Self-organization critical state identification method for power grid containing high-proportion wind power generation

Abstract

The invention belongs to the technical field of electric power, and discloses a self-organizing critical state identification method for a power grid containing high-proportion wind power generation. According to the method, the influence of wind power generation on the self-organizing critical state of the power grid is considered, the index weight can be dynamically adjusted according to real-time working conditions such as topological structure change, load fluctuation and wind power output of the power grid through multi-index collaborative analysis, and meanwhile, a comprehensive judgment index capable of more comprehensively reflecting the self-organizing critical state of the power grid is provided; and the reliability of critical state identification is improved. Compared with a traditional neural network method with a fixed weight or needing training and learning and a method depending on historical data, the method has better environmental adaptability and scene generalization ability, and can be suitable for power grid systems of different scales and different structures without depending on the historical data. Compared with an existing probabilistic evaluation method, the deterministic index system provided by the invention can directly output a quantitative evaluation result, a clear operation basis is provided for a scheduling decision, and the risk of misjudgment is effectively reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power, and relates to a method for identifying the self-organized critical state of a power grid with a high proportion of wind power generation. Background Art

[0002] With the rapid development of renewable energy, wind power generation has become an important power source. However, after a high proportion of wind power generation is connected to the power grid, new challenges have been brought to the stability and security of the power grid.

[0003] Due to the intermittent and uncertain characteristics of wind power generation, problems such as power fluctuations and voltage instability may occur during the operation of the power grid, which may cause the power grid to enter the self-organized critical state and further lead to power outages.

[0004] Relevant personnel have conducted various studies on identifying the self-organized critical state of the power grid. These studies have the following defects:

[0005] On the one hand, the existing methods only focus on the probability of the power grid being in the self-organized critical state or the evolution law of the self-organized critical state of the power grid, but lack the identification criteria for whether the currently operating power grid is in the self-organized critical state.

[0006] On the other hand, the influence of wind power generation characteristics on the critical state of the power grid is not fully considered, or only starting from the wind power generation characteristics, without comprehensively considering other factors of the power grid operation. Therefore, it is difficult to achieve accurate critical state assessment and prediction. Summary of the Invention

[0007] The purpose of the present invention is to propose a method for identifying the self-organized critical state of a power grid with a high proportion of wind power generation. By comprehensively considering the characteristics of wind power generation, this method improves the recognition ability of the self-organized critical state of the power grid under the condition of high proportion of wind power generation access, so as to accurately judge whether the power grid is in the self-organized critical state, with high reliability and good adaptability.

[0008] In order to achieve the above purpose, the present invention adopts the following technical solutions: A method for identifying the self-organized critical state of a power grid with a high proportion of wind power generation, comprising the following steps: Step 1. Calculate the output power of the wind farm according to the wind speed value in the area where the wind farm is located in the power grid; Step 2. Define six indicators including the proportion of wind power output, wind power penetration rate, power grid voltage deviation rate, power grid line load rate, node degree, and node betweenness, and give the calculation formulas for the corresponding indicators; Step 3. Real-time obtain the network parameters of the power grid with a high proportion of wind power generation, set the output power of the wind farm in the power grid according to the output power of the wind farm, and calculate the proportion of wind power output index of the current power grid according to Step 2; Perform simulations under the current proportion of wind power output in the power grid, add random load disturbances to the power grid until a power outage accident occurs in the power grid. When a power outage accident occurs, it indicates that the power grid has evolved to the self-organized critical state; Use the network parameters when a power outage accident occurs in the power grid each time to perform power flow calculations, and obtain the parameters required for calculating each index according to the power flow calculation results. Then calculate the values of each index according to the calculation formula of the index; Step 4. According to the values of each index when a power outage accident occurs in the power grid each time, use the entropy weight method to assign weights to the indexes, and perform weighted calculations based on the weighting results and the values of each index to obtain the comprehensive index value of each power outage accident in the power grid; calculate the mean value of the comprehensive index values under multiple power outage accidents, and use this mean value as the power grid self-organized critical state judgment index; Step 5. Use the network parameters at the current operating moment of the power grid to perform power flow calculations, obtain the required parameters from the power flow calculation results to calculate the values of each index, perform weighted calculations on each index according to Step 4 to obtain the comprehensive index value at the current operating moment of the power grid, and compare it with the power grid self-organized critical state judgment index to determine whether the power grid is close to or enters the self-organized critical state.

[0009] In addition, based on the above power grid self-organized critical state identification method, the present invention also proposes a computer device, which includes a memory and one or more processors. Executable code is stored in the memory. When the processor executes the executable code, it is used to implement the steps of the power grid self-organized critical state identification method with a high proportion of wind power generation.

[0010] In addition, based on the above power grid self-organized critical state identification method, the present invention also proposes a computer-readable storage medium, on which a program is stored. When the program is executed by a processor, it is used to implement the steps of the power grid self-organized critical state identification method with a high proportion of wind power generation.

[0011] The present invention has the following advantages: As described above, the present invention proposes a method for identifying the self-organized critical state of a power grid with a high proportion of wind power generation, which takes into account the impact of wind power generation on the self-organized critical state of the power grid. By proposing a comprehensive judgment index for the self-organized critical state, it can accurately identify the self-organized critical state of a power grid with a high proportion of wind power generation. Through multi-index collaborative analysis, the index weights can be dynamically adjusted according to real-time conditions such as the change of power grid topology structure, load fluctuation, and wind power output, improving the reliability of identification. Compared with traditional methods with fixed weights or neural network methods that require training and learning and methods that rely on historical data, the method of the present invention has better adaptability and scenario generalization ability, and can be applied to power grid systems with different wind power outputs and different topology structures without relying on historical data. When the wind power output of the power grid increases or decreases or the topology structure changes, the calculated wind power output ratio, wind power penetration rate, node degree, and node betweenness indexes will increase or decrease accordingly. Especially for power grids lacking historical operation data or with data loss, only the existing real-time operation data needs to be obtained to identify the self-organized critical state through simulation. Compared with existing probabilistic evaluation methods, the deterministic index system proposed by the present invention can directly output a quantitative evaluation result, providing a clear operation basis for dispatching decisions, effectively reducing the risk of misjudgment, and can also provide a reference for the planning of wind farms in the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a flowchart of the method for identifying the self-organized critical state of a power grid with a high proportion of wind power generation in an embodiment of the present invention.

[0013] Figure 2 It is a simulation diagram of the method for identifying the self-organized critical state of a power grid with a high proportion of wind power generation in an embodiment of the present invention.

[0014] Figure 3 It is the IEEE 39-node power grid diagram in an embodiment of the present invention.

[0015] Figure 4 It is a statistical chart of the wind power generation output power with a time scale of 500 days in an embodiment of the present invention.

[0016] Figure 5 It is a statistical chart of the load loss of 500 power outage accidents (wind farms No. 32 and No. 33) in an embodiment of the present invention.

[0017] Figure 6 It is a double logarithmic coordinate diagram of scale-frequency based on load loss statistical data in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The present invention will be further described in detail below with reference to the drawings and specific embodiments: Embodiment 1 Embodiment 1 of the present invention describes a method for identifying the self-organized critical state of a power grid with a high proportion of wind power generation, as follows: Figure 1 As shown, the method for identifying the self-organized critical state of a power grid with a high proportion of wind power generation includes the following steps: Step 1. Calculate the output power of the wind farm according to the wind speed value in the area where the wind farm is located in the power grid.

[0019] The functional relationship between the output power of the wind turbine and the wind speed is: .

[0020] In the formula, is the actual output power of the th wind farm turbine; is the wind speed at the location of the th wind farm; is the cut-in wind speed of the th wind farm turbine; is the cut-out wind speed of the th wind farm turbine; is the rated wind speed of the th wind farm turbine; is the rated output power of the th wind farm turbine.

[0021] The output power of the wind farm is expressed as: .

[0022] In the formula, is the output power of the th wind farm; is the number of wind turbines in the th wind farm.

[0023] Step 2. Define six indicators including the proportion of wind power output, wind power penetration rate, power grid voltage deviation rate, power grid line load rate, node degree, and node betweenness, and give the calculation formulas for the corresponding indicators.

[0024] Among them, the proportion of output and the wind power penetration rate are two wind power generation indicators, the voltage deviation rate and the line load rate are two power grid operation safety indicators, and the node degree and the node betweenness are two power grid topology structure indicators.

[0025] The present invention starts from three aspects of wind power generation, power grid operation safety, and power grid topology structure. Through multi-index collaborative analysis, the index weights can be dynamically adjusted according to real-time conditions such as power grid topology structure changes, load fluctuations, and wind power output, which can more comprehensively reflect the comprehensive characteristics of the self-organized critical state of the power grid and improve the reliability of identification.

[0026] The calculation formula for the proportion p index of wind power output is as follows: 。

[0027] In the formula, represents the sum of the output powers of all wind farms; represents the sum of the output powers of all power plants including wind farms in the power grid.

[0028] The calculation formula for the wind power penetration rate α index is as follows: 。

[0029] In the formula, represents the sum of the output powers of all wind farms per unit time; represents the sum of the power loads of the entire network during the same period.

[0030] Grid voltage deviation rate The calculation formula for the index is as follows: , 。

[0031] In the formula, is the actual voltage of node ; is the rated voltage of node ; is the total number of grid nodes; is the voltage deviation rate of node 。

[0032] Grid line load rate The calculation formula for the index is as follows: , 。

[0033] In the formula, represents the load power borne by the th line; represents the maximum load power that the th line can bear; represents the load rate of the th line; represents the number of lines operating normally in the power grid.

[0034] The calculation formula for the node degree index is as follows: , 。

[0035] In the formula, is the node adjacent to node ; is the node adjacent to node The total number of adjacent nodes; For node and the number of connecting edges between them; is the total number of grid nodes; is the average degree of the power grid.

[0036] The calculation formula of the node betweenness index is as follows: .

[0037] In the formula, is the node and node the number of all shortest paths between them; is the number of shortest paths passing through node ; represents the betweenness of node ; is the total number of grid nodes; is the average betweenness of the power grid.

[0038] Step 3. Obtain the network parameters of the power grid with a high proportion of wind power generation in real time, set the output power of the wind farms in the power grid according to the output power of the wind farms, and calculate the wind power output ratio index of the current power grid according to Step 2. Under the current wind power output ratio of the power grid, add random load disturbances to the power grid until a power outage accident occurs in the power grid. When a power outage accident occurs, it indicates that the power grid has experienced a large number of small disturbances in the long term, and these disturbances accumulate and finally lead the system to evolve to the self-organized critical state.

[0039] Perform power flow calculations using the network parameters when a power outage accident occurs in the power grid each time, obtain the parameters required for calculating each index according to the power flow calculation results, and then calculate the values of each index according to the calculation formula of the index.

[0040] Step 3.1. Set the output power of the current wind farm according to the output power of the wind farm, and calculate the wind power output ratio index of the power grid according to Step 2; set the total number of power outage accidents to , and initialize the current number of power outage accidents .

[0041] To avoid inaccurate index calculations caused by errors and accidental factors, for example, a value above 100 can be taken, but it should not be too large. For example, the maximum value of N can be set to 800. Of course, 800 is only exemplary and does not constitute a limitation to the present invention.

[0042] Step 3.2. Randomly select a node from the power grid to add a load disturbance to increase its load by , and then calculate the power flow of the power grid; where Set to the total grid load 。

[0043] Step 3.3. Determine whether there is any line overload in the power grid according to the power flow calculation results after adding load disturbances to the power grid.

[0044] If there is a line power flow limit, disconnect this line, and the remaining lines are disconnected according to the line hidden fault model with probability, and go to Step 3.4; if there is no line overload, return to Step 3.2. Among them, the line hidden fault model is the relay protection hidden fault probability model for line power flow limit, which is used to judge the operating state of the line.

[0045] Step 3.4. After disconnecting the overloaded line, determine whether there is any island formation or load shedding in the power grid.

[0046] If there is, let the current number of power outage accidents , at this time, the network parameters of the power grid are the network parameters when the power grid enters the self-organized critical state. Use the network parameters of the power grid at this time to perform power flow calculation, and obtain the parameters required for each index in Step 2 except the wind power output ratio according to the power flow calculation results, and then go to Step 3.5.

[0047] Here, each index except the wind power output ratio includes two power grid operation safety and two power grid topology structure indexes. The network parameters of the power grid include the node parameters, line parameters, topology structure parameters and load parameters of the power grid.

[0048] If not, modify the network parameters of the power grid, remove the disconnected lines and failed nodes from the power grid, and return to Step 3.2.

[0049] Step 3.5. Compare and judge whether the current number of power outage accidents reaches the preset total number of power outage accidents ; if so, end this process; if not, reset the power grid parameters to the network parameters when the power grid has not added load disturbances, that is, reconnect the disconnected lines and failed nodes of the power grid, reset the load parameters to the load when the power grid has not added load disturbances, and then return to Step 3.2.

[0050] Step 4. According to the values of each index when each power outage accident occurs in the power grid, use the entropy weight method to assign weights to the indexes, and perform weighted calculation according to the weighting results and the values of each index to obtain the comprehensive index value of each power outage accident in the power grid.

[0051] Calculate the mean value of the comprehensive index values under multiple power outage accidents, and use this mean value as the power grid self-organized critical state judgment index.

[0052] Use the entropy weight method to first calculate the weights of each index in Step 2, and the process is as follows: Step 4.1. Construct an index evaluation matrix: 。

[0053] In the formula, is the value of the th index during the th power outage accident in the power grid, is the number of indicators, ; is the number of power outage accidents in the power grid.

[0054] Normalize the indicators, perform dimensionless processing on heterogeneous indicators, and obtain the standardized indicator 。

[0055] The calculation formula for the standardized indicator is as follows: 。

[0056] In the formula, is the value of the th index during the th power outage accident in the power grid; 、 are the maximum and minimum values of the th index in all power outage accidents in the power grid; represents the normalization result, and 。

[0057] Step 4.2. Calculate the index weights.

[0058] First, define an intermediate variable , and its calculation formula is as follows: 。

[0059] Then, calculate the entropy value of the th index during the th power outage accident in the power grid, the information redundancy of the th index, and the weight value of the th index in turn. The formulas are as follows: , , 。

[0060] Perform weighted calculation on each indicator defined in Step 2 according to the weight value, and obtain the comprehensive indicator of the power grid during the th power outage accident. The calculation formula is as follows: 。

[0061] In the formula: respectively represent the weights of each index described in Step 2 obtained by using the entropy weight method.

[0062] Power grid self-organized critical state judgment index The calculation formula is as follows: .

[0063] In the formula, represents the comprehensive index of the power grid when the th power outage accident occurs; represents the total number of power grid outage accidents.

[0064] Step 5. Use the network parameters at the current operating moment of the power grid to perform power flow calculation, obtain the required parameters from the power flow calculation results to calculate the values of each index, perform weighted calculation on each index according to Step 4 to obtain the comprehensive index value at the current operating moment of the power grid, and compare it with the power grid self-organized critical state judgment index to determine whether the power grid is close to or enters the self-organized critical state.

[0065] Specifically, define the comprehensive index value of the power grid at the current operating moment calculated according to the weights of each index obtained in Step 4 and compare it with the power grid self-organized critical state judgment index . If reaches , it indicates that the power grid is close to the self-organized critical state at this time; if reaches or exceeds

[0066] The IEEE 39-bus power grid is taken as an example below to prove the effectiveness and rationality of the method of the present invention. In this example, the Weibull distribution is used to simulate the wind speed in the area where the wind farm is located. The cumulative probability density function of the Weibull distribution is defined as: .

[0067] In the formula: F(v) represents the Weibull wind frequency cumulative probability density (%); v represents the wind speed (m / s); c represents the Weibull scale parameter (m / s); k represents the Weibull shape parameter.

[0068] Subsequently, the following formula is used to calculate the simulated wind speed in the area where the fan is located: .

[0069] In the formula, is the simulated wind speed in the area where the fan is located; is in the interval A random variable that follows a uniform distribution above.

[0070] First, perform 500 power outage accident simulations on the IEEE 39 - node power grid. The selected wind farms are at nodes 32 and 33. Use the Weibull distribution to generate simulated wind speeds for 500 days.

[0071] According to the recorded wind speed values, use the functional relationship between the output power of the wind turbine and the wind speed in the method of the present invention: .

[0072] Calculate the output power of the wind farm at different wind speeds.

[0073] Figure 4 It is a wind power output model with a time scale of 500 days established using the Weibull distribution. When introducing the wind power generation model into the power grid model, the wind farm is regarded as a PV node with adjustable reactive power output and active power output determined by the wind power generation model. The total installed capacity of the entire IEEE 39 - node power grid is 6297.871 MW. Set the installed capacity of the power plants at nodes 32 and 33 as the installed capacity of the wind farm. According to the high - proportion wind power judgment method described in step 3, after calculation, the proportion of the wind power installed capacity at this time is 20.36%, which has reached the standard of high - proportion wind power generation.

[0074] After setting up the high - proportion wind power generation scenario in the power grid, according to Figure 2 the simulation process shown, under the condition of random output of the wind farm, randomly add 1% - 2% load disturbance to the power grid each time until load loss occurs.

[0075] Load loss can be expressed as: .

[0076] In the formula: is the total number of power grid nodes; is the original load of node ; is the actual load of node after a power outage accident occurs. In this embodiment, a power outage accident is set to occur once a day, and 500 power outage accidents are accumulated in total. The load loss generated by each power outage accident is statistically analyzed, as shown in Figure 5 .

[0077] To simplify the calculation of power grid topology indicators, the power grid topology structure is simplified, that is, it is processed as an unweighted undirected graph, and there is only one connecting edge between two nodes. Calculate and record the comprehensive index value of the power grid at the time of each power outage accident, and perform statistics after the simulation is completed. Table 1 only lists the index values for the first 10 power outage accidents.

[0078] Table 1 Index values during the first ten power outage accidents

[0079] Then, weighted processing is performed using the comprehensive index values during 500 power outage accidents of the power grid.

[0080] The topological indexes obtained in this embodiment are the degree and betweenness of the initial state of the power grid, and their weights are set to 1. The other four indexes are weighted using the entropy weight method. An index evaluation matrix X is formed according to the comprehensive indexes during all power outage accidents: .

[0081] According to the weighting method of the present invention, the weights of the other four indexes are obtained. The weights assigned to the other four indexes are shown in Table 2.

[0082] Table 2 Weights assigned to the indexes

[0083] Take the comprehensive index values of 500 power outage accidents and take the average value = 1.0993.

[0084] After re - sorting the scale of each power outage accident from small to large, set the loss load statistical scale and loss load statistical frequency, and plot the scale - frequency data of the loss load in a double - logarithmic coordinate graph. Use the least - squares method to perform linear regression on each data point. The obtained double - logarithmic coordinate graph of scale - frequency is as Figure 6 shown.

[0085] When the wind farms are at nodes 32 and 33, the regression fitting equation of the scale - frequency of the load loss in 500 power outage accidents is y = - 9.72x + 34.23, and the correlation coefficient is 0.9482 (significant at β = 0.01), that is, the scale - frequency of the load loss in power outage accidents follows a power - law distribution, indicating that the power grid has entered the self - organized critical state at this time.

[0086] Therefore, the self - organized critical state of the power grid can be identified according to the relationship between the comprehensive index and of the power grid at this time.

[0087] In addition, this example also conducts simulations under different proportions of wind power generation. Among them, the proportions of wind power generation are 3.97%, 10.32%, 20.356%, 24.326%, 28.42%, 40.2%, and 48.27% respectively. The power grids with different proportions of wind power generation are simulated 3 times each, and 500 power outage accidents are generated each time.

[0088] Comprehensive index value when a power outage occurs in the statistical power grid mean value As shown in Table 3

[0089] As can be seen from Table 3, the mean value of the comprehensive index when the power grid enters the self-organized critical state during each simulation increases with the increase of the proportion of wind power generation, indicating that the method of the present invention has good adaptability to the change of the proportion of wind power generation

[0090] Table 3 Comparison of simulation results with different wind power generation ratios

[0091] Figure 6 The regression fitting of the scale-frequency of load loss of the power grid at different wind power generation ratios is given in

[0092] Table 4 gives the regression equations and correlation coefficients fitted at each ratio during the first simulation

[0093] Table 5 is the critical value table of the correlation coefficient in the embodiment of the present invention, as follows

[0094] Table 5 Critical value table of correlation coefficient

[0095] According to the correlation coefficient test table in Table 5, all the fitted regression equations are significant at and all follow the power-law distribution. That is, under these conditions of the proportion of wind power generation, the power grid enters the self-organized critical state after being processed in step 3. The results in Table 3 can prove that the method of the present invention has good adaptability to the change of the proportion of wind power generation in the power grid and can adjust accordingly. Therefore, only by judging whether the comprehensive index of the power grid during current operation reaches or exceeds the self-organized critical state judgment index calculated according to the network parameters of the power grid at this time , it can be judged whether it enters the self-organized critical state. In practical applications, it can be set to send a prompt to the operator when reaches , leaving a certain margin

[0096] The novel power grid self-organized critical state identification method proposed in this embodiment can comprehensively consider the characteristics of wind power generation, improve the ability to identify the self-organized critical state of the power grid under the condition of high-proportion wind power generation access, so as to accurately judge whether the power grid is in the self-organized critical state. This will help monitor and warn the power grid state, ensure the stable operation of the power grid, reduce the risks of large-scale power outages and economic losses caused by emergencies, and provide a basis for optimizing decisions.

[0097] Embodiment 2 This Embodiment 2 describes a computer device, which includes a memory and one or more processors. Executable code is stored in the memory. When the processor executes the executable code, it is used to implement the steps of the power grid self-organized critical state identification method with high-proportion wind power generation in the above-mentioned Embodiment 1.

[0098] In this embodiment, the computer device is any device or apparatus with data processing ability, which will not be elaborated here.

[0099] Embodiment 3 This Embodiment 3 describes a computer-readable storage medium, on which a program is stored. When the program is executed by the processor, it is used to implement the steps of the power grid self-organized critical state identification method with high-proportion wind power generation in the above-mentioned Embodiment 1.

[0100] The computer-readable storage medium can be an internal storage unit of any device or apparatus with data processing ability, such as a hard disk or memory, or an external storage device of any device with data processing ability, such as a plug-in hard disk, a Smart Media Card (SMC), an SD card, a Flash Card, etc. equipped on the device.

[0101] Of course, the above description is only the preferred embodiments of the present invention. The present invention is not limited to listing the above embodiments. It should be noted that all equivalent substitutions and obvious deformation forms made by any person skilled in the art under the teaching of this specification fall within the substantial scope of this specification and should be protected by the present invention.

Claims

1. A method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation, characterized in that: The steps include: Step 1. Calculate the output power of the wind farm according to the wind speed value in the area where the wind farm is located in the power grid; Step 2. Define six indicators: wind power output ratio, wind power penetration rate, grid voltage deviation rate, grid line load rate, node degree and node betweenness, and give the calculation formulas for the corresponding indicators; Step 3. Obtain network parameters of the power grid with a high proportion of wind power generation in real time, set the output power of the wind farm in the power grid according to the output power of the wind farm, and calculate the current wind power output ratio index of the power grid according to step 2; The simulation is carried out under the current wind power output ratio of the power grid, and random load disturbances are added to the power grid until a power outage occurs. When a power outage occurs, it indicates that the power grid has evolved to a self-organized critical state. The network parameters of each power outage accident are used to calculate the power flow, and the parameters required for calculating each indicator are obtained according to the power flow calculation results, and then the values ​​of each indicator are calculated according to the calculation formula of the indicator; Step 4. According to the index values ​​of each power outage accident in the power grid, the entropy weight method is used to assign weights to the indexes, and the weighted calculation is performed based on the weighted results and the index values ​​to obtain the comprehensive index value of each power outage accident in the power grid; Calculate the average of the comprehensive index values ​​under multiple power outage accidents, and use the average as the judgment index of the self-organized critical state of the power grid; Step 5. Use the network parameters of the current operation time of the power grid to perform power flow calculation, obtain the required parameters from the power flow calculation results to calculate the values ​​of each index, perform weighted calculation on each index according to step 4 to obtain the comprehensive index value of the current operation time of the power grid, and compare it with the power grid self-organized critical state judgment index to determine whether the power grid is approaching or entering the self-organized critical state.

2. The method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation according to claim 1, characterized in that: The step 3 is specifically as follows: Step 3.

1. Set the current wind farm output power according to the wind farm output power, calculate the wind power output ratio of the power grid according to step 2; set the total number of power outage accidents to , initialize the current power outage number ; Step 3.

2. Randomly select a node from the power grid and add a load disturbance to increase its load , and then calculate the power grid flow; where, Set to the total grid load ; Step 3.

3. Determine whether there is a line overload in the power grid based on the power flow calculation result after adding load disturbance to the power grid; If the power flow of a line exceeds the limit, disconnect the line, and the remaining lines are disconnected according to the probability of the line invisible fault model, and go to step 3.4; if no line is overloaded, return to step 3.2; Step 3.

4. After disconnecting the overloaded line, determine whether the grid has island formation or load shedding; If yes, then set the current power outage number , the network parameters of the power grid at this time are the network parameters when the power grid enters the self-organized critical state, and the network parameters of the power grid at this time are used to perform power flow calculation, and the parameters required for the remaining indicators in step 2 except for the proportion of wind power output are obtained according to the power flow calculation results, and then step 3.5 is entered; If not, modify the network parameters of the power grid, remove the disconnected lines and failed nodes from the power grid, and return to step 3.2; Step 3.

5. Compare and determine the current power outage frequency Whether the preset total number of power outage accidents has been reached ; If yes, end this process; if not, reset the grid parameters to the network parameters when no load disturbance is added to the grid, that is, put the disconnected lines and failed nodes of the grid back into operation, reset the load parameters to the load when no load disturbance is added to the grid, and then return to step 3.

2.

3. The method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation according to claim 1, characterized in that: In step 4, the entropy weight method is used to first calculate the weight of each indicator in step 2, and the process is as follows: Step 4.

1. Construct indicator evaluation matrix: ； In the formula, It is The indicator occurs in the power grid The value of the power outage accident, is the number of indicators, ; is the number of power outages in the power grid; Normalize the indicators and remove the dimensions of heterogeneous indicators to obtain standardized indicators. ; Step 4.

2. Calculate the indicator weights; First define the intermediate variables , and its calculation formula is as follows: ； Then calculate the The indicator occurs in the power grid Entropy value during power outage , No. The information redundancy of an indicator and The weight of the indicator , the formulas are as follows: ， ， 。 4. The method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation according to claim 3, characterized in that: In step 4.1, standardization index The calculation formula is as follows: ； In the formula, The power grid is The first power outage The value of an indicator; , It is The maximum and minimum values ​​of an indicator in all power outages in the power grid; represents the normalized result, and .

5. The method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation according to claim 3, characterized in that: In step 4, the calculation process of the comprehensive index value of each power outage accident in the power grid is as follows: Perform weighted calculation on each indicator defined in step 2 according to the weight value, and calculate the power grid Comprehensive indicators of power outage accidents , the calculation formula is as follows: ; Power grid critical state judgment index The calculation formula is as follows: ； In the formula, Indicates that the power grid is in the Comprehensive indicators during power outages; Indicates the total number of power outages in the power grid.

6. The method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation according to claim 1, characterized in that: The step 5 is specifically as follows: The network parameters of the current operation time of the power grid are used to calculate the power flow, and the index values ​​in step 2 of the required parameter calculation are obtained from the power flow calculation results. According to step 4, the weighted calculation of each index is performed to obtain the comprehensive index value of the current operation time of the power grid. and compare it with the grid self-organized critical state judgment index For comparison; if achieve , it indicates that the power grid is close to the self-organized critical state; if Meet or exceed , it indicates that the power grid has entered the self-organized critical state.

7. The method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation according to claim 1, characterized in that: The step 1 is specifically as follows: The functional relationship between wind turbine output power and wind speed is: ； In the formula, For the The actual output power of wind turbines in wind farm No. For the The wind speed at the location of the wind farm; For the Cut-in wind speed of wind turbines in wind farm No. For the Cut-out wind speed of wind turbines in wind farm No. For the Rated wind speed of wind turbines in wind farm No. For the Rated output power of wind turbines in wind farm No. The output power of a wind farm is expressed as: ; In the formula, For the Output power of wind farm No.; For the The number of wind turbines in wind farm No.

8. The method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation according to claim 1, characterized in that: In step 2, the calculation formulas of the defined indicators are as follows: The calculation formula of wind power output ratio p is as follows: ； In the formula, Represents the sum of the output power of all wind farms; It represents the sum of the output power of all power plants in the grid, including wind farms; The calculation formula of wind power penetration rate α index is as follows: ； In the formula, It represents the sum of the output power of all wind farms per unit time; It represents the total power load of the entire network during the same period; Grid voltage deviation rate The indicator is calculated using the following formula: ， ； In the formula, For Node The actual voltage of For Node Rated voltage; is the total number of grid nodes; For Node The voltage deviation rate; Grid line load rate The indicator is calculated using the following formula: ， ； In the formula, Indicates The load power borne by the line; Indicates The maximum load power that the line can bear; Indicates The load factor of the line; Indicates the number of normal operating lines of the power grid; The calculation formula of node degree index is as follows: ， ； In the formula, For the node Adjacent nodes; For the node The total number of adjacent nodes; For Node and The number of connected edges between ; is the total number of grid nodes; is the average degree of the power grid; The calculation formula of the node betweenness index is as follows: ； In the formula, For Node and nodes The number of all shortest paths between ; Passing Node The number of shortest paths; Representation Node Betweenness of; is the total number of grid nodes; is the average betweenness of the power grid.

9. A computer device comprising a memory and one or more processors; an executable code is stored in the memory; characterized in that: When the processor executes the executable code, it is used to implement the steps of the method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation according to any one of claims 1 to 8.

10. A computer-readable storage medium having a program stored thereon; characterized in that: When the program is executed by a processor, it is used to implement the steps of the method for identifying the self-organized critical state of a power grid containing a high proportion of wind power generation as described in any one of claims 1 to 8.

Images