. The permanent magnet motor drive system 60 of the present embodiment comprises a controller 62, a permanent magnet motor 64, and a power factor angle sensor 67. In the present embodiment, as a load 70, for example, an air conditioner or a fan is assumed. When a user inputs a room temperature command to a host CPU 50, the host CPU 50 inputs a frequency command which depends on the room temperature command, to the controller 62 inside the permanent magnet motor drive system 60. The controller 62 generates a drive signal which depends on the frequency command, and supplies it to the permanent magnet motor 64. The permanent magnet motor 64 operates in accordance with the drive signal, and supplies a speed and a torque to the load 70. The load (air conditioner or fan) 70 operates in accordance with the speed and the torque, and the room temperature changes.

(185) In the present embodiment, a power factor angle is controlled by regulating a voltage in an inner loop. That is, the power factor angle sensor 67 measures the power factor angle of the permanent magnet motor 64, and supplies it to the controller 62. The controller 62 generates the drive signal based on the frequency command and the measured power factor angle. Here, for the number of rotations of the permanent magnet motor 64, in accordance with the magnitude of the load, there exists the value of the power factor angle (target power factor angle) at which the efficiency of the permanent magnet motor 64 becomes optimal. The controller 62 generates the drive signal by controlling (adjusting) the applied voltage independently of the applied frequency so that the supplied measured power factor angle approaches the target power factor angle, and thereby the optimal efficiency can be achieved.

(186) Further, in the present embodiment, frequency control is conducted in an outer loop. That is, a room temperature sensor 80 measures the room temperature, and supplies it to the host CPU 50. The host CPU 50 controls (adjusts) the frequency command in accordance with the supplied room temperature.

(187) Further, it is noted that a compressor, etc. may be employed as the load 70.

(188) FIG. 27 is a diagram showing one example of an optimal efficiency data table obtained as a result of conducting a load test on a motor. We will explain this further later.

(189) The power factor angle measurement and control of the present embodiment are not always executed during the entire motor drive. They are operational within a certain motor speed range in which the motor has entered into steady operation. For the motor drive in a transient state of increasing or decreasing the speed of the motor, an open loop control is executed by fully utilizing the character of a PCWM scheme employed by the present embodiment. The frequency of the power factor angle measurement and control of the present embodiment may be control of an extremely long interval such as on a “minute” basis. However, at the time of detecting the power factor angle, real time processing of a short interval by a counting signal PCK output from the PCWM signal encoder, described later, is preferable.

(190) A 24-poles/18-slots external rotor type motor is preferable for practical use. A method for measuring the motor power factor angle for such motor will be explained by using FIGS. 28 and 29 which show one embodiment.

(191) The measurement of the motor power factor angle is conducted by measuring the phase difference between a terminal voltage (In the present embodiment, the U phase terminal voltage is used as a representative of the three phases.) of the motor and a terminal current (In the present embodiment, the U phase terminal current is used as a representative of the three phases.) of the motor. Terminals of the motor and terminals of an inverter are connected each other. Therefore, the terminal voltage and the terminal current of the motor are identical to the terminal voltage and the terminal current of the inverter, respectively. In the present embodiment, a current sensor 144 detects the terminal current of the inverter, and thereby the terminal current of the motor is detected. Specifically, the measurement of the motor power factor angle is made by converting the sine wave of each of the terminal voltage and the terminal current to 50% duty on/off signal, and measuring the phase difference between the signals after the conversion. Thereby, an ND converter for measuring an amplitude becomes unnecessary, and a signal processing circuit which is tolerant to external noises can be realized. Further, in contrast with the first carrier (CK3 described later) used in the PCWM scheme which has a constant frequency, the period of the sine wave of the voltage applied to the motor changes in accordance with a drive frequency, and therefore by using the second carrier (the counting signal PCK described later) which is synchronized to the sine wave frequency, what percentage of the wavelength at the current drive frequency the magnitude of the power factor angle becomes, is digitally counted.

(192) In FIGS. 28 and 29, the sine wave of the terminal voltage is converted to an on/off signal δV representing the magnitude of the terminal voltage. The conversion is conducted inside a PCWM signal encoder 116, which is described later. That is, the sine wave is converted to a signal which becomes on when the magnitude of the terminal voltage is zero or more, and off when the magnitude of the terminal voltage is less than zero. Secondly, the sine wave of the terminal current is converted to an on/off signal δI representing the magnitude of the terminal current. The conversion is conducted inside the current sensor 144, which is described later. That is, the sine wave is converted to a signal which becomes on when the magnitude of the terminal current is zero or more, and off when the magnitude of the terminal current is less than zero. In the present embodiment, both δV and δI are made to be 50% duty on/off signals.

(193) FIG. 28 shows a case where the voltage phase is advanced from the current phase. In FIG. 28, the sinusoidal wave table position is identical to the table position of FIG. 3. Thus, in one wavelength (one period) of the drive frequency, the PCWM signal encoder outputs 720 pulses as the counting signal. For example, as a result of digital measurement, if the phase difference (measured power factor angle δP) between the voltage phase and the current phase is 36 of the above-described pulses, this is 5% (=36/720) of the wavelength, and therefore the measured power factor angle δP is 18° (=360°×0.05).

(194) FIG. 29 shows a case where the voltage phase is delayed from the current phase. In FIG. 29, the sinusoidal wave table position is identical to the table position of FIG. 3. Thus, in one wavelength (one period) of the drive frequency, the PCWM signal encoder outputs 720 pulses as the counting signal. For example, as a result of digital measurement, if the phase difference (measured power factor angle δP) between the voltage phase and the current phase is 36 of the above-described pulses, this is 5% (=36/720) of the wavelength, and therefore the measured power factor angle δP is −18° (=−360°×0.05).

(195) There exists the power factor angle value at which the efficiency becomes optimal for the varying load magnitude at the given number of rotations of the drive motor. A model follower adaptive control method is employed by adjusting the motor applied voltage to have the value of the counted power factor angle become the ideal value stored in the table. For this purpose, a load test is conducted on the motor in advance, to get the table shown in FIG. 27. The relationship between the voltage command VC and the target power factor angle δS shown in FIG. 27 is prepared for each of the certain frequency ranges, and stored in the target power factor angle table 105. Thus, the target power factor angle table 105 stores the target power factor angle (OS) to be targeted for the frequency (FC) and the voltage (VC) applied to the permanent magnet motor 64.

(196) As shown in FIG. 23, in use for a fan or a blower, a Hall sensor can be installed on the motor stator. However, for an air conditioner compressor application, the Hall sensor cannot be installed due to environmental restriction inside the compressor. Therefore, the current sensor shown in FIG. 30 is employed instead.

(197) Next, by referring to FIG. 30 which shows one embodiment of power factor angle control, further detailed explanation will be given. FIG. 30 specifically shows the constitution of the permanent magnet motor drive system 60 of the present embodiment shown in FIG. 26.

(198) In the present embodiment, the power factor angle sensor 67 and the power factor angle control block 111 perform the power factor angle measurement and control. However, as described previously, the power factor angle measurement and control are not always done during the entire motor drive. They are performed within certain motor speed ranges in which the motor enters into steady operation.

(199) The power factor angle control block 111 generates the voltage command VC based on the frequency command FC and the measured power factor angle δP to control the power factor angle.

(200) The frequency command FC supplied from the outside of FIG. 30, and the voltage command VC which is calculated on real time inside the figure and described later, are input to a target power factor angle determination block 103 inside the power factor angle control block 111. The target power factor angle determination block 103 determines the target power factor angle δS based on the frequency command FC and the voltage command VC by referring to the target power factor angle table 105. The target power factor angle table 105 is a set of values of power factor angles at which the efficiency becomes optimal, and which are obtained by conducting load test on the motor in advance. The target power factor angle table 105 is given in the format shown in FIG. 27 as described previously.

(201) The power factor angle sensor 67 in FIG. 30 subtracts the current phase signal δI output from the current sensor 144, from the voltage phase signal δV output from the PCWM signal encoder 116, which is described later, to obtain the measured power factor angle δP as the output.

(202) The power factor angle sensor 67 outputs the measured power factor angle δP of which measurement is made by the method described in FIGS. 28 and 29. That is, it subtracts the current phase signal (signal representing the phase of the terminal current) δI, from the voltage phase signal (signal representing the phase of the terminal voltage) δV, to obtain the output which is the measured power factor angle δP. The measured power factor angle δP is output to a power factor angle error calculator 109.

(203) The power factor angle error calculator 109 subtracts the measured power factor angle δP from the target power factor angle δS to obtain a power factor angle error OF. The power factor angle error OF is input to a voltage command accumulator 112 inside a voltage command generator 107.

(204) The voltage command generator 107 generates a voltage command VC. The voltage command generator 107 comprises a base voltage determination block 108, a V/F base voltage table 110, and the voltage command accumulator 112.

(205) The frequency command FC is input to the base voltage determination block 108. The base voltage determination block 108 determines a base voltage VB by referring to the V/F base voltage table 110. The V/F base voltage table 110 is a one obtained by conducting a load test on the motor in advance, and given in the format shown in FIG. 25. The base voltage VB output from the base voltage determination block 108, is input to the voltage command accumulator 112 having an addition/subtraction/storage function.

(206) At the time of entering into the power factor angle measurement and control loop, the voltage command accumulator 112 outputs the base voltage VB as the initial value of the voltage command VC, to the target power factor angle determination block 103 and the PCWM encoder 116. Thereafter, the voltage command accumulator 112 receives the power factor angle error OF from the power factor angle error calculator 109, and adjusts the voltage command VC based on the power factor angle error OF. Specifically, when the power factor angle error OF is plus, it means the voltage phase is delayed from the current phase beyond the target value, and the load is light. Therefore, it works so that the voltage command VC is decreased. Contrary, when the power factor angle error OF is minus, it means the voltage phase is advanced from the current phase beyond the target value, and the load is heavy. Therefore, it works so that the voltage command VC is increased.

(207) On the other hand, at the time of exiting from the power factor angle measurement and control loop, the voltage command accumulator 112 continues to renew the value held by itself toward the value of the base voltage VB so that the held value matches the base voltage VB in the end.

(208) A PWM signal generator 114 comprises the PCWM signal encoder 116, a PCWM signal decoder 128, and the sine wave 360° function table 120. For example, the PWM signal generator 114 can be realized as an LSI or an ASIC. A logic part DC voltage 138 is supplied to the PWM signal generator 114. The PWM signal generator 114 generates a PWM signal based on the frequency command FC and the voltage command VC.

(209) Here, the PWM signal generator 114 can be configured similarly to the ASIC 06 shown in FIGS. 1 and 2. However, in the present embodiment, the clocks CK1, CK3, CK4 and CK5 shown in FIG. 2 are made eight times longer. Namely, CK1, CK3, CK4 and CK5 are made as 1.6 μs, 409.6 μs, 409.6 μs and 29.4912 ms, respectively. Further, until the machine frequency FM reaches the frequency command (command frequency) FC, the process shown in FIG. 6 is conducted, and the power factor angle control is not conducted. After the machine frequency FM reaches the frequency command (command frequency) FC, the voltage command VC is input as the machine voltage VM to the PCWM signal encoder 116, and power factor angle control is conducted.

(210) The PCWM signal encoder 116 receives the frequency command FC and the voltage command VC as inputs, and receives data stored in the sine wave 360° function table 120 shown in FIG. 3 as a write signal 122 by a read signal 118 of a prescribed period. The PCWM signal encoder 116 outputs an encoded PCWM signal 126 by real time processing the write signal 122 and the information of the frequency command FC and the voltage command VC.

(211) The format of the sine wave 360° function table 120 is the same as the unit sine function table in FIG. 3. The table consists of 720 8-bit binary signals. 720 is selected as a multiple of 6. In the present embodiment, a three phase permanent magnet type motor is used, and therefore it is preferable to select a multiple of 6.

(212) A figure showing how the information of the sine wave 360° function table 120 is processed in the PCWM signal encoder 116, is the fractional sine function numeric nf representing the instantaneous amplitude value and the pulse width numeric pw of FIG. 4. It shows that after the inverter starts at the origin of the circle, the scanning speed increases with the operation point moving toward the outside as it is accelerated, and when it reaches the maximum speed, the operation point circulates on the outer periphery of the circle.

(213) A figure showing how the fractional sine function numeric nf in FIG. 4 is converted to the pulse width numeric pw, is the relationship between the pulse width numeric pw and the fractional sine function numeric nf of FIG. 5. However, as described above, in the present embodiment, CK1 and CK3 are made as 1.6 μs and 409.6 μs, respectively.

(214) As shown in the left side of FIG. 5, based on the calculation formula shown in FIG. 5, the encoded pulse width numeric pw in the unit PWM pulse interval is obtained, and this is output as the PCWM signal 126.

(215) The PCWM signal decoder 128 decodes the PCWM signal 126 input from the PCWM signal encoder 116 on real time as a PWM signal 130, and outputs it to the next stage inverter (gate drive) 132. The decoding method of the PCWM signal decoder 128 is described previously with reference to FIGS. 10-14.

(216) The inverter 132 generates a motor drive signal 134 based on the PWM signal 130. The inverter 132 can be configured similarly to the gate drive and power transistor circuitry 08 shown in FIGS. 1 and 2. The motor drive signal 134 output from the inverter 132 drives the three phase permanent magnet motor 64 which is directly coupled to the load 70. A main circuit DC voltage 140 is supplied to the inverter 132.

(217) The controller 62 applies the voltage to the permanent magnet motor 64, and therefore it knows the state of the terminal voltage at each time. The PCWM signal encoder 116 of the controller 62 outputs, among its outputs, the voltage phase signal δV which is an on/off signal of 50% duty representing the sine wave signal magnitude (zero or more, or less than zero) of the terminal voltage.

(218) In the case of FIG. 30, the current sensor 144 is installed on the inverter board of the controller 62, and is coupled to the drive signal 134 of the three phase permanent magnet motor 64 without contact by the Hall effect. The current sensor 144 converts the drive current (terminal current) signal of the three phase permanent magnet motor 64 to the current phase signal δI which is an on/off signal of 50% duty representing the magnitude (zero or more, or less than zero) of the signal by a comparator in the current sensor 144, and outputs the current phase signal δI.

(219) The PCWM signal encoder 116 outputs the counting signal (read signal) PCK which becomes on/off at every occurrence of the write signal 122. The counting signal PCK outputs as many pulses as the number of data stored in the sine wave 360° function table 120 during one period of the drive frequency signal regardless of the magnitude of the drive frequency. This can be called the second carrier which is synchronized to the period of the drive frequency. When the phase difference between the voltage phase signal δV and the current phase signal δI is digitally measured, the phase difference becomes an effective means as an index showing the ratio to the wavelength of the drive frequency signal.

(220) The power factor angle sensor 67 receives the voltage phase signal δV and the current phase signal δI as inputs, counts the phase difference between the two by the counting signal PCK, and outputs the resulting number of counts as the measured power factor angle δP.

(221) In this way, it is possible to count digitally what percentage of the wavelength at the current drive frequency the magnitude of the power factor angle becomes. Namely, the inverter 132 of the controller 62 applies the sine wave voltage to the permanent magnet motor 64. Further, the phase of the voltage is expressed in n=720 ways (see FIGS. 3 and 4). Further, the PCWM signal encoder 116 of the controller 62 sends n=720 counting signals PCK (pulses) to the power factor angle sensor 67 during one period (one wavelength) of the voltage. Then, the power factor angle sensor 67 measures the power factor angle by measuring the number of pulses which corresponds to the phase difference between the terminal voltage and the terminal current corresponds to. Thus, the power factor angle sensor 67 can count digitally what percentage of the wavelength at the current drive frequency the magnitude of the power factor angle becomes.

(222) In the present embodiment, the phase of the voltage is expressed in n=720 ways. However, another value (integer equal to or greater than 2) can be employed as n. Here, as the value of n, 6 or more is preferable. Specifically, a multiple of 6 which is 6 or more is preferable.

OTHERS

(223) In the above-described embodiments, a permanent magnet motor (three phase permanent magnet motor) is used as a synchronous motor, but the present disclosure can be applied to other synchronous motors. Further, an outer rotor type motor is used as a motor, but the present disclosure can be applied to an inner rotor type motor.

(224) Those having skill in this art will understand that many changes may be made to the details of the above-described embodiments without departing from the underlying principles of the present disclosure. Therefore, the scope of the present invention is determined only by the claims.