A dual harmonic current detection method for active power filters

Abstract

The application discloses a double-harmonic current detection method for an active power filter, and comprises the following steps: obtaining a first three-phase harmonic current and a second three-phase harmonic current to be compensated by respectively adopting a p-q method harmonic detection method based on instantaneous reactive power, and a harmonic detection method based on Prak transformation and sliding average filtering; giving weights to the two kinds of harmonic currents to obtain a final instruction current; taking the deviation between an actual output current and the instruction current as an input of a hysteresis comparator, generating a PWM modulation wave to control the switching action of a power device through the output of the hysteresis comparator, generating a compensation current opposite to the harmonic current on the load side, and compensating the harmonic current. According to the detection error, the dynamic weight is determined, the two kinds of harmonic detection methods are efficiently combined together, the detection precision of the load harmonic current is remarkably improved, and the compensation effect of the harmonic current caused by the nonlinear load is improved.

Description

Technical Field

[0001] This invention belongs to the field of three-phase parallel active power filter technology, specifically relating to a dual harmonic current detection method for active power filters. It is applicable to harmonic pollution problems in power systems and is mainly used in distributed generation systems. It is particularly suitable for harmonic pollution problems caused by the distortion of distributed generation supply current due to nonlinear loads used by power users. Background Technology

[0002] Distributed generation has a weaker load regulation capability compared to the main power grid, and current harmonics can easily cause power instability. In recent years, with the rapid development of power electronics technology, various power electronic devices have been widely used in power systems. Rectifiers, various switching devices, and other nonlinear loads have been connected to the grid and distributed generation systems in large quantities, leading to increasingly serious power quality problems such as harmonic pollution and reactive power, even threatening the safe and stable operation of the power system. Compared to passive filters, parallel active power filters have significant advantages in stability and flexibility, and are currently one of the effective ways to control harmonic pollution in power systems.

[0003] Harmonic current detection is a crucial component of an APF (Active Power Filter) system. The ability to quickly and accurately detect harmonic signals in the load current significantly impacts the power quality management effectiveness of the APF. Currently, harmonic current detection algorithms are mainly divided into two categories: frequency domain and time domain. Frequency domain detection methods include several analog filter methods, notch filter methods, and Fourier transform-based methods. Time domain detection methods mainly include those based on instantaneous reactive power, Fryze power theory, and conservative power theory.

[0004] Fourier analysis is a commonly used method for harmonic current detection. Its advantage lies in obtaining a compensation reference value for any harmonic current. However, its disadvantages include high computational complexity, poor real-time performance, and inability to detect dynamic signals and non-integer harmonics. Harmonic detection methods based on Fryze power theory cannot extract harmonic components independently. Since they are based on average power, an integration over one cycle is required to extract active current, resulting in a one-cycle delay. Harmonic detection methods based on instantaneous reactive power theory mainly include: pq detection method, i... p -i q The detection method includes the dq detection method and the pq detection method. The pq detection method can detect all harmonic currents when the three-phase voltage is symmetrical and undistorted. p -i q The detection method and the dq detection method have relatively small detection errors and a wide range of applications, but the detection accuracy mainly depends on the performance of the phase-locked loop and the low-pass filter, and has high requirements for hardware precision.

[0005] In addition, several novel methods have emerged in recent years and have been applied to harmonic current detection, such as wavelet transform, adaptive filtering, and neural networks. However, the universality and effectiveness of these methods still need to be verified through experiments. Summary of the Invention

[0006] Technical Problem Solved: The main objective of this invention is to improve the harmonic current detection accuracy of active power filters (APFs). To address the problem of poor harmonic compensation in existing APFs due to low harmonic current detection accuracy, this invention proposes an APF based on a dual harmonic current detection method. Simply combining two methods with fixed weights can lead to a large overall detection error when one method has a significant error. However, this invention selects two high-accuracy detection methods for harmonic currents caused by nonlinear loads. These two methods can share a single acquisition circuit, avoiding complex hardware structures. Furthermore, dynamic weights are calculated based on the rate of change of the detection errors of the two methods, effectively solving the aforementioned problem. This efficient combination of two harmonic detection methods significantly improves the detection accuracy of load harmonic currents, thereby enhancing the compensation effect for harmonic currents caused by nonlinear loads.

[0007] Technical solution:

[0008] A method for detecting dual harmonic currents in an active power filter, the method comprising the following steps:

[0009] S1, connect the LCL type active power filter to the distributed generation system to collect the distributed generation supply voltage v. abc Load current i L The compensation current i generated by the inverter c and inverter DC bus capacitor voltage V dc Set the reference value for the DC bus capacitor voltage. Subtract the collected DC bus capacitor voltage V from the reference value of the DC bus capacitor voltage. dc The error of the DC bus capacitor voltage is obtained, and the DC bus voltage adjustment amount is obtained through the PI controller;

[0010] S2 transforms the detected distributed generation supply voltage and load current to an αβ two-phase orthogonal coordinate system. Based on the instantaneous reactive power theory, the instantaneous active power p and instantaneous reactive power q are calculated. The fundamental active power is then obtained through a low-pass filter. and reactive power The harmonic power is then obtained by subtracting it from p and q respectively. Based on this harmonic power, the first three-phase harmonic current i to be compensated is calculated. habc1 ;

[0011] S3, the distributed generation supply voltage obtains phase wt through a three-phase phase-locked loop, the detected load current is transformed by Park to obtain d-axis and q-axis components, and then the dq-axis components of the load fundamental current are obtained through a moving average filter. The fundamental component of the load current is obtained through an inverse Park transformation, and the difference between the fundamental component and the load current is used to calculate the second three-phase harmonic current i to be compensated. habc2 ;

[0012] S4, respectively supplying the first and third phase harmonic currents i habc1 Second and third phase harmonic current i habc2 Assign weights and define the final command current as follows: k1 and k2 are the first three-phase harmonic currents i habc1 Second and third phase harmonic current i habc2 The weight values ​​are set, and k1+k2=1. k1 and k2 are updated in real time according to the deviation between the detected harmonic current signal and the actual output compensation current value. Through hysteresis control, the deviation between the actual output current and the command current is used as the input of the hysteresis comparator. The output of the hysteresis comparator generates a PWM modulation wave to control the switching action of the power device, generating a compensation current opposite to the harmonic current on the load side, and compensating for the harmonic current.

[0013] Furthermore, in step S2, the first three-phase harmonic current i to be compensated is calculated. habc1 The process includes the following steps:

[0014] S21, the distributed generation supply voltage and load current are transformed into a two-phase orthogonal coordinate system αβ using the following formula:

[0015]

[0016] The voltage v in the αβ coordinate system was calculated. α v β and current i Lα i Lβ :

[0017]

[0018] S22, Calculate the instantaneous active power p and instantaneous reactive power q according to the instantaneous reactive power theory:

[0019]

[0020] S23, after passing through a low-pass filter, obtains the fundamental power, and the difference between this fundamental power and the instantaneous active power p and instantaneous reactive power q yields the harmonic power. and Next, the components i of the harmonic current in the αβ coordinate system are calculated.hα and i hβ :

[0021]

[0022]

[0023]

[0024] S24, based on the transformation in step 23, finally obtains the first harmonic current signal i. habc1 :

[0025]

[0026] Furthermore, in step S3, the second and third phase harmonic currents i to be compensated are calculated. habc2 The process includes the following steps:

[0027] S31, the phase wt of the distributed generation supply voltage is obtained through a phase-locked loop circuit, and the load current in the abc coordinate system is transferred through c abc-dq Transforming to the dq coordinate system, we obtain the d-axis and q-axis components i of the load current. Ld and i Lq :

[0028]

[0029]

[0030] S32, respectively obtain i Ld and i Lq As the input to the moving average filter, the filtered output is used as the d-axis and q-axis components of the fundamental frequency of the load current:

[0031]

[0032]

[0033] in and is the filtered output, and T is the window width of the MAF.

[0034] Furthermore, in step S4, the update process for weight values ​​k1 and k2 includes the following steps:

[0035] S41, calculate the first and third phase harmonic currents i respectively. habc1 Second and third phase harmonic current i habc2 Deviation from actual output compensation current:

[0036]

[0037] Among them, i c σ(t) represents the actual compensation current at time t, and σ1(t) and σ2(t) represent the first three-phase harmonic currents i, respectively. habc1 Second and third phase harmonic currents i habc2 The deviation between the actual compensation current and the actual compensation current;

[0038] S42, differentiate the deviation values ​​calculated in step S41 to obtain the slopes η1(t) and η2(t) of the deviation values ​​as a function of time:

[0039]

[0040] S43, Update the weight values ​​k1 and k2 based on the slope of the deviation value calculated in step S42:

[0041]

[0042] Where k1(t) and k2(t) are the first three-phase harmonic currents i at time t, respectively. habc1 Second and third phase harmonic current i habc2 The corresponding weight values ​​are k1(t) + k2(t) = 1.

[0043] The present invention also discloses an active power filter based on the aforementioned dual harmonic current detection method. The active power filter includes a parallel LCL active power filter main circuit, a first harmonic detection module based on the pq method of instantaneous reactive power, a second harmonic detection module based on Prak transform and moving average filtering, a weighting coefficient calculation module, and a hysteresis controller.

[0044] The main circuit of the parallel-type LCL active power filter is connected to the distributed generation system, using the distributed generation supply voltage V. abc Load current i L The compensation current i generated by the inverter c and inverter DC bus capacitor voltage V dc Set the reference value for the DC bus capacitor voltage. Subtract the collected DC bus capacitor voltage V from the reference value of the DC bus capacitor voltage. dc The error of the DC bus capacitor voltage is obtained, and the DC bus voltage adjustment amount is obtained through the PI controller;

[0045] The first harmonic detection module transforms the detected distributed generation supply voltage and load current into an αβ two-phase orthogonal coordinate system. Based on the instantaneous reactive power theory, it calculates the instantaneous active power p and instantaneous reactive power q, and obtains the fundamental active power through a low-pass filter. and reactive power The harmonic power is then obtained by subtracting it from p and q respectively. Based on this harmonic power, the first three-phase harmonic current i to be compensated is calculated. habc1 ;

[0046] The second harmonic detection module performs a Park transform on the detected load current to obtain the d-axis and q-axis components, then uses a moving average filter to obtain the dq-axis components of the load fundamental current, performs an inverse Park transform to obtain the fundamental component of the load current, and finally calculates the second three-phase harmonic current i to be compensated by subtracting it from the load current. habc2 ;

[0047] The weighting coefficient calculation module respectively assigns weights to the first three-phase harmonic current i habcl Second and third phase harmonic current i habc2 Assign weights and define the final command current as follows: k1 and k2 are the first three-phase harmonic currents i habc1 Second and third phase harmonic current i habc2 The weight values ​​are determined by k1+k2=1, and k1 and k2 are updated in real time based on the deviation between the detected harmonic current signal and the actual output compensation current value.

[0048] The hysteresis controller uses the deviation between the actual output current and the commanded current as the input of the hysteresis comparator, and outputs a PWM modulation wave to control the switching action of the power device, generating a compensation current opposite to the harmonic current on the load side to compensate for the harmonic current.

[0049] Beneficial effects:

[0050] First, the dual harmonic current detection method for active power filters of the present invention employs digital filtering technology, using a moving average filtering algorithm instead of a traditional low-pass filter to more accurately filter out harmonic components in the load current and obtain the d-axis and q-axis components of the fundamental frequency of the load current. Simultaneously, by employing digital filtering and implementing the filtering function in code, it avoids complex hardware structures and delay issues.

[0051] Secondly, the dual harmonic current detection method for active power filters of the present invention assigns a dynamically changing weight value to the harmonic current signals obtained by the two detection methods, thereby obtaining the final harmonic current command signal. The weight value is determined according to the error change rate, which effectively improves the detection accuracy of the load-side harmonic current, resulting in better compensation effect of the APF. After compensation, the distortion rate of the distributed generation supply current is significantly reduced compared with the traditional harmonic current detection method. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the overall system structure of an active power filter employing the dual harmonic current detection method of the present invention.

[0053] Figure 2 This is a structural block diagram of the first harmonic detection module of the present invention.

[0054] Figure 3 This is a structural block diagram of the second harmonic detection module of the present invention.

[0055] Figure 4 This is a structural block diagram of the hybrid dual harmonic detection of the present invention.

[0056] Figure 5 This is a simulation diagram of the load current phase A waveform after a nonlinear load is connected to a distributed generation system.

[0057] Figure 6 The figure shows the FFT analysis results of the A-phase current of the distributed generation system without the active power filter of this invention.

[0058] Figure 7 The output A-phase tracking current waveform of the active power filter of the present invention is shown.

[0059] Figure 8 The simulation diagram shows the A-phase current waveform after the active power filter of the present invention is put into a distributed generation system.

[0060] Figure 9 The figure shows the FFT analysis results of the A-phase current of the distributed generation system after compensation based on the traditional harmonic current detection method.

[0061] Figure 10 The figure shows the FFT analysis results of the A-phase current after the active power filter of the present invention is put into a distributed generation system. Detailed Implementation

[0062] The following embodiments are provided to enable those skilled in the art to more fully understand the present invention, but do not limit the invention in any way.

[0063] A method for detecting dual harmonic currents in an active power filter, the method comprising the following steps:

[0064] S1, connect the LCL type active power filter to the distributed generation system to collect the distributed generation supply voltage v. abc Load current i L The compensation current i generated by the inverter c and inverter DC bus capacitor voltage V dc Set the reference value for the DC bus capacitor voltage. Subtract the collected DC bus capacitor voltage V from the reference value of the DC bus capacitor voltage. dc The error of the DC bus capacitor voltage is obtained, and the DC bus voltage adjustment amount is obtained through the PI controller;

[0065] S2 transforms the detected distributed generation supply voltage and load current to an αβ two-phase orthogonal coordinate system. Based on the instantaneous reactive power theory, the instantaneous active power p and instantaneous reactive power q are calculated. The fundamental active power is then obtained through a low-pass filter. and reactive power The harmonic power is then obtained by subtracting it from p and q respectively. Based on this harmonic power, the first three-phase harmonic current ih to be compensated is calculated. abc1 ;

[0066] S3, the distributed generation supply voltage obtains phase wt through a three-phase phase-locked loop, the detected load current is transformed by Park to obtain d-axis and q-axis components, and then the dq-axis components of the load fundamental current are obtained through a moving average filter. The fundamental component of the load current is obtained through an inverse Park transformation, and the difference between the fundamental component and the load current is used to calculate the second three-phase harmonic current i to be compensated. habc2 ;

[0067] S4, respectively supplying the first and third phase harmonic currents i habc1 Second and third phase harmonic current i habc2 Assign weights and define the final command current as follows: k1 and k2 are the first three-phase harmonic currents i habc1 Second and third phase harmonic current i habc2 The weight values ​​are set, and k1+k2=1. k1 and k2 are updated in real time according to the deviation between the detected harmonic current signal and the actual output compensation current value. Through hysteresis control, the deviation between the actual output current and the command current is used as the input of the hysteresis comparator. The output of the hysteresis comparator generates a PWM modulation wave to control the switching action of the power device, generating a compensation current opposite to the harmonic current on the load side, and compensating for the harmonic current.

[0068] The structure of the active power filter based on the dual harmonic current detection method of the present invention is as follows: Figure 1 As shown, the active power filter mainly includes a nonlinear load, a main circuit, and a control circuit.

[0069] The main circuit mainly includes:

[0070] Distributed generation system: mainly includes wind power generation, photovoltaic power generation and energy storage system; Three-phase inverter: converts the electrical energy stored in the energy storage element into a three-phase current with the same magnitude but opposite phase to the harmonic current on the load side through the input pulse signal; Energy storage element: energy storage capacitors are selected and connected to the DC side of the three-phase inverter to provide it with a stable DC voltage; Output filter circuit: adopts LCL topology, with the input connected to the AC side of the three-phase inverter and the output connected to the distributed generation system.

[0071] The control circuit mainly includes:

[0072] Voltage and current acquisition circuit: includes DC-side capacitor voltage detection circuit, distributed generation power supply voltage and current detection circuit, load current detection circuit, and inverter output current detection circuit; Harmonic current detection circuit: detects the harmonic current components contained in the acquired load current, and the output is connected to the current tracking control circuit; Current tracking control circuit: the input is connected to the harmonic current detection circuit, and the output is connected to the drive circuit, mainly to achieve accurate tracking of harmonic compensation current.

[0073] Before the LCL-type active power filter involved in this invention is connected to the distributed generation system, the waveform of the load current phase A is as follows: Figure 5 As shown in the figure, the FFT analysis results of the A-phase current of the distributed generation system are as follows: Figure 6 As shown. Figure 2 , 3 As shown in Figure 4, the dual harmonic current detection method of the present invention includes the following steps:

[0074] Step 1: First, use a three-phase voltage and current sensor to collect the distributed generation supply voltage V. abc Load current i L The compensation current i generated by the inverter c and inverter DC bus capacitor voltage V dc Set the reference value for the DC bus capacitor voltage. Subtract the collected DC bus capacitor voltage V from the reference value of the DC bus capacitor voltage. dc The error of the DC bus capacitor voltage is obtained, and then the DC bus voltage adjustment amount is obtained through the PI controller and input to the harmonic current detection module.

[0075] Step 2: Harmonic Detection Scheme 1: First, transform the detected distributed generation supply voltage and load current to an αβ two-phase orthogonal coordinate system. Based on the instantaneous reactive power theory, the instantaneous active power p and instantaneous reactive power q can be calculated. Then, the fundamental active power is obtained through a low-pass filter (LPF). and reactive power The harmonic power can be obtained by subtracting p and q respectively. Based on this power, the first three-phase harmonic current i to be compensated can be calculated. habe1 The specific implementation method is as follows:

[0076] Step a1: According to formula (1), the distributed generation supply voltage and load current are transformed to the αβ two-phase orthogonal coordinate system through a 32 transformation:

[0077]

[0078] Obtain the voltage v in the αβ coordinate system α vβ and current i Lα i Lβ As shown in formula (2):

[0079]

[0080] Step a2: Calculate the instantaneous active power p and instantaneous reactive power q according to the instantaneous reactive power theory, as shown in equation (3):

[0081]

[0082] Step a3: Obtain the fundamental power through a low-pass filter, and obtain the harmonic power by subtracting it from the instantaneous active power p and the instantaneous reactive power q. and Then, the components i of the harmonic current in the αβ coordinate system can be calculated. hα and i hβ As shown in equation (6):

[0083]

[0084]

[0085]

[0086] Finally, the first harmonic current signal i is obtained based on the 23 transformation. habc1 As shown in equation (7):

[0087]

[0088] Step 3: Harmonic Detection Scheme Two: The phase wt of the distributed generation supply voltage is obtained through a three-phase phase-locked loop. The detected load current is transformed by Park to obtain the d-axis and q-axis components. After passing through a moving average filter (MAF) stage, the dq-axis components of the load fundamental current are obtained. Then, the fundamental component of the load current is obtained through an inverse Park transform. Finally, the difference between the fundamental component and the load current is used to calculate the second three-phase harmonic current i to be compensated. habc2 .

[0089] The specific method for implementing the moving average filtering of the load current in this harmonic current detection method is as follows:

[0090] Step b1: Obtain the phase wt of the distributed generation supply voltage through a phase-locked loop circuit, and transfer the load current in the abc coordinate system through c abc-dq Transforming (Park transform) to the dq coordinate system, we can obtain the d-axis and q-axis components i of the load current. Ld and i Lq As shown in equation (9):

[0091]

[0092]

[0093] Step b2: The obtained i Ld and i Lq As the input to the moving average filter (MAF), the MAF function is shown in equations (10) and (11). The filtered output is the d-axis and q-axis components of the fundamental frequency of the load current.

[0094]

[0095]

[0096] in and is the filtered output, and T is the window width of the MAF.

[0097] Step 4: Assign a weight to the harmonic currents detected by the two schemes respectively, and define the final command current as... k1 and k2 are the weight values ​​of the two detection schemes, respectively, where k1 + k2 = 1. After obtaining the command current, the deviation between the actual output current and the command current is used as the input of a hysteresis comparator through hysteresis control. The output of the hysteresis comparator generates a PWM modulation wave to control the switching action of the power device, generating a compensation current opposite to the harmonic current on the load side, thus achieving harmonic current compensation. The specific method for determining the weights of the harmonic current signals obtained by the two detection methods is as follows:

[0098] Step c1: Calculate the deviation between the harmonic current signal detected by the two methods and the actual output compensation current. The specific calculation formula is as follows:

[0099]

[0100] Among them, i c (t) represents the actual compensation current, and σ1(t) and σ2(t) represent the deviations between the harmonic current signals detected by the two detection methods and the actual compensation current, respectively.

[0101] Step c2: After obtaining the formulas for calculating the deviation values ​​of the two methods, differentiate them respectively to obtain the slopes η1(t) and η2(t) of the deviation values ​​as a function of time, as shown in formula (13):

[0102]

[0103] Step c3: Based on the slope of the deviation values ​​of the two methods, define the weights of the two methods as follows:

[0104]

[0105] Where k1(t) and k2(t) are the weight values ​​of the two detection schemes, respectively, where k1(t) + k2 ( When t) = 1, if the slope of the deviation value of Scheme 1 is too large, the weight of Scheme 1 will be reduced, and the weight of Scheme 2 will be increased; similarly, if the slope of the deviation value of Scheme 1 is too large, the weight of Scheme 2 will be reduced, and the weight of Scheme 1 will be increased.

[0106] Step c4: The final harmonic current reference signal can be obtained based on the weighting coefficients of both. Specifically, as shown in formula (15):

[0107]

[0108] Finally, hysteresis control is used to track the compensation current. The tracking compensation current A-phase and harmonic command signal output by the active power filter are as follows: Figure 7 As shown; the compensated A-phase current waveform of the distributed generation system is as follows. Figure 8 As shown; Figure 9 and Figure 10 The figures shown are the FFT analysis results of the A-phase current of the distributed generation system after compensation based on the traditional harmonic current detection method and the FFT analysis results of the A-phase current after the active power filter of the present invention is put into the distributed generation system. It can be seen that the compensation effect is greatly improved.

[0109] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Furthermore, the embodiments described above are not intended to limit the present invention; various improvements and modifications can be made without departing from the spirit and scope of the invention, and all such improvements and modifications shall fall within the scope of the claimed invention.

Claims

1. A method for detecting dual harmonic currents in an active power filter, characterized in that, The dual harmonic current detection method includes the following steps: S1, Connect the LCL type active power filter to the distributed generation system to collect the distributed generation supply voltage. Load current The compensation current generated by the inverter and inverter DC bus capacitor voltage Set the reference value for the DC bus capacitor voltage. Subtract the collected DC bus capacitor voltage from the reference value of the DC bus capacitor voltage. The error of the DC bus capacitor voltage is obtained, and the DC bus voltage adjustment amount is obtained through the PI controller; S2 converts the detected distributed generation supply voltage and load current to... In a two-phase orthogonal coordinate system, based on the theory of instantaneous reactive power, the instantaneous active power is calculated. and instantaneous reactive power The fundamental active power is obtained through a low-pass filter. and reactive power Then, respectively with and The harmonic power is obtained by subtracting the harmonics, and the first three-phase harmonic current to be compensated is calculated based on this harmonic power. ; In step S2 above, the first three-phase harmonic current to be compensated is calculated. The process includes the following steps: S21, the distributed generation supply voltage and load current are transformed using the following formula through a 32-transformation process. In two orthogonal coordinate systems: ； Calculated Voltage in coordinate system , and current , : ； S22, Calculate the instantaneous active power based on the instantaneous reactive power theory. and instantaneous reactive power : ； S23, after passing through a low-pass filter, yields the fundamental power, which is then compared with the instantaneous active power. and instantaneous reactive power Harmonic power is obtained by subtraction. and Next, the harmonic current is calculated. Components in coordinate system and : ； ； ； S24, based on the transformation in step 23, finally obtains the first harmonic current signal. : ； S3, the distributed generation supply voltage obtains phase wt through a three-phase phase-locked loop, the detected load current is transformed by Park to obtain d-axis and q-axis components, and then the dq-axis components of the load fundamental current are obtained through a moving average filter. The fundamental component of the load current is obtained through an inverse Park transformation, and the difference between the fundamental component and the load current is used to calculate the second three-phase harmonic current to be compensated. ; In step S3 above, the second and third phase harmonic currents to be compensated are calculated. The process includes the following steps: S31, the phase wt of the distributed generation supply voltage is obtained through a phase-locked loop circuit, and the load current in the abc coordinate system is transferred through... Transforming to the dq coordinate system yields the d-axis and q-axis components of the load current. and : ； ； S32, respectively obtain and As the input to the moving average filter, the filtered output is used as the d-axis and q-axis components of the fundamental frequency of the load current: ； ； in and The output is the filtered value, and T is the window width of the MAF. S4, respectively supplying the first and third phase harmonic currents. Second and third phase harmonic currents Assign weights and define the final command current as follows: , and These are the first and third phase harmonic currents, respectively. Second and third phase harmonic currents The weight value, and , and The data is updated in real time based on the deviation between the detected harmonic current signal and the actual output compensation current value. Through hysteresis control, the deviation between the actual output current and the command current is used as the input of the hysteresis comparator. The output of the hysteresis comparator generates a PWM modulation wave to control the switching action of the power device, generating a compensation current opposite to the harmonic current on the load side, thereby compensating for the harmonic current. In step S4 above, the weight value and The update process includes the following steps: S41, calculate the first and third phase harmonic currents respectively. Second and third phase harmonic currents Deviation from actual output compensation current: ； in, Let be the actual compensation current at time t. and These are the first and third phase harmonic currents, respectively. Second and third phase harmonic currents The deviation between the actual compensation current and the actual compensation current; S42, differentiate the deviation values ​​calculated in step S41 to obtain the slope of the deviation values ​​as a function of time. and : ； S43, Based on the slope of the deviation value calculated in step S42, adjust the weight value... and Update: ； in, and The first three-phase harmonic currents at time t are respectively Second and third phase harmonic currents The corresponding weight value, .

2. An active power filter based on the dual harmonic current detection method of claim 1, characterized in that, The active power filter includes a parallel LCL active power filter main circuit, a first harmonic detection module based on the pq method of instantaneous reactive power, a second harmonic detection module based on Park transform and moving average filtering, a weighting coefficient calculation module, and a hysteresis controller. The main circuit of the parallel-type LCL active power filter is connected to the distributed generation system, using the distributed generation supply voltage as input. Load current The compensation current generated by the inverter and inverter DC bus capacitor voltage Set the reference value for the DC bus capacitor voltage. Subtract the collected DC bus capacitor voltage from the reference value of the DC bus capacitor voltage. The error of the DC bus capacitor voltage is obtained, and the DC bus voltage adjustment amount is obtained through the PI controller; The first harmonic detection module converts the detected distributed generation supply voltage and load current to... In a two-phase orthogonal coordinate system, based on the theory of instantaneous reactive power, the instantaneous active power is calculated. and instantaneous reactive power The fundamental active power is obtained through a low-pass filter. and reactive power Then, respectively with and The harmonic power is obtained by subtracting the harmonics, and the first three-phase harmonic current to be compensated is calculated based on this harmonic power. ; The second harmonic detection module performs a Park transform on the detected load current to obtain the d-axis and q-axis components, then uses a moving average filter to obtain the dq-axis components of the load fundamental current, performs an inverse Park transform to obtain the fundamental component of the load current, and finally calculates the second and third phase harmonic currents to be compensated by subtracting them from the load current. ; The weighting coefficient calculation module respectively assigns weights to the first three-phase harmonic currents. Second and third phase harmonic currents Assign weights and define the final command current as follows: , and These are the first and third phase harmonic currents, respectively. Second and third phase harmonic currents The weight value, and , and The data is updated in real time based on the deviation between the detected harmonic current signal and the actual output compensation current value. The hysteresis controller uses the deviation between the actual output current and the commanded current as the input of the hysteresis comparator, and outputs a PWM modulation wave to control the switching action of the power device, generating a compensation current opposite to the harmonic current on the load side to compensate for the harmonic current.

Images