Optimization control method applied to photovoltaic inverter of on-load capacity regulation distribution network

Abstract

The invention discloses an optimal control method applied to an on-load capacity-regulating distribution network photovoltaic inverter, which belongs to the technical field of distribution network photovoltaic inverter control, and comprises the following steps: S1, identifying the operation capacity and the network side equivalent impedance of an on-load capacity-regulating transformer; s2, detecting the voltage state quantity of a grid-connected node of the photovoltaic inverter; s3, detecting harmonic waves of the on-load capacity-regulating distribution network photovoltaic grid-connected system; and S4, harmonic resonance, voltage out-of-limit and a photovoltaic inverter of the distribution network are controlled. Aiming at photovoltaic inverter harmonic resonance and voltage out-of-limit multi-power quality control requirements of an on-load capacity-regulating distribution network area, according to a detected grid-connected point voltage amplitude, a voltage out-of-limit control compensation amount is generated; according to the extracted harmonic voltages, given harmonic currents are formed, and the dual functions of weak network harmonic resonance suppression and voltage out-of-limit compensation are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic inverter control in distribution networks, and in particular relates to an optimization control method applied to photovoltaic inverters in distribution networks containing load regulation and capacity distribution. Background Art

[0002] In the future, distributed renewable energy will gradually be connected to the distribution network on a large scale, and is expected to become an important energy source for the new power system. Most distributed renewable energy uses power electronic converters to connect to the distribution network. In the future, the distribution network will develop towards a high proportion of renewable energy generation and a high proportion of power electronic equipment (double high). In addition to the conventional safe and stable operation problems, the harmonic resonance problems caused by the double high characteristics and the voltage limit problems caused by the reverse flow, as unsafe operation phenomena derived from the new distribution network system, will bring huge challenges to the operation and control of the distribution network.

[0003] In terms of voltage over-limit, the document "The Impact of Distributed Photovoltaic Generation on Distribution Network Voltage and Solutions to Voltage Over-limit" points out from the photovoltaic power generation system itself that when a photovoltaic power generation system with high penetration rate is connected to the distribution network, it will generate a large amount of power under the condition of sufficient light intensity and suitable temperature, which will impact the voltage of the grid-connected node of the distribution network, and even cause voltage over-limit, resulting in reduced equipment life and burning. The document "The Impact of Distributed Photovoltaic Generation on Distribution Network Voltage and Solutions to Voltage Over-limit" uses formulas to calculate the node voltage fluctuation caused by photovoltaic access, and then verifies the impact of photovoltaic access on node voltage distribution, bus voltage fluctuation and terminal voltage increase from three perspectives of photovoltaic capacity, access distribution network location and power line length through simulation. It also proposes to use a photovoltaic inverter reactive control scheme to improve it. The document "Research on Static Voltage Stability and Reactive Control Strategy of Large Photovoltaic Power Stations" uses the eigenvalue analysis method to explore the voltage stability of the photovoltaic power generation system, obtains the minimum eigenvalue of voltage sensitivity to obtain the stability margin and criterion of voltage over-limit, and more effectively adjusts the grid-connected node voltage by analyzing the degree of reactive power regulation in the photovoltaic power station.

[0004] In terms of inverter harmonic resonance suppression, the paper "A Series-LC-Filtered Active Damper WithGrid Disturbance Rejection for AC Power-Electronics-Based Power Systems" proposed an active damper with a series LC filter connected in parallel at the PCC point to suppress the harmonic problem caused by the interaction between multiple inverters in parallel and the grid impedance. This active damper not only has a lower voltage level and a smaller inductance, but also can suppress harmonics at higher frequencies. The paper "Investigation and Active Damping of Multiple Resonances in a Parallel-Inverter-Based Microgrid" established a Norton equivalent circuit model in the discrete time domain to address the multiple resonance problems caused by multiple inverters in parallel, and used the virtual harmonic resistance active damping method to solve the instability phenomenon. The active damping method can effectively handle transient and steady-state resonances. The paper "Research on Adaptive Impedance Amplitude Reshaping Scheme for Differentiated Local Loads in Regional Microgrids" analyzes the impact of different local loads on the grid-side impedance. Aiming at the diversified operating characteristics caused by the access of differentiated local loads to the grid, a differentiated local load adaptive impedance amplitude reshaping scheme is proposed by distinguishing the grid-side impedance characteristics and considering multi-position feedback nodes.

[0005] In the patent document with the publication number CN108964120A, a low-voltage distributed photovoltaic access capacity optimization control method is disclosed. The method uses a controller and a capacity-adjusting and voltage-adjusting transformer to improve the optimization control of photovoltaic access capacity. The method solves the voltage positive deviation problem caused by the poor photovoltaic absorption capacity of the substation to a certain extent, improves the access capacity of low-voltage distributed photovoltaics in the low-voltage substation, and reduces the cost of construction and transformation of the power quality of the low-voltage substation. The general film is suitable for low-voltage substations containing distributed photovoltaics; in the patent document with the publication number CN108321810A, a distribution network for suppressing voltage fluctuations at photovoltaic grid-connected points is disclosed. The specific steps of the multi-time scale reactive power control method are as follows: first, the distribution network grid, regulation equipment parameters and the predicted values ​​and prediction errors of the load and photovoltaic output at each hour of the next day are collected; then, with the goal of minimizing the total operating cost of the distribution network, a distribution network reactive power optimization scheduling model based on chance constrained programming is established; then, the disaster genetic algorithm is used to solve the distribution network reactive power optimization scheduling model and issue scheduling plan instructions; finally, the load active, reactive and photovoltaic active output data are collected in real time, the real-time reactive adjustable capacity of the photovoltaic inverter is calculated, and the real-time reactive output of the inverter is controlled by comparing the reactive adjustable capacity and the reactive output target value.

[0006] As distributed renewable energy sources are connected to the grid from the end of the substation, the traditional low-voltage distribution network that supplies power to the load has gradually evolved into an active distribution network. At the same time, considering the fluctuation of the load in the substation, some substations are equipped with on-load capacity-adjusting transformers. Such substations will become more complex with the access of distributed energy. One is the sudden change of grid-side impedance caused by the switching process of the on-load capacity-adjusting transformer; the other is the dynamic interaction between the reverse flow and the source network impedance in the dynamic scenario of the on-load capacity-adjusting transformer switching after the access of distributed renewable energy sources. The combined influence of the above factors will lead to frequent voltage over-limit and harmonic resonance at the end of the on-load capacity-adjusting distribution network, which urgently needs to be managed. Summary of the invention

[0007] In order to solve the problem of sudden change of grid-side impedance caused by the switching process of on-load capacity-adjusting transformers in the substation area equipped with on-load capacity-adjusting transformers; at the same time, in order to solve the problem of dynamic interaction between reverse power flow and source-grid impedance in the dynamic scene of on-load capacity-adjusting transformer switching after the distributed new energy is connected to the substation area. The present invention provides an optimization control method applied to a photovoltaic inverter containing a load-adjusting capacity distribution network, which solves the multi-power quality lossless optimization control problems of harmonic resonance and voltage over-limit of photovoltaic inverters containing load-adjusting capacity distribution networks.

[0008] The technical solution adopted by the optimization control method of the present invention applied to a photovoltaic inverter with a load regulation and capacity distribution network is:

[0009] An optimization control method for a photovoltaic inverter in a load-adjusting capacity distribution network is characterized by comprising the following steps:

[0010] S1. Identify the operating capacity of the on-load capacity-changing transformer and the grid-side equivalent impedance;

[0011] S2, detecting the voltage state quantity of the photovoltaic inverter grid-connected node;

[0012] S3, detect the harmonics of photovoltaic grid-connected system including load regulation and capacity distribution;

[0013] S4: Control the harmonic resonance, voltage over-limit and photovoltaic inverter of the distribution network.

[0014] A further improvement of the technical solution of the present invention is that: in step S1, the operating capacity of the on-load capacity-adjusting transformer takes the operating loss and the number of switch switching of the on-load capacity-adjusting transformer as the preferred indicators, and calculates the optimal value of the capacity of the capacity-adjusting transformer under the two preferred indicators respectively, and the calculation formula is as follows:

[0015] P ZH =P OZH +(S / S NH ) 2 P KZH

[0016] P ZL =P OZL+(S / S NL ) 2 P KZL

[0017] Where P ZH , P OZH , P KZH , S NH They represent the comprehensive power loss of the large capacity of the on-load capacity-adjusting transformer, the comprehensive power no-load loss, the comprehensive power load loss and the rated capacity respectively; P ZL , P OZL , P KZL , S NL Respectively represent the comprehensive power loss of small capacity of on-load capacity-changing transformer, comprehensive power no-load loss, comprehensive power load loss and rated capacity;

[0018] Combining the above two equations, the theoretical optimal capacity adjustment point S of the on-load capacity-adjusting transformer can be calculated. c , the formula is as follows,

[0019]

[0020] A further improvement of the technical solution of the present invention is that the grid-side equivalent impedance of the on-load capacity-changing transformer in step S1 is calculated by injecting non-characteristic subharmonic voltage at the public access point and extracting the corresponding harmonic corresponding current, and the grid characteristic impedance is calculated as follows:

[0021]

[0022] In the formula, ω hx is the angular frequency of harmonics injected into the power grid; V ghx and θ v are the amplitude and phase of the common access point voltage at the injection frequency; I hx and θ i They are respectively the amplitude and phase of the grid-side harmonic response current under the harmonic voltage excitation extracted after the harmonic voltage frequency is injected.

[0023] A further improvement of the technical solution of the present invention is that the linearization calculation formula of the photovoltaic inverter grid-connected node voltage signal in step S2 is as follows:

[0024] y(t)≈B1cos(ω0t)-B2sin(ω0t)-B3tcos(ω0t)-B4tsin(ω0t)+B5t 2 cos(ω0t)

[0025] +B6t 2 sin(ω0t)-B7t 3 cos(ω0t)+B8t 3sin(ω0t)-B9t 4 cos(ω0t)-B 10 t 4 sin(ω0t)

[0026] In the formula, each coefficient B 1～10 When the signal is unknown, the coefficient B is unknown and needs to be calculated using the least squares method.

[0027] A further improvement of the technical solution of the present invention is that the calculation formula for estimating the voltage signal parameter of the photovoltaic inverter grid-connected node in step S2 is as follows:

[0028] The Hall element is used to measure the AC voltage V at the grid point pcc Sampling is performed, where the sampling frequency is f s , the sampling interval is t s =1 / f s , the sampling point is an odd number, expressed as 2N+1;

[0029] Y=XB

[0030] B=(X T W T WX) -1 X T W T WY

[0031] In the formula, the Y matrix is ​​obtained after the signal is sampled, and the middle time of the data window is used as the reference time. In the X matrix, let c(n) = cos(ω0nt s ), s(n) = sin(ω0nt s ), where the weight matrix W = diag{w -N ,…,w0,…,w N};

[0032] The coefficient matrix B corresponding to each data window is obtained by performing the least squares calculation on each data window, and the amplitude A(t) and phase angle θ(t) of the signal at time t are calculated according to the following formula:

[0033]

[0034] A further improvement of the technical solution of the present invention is that the calculation method for decomposing and reconstructing the detected harmonic signal in step S3 is as follows:

[0035] Using the multi-scale algorithm of wavelet transform, the fundamental wave and each harmonic component in the sinusoidal signal are obtained by decomposing the harmonic signal multiple times, and then the decomposed signal is calculated to obtain the effective value and phase parameters of the corresponding signal. The reconstructed form of the signal is:

[0036]

[0037] Where a is the expansion factor, h0(t) and h1(t) are the two sets of filtering signals of the low-pass filter and the high-pass filter.

[0038] A further improvement of the technical solution of the present invention is that the harmonic calculation formula in step S3 is as follows:

[0039] For a frequency of f0, amplitude of C, and initial phase of The single frequency signal x(t) is used to sample the AC harmonic voltage signal of the grid connection point using Hall elements, where the sampling frequency f s , uniformly sampled, and through fast Fourier decomposition, the discrete time signals of each harmonic voltage are obtained:

[0040]

[0041] The correction formula for frequency, initial phase and amplitude is:

[0042] f0=kΔf=(ε+k p +0.5)Δf

[0043]

[0044] C=N -1 (X(k p-1 Δf)-2X(k p Δf)+2X(k p+1 Δf)-X(k p+2 Δf))g(ε)

[0045] Where Δf = f s / N, k=0, 1, 2..., N-1.

[0046] A further improvement of the above technical solution of the present invention is that the control method of the photovoltaic inverter in step S4 is as follows:

[0047] According to the grid-side equivalent impedance of the on-load capacity-adjusting transformer and the real-time power generation capacity P of the photovoltaic inverter g Calculate the real-time short-circuit ratio of SCR,

[0048]

[0049] If the grid-connected system short-circuit ratio SCR is less than 3, it is in weak grid mode, and the inverter control method adopts grid-connected control mode without resonance problem;

[0050] If the short-circuit ratio SCR of the grid-connected system is greater than 3, the inverter adopts the grid-following control mode.

[0051] A further improvement of the technical solution of the present invention is that the voltage over-limit control method in step S4 is as follows:

[0052] According to step S2, the voltage amplitude V of the photovoltaic inverter grid connection point is detected. pcc and with the rated reference voltage V ref Compare and generate voltage over-limit control compensation through the controller, which is added to the inverter q-axis current setting to achieve voltage compensation.

[0053] i qref =G PI (V ref -V pcc ).

[0054] A further improvement of the technical solution of the present invention is that: the harmonic resonance control method in step S4 is as follows:

[0055] According to step S3, each harmonic voltage V is extracted th , divide each harmonic voltage by the virtual resistance to form each harmonic current setting, and superimpose it on the dq axis current setting to achieve the weak network harmonic resonance suppression function.

[0056]

[0057] Due to the adoption of the above technical solution, the technical advances achieved by the present invention include:

[0058] Aiming at the requirements of on-load capacity-regulating distribution network operation scenario division and impedance identification, the present invention determines the optimal capacity adjustment node of the capacity-regulating transformer and divides the operation scenarios through the excellence value of the on-load capacity-regulating transformer; and calculates the complex frequency domain components of the public access point voltage and the grid-side current at the injected harmonic frequency to obtain the grid impedance.

[0059] Aiming at the harmonic detection demand of photovoltaic grid-connected system in on-load capacity regulation distribution network area, the present invention utilizes the multi-scale algorithm of wavelet transform to obtain the fundamental wave and each harmonic component in the sinusoidal signal through multiple decompositions, and calculates the decomposed signal to obtain parameters such as the effective value and phase of the signal.

[0060] The present invention aims at the multi-mode switching requirements of photovoltaic inverters in the on-load capacity distribution grid area, calculates the real-time short-circuit ratio according to the detected grid-side equivalent impedance and the real-time power generation capacity of the inverter, and determines the inverter working mode. In the extremely weak grid mode, the inverter control method adopts the grid-building control mode, and improves the voltage over-limit problem by giving an amplitude through the grid-building reactive control loop. In the scenario between strong and weak grids, the inverter adopts grid-following control, superimposing multiple power quality additional control quantities to ensure stable operation of the system.

[0061] Aiming at the requirements of harmonic resonance and voltage over-limit control of photovoltaic inverters in load-controlled capacity distribution grid areas, the present invention generates a voltage over-limit control compensation amount according to the detected grid connection point voltage amplitude; and forms a given harmonic current according to the extracted harmonic voltages, thereby realizing the dual functions of weak grid harmonic resonance suppression and voltage over-limit compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic diagram of the overall process of an optimization control method of the present invention applied to a photovoltaic inverter having a load regulation and capacity distribution network;

[0063] Figure 2 The present invention is a circuit connection diagram of an optimization control method applied to a photovoltaic inverter having a load-regulating capacity distribution network. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail below in combination with specific implementation methods and with reference to the accompanying drawings. In the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0065] The present invention provides an optimization control method for a photovoltaic inverter having a load regulation capacity distribution network, such as Figure 1 As shown, the following steps are included:

[0066] S1. Identify the operating capacity of the on-load capacity-changing transformer and the grid-side equivalent impedance.

[0067] In the above step S1, the operating loss and the number of switching times of the capacity regulating transformer are used as the preferred indicators, and the optimality values ​​of the capacity regulating transformer capacity under the two preferred indicators are calculated respectively.

[0068]

[0069] Where P ZH , P OZH , P KZH , S NH Respectively represent the comprehensive power loss of the transformer's large capacity, comprehensive power no-load loss, comprehensive power load loss, and rated capacity; P ZL , P OZL , P KZL , S NL They respectively represent the comprehensive power loss of the transformer with small capacity, the comprehensive power no-load loss, the comprehensive power load loss and the rated capacity.

[0070] Combining the above two equations, we can calculate the theoretical optimal capacity adjustment point S of the on-load capacity-adjusting transformer. c , the formula is as follows,

[0071]

[0072] By injecting non-characteristic subharmonic voltage at the public access point and extracting the corresponding harmonic current, the Goertzel algorithm is used to calculate the grid characteristic impedance. The formula is as follows:

[0073]

[0074] In the formula, ω hx is the angular frequency of harmonics injected into the power grid; V ghx and θ v are the amplitude and phase of the common access point voltage at the injection frequency; I hx and θ i They are respectively the amplitude and phase of the grid-side harmonic response current under the harmonic voltage excitation extracted after the harmonic voltage frequency is injected.

[0075] S2. Detect the voltage state quantity of the grid-connected node of the photovoltaic inverter.

[0076] The above step S2 includes the following parts:

[0077] (1) Generation of voltage signal. Generally, the phase angle modulation signal is taken as the electrical signal when the system oscillates, which can be expressed as y(t):

[0078] y(t)=Acos(ω0t+θ(t)) (4)

[0079] Where ω0 is the fundamental angular velocity of the electrical signal, A is the signal amplitude, and θ(t) is the signal phase angle:

[0080] θ(t)=θ0+k b cos(ω b t+θ b ) (5)

[0081] k b is the phase modulation factor, ω b is the phase modulation angular frequency, θ b is the initial phase of phase angle modulation, A is the signal fundamental amplitude, and θ0 is the initial phase of the signal. The above formula can be written as:

[0082] y(t)=Acos(ω0t+θ0+k b cos(ω b t+θ b )) (6).

[0083] (2) Signal linearization. The signal is linearized by Taylor series expansion, and we get:

[0084] y(t)≈B1cos(ω0t)-B2sin(ω0t)-B3tcos(ω0t)-B4tsin(ω0t)+B5t 2 cos(ω0t)

[0085] +B6t 2 sin(ω0t)-B7t 3 cos(ω0t)+B8t 3 sin(ω0t)-B9t 4 cos(ω0t)-B 10 t 4 sin(ω0t)(7)

[0086] In the formula, each coefficient B 1～10 When the signal is unknown, the coefficient B is unknown and needs to be calculated using the least squares method.

[0087] (3) Parameter estimation. The Hall element is used to measure the grid-connected AC voltage V pcc Sampling is performed, where the sampling frequency is f s , so the sampling interval is t s =1 / f s , the sampling point is an odd number, expressed as 2N+1. Therefore, formula (7) is written in matrix form:

[0088] Y=XB (8)

[0089] B=(X T W T WX) -1 X T W T WY (9)

[0090] In the formula, the Y matrix is ​​obtained after the signal is sampled, and the middle time of the data window is used as the reference time. In the X matrix, let c(n) = cos(ω0nt s ), s(n) = sin(ω0nt s ). The weight matrix W = diag{w -N ,…,w0,…,w N}.

[0091] (4) State quantity output. The coefficient matrix B corresponding to each data window can be obtained by performing the least squares calculation on each data window. The amplitude A(t) and phase angle θ(t) of the signal at time t can be calculated according to the following formula.

[0092]

[0093] S3. Detect the harmonics of the photovoltaic grid-connected system including load regulation and capacity distribution.

[0094] The above step S3 includes the following parts:

[0095] (1) Input signal. Assume that the mathematical model of the target detection signal is:

[0096] y(i)=x(i)+σ·f(i),i=0,1,…,N-1 (12)

[0097] In the formula, x(i) is the basic signal, also called the true signal, f(i) is the interference signal, σ is an interference standard deviation, and y(i) is the original signal.

[0098] (2) Signal decomposition and reconstruction. Using the multi-scale algorithm of wavelet transform, the fundamental wave and harmonic components of the sinusoidal signal are obtained through multiple decompositions. The decomposed signal is calculated to obtain parameters such as the effective value and phase of the signal. The reconstructed form of the signal is:

[0099]

[0100] Where a is the expansion factor, h0(t) and h1(t) are the two sets of filtering signals of the low-pass filter and the high-pass filter.

[0101] (3) Harmonic calculation. For a frequency of f0, amplitude of C, and initial phase of The single frequency signal x(t) is used to sample the AC harmonic voltage signal of the grid connection point using Hall elements, where the sampling frequency f s , uniformly sampled, and through fast Fourier decomposition, the discrete time signals of each harmonic voltage are obtained:

[0102]

[0103] The correction formulas for frequency, initial phase and amplitude are:

[0104]

[0105] Where Δf = f s / N, k=0, 1, 2..., N-1.

[0106] S4: Control the harmonic resonance, voltage over-limit and photovoltaic inverter of the distribution network.

[0107] The above step S4 is as follows Figure 2 As shown:

[0108] (1) Determine the inverter control mode. According to the grid-side equivalent impedance detected by S1 and the real-time power generation capacity P of the inverter g Calculate real-time short circuit ratio of SCR.

[0109]

[0110] If the short-circuit ratio SCR of the grid-connected system is less than 3, it is in weak grid mode, and the inverter control method should adopt the grid control mode. In this mode, there is no resonance problem. The voltage limit detected by S2 can be inferred from the reactive power by the following formula. Improvement can be achieved by giving an amplitude to the grid reactive control loop.

[0111]

[0112] If the short-circuit ratio SCR of the grid-connected system is greater than 3, the inverter adopts the grid-following control mode. In this mode, the voltage over-limit and resonance problems caused by the reverse flow of new energy can be improved by the following parts.

[0113] (2) Voltage over-limit control. According to the grid voltage amplitude V detected by S2 pcc , with the rated reference voltage V ref A comparison is made and, after passing through the controller, a voltage over-limit control compensation amount is generated and superimposed on the inverter q-axis current setting to achieve voltage compensation.

[0114] i qref =G PI (V ref -V pcc ) (18)

[0115] (3) Harmonic resonance suppression. The harmonic voltage V can be extracted according to S3. th , divide each harmonic voltage by the virtual resistance to form each harmonic current given, which is superimposed on the dq axis current given to achieve the weak network harmonic resonance suppression function.

[0116]

[0117] In the above embodiment, the present invention provides an optimization control method for a photovoltaic inverter with a load-adjusting capacity distribution network. The present invention calculates the complex frequency domain components of the common access point voltage and the grid-side current at the injected harmonic frequency by using the Goertzel algorithm to obtain the grid impedance. The least squares method is used to detect the voltage state quantity of the photovoltaic inverter grid-connected node; the multi-scale algorithm of wavelet transform is used to obtain the fundamental wave and each harmonic component in the sinusoidal voltage signal. And the real-time short-circuit ratio is calculated according to the detected grid-side equivalent impedance and the real-time power generation capacity of the inverter to determine the inverter working mode. In the extremely weak grid mode, the inverter control method adopts the grid-building control mode, and the voltage over-limit problem is improved by setting the amplitude of the grid-building reactive control loop. In the scenario between strong and weak networks, the inverter adopts grid-following control, and generates the voltage over-limit control compensation according to the detected grid-connected point voltage amplitude; according to the extracted harmonic voltages, each harmonic current is set to achieve the dual functions of weak network harmonic resonance suppression and voltage over-limit compensation. This control method can achieve high power quality operation of distribution networks containing load-regulating transformers and grid-connected inverters, avoid voltage over-limit and excessive harmonic content in the terminal area, and fundamentally avoid grid relay chain accidents induced by voltage over-limit and harmonic resonance, which is conducive to the widespread and friendly access of distributed new energy and comprehensive power quality management.

[0118] The above-described embodiments are merely descriptions of preferred implementations of the present invention, and are not intended to limit the concept and scope of the present invention. Without departing from the design concept of the present invention, various modifications and improvements made by ordinary persons in the art to the technical solution of the present invention should fall within the protection scope of the present invention, and the technical contents for which protection is sought in the present invention have been fully recorded in the claims.

Claims

1. An optimization control method for a photovoltaic inverter in a load-adjusting and capacity-distributing network, characterized in that: The following steps are included: S1. Identify the operating capacity of the on-load capacity-changing transformer and the grid-side equivalent impedance; S2, detecting the voltage state quantity of the photovoltaic inverter grid-connected node; S3, detect the harmonics of photovoltaic grid-connected system including load regulation and capacity distribution; S4: Control the harmonic resonance, voltage over-limit and photovoltaic inverter of the distribution network.

2. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 1 is characterized in that: In step S1, the operating capacity of the on-load capacity-adjusting transformer takes the operating loss and the number of switching times of the on-load capacity-adjusting transformer as the preferred indicators, and calculates the optimal value of the capacity of the capacity-adjusting transformer under the two preferred indicators respectively. The calculation formula is as follows: P ZH =P OZH +(S / S NH ) 2 P KZH P ZL =P OZL +(S / S NL ) 2 P KZL Where P ZH , P OZH , P KZH , S NH They represent the comprehensive power loss of the large capacity of the on-load capacity-adjusting transformer, the comprehensive power no-load loss, the comprehensive power load loss and the rated capacity respectively; P ZL , P OZL , P KZL , S NL Respectively represent the comprehensive power loss of small capacity of on-load capacity-changing transformer, comprehensive power no-load loss, comprehensive power load loss and rated capacity; Combining the above two equations, we can calculate the theoretical optimal capacity adjustment point S of the on-load capacity-adjusting transformer. c , the formula is as follows, 3. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 1 is characterized in that: The grid-side equivalent impedance of the on-load capacity-changing transformer in step S1 is calculated by injecting non-characteristic subharmonic voltage at the public access point and extracting the corresponding harmonic corresponding current, and the grid characteristic impedance is calculated as follows: In the formula, ω hx is the angular frequency of harmonics injected into the power grid; V ghx and θ v are the amplitude and phase of the common access point voltage at the injection frequency; I hx and θ i They are respectively the amplitude and phase of the grid-side harmonic response current under the harmonic voltage excitation extracted after the harmonic voltage frequency is injected.

4. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 1 is characterized in that: The linearization calculation formula of the photovoltaic inverter grid-connected node voltage signal in step S2 is as follows: y(t)≈B1cos(ω0t)-B2sin(ω0t)-B3tcos(ω0t)-B4tsin(ω0t)+B5t 2 cos(ω0t) +B6t 2 sin(ω0t)-B7t 3 cos(ω0t)+B8t 3 sin(ω0t)-B9t 4 cos(ω0t)-B 10 t 4 sin(ω0t) In the formula, each coefficient B 1～10 When the signal is unknown, the coefficient B is unknown and needs to be calculated using the least squares method.

5. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 1 is characterized in that: The calculation formula for estimating the voltage signal parameter of the photovoltaic inverter grid-connected node in step S2 is as follows: The Hall element is used to measure the AC voltage V at the grid point pcc Sampling is performed, where the sampling frequency is f s , the sampling interval is t s =1 / f s , the sampling point is an odd number, expressed as 2N+1; Y=XB B=(X T W T WX) -1 X T W T WY In the formula, the Y matrix is ​​obtained after the signal is sampled, and the middle time of the data window is used as the reference time. In the X matrix, let c(n) = cos(ω0nt s ), s(n) = sin(ω0nt s ), where the weight matrix W = diag{w -N ,…,w0,…,w N }; The coefficient matrix B corresponding to each data window is obtained by performing the least squares calculation on each data window, and the amplitude A(t) and phase angle θ(t) of the signal at time t are calculated according to the following formula:

6. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 1, characterized in that: The calculation method for decomposing and reconstructing the detected harmonic signal in step S3 is as follows: Using the multi-scale algorithm of wavelet transform, the fundamental wave and each harmonic component in the sinusoidal signal are obtained by decomposing the harmonic signal multiple times, and then the decomposed signal is calculated to obtain the effective value and phase parameters of the corresponding signal. The reconstructed form of the signal is: Where a is the expansion factor, h0(t) and h1(t) are the two sets of filtering signals of the low-pass filter and the high-pass filter.

7. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 1, characterized in that: The harmonic calculation formula in step S3 is as follows: For a frequency of f0, amplitude of C, and initial phase of The single frequency signal x(t) is used to sample the AC harmonic voltage signal of the grid connection point using Hall elements, where the sampling frequency f s , uniformly sampled, and through fast Fourier decomposition, the discrete time signals of each harmonic voltage are obtained: The correction formula for frequency, initial phase and amplitude is: f0=kΔf=(ε+k p +0.5)Δf C=N -1 (X(k p-1 Δf)-2X(k p Δf)+2X(k p+1 Δf)-X(k p+2 Δf))g(ε) Where Δf = f s / N, k=0, 1, 2..., N-1.

8. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 3 is characterized in that: The control method of the photovoltaic inverter in step S4 is as follows: According to the grid-side equivalent impedance of the on-load capacity-adjusting transformer and the real-time power generation capacity P of the photovoltaic inverter g Calculate the real-time short-circuit ratio of SCR, If the grid-connected system short-circuit ratio SCR is less than 3, it is in weak grid mode, and the inverter control method adopts grid-connected control mode without resonance problem; If the short-circuit ratio SCR of the grid-connected system is greater than 3, the inverter adopts the grid-following control mode.

9. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 1, characterized in that: The voltage over-limit control method in step S4 is as follows: According to step S2, the voltage amplitude V of the photovoltaic inverter grid connection point is detected. pcc and with the rated reference voltage V ref Compare and generate voltage over-limit control compensation through the controller, which is added to the inverter q-axis current setting to achieve voltage compensation. i qref =G PI (V ref -V pcc )。 10. The optimization control method for a photovoltaic inverter in a load-regulating capacity distribution network according to claim 1, characterized in that: The harmonic resonance control method in step S4 is as follows: According to step S3, each harmonic voltage V is extracted th , divide each harmonic voltage by the virtual resistance to form each harmonic current setting, and superimpose it on the dq axis current setting to achieve the weak network harmonic resonance suppression function.

Images