Optimized scheduling method for power distribution network and distributed power supply grid connection

Abstract

The invention discloses an optimal scheduling method for grid connection of a power distribution network and a distributed power supply, and particularly relates to the technical field of optimal scheduling, which comprises the following steps of: determining employment performance information during grid connection of the distributed photovoltaic power supply according to the difference between real-time and historical solar photovoltaic output models; based on charge state data of an energy storage system connected with the distributed photovoltaic power supply, determining energy storage performance information during grid connection of the distributed photovoltaic power supply, determining a grid connection strategy of the distributed photovoltaic power supply, and determining a grid connection strategy of the distributed photovoltaic power supply based on consideration of a node system topology connection relation of the power distribution network and related data of nodes and branches; possible risks existing in distributed photovoltaic power grid-connected nodes in the power distribution network are simulated, grid-connected risk information of the power distribution network is determined, the matching degree of output power and required power of the distributed photovoltaic power is considered, grid-connected load matching information of the power distribution network is determined, and the overall grid-connected performance of the power distribution network is quantified. According to the invention, stable operation of the power distribution network can be ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of optimized scheduling, and more specifically, to a method for optimized scheduling of a power distribution network and a distributed power source grid connection. Background Art

[0002] With the rapid development of renewable energy, distributed photovoltaic power sources (including household photovoltaic systems and small photovoltaic power plants) have gradually become an important component of modern distribution networks. Traditional photovoltaic power grid-connection strategies often address the volatility of photovoltaic output through simple technical standards or excessive reliance on energy storage systems. However, these approaches often ignore the complex interactions between photovoltaic power sources, energy storage systems, and distribution networks, as well as the changing external conditions. Existing technologies make it difficult to easily select photovoltaic power sources suitable for direct grid connection, and it is difficult to consider the impact of photovoltaic power sources on distribution network stability. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an optimization scheduling method for a distribution network and a distributed power source grid connection, so as to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A method for optimizing the scheduling of a distribution network and a distributed power supply grid connection comprises the following steps:

[0006] S1: Based on historical data of distributed photovoltaic power sources, determine the solar photovoltaic output model. By comparing the difference between the real-time and historical solar photovoltaic output models, determine the labor performance information of the distributed photovoltaic power source when it is connected to the grid. Based on the state of charge data of the energy storage system connected to the distributed photovoltaic power source, determine the energy storage performance information of the distributed photovoltaic power source when it is connected to the grid.

[0007] S2: Evaluate the grid-connected performance of distributed PV power sources and determine their grid-connected strategy by comprehensively analyzing their labor performance and energy storage performance during grid-connection.

[0008] S3: Based on the node-related data of the distribution network, simulate the possible risks of the distributed photovoltaic power grid-connected nodes in the distribution network, determine the grid-connected risk information of the distribution network, and consider the matching degree between the output power of the distributed photovoltaic power and the required power to determine the grid-connected load matching information of the distribution network;

[0009] S4: When using the same type of distributed photovoltaic power sources to access the distribution network, the grid-connected risk information and grid-connected load matching information of the distribution network are integrated and evaluated to quantify the overall grid-connected performance of the distribution network, generate early warning signals, and optimize the scheduling of distributed photovoltaic power sources.

[0010] In a preferred embodiment, the employment performance information and energy storage performance information of the distributed photovoltaic power source when connected to the grid include:

[0011] The labor performance information of distributed photovoltaic power generation when connected to the grid is represented by the output distribution difference coefficient, and the energy storage performance information of distributed photovoltaic power generation when connected to the grid is represented by the energy storage change fluctuation coefficient, where: is the output distribution difference coefficient, is the energy storage fluctuation coefficient.

[0012] In a preferred embodiment, the logic for obtaining the output distribution difference coefficient is:

[0013] Based on the historical data of different distributed photovoltaic power sources, the probability density functions of different distributed photovoltaic power sources in different scenarios are determined, and the probability density functions of distributed photovoltaic power sources in the monitoring interval under different scenarios are calculated. The calculation formula is: ;in, is the probability density function of distributed photovoltaic power in the monitoring interval, g is the actual light intensity at a certain moment, is the maximum light intensity in the monitoring area, is the Gamma function, , is the parameter of Beta distribution, , , is the mean value of sunlight intensity in the monitoring interval, is the variance of sunlight intensity within the monitoring interval;

[0014] By obtaining the photoelectric conversion efficiency of different distributed photovoltaic power sources and the area of ​​photovoltaic solar panels, the output power of different distributed photovoltaic power supply photovoltaic power generation systems is determined. The calculation formula for the output power of different distributed photovoltaic power supply photovoltaic power generation systems is: ; Among them, P is the output power of the photovoltaic power generation system, A is the area of ​​the photovoltaic solar panel, is the photoelectric conversion efficiency;

[0015] According to the probability distribution model of solar photovoltaic, it can be concluded that the output model of solar photovoltaic approximately obeys the Beta distribution. The probability distribution model of solar photovoltaic is expressed as: ;in, is the probability distribution model of historical solar photovoltaics, is the maximum output power of solar photovoltaic power generation under the light intensity within the monitoring interval;

[0016] By real-time acquisition of the probability distribution model of different distributed photovoltaic power sources within the monitoring interval, the probability distribution model of different distributed photovoltaic power sources within the real-time acquisition monitoring interval is marked as: ;

[0017] The Kullback-Leibler divergence is used to compare the real-time probability distribution model of different distributed photovoltaic power sources with the probability distribution model of historical solar photovoltaic power generation, and the output distribution difference coefficient is calculated. The calculation formula is: .

[0018] In a preferred embodiment, the logic for obtaining the energy storage change fluctuation coefficient is:

[0019] Based on the energy storage system connected to the distributed photovoltaic power generation, the state of charge change differential equation of the distribution network energy storage system is obtained. The state of charge change rate of the distribution network energy storage system is calculated based on the state of charge change differential equation, and the state of charge change rate is marked as: , where t=1, 2, 3, ..., T, T is a positive integer, and t is the number of different moments in the monitoring interval;

[0020] The difference equation for the charge state change of the distribution network energy storage system is expressed as: ;in, represents the state of charge at time t, represents the self-discharge rate, 、 Respectively represent the charge and discharge flags of the energy storage system, Represents the actual charging power of the energy storage system, Represents the actual discharge power of the energy storage system, and Represent the charging and discharging efficiency of the energy storage system, Represents the capacity of the energy storage system;

[0021] Obtain the average value and standard deviation of the state of charge change rate within the monitoring interval, and mark the average value and standard deviation of the state of charge change rate within the monitoring interval as: and ,in, , ;

[0022] Calculate the energy storage change fluctuation coefficient using the following formula: .

[0023] In a preferred embodiment, evaluating the grid-connected performance of distributed photovoltaic power sources includes:

[0024] A comprehensive analysis of the labor performance information and energy storage performance information of distributed photovoltaic power generation when connected to the grid is conducted. By performing weighted calculations on the normalized output distribution difference coefficient and the energy storage change fluctuation coefficient, a power supply evaluation model is constructed to generate a power supply evaluation coefficient. The calculation formula for the power supply evaluation coefficient is: ;in, is the power supply evaluation coefficient, 、 are the proportional coefficients of output distribution difference coefficient and energy storage change fluctuation coefficient, respectively. 、 Both are greater than 0.

[0025] In a preferred embodiment, determining the grid connection risk information and grid connection load matching information of the distribution network includes:

[0026] The grid-connected risk information of the distribution network is represented by the distributed photovoltaic risk coefficient, and the grid-connected load matching information of the distribution network is represented by the load mismatch coefficient, where: is the risk coefficient of distributed photovoltaics, is the load mismatch coefficient.

[0027] In a preferred embodiment, the logic for obtaining the distributed photovoltaic risk coefficient is:

[0028] Based on the node locations of the distribution network, distributed photovoltaic power sources connected to the grid are connected at the node locations. That is, the node topology of the distribution network is modeled, and the distributed photovoltaic power sources connected to different nodes are calibrated. By simulating different fault scenarios, the fault set and failure rate caused by the access of distributed photovoltaic power sources at different node locations are obtained. In other words, Monte Carlo simulation is performed on different nodes and the access conditions of distributed photovoltaic power sources to randomly sample different system states and fault conditions, estimate the probability of failure in each scenario, and determine the fault impact score of the distributed photovoltaic power source connected to the grid at the node location;

[0029] Based on the current operation status of the distribution network, the distributed photovoltaic power nodes that have been connected to the grid in the current distribution network are determined, and the load at the distributed photovoltaic power nodes is obtained. The load at different types of nodes and the fault impact score of different types of nodes connected to distributed photovoltaic power sources are used as input features, and the distributed photovoltaic risk coefficient is used as the output feature to construct a regression model. The calculation formula of the distributed photovoltaic risk coefficient is:

[0030] ; Among them, 1, 2, 3, ..., J are the numbers of different types of nodes, is the load at different types of nodes, The fault impact scores of different types of nodes connected to distributed photovoltaic power sources are calculated. When the node is not connected to distributed photovoltaic power sources, the fault impact score is marked as a fixed constant. is the weight of different types of nodes, and e is the base.

[0031] In a preferred embodiment, the load mismatch coefficient is obtained by:

[0032] Obtain the demand load at the distributed photovoltaic power grid-connected node within a unit time period, and mark the demand load at the distributed photovoltaic power grid-connected node within a unit time period as: ; Wherein, i=1, 2, 3, ..., 1, I is a positive integer, i is the number of different time points in a unit time period, n=1, 2, 3, ..., N, N is a positive integer, n is the number of the distributed photovoltaic power grid-connected node;

[0033] Obtain the output power of the distributed photovoltaic power source at the distributed photovoltaic power source grid-connected node within a unit time period, and mark the output power of the distributed photovoltaic power source at the distributed photovoltaic power source grid-connected node within a unit time period as: ;

[0034] Determine the load matching degree of a single distributed photovoltaic power grid-connected node within a unit time period. The calculation formula is: ;in, is the load matching degree of the nth distributed photovoltaic power grid-connected node;

[0035] Calculate the load mismatch coefficient using the following formula: .

[0036] In a preferred embodiment, quantifying the overall grid-connected performance of the distribution network and generating an early warning signal includes:

[0037] A comprehensive analysis of the grid-connected risk information and grid-connected load matching information of the distribution network is conducted. By weighted calculation of the normalized distributed photovoltaic risk coefficient and load mismatch coefficient, a distribution network optimization evaluation model is constructed to generate the distribution network optimization evaluation coefficient. The calculation formula of the distribution network optimization evaluation coefficient is: ;in, Optimization evaluation coefficient for distribution network, 、 are the proportional coefficients of distributed photovoltaic risk coefficient and load mismatch coefficient, 、 Both are greater than 0.

[0038] Technical effects and advantages of the present invention:

[0039] The present invention combines multiple factors such as historical data of photovoltaic power sources, the charge state of the energy storage system, and the topological structure of the distribution network to comprehensively evaluate the risks and load matching of distributed photovoltaic power grid connection, and then optimize the photovoltaic power grid connection strategy. This includes establishing a solar photovoltaic output model based on historical data and real-time monitoring data, and evaluating the possible risks of distributed photovoltaic power grid connection by simulating different fault scenarios based on the node topology of the distribution network and related branch data. The present invention helps to ensure the stable operation of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0041] Figure 1 The present invention is a flow chart of an optimization scheduling method for connecting a distribution network and distributed power sources to the grid. DETAILED DESCRIPTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0043] Example 1

[0044] Figure 1 The present invention is a flow chart of an optimization scheduling method for a distribution network and a distributed power supply grid connection, which specifically includes the following steps:

[0045] S1: Based on historical data of distributed photovoltaic power sources, determine the solar photovoltaic output model. By comparing the difference between the real-time and historical solar photovoltaic output models, determine the labor performance information of the distributed photovoltaic power source when it is connected to the grid. Based on the state of charge data of the energy storage system connected to the distributed photovoltaic power source, determine the energy storage performance information of the distributed photovoltaic power source when it is connected to the grid.

[0046] S2: Evaluate the grid-connected performance of distributed PV power sources and determine their grid-connected strategy by comprehensively analyzing their labor performance and energy storage performance during grid-connection.

[0047] S3: Based on the node-related data of the distribution network, simulate the possible risks of the distributed photovoltaic power grid-connected nodes in the distribution network, determine the grid-connected risk information of the distribution network, and consider the matching degree between the output power of the distributed photovoltaic power and the required power to determine the grid-connected load matching information of the distribution network;

[0048] S4: When using the same type of distributed photovoltaic power sources to access the distribution network, the grid-connected risk information and grid-connected load matching information of the distribution network are integrated and evaluated to quantify the overall grid-connected performance of the distribution network, generate early warning signals, and optimize the scheduling of distributed photovoltaic power sources.

[0049] The relationship between distributed photovoltaic power sources, energy storage systems, and distribution networks is a complex system that requires multi-party coordination and interdependence. When the photovoltaic power source performs well, can generate stable power, and meets grid connection requirements (such as voltage and current fluctuation range), direct grid connection is the simplest and most economical option. If the photovoltaic power source's power generation fluctuates greatly, especially during weather changes (such as cloudy days, cloud cover, etc.), the energy storage system can smooth out power fluctuations, ensure stable power output, and reduce the impact on the grid. Therefore, the system can dynamically choose whether the distributed photovoltaic power source should be connected to the grid through the energy storage system or directly.

[0050] The labor performance information and energy storage performance information of the distributed photovoltaic power generation when it is connected to the grid are collected. The labor performance information of the distributed photovoltaic power generation when it is connected to the grid is represented by the output distribution difference coefficient, and the energy storage performance information of the distributed photovoltaic power generation when it is connected to the grid is represented by the energy storage change fluctuation coefficient.

[0051] The advantages of the output distribution difference coefficient are:

[0052] The output distribution difference coefficient can provide a quantitative assessment method by quantifying the difference between the real-time output of distributed photovoltaic power sources and historical data. By comparing the output differences in different time periods, it can reveal whether the photovoltaic power source has experienced abnormal fluctuations or abnormal behavior.

[0053] By comparing it with historical data, the output distribution variance coefficient can not only identify abnormal behavior of PV power sources within specific monitoring intervals, but also reveal the consistency of their long-term output characteristics. Stable PV power sources can maintain a relatively consistent output pattern under different meteorological conditions, thus avoiding unnecessary impacts on the power grid. This is very important for grid scheduling and stability, especially when considering large-scale grid connection. Stable output can reduce grid frequency and voltage.

[0054] The output distribution variance coefficient can help identify potential issues with PV power sources during real-time monitoring. A large variance coefficient can indicate significant changes in the PV power source's output characteristics, providing early warning of equipment failures, shading issues, and other issues. This early warning mechanism helps prevent grid frequency fluctuations and voltage instability caused by unstable PV power sources, thereby improving grid connection security.

[0055] The modeling method of photovoltaic output in the prior art usually adopts a statistical modeling method, that is, assuming the renewable energy output uncertainty as a statistical model, solving the specific parameters of the model through historical data, and obtaining a specific output scene by combining a sampling method. The distributed power output modeling method based on the statistical model mainly includes a probability modeling method, a Markov chain method, a time series method, and a scene tree generation method.

[0056] The acquisition logic of the output distribution difference coefficient is: based on historical data of different distributed photovoltaic power sources, determining probability density functions of different distributed photovoltaic power sources under different scenes, and calculating the probability density functions of the distributed photovoltaic power sources in the monitoring interval under different scenes, and the calculation formula is: ; wherein, is the probability density function of the distributed photovoltaic power source in the monitoring interval, g is the actual light intensity at a certain moment, is the maximum light intensity in the monitoring interval, is a Gamma function, , is a parameter of the Beta distribution, , , is the mean of the sunlight intensity in the monitoring interval, is the variance of the sunlight intensity in the monitoring interval;

[0057] It should be noted that the output of photovoltaic power sources with the same geographical position will be significantly affected by meteorological conditions. Even if the photovoltaic power stations are of the same type, the power sources located in different positions may exhibit different output characteristics. By analyzing different photovoltaic power sources and their output performances under different scenes, the output fluctuations and characteristics of each photovoltaic power source can be more accurately captured.

[0058] The monitoring interval is set by professional staff, and is usually 60 minutes. By setting a fixed monitoring interval, the performance of the photovoltaic power source can be evaluated in the interval. If the output performance of the photovoltaic system in a certain time period is significantly different from the expected value, and the difference does not conform to the probability distribution of the historical performance (for example, the difference between the predicted and actual light intensity is too large), it may indicate that the photovoltaic power source has an abnormal condition, such as equipment failure, shading, connection problem, etc. The output performance in the fixed monitoring interval can be compared with the historical data. By establishing a long-term data set, the output performance of the photovoltaic system under the same meteorological conditions and the same time period is analyzed, so as to evaluate its consistency.

[0059] By obtaining the photoelectric conversion efficiency of different distributed photovoltaic power sources and the area of the photovoltaic solar panel, the output power of the photovoltaic power generation system of different distributed photovoltaic power sources is determined, and the calculation formula of the output power of the photovoltaic power generation system of different distributed photovoltaic power sources is: ; Among them, P is the output power of the photovoltaic power generation system, A is the area of ​​the photovoltaic solar panel, is the photoelectric conversion efficiency;

[0060] It should be noted that the output power of a photovoltaic power generation system is a function model of the sunlight intensity felt by the solar panels.

[0061] According to the probability distribution model of solar photovoltaic, it can be concluded that the output model of solar photovoltaic approximately obeys the Beta distribution. The probability distribution model of solar photovoltaic is expressed as: ;in, is the probability distribution model of historical solar photovoltaics, is the maximum output power of solar photovoltaic power generation under the light intensity within the monitoring interval;

[0062] By real-time acquisition of the probability distribution model of different distributed photovoltaic power sources within the monitoring interval, the probability distribution model of different distributed photovoltaic power sources within the real-time acquisition monitoring interval is marked as: ;

[0063] The Kullback-Leibler divergence is used to compare the real-time probability distribution model of different distributed photovoltaic power sources with the probability distribution model of historical solar photovoltaic power generation, and the output distribution difference coefficient is calculated. The calculation formula is: ;in, is the output distribution difference coefficient.

[0064] As the formula shows, a larger output distribution variance coefficient indicates a significant deviation between the current PV system's performance and historical data, suggesting a possible anomaly or fault. For example, system failures, inverter issues, shading, wiring problems, and other factors can cause a large discrepancy between the real-time and historical distributions, indicating unstable power output from the distributed PV power source. This may require regulation through an energy storage system to reduce impact on the grid. Conversely, a smaller output distribution variance coefficient indicates that the current PV system's output fluctuations are consistent with historical data, indicating normal system operation.

[0065] The distribution network uses an intelligent control system to coordinate the operation of photovoltaic power sources and energy storage systems. It should be noted that the system can automatically adjust the energy storage system's charging and discharging strategies based on parameters such as real-time power demand, photovoltaic power generation, and energy storage status to ensure grid load balance and stable power quality. When the power generated by the photovoltaic power generation system exceeds local demand, the system can feed the excess power back to the distribution network, maximizing power utilization through the energy storage system or direct grid connection.

[0066] If the energy storage system has large fluctuations, it may cause instability in the voltage and frequency of the power grid. In particular, the output of photovoltaic power sources itself is also volatile. When this fluctuation is superimposed on the fluctuation of the energy storage system, it may cause the load of the power grid to be unbalanced, or even cause a power grid failure.

[0067] Frequent charging and discharging of energy storage systems increases the system's workload, potentially accelerating battery aging and affecting the system's long-term stability. When photovoltaic power is connected to the grid, the energy storage system is supposed to provide stable power output by smoothing out photovoltaic fluctuations. However, if the energy storage system itself is unstable, this regulatory role may be weakened.

[0068] The logic for obtaining the energy storage change fluctuation coefficient is as follows: based on the energy storage system connected to the distributed photovoltaic power source, the state of charge change differential equation of the distribution network energy storage system is obtained, and the state of charge change rate of the distribution network energy storage system is calculated based on the state of charge change differential equation, and the state of charge change rate is marked as: , where t=1, 2, 3, ..., T, T is a positive integer, and t is the number of different moments in the monitoring interval;

[0069] The difference equation for the charge state change of the distribution network energy storage system is expressed as: ;in, represents the state of charge at time t, represents the self-discharge rate, 、 Respectively represent the charge and discharge flags of the energy storage system, Represents the actual charging power of the energy storage system, Represents the actual discharge power of the energy storage system, and Represent the charging and discharging efficiency of the energy storage system, Represents the capacity of the energy storage system;

[0070] Obtain the average value and standard deviation of the state of charge change rate within the monitoring interval, and mark the average value and standard deviation of the state of charge change rate within the monitoring interval as: and ,in, , ;

[0071] Calculate the energy storage change fluctuation coefficient using the following formula: ;in, is the energy storage fluctuation coefficient.

[0072] As can be seen from the formula, the larger the energy storage fluctuation coefficient, the greater the change in the charge state of the energy storage system, the poorer the system stability, and there may be frequent charging and discharging fluctuations, or the charging and discharging efficiency of the energy storage system is unstable, affecting the stability of the power grid. Therefore, distributed photovoltaic power sources cannot be directly connected to the grid. PV power grid connection needs to be considered after the stability of the energy storage system is improved.

[0073] A comprehensive analysis of the labor performance information and energy storage performance information of distributed photovoltaic power generation when connected to the grid is conducted. By performing weighted calculations on the normalized output distribution difference coefficient and the energy storage change fluctuation coefficient, a power supply evaluation model is constructed to generate a power supply evaluation coefficient. The calculation formula for the power supply evaluation coefficient is: ;in, is the power supply evaluation coefficient, 、 are the proportional coefficients of output distribution difference coefficient and energy storage change fluctuation coefficient, respectively. 、 Both are greater than 0.

[0074] A power evaluation coefficient threshold is set, and the power evaluation coefficients of all distributed photovoltaic power sources are compared with the power evaluation coefficient threshold. If the power evaluation coefficient is greater than the power evaluation coefficient threshold, a first warning is generated for the distributed photovoltaic power source, indicating that the distributed photovoltaic power source cannot be directly connected to the grid, and the photovoltaic power generation is stored in the energy storage system. If the power evaluation coefficient is less than the power evaluation coefficient threshold, no warning is issued for the distributed photovoltaic power source, indicating that the distributed photovoltaic power source may be able to be connected to the grid through the energy storage system or directly connected to the grid.

[0075] When the same type of distributed photovoltaic power sources are connected to the distribution network, the impact of a single distributed photovoltaic power source on the overall distribution network will not be considered. It is necessary to comprehensively analyze the performance of the distribution network when multiple distributed photovoltaic power sources are connected at the same time, collect the grid-connected risk information and grid-connected load matching information of the distribution network, and express the grid-connected risk information of the distribution network through the distributed photovoltaic risk coefficient, and express the grid-connected load matching information of the distribution network through the load mismatch coefficient.

[0076] Among them, the advantages of distributed photovoltaic risk coefficient are:

[0077] The distributed photovoltaic risk factor comprehensively considers factors such as the number, distribution, capacity, and fault impact score of distributed photovoltaic power sources, thereby comprehensively assessing the potential risks of photovoltaic power sources connected to the distribution network. The risk factor not only considers the output fluctuation and randomness of photovoltaic power sources, but also factors such as their fault propagation path and the difficulty of system recovery. It quantifies the specific impact of photovoltaic power generation on distribution network stability and helps identify possible grid failure modes and their propagation processes.

[0078] Whether it is a small distribution network or a large regional power grid, the distributed photovoltaic risk coefficient can be flexibly applied. It can not only be applied to distribution networks with different topologies, but also provide accurate risk assessment for different types of photovoltaic power sources (such as centralized, distributed, small photovoltaic power stations, etc.), thereby adapting to the needs of different grid sizes.

[0079] The logic for obtaining the distributed photovoltaic risk coefficient is as follows: based on the node locations of the distribution network, the distributed photovoltaic power sources connected to the grid at the node locations are connected, that is, the node topology of the distribution network is modeled, the distributed photovoltaic power sources connected to different nodes are calibrated, and by simulating different fault scenarios, the fault set and failure rate caused by the access of distributed photovoltaic power sources at different node locations are obtained. In other words, Monte Carlo simulation is performed on different nodes and the access conditions of distributed photovoltaic power sources to randomly sample different system states and fault conditions, estimate the probability of failure in each scenario, and determine the fault impact score of the distributed photovoltaic power source connected to the grid at the node location;

[0080] It should be noted that due to the reliability of the power system, the vast majority of samples are reliable and do not require load shedding. Therefore, the number of samples that cannot meet power supply requirements is relatively small. To further improve training accuracy, a correction mechanism is needed to train samples with unreliable power supply: artificially setting fault sets for specific scenarios in the network can better reflect actual engineering practices. Secondly, the probability of failure is increased to generate more fault samples. Through the above two methods, the problem of unbalanced training samples is alleviated;

[0081] The fault impact score is based on the failure probability, fault propagation path and impact range, system recovery time and difficulty, and the capacity and output characteristics of the photovoltaic power source and whether it can effectively support load demand under each fault scenario.

[0082] Based on the current operation status of the distribution network, the distributed photovoltaic power nodes that have been connected to the grid in the current distribution network are determined, and the load at the distributed photovoltaic power nodes is obtained. The load at different types of nodes and the fault impact score of different types of nodes connected to distributed photovoltaic power sources are used as input features, and the distributed photovoltaic risk coefficient is used as the output feature to construct a regression model. The calculation formula of the distributed photovoltaic risk coefficient is:

[0083] ;in, is the risk coefficient of distributed photovoltaics, 1, 2, 3, ..., J are the numbers of different types of nodes, is the load at different types of nodes, The fault impact scores of different types of nodes connected to distributed photovoltaic power sources are calculated. When the node is not connected to distributed photovoltaic power sources, the fault impact score is marked as a fixed constant. wherein, w is the weight of different types of nodes, e is the base number;

[0084] It should be noted that different types of nodes are determined based on the structure of the power distribution network, such as load nodes, power generation nodes, and balance nodes, etc. The impact of distributed photovoltaic power grid connection on the power grid system is different, and the fault impact score at the node without access to distributed photovoltaic power is obtained by training, which is usually 0. Because at this time, photovoltaic power does not participate in system operation, there is no potential impact of photovoltaic power, and the complexity of the model is reduced.

[0085] As can be seen from the formula, the greater the risk coefficient of distributed photovoltaic power, the higher the impact and risk brought by the distributed photovoltaic power connected at each node of the current power distribution network. If multiple high-risk distributed photovoltaic power is connected for a long time, the power grid may face a greater stability risk, increasing the difficulty of overload, failure, recovery, and light abandonment.

[0086] Among them, the advantage of the load mismatch degree coefficient is:

[0087] The load mismatch degree coefficient can quantify the operational stability of the power grid. A lower load mismatch degree means better coordination between the power grid's load and photovoltaic power output, and the system can operate more stably in the face of load fluctuations, avoiding problems such as overload and frequent start-stop of equipment. When photovoltaic output and load are mismatched, the power grid may need to cut off the load to ensure stability. Through the load mismatch degree coefficient, the matching situation can be identified and optimized in advance, reducing the dependence on load shedding and standby power;

[0088] When the output of photovoltaic power and the demand of load are highly matched, the power grid can more efficiently absorb photovoltaic power and reduce the phenomenon of light abandonment. When the load mismatch degree coefficient is low, photovoltaic power can be maximally absorbed by the power grid, ensuring the full use of clean energy.

[0089] The acquisition logic of the load mismatch degree coefficient is: obtaining the demand load at the distributed photovoltaic power grid connection node within a unit time period, and marking the demand load at the distributed photovoltaic power grid connection node within a unit time period as: ; wherein, i=1, 2, 3, …, I, I is a positive integer, i is the number of different time points within a unit time period, n=1, 2, 3, …, N, N is a positive integer, and n is the number of distributed photovoltaic power grid connection nodes;

[0090] Obtaining the output power of the distributed photovoltaic power at the distributed photovoltaic power grid connection node within a unit time period, and marking the output power of the distributed photovoltaic power at the distributed photovoltaic power grid connection node within a unit time period as: ;

[0091] Determine the load matching degree of the single distributed photovoltaic power grid-connected node in a unit time period, and the calculation formula is: ; wherein, is the load matching degree of the nth distributed photovoltaic power grid-connected node.

[0092] It should be noted that the unit time period is a specific length of time set by professional staff. When the photovoltaic power output is consistent with the load, the power grid does not need additional scheduling or load shedding, which can avoid frequent start-stop equipment or adjustment of other power sources, which can greatly improve the stability of the power grid.

[0093] Calculate the load mismatch coefficient, and the calculation formula is: ; wherein, is the load mismatch coefficient.

[0094] As can be seen from the formula, the smaller the load mismatch coefficient, the better the output power of the distributed photovoltaic power at the grid-connected node can meet the load demand, the higher the matching degree between the photovoltaic power and the load, and the more stable the power grid can operate, and the overall distribution network performs better.

[0095] Comprehensively analyze the grid-connected risk information and grid-connected load matching information of the distribution network, and calculate the normalized distributed photovoltaic risk coefficient and load mismatch coefficient by weighting, construct a distribution network optimization evaluation model, and generate a distribution network optimization evaluation coefficient. The calculation formula of the distribution network optimization evaluation coefficient is: ; wherein, is the distribution network optimization evaluation coefficient, , are the proportional coefficients of the distributed photovoltaic risk coefficient and the load mismatch coefficient, , respectively.

[0096] Set the distribution network optimization evaluation coefficient threshold, obtain the distribution network optimization evaluation coefficient of the distribution network in the running state, and compare the distribution network optimization evaluation coefficient with the distribution network optimization evaluation coefficient threshold. If the distribution network optimization evaluation coefficient is greater than the distribution network optimization evaluation coefficient threshold, a second warning signal is generated, indicating that the current operation of the distribution network may be poor, and measures such as load shedding or standby power supply need to be taken to alleviate the risk of unstable power grid. If the power grid optimization evaluation coefficient is less than the distribution network optimization evaluation coefficient threshold, no second warning signal is generated.

[0097] It should be noted that the weights of each coefficient in the power supply evaluation model and the distribution network optimization evaluation model are obtained through simulation training of machine learning algorithms, and the weights of each coefficient are optimized using historical data and simulation results. The power supply evaluation coefficient thresholds and the distribution network optimization evaluation coefficient thresholds are set according to historical data, experimental results, or the specific needs and standards of the power grid, and can be analyzed through statistical analysis of historical operation data.

[0098] The present invention combines multiple factors such as historical data of photovoltaic power sources, the charge state of the energy storage system, and the topological structure of the distribution network to comprehensively evaluate the risks and load matching of distributed photovoltaic power grid connection, and then optimize the photovoltaic power grid connection strategy. This includes establishing a solar photovoltaic output model based on historical data and real-time monitoring data, and evaluating the possible risks of distributed photovoltaic power grid connection by simulating different fault scenarios based on the node topology of the distribution network and related branch data. The present invention helps to ensure the stable operation of the distribution network.

[0099] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0100] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0101] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0102] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0103] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0104] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling 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 method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0105] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for optimizing the scheduling of a distribution network and distributed power supply grid connection, characterized in that: The specific steps include: S1: Based on historical data of distributed photovoltaic power sources, determine the solar photovoltaic output model. By comparing the difference between the real-time and historical solar photovoltaic output models, determine the labor performance information of the distributed photovoltaic power source when it is connected to the grid. Based on the state of charge data of the energy storage system connected to the distributed photovoltaic power source, determine the energy storage performance information of the distributed photovoltaic power source when it is connected to the grid. S2: Evaluate the grid-connected performance of distributed PV power sources and determine their grid-connected strategy by comprehensively analyzing their labor performance and energy storage performance during grid-connection. S3: Based on the node-related data of the distribution network, simulate the possible risks of the distributed photovoltaic power grid-connected nodes in the distribution network, determine the grid-connected risk information of the distribution network, and consider the matching degree between the output power of the distributed photovoltaic power and the required power to determine the grid-connected load matching information of the distribution network; S4: When using the same type of distributed photovoltaic power sources to access the distribution network, the grid-connected risk information and grid-connected load matching information of the distribution network are integrated and evaluated to quantify the overall grid-connected performance of the distribution network, generate early warning signals, and optimize the scheduling of distributed photovoltaic power sources.

2. The method for optimizing the scheduling of a distribution network and distributed power supply grid connection according to claim 1, characterized in that: The labor performance information and energy storage performance information of distributed photovoltaic power generation when it is connected to the grid include: The labor performance information of distributed photovoltaic power generation when connected to the grid is represented by the output distribution difference coefficient, and the energy storage performance information of distributed photovoltaic power generation when connected to the grid is represented by the energy storage change fluctuation coefficient, where: is the output distribution difference coefficient, is the energy storage fluctuation coefficient.

3. The method for optimizing the scheduling of a distribution network and distributed power supply grid connection according to claim 2, characterized in that: The logic for obtaining the output distribution difference coefficient is: Based on the historical data of different distributed photovoltaic power sources, the probability density functions of different distributed photovoltaic power sources in different scenarios are determined, and the probability density functions of distributed photovoltaic power sources in the monitoring interval under different scenarios are calculated. The calculation formula is: ;in, is the probability density function of distributed photovoltaic power in the monitoring interval, g is the actual light intensity at a certain moment, is the maximum light intensity in the monitoring area, is the Gamma function, , is the parameter of Beta distribution, , , is the mean value of sunlight intensity in the monitoring interval, is the variance of sunlight intensity within the monitoring interval; By obtaining the photoelectric conversion efficiency of different distributed photovoltaic power sources and the area of ​​photovoltaic solar panels, the output power of different distributed photovoltaic power supply photovoltaic power generation systems is determined. The calculation formula for the output power of different distributed photovoltaic power supply photovoltaic power generation systems is: ; Among them, P is the output power of the photovoltaic power generation system, A is the area of ​​the photovoltaic solar panel, is the photoelectric conversion efficiency; According to the probability distribution model of solar photovoltaic, it can be concluded that the output model of solar photovoltaic approximately obeys the Beta distribution. The probability distribution model of solar photovoltaic is expressed as: ;in, is the probability distribution model of historical solar photovoltaics, is the maximum output power of solar photovoltaic power generation under the light intensity within the monitoring interval; By real-time acquisition of the probability distribution model of different distributed photovoltaic power sources within the monitoring interval, the probability distribution model of different distributed photovoltaic power sources within the real-time acquisition monitoring interval is marked as: ; The Kullback-Leibler divergence is used to compare the real-time probability distribution model of different distributed photovoltaic power sources with the probability distribution model of historical solar photovoltaic power generation, and the output distribution difference coefficient is calculated. The calculation formula is: .

4. The method for optimizing the scheduling of a distribution network and distributed power supply grid connection according to claim 3, characterized in that: The logic for obtaining the energy storage change fluctuation coefficient is: Based on the energy storage system connected to the distributed photovoltaic power generation, the state of charge change differential equation of the distribution network energy storage system is obtained. The state of charge change rate of the distribution network energy storage system is calculated based on the state of charge change differential equation, and the state of charge change rate is marked as: , where t=1, 2, 3, ..., T, T is a positive integer, and t is the number of different moments in the monitoring interval; The difference equation for the charge state change of the distribution network energy storage system is expressed as: ;in, represents the state of charge at time t, represents the self-discharge rate, 、 Respectively represent the charge and discharge flags of the energy storage system, Represents the actual charging power of the energy storage system, Represents the actual discharge power of the energy storage system, and Represent the charging and discharging efficiency of the energy storage system, Represents the capacity of the energy storage system; Obtain the average value and standard deviation of the state of charge change rate within the monitoring interval, and mark the average value and standard deviation of the state of charge change rate within the monitoring interval as: and ,in, , ; Calculate the energy storage change fluctuation coefficient using the following formula: .

5. The method for optimizing the scheduling of the power distribution network and the distributed power supply grid connection according to claim 4, characterized in that: Evaluate the grid-connected performance of distributed photovoltaic power sources, including: A comprehensive analysis of the labor performance information and energy storage performance information of distributed photovoltaic power generation when connected to the grid is conducted. By performing weighted calculations on the normalized output distribution difference coefficient and the energy storage change fluctuation coefficient, a power supply evaluation model is constructed to generate a power supply evaluation coefficient. The calculation formula for the power supply evaluation coefficient is: ;in, is the power supply evaluation coefficient, 、 are the proportional coefficients of output distribution difference coefficient and energy storage change fluctuation coefficient, respectively. 、 Both are greater than 0.

6. The method for optimizing the scheduling of the power distribution network and the grid connection of distributed power sources according to claim 5, characterized in that: Determine the grid connection risk information and grid connection load matching information of the distribution network, including: The grid-connected risk information of the distribution network is represented by the distributed photovoltaic risk coefficient, and the grid-connected load matching information of the distribution network is represented by the load mismatch coefficient, where: is the risk coefficient of distributed photovoltaics, is the load mismatch coefficient.

7. The method for optimizing the scheduling of a distribution network and distributed power supply grid connection according to claim 6, characterized in that: The logic for obtaining the distributed photovoltaic risk coefficient is as follows: Based on the node locations of the distribution network, distributed photovoltaic power sources connected to the grid are connected at the node locations. That is, the node topology of the distribution network is modeled, and the distributed photovoltaic power sources connected to different nodes are calibrated. By simulating different fault scenarios, the fault set and failure rate caused by the access of distributed photovoltaic power sources at different node locations are obtained. In other words, Monte Carlo simulation is performed on different nodes and the access conditions of distributed photovoltaic power sources to randomly sample different system states and fault conditions, estimate the probability of failure in each scenario, and determine the fault impact score of the distributed photovoltaic power source connected to the grid at the node location; Based on the current operation status of the distribution network, the distributed photovoltaic power nodes that have been connected to the grid in the current distribution network are determined, and the load at the distributed photovoltaic power nodes is obtained. The load at different types of nodes and the fault impact score of different types of nodes connected to distributed photovoltaic power sources are used as input features, and the distributed photovoltaic risk coefficient is used as the output feature to construct a regression model. The calculation formula of the distributed photovoltaic risk coefficient is: ; Among them, 1, 2, 3, ..., J are the numbers of different types of nodes, is the load at different types of nodes, The fault impact scores of different types of nodes connected to distributed photovoltaic power sources are calculated. When the node is not connected to distributed photovoltaic power sources, the fault impact score is marked as a fixed constant. is the weight of different types of nodes, and e is the base.

8. The method for optimizing the scheduling of a distribution network and distributed power supply grid connection according to claim 7, characterized in that: The logic for obtaining the load mismatch coefficient is: Obtain the demand load at the distributed photovoltaic power grid-connected node within a unit time period, and mark the demand load at the distributed photovoltaic power grid-connected node within a unit time period as: ; Wherein, i=1, 2, 3, ..., 1, I is a positive integer, i is the number of different time points in a unit time period, n=1, 2, 3, ..., N, N is a positive integer, n is the number of the distributed photovoltaic power grid-connected node; Obtain the output power of the distributed photovoltaic power source at the distributed photovoltaic power source grid-connected node within a unit time period, and mark the output power of the distributed photovoltaic power source at the distributed photovoltaic power source grid-connected node within a unit time period as: ; Determine the load matching degree of a single distributed photovoltaic power grid-connected node within a unit time period. The calculation formula is: ;in, is the load matching degree of the nth distributed photovoltaic power grid-connected node; Calculate the load mismatch coefficient using the following formula: .

9. The method for optimizing the scheduling of the power distribution network and the distributed power supply grid connection according to claim 8, characterized in that: Quantify the overall grid performance of the distribution network, including: A comprehensive analysis of the grid-connected risk information and grid-connected load matching information of the distribution network is conducted. By weighted calculation of the normalized distributed photovoltaic risk coefficient and load mismatch coefficient, a distribution network optimization evaluation model is constructed to generate the distribution network optimization evaluation coefficient. The calculation formula of the distribution network optimization evaluation coefficient is: ;in, Optimization evaluation coefficient for distribution network, 、 are the proportional coefficients of distributed photovoltaic risk coefficient and load mismatch coefficient, 、 Both are greater than 0.

Images