Power factor correcting method

Abstract

A power factor correcting method for correcting a power factor in an alternating current power supply includes detecting zero-crossing points of an input voltage V.sub.AC of the AC power supply to obtain a period T of the input voltage V.sub.AC; generating a half-wave symmetric periodic function f(t) by using the period T; implementing a recursive low-pass filter y(t) by using the half-wave symmetric periodic function f(t), so that an output of the recursive low-pass filter y(t) matches a target reactive current I.sub.xcap(t) flowing through an X-capacitor; generating a reference current I′.sub.ref(t) for correcting the power factor by using a given reference current I.sub.ref(t) and the recursive low-pass filter y(t); and correcting the power factor by using the reference current I′.sub.ref(t). Accordingly, the reactive current of the EMI capacitor can be compensated in a high code efficiency manner.

Claims

1. A power factor correcting method for correcting a power factor of an alternating current (AC) power supply, comprising: detecting zero-crossing points of an input voltage V.sub.AC of the AC power supply to obtain a period T of the input voltage V.sub.AC; generating a half-wave symmetric periodic function f(t) by using the period T; implementing a recursive low-pass filter y(t) by using the half-wave symmetric periodic function f(t), so that an output of the recursive low-pass filter y(t) matches a target reactive current I.sub.xcap(t) flowing through an X-capacitor; generating a reference current I′.sub.ref(t) to correct the power factor by using a given reference current I.sub.ref(t) and the recursive low-pass filter y(t); and correcting the power factor by using the reference current I′.sub.ref(t).

2. The power factor correcting method according to claim 1, wherein an output of the half-wave symmetric periodic function f(t) is a square wave, a triangular wave or a trapezoidal wave.

3. The power factor correcting method according to claim 1, wherein the half-wave symmetric periodic function f(t) is generated using the following equation: $f\left(t\right)=-f\left(t-\frac{T}{2}\right).$

4. The power factor correcting method according to claim 1, wherein the half-wave symmetric periodic function f(t) is generated using the following equation: $f\left(t\right)=2.Math.2\left(\frac{t}{p}-\mathrm{floor}\left(\frac{t}{p}+\frac{1}{2}\right)\right).Math.-1,$ wherein p is a periodic time of a period preceding the period T.

5. The power factor correcting method according to claim 1, wherein an amplitude and a phase of the recursive low-pass filter y(t) match the target reactive current I.sub.xcap(t).

6. The power factor correcting method according to claim 1, further comprising adjusting an amplitude and a phase of the recursive low-pass filter y(t), so that the amplitude and the phase of the recursive low-pass filter y(t) match the target reactive current I.sub.xcap(t).

7. The power factor correcting method according to claim 6, wherein the adjusting the amplitude and the phase of the recursive low-pass filter y(t), so that the amplitude and the phase of the recursive low-pass filter y(t) match the target reactive current I.sub.xcap(t) comprises: adjusting a gain; re-implementing the recursive low-pass filter y(t) by using the adjusted gain and the f(t); adjusting the phase of the recursive low-pass filter y(t) to advance the input voltage V.sub.AC by a given threshold; and determining whether the amplitude and the phase of the recursive low-pass filter y(t) match the target reactive current I.sub.xcap(t).

8. The power factor correcting method according to claim 7, wherein the given threshold is in a range of about 85 degrees to about 95 degrees.

9. The power factor correcting method according to claim 7, wherein the given threshold is about 90 degrees.

10. The power factor correcting method according to claim 1, wherein the recursive low-pass filter y(t) is implemented by the following equation: $y\left(t\right)=\frac{\left(2^n-1\right)y\left(t-1\right)+A.Math.f\left(t\right)}{2^n},$ wherein n is a natural number and A is a given gain.

11. The power factor correcting method according to claim 1, wherein the target reactive current I.sub.xcap(t) is calculated by the following equation: $I_{xcap}=C_{xcap}\times \frac{d}{\mathrm{dt}}{V}_{AC}\left(t\right),$ wherein C.sub.xcap is a capacitance of the X-capacitor, $\frac{d}{\mathrm{dt}}$ is a time derivative, and V.sub.AC(t) is an instantaneous input voltage of the AC power supply.

12. The power factor correcting method according to claim 1, wherein when the given reference current I.sub.ref(t) is greater than an absolute value of the recursive low-pass filter y(t), the reference current I′.sub.ref(t) satisfies I′.sub.ref(t)=I.sub.ref(t)−y(t); and when the given reference current I.sub.ref(t) is equal to the absolute value of the recursive low-pass filter y(t), the reference current I′.sub.ref(t) is equals to zero.

13. The power factor correcting method according to claim 1, wherein the given reference current I.sub.ref(t) is calculated based on a loop output voltage, $\frac{1}{{\left(\mathrm{RMS}\left[V_{AC}\right]\right)}^2}$ and an instantaneous input voltage V.sub.AC(t) of the AC power supply.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 illustrates a schematic diagram of a power factor correcting configuration according to the prior art.

(2) FIG. 2 schematically illustrates an equivalent circuit diagram of main portions of a power supply including an X-cap.

(3) FIG. 3 illustrates a schematic diagram of a composite input waveform without X-cap compensation.

(4) FIG. 4 illustrates a schematic diagram of a composite input waveform with X-cap compensation.

(5) FIG. 5 illustrates a flowchart of generating a new reference current using X-cap compensation according to a preferred embodiment of the present invention.

(6) FIG. 6 schematically illustrates a diagram of a detailed graphical description of X-cap compensation according to a preferred embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(7) In order to facilitate understanding of the present invention, before describing preferred embodiments of the present invention, an existing active PFC will be described first.

(8) FIG. 1 illustrates a schematic diagram of a power factor correcting configuration according to the prior art. In FIG. 1, 101 denotes an alternating current power supply (AC power supply) that supplies power to the PFC, 102 denotes an EMI filter that reduces or prevents conductive noises (including common mode and differential mode), 103 denotes a PFC circuit, which changes a waveform of the current supplied to a load to increase the power factor, and 104 denotes a load corresponding to an output of the PFC. Further, a rectifier circuit is provided inside the PFC circuit 103.

(9) FIG. 2 schematically illustrates an equivalent circuit diagram of main portions of a power supply including an X-cap. A simplified EMI filter is shown in FIG. 2. An equivalent circuit of a differential input of the PFC is similar to a capacitor 202 before the PFC 203, because the common mode choke is not affected. In the circuit illustrated in FIG. 2, the current has a relationship shown in Equation 1 below:
I.sub.AC=I.sub.xcap+I.sub.L  [Equation 1]
wherein I.sub.AC denotes a current provided by AC power supply 201, I.sub.xcap denotes a reactive current flowing through the X-cap, and I.sub.L denotes a current flowing through the PFC.

(10) Generally, in the digitally controlled PFC, based on current loop control, I.sub.L should follow a reference current I.sub.ref generated by Equation 2 below:
I.sub.ref=A×B×C  [Equation 2]
wherein A denotes a loop output voltage that changes with the load, B denotes

(11) $\frac{1}{{\left(\mathrm{RMS}\left[V_{AC}\right]\right)}^2},$
and C denotes an instantaneous input voltage, that is, C=V.sub.AC(t). I.sub.ref denotes a function that changes over time.

(12) Accordingly, I.sub.L is in phase with V.sub.AC.

(13) In additional, I.sub.xcap may be calculated using Equation 3 below:

(14) $\begin{array}{cc}{I}_{xcap}=C_{xcap}\times \frac{d}{\mathrm{dt}}{V}_{AC}\left(t\right)& \left[\mathrm{Equation}3\right]\end{array}$
wherein, C.sub.xcap denotes a capacitance of the X-cap,

(15) $\frac{d}{\mathrm{dt}}$
denotes a time derivative, and V.sub.AC(t) denotes an instantaneous input voltage.

(16) FIG. 3 illustrates a schematic diagram of a composite input waveform without X-cap compensation. As illustrated in FIG. 3, I.sub.AC is ahead of V.sub.AC due to the effect of I.sub.xcap, which would result in a poor power factor calculated by Equation 4 below:

(17) $\begin{array}{cc}\mathrm{Power}\mathrm{factor}=\frac{\cos \theta }{\sqrt{1+{\mathrm{iTHD}}^2}}& \left[\mathrm{Equation}4\right]\end{array}$
wherein, cos θ denotes a phase shift power factor,

(18) $\frac{1}{\sqrt{1+{\mathrm{iTHD}}^2}}$
denotes a distortion power factor, and iTHD denotes a total harmonic distortion of the load current.

(19) In order to enable I.sub.AC to be in phase with the voltage input V.sub.AC of the AC power supply, I.sub.L may be modified to achieve a unit power factor. As mentioned above, a PFC current I′.sub.L compensated with X-cap should follow a reference current I′.sub.ref generated by Equation 5 below:
I′.sub.ref(t)=I.sub.ref(t)−I.sub.xcap(t)  [Equation 5]
wherein I′.sub.ref(t) denotes a new reference current that changes over time, and I.sub.xcap(t) denotes a reactive current flowing through the X-cap calculated by the microcontroller using Equation 3.

(20) FIG. 4 illustrates a schematic diagram of a composite input waveform with X-cap compensation. The results after performing the X-cap compensation are illustrated in FIG. 4. As illustrated in FIG. 4, composite I.sub.AC is in phase with V.sub.AC and a power factor is close to 1.

(21) Thus, it can be determined from the above that a key factor to implement the X-cap compensation is to generate, within the microcontroller, a reactive current flowing through the X-cap, that is, I.sub.xcap(t). Certainly, the reactive current I.sub.xcap(t) may also be detected by an isolated current meter, such as a Hall effect sensor, for example. In the present preferred embodiment, however, the reactive current I.sub.xcap(t) is preferably obtained by an algorithm in the microcontroller.

(22) In this regard, in Increase power factor by digitally compensating for PFC EMI-capacitor reactive current, http://www.ti.com/lit/an/slyt673/slyt673.pdf, it is proposed to use an internal random access memory (RAM) to continuously record V.sub.AC to obtain cos(2πft). However, in the method disclosed in Increase power factor by digitally compensating for PFC EMI-capacitor reactive current, http://www.ti.com/lit/an/slyt673/slyt673.pdf, a large amount of RAM storage space is occupied, and it is proportional to a sampling rate in a half period of V.sub.AC. Moreover, if a line frequency is a time-varying variable, then it can be determined from Equation 6 below that:
I.sub.xcap(t)=2πfV.sub.AC cos(2πft)  [Equation 6]
because the line frequency f is a variable, a periodic time between V.sub.AC zero-crossing points must be obtained, and then f is calculated from

(23) $f=\frac{1}{\mathrm{periodic}\mathrm{time}},$
the division calculation of which would take a long time in the microcontroller unit, so the microcontroller unit needs to use considerable computing resources to calculate I.sub.xcap(t).

(24) To this end, the inventor of preferred embodiments of the present invention, after research, has developed a new power factor correcting method. By using the methods according to preferred embodiments of the present invention to calculate the reference current, the reactive current of the EMI capacitor can be effectively compensated, and accordingly, the power factor can be improved and the current harmonic distortion can be reduced, and compared with the existing methods, code efficiency can be increased, and the microcontroller can generate a continuous reactive current of the EMI capacitor with less computing resources.

(25) The technical ideas and preferred embodiments of the present invention will be specifically described below.

(26) Power factor correcting methods according to preferred embodiments of the present invention are described below. By applying the power factor correcting methods according to preferred embodiments of the present invention to a reference current calculation to compensate for the reactive current of an EMI capacitor, the power factor of a switching AC power supply can be improved, and the current harmonic distortion can be reduced.

(27) FIG. 5 illustrates a flowchart of generating a new reference current using X-cap compensation according to a preferred embodiment of the present invention.

(28) In order to facilitate understanding of the present preferred embodiment and avoid repetitive descriptions, a portion of a current loop control flow is illustrated in FIG. 5, in which the calculation of I.sub.ref using Equation 2 is skipped. An output of the flowchart as illustrated in FIG. 5 is a new reference current I′.sub.ref, which is periodically updated after I.sub.ref.

(29) As illustrated in FIG. 5, in Step 501, V.sub.AC zero-crossing points are digitally detected from a AC power supply (referring to the AC power supply 101 of the circuit illustrated in FIG. 1 and the AC power supply 201 of the circuit illustrated in FIG. 2), to obtain a period of V.sub.AC and set the period of V.sub.AC to T. In Step 502, the period T is used to generate a periodic function with half-wave symmetry, that is, a half-wave symmetric function ƒ(t). Any type of periodic function may be generated in Step 502, as long as it has half-wave symmetry, such as, for example, square wave, triangular wave, trapezoidal wave, etc. The simpler the function is, the higher the code efficiency can be. For example, in the present preferred embodiment, the half-wave symmetric function ƒ(t) may be generated using Equation 7 below, and the period of the half-wave symmetric function generated in the present preferred embodiment is equal to the period of V.sub.AC:

(30) $\begin{array}{cc}f\left(t\right)=-f\left(t-\frac{T}{2}\right)& \left[\mathrm{Equation}7\right]\end{array}$
wherein T denotes the period of V.sub.AC.

(31) In Step 503, a recursive filter y(t) is applied to the half-wave symmetric function ƒ(t) to generate an approximately sine waveform. In the present preferred embodiment, the recursive filter y(t) is preferably a low-pass filter, for example. That is, the recursive low-pass filter y(t) is implemented using the half-wave symmetric function ƒ(t). It should be noted that y(t) can be any type of recursive filter using one or more outputs as input. For example, Equation 9 as described in the following may be used to obtain the recursive filter y(t), by a simple replacement of τ(t) in Equation 9 with ƒ(t). The characteristic of the recursive filter y(t) is to smooth the periodic function generated in Step 502 and is preferably capable of significantly attenuating relatively high frequency components after many V.sub.AC periods. In the present preferred embodiment, the obtained recursive filter y(t) can be represented as a sine wave with a baseline frequency, which looks like a reactive current I.sub.xcap flowing through the X-cap, but may have different amplitudes and phase shifts.

(32) When the amplitude and phase of the obtained recursive filter y(t) do not match the target I.sub.xcap, a complex mathematical method may be used to calculate a gain and phase shift to adjust the amplitude and phase of y(t) so as to match the amplitude and phase shift of I.sub.xcap.

(33) However, in the present preferred embodiment, steps 504, 505, 506, and 507 in FIG. 5 are used to provide a preferred and simpler iterative method to adjust the amplitude and phase shift of y(t).

(34) In the present preferred embodiment, in order to monitor y(t) and I.sub.ref, in Step 506, the microcontroller unit MCU may derive y(t), as well as I.sub.ref(t) which is calculated from the voltage of the AC power supply and the loop output voltage using the above Equation 2, and verify whether the amplitude and phase of y(t) match the target I.sub.xcap(t) calculated using the above Equation 3. If it is No in Step 506, then in Step 507, a gain is adjusted, and in Step 504, the adjusted gain is used as a multiplication coefficient to compensate for y(t) to adjust the amplitude of y(t). In Step 505, the phase of y(t) is adjusted to advance V.sub.AC by a given threshold. The amplitude and phase of y(t) can be approximately consistent with the target I.sub.xcap(t) through several iterations. The given threshold is preferably, for example, in a range of about 85 degrees to about 95 degrees, and more preferably is about 90 degrees.

(35) Similar to the prior art using Equation 5 as described above, in the present preferred embodiment, a new reference current I′.sub.ref(t) may be obtained through steps 508, 509, 510, and 511 shown in FIG. 5. Specifically, in Step 508, it is determined whether the target I.sub.ref(t) is greater than the absolute value of y(t). In this example, the comparison may be performed using the values of the target I.sub.ref(t) and y(t) at a current time point. If it is YES in Step 508, then as shown in Step 509, the reference current I′.sub.ref(t)=I.sub.ref(t)−y(t) is set, wherein I.sub.ref(t) is calculated by the microcontroller unit MCU in Step 506. If it is NO in Step 508, then it is determined in Step 510 whether the target I.sub.ref(t) is less than an absolute value of y(t). If it is NO in Step 510, then as shown in Step 511, the reference current I′.sub.ref(t)=0 is set. If it is YES in Step 510, then the current control cycle is repeated to update I′.sub.ref(t).

(36) Further, after optimizing the gain and phase shift to obtain a proper power factor and lower current harmonic distortion (iTHD), steps 504, 505, 506, and 507 in FIG. 5 may be omitted.

(37) In addition, in order to support an entire range of the AC input, several sets of gains and phase shifts may be applied discretely based on the line frequency and amplitude (e.g., about 230 Vac at 50 Hz or about 110 AC at 60 Hz).

(38) Further, in the preferred embodiments of the present invention, the half-wave symmetric periodic function ƒ(t) may include, for example, but not limited to, a square wave, a triangular wave, a trapezoidal wave, or the like. The digital low-pass filter y(t) may include, for example, but not limited to, a recursive or non-recursive filter.

(39) Next, a specific preferred embodiment of the present invention will be described.

(40) FIG. 6 schematically illustrates a diagram of a detailed graphical description of X-cap compensation according to a preferred embodiment of the present invention. In the present preferred embodiment, a triangular wave τ(t) is generated as an example of the half-wave symmetric periodic function ƒ(t) in the above-described preferred embodiments. For a mathematical expression of the triangular wave τ(t), it may be expressed as a trigonometric function τ(t) as shown in Equation 8 below:

(41) $\begin{array}{cc}\tau \left(t\right)=2.Math.2\left(\frac{t}{p}-\mathrm{floor}\left(\frac{t}{p}+\frac{1}{2}\right)\right).Math.-1& \left[\mathrm{Equation}8\right]\end{array}$
wherein τ(t) is in a range from −1 to +1. p is a periodic time of a period preceding V.sub.AC and is reset and updated at the end of the V.sub.AC period.

(42) In the present preferred embodiment, in order to use the microcontroller unit MCU more conveniently, the same input and output as those in Equation 8 are used to generate a triangular wave.

(43) First, zero-crossing points of V.sub.AC are detected to obtain the period of V.sub.AC. To adjust the phase, an instantaneous value of the trigonometric function τ(t) is reset to

(44) $\frac{p}{4}-s,$
wherein p is the periodic duration of a preceding period, which is reset and updated at the end of the V.sub.AC period, and s is a given phase shift, which is shown by the symbol 603 in FIG. 6. An upper limit (shown by the symbol 601 in FIG. 6) of the value of the trigonometric function τ(t) is set to

(45) 0$\frac{p}{4}.$
A lower limit (shown by the symbol 602 in FIG. 6) of the value of the trigonometric function τ(t) is set to

(46) $\left(\mathrm{the}\mathrm{upper}\mathrm{limit}-\frac{p}{2}\right).$
If a triangular wave at the current time is towards the lower limit, then

(47) $\tau \left(t\right)=\frac{p}{4}-s-t,$
wherein

(48) $0\le t<\frac{p}{2}-s;$
if the lower limit is reached, then

(49) $\tau \left(t\right)=s-\frac{3p}{4}+t,$
wherein

(50) $\frac{p}{2}-s\le t<p-s;$
if the upper limit is reached, then

(51) $\tau \left(t\right)=\frac{5p}{4}-s-t,$
wherein p−s≥t<p. Thus, a triangular wave is generated.

(52) Next, in the present preferred embodiment, a simple recursive low-pass filter R(t) can be obtained using Equation 9 below. R(t) is an example of y(t) in the present preferred embodiment:

(53) $\begin{array}{cc}R\left(t\right)=\frac{\left(2^n-1\right)R\left(t-1\right)+A.Math.\tau \left(t\right)}{2^n}& \left[\mathrm{Equation}9\right]\end{array}$
wherein n is a natural number and A is a given gain.

(54) In the present preferred embodiment, R(t) may be any type of digital recursive low-pass filter, which corresponds to y(t) in the present preferred embodiment.

(55) Next, the obtained recursive low-pass filter R(t) is applied to τ(t). If I.sub.ref(t)>|R(t)|, then I′.sub.ref(t)=I.sub.ref(t)−R(t); and if I.sub.ref(t)<|R(t)|, then I′.sub.ref(t)=0.

(56) I.sub.ref(t) and R(t) are derived to verify whether R(t) matches the amplitude and phase of the calculated target I.sub.xcap(t).

(57) After several iterations, an appropriate amplitude (shown by the symbol 605 in FIG. 6), phase (shown by the symbol 604 in FIG. 6), and n in Equation 9 can be obtained. In the present preferred embodiment, R(t) should be close to the calculated target I.sub.xcap(t).

(58) If a preferred embodiment of the present invention is applied to a 1600 watt AC-DC power supply unit of a 230 Vac, 50 Hz line, then result measurements of iTHD and power factor may be shown in Table 1 below.

(59) TABLE-US-00001 TABLE 1 Load Before After ratio iTHD Power factor iTHD Power factor 10 15.6 0.86 8.5 0.991 20 9.3 0.93 7.1 0.992 50 5.7 0.97 4.4 0.997 100 4.2 0.98 3.5 0.999

(60) Table 1 shows results before and after implementing a preferred embodiment of the present invention at different percentages of full load. It can be seen from Table 1 that significant improvements can be obtained after the implementation of the preferred embodiment, especially at light loads, since the X-cap current is dominant.

(61) By applying preferred embodiments of the present invention, compared with the prior art methods, iTHD and power factor can be improved equivalently to or better than the prior art. In addition, compared with the prior art methods, which require more RAM and computing resources, a method according to a preferred embodiment of the present invention can improve code efficiency, and can compensate for the reactive current of the EMI capacitor by the microcontroller unit MCU with less computing resources. For example, if 2000 updates occur in one AC period, then compared with the prior art methods, the method according to a preferred embodiment can save about 90% of RAM and 80% of computing time, for example.

(62) In addition, the algorithm in the above-described preferred embodiment can be implemented by, for example, but not limited to, a microcontroller unit MCU, a digital signal processor, a processor, an object, an executable program, an execution thread, a program, and/or a computer.

(63) While preferred embodiments of the present invention have been described above, it is to be understood that variations and modifications will be apparent to those skilled in the art without departing from the scope and spirit of the present invention. The scope of the present invention, therefore, is to be determined solely by the following claims.