Power grid cross-correlation harmonic suppression method, device, equipment and storage medium

Abstract

The invention discloses a power grid cross-correlation harmonic suppression method, device and equipment and a storage medium, and the method comprises the steps: carrying out the wavelet packet decomposition of a subnet electric signal, and obtaining the harmonic component of a subnet at each frequency band; calculating the harmonic phase difference between the subnets according to the cross-correlation function between the harmonic components of the subnets and the topological impedance matrix, and calculating the total harmonic distortion rate according to the harmonic components of the subnets in the frequency bands; constructing a harmonic transfer equation according to the harmonic phase difference between the subnets and the equivalent output impedance of the inverter so as to calculate the harmonic responsibility proportion of the subnets; solving the double-layer game model based on the total harmonic distortion rate, the reference value of the total harmonic distortion rate and the harmonic responsibility proportion of the subnet, and obtaining the output of the governance equipment; and inputting the output of the treatment equipment into the fuzzy controller to obtain the impedance adjusting quantity of the subnet, and adjusting the virtual impedance according to the impedance adjusting quantity to update the harmonic backflow path. According to the invention, harmonic suppression among multiple subgrids during high-permeability distributed grid connection is realized, and the operation stability of the power grid is improved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and particularly to a method, device, equipment and storage medium for harmonic governance of grid cross-correlation. Background Art

[0002] Harmonics are important parameters of the power quality of the power grid. To ensure the stable operation of the power grid, harmonics must be controlled within a certain level. With the grid connection of high-penetration distributed photovoltaics, the switching of their power electronic devices will generate more harmonics, and the harmonics between subnets have cross-correlation properties. The harmonic governance of a single subnet often fails to achieve the expected goal. Therefore, how to provide a method for harmonic governance between multiple subnets during high-penetration distributed grid connection is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0003] The present application provides a method, device, equipment and storage medium for harmonic governance of grid cross-correlation, which is used to govern the harmonics between multiple subnets during the grid connection of high-penetration distributed photovoltaics and improve the stability of the power grid operation.

[0004] In view of this, in the first aspect of the present application, a method for harmonic governance of grid cross-correlation is provided, including:

[0005] Performing wavelet packet decomposition on the acquired electrical signals of the photovoltaic subnets to obtain the harmonic components of each photovoltaic subnet in each frequency band;

[0006] Calculating the cross-correlation function according to the harmonic components of each photovoltaic subnet in the same frequency band, calculating the harmonic phase difference between each photovoltaic subnet according to the cross-correlation function between each photovoltaic subnet and the topological impedance matrix of the photovoltaic subnet, and calculating the total harmonic distortion rate of each photovoltaic subnet according to the harmonic components of each photovoltaic subnet in each frequency band;

[0007] Constructing a harmonic transfer equation according to the harmonic phase difference, total harmonic distortion rate and inverter equivalent output impedance between each photovoltaic subnet, and calculating the harmonic responsibility ratio of each photovoltaic subnet based on the harmonic transfer equation;

[0008] Constructing a two-layer game model, solving the two-layer game model based on the total harmonic distortion rate of each photovoltaic subnet, the total harmonic distortion rate reference value and the harmonic responsibility ratio of each photovoltaic subnet, and obtaining the output of the governance equipment in each photovoltaic subnet;

[0009] Inputting the output of the governance equipment in each photovoltaic subnet into a fuzzy controller to obtain the impedance adjustment amount of each photovoltaic subnet, and adjusting the virtual impedance according to the impedance adjustment amount of each photovoltaic subnet to update the harmonic reflux path.

[0010] Optionally, after calculating the total harmonic distortion rate of each photovoltaic subnet according to the harmonic components of each photovoltaic subnet in each frequency band, it further includes:

[0011] Obtain the admittance matrix based on the topological impedance matrix of the photovoltaic subnets, and calculate the distribution weights of the harmonic electrical signals in each photovoltaic subnet through the admittance matrix;

[0012] Correct the total harmonic distortion rate of each photovoltaic subnet through the distribution weights of the harmonic electrical signals in each photovoltaic subnet.

[0013] Optionally, calculate the harmonic phase difference between each photovoltaic subnet according to the cross-correlation function between each photovoltaic subnet and the topological impedance matrix of the photovoltaic subnet, including:

[0014] Extract the extreme points from the cross-correlation function between each photovoltaic subnet, and calculate the path difference between each photovoltaic subnet through the extreme points and the harmonic propagation speed between each photovoltaic subnet;

[0015] According to the relationship between the equivalent impedance difference and the path difference between each photovoltaic subnet, combined with the topological impedance matrix of the photovoltaic subnet, calculate the equivalent impedance between the nodes of each photovoltaic subnet, and calculate the harmonic phase difference between each photovoltaic subnet through the equivalent impedance between the nodes of each photovoltaic subnet.

[0016] Optionally, the calculation formula for the harmonic responsibility ratio of each photovoltaic subnet is:

[0017]

[0018] In the formula, is the h-th harmonic responsibility ratio of the photovoltaic subnet i; is the harmonic transfer impedance from node i to node j; is the h-th harmonic current component of node j; is the total harmonic distortion rate of the photovoltaic subnet i; N is the number of photovoltaic subnets.

[0019] Optionally, the optimization objective function of the main game layer in the double-layer game model is:

[0020]

[0021] In the formula, is the output vector of the governance equipment in the photovoltaic subnet, N is the number of photovoltaic subnets; is the total harmonic distortion rate of the photovoltaic subnet i; is the reference value of the total harmonic distortion rate; is the penalty coefficient;

[0022] The utility function of the subordinate game layer in the double-layer game model is:

[0023]

[0024] In the formula, is the utility function of the photovoltaic subnet i; ,, All are weight coefficients; is the harmonic responsibility ratio of the i-th photovoltaic subnet; is the output of the governance device in the i-th photovoltaic subnet.

[0025] Optionally, the method further includes:

[0026] Taking the output, impedance adjustment amount, and total harmonic distortion rate of the governance devices in each photovoltaic subnet as states, and the control parameters in the two-layer game model and the fuzzy controller as actions, using the reinforcement learning method to obtain the optimal actions; updating the control parameters in the two-layer game model and the fuzzy controller through the optimal actions.

[0027] The second aspect of the present application provides a device for harmonic governance of grid intercorrelation, including:

[0028] A wavelet packet decomposition unit, configured to perform wavelet packet decomposition on the acquired electrical signals of the photovoltaic subnets to obtain harmonic components of each photovoltaic subnet in each frequency band;

[0029] A first calculation unit, configured to calculate the cross-correlation function according to the harmonic components of each photovoltaic subnet in the same frequency band, calculate the harmonic phase difference between each photovoltaic subnet according to the cross-correlation function between each photovoltaic subnet and the topological impedance matrix of the photovoltaic subnet, and calculate the total harmonic distortion rate of each photovoltaic subnet according to the harmonic components of each photovoltaic subnet in each frequency band;

[0030] A second calculation unit, configured to construct a harmonic transfer equation according to the harmonic phase difference, total harmonic distortion rate between each photovoltaic subnet, and the equivalent output impedance of the inverter, and calculate the harmonic responsibility ratio of each photovoltaic subnet based on the harmonic transfer equation;

[0031] A model construction and solution unit, configured to construct a two-layer game model, solve the two-layer game model based on the total harmonic distortion rate of each photovoltaic subnet, the total harmonic distortion rate reference value, and the harmonic responsibility ratio of each photovoltaic subnet, and obtain the output of the governance device in each photovoltaic subnet;

[0032] An adjustment unit, configured to input the output of the governance device in each photovoltaic subnet into the fuzzy controller to obtain the impedance adjustment amount of each photovoltaic subnet, and adjust the virtual impedance according to the impedance adjustment amount of each photovoltaic subnet to update the harmonic reflux path.

[0033] Optionally, the device further includes:

[0034] A feedback adjustment unit, configured to take the output, impedance adjustment amount, and total harmonic distortion rate of the governance devices in each photovoltaic subnet as states, and the control parameters in the two-layer game model and the fuzzy controller as actions, using the reinforcement learning method to obtain the optimal actions; updating the control parameters in the two-layer game model and the fuzzy controller through the optimal actions.

[0035] A third aspect of the present application provides an electronic device, the device comprising a processor and a memory;

[0036] The memory is used to store program code and transmit the program code to the processor;

[0037] The processor is used to execute the power grid cross-correlation harmonic control method described in any one of the first aspects according to the instructions in the program code.

[0038] In a fourth aspect, the present application provides a computer-readable storage medium, which is used to store program code. When the program code is executed by a processor, it implements the method for controlling cross-correlated harmonics in a power grid as described in any one of the first aspects.

[0039] It can be seen from the above technical solutions that this application has the following advantages:

[0040] The grid cross-correlation harmonic control method provided in the present application realizes accurate separation of multi-band harmonic energy by performing wavelet packet decomposition on electrical signals, thereby avoiding the spectrum leakage problem of traditional FFT; performs cross-correlation analysis on wavelet packets instead of cross-correlation analysis on original electrical signals, thereby avoiding noise and frequency band aliasing interference in original electrical signals and improving the accuracy of phase difference detection; constructs inter-subgrid harmonic transfer equations based on the phase difference between subgrids, thereby quantifying the harmonic responsibility weights of each subgrid; realizes dynamic coordination between subgrids by adopting a two-layer game model, obtains the control output of each subgrid, and then adjusts the virtual impedance, changes the equivalent impedance characteristics of the local power grid, forces the harmonic current to choose a low-impedance path, breaks the original harmonic propagation path, and forms a new return network with control equipment as the core, thereby realizing harmonic control between multiple subgrids during high-penetration distributed grid connection;

[0041] Furthermore, the present application dynamically adjusts the game control strategy and virtual impedance parameters through a feedback mechanism, thereby improving the problem of operating condition changes caused by photovoltaic fluctuations, and achieving Pareto optimality among multiple objectives such as harmonic suppression, equipment loss, and stability, thereby improving the system's anti-interference ability and realizing adaptive management of harmonics between multiple subgrids during high-penetration distributed grid connection. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1A schematic diagram of a flow chart of a method for controlling cross-correlated harmonics in a power grid provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the structure of a power grid cross-correlated harmonic control device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] For easier understanding, please refer to Figure 1 The present application embodiment provides a method for controlling cross-correlated harmonics in a power grid, including:

[0047] Step 110, performing wavelet packet decomposition on the acquired electrical signals of the photovoltaic subgrid to obtain the harmonic components of each photovoltaic subgrid in each frequency band;

[0048] Obtain the voltage signal or current signal of each subnet in the high-penetration distributed photovoltaic grid-connected system, perform wavelet packet decomposition on it, and obtain the harmonic components of each photovoltaic subnet in each frequency band.

[0049]

[0050] In the formula, Represents the voltage signal / current signal of the current photovoltaic subgrid; is the harmonic component of the rth photovoltaic node in the mth layer, is the wavelet basis function of the rth PV node in the mth layer, S is the number of wavelet decomposition layers, and M is the number of PV nodes in the current PV subnet.

[0051] The energy projection of the electrical signal in each frequency band can be extracted by inner product operation to obtain the harmonic components of each frequency band. :

[0052]

[0053] Wavelet basis functions It is generated by scaling and translating the mother wavelet to cover a specific frequency range and time window. The mother wavelet can be selected according to the actual situation and is not specifically limited here.

[0054] Step 120: Calculate the cross-correlation function based on the harmonic components of each PV subnet at the same frequency band, calculate the harmonic phase difference between each PV subnet according to the cross-correlation function between each PV subnet and the topological impedance matrix of the PV subnet, and calculate the total harmonic distortion rate of each PV subnet according to the harmonic components of each PV subnet at each frequency band;

[0055] If the cross-correlation analysis is directly performed on the original electrical signal, it is vulnerable to noise and frequency band aliasing interference, and the phase difference detection error will be caused by the aliased multi-frequency band harmonics in the electrical signal. Therefore, the cross-correlation analysis of the wavelet packet is performed in the embodiment of the present application:

[0056]

[0057] wherein, is the cross-correlation function; is the integration variable in the time-domain integration, representing the instantaneous moment on the time axis; and are respectively the harmonic components of the electrical signal of the current PV subnet and the electrical signal

[0058] of other PV subnets at the same frequency band. By suppressing the interference of non-target frequency bands through wavelet packet decomposition, the phase difference detection error can be reduced by about 30%. Extract the extreme points from the cross-correlation function between each PV subnet to determine the harmonic propagation delay, and calculate the path difference

[0059] between each PV subnet in combination with the harmonic propagation speed v (usually taking the electromagnetic wave speed m / s) between the measurement points of each PV subnet.

[0060] The path difference of the harmonic calculated through the time delay can locate the harmonic position, and after the location, the key treatment of the harmonic source can be carried out. This method can avoid the need for one-by-one testing of all test points. In the embodiment of the present application, according to the relationship between the equivalent impedance difference and the path difference between each PV subnet, combined with the topological impedance matrix of the PV subnet (the inverse matrix of the nodal admittance matrix ), calculate the equivalent impedance between the nodes of each PV subnet, and calculate the harmonic phase difference between each PV subnet through the equivalent impedance between the nodes of each PV subnet. For the PV subnet (i.e., node i) and the PV subnet

[0061]

[0062] Wherein, is the system reference impedance, which is determined by the rated voltage and capacity of the power grid.

[0063] Using the topological impedance matrix the node impedance in it, calculate the equivalent reactance (the imaginary part of the impedance) between nodes, and then deduce the phase difference:

[0064]

[0065] Wherein, is the phase difference between node and node ; is the impedance between node i and harmonic source node k, is the impedance between node j and harmonic source node k, and are the resistance and reactance values from node to harmonic source node respectively, which can be determined by solving the system of equations using the least squares method position; and are the resistance and reactance values from node to harmonic source node respectively.

[0066] It should be noted that when , is 0, at this time , is infinite or infinitesimal, and this is also meaningful. Taking as an example, when is infinite, is equal to 90°, and when is infinitesimal, is equal to -90°.

[0067] The determination process of the harmonic source position is as follows: According to the harmonic transfer impedance from node i to node j in the impedance matrix , construct a system of equations for the harmonic current propagation path:

[0068]

[0069] Wherein, is the node where the harmonic source is located, and are the measurement point nodes. By solving the system of equations using the least squares method, the position can be determined.

[0070] Time delay It reflects the difference in the propagation path of harmonics between subnets and is directly related to the power grid topology. Impedance matrix Describes the impedance relationship between grid nodes and is used to quantify the impedance difference in the propagation path of harmonic currents. The signal after wavelet packet decomposition ( and frequency band components) is used to suppress noise interference and helps to improve detection accuracy. In the embodiments of the present application, the time difference of harmonic propagation is captured through cross-correlation analysis, and combined with the grid topology impedance matrix, the physical path difference is converted into an electrical phase difference.

[0071] After obtaining the harmonic components of the photovoltaic subnet i, calculate the harmonic energy of each frequency band:

[0072]

[0073] In the formula, is the h-th harmonic energy of the photovoltaic subnet i; is the h-th harmonic component of the r-th photovoltaic node in the m-th layer, which can be obtained by filtering through a band-pass filter in the h-th harmonic frequency band ; S is the number of wavelet decomposition layers, and M is the total number of photovoltaic nodes in the photovoltaic subnet i.

[0074] It should be noted that a single frequency band may contain multiple harmonics. For example, 200 - 300 Hz includes the 4th, 5th, and 6th harmonics (the fundamental frequency is usually 50 Hz); the h-th harmonic component can be obtained through a band-pass filter in the h-th harmonic frequency band, and this process belongs to the prior art, so the specific process will not be elaborated here.

[0075] Then, calculate the total harmonic distortion rate of the photovoltaic subnet i according to the harmonic energy of each frequency band of the photovoltaic subnet i:

[0076]

[0077] In the formula, is the fundamental wave energy of the photovoltaic subnet i, and H is the highest harmonic order (usually taken as 50).

[0078] Furthermore, after calculating the total harmonic distortion rate of each photovoltaic subnet in the embodiments of the present application, according to the topological impedance matrix of the photovoltaic subnet, obtain the admittance matrix ( is the inverse matrix of the admittance matrix ), and through the admittance matrix Calculate the distribution weights of the harmonic electrical signal in each photovoltaic subnet; correct the total harmonic distortion rate of each photovoltaic subnet according to the distribution weights of the harmonic electrical signal in each photovoltaic subnet. The distribution weight of the harmonic electrical signal in photovoltaic subnet i is calculated by the following formula:

[0079]

[0080] wherein, is the self-admittance of node i (the sum of the admittance to the ground and all branch admittances), is the admittance between nodes i and j;

[0081] The total harmonic distortion rate of the corrected photovoltaic subnet i .

[0082] The grid topology impedance matrix determines the distribution characteristics of harmonic currents. High-impedance nodes have a more significant inhibitory effect on harmonic currents. It is necessary to correct the total harmonic distortion rate THD of photovoltaic subnet i to account for its actual impact. i

[0083] In the embodiments of the present application, precise separation of multi-band harmonic energy is achieved through wavelet packet decomposition, which can avoid the spectrum leakage problem of the traditional FFT (Fast Fourier Transform). By adopting a multi-scale cross-correlation harmonic detection method, the limitations of traditional FFT spectrum analysis for non-steady-state harmonics can be overcome, and a composite detection model combining wavelet packet decomposition and cross-correlation function is introduced to achieve precise positioning of the harmonic phase difference and propagation path between subnets.

[0084] Step 130: Construct a harmonic transfer equation based on the harmonic phase difference, total harmonic distortion rate, and inverter equivalent output impedance between each photovoltaic subnet, and calculate the harmonic responsibility ratio of each photovoltaic subnet based on the harmonic transfer equation;

[0085] The harmonic phase difference between each photovoltaic subnet can be generated in matrix form to obtain the harmonic phase difference matrix . According to the harmonic phase difference matrix , the inverter equivalent output impedance (known parameter), and the harmonic voltage of each node, construct a harmonic transfer equation:

[0086]

[0087] wherein, is the harmonic current vector, which is the set of harmonic currents of each subnet; is the inverse matrix of the admittance matrix; is the harmonic voltage vector, which is the set of harmonic voltages of each subnet;​ is the inverse matrix of the equivalent output impedance matrix of the inverter; is the internal harmonic voltage source vector of the inverter.

[0088] The admittance matrix in the harmonic transfer equation needs to be adjusted according to the phase difference to reflect the time-delay effect of the harmonic propagation path:

[0089]

[0090] where is the phase rotation factor, which corrects the phase relationship of the harmonic current between nodes. The harmonic current vector needs to introduce a phase difference correction term in the calculation:

[0091]

[0092] is the phase difference of the harmonic source node k (derived from ), and the phase difference directly affects the superposition effect of the harmonic current. A high phase difference may lead to harmonic cancellation or enhancement.

[0093] indirectly affects the harmonic transfer equation through the harmonic voltage amplitude . Assuming the fundamental voltage is , then:

[0094]

[0095] In the formula, is the h-th harmonic current component of node i; is the fundamental current of node i; is the fundamental voltage of node i; is the h-th harmonic voltage component of node i. The above formula transforms into the harmonic voltage components of each order and injects them into the vector on the right side of the harmonic transfer equation.

[0096] The calculation formula for the harmonic responsibility ratio of the photovoltaic subnet i is:

[0097]

[0098] In the formula, is the h-th harmonic responsibility ratio of the photovoltaic subnet i; is the harmonic transfer impedance from node i to node j; is the h-th harmonic current component of node j; N is the number of photovoltaic subnets. The photovoltaic subnet with a high THD value will be assigned a higher governance responsibility.

[0099] After calculating the harmonic liability ratios of each PV subnet, one of the harmonic liability ratios can be selected from , ,..., as the final harmonic liability ratio of each PV subnet , that is = , , and finally output the liability degree vector .

[0100] Step 140: Construct a two-layer game model, solve the two-layer game model based on the total harmonic distortion rate, the reference value of the total harmonic distortion rate, and the harmonic liability ratio of each PV subnet, and obtain the output of the governance equipment in each PV subnet;

[0101] The embodiment of the present application adopts hierarchical cooperative game control and uses a two-layer game model (including a main game layer, a sub-game layer, and a Nash equilibrium strategy) to achieve dynamic coordination between subnets. The optimization objective function of the main game layer in the two-layer game model is:

[0102]

[0103] In the formula, is the output vector of the governance equipment in the PV subnet, and N is the number of PV subnets; is the total harmonic distortion rate of PV subnet i; is the reference value of the total harmonic distortion rate; is the penalty coefficient;

[0104] The utility function of the sub-game layer in the two-layer game model is:

[0105]

[0106] In the formula, is the utility function of PV subnet i; is the weight coefficient of the harmonic liability ratio, is the weight coefficient of the marginal effect of the governance equipment output on harmonic suppression; is the harmonic liability ratio of PV subnet i; is the output of the governance equipment in PV subnet i; represents the marginal effect of the governance equipment output on harmonic suppression.

[0107] The main game layer sets the global optimization goal, and the sub-game layer feeds back the local optimal solution through the Nash equilibrium strategy, forming a two-layer iterative optimization structure. The governance output vector P output by the main game layer is used as the constraint condition of the sub-game layer to limit the output range of each subnet. The utility function Act on the objective function of the main game layer through the Lagrange multiplier method to adjust the penalty coefficient and the weight coefficient .

[0108] The bilevel game model can be solved by the Lagrange multiplier method to obtain the Pareto optimal solution. The main game layer decomposes the global optimization problem into multiple sub-problems, and each sub-network solves:

[0109]

[0110] are the minimum output and maximum output of the governance equipment in the photovoltaic sub-network i respectively. Through iterative update , the system reaches the Nash equilibrium state, that is, any sub-network changing its output alone cannot improve its own utility. For example, when a certain photovoltaic sub-network has a higher , increases, driving this photovoltaic sub-network to allocate more governance resources ( increases), thereby reducing the global

[0111] This application adopts a global-local collaborative control method. Compared with the centralized control (such as PID tuning), the bilevel game model reduces the computational complexity from to .

[0112] The generation process of the output vector P of the governance equipment is as follows:

[0113] 1. Initial solution of the main game layer:

[0114] Based on the responsibility degree vector and system constraints, initially calculate the reference value of .

[0115] 2. Fine adjustment of the sub-game layer:

[0116] Each photovoltaic sub-network adjusts its output according to :

[0117] If (that is, increasing can reduce ), then increase ;

[0118] Conversely, reduce to avoid waste of resources.

[0119] 3. Dynamic convergence process:

[0120] The main game layer controls the total output by adjusting the penalty coefficient ​​ , from the game layer, optimize the local parameters through the utility function until the system reaches a stable state, and finally output the output vector P of the governance device.

[0121] Step 150: Input the output of the governance device in each photovoltaic subnet into the fuzzy controller, obtain the impedance adjustment amount of each photovoltaic subnet, and adjust the virtual impedance according to the impedance adjustment amount of each photovoltaic subnet to update the harmonic reflux path;

[0122] This application adopts the virtual impedance adaptive adjustment method, and adjusts the virtual impedance characteristics in real time through power electronic devices to change the harmonic reflux path. Specifically, the dynamic virtual impedance equation in this application is:

[0123]

[0124] In the formula, are the proportional gain, integral gain, and derivative gain respectively, which are tuned online by the fuzzy controller. s is the complex frequency variable (Laplace domain variable), and F is the cut-off frequency coefficient of the derivative filter, which is used to suppress high-frequency noise;

[0125] The corresponding stability constraint is:

[0126]

[0127] In the formula, Re( ) represents taking the real part of a complex number; is the grid impedance.

[0128] Each corresponds to the governance power demand of the photovoltaic subnet . For example, if is the compensation current amplitude of the active filter, its magnitude directly affects the adjustment intensity of the virtual impedance.

[0129] As the input variable of the fuzzy controller, it drives the dynamic adjustment of the gain parameter of the fuzzy controller, and then outputs the impedance adjustment amount . For example: when increases, the fuzzy rule will increase to quickly respond to the harmonic suppression demand. When fluctuates continuously, increases to suppress overshoot and ensure stability. The output impedance adjustment amount is the increment of in the current adjustment period t relative to the previous adjustment period t-1, that is:

[0130]

[0131] According to the impedance adjustment amount Update the harmonic reflux path, such as through the impedance adjustment amount Adjust the virtual impedance , change the equivalent impedance characteristics of the local power grid, and force the harmonic current to select a low-impedance path. Through dynamic adjustment, the original harmonic propagation path is broken, and a new reflux network with the governance device as the core is formed.

[0132] In another embodiment, after step 150, it further includes:

[0133] Step 160: Using the output power, impedance adjustment amount, and total harmonic distortion rate of the governance devices in each photovoltaic subnet as the state, and the control parameters in the double-layer game model and the fuzzy controller as the actions, adopt the reinforcement learning method to obtain the optimal actions; update the control parameters in the double-layer game model and the fuzzy controller through the optimal actions.

[0134] In the embodiment of the present application, the output vector P of the governance device, the impedance adjustment amount and the total harmonic distortion rate THD are used as the state, and the penalty coefficient , weight coefficient in the double-layer game model and the impedance adjustment parameter in the fuzzy controller are used as the actions to construct a DQN network (Deep Q-Network) to optimize the control parameters in real time and solve the multi-objective conflict problem.

[0135] The Q-value function in the DQN network is:

[0136]

[0137] The reward function is:

[0138]

[0139] The update process of the policy network is:

[0140]

[0141] The loss function is:

[0142]

[0143] In the formula, is the Q-value function, indicating the expected cumulative reward for executing the action in the state ; E represents taking the expected value; is the reinforcement learning discount factor at time t, used to balance the weights of the current reward and future rewards; T is the termination time; is the weight parameter for the total harmonic distortion rate suppression effect in the reward function, used to enhance the priority of THD suppression; is the weight parameter for the energy consumption of the governance device in the reward function, used to limit the energy consumption of the governance device output; is the weight parameter for the impedance adjustment amount in the reward function, used to enhance the system stability constraint; is the parameter of the policy network, used to generate the action policy; is the learning rate, controlling the step size of the policy network parameter update; is the loss function, measuring the gap between the predicted Q value and the target Q value ( ), and updating the network parameters by minimizing the gap between the predicted Q value and the target Q value. is the state vector, including parameters such as the current impedance adjustment amount ΔZ v , THD value, output vector P of the governance device, etc.; is the action vector, representing the adjustment amount of the weight parameter .

[0144] The DQN network calculates the Q value according to the real-time reward , updates the policy network parameters , and finally outputs the optimal action, that is, the optimized parameter . The optimized parameter acts on the master-slave game layer to adjust the objective function and the utility function respectively, and acts on the virtual impedance adjustment layer to correct the fuzzy rule base of the virtual impedance adjustment, forming a closed-loop control. The updated control parameters act on the harmonic governance device to generate new and , and re-enter the DQN network for iterative optimization.

[0145] Among them, the parameter acts on the penalty coefficient in the optimization objective function of the master game layer to update the penalty coefficient in this objective function, that is ;

[0146] acts on the weight parameter of the utility function of the slave game layer, updating the weight parameter in this utility function, that is ;

[0147] acts on the gain tuning rule of the fuzzy controller. The gain adjustment amount in the fuzzy controller is positively correlated with , and the impedance matching condition is preferentially satisfied: , where, , for quickly responding to impedance mutations; = , for eliminating steady-state errors; = , for suppressing high-frequency oscillations; , 、 、 are all learning rate coefficients (usually taken as 0.1~1.0), controlling the adjustment amplitude to ensure the stability of the total gain. The larger it is, the higher the priority of the virtual impedance adjustment for stability constraints.

[0148] This application optimizes the control weight parameters in real time through reinforcement learning, dynamically adjusts the game control strategy and virtual impedance parameters through a feedback mechanism, and realizes the maximization of the global harmonic suppression efficiency and adaptive governance. By adjusting the weight parameters in real time, the problem of operating condition changes caused by photovoltaic fluctuations can be improved; through the collaborative feedback, Pareto optimality is achieved among multiple objectives such as harmonic suppression, device loss, and stability. The parameter closed-loop mechanism in this application increases the tolerance of the system to disturbances such as grid impedance mutations by 30%.

[0149] The grid cross-correlation harmonic governance method provided by this application realizes the precise separation of multi-band harmonic energy through wavelet packet decomposition of electrical signals, avoiding the spectral leakage problem of traditional FFT; through cross-correlation analysis of wavelet packets instead of the original electrical signals, noise and frequency band aliasing interference in the original electrical signals are avoided, and the accuracy of phase difference detection is improved; according to the phase difference between subnets, a harmonic transfer equation between subnets is constructed, and then the harmonic responsibility weights of each subnet are quantified; through the adoption of a two-layer game model, dynamic coordination between subnets is realized, the governance output of each subnet is obtained, and then the virtual impedance is adjusted to change the equivalent impedance characteristics of the local grid, forcing harmonic currents to choose low-impedance paths, breaking the original harmonic propagation path, and forming a new reflux network with the governance equipment as the core, realizing the harmonic governance between multiple subnets during high-penetration distributed grid connection;

[0150] Further, this application dynamically adjusts the game control strategy and virtual impedance parameters through a feedback mechanism, improves the problem of operating condition changes caused by photovoltaic fluctuations, and achieves Pareto optimality among multiple objectives such as harmonic suppression, device loss, and stability, improves the anti-interference ability of the system, and realizes the adaptive governance of harmonics between multiple subnets during high-penetration distributed grid connection.

[0151] The above is an embodiment of a grid cross-correlation harmonic governance method provided by this application. The following is an embodiment of a grid cross-correlation harmonic governance device provided by this application.

[0152] Please refer to Figure 2, an embodiment of the present application provides a power grid cross-correlation harmonic control device, comprising:

[0153] The wavelet packet decomposition unit 210 is used to perform wavelet packet decomposition on the acquired electrical signals of the photovoltaic subgrid to obtain the harmonic components of each photovoltaic subgrid in each frequency band;

[0154] The first calculation unit 220 is used to calculate the cross-correlation function according to the harmonic components of each photovoltaic subgrid in the same frequency band, calculate the harmonic phase difference between each photovoltaic subgrid according to the cross-correlation function between each photovoltaic subgrid and the topological impedance matrix of the photovoltaic subgrid, and calculate the total harmonic distortion rate of each photovoltaic subgrid according to the harmonic components of each photovoltaic subgrid in each frequency band;

[0155] The second calculation unit 230 is used to construct a harmonic transfer equation according to the harmonic phase difference, total harmonic distortion rate and inverter equivalent output impedance between the photovoltaic subgrids, and calculate the harmonic responsibility ratio of each photovoltaic subgrid based on the harmonic transfer equation;

[0156] The model building and solving unit 240 is used to build a double-layer game model, solve the double-layer game model based on the total harmonic distortion rate of each photovoltaic subgrid, the total harmonic distortion rate reference value and the harmonic responsibility ratio of each photovoltaic subgrid, and obtain the output of the governance equipment in each photovoltaic subgrid;

[0157] The regulating unit is used to input the output of the treatment equipment in each photovoltaic subnet into the fuzzy controller, obtain the impedance adjustment amount of each photovoltaic subnet, and adjust the virtual impedance according to the impedance adjustment amount of each photovoltaic subnet to update the harmonic return path.

[0158] As a further improvement, the device further comprises:

[0159] The correction unit 250 is used to obtain an admittance matrix according to the topological impedance matrix of the photovoltaic subgrid, calculate the distribution weight of the harmonic electrical signal in each photovoltaic subgrid through the admittance matrix, and correct the total harmonic distortion rate of each photovoltaic subgrid according to the distribution weight of the harmonic electrical signal in each photovoltaic subgrid.

[0160] As a further improvement, the calculation formula for the harmonic responsibility ratio of each PV subgrid is:

[0161]

[0162] In the formula, is the hth harmonic responsibility proportion of PV subgrid i; is the harmonic transfer impedance from node i to node j; is the hth harmonic current component of node j; is the total harmonic distortion rate of PV subgrid i; N is the number of PV subgrids.

[0163] As a further improvement, the optimization objective function of the master game layer in the two-layer game model is as follows:

[0164]

[0165] where is the output vector of the governance equipment in the photovoltaic sub-network, and N is the number of photovoltaic sub-networks; is the total harmonic distortion rate of the i-th photovoltaic sub-network; is the reference value of the total harmonic distortion rate; is the penalty coefficient;

[0166] The utility function of the slave game layer in the two-layer game model is as follows:

[0167]

[0168] where is the utility function of the i-th photovoltaic sub-network; , are both weight coefficients; is the proportion of harmonic responsibility of the i-th photovoltaic sub-network; is the output of the governance equipment in the i-th photovoltaic sub-network.

[0169] As a further improvement, the device further includes:

[0170] A feedback adjustment unit 260, which takes the output of the governance equipment, the impedance adjustment amount, and the total harmonic distortion rate in each photovoltaic sub-network as states, takes the control parameters in the two-layer game model and the fuzzy controller as actions, and uses the reinforcement learning method to obtain the optimal action; updates the control parameters in the two-layer game model and the fuzzy controller through the optimal action.

[0171] This application realizes the accurate separation of multi-band harmonic energy by performing wavelet packet decomposition on the electrical signal, avoiding the spectrum leakage problem of the traditional FFT; by performing cross-correlation analysis on the wavelet packet instead of the original electrical signal, so as to avoid the noise and frequency band aliasing interference in the original electrical signal and improve the accuracy of phase difference detection; constructs an inter-subnetwork harmonic transfer equation according to the phase difference between the sub-networks, and then quantifies the harmonic responsibility weights of each sub-network; realizes the dynamic coordination between the sub-networks by using a two-layer game model, obtains the governance output of each sub-network, and then adjusts the virtual impedance to change the equivalent impedance characteristics of the local power grid, forcing the harmonic current to select a low-impedance path, breaking the original harmonic propagation path, and forming a new reflux network with the governance equipment as the core, realizing the harmonic governance between multiple sub-networks during high-penetration distributed grid connection;

[0172] Furthermore, the present application dynamically adjusts the game control strategy and virtual impedance parameters through a feedback mechanism, improves the problem of operating condition changes caused by photovoltaic fluctuations, and achieves Pareto optimality among multiple objectives such as harmonic suppression, equipment loss, and stability, enhances the anti-interference ability of the system, and realizes the adaptive governance of harmonics between multiple subnets during high-penetration distributed grid connection.

[0173] An embodiment of the present application further provides an electronic device, which includes a processor and a memory;

[0174] The memory is used to store program codes and transmit the program codes to the processor;

[0175] The processor is configured to execute the method for cross-correlation harmonic governance of the power grid in the foregoing method embodiment according to the instructions in the program codes.

[0176] An embodiment of the present application further provides a computer-readable storage medium, which is used to store program codes, and when the program codes are executed by a processor, the method for cross-correlation harmonic governance of the power grid in the foregoing method embodiment is realized.

[0177] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices and units can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0178] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0179] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expressions refer to any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0180] In several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of devices or units can be in electrical, mechanical or other forms.

[0181] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0182] In addition, each functional unit in each embodiment of this application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0183] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (English full name: Read-Only Memory, English abbreviation: ROM), random access memory (English full name: Random Access Memory, English abbreviation: RAM), magnetic disks, or optical discs.

[0184] As described above, the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of various embodiments of this application.

Claims

1. A method for harmonic governance of grid cross-correlation, characterized in that, Including: Performing wavelet packet decomposition on the acquired electrical signals of the photovoltaic subnets to obtain the harmonic components of each photovoltaic subnet in each frequency band; Calculating the cross-correlation function according to the harmonic components of each photovoltaic subnet in the same frequency band, calculating the harmonic phase difference between each photovoltaic subnet according to the cross-correlation function between each photovoltaic subnet and the topological impedance matrix of the photovoltaic subnet, and calculating the total harmonic distortion rate of each photovoltaic subnet according to the harmonic components of each photovoltaic subnet in each frequency band; Constructing a harmonic transfer equation based on the harmonic phase difference, total harmonic distortion rate between each photovoltaic subnet, and the equivalent output impedance of the inverter, and calculating the harmonic responsibility ratio of each photovoltaic subnet based on the harmonic transfer equation; Constructing a two-layer game model, solving the two-layer game model based on the total harmonic distortion rate of each photovoltaic subnet, the reference value of the total harmonic distortion rate, and the harmonic responsibility ratio of each photovoltaic subnet, and obtaining the output of the governance equipment in each photovoltaic subnet; Inputting the output of the governance equipment in each photovoltaic subnet into a fuzzy controller to obtain the impedance adjustment amount of each photovoltaic subnet, and adjusting the virtual impedance according to the impedance adjustment amount of each photovoltaic subnet to update the harmonic reflux path.

2. The method for controlling cross-correlated harmonics in a power grid according to claim 1, characterized in that: Calculating the total harmonic distortion rate of each photovoltaic subnet according to the harmonic components of each photovoltaic subnet in each frequency band, and then further including: Obtaining the admittance matrix according to the topological impedance matrix of the photovoltaic subnet, and calculating the distribution weight of the harmonic electrical signal in each photovoltaic subnet through the admittance matrix; Correcting the total harmonic distortion rate of each photovoltaic subnet through the distribution weight of the harmonic electrical signal in each photovoltaic subnet.

3. The method for controlling cross-correlated harmonics in a power grid according to claim 1, characterized in that: Calculating the harmonic phase difference between each photovoltaic subnet according to the cross-correlation function between each photovoltaic subnet and the topological impedance matrix of the photovoltaic subnet, including: Extracting the extreme points from the cross-correlation function between each photovoltaic subnet, and calculating the path difference between each photovoltaic subnet through the extreme points and the harmonic propagation speed between each photovoltaic subnet; According to the relationship between the equivalent impedance difference and the path difference between each photovoltaic subnet, combining the topological impedance matrix of the photovoltaic subnet to calculate the equivalent impedance between the nodes of each photovoltaic subnet, and calculating the harmonic phase difference between each photovoltaic subnet through the equivalent impedance between the nodes of each photovoltaic subnet.

4. The method for harmonic governance of grid cross-correlation according to claim 1, characterized in that The calculation formula for the harmonic responsibility ratio of each photovoltaic subnet is: Wherein, is the proportion of the h-th harmonic liability of the photovoltaic sub-network i; is the harmonic transfer impedance from node i to node j; is the h-th harmonic current component of node j; is the total harmonic distortion rate of the photovoltaic sub-network i; N is the number of photovoltaic sub-networks.

5. The method for governing cross-correlation harmonics of a power grid according to claim 1, wherein The optimization objective function of the main game layer in the two-layer game model is: In the formula, is the output vector of the governance equipment in the photovoltaic subnet, and N is the number of photovoltaic subnets; is the total harmonic distortion rate of the i-th photovoltaic subnet; is the reference value of the total harmonic distortion rate; is the penalty coefficient; The utility function of the subordinate game layer in the two-layer game model is: In the formula, is the utility function of the photovoltaic subnet i; , are both weight coefficients; is the proportion of harmonic responsibility of the photovoltaic subnet i; is the output of the governance equipment in the photovoltaic subnet i.

6. The method for controlling cross-correlated harmonics in a power grid according to claim 1, characterized in that: The method further includes: Taking the output, impedance adjustment amount, and total harmonic distortion rate of the governance equipment in each photovoltaic subnet as the state, taking the control parameters in the two-layer game model and the fuzzy controller as the actions, and using the reinforcement learning method to obtain the optimal action; updating the control parameters in the two-layer game model and the fuzzy controller through the optimal action.

7. A device for controlling cross-correlated harmonics in a power grid, characterized in that: Including: A wavelet packet decomposition unit for performing wavelet packet decomposition on the acquired electrical signals of the photovoltaic subnet to obtain the harmonic components of each photovoltaic subnet in each frequency band; A first calculation unit for calculating the cross-correlation function according to the harmonic components of each photovoltaic subnet in the same frequency band, calculating the harmonic phase difference between each photovoltaic subnet according to the cross-correlation function between each photovoltaic subnet and the topological impedance matrix of the photovoltaic subnet, and calculating the total harmonic distortion rate of each photovoltaic subnet according to the harmonic components of each photovoltaic subnet in each frequency band; A second calculation unit, configured to construct a harmonic transfer equation based on the harmonic phase difference, total harmonic distortion rate, and inverter equivalent output impedance between each photovoltaic subnet, and calculate the harmonic responsibility ratio of each photovoltaic subnet based on the harmonic transfer equation; A model construction and solution unit, configured to construct a two-layer game model, solve the two-layer game model based on the total harmonic distortion rate, total harmonic distortion rate reference value, and harmonic responsibility ratio of each photovoltaic subnet, and obtain the output of the governance device in each photovoltaic subnet; An adjustment unit, configured to input the output of the governance device in each photovoltaic subnet into a fuzzy controller, obtain the impedance adjustment amount of each photovoltaic subnet, and adjust the virtual impedance according to the impedance adjustment amount of each photovoltaic subnet to update the harmonic reflux path.

8. The power grid mutual-correlation harmonic control device according to claim 7, characterized in that: The device further includes: A feedback adjustment unit, configured to use the output of the governance device, impedance adjustment amount, and total harmonic distortion rate in each photovoltaic subnet as states, use the control parameters in the two-layer game model and fuzzy controller as actions, and adopt a reinforcement learning method to obtain the optimal action; update the control parameters in the two-layer game model and fuzzy controller through the optimal action.

9. An electronic device, characterized in that, The device includes a processor and a memory; The memory is configured to store program code and transmit the program code to the processor; The processor is configured to execute the grid cross-correlation harmonic governance method according to any one of claims 1-6 based on the instructions in the program code.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is configured to store program code, and when the program code is executed by a processor, the grid cross-correlation harmonic governance method according to any one of claims 1-6 is implemented.

Images