METHOD FOR STARTING AND STOPPING AN ASYNCHRONOUS MOTOR

Abstract

A method for starting and stopping an asynchronous motor by way of a soft starter. The method includes the following steps: determining ignition options of one or more thyristors of the soft starter that are possible at a future calculation time; predicting the motor behavior for the determined ignition options, if an ignition of one or more thyristors of the soft starter is carried out; based on the predicted motor behavior, deciding whether an ignition option is to be selected and which is to be selected; and generating one or more ignition signals for one or more thyristors, if the decision for an ignition option has been made.

Claims

1-12. (canceled)

13. A method for starting and stopping an asynchronous motor by a soft starter, the method comprising the following steps: ascertaining firing opportunities available for one or more thyristors of the soft starter at a calculation time in the future; forecasting, for the firing opportunities thus ascertained, an electrical and/or mechanical motor behavior if a firing of one or more thyristors of the soft starter were performed; deciding, based on the motor behavior thus forecast, whether and which firing opportunity is to be chosen; and generating one or more firing signals for one or more thyristors if a decision has been made to take a firing opportunity.

14. The method according to claim 13, which comprises ascertaining the firing opportunities for a respectively following time step.

15. The method according to claim 13, which comprises modelling the motor behavior by a transient motor model.

16. The method according to claim 13, wherein the deciding step comprises making a decision based on one or more decision criteria selected from the group consisting of: maximum stator current, maximum torque, minimum on-time for thyristors, minimum average torque, maximum rotor flux, minimum rotor flux growth as a result of the firing, thyristor losses, and motor losses.

17. The method according to claim 13, which comprises not firing at an immediate next calculation time and awaiting the forecast for a calculation time that follows the immediate next calculation time if a decision is made that a firing opportunity is not useful.

18. The method according to claim 13, which comprises ascertaining only firing opportunities that result in negative torques that actively slow down the asynchronous motor.

19. The method according to claim 13, which comprises: repeatedly ascertaining a present rotor angle, a present rotor speed, a present grid phase angle, and present stator phase currents; using the present rotor angle, the present rotor speed, the present grid phase angle, and the present stator phase currents to anticipate a torque response for a torque acting on a rotor for first firing opportunities, in which a firing of thyristors turns on two phases, and for a second firing opportunity, in which firing of thyristors turns on three phases; and using the torque responses thus anticipated to decide whether each thyristor is fired.

20. The method according to claim 13, which comprises preceding the step of ascertaining firing opportunities with a step of magnetic flux generation in the rotor by way of firings of the thyristors.

21. The method according to claim 20, which comprises generating the magnetic flux in the rotor by way of two-phase firings of the thyristors at always the same grid phase angle.

22. A soft starter, comprising a firing signal unit and a device configured to carry out the method according to claim 13.

23. A computer program product, comprising computer-readable instructions that cause a soft starter with a firing signal unit and a processor to carry out the method according to claim 13.

24. A non-transitory computer-readable medium containing a computer program product with instructions for a processor of a soft starter to carry out the method steps according to claim 13.

Description

[0071] The above-described properties, features and advantages of this invention and the way in which they are achieved will become clearer and more clearly understood in association with the description below of the exemplary embodiments that are explained in greater detail in association with the figures. Here in schematic illustration:

[0072] FIG. 4 shows a soft starter according to one configuration of the invention;

[0073] FIG. 6a shows a start-up using the conventional starting algorithm with a constant load torque M.sub.L of 30 Nm;

[0074] FIG. 6b shows a start-up using an embodiment of the predictive starting algorithm according to the invention with a constant load torque M.sub.L of 30 Nm;

[0075] FIG. 7a shows a start-up using the conventional starting algorithm with a constant load torque M.sub.L of 70 Nm;

[0076] FIG. 7b shows a start-up using an embodiment of the predictive starting algorithm according to the invention with a constant load torque M.sub.L of 70 Nm; and

[0077] FIG. 8 shows a flowchart according to an embodiment of the method according to the invention.

[0078] FIG. 4 shows the structure of a soft starter 1 according to one configuration of the invention. The soft starter 1 may be used for example in a topology as shown in FIG. 1. The soft starter 1 has a control unit 41 having a computing unit 46, also referred to as a processor, and a storage unit 43. The computing unit 46 executes a computer program that is stored in the storage unit 43 and contains an algorithm for carrying out the method. Execution of the algorithm results in firing opportunities being ascertained for at least one next time step, a motor model being used to anticipate the electrical and/or mechanical motor behavior corresponding to the firing opportunities, and finally the forecast motor behavior being taken as a basis for deciding whether and which firing opportunity is supposed to be chosen. To initialize the algorithm, state variables of the system are measured or calculated. The computing unit 46 receives a series of measured values 44, e.g. the motor current I.sub.1,RMS, the motor voltage U.sub.1,RMS and the speed n of the rotor, as input values. Grid voltages U.sub.A, U.sub.B, U.sub.C are measured in order to calculate the grid angle φ.sub.grid and the grid voltage amplitude U.sub.grid. The motor currents i.sub.a, i.sub.b, i.sub.c and motor voltages u.sub.a, u.sub.b, u.sub.c are also used for a model-based calculation of the rotor flux. Together with the measured mechanical rotor speed n and the grid frequency f.sub.grid, all system variables are defined. After the prediction, it is clear which opportunities for firing there are at the time under consideration in the future, i.e which firing opportunities. These firing opportunities may then be checked for a multiplicity of decision criteria 45, e.g. a maximum torque or a maximum stator current, and rated. If the decision has been made for a specific firing opportunity, the controller 41 generates one or more initiation signals 47 to a firing block 42 of the soft starter, which has thyristors and a firing signal unit 48 for generating firing signals for the thyristors. The initiation signals 47 cause the firing signal unit 48 to generate firing signals for one or more thyristors, as a result of which the anticipated motor behavior is established.

[0079] A standard model for induction motors is used for prediction below as an exemplary embodiment of the method according to the invention. In principle, it will be pointed out here that it is by no means imperative for it to be the motor model shown here that is used to predict the behavior of the motor.

[0080] Simplified models that e.g. ignore factors such as leakage fluxes, allow for current displacement effects in the rotor or have variables other than state variables may likewise be used here, of course. To calculate two-phase firings, the model states need to be rotated in a suitable manner or the missing third conductor voltage needs to be calculated separately.

[0081] The subscript index 1 indicates that a value is a stator parameter and the subscript index 2 indicates that a value is a rotor parameter. The superscript index K indicates that a value has been rotated through the rotation angle φ.sub.K in the reference system:

[00004] ${\overset{.fwdarw.}{I}}_{1}^{K} = {\overset{.fwdarw.}{I}}_{1}^{S} \cdot e^{- j φ κ}$

[0082] The motor model comprises the stator resistance R.sub.1, the stator leakage inductances L.sub.1σ, the mutual inductance L.sub.h, the rotor resistance R.sub.2 and the rotor leakage inductances L.sub.2σ. For the sake of simplicity, the rotor parameters are referenced to the stator side. The electrical rotor speed is defined as Ω.sub.L and the rotation speed of the reference system as Ω.sub.K. The inductances are defined as:

[00005] $L_{1} = L_{1 σ} + L_{h}$

[00006] $L_{2} = L_{2 σ} + L_{h}$

[0083] Equations (6) and (7) show a model of a general induction motor, based on the rotor magnetic flux linkage

[00007] ${\overset{.fwdarw.}{Ψ}}_{2}^{K}$

and the stator magnetic flux linkage

[00008] ${\overset{.fwdarw.}{Ψ}}_{1}^{K}$

as state variables:

[00009] ${\overset{.fwdarw.}{U}}_{1}^{K} = R_{1} \cdot {\overset{.fwdarw.}{I}}_{1}^{K} + \frac{d {\overset{.fwdarw.}{Ψ}}_{1}^{K}}{d t} + j \cdot Ω_{K} \cdot {\overset{.fwdarw.}{Ψ}}_{1}^{K}$

[00010] ${\overset{.fwdarw.}{U}}_{2}^{K} = R_{2} \cdot {\overset{.fwdarw.}{I}}_{2}^{K} + \frac{d {\overset{.fwdarw.}{Ψ}}_{2}^{K}}{d t} + j (Ω_{K} - Ω_{L}) \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K}$

[0084] The magnetic flux linkage can be calculated on the basis of the stator and rotor currents:

[00011] ${\overset{.fwdarw.}{Ψ}}_{1}^{K} = L_{1} \cdot {\overset{.fwdarw.}{I}}_{1}^{K} + L_{h} \cdot {\overset{.fwdarw.}{I}}_{2}^{K}$

[00012] ${\overset{.fwdarw.}{Ψ}}_{2}^{K} = L_{h} \cdot {\overset{.fwdarw.}{I}}_{1}^{K} + L_{2} \cdot {\overset{.fwdarw.}{I}}_{2}^{K}$

[0085] The motor torque is defined as:

[00013] $M_{M} = \frac{3}{2} \cdot p \cdot {\overset{.fwdarw.}{Ψ}}_{1}^{K} \times {\overset{.fwdarw.}{I}}_{1}^{K}$

[0086] Assuming a squirrel-cage induction motor, a short circuit on the rotor side can be estimated:

[00014] ${\overset{.fwdarw.}{U}}_{2}^{K} = 0$

[0087] Additionally, there is no advantage in continuously rotating the reference frame for predictive calculations, as is often done in the motor controller. The rotation speed of the reference system is therefore set equal to zero:

[00015] $\frac{d φ_{K}}{d t} = Ω_{K} = 0$

[0088] For a compact description of the model, it is additionally useful to define the leakage factor σ:

[00016] $σ = 1 - \frac{L_{h}^{2}}{L_{1} \cdot L_{2}}$

[0089] From equation (9), it follows that:

[00017] ${\overset{.fwdarw.}{I}}_{2}^{K} = - \frac{L_{h}}{L_{2}} \cdot {\overset{.fwdarw.}{I}}_{1}^{K} + \frac{1}{L_{2}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K}$

[0090] In equations (6) and (7), the rotor magnetic flux linkage

[00018] ${\overset{.fwdarw.}{Ψ}}_{2}^{K}$

and the stator magnetic flux linkage

[00019] ${\overset{.fwdarw.}{Ψ}}_{1}^{K}$

are defined as state variables. Only the stator current are measured in the soft starter shown in FIG. 1. These currents are a system property that is necessary for initializing the prediction model. In order to reduce the effort for calculating the magnetic flux on the basis of the measured currents, it is advantageous to have the current space vector as a stator state variable. Equations (6) and (9) are therefore used to obtain the stator current

[00020] ${\overset{.fwdarw.}{I}}_{1}^{K}$

as a new system-dynamic state variable:

[00021] $\begin{array}{l} \frac{d {\overset{.fwdarw.}{I}}_{1}^{K}}{d t} = \frac{1}{σ \cdot L_{1}} \cdot {\overset{.fwdarw.}{U}}_{1}^{K} - \frac{R_{1}}{σ \cdot L_{1}} \cdot {\overset{.fwdarw.}{I}}_{1}^{K} + \\ \frac{R_{2} \cdot L_{h}}{σ \cdot L_{1} \cdot L_{2}} \cdot {\overset{.fwdarw.}{I}}_{2}^{K} - j \cdot \frac{Ω_{L}}{σ \cdot L_{1}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K} \end{array}$

[00022] $\frac{{\overset{.fwdarw.}{Ψ}}_{2}^{K}}{d t} = - R_{2} \cdot {\overset{.fwdarw.}{I}}_{2}^{K} + j \cdot Ω_{L} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K}$

[0091] To reduce the computation complexity, equation (14) is integrated into equations (15) and (16):

[00023] $\begin{array}{l} \frac{D {\overset{.fwdarw.}{I}}_{1}^{K}}{d t} = \frac{1}{σ \cdot L_{1}} \cdot {\overset{.fwdarw.}{U}}_{1}^{K} - \frac{R_{1} \cdot L_{2}^{2} - R_{2} \cdot L_{h}^{2}}{σ \cdot L_{1} \cdot L_{2}^{2}} \cdot {\overset{.fwdarw.}{I}}_{1}^{K} \\ \frac{R_{2} \cdot L_{h}}{σ \cdot L_{12} \cdot L_{2}^{2}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K} - j \cdot \frac{Ω_{L}}{σ \cdot L_{1}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K} \end{array}$

[00024] $\frac{d {\overset{.fwdarw.}{Ψ}}_{2}^{K}}{d t} = \frac{R_{2} \cdot L_{h}}{L_{2}} \cdot {\overset{.fwdarw.}{I}}_{1}^{K} - \frac{R_{2}}{L_{2}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K} + j \cdot Ω_{L} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K}$

[0092] Torque equation (10) is remodelled to produce the new system state variables using equations (8) and (9):

[00025] $M_{M} = \frac{3}{2} \cdot p \cdot \frac{L_{h}}{L_{2}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{K} \times {\overset{.fwdarw.}{I}}_{1}^{K}$

[0093] Equations (17), (18) and (19) represent the whole system model. For a 3-phase firing case, this model can be used to calculate the electrical and mechanical motor behavior by taking the grid voltage

[00026] ${\overset{.fwdarw.}{U}}_{g r i d}^{S}$

as stator voltage

[00027] ${\overset{.fwdarw.}{U}}_{1}^{K} \cdot$

[0094] It is likewise possible to calculate a 2-phase firing on the basis of this model, but additional effort needs to be made for this. If e.g. the thyristors in phases a and b are fired, the voltage across the thyristor in phase C needs to be computed in parallel. This calculation must satisfy the condition i.sub.c = 0 because the thyristor in phase c is still blocking the current. The calculation therefore becomes even more time-consuming because the voltage needs to be calculated in each prediction step.

[0095] To reduce this computation complexity, the following approach may be chosen. If the thyristors in two of the three grid phases are fired, they turn on, whereas the thyristor in the third phase remains off. Because the sum of the currents in all three phases must always be zero, the current flowing to the motor in one of the conductive phases needs to be identical to the current flowing away from the motor in the other conductive phase; the current levels in the two conductive phases are therefore exactly the same, only with opposite arithmetic signs. FIG. 5a and FIG. 5b relate to a 2-phase firing of phases a and b. In FIG. 5a, in a fixed α-β reference frame, it is evident that this condition leads to a pulsating current vector that varies in length but not in direction. Therefore, only the absolute value of the current vector

[00028] $|{\overset{.fwdarw.}{I}}_{1}^{K}|$

varies. FIG. 5b shows a graph containing the absolute values of the generated current vector.

[0096] If the direction is fixed and only the absolute value of the current state vector varies, it is no longer necessary to describe the current as a complex variable. By rotating all system state variables through an angle φ.sub.P that is dependent on the firing combination, the equations may be simplified. A prediction reference frame for this rotation is defined using the rotation angle φ.sub.P:

[00029] ${\overset{.fwdarw.}{I}}_{1}^{P} = {\overset{.fwdarw.}{I}}_{1}^{S} \cdot e^{- j φ P}$

[0097] System equations (17), (18) and (19) are defined in the prediction reference frame as follows:

[00030] $\begin{array}{l} \frac{d {\overset{.fwdarw.}{I}}_{1}^{P}}{d t} = \frac{1}{σ \cdot L_{1}} \cdot {\overset{.fwdarw.}{U}}_{1}^{K} - \frac{R_{1} \cdot L_{2}^{2} - R_{2} \cdot L_{h}^{2}}{σ \cdot L_{1} \cdot L_{2}^{2}} \cdot {\overset{.fwdarw.}{I}}_{1}^{P} + \\ \frac{R_{2} \cdot L_{h}}{σ L_{1} \cdot L_{2}^{2}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{P} - j \cdot \frac{Ω_{L}}{σ \cdot L_{1}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{P} \end{array}$

[00031] $\frac{d {\overset{.fwdarw.}{Ψ}}_{2}^{P}}{d t} = \frac{R_{2} \cdot L_{h}}{L_{2}} \cdot {\overset{.fwdarw.}{I}}_{1}^{P} - \frac{R_{2}}{L_{2}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{p} + j \cdot Ω_{L} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{P}$

[00032] $M_{M} = \frac{3}{2} \cdot p \cdot \frac{L_{h}}{L_{2}} \cdot {\overset{.fwdarw.}{Ψ}}_{2}^{P} \times {\overset{.fwdarw.}{I}}_{1}^{P}$

[0098] The rotation angle φ.sub.P is dependent on the firing combination in the thyristors. For the three possible 2-phase firing combinations: [0099] 1. firing of phases a and b, [0100] 2. firing of phases b and c, [0101] 3. firing of phases a and c, the rotation angles φ.sub.P in the prediction reference frame are:

[00033] $φ P, a - b = \frac{1}{6} π, φ P, b - c = \frac{1}{2} π, φ P, c - a = - \frac{5}{6} π$

[0102] This method has a few advantages compared to calculation of the third voltage. Firstly, the third additional voltage of the zero-current phase does not need to be calculated for each prediction step. Secondly, the motor model as such also becomes simpler because equation (17) does not need to be solved twice, once for the real part and once for the imaginary part. Owing to the rotation through the constant rotation angle φ.sub.P, only the real part in equation (17) is left for calculating the 2-phase firing. If one of the on states of one of the thyristors is changed, the states of equations (17) and (18) and the grid input voltage vector U.sub.grid need to be rotated through the new angle φ.sub.P.

[0103] Calculation of the electrical and mechanical motor behavior requires the actual rotor magnetic flux linkage

[00034] ${\overset{.fwdarw.}{Ψ}}_{2}^{K}$

as a state variable of the motor, see equation (18). As it is not possible to measure the rotor magnetic flux linkage

[00035] ${\overset{.fwdarw.}{Ψ}}_{2}^{K}$

using conventional hardware equipment of a soft starter, a magnetic flux model is required in order to calculate the rotor magnetic flux linkage. Therefore, equation (18) is implemented using the measured stator currents

[00036] ${\overset{.fwdarw.}{I}}_{1}^{K}$

as input value for tracking the rotor magnetic flux linkage. The value of the rotor magnetic flux linkage is also used in the next section to calculate a torque and a magnetic flux generation current (I.sub.1q and I.sub.1d). Together, they represent the stator current state vector, while φ.sub.K is equal to the angle of

[00037] ${\overset{.fwdarw.}{Ψ}}_{2}^{P} :$

[00038] ${\overset{.fwdarw.}{I}}_{1}^{K} = I_{1 q} + j I_{1 d}$

[0104] This approach can be used to simplify torque equation (10):

[00039] $M_{M} = \frac{3}{2} \cdot p \cdot Ψ_{1 d} \cdot I_{1 q}$

[0105] For all calculations, it is assumed that the rotor speed Ω.sub.L is constant up to the forecast horizon. This assumption results in acceptably small errors in the forecast, because the mass inertias of the motor and the load are usually large enough to keep the rotor speed stable up to the forecast horizon of less than 20 ms.

[0106] All firing opportunities are calculated in the prediction step. The different firing opportunities result in different time responses for the torque and the currents. Based on these time responses and further decision criteria, it is then necessary to decide whether or not a specific firing opportunity is useful.

[0107] The main reason for using a soft starter is the resultant possible limiting of current and torque, as a result of which the grid and the application are not damaged by the motor starting. The decision criteria should therefore include a maximum current amplitude i.sub.p,.sub.max and a maximum torque amplitude M.sub.p,.sub.max. Additionally, a minimum mean torque M.sub.p,.sub.avgmin ensures that only firing opportunities that accelerate the rotor are accepted. Down times in the firing hardware and inaccuracies in the voltage zero crossing detection mean that it is additionally useful to define a minimum on-time t.sub.p,mc for the thyristors. Firing opportunities that result in conduction ending before this minimum on-time t.sub.p,mc has elapsed are rejected.

[0108] Additionally, it is necessary to keep the amplitude of the rotor magnetic flux linkage at a useful level in order to have the opportunity to generate a torque using the stator currents. To this end, a minimum value for the rotor flux linkage Ψ.sub.p,free is defined in order to ensure a specific level of magnetic flux in the rotor. At this level, each firing needs to yield an increase in the rotor magnetic flux of Ψ.sub.p,Δmin • This increases the rotor magnetic flux linkage until the minimum value for the rotor flux linkage Ψ.sub.p,free has been reached again. Using these rules, it is possible to start up the induction motor.

[0109] When start-up of the induction motor begins, the rotor is idle and is not magnetized. To produce a torque, it is necessary for there to be a specific level of magnetic flux in the rotor. The thyristors are therefore fired with a 2-phase firing one or more times at the beginning of start-up in order to generate a defined level of rotor magnetic flux, see FIGS. 5a and 5b.

[0110] Measurements to demonstrate performance

[0111] The section below shows two measurements that illustrate a different behavior between the classic and predictive methods.

[0112] This measurement shows starting using a classic starting method. Here, the current passes through the predefined firing angle in a recurring symmetrical pattern. The speed also rises continuously.

[0113] This measurement shows starting using a predictive method. The form of the decision criterion means that the firing of the thyristors behaves completely differently.

[0114] First, the rotor is magnetized, characterized by uniform current pulses at the beginning. Then, after approximately 180 ms, the actual starting algorithm begins. Here, the current is unstable and occurs more or less “in packets”. The acceleration is also reminiscent of steps, in particular for low speeds. Allowing for the fact that forward losses occur in thyristors only when a current flows, it becomes clear that the losses in the thyristors are much smaller in the case of this starting than in the case of the comparable classic starting process. The same applies at least for the stator winding of the motor.

[0115] FIGS. 6a and 6b show a comparison of the behavior of a conventional algorithm and that of an algorithm according to the invention that run on a soft starter motor system as shown in FIG. 1. The measurements of the voltages and currents are taken using Hall sensors connected to an RCP system, comprising a CPU for handling the algorithm (CPU = central processing unit; RCP = rapid controller prototyping). The measured values shown were recorded using a commercially available multichannel power meter for measuring physical values. The rotor speed is measured using an inductive position sensor. Table 1 contains the motor data of the normal squirrel-cage induction motor that is used.

TABLE-US-00001 motor data for the standard model of the squirrel-cage induction motor that is used nominal power, P.sub.N 15 kW nominal voltage, U.sub.N 230 V nominal current, U.sub.N 28.5 A nominal frequency, ƒ.sub.N 50 Hz number of pole pairs, p 2 motor inertia, J.sub.M 0.0850 kgm.sup.2 load inertia,J.sub.L 0.0955 kgm.sup.2 stator DC resistance, R.sub.1 0.150 Ohm

[0116] The induction motor is mechanically connected to a permanent-magnet servo motor powered by a high-performance servo inverter. This inverter may be used to simulate mechanical loads having different characteristics. The moment of inertia from the servo load simulation is relatively low compared to typical industrial applications; this leads to relatively fast start-up of the motor and a few oscillations, in particular for the classic algorithm measurements in FIGS. 6a and 7a.

[0117] FIG. 6a shows a measurement for a start-up of a soft starter motor system, wherein a conventional algorithm is used and an application with a constant load torque M.sub.L of 30 Nm is actuated.

[0118] The voltage value is predefined by a ramp generator that begins at a relative starting voltage of 55% of the nominal motor voltage U.sub.N. The voltage is then increased with a continuous gradient of 25% per sec. The inertial mass of the motor and the load is accelerated using the difference between the motor torque M.sub.M and the load torque M.sub.L. This acceleration increases with the motor speed, i.e. the speed n, and has a more or less constant shape: the speed n increases continuously. The current profile i.sub.a, i.sub.b, i.sub.c is a constantly repeating sequence that differs only in the firing angle and a few oscillations. The current passes through the predefined firing angle in a recurring symmetrical pattern. Additionally, the d currents I.sub.1d and q currents I.sub.1q in the rotor magnetic-flux-oriented stator currents are indicated. These values are calculated from the measured stator currents and the reconstructed flux values of the rotor magnetic flux model. The current vector is divided into the field-generating current I.sub.1d and the torque-producing current I.sub.1q. Ignoring the oscillations on the currents, the field-generating current is higher than the torque-producing current. Owing to the more or less identical maximum current value per phase, the mean values of the d and q currents also remain constant.

[0119] Equation (26) shows that the present motor torque M.sub.M is dependent on the q current and the level of the rotor magnetic flux. Even if the q current remains at the same level, more or less torque can be produced by influencing the rotor magnetic flux. This is the reason for the rising acceleration at higher rotor speeds without a significantly higher q current occurring.

[0120] FIG. 7a shows a further start-up using the conventional algorithm, with a higher constant load torque M.sub.L of 70 Nm. The other constraints are as described for FIG. 6a.

[0121] FIG. 6b shows a measurement for a start-up of a soft starter motor system, wherein a prediction algorithm according to the invention is used. The load and grid conditions correspond to those of the conventional algorithm from FIG. 6a. The load torque M.sub.L is constantly 30 Nm.

[0122] Before the prediction algorithm starts, the start-up sequence has a special section added: first, the rotor is magnetized, characterized by uniform current pulses at the beginning. A few 2-phase firings take place in order to generate the rotor magnetic flux linkage This leads to a current space vector, the direction of which is rigid, but the amplitude of which pulsates, see FIGS. 5a and 5b, which continuously increases the amplitude of the rotor magnetic flux linkage Since only rotor magnetic flux is generated and no motor torque M.sub.M is produced, a pure d current is present in this section of the start-up sequence. In the time period from t=0 s to t=0.18 s, said current serves as a prerequisite for the production of torque in the next part of the start-up sequence.

[0123] The algorithm according to the invention then begins to run: here, the current is unstable and occurs more or less “in packets”. Compared to start-up using the conventional algorithm, there is no continuous firing profile, but rather time periods alternately with and without current. This also leads to a difference in the acceleration behavior. The acceleration is not as uniform as with a soft starter controlled by a conventional algorithm, but rather has a stepped shape, in particular for low speeds: there are time periods during start-up in which the acceleration is relatively high (time periods with a flow of current), and time periods without acceleration (time periods without a flow of current). Additionally, the d and q currents have a different shape. Here too, there are time periods with and without current.

[0124] In addition, the relationship between the levels of the d and q currents is different than that in the case of the conventional start-up. The d current is much smaller compared to the conventional algorithm. The d current is also smaller than during start-up using the conventional algorithm. This indicates that the linked rotor magnetic flux is higher when the prediction algorithm is used.

[0125] FIG. 7b shows a further start-up using the prediction algorithm, with a higher constant load torque M.sub.L of 70 Nm. The main parts of the start-up sequence are similar to those of the sequence using the prediction algorithm with the lower load torque M.sub.L of 30 Nm. In the time period from t=0 s to t=0.18 s, the absolute rotor magnetic flux linkage increases. The same pulsating behavior as in FIG. 6b is then evident. Starting at t=0.88 s up until synchronization at t=1.1 s, however, a behavior is visible that differs significantly from the behavior previously. In this time period of the start-up, the response changes and the current response is similar to the current response with the conventional algorithm. This is possible only because the requirements of the decision criteria are also met by the behavior with the conventional algorithm. Before this time period of the start-up, the firings using the conventional algorithm do not fit the decision criteria. For example, the criterion of minimum magnetic flux cannot be satisfied with a firing using the conventional algorithm. The prediction algorithm achieves the same result in a different way than the simpler firing-angle-based solution, because the conventional solution meets the requirements of the decision criteria in the last time period of the start-up.

[0126] When comparing the current curves of the conventional algorithm and the prediction algorithm according to the invention, it becomes clear that the prediction algorithm leads to a start-up with a shorter current conduction time. As a result, the losses in the thyristors of the soft starter and in the power lines of the induction motor are also lower. To assess the losses P.sub.thy in the thyristors, a simple model is used that is based on the forward voltage U.sub.f and the thyristor-internal resistance R.sub.on (thyristor dynamic resistance):

[00040] $P_{thy} (I_{thy}) = I_{thy} \cdot U_{f} + R_{o n} \cdot I_{thy}^{2}$

[0127] The parameters of the thyristor module used are summarized in table 2.

TABLE-US-00002 technical data for the thyristor module used forward voltage, U.sub.ƒ 0.9 V thyristor dynamic resistance. R.sub.on 2 mOhm

[0128] For the sake of simplicity, only the resistive stator losses are compared, because they can be calculated relatively easily using the resistance R.sub.1 of the stator windings.

[0129] Tables 3 and 4 provide the calculated losses, separated as thyristor losses and resistive stator losses, for a start-up of the motor with a constant load torque of 30 Nm and 70 Nm, for start-up using a conventional algorithm (classic start-up) and start-up using the predictive algorithm (predictive start-up). The values were calculated on the basis of the measurements shown in FIGS. 5a, 5b, 6a and 6b. The time period for the energy measurement starts with the first firing signal and ends when a steady-state current is reached.

TABLE-US-00003 calculated losses for the measurement with a constant load torque of 30 Nm Classic Start-up Predictive Start-up Thyristor losses 213 J 108 J Resistive stator losses 3537 J 1741 J

TABLE-US-00004 calculated losses for the measurement with a constant load torque of 70 Nm Classic Start-up Predictive Start-up Thyristor losses 414 J 209 J Resistive stator losses 7614 J 3585 J

[0130] Tables 3 and 4 show that the thyristor losses can be reduced by almost 50% under both load conditions using the predictive algorithm compared to the conventional algorithm. The resistive stator losses were also halved.

[0131] Both cases show that the invention makes it possible to reduce the losses in the soft starter and in the induction motor during the starting process. This allows more starts to be performed per unit time, cooling to be reduced and therefore smaller soft starters to be built and, if an application is started often, energy to be saved.

[0132] FIG. 8 shows a flowchart for a method comprising method steps S1 to S4 for operating an ASM with a soft starter.

[0133] A first method step S1 comprises ascertaining firing opportunities possible for one or more thyristors of the soft starter at a calculation time in the future. After the first method step S1, a second method step S2 is carried out.

[0134] The second method step S2 comprises forecasting for the ascertained firing opportunities an electrical and/or mechanical motor behavior if a firing of one or more thyristors of the soft starter were performed. After the second method step S2, a third method step S3 is carried out.

[0135] The third method step S3 comprises using the forecast motor behavior to decide whether and which firing opportunity is supposed to be chosen.

[0136] The fourth method step S4 comprises generating one or more firing signals for one or more thyristors if the decision has been made to take a firing opportunity. After the fourth method step S4, the method is continued with the first method step S1.

METHOD FOR STARTING AND STOPPING AN ASYNCHRONOUS MOTOR

Inventors

Cpc classification

Classification Explorer

H02P21/20

ELECTRICITY

Classification Explorer

H02P1/26

ELECTRICITY

Classification Explorer

H02P21/141

ELECTRICITY

Classification Explorer

H02P21/0017

ELECTRICITY

Classification Explorer

H02P27/024

ELECTRICITY

Classification Explorer

H02P2207/01

ELECTRICITY

Classification Explorer

H02P21/13

ELECTRICITY

International classification

Classification Explorer

H02P1/26

ELECTRICITY

Abstract

Claims

Description