Distributed photovoltaic power distribution network voltage optimization control method

Abstract

The invention discloses a distributed photovoltaic power distribution network voltage optimization control method, which relates to the technical field of renewable energy, and comprises the following steps: analyzing environmental factors influencing photovoltaic output in monitoring data, and evaluating the power capacity of a photovoltaic power distribution network; analyzing the power load change trend in the power consumption data, and evaluating the power load of the power consumption terminal; judging whether the output power of the distributed photovoltaic power distribution network meets power load supply and demand balance supplied to the power utilization terminal or not, if yes, screening out distributed photovoltaic power distribution network nodes with power redundancy values, and if not, marking distributed photovoltaic power distribution network nodes with power shortage values; and based on the distributed photovoltaic power distribution network with the power redundancy value and the distributed photovoltaic power distribution network with the power shortage value, constructing a distributed photovoltaic power distribution network power intelligent distribution model, and generating an optimal power scheduling strategy. The method has the advantages that voltage optimization control is realized without equipment assistance, and the stability of the photovoltaic power distribution network is improved.

Description

Technical Field

[0001] The present invention relates to the field of renewable energy technology, and in particular to a distributed photovoltaic power distribution network voltage optimization control method. Background Art

[0002] Distributed photovoltaic distribution network utilizes the power resources distributed in multiple small-scale photovoltaic power generation systems, converts direct current into alternating current through inverters, and realizes renewable energy power supply in local areas.

[0003] When the existing distributed photovoltaic distribution network faces excessively high voltage, it usually uses smart inverters for regulation or connects excess power to the grid. However, under the new policy, low-voltage power cannot be directly connected to the grid, and at the current stage, smart inverters still have defects. When an abnormality occurs in the smart inverter, it is very easy to cause electrical burns. Therefore, a method is proposed to reduce the voltage increase problem by allocating excess power to a nearby point. Summary of the invention

[0004] In order to solve the above technical problems, a distributed photovoltaic distribution network voltage optimization control method is provided. This technical solution solves the problem that when the existing distributed photovoltaic distribution network faces excessively high voltage, it usually adopts intelligent inverters for adjustment or connects excess power to the grid. However, under the new policy, low-voltage power cannot be directly connected to the grid, and at the current stage, intelligent inverters still have defects. When an abnormality occurs in the intelligent inverter, it is very easy to cause electrical burnout. Therefore, a method is proposed to reduce the problem of voltage increase by allocating excess power to a nearby point.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A distributed photovoltaic power distribution network voltage optimization control method, comprising:

[0007] Obtain monitoring data of distributed photovoltaic distribution networks, analyze environmental factors affecting photovoltaic output in the monitoring data, and evaluate the power production capacity of photovoltaic distribution networks;

[0008] Obtain power consumption data of power consumption terminals, analyze the power load change trend in the power consumption data, and evaluate the power load of power consumption terminals;

[0009] Determine whether the output power of the distributed photovoltaic distribution network meets the supply and demand balance of the power load supplied to the power consumption terminal. If so, select the distributed photovoltaic distribution network nodes with power redundancy values. If not, mark the distributed photovoltaic distribution network nodes with power shortage values.

[0010] Based on the distributed photovoltaic distribution network with power redundancy value and the distributed photovoltaic distribution network with power shortage value, an intelligent power distribution model of the distributed photovoltaic distribution network is constructed to generate the optimal power dispatching strategy.

[0011] Preferably, obtaining monitoring data of a distributed photovoltaic distribution network, analyzing environmental factors affecting photovoltaic output in the monitoring data, and evaluating the power production capacity of the photovoltaic distribution network specifically include:

[0012] Determining the structural parameters of the distributed photovoltaic distribution network; the structural parameters include: the area of ​​the photovoltaic module, the ideal photoelectric conversion efficiency of the photovoltaic module;

[0013] Based on the several environmental parameters in the meteorological data that affect the corresponding distributed photovoltaic distribution network at all times, as environmental influencing factors; the environmental influencing factors include: sunshine intensity, haze, sand and dust, temperature and humidity;

[0014] Analyze the power production efficiency of the photovoltaic distribution network based on the structural parameters and environmental impact factors of the distributed photovoltaic distribution network;

[0015] Based on the operation data of the distributed photovoltaic distribution network, the operation status of the photovoltaic components of the distributed photovoltaic distribution network is evaluated;

[0016] The power production efficiency of the photovoltaic distribution network is corrected by the operating state of the distributed photovoltaic distribution network to obtain the power production capacity of the photovoltaic distribution network;

[0017] Among them, the calculation expression of the power capacity of the photovoltaic distribution network is:

[0018]

[0019] Where P is the power capacity of the photovoltaic distribution network, F is the structural parameters of the photovoltaic module, G i is the i-th environmental impact factor corresponding to the distributed photovoltaic distribution network, U k is the operating status index of the kth PV module in the distributed PV distribution network, and L is the total number of PV modules.

[0020] Preferably, based on the operation data of the distributed photovoltaic distribution network, evaluating the operation status of the photovoltaic components of the distributed photovoltaic distribution network specifically includes:

[0021] Based on the maintenance task index of historical maintenance data, the historical maintenance data of the distributed photovoltaic distribution network is divided and reorganized according to the maintenance timestamp to obtain the historical operation time series data of the photovoltaic components;

[0022] The historical maintenance tasks that affect the output of PV modules are regarded as abnormal events of concern, and the time series data associated with the abnormal events of concern are screened out from the historical operation time series data of PV modules as the monitoring target data of the abnormal events of concern;

[0023] Based on the monitoring target parameters of the abnormal events of concern, the abnormal events of concern are taken as the root node, and the abnormal monitoring target data are taken as the leaf nodes. A decision tree corresponding to each abnormal event of concern is built and combined into an abnormal recognition model that affects the output of photovoltaic modules;

[0024] Substitute the operating data of the distributed photovoltaic distribution network into the anomaly recognition model that affects the output of photovoltaic modules, and perform several rounds of recursive division until all the monitoring target data in each sub-leaf node belong to the same category and have the same attribute values, and obtain the probability of occurrence of the predicted abnormal event of concern;

[0025] Analyze the loss value of the corresponding photovoltaic module under the probability of occurrence of each predicted abnormal event;

[0026] Based on the cumulative operation time of the distributed photovoltaic distribution network, determine the basic loss value of each photovoltaic module;

[0027] Based on the probability of occurrence of abnormal events of concern, the loss value of the corresponding photovoltaic components under each predicted probability of occurrence of abnormal events of concern, and the basic loss of each photovoltaic component, the operating status of each distributed photovoltaic distribution network is evaluated through the photovoltaic status evaluation formula;

[0028] Among them, the photovoltaic status assessment formula is:

[0029]

[0030] Where U k is the operating status index of the kth photovoltaic module in the distributed photovoltaic distribution network, B k is the basic loss value of the kth photovoltaic module, E j is the probability of occurrence of the jth abnormal event of interest, D jk is the loss value of the kth PV module when the jth abnormal event of concern occurs, α is the influencing factor, and is assigned a value based on the probability of occurrence of the abnormal event of concern.

[0031] Preferably, acquiring the power consumption data of the power consumption terminal, analyzing the power load change trend in the power consumption data, and evaluating the power load of the power consumption terminal specifically includes:

[0032] Based on LSTM long short-term memory network, a terminal load prediction model is constructed;

[0033] Based on the power consumption data of each distributed photovoltaic distribution network supplying the power consumption end, the characteristics of the factors affecting the change of power consumption in the power consumption data are marked, and the characteristic parameters of the factors affecting the change of power consumption of each supplying the power consumption end are obtained;

[0034] The characteristic parameters of each factor affecting the change of electricity consumption supplied to the electricity user are substituted into the load prediction model of the electricity user terminal to predict the power load of the electricity user terminal in the future.

[0035] Preferably, based on a distributed photovoltaic distribution network with power redundancy value and a distributed photovoltaic distribution network with power shortage value, a distributed photovoltaic distribution network power intelligent distribution model is constructed, and the optimal dispatching strategy for generating power specifically includes:

[0036] Based on the difference calculation between the power production capacity of the distributed photovoltaic distribution network and the power load supplied to the power terminal, the power surplus or shortage state of the distributed photovoltaic distribution network at each moment is determined;

[0037] Based on the power surplus or shortage status of the distributed photovoltaic distribution network at each moment, when there is a power shortage value in the distributed photovoltaic distribution network, it is recorded as a distributed photovoltaic distribution network node to be supplemented;

[0038] Mark each dispatchable distributed photovoltaic distribution network node around the distributed photovoltaic distribution network node to be supplemented;

[0039] Calculate the dispatching cost of each dispatchable distributed photovoltaic distribution network node when distributing electricity to the distributed photovoltaic distribution network node to be supplemented;

[0040] Determine whether the power redundancy of each dispatchable distributed photovoltaic distribution network node meets the power shortage of the distributed photovoltaic distribution network node to be supplemented. If so, implement single-point dispatch; if not, implement multi-point dispatch;

[0041] Whether to implement single-point dispatch or multi-point dispatch is used as the first constraint, and the minimum dispatch cost during power distribution is used as the second constraint;

[0042] Construct an intelligent power distribution model for distributed photovoltaic distribution networks;

[0043] Based on the distributed photovoltaic distribution network power intelligent distribution model, under the first constraint and the second constraint, the power distribution cost under single-point dispatching is minimized and the power distribution cost under multi-point dispatching is minimized to generate a power distribution plan;

[0044] Among them, the expression of the distributed photovoltaic distribution network power intelligent distribution model is:

[0045]

[0046] Where C() is the cost function, q xy,t is the amount of power distributed from node x to node y at time t, V xy,t is a binary decision variable, indicating whether to transfer power from node x to node y at time t, V xy,t =1 indicates power transmission, V xy,t =0 means no power transmission, ΔR x,tis the power surplus or shortage value of node x at time t, They represent the power generation and load of node y at time t respectively, H is the total number of nodes in the distributed photovoltaic distribution network, S is the total number of indexes of dispatchable distributed photovoltaic distribution network nodes, and t is the time index.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The present invention proposes a distributed photovoltaic distribution network voltage optimization control method. By analyzing the power generation and power consumption of each distributed photovoltaic distribution network, the power generation redundancy is allocated to the photovoltaic distribution network nodes with power generation shortage, and the distributed photovoltaic distribution network voltage optimization control is realized without the help of external reactive power compensation, so as to reduce the damage to the equipment caused by the voltage increase at the power generation end when the power generation at the power generation end is greater than the power load at the power generation end. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a flow chart of a voltage optimization control method for a distributed photovoltaic distribution network;

[0050] Figure 2 A flow chart of the method for evaluating the power production capacity of a photovoltaic distribution network;

[0051] Figure 3 A flow chart of a method for evaluating the operating status of photovoltaic modules in a distributed photovoltaic distribution network;

[0052] Figure 4 A flow chart of a method for evaluating the power load of a power consumption terminal;

[0053] Figure 5 Flowchart of an optimal dispatch strategy method for generating electricity. DETAILED DESCRIPTION

[0054] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0055] Reference Figure 1 As shown, a distributed photovoltaic distribution network voltage optimization control method includes:

[0056] Obtain monitoring data of distributed photovoltaic distribution networks, analyze environmental factors affecting photovoltaic output in the monitoring data, and evaluate the power production capacity of photovoltaic distribution networks;

[0057] Obtain power consumption data of power consumption terminals, analyze the power load change trend in the power consumption data, and evaluate the power load of power consumption terminals;

[0058] Determine whether the output power of the distributed photovoltaic distribution network meets the supply and demand balance of the power load supplied to the power consumption terminal. If so, select the distributed photovoltaic distribution network nodes with power redundancy values. If not, mark the distributed photovoltaic distribution network nodes with power shortage values.

[0059] Based on the distributed photovoltaic distribution network with power redundancy value and the distributed photovoltaic distribution network with power shortage value, an intelligent power distribution model of the distributed photovoltaic distribution network is constructed to generate the optimal power dispatching strategy.

[0060] This solution determines the supply and demand balance by analyzing the power production capacity of the distributed photovoltaic distribution network and the power load of the power consumption terminal, marking the distributed grid nodes where the power production capacity of the distributed photovoltaic power grid is greater or less than the power load of the power consumption terminal, and allocating the distributed photovoltaic distribution network with redundant power generation to the distributed photovoltaic distribution network with insufficient power generation to reduce the occurrence of overvoltage characteristics caused by less power consumption at the power consumption end and more power generation at the power generation end.

[0061] Balance the power supply and demand between the power consumption terminals and reduce the occurrence of overvoltage characteristics.

[0062] Reference Figure 2 As shown in the figure, the monitoring data of the distributed photovoltaic distribution network is obtained, the environmental factors affecting the photovoltaic output in the monitoring data are analyzed, and the power production capacity of the photovoltaic distribution network is evaluated, which specifically includes:

[0063] Determining the structural parameters of the distributed photovoltaic distribution network; the structural parameters include: the area of ​​the photovoltaic module, the ideal photoelectric conversion efficiency of the photovoltaic module;

[0064] Based on the several environmental parameters in the meteorological data that affect the corresponding distributed photovoltaic distribution network at all times, as environmental influencing factors; the environmental influencing factors include: sunshine intensity, haze, sand and dust, temperature and humidity;

[0065] Analyze the power production efficiency of the photovoltaic distribution network based on the structural parameters and environmental impact factors of the distributed photovoltaic distribution network;

[0066] Based on the operation data of the distributed photovoltaic distribution network, the operation status of the photovoltaic components of the distributed photovoltaic distribution network is evaluated;

[0067] The power production efficiency of the photovoltaic distribution network is corrected by the operating state of the distributed photovoltaic distribution network to obtain the power production capacity of the photovoltaic distribution network;

[0068] Among them, the calculation expression of the power capacity of the photovoltaic distribution network is:

[0069]

[0070] Where P is the power capacity of the photovoltaic distribution network, F is the structural parameters of the photovoltaic module, Gi is the i-th environmental impact factor corresponding to the distributed photovoltaic distribution network, U k is the operating status index of the kth PV module in the distributed PV distribution network, and L is the total number of PV modules.

[0071] It is understandable that by analyzing the current state of the PV modules and making corrections to the power production efficiency, the accuracy of the analysis and prediction of future power generation can be guaranteed to the greatest extent. Secondly, when the state of the PV modules is poor, the power generation fluctuates greatly. This part of the fluctuating power generation cannot be considered in the overall scheduling to ensure the stability of each scheduling.

[0072] Reference Figure 3 As shown in the figure, based on the operating data of the distributed photovoltaic distribution network, the evaluation of the operating status of the photovoltaic components of the distributed photovoltaic distribution network specifically includes:

[0073] Based on the maintenance task index of historical maintenance data, the historical maintenance data of the distributed photovoltaic distribution network is divided and reorganized according to the maintenance timestamp to obtain the historical operation time series data of the photovoltaic components;

[0074] The historical maintenance tasks that affect the output of PV modules are regarded as abnormal events of concern, and the time series data associated with the abnormal events of concern are screened out from the historical operation time series data of PV modules as the monitoring target data of the abnormal events of concern;

[0075] Based on the monitoring target parameters of the abnormal events of concern, the abnormal events of concern are taken as the root node, and the abnormal monitoring target data are taken as the leaf nodes. A decision tree corresponding to each abnormal event of concern is built and combined into an abnormal recognition model that affects the output of photovoltaic modules;

[0076] Substitute the operating data of the distributed photovoltaic distribution network into the anomaly recognition model that affects the output of photovoltaic modules, and perform several rounds of recursive division until all the monitoring target data in each sub-leaf node belong to the same category and have the same attribute values, and obtain the probability of occurrence of the predicted abnormal event of concern;

[0077] Analyze the loss value of the corresponding photovoltaic module under the probability of occurrence of each predicted abnormal event;

[0078] Based on the cumulative operation time of the distributed photovoltaic distribution network, determine the basic loss value of each photovoltaic module;

[0079] Based on the probability of occurrence of abnormal events of concern, the loss value of the corresponding photovoltaic components under each predicted probability of occurrence of abnormal events of concern, and the basic loss of each photovoltaic component, the operating status of each distributed photovoltaic distribution network is evaluated through the photovoltaic status evaluation formula;

[0080] Among them, the photovoltaic status assessment formula is:

[0081]

[0082] Where U k is the operating status index of the kth photovoltaic module in the distributed photovoltaic distribution network, B k is the basic loss value of the kth photovoltaic module, E j is the probability of occurrence of the jth abnormal event of interest, D jk is the loss value of the kth PV module when the jth abnormal event of concern occurs, α is the influencing factor, and is assigned a value based on the probability of occurrence of the abnormal event of concern.

[0083] It should be noted that in the process of constructing the random forest, the use of Bikini minimum or information entropy increase for the split nodes can meet the classification of the monitoring target data in this scheme, and the pre-pruning in the construction of the random forest can be preset according to different events, which will not be described in detail here.

[0084] Reference Figure 4 As shown, obtaining the power consumption data of the power consumption terminal, analyzing the power load change trend in the power consumption data, and evaluating the power load of the power consumption terminal specifically include:

[0085] Based on LSTM long short-term memory network, a terminal load prediction model is constructed;

[0086] Based on the power consumption data of each distributed photovoltaic distribution network supplying the power consumption end, the characteristics of the factors affecting the change of power consumption in the power consumption data are marked, and the characteristic parameters of the factors affecting the change of power consumption of each supplying the power consumption end are obtained;

[0087] The characteristic parameters of each factor affecting the change of electricity consumption supplied to the electricity user are substituted into the load prediction model of the electricity user terminal to predict the power load of the electricity user terminal in the future.

[0088] Reference Figure 5 As shown in the figure, based on the distributed photovoltaic distribution network with power redundancy value and the distributed photovoltaic distribution network with power shortage value, a distributed photovoltaic distribution network power intelligent distribution model is constructed, and the optimal dispatching strategy for generating power specifically includes:

[0089] Based on the difference calculation between the power production capacity of the distributed photovoltaic distribution network and the power load supplied to the power terminal, the power surplus or shortage state of the distributed photovoltaic distribution network at each moment is determined;

[0090] Based on the power surplus or shortage status of the distributed photovoltaic distribution network at each moment, when there is a power shortage value in the distributed photovoltaic distribution network, it is recorded as a distributed photovoltaic distribution network node to be supplemented;

[0091] Mark each dispatchable distributed photovoltaic distribution network node around the distributed photovoltaic distribution network node to be supplemented;

[0092] Calculate the dispatching cost of each dispatchable distributed photovoltaic distribution network node when distributing electricity to the distributed photovoltaic distribution network node to be supplemented;

[0093] Determine whether the power redundancy of each dispatchable distributed photovoltaic distribution network node meets the power shortage of the distributed photovoltaic distribution network node to be supplemented. If so, implement single-point dispatch; if not, implement multi-point dispatch;

[0094] Whether to implement single-point dispatch or multi-point dispatch is used as the first constraint, and the minimum dispatch cost during power distribution is used as the second constraint;

[0095] Construct an intelligent power distribution model for distributed photovoltaic distribution networks;

[0096] Based on the distributed photovoltaic distribution network power intelligent distribution model, under the first constraint and the second constraint, the power distribution cost under single-point dispatching is minimized and the power distribution cost under multi-point dispatching is minimized to generate a power distribution plan;

[0097] Among them, the expression of the distributed photovoltaic distribution network power intelligent distribution model is:

[0098]

[0099] Where C() is the cost function, q xy,t is the amount of power distributed from node x to node y at time t, V xy,t is a binary decision variable, indicating whether to transfer power from node x to node y at time t, V xy,t =1 indicates power transmission, V xy,t =0 means no power transmission, ΔR x,t is the power surplus or shortage value of node x at time t, They represent the power generation and load of node y at time t respectively, H is the total number of nodes in the distributed photovoltaic distribution network, S is the total number of indexes of dispatchable distributed photovoltaic distribution network nodes, and t is the time index.

[0100] It can be understood that by analyzing the power generation and power consumption of each distributed photovoltaic distribution network, the power generation redundancy is allocated to the photovoltaic distribution network nodes with power generation shortages to reduce the damage to equipment caused by the voltage increase at the power generation end when the power generation at the power generation end is greater than the power load at the power end.

[0101] It should be noted that the distributed photovoltaic distribution network in this scheme can still be used as an implementation plan at nodes with storage capacity. When the power generation of the distributed photovoltaic distribution network is greater than the storage capacity, power distribution is carried out to achieve voltage optimization control of the distributed photovoltaic distribution network without the help of external reactive power compensation.

[0102] In summary, the advantages of the present invention are: by analyzing the power generation and power consumption of each distributed photovoltaic distribution network, the power generation redundancy is allocated to the photovoltaic distribution network nodes with power generation shortage, and the voltage optimization control of the distributed photovoltaic distribution network is realized without the help of smart inverters, so as to reduce the damage to the equipment caused by the voltage increase at the power generation end when the power generation at the power generation end is greater than the power load at the power end.

[0103] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A distributed photovoltaic distribution network voltage optimization control method, characterized in that: include: Obtain monitoring data of distributed photovoltaic distribution networks, analyze environmental factors affecting photovoltaic output in the monitoring data, and evaluate the power production capacity of photovoltaic distribution networks; Obtain power consumption data of power consumption terminals, analyze the power load change trend in the power consumption data, and evaluate the power load of power consumption terminals; Determine whether the output power of the distributed photovoltaic distribution network meets the supply and demand balance of the power load supplied to the power consumption terminal. If so, select the distributed photovoltaic distribution network nodes with power redundancy values. If not, mark the distributed photovoltaic distribution network nodes with power shortage values. Based on the distributed photovoltaic distribution network with power redundancy value and the distributed photovoltaic distribution network with power shortage value, an intelligent power distribution model of the distributed photovoltaic distribution network is constructed to generate the optimal power dispatching strategy.

2. A distributed photovoltaic distribution network voltage optimization control method according to claim 1, characterized in that: Obtain monitoring data of distributed photovoltaic distribution networks, analyze environmental factors affecting photovoltaic output in the monitoring data, and evaluate the power production capacity of photovoltaic distribution networks, including: Determining the structural parameters of the distributed photovoltaic distribution network; the structural parameters include: the area of ​​the photovoltaic module, the ideal photoelectric conversion efficiency of the photovoltaic module; Based on the several environmental parameters in the meteorological data that affect the corresponding distributed photovoltaic distribution network at all times, as environmental influencing factors; the environmental influencing factors include: sunshine intensity, haze, sand and dust, temperature and humidity; Analyze the power production efficiency of the photovoltaic distribution network based on the structural parameters and environmental impact factors of the distributed photovoltaic distribution network; Based on the operation data of the distributed photovoltaic distribution network, the operation status of the photovoltaic components of the distributed photovoltaic distribution network is evaluated; The operating status of the distributed photovoltaic distribution network is used to correct the power production efficiency of the photovoltaic distribution network to obtain the power production capacity of the photovoltaic distribution network.

3. A distributed photovoltaic distribution network voltage optimization control method according to claim 2, characterized in that: The power capacity calculation expression of the photovoltaic distribution network is: Where P is the power capacity of the photovoltaic distribution network, F is the structural parameters of the photovoltaic module, G i is the i-th environmental impact factor corresponding to the distributed photovoltaic distribution network, U k is the operating status index of the kth PV module in the distributed PV distribution network, and L is the total number of PV modules.

4. A distributed photovoltaic distribution network voltage optimization control method according to claim 3, characterized in that: Based on the operating data of the distributed photovoltaic distribution network, the evaluation of the operating status of the photovoltaic components of the distributed photovoltaic distribution network specifically includes: Based on the maintenance task index of historical maintenance data, the historical maintenance data of the distributed photovoltaic distribution network is divided and reorganized according to the maintenance timestamp to obtain the historical operation time series data of the photovoltaic components; The historical maintenance tasks that affect the output of PV modules are regarded as abnormal events of concern, and the time series data associated with the abnormal events of concern are screened out from the historical operation time series data of PV modules as the monitoring target data of the abnormal events of concern; Based on the monitoring target parameters of the abnormal events of concern, the abnormal events of concern are taken as the root node, and the abnormal monitoring target data are taken as the leaf nodes. A decision tree corresponding to each abnormal event of concern is built and combined into an abnormal recognition model that affects the output of photovoltaic modules; Substitute the operating data of the distributed photovoltaic distribution network into the anomaly recognition model that affects the output of photovoltaic modules, and perform several rounds of recursive division until all the monitoring target data in each sub-leaf node belong to the same category and have the same attribute values, and obtain the probability of occurrence of the predicted abnormal event of concern; Analyze the loss value of the corresponding photovoltaic module under the probability of occurrence of each predicted abnormal event; Based on the cumulative operation time of the distributed photovoltaic distribution network, determine the basic loss value of each photovoltaic module; Based on the probability of occurrence of abnormal events of concern, the loss value of the corresponding PV module under each predicted probability of occurrence of abnormal events of concern and the basic loss of each PV module, the operating status of each distributed PV distribution network is evaluated through the photovoltaic status evaluation formula.

5. A distributed photovoltaic distribution network voltage optimization control method according to claim 4, characterized in that: The photovoltaic status assessment formula is: Where U k is the operating status index of the kth photovoltaic module in the distributed photovoltaic distribution network, B k is the basic loss value of the kth photovoltaic module, E j is the probability of occurrence of the jth abnormal event of interest, D jk is the loss value of the kth PV module when the jth abnormal event of concern occurs, α is the influencing factor, and is assigned a value based on the probability of occurrence of the abnormal event of concern.

6. A distributed photovoltaic distribution network voltage optimization control method according to claim 5, characterized in that: Obtaining the power consumption data of the power consumption terminal, analyzing the power load change trend in the power consumption data, and evaluating the power load of the power consumption terminal specifically include: Based on LSTM long short-term memory network, a terminal load prediction model is constructed; Based on the power consumption data of each distributed photovoltaic distribution network supplying the power consumption end, the characteristics of the factors affecting the change of power consumption in the power consumption data are marked, and the characteristic parameters of the factors affecting the change of power consumption of each supplying the power consumption end are obtained; The characteristic parameters of each factor affecting the change of electricity consumption supplied to the electricity user are substituted into the load prediction model of the electricity user terminal to predict the power load of the electricity user terminal in the future.

7. A distributed photovoltaic distribution network voltage optimization control method according to claim 6, characterized in that: Based on the distributed photovoltaic distribution network with power redundancy value and the distributed photovoltaic distribution network with power shortage value, a distributed photovoltaic distribution network power intelligent distribution model is constructed to generate the optimal dispatching strategy of power, which specifically includes: Based on the difference calculation between the power production capacity of the distributed photovoltaic distribution network and the power load supplied to the power terminal, the power surplus or shortage state of the distributed photovoltaic distribution network at each moment is determined; Based on the power surplus or shortage status of the distributed photovoltaic distribution network at each moment, when there is a power shortage value in the distributed photovoltaic distribution network, it is recorded as a distributed photovoltaic distribution network node to be supplemented; Mark each dispatchable distributed photovoltaic distribution network node around the distributed photovoltaic distribution network node to be supplemented; Calculate the dispatching cost of each dispatchable distributed photovoltaic distribution network node when distributing electricity to the distributed photovoltaic distribution network node to be supplemented; Determine whether the power redundancy of each dispatchable distributed photovoltaic distribution network node meets the power shortage of the distributed photovoltaic distribution network node to be supplemented. If so, implement single-point dispatch; if not, implement multi-point dispatch; Whether to implement single-point dispatch or multi-point dispatch is used as the first constraint, and the minimum dispatch cost during power distribution is used as the second constraint; Construct an intelligent power distribution model for distributed photovoltaic distribution networks; Based on the intelligent power distribution model of distributed photovoltaic distribution network, the power distribution plan is generated by minimizing the dispatching cost of power distribution under single-point dispatch and the dispatching cost of power distribution under multi-point dispatch under the first constraint and the second constraint.

8. A distributed photovoltaic power distribution network voltage optimization control method according to claim 7, characterized in that: The expression of the distributed photovoltaic distribution network power intelligent distribution model is: Where C() is the cost function, q xy，t is the amount of power distributed from node x to node y at time t, V xy，t is a binary decision variable, indicating whether to transfer power from node x to node y at time t, V xy，t =1 indicates power transmission, V xy，t =0 means no power transmission, ΔR x，t is the power surplus or shortage value of node x at time t, They represent the power generation and load of node y at time t respectively, H is the total number of nodes in the distributed photovoltaic distribution network, S is the total number of indexes of dispatchable distributed photovoltaic distribution network nodes, and t is the time index.

Images