Harmonic current detection method for parallel filter

Abstract

The invention discloses a parallel filter harmonic current detection method. The method specifically comprises the following steps: (1) adding a feedback link of a PI controller with energy feedback optimization behind a low-pass filter to obtain a harmonic component of a detection current, wherein the PI controller with energy feedback optimization is divided into a PI control unit and an energy feedback optimization processing unit; (2) three-phase load current passes through an output signal uc (k) of the PI control unit, the output signal uc (k) serves as an input excitation signal of an energy feedback unit and is processed through an energy feedback optimization processing unit, and then the output signal uc (k) and the output signal uc (k) of the PI control unit are combined to generate compensation current # imgabs0 #; and the output power driving module of the parallel filter outputs compensation current for harmonic compensation. The method has the beneficial effects that the method is used for detecting the harmonic current of the power distribution network containing the micro-grid, so that the anti-interference performance of harmonic detection of the electric energy quality controller is improved.

Description

Technical Field

[0001] The present invention relates to the technical field related to power distribution networks, and in particular to a parallel filter harmonic current detection method. Background Art

[0002] With the continuous increase in the penetration rate of microgrids, a large number of power electronic generating devices are connected to the distribution network. The operation of these devices can easily cause the distortion of the grid current and voltage waveforms, resulting in power quality problems such as harmonic pollution and reactive power loss in the distribution network. Power quality problems such as harmonics generated by nonlinear loads on the user side will also affect distributed power sources. As a comprehensive power quality regulation device, the power quality controller can be used to compensate for reactive power, harmonics, negative sequence current, etc. generated by nonlinear loads, and has become a research hotspot for many experts and scholars. For the distribution network harmonic detection method, it is mainly divided into Fourier transform method, wavelet transform method, harmonic detection algorithm based on instantaneous theory, adaptive filtering algorithm and other methods.

[0003] The Fourier transform method detects harmonic currents based on the discrete Fourier transform method and uses a triangular carrier comparison method for current tracking. The Fourier FFT algorithm is more suitable for loads that change slowly due to its delay, but it is difficult to be effective for microgrids with complex dynamic changes. The wavelet analysis method basically inherits all the advantages of the Fourier analysis method, but its disadvantage is that the amount of calculation is large. How to apply the wavelet transform to an asymmetric power system is an unresolved problem. According to an extensive and in-depth survey of existing literature, in the process of UPQC harmonic detection research, the most widely used method is based on instantaneous power theory, which now mainly includes the pq method and i p -i q Method.i p -i q This method is applicable to various situations such as grid voltage distortion and grid voltage asymmetry. Since various power devices in microgrids are easily affected by weather and other environmental factors, it is very easy to generate random, unstable, and fluctuating harmonic voltages, which increases the difficulty of harmonic control in distribution networks. For multivariable, nonlinear, time-varying control systems, traditional PI control parameters cannot be optimized as the state of the controlled object changes, resulting in poor control performance. Summary of the invention

[0004] The present invention aims to overcome the above-mentioned deficiencies in the prior art and provides a parallel filter harmonic current detection method capable of improving anti-interference performance.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A parallel filter harmonic current detection method specifically comprises the following steps:

[0007] (1) A feedback link of a PI controller with energy feedback optimization is added after the low-pass filter to obtain the harmonic component of the detection current, wherein the PI controller with energy feedback optimization is divided into a PI control unit and an energy feedback optimization processing unit;

[0008] (2) The three-phase load current passes through the output signal u of the PI control unit c (k), output signal u c (k) As the input excitation signal of the energy feedback unit, after being processed by the energy feedback optimization processing unit, it is compared with the output signal u of the PI control unit c (k) Combination generates compensation current

[0009] (3) After the hysteresis loop control, the PWM signal is output, and the output power drive module of the parallel filter outputs the compensation current to perform harmonic compensation.

[0010] i p -i q The harmonic current detection method overcomes the problem of detecting distorted grid voltage, which further improves the harmonic detection theory. p -i q The detection technology is improved by adding a feedback link of the PI controller before the parallel filter, and then the harmonic component of the detection current is obtained. For the power quality controller taking the microgrid into account to comprehensively manage the power quality, due to the intermittent and volatile nature of the photovoltaic array, its inverter is connected to the distribution network, which is prone to increase the harmonics of the power grid. The changes of various nonlinear and impact loads can easily lead to changes in the harmonics of the power grid. For multi-variable, nonlinear and time-varying power quality control systems of distribution networks, traditional PI control parameters cannot be optimized as the state of the controlled object changes, resulting in poor control performance. Therefore, the present invention proposes a PI control algorithm based on energy feedback to improve the harmonic current detection of the parallel filter APF. That is, the present invention proposes a harmonic current detection method based on energy feedback, by using the existing i p -i q Based on the proposed method, an energy feedback controller is inserted to detect the harmonic current of the distribution network containing the microgrid, thereby improving the anti-interference performance of the harmonic detection of the power quality controller.

[0011] Preferably, in step (2), the specific optimization principle of the PI controller with energy feedback optimization is:

[0012] Setting nonlinear function As the feedback optimization processing function in the energy feedback optimization processing unit, the voltage drop compensation instruction e(k) caused by the disturbance of the distributed photovoltaic panel and the PI control unit output control signal Δu c(k) Optimize the output auxiliary control adjustment signal F(Δu c (k)) is:

[0013]

[0014] In the formula, α and n are factor coefficients, α is an integer, n≥1; Δu c (k) is the control signal output by the PI control unit, which is used as the input excitation signal of the energy feedback optimization processing unit; the voltage drop compensation instruction satisfies Δe(k)=e(k)-e((k-1), k is the number of calculation steps; L1 is the control parameter, which represents the inverse sign of the product of the positive and negative properties of the control deviation and the positive and negative properties of the control deviation change rate; L2 represents the control parameter of the change direction of the control output; the two control parameters L1 and L2 are calculated using Δe(k), e(k), Δu c The information provided by (k) determines the value of 1 or -1, thus ensuring that F(Δu c (k)) changes towards the set value;

[0015] According to the discretization formula of the PI controller with energy feedback optimization in the parallel filter detection circuit:

[0016]

[0017] In the formula, K p0 Indicates the proportional gain when the step number is 0, T s represents the discrete sampling time, T0 represents the integration time when the step number is 0, and τ represents the continuous sampling time;

[0018] The auxiliary control adjustment signal F(Δu c (k)) and the control signal Δu output by the PI control unit c (k) The following formula is obtained by superposition:

[0019] Δu(k)=Δu c (k)+f(Au c (k))

[0020] where u c (k) is the output change value of the PI control unit, expressed as:

[0021]

[0022] Where, T s is the discrete sampling time, K P and T i are proportional gain and integral time respectively;

[0023] The output expression of the current harmonic compensation command after passing through the PI controller with energy feedback optimization is as follows:

[0024]

[0025] The output value enters the hysteresis control module to generate a PWM waveform, where T s is the discrete sampling time, K P and T i are proportional gain and integral time respectively. Due to the delay and lag of current harmonic signal, which leads to the lag and incompleteness of e(k) and Δe(k), the control parameter L1 in the above formula is improved to

[0026] L1=sign((e(k))·sign(Δe(k))

[0027]

[0028] Where ε is the adjustment threshold.

[0029] As a preferred embodiment, in the ideal state where the voltage of the distributed photovoltaic panel does not change, the voltage drop compensation instruction satisfies Δe(k) tends to zero, and at this time the feedback optimization processing function F(Δu c (k)) also tends to zero synchronously. The energy feedback optimization processing unit does not actually do any processing and is automatically shut down. Once the output voltage of the distributed photovoltaic panel is disturbed and the voltage drops instantly or the power load changes, the energy feedback optimization processing unit automatically intervenes and uses the feedback optimization processing function F(Δu c (k)) convergence, optimizing the output signal value of the PI controller with energy feedback optimization.

[0030] As a preference, in practical applications, in order to reduce the influence of the energy feedback optimization processing unit on the output of the PI controller containing energy feedback optimization and to improve the efficiency of circuit optimization, the influence coefficient of the energy feedback optimization processing unit is weighted, that is, the weight coefficient β = F(Δu c (k)), the feedback optimization processing function F(Δu c (k)) is weighted, and the weight coefficient β is:

[0031]

[0032] By using the above weight coefficient, the feedback optimization processing function F(Δu c (k)) intervention weight ratio, thereby accelerating the speed of current harmonic detection and achieving the purpose of rapid harmonic control.

[0033] The beneficial effect of the present invention is that it is used to detect harmonic currents of a distribution network including a microgrid, thereby improving the anti-interference performance of harmonic detection of a power quality controller. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the method for generating compensation current of the present invention;

[0035] Figure 2 It is the current control block diagram of the PI controller with energy feedback optimization. DETAILED DESCRIPTION

[0036] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0037] i p -i q The harmonic current detection method overcomes the problem of detecting distorted grid voltage, which further improves the harmonic detection theory. p -i q The detection technology is improved by adding a feedback link of the PI controller after the parallel filter, and then the harmonic component of the detection current is obtained. For the power quality controller taking the microgrid into account to comprehensively manage the power quality, due to the intermittent and volatile nature of the photovoltaic array, its inverter is connected to the distribution network, which is prone to increase the harmonics of the power grid. The changes of various nonlinear and impact loads can easily lead to changes in the harmonics of the power grid. For multi-variable, nonlinear and time-varying distribution network power quality control systems, traditional PI control parameters cannot be optimized as the state of the controlled object changes, resulting in poor control performance. Therefore, the present invention proposes a PI control algorithm based on energy feedback to improve the parallel filter APF harmonic current detection. The principle of this method is as follows Figure 1 As shown, the present invention proposes a harmonic current detection method based on energy feedback. p -i q Based on the proposed method, an energy feedback controller is inserted to detect the harmonic current of the distribution network containing the microgrid, thereby improving the anti-interference performance of the harmonic detection of the power quality controller.

[0038] Figure 1 Indicates compensation current The generation method of Figure 2 Refers to the current control block diagram of the PI controller with energy feedback optimization, the purpose is to control the compensation current of the parallel filter, Figure 1 and Figure 2 Need to combine description, Figure 2 i LF Represents the fundamental component of the three-phase load current, i af ,i bf ,icf ; Figure 2 i L That is, the three-phase load current i a ,i b ,i c ;Depend on Figure 1 As shown, the three-phase load current i a ,i b ,i c Through the PI controller of energy feedback, the final generated Right now Figure 2 Compensation current Figure 2 The control object in refers to the parallel filter.

[0039] exist Figure 2 The PI controller with energy feedback optimization is divided into a PI control unit and an energy feedback optimization processing unit. The output signal u of the PI control unit is c (k), after being processed by the energy feedback optimization processing unit, is combined with the output signal of the PI control unit to generate current The PWM signal is then output through hysteresis control, and the output power drive module of the parallel filter outputs a compensation current for harmonic compensation.

[0040] It specifically includes the following steps:

[0041] (1) A feedback link of a PI controller with energy feedback optimization is added after the low-pass filter to obtain the harmonic component of the detection current, wherein the PI controller with energy feedback optimization is divided into a PI control unit and an energy feedback optimization processing unit;

[0042] (2) The three-phase load current passes through the output signal u of the PI control unit c (k), output signal u c (k) As the input excitation signal of the energy feedback unit, after being processed by the energy feedback optimization processing unit, it is compared with the output signal u of the PI control unit c (k) Combination generates compensation current

[0043] The specific optimization principle of the PI controller with energy feedback optimization is:

[0044] Setting nonlinear function As the feedback optimization processing function in the energy feedback optimization processing unit, the voltage drop compensation instruction e(k) caused by the disturbance of the distributed photovoltaic panel and the PI control unit output control signal Δu c (k) Optimize the output auxiliary control adjustment signal F(Δu c(k) is: (wherein, the voltage drop compensation instruction e(k) refers to the control deviation, and a Δ on the left represents the control deviation change rate. The following formula uses the inverse sign of the product of the positive and negative properties of the control deviation and the positive and negative properties of the control deviation change rate to represent the control parameter L1, Δu c (k) Similarly)

[0045]

[0046] In the formula, α and n are factor coefficients, α is an integer, n≥1; Δu c (k) is the control signal output by the PI control unit, which is used as the input excitation signal of the energy feedback optimization processing unit; the voltage drop compensation instruction satisfies Δe(k)=e(k)-e(k-1), k is the number of calculation steps; L1 is the control parameter, which represents the inverse sign of the product of the positive and negative properties of the control deviation and the positive and negative properties of the control deviation change rate; L2 represents the control parameter of the change direction of the control output; the two control parameters L1 and L2 are calculated using Δe(k), e(k), Δu c The information provided by (k) determines the value of 1 or -1, thus ensuring that F(Δu c (k)) changes towards the set value;

[0047] According to the discretization formula of the PI controller with energy feedback optimization in the parallel filter detection circuit:

[0048]

[0049] In the formula, K p0 Indicates the proportional gain when the step number is 0, T s represents the discrete sampling time, T0 represents the integration time when the step number is 0, and τ represents the continuous sampling time;

[0050] The auxiliary control adjustment signal F(Δu c (k)) and the control signal Δu output by the PI control unit c (k) The following formula is obtained by superposition:

[0051] Δu(k)=Δu c (k)+f(Δu c (k))

[0052] where u c (k) is the output change value of the PI control unit, expressed as:

[0053]

[0054] Where, T s is the discrete sampling time, K P and T iare proportional gain and integral time respectively;

[0055] The output expression of the current harmonic compensation command after passing through the PI controller with energy feedback optimization is as follows:

[0056]

[0057] The output value enters the hysteresis control module to generate a PWM waveform, where T s is the discrete sampling time, K P and T i are proportional gain and integral time respectively. Due to the delay and lag of current harmonic signal, which leads to the lag and incompleteness of e(k) and Δe(k), the control parameter L1 in the above formula is improved to

[0058] L1=sign((e(k))·sign(Δe(k))

[0059]

[0060] Where ε is the adjustment threshold.

[0061] according to Figure 2 In the ideal state where the voltage of the distributed photovoltaic panel does not change, the voltage drop compensation instruction satisfies Δe(k) tends to zero. At this time, the feedback optimization processing function F(Δu c (k)) also tends to zero simultaneously, Figure 2 The energy feedback optimization processing unit in the system does not actually do any processing and is automatically shut down. Once the output voltage of the distributed photovoltaic panel is disturbed and the voltage drops instantly or the power load changes, the energy feedback optimization processing unit automatically intervenes and uses the feedback optimization processing function F(Δu c (k)) convergence, optimize the output signal value of the PI controller containing energy feedback optimization. However, in practical applications, in order to reduce the influence of the energy feedback optimization processing unit on the output of the PI controller containing energy feedback optimization and improve the efficiency of circuit optimization, the influence coefficient of the energy feedback optimization processing unit is weighted, that is, the weight coefficient β = F(Δu c (k)), the feedback optimization processing function F(Δu c (k)) is weighted, and the weight coefficient β is:

[0062]

[0063] When β approaches 0, the voltage drop amplitude approaches 0. When the β value is larger, the voltage drop amplitude is larger. Using the above weight coefficient, in the case of voltage drop, the feedback optimization processing function F(Δu c(k)) intervention weight ratio, thereby accelerating the speed of current harmonic detection and achieving the purpose of rapid harmonic control.

[0064] (3) After the hysteresis loop control, the PWM signal is output, and the output power drive module of the parallel filter outputs the compensation current to compensate the harmonics.

Claims

1. A parallel filter harmonic current detection method, characterized in that: The specific steps include: (1) A feedback link of a PI controller with energy feedback optimization is added after the low-pass filter to obtain the harmonic component of the detection current, wherein the PI controller with energy feedback optimization is divided into a PI control unit and an energy feedback optimization processing unit; (2) The three-phase load current passes through the output signal u of the PI control unit c (k), output signal u c (k) As the input excitation signal of the energy feedback unit, after being processed by the energy feedback optimization processing unit, it is compared with the output signal u of the PI control unit c (k) Combination generates compensation current (3) After the hysteresis loop control, the PWM signal is output, and the output power drive module of the parallel filter outputs the compensation current to perform harmonic compensation.

2. A parallel filter harmonic current detection method according to claim 1, characterized in that: In step (2), the specific optimization principle of the PI controller with energy feedback optimization is: Setting nonlinear function As the feedback optimization processing function in the energy feedback optimization processing unit, the voltage drop compensation instruction e(k) caused by the disturbance of the distributed photovoltaic panel and the PI control unit output control signal Δu c (k) Optimize the output auxiliary control adjustment signal F(Δu c (k)) is: In the formula, α and n are factor coefficients, α is an integer, n≥1; Δu c (k) is the control signal output by the PI control unit, which is used as the input excitation signal of the energy feedback optimization processing unit; the voltage drop compensation instruction satisfies Δe(k)=e(k)-e(k-1), k is the number of calculation steps; L1 is the control parameter, which represents the inverse sign of the product of the positive and negative properties of the control deviation and the positive and negative properties of the control deviation change rate; L2 represents the control parameter of the change direction of the control output; the two control parameters L1 and L2 are calculated using Δe(k), e(k), Δu c The information provided by (k) determines the value of 1 or -1, thus ensuring that F(Δu c (k)) changes towards the set value; According to the discretization formula of the PI controller with energy feedback optimization in the parallel filter detection circuit: In the formula, K p0 Indicates the proportional gain when the step number is 0, T s represents the discrete sampling time, T0 represents the integration time when the step number is 0, and τ represents the continuous sampling time; The auxiliary control adjustment signal F(Δu c (k)) and the control signal Δu output by the PI control unit c (k) The following formula is obtained by superposition: Δu(k)=Δu c (k)+f(Δu c (k)) where u c (k) is the output change value of the PI control unit, expressed as: Where, T s is the discrete sampling time, K P and T i are proportional gain and integral time respectively; The output expression of the current harmonic compensation command after passing through the PI controller with energy feedback optimization is as follows: The output value enters the hysteresis control module to generate a PWM waveform, where T s is the discrete sampling time, K P and T i are proportional gain and integral time respectively. Due to the delay and lag of current harmonic signal, which leads to the lag and incompleteness of e(k) and Δe(k), the control parameter L1 in the above formula is improved to L1=sign(e(k))·sign(Δe(k)) Where ε is the adjustment threshold.

3. A parallel filter harmonic current detection method according to claim 2, characterized in that In the ideal state where the voltage of the distributed photovoltaic panel does not change, the voltage drop compensation instruction satisfies Δe(k) tends to zero. At this time, the feedback optimization processing function F(Δu c (k)) also tends to zero synchronously. The energy feedback optimization processing unit does not actually do any processing and is automatically shut down. Once the output voltage of the distributed photovoltaic panel is disturbed and the voltage drops instantly or the power load changes, the energy feedback optimization processing unit automatically intervenes and uses the feedback optimization processing function F(Δu c (k)) convergence, optimizing the output signal value of the PI controller with energy feedback optimization.

4. A parallel filter harmonic current detection method according to claim 2 or 3, characterized in that: In practical applications, in order to reduce the influence of the energy feedback optimization processing unit on the output of the PI controller containing energy feedback optimization and to improve the efficiency of circuit optimization, the influence coefficient of the energy feedback optimization processing unit is weighted, that is, the weight coefficient β = F(Δu c (k)), the feedback optimization processing function F(Δu c (k)) is weighted, and the weight coefficient β is: By using the above weight coefficient, the feedback optimization processing function F(Δu c (k)) intervention weight ratio, thereby accelerating the speed of current harmonic detection and achieving the purpose of rapid harmonic control.

Images