A reliable capacity optimization method for wind-storage system based on chaotic particle swarm optimization

Abstract

The present invention provides a method for optimizing the credible capacity of a wind-storage system based on a chaotic particle swarm, including: constructing a new energy output characteristic calculation model and an index system from four dimensions: daily average output characteristic, monthly average output characteristic, credible capacity and effective capacity; configuring energy storage for wind power generation can make its output smooth, and constructing a credible capacity optimization model for the wind-storage system with the goal of minimizing the sum of squares of deviations between the total power generation power of the wind-storage system and the planned output power; using an improved chaotic particle swarm optimization algorithm to solve the model to obtain energy storage charging and discharging strategies under different scenarios, until the optimal credible capacity of the wind-storage system that meets the set accuracy requirements is obtained. The credible capacity optimization model for the wind-storage system of the present invention fully considers the impact of the output characteristics of wind farms in various regions on the credible capacity optimization, making the optimization results more comprehensive, more consistent with reality, and more instructive.

Description

Technical Field

[0001] The present invention relates to the technical field of power system planning, and in particular to a method for optimizing the credible capacity of a wind-storage system based on chaotic particle swarms. Background Art

[0002] With the rapid development of the economy and society, problems such as energy shortage and environmental pollution in my country have become increasingly prominent. Developing and utilizing clean energy and improving the energy structure have become a general consensus. Hydropower, wind power, and photovoltaic power generation, as the main forms of clean energy development and utilization, have the potential to compete with traditional power sources and have developed rapidly.

[0003] Vigorously promoting the development of green energy sources such as wind power and photovoltaic power generation is an inevitable choice for achieving carbon peak in the power industry as soon as possible. However, large-scale wind power and photovoltaic power generation are characterized by volatility and intermittentness, and their low installed capacity replacement efficiency is a significant issue. Energy storage has become a key supporting technology for ensuring the safe and reliable operation of new power systems. Therefore, research on optimizing the reliable capacity of new energy power generation systems with energy storage has important guiding significance for subsequent work such as grid design and new energy planning.

[0004] The technology for evaluating and calculating the reliable capacity of wind-storage systems is one of the key exploration directions of researchers in this field. Existing technologies, such as the Chinese invention patent "Reliable Capacity Evaluation Method and Device for Wind-Storage Power Generation System", publication number CN111753437A, include: obtaining the initial reliability index of the original power generation system at the initial load level, the original power generation system does not include wind turbines; determining the reference load level based on the initial reliability index, the first reliability index of the wind-storage power generation system at the initial load level and the second reliability index at the boundary load level, the wind-storage power generation system is obtained by adding wind turbines to the original power generation system, and the boundary load level is obtained based on the initial load level and the rated capacity of the wind turbines; calculating the third reliability index of the wind-storage power generation system at the reference load level; and determining the reliable capacity of the wind-storage power generation system based on at least the initial reliability index and the third reliability index. Another example is “A method for optimizing the daily operation strategy of a wind-storage hybrid power station based on a genetic algorithm”, publication number N113361715A, which includes the following steps: S1: Given relevant parameters, including: real-time electricity price, output deviation penalty electricity price, maximum allowable probability of output deviation of the wind-storage hybrid power station, battery energy storage system capacity, initial state of charge and maximum charge and discharge power, wind power installed capacity, wind power forecast value, and number of scheduling periods; S2: Establishing a daily operation strategy optimization model for the wind-storage hybrid power station, where the optimization variables are the planned charge and discharge sequence and power of the battery energy storage system within the scheduling day, and the optimization goal is to maximize the daily operation profit of the wind-storage hybrid power station; S3: Based on the Monte Carlo simulation of the daily operating conditions of the wind-storage hybrid power station, a genetic algorithm is used to solve the daily operation strategy optimization model of the wind-storage hybrid power station, and the optimal planned charge and discharge sequence and power of the battery energy storage system in the wind-storage hybrid power station are given; however, the model basis adopted by the above-mentioned existing technologies does not have a high degree of randomness, and some characteristic index data of the wind farm are not taken into consideration, and there are still certain discrepancies in reality. Summary of the Invention

[0005] In view of the above analysis, the present invention aims to provide a method for optimizing the credible capacity of a wind-storage system based on a chaotic particle swarm optimization algorithm to solve the above problems. The chaotic particle swarm optimization algorithm has good timeliness and can incorporate more characteristic index factors, so the results obtained are closer to the actual situation.

[0006] The purpose of the present invention is mainly achieved through the following technical solutions:

[0007] A method for optimizing the credible capacity of a wind-storage system based on chaotic particle swarm optimization comprises the following steps:

[0008] Step 1: Collect historical actual active power data of wind farms in operation in the region;

[0009] Step 2: Construct a new energy output characteristic calculation model and indicator system based on the four dimensions of daily average output characteristics, monthly average output characteristics, reliable capacity, and effective capacity;

[0010] Step 3: A wind-storage system credible capacity optimization model is constructed with the goal of minimizing the sum of squares of the deviations between the total power generation power of the wind-storage system and the planned output power;

[0011] Step 4: Use the improved chaotic particle swarm optimization algorithm to solve the model and obtain the energy storage charging and discharging strategies under different scenarios;

[0012] Step 5: Taking actual cases as an example, the sensitivity curve of wind power credible capacity under different energy storage capacity configurations is studied;

[0013] in:

[0014] In step 1, the actual active power data of the wind farm over the past years is mainly derived from the regional power grid dispatching and operation management system according to the conditions such as the wind farm name, grid voltage level, and location;

[0015] The new energy output characteristic calculation model and index system constructed in step 2 are as follows:

[0016] (1) Daily average output characteristic index , its calculation expression is as follows:

[0017]

[0018] in,

[0019]

[0020] Where: is the daily average output characteristic sequence of the wind farm; For wind farms within one day h The ratio of the average active power to the installed capacity at a certain time; H Indicates the time of day at 23:00; N Indicates the whole year h The total number of active power data at a certain time; is the installed capacity of the wind farm; for h Point in time j Active power value;

[0021] (2) Monthly average output characteristic index , its calculation expression is as follows:

[0022]

[0023] in,

[0024]

[0025] Where: is the monthly average output characteristic sequence of the wind farm; For wind farms within one year m Ratio of monthly average active power to installed capacity; M Indicates December of the year; L Indicates the whole year m The total number of monthly active power data; is the installed capacity of the wind farm; for m Month j Active power value;

[0026] (3) Trusted Capacity Index , its calculation expression is as follows:

[0027]

[0028] in,

[0029]

[0030] Where: is the credible capacity sequence of the wind farm; For wind farms in guaranteed rate r The corresponding effective capacity is as follows; Indicates the guaranteed rate of the sequence of all active power values ​​in the evening peak period of the load within a year, which is arranged in descending order. r The corresponding active power value is as follows; R Indicates the maximum guarantee rate is 100%; is the installed capacity of the wind farm;

[0031] (4) Effective capacity index , its calculation expression is as follows:

[0032]

[0033] in,

[0034]

[0035] Where: is the effective capacity sequence of the wind farm; For wind farms in guaranteed rate r The corresponding effective capacity is: Indicates the guaranteed rate of the sequence of all active power values ​​in the low load period within a year arranged in ascending order. r The corresponding active power value is as follows; R Indicates the maximum guarantee rate is 100%; is the installed capacity of the wind farm;

[0036] In step 3, the specific steps of constructing the wind-storage system reliable capacity optimization model with the goal of minimizing the sum of squares of the deviations between the total power generation power of the wind-storage system and the planned output power are as follows:

[0037] (1) Wind power is an intermittent renewable energy source. When combined with energy storage, it can smoothen its output and improve the “friendliness” of the renewable energy output. The objective function of the energy storage strategy is:

[0038]

[0039] in,

[0040]

[0041] Where, is the total power on the power supply side; Indicates the energy storage discharge power; Indicates the energy storage charging power; is wind power; is the power of conventional units; is the load power; The credible capacity value after the specified energy storage strategy acts on wind power;

[0042] (2) In the process of smoothing the output of the wind farm, the energy storage system mainly considers the charging and discharging constraints and the power constraints of the energy storage system. In order to protect the battery, the minimum power stored in the battery shall not be less than 5% of the battery capacity, and the maximum power shall not be higher than 95% of the battery capacity. In addition, considering the constraints of the system on the wind power acceptance capacity, the constraints of the above optimization model are as follows:

[0043]

[0044] Where, For the Power for the time period; is the capacity of the battery; is the charge-discharge conversion efficiency;

[0045] (3) The constraints of energy storage are processed using the exterior point penalty function method. The objective function of the battery is expressed as follows:

[0046]

[0047] The corresponding penalty function constructed by the exterior point method is:

[0048]

[0049] First, the credible capacity value calculated by wind power alone is used as , bringing the specified objective function into the system calculation for optimization, solving for the energy storage charging and discharging strategy at each point. This strategy is applied to the wind power system to obtain a new power curve on the power source side. The credible capacity of the wind power curve after energy storage optimization is calculated and added to the objective function in the next step. This iterative calculation continues until a credible capacity value for the wind power energy storage system that meets the accuracy requirements is found, and an energy storage strategy that maximizes the credible capacity is obtained.

[0050] In step 4, the improved chaotic particle swarm optimization algorithm is used to solve the model to obtain the specific steps of energy storage charging and discharging strategies under different scenarios:

[0051] (1) Assume that the particle swarm Particles in D The position and velocity in the dimensional solution space can be expressed as and In the optimization process, particles keep tracking their own individual optimal position. and the global optimal position of the population To update its own state, through the personalized learning of particles and the collaborative interaction between each other, the entire group is pushed to continuously approach the optimal solution to the problem. Its velocity and position will be updated according to the following formula:

[0052]

[0053] Where, =1, 2, 3, …, M ; d =1, 2, 3, …, D ; k For evolutionary algebra; w is the inertia weight factor; c 1 and c 2 is the acceleration factor, usually set to 2.0; r 1 and r 2 is a random number between [0,1]. Furthermore, to improve the standard PSO algorithm's tendency to fall into local optimal values ​​and thus experience premature convergence, the idea of ​​chaotic optimization is introduced into the standard PSO algorithm. A chaotic particle swarm optimization algorithm (CPSO) is proposed. This algorithm utilizes the ergodic and random characteristics of chaotic motion to perform chaotic mutation on the optimal position of the swarm, thereby increasing population diversity and improving the algorithm's global convergence performance.

[0054] (2) In the actual optimization process, the inertia weight wIt is mainly used to balance the global search and local search capabilities in the CPSO algorithm. This paper will adopt a strategy of adaptively adjusting the inertia weight based on the group's premature convergence degree and individual fitness value. The calculation formula is:

[0055]

[0056] Where, and yes w The upper and lower limits of the value; For particles Adaptability; is the mean of all particle fitness; The fitness of the particle swarm is better than The average fitness value of the particles; Used to evaluate the degree of premature convergence in the particle evolution process; generally k 1=1.5, k 2=1.8. In order to further improve the efficiency of the algorithm, the convergence factor is introduced The particle position equation is modified as follows:

[0057]

[0058] in,

[0059]

[0060] (3) Compared with the Logistic map, the Tent map has better traversal uniformity and higher optimization efficiency. Therefore, the Tent chaotic mutation process is introduced to improve the problem that the algorithm is prone to falling into local optimality. The dynamic equation of the Tent chaotic map is:

[0061]

[0062] (4) In order to avoid the problem that the iterative sequence falls into the small period point (0.2, 0.4, 0.6, 0.8) or the fixed point (0, 0.25, 0.5, 0.75), the perturbation equation is introduced to correct Equation (10) as follows:

[0063]

[0064] (5) When the global optimal fitness value of the particle swarm stops updating for several consecutive times, the optimal particle position of the swarm is Perform chaotic mutation processing:

[0065]

[0066] Where, It is called the shrinkage factor, which determines the variation space of the particle position.

[0067] In step 5, the sensitivity change curve of wind power credible capacity under different energy storage capacity configurations is studied using actual cases as an example. Specifically, the basic situation of the power system of the example, the analysis of wind power output characteristics and the comparison with the change trend of the regional load curve are included. By configuring different proportions of energy storage capacity, the ICPSO optimization algorithm is used to solve the model one by one to obtain the corresponding wind power credible capacity under each energy storage ratio.

[0068] A calculation model and index system for the output characteristics of new energy are constructed from four dimensions: daily average output characteristics, monthly average output characteristics, credible capacity and effective capacity; configuring wind power generation with energy storage can make its output smooth, and a credible capacity optimization model for the wind-storage system is constructed with the goal of minimizing the sum of squares of the deviations between the total power generation power of the wind-storage system and the planned output power.

[0069] An improved chaotic particle swarm optimization algorithm is used to solve the model to obtain the energy storage charging and discharging strategies under different scenarios. This strategy is then applied to the wind-storage combined power generation system to obtain a new active power curve, and the credible capacity of the wind-storage system combined power generation curve is calculated. This iterative calculation is repeated until the optimal credible capacity of the wind-storage system that meets the set accuracy requirements is obtained.

[0070] The wind-storage system credible capacity optimization model of the present invention fully considers the impact of wind farm output characteristics in various regions on credible capacity optimization, making the optimization results more comprehensive, more consistent with reality, and more guiding. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a schematic diagram of the process of the present invention;

[0072] Figure 2 Analysis results of wind power output characteristics in Yunnan Province Figure 1 , where (a) is the daily average output characteristic curve of wind power in the abundant season;

[0073] Figure 3 Analysis results of wind power output characteristics in Yunnan Province Figure 2 , where (b) is the daily average output characteristic curve of wind power in the dry season;

[0074] Figure 4 Analysis results of wind power output characteristics in Yunnan Province Figure 3 , where (a) is the monthly average output characteristic curve of wind power;

[0075] Figure 5 Analysis results of wind power output characteristics in Yunnan Province Figure 4 , where (a) is the credible capacity characteristic curve of wind power in the abundant season;

[0076] Figure 6Analysis results of wind power output characteristics in Yunnan Province Figure 5 , where (b) is the wind power dry season credible capacity characteristic curve;

[0077] Figure 7 Analysis results of wind power output characteristics in Yunnan Province Figure 6 , where (a) is the effective capacity characteristic curve of wind power in the abundant season;

[0078] Figure 8 Analysis results of wind power output characteristics in Yunnan Province Figure 7 , where (b) is the effective capacity characteristic curve of wind power in the abundant season;

[0079] Figure 9 This is a comparison chart of wind power output and load curves in a certain area of ​​Yunnan;

[0080] Figure 10 To ensure the output characteristics of wind power in a certain area of ​​Yunnan during the wet and dry seasons;

[0081] Figure 11 This is a curve showing the sensitivity change of wind power credible capacity under different energy storage capacity configurations in a certain area of ​​Yunnan. DETAILED DESCRIPTION

[0082] In order to make the technical problems and technical solutions solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0083] An embodiment of the present invention will be described in detail below with reference to the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used to illustrate the principles of the present invention together with the embodiments of the present invention.

[0084] like Figure 1 As shown, the present invention discloses a method for optimizing the credible capacity of a wind-storage system based on chaotic particle swarm optimization, comprising the following steps:

[0085] Step 1: Collect historical actual active power data of wind farms in operation in the region;

[0086] Step 2: Construct a new energy output characteristic calculation model and indicator system based on the four dimensions of daily average output characteristics, monthly average output characteristics, reliable capacity, and effective capacity;

[0087] Step 3: A wind-storage system credible capacity optimization model is constructed with the goal of minimizing the sum of squares of the deviations between the total power generation power of the wind-storage system and the planned output power;

[0088] Step 4: Use the improved chaotic particle swarm optimization algorithm to solve the model and obtain the energy storage charging and discharging strategies under different scenarios;

[0089] Step 5: Taking actual cases as an example, the sensitivity curve of wind power credible capacity under different energy storage capacity configurations is studied;

[0090] in:

[0091] In step 1, the actual active power data of the wind farm over the past years is mainly derived from the regional power grid dispatching and operation management system according to conditions such as the wind farm name, grid-connected voltage level, and location.

[0092] This embodiment collects the actual active power generation data of all wind farms in Yunnan Province from 2016 to 2019. The data format is one active power sampling data every 15 minutes for each wind farm. There are a total of 365×24×4=35040 data values ​​throughout the year.

[0093] The new energy output characteristic calculation model and index system constructed in step 2 are as follows:

[0094] (1) Daily average output characteristic index , its calculation expression is as follows:

[0095]

[0096] in,

[0097]

[0098] Where: is the daily average output characteristic sequence of the wind farm; For wind farms within one day h The ratio of the average active power to the installed capacity at a certain time; H Indicates the time of day at 23:00; N Indicates the whole year h The total number of active power data at a certain time; is the installed capacity of the wind farm; for h Point in time j Active power value;

[0099] Based on the above historical power generation data, the analysis and calculation results of the provincial wind power daily average output characteristic curve after considering the wind power cluster effect are as follows: Figure 2 and Figure 3As shown in the figure, the trend of the change in the daily average output characteristics of wind farms in the province during the wet and dry seasons from 2016 to 2019 is basically similar. The wind power output is the lowest between 10 and 12 in the morning, and then gradually increases, reaching the maximum at around 21 o'clock in the evening. The province's average daily wind power output during the dry season is relatively large, and with the increase in installed capacity, it has increased year by year, at around 30-50%. The wider the distribution of wind power and the larger the installed capacity, the greater the average output; the output during the wet season is relatively small, with an average output below 30%. Overall, the province's wind power generates more power in the afternoon and evening within a day, which matches Yunnan's daily load curve, as shown in the figure below. Figure 2 and Figure 3 shown.

[0100] (2) Monthly average output characteristic index , its calculation expression is as follows:

[0101]

[0102] in,

[0103]

[0104] Where: is the monthly average output characteristic sequence of the wind farm; For wind farms within one year m Ratio of monthly average active power to installed capacity; M Indicates December of the year; L Indicates the whole year m The total number of monthly active power data; is the installed capacity of the wind farm; for m Month j Active power value;

[0105] Based on the above historical power generation data, the analysis and calculation results of the provincial wind power daily average output characteristic curve after considering the wind power cluster effect are as follows: Figure 4 As shown in the data, the trend of the average monthly wind power output characteristics in the province remained basically the same from 2016 to 2019, showing the overall characteristics of low output in the wet season and high output in the dry season. The average monthly wind power output from May to October was between 11% and 33%; the average monthly output from November, December, and January to April was between 31% and 46%. The annual generating hours were about 2,600 hours, and the wet-dry ratio was about 33:67. Figure 4 shown.

[0106] (3) Trusted Capacity Index , its calculation expression is as follows:

[0107]

[0108] in,

[0109]

[0110] Where: is the credible capacity sequence of the wind farm; For wind farms in guaranteed rate r The corresponding effective capacity is: Indicates the guaranteed rate of the sequence of all active power values ​​in the evening peak period of the load within a year, which is arranged in descending order. r The corresponding active power value is as follows; R Indicates the maximum guarantee rate is 100%; is the installed capacity of the wind farm;

[0111] Based on the above historical power generation data, the results of the analysis and calculation of the guaranteed output characteristic curve of wind power in the province during the wet and dry seasons after considering the wind power cluster effect are as follows: Figure 5 and Figure 6 As shown in Figure 2, from the perspective of the minimum output that wind power can provide when the system load is large, when the guaranteed rate is 95%, the guaranteed output in the wet season is less than 10%, while the guaranteed output in the dry season can reach about 20%. Figure 5 and Figure 6 shown.

[0112] (4) Effective capacity index , its calculation expression is as follows:

[0113]

[0114] in,

[0115]

[0116] Where: is the effective capacity sequence of the wind farm; For wind farms in guaranteed rate r The corresponding effective capacity is: Indicates the guaranteed rate of the sequence of all active power values ​​in the low load period within a year arranged in ascending order. r The corresponding active power value is as follows; R Indicates the maximum guarantee rate is 100%; is the installed capacity of the wind farm; Figure 7 and Figure 8 shown.

[0117] In step 3, the specific steps of constructing the wind-storage system reliable capacity optimization model with the goal of minimizing the sum of squares of the deviations between the total power generation power of the wind-storage system and the planned output power are as follows:

[0118] (1) Wind power is an intermittent renewable energy source. When combined with energy storage, it can smoothen its output and improve the “friendliness” of the renewable energy output. The objective function of the energy storage strategy is:

[0119]

[0120] in,

[0121]

[0122] Where, is the total power on the power supply side; Indicates the energy storage discharge power; Indicates the energy storage charging power; is wind power; is the power of conventional units; is the load power; The credible capacity value after the specified energy storage strategy acts on wind power;

[0123] (2) In the process of smoothing the output of the wind farm, the energy storage system mainly considers the charging and discharging constraints and the power constraints of the energy storage system. In order to protect the battery, the minimum power stored in the battery shall not be less than 5% of the battery capacity, and the maximum power shall not be higher than 95% of the battery capacity. In addition, considering the constraints of the system on the wind power acceptance capacity, the constraints of the above optimization model are as follows:

[0124]

[0125] Where, For the Power for the time period; is the capacity of the battery; is the charge-discharge conversion efficiency;

[0126] (3) The constraints of energy storage are processed using the exterior point penalty function method. The objective function of the battery is expressed as follows:

[0127]

[0128] The corresponding penalty function constructed by the exterior point method is:

[0129]

[0130] First, the credible capacity value calculated by wind power alone is used as , bringing the specified objective function into the system calculation for optimization, solving for the energy storage charging and discharging strategy at each point. This strategy is applied to the wind power system to obtain a new power curve on the power source side. The credible capacity of the wind power curve after energy storage optimization is calculated and added to the objective function in the next step. This iterative calculation continues until a credible capacity value for the wind power energy storage system that meets the accuracy requirements is found, and an energy storage strategy that maximizes the credible capacity is obtained.

[0131] In step 4, the improved chaotic particle swarm optimization algorithm is used to solve the model to obtain the specific steps of energy storage charging and discharging strategies under different scenarios:

[0132] (1) Assume that the particle swarm Particles in D The position and velocity in the dimensional solution space can be expressed as and In the optimization process, particles keep tracking their own individual optimal position. and the global optimal position of the population To update its own state, through the personalized learning of particles and the collaborative interaction between each other, the entire group is pushed to continuously approach the optimal solution to the problem. Its velocity and position will be updated according to the following formula:

[0133]

[0134] Where, i =1, 2, 3, …, M ; d =1, 2, 3, …, D ; k For evolutionary algebra; w is the inertia weight factor; c 1 and c 2 is the acceleration factor, usually set to 2.0; r 1 and r 2 is a random number between [0,1]. Furthermore, to improve the standard PSO algorithm's tendency to fall into local optimal values ​​and thus experience premature convergence, the idea of ​​chaotic optimization is introduced into the standard PSO algorithm. A chaotic particle swarm optimization algorithm (CPSO) is proposed. This algorithm utilizes the ergodic and random characteristics of chaotic motion to perform chaotic mutation on the optimal position of the swarm, thereby increasing population diversity and improving the algorithm's global convergence performance.

[0135] (2) In the actual optimization process, the inertia weight w It is mainly used to balance the global search and local search capabilities in the CPSO algorithm. This paper will adopt a strategy of adaptively adjusting the inertia weight based on the group's premature convergence degree and individual fitness value. The calculation formula is:

[0136]

[0137] Where, and yes w The upper and lower limits of the value; For particles Adaptability; is the mean of all particle fitness; The fitness of the particle swarm is better than The average fitness value of the particles; Used to evaluate the degree of premature convergence in the particle evolution process; generally k 1=1.5, k 2=1.8. In order to further improve the efficiency of the algorithm, the convergence factor is introduced δ The particle position equation is modified as follows:

[0138]

[0139] in,

[0140]

[0141] (3) Compared with the Logistic map, the Tent map has better traversal uniformity and higher optimization efficiency. Therefore, the Tent chaotic mutation process is introduced to improve the problem that the algorithm is prone to falling into local optimality. The dynamic equation of the Tent chaotic map is:

[0142]

[0143] (4) In order to avoid the problem that the iterative sequence falls into the small period point (0.2, 0.4, 0.6, 0.8) or the fixed point (0, 0.25, 0.5, 0.75), the perturbation equation is introduced to correct Equation (10) as follows:

[0144]

[0145] (5) When the global optimal fitness value of the particle swarm stops updating for several consecutive times, the optimal particle position of the swarm is Perform chaotic mutation processing:

[0146]

[0147] Where, It is called the shrinkage factor, which determines the variation space of the particle position.

[0148] In step 5, the sensitivity change curve of wind power credible capacity under different energy storage capacity configurations is studied using actual cases as an example. Specifically, the basic situation of the power system of the example, the analysis of wind power output characteristics and the comparison with the change trend of the regional load curve are included. By configuring different proportions of energy storage capacity, the ICPSO optimization algorithm is used to solve the model one by one to obtain the corresponding wind power credible capacity under each energy storage ratio.

[0149] Taking a power grid in southern Yunnan Province as an example, the total installed capacity of the system is 2156MW, the current maximum load is 1400MW, and the wind power installed capacity is 339MW. The current wind power output and load change curve for 8760 hours a year is as follows: Figure 9 and Figure 10 As shown in the figure, it is obvious that wind power output is more volatile and intermittent, and wind power has a strong anti-peaking characteristic; the load has a strong diurnal variation, and apart from the stable output of conventional units, there is a low degree of synchronization between wind power and load. Therefore, it is necessary to configure energy storage devices to improve the temporal matching between wind power output and load demand, that is, to increase the reliable capacity of wind power, improve wind power output characteristics, and enhance the contribution of wind power to the power system; Figure 9 and Figure 10 shown.

[0150] According to the above analysis, due to the inherent characteristics of wind power fluctuation and intermittency, its reliable capacity remains at a low level. To further improve the reliable capacity level of wind power, a strategy of configuring energy storage systems in wind farms is proposed. Based on the aforementioned optimization model of wind power bundled with energy storage, the basic parameters of Wenshan Prefecture, such as power source and load, are substituted into the model. By configuring different proportions of energy storage capacity, the ICPSO optimization algorithm is used to solve the model one by one to obtain the corresponding wind power reliable capacity under each energy storage ratio. The calculation results are shown in Table 1 (Calculation table of different energy storage capacities and wind power reliable capacity) and Figure 11 shown.

[0151] Table 1 Calculation table of different energy storage capacities and wind power reliable capacity

[0152]

[0153] According to the above calculation results, the reliable capacity of wind power has been significantly improved after the addition of energy storage. Energy storage has a significant impact on the reliable capacity of wind power, especially when the reliable capacity of wind power for a specific load is small. There is a certain positive relationship between the energy storage capacity and the charging and discharging power and the reliable capacity of wind power. However, as the energy storage gradually increases, the improvement in the reliable capacity of wind power becomes smaller and smaller, and will tend to saturate when a certain ratio is reached. Because current energy storage applications still face the problem of high costs, over-investing in energy storage and increasing energy storage capacity to improve reliable capacity is not very meaningful. Planning decision makers can use the above sensitivity curve to select the appropriate energy storage system installation capacity according to the actual demand for reliable capacity. This will not only improve the reliable capacity level of wind power, but also meet the economic efficiency of energy storage system construction.

[0154] Yunnan boasts significant advantages in green energy resources and enormous development potential. The province strives to be the first in the nation to achieve carbon peak, helping accelerate progress in the eastern region and better serve and integrate into the national carbon peak and carbon neutrality strategies. Yunnan boasts a total wind energy reserve of 122.91 million kilowatts, with approximately 28 million kilowatts of exploitable wind energy reserves. Fifty-nine counties in Yunnan province have annual total solar radiation exceeding 5,500 MJ / m², and 12 counties exceed 6,000 MJ / m². The prospects for the development and utilization of wind and solar energy are promising. The proportion of wind and photovoltaic power generation capacity in Yunnan is also expected to gradually increase. By the end of the 14th Five-Year Plan and during the 15th Five-Year Plan, combined wind and photovoltaic capacity is projected to reach approximately 15% of the province's total power generation capacity. Yunnan's total clean energy capacity is projected to reach 130 million kilowatts by 2030. This will increase the credible capacity of wind power and play a crucial role in subsequent grid design, new energy planning, and other related tasks.

[0155] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing the credible capacity of a wind-storage system based on chaotic particle swarm optimization, characterized in that: The method comprises the following steps: Step 1: Collect the actual active power data of wind farms in operation in the region over the past years; Step 2: Construct a new energy output characteristic calculation model and indicator system from four dimensions: daily average output characteristics, monthly average output characteristics, reliable capacity and effective capacity; Step 3: The wind-storage system credible capacity optimization model is constructed with the goal of minimizing the sum of squares of the deviations between the total power generation of the wind-storage system and the planned output power; Step 4: Use the improved chaotic particle swarm optimization algorithm to solve the model and obtain the energy storage charging and discharging strategies under different scenarios; Step 5: Taking actual cases as an example, the sensitivity curve of wind power reliable capacity under different energy storage capacity configurations is studied; in: The new energy output characteristic calculation model and index system constructed in step 2 are as follows: (1) Daily average output characteristic index , and its calculation expression is as follows: in, Where: is the daily average output characteristic sequence of the wind farm; For wind farms within one day h The ratio of the mean active power to the installed capacity at a certain time; H Indicates the time of day: 23:00; N Indicates the whole year h The total number of active power data at a certain time; is the installed capacity of the wind farm; for h Point time j Active power value; (2) Monthly average output characteristic index , and its calculation expression is as follows: in, Where: is the monthly average output characteristic sequence of the wind farm; For wind farms within one year m The ratio of the monthly average active power to installed capacity; M Indicates December of a year; L Indicates the whole year m The total number of monthly active power data; is the installed capacity of the wind farm; for m Month j Active power value; (3) Trusted Capacity Indicator , and its calculation expression is as follows: in, Where: is the credible capacity sequence of the wind farm; For wind farms to ensure r The corresponding effective capacity is as follows; It indicates that the sequence of all active power values ​​in the evening peak period of the load within a year, arranged in descending order, is within the guaranteed rate r The corresponding active power value is as follows; R Indicates the maximum guarantee rate is 100%; is the installed capacity of the wind farm; (4) Effective capacity index , and its calculation expression is as follows: in, Where: is the effective capacity sequence of the wind farm; For wind farms to ensure r The corresponding effective capacity is as follows; Indicates the sequence of all active power values ​​in the low load period within a year in ascending order within the guaranteed rate r The corresponding active power value is as follows; R Indicates the maximum guarantee rate is 100%; is the installed capacity of the wind farm; In step 3, the specific steps of constructing the wind-storage system credible capacity optimization model with the goal of minimizing the sum of squares of the deviations between the total power generation power of the wind-storage system and the planned output power are as follows: (1) Wind power is an intermittent renewable energy source. With energy storage, its output can be smoothed and the "friendliness" of the renewable energy output can be improved. The objective function of the energy storage strategy is: in, In the formula, is the total power on the power supply side; Indicates the energy storage discharge power; Indicates the energy storage charging power; is wind power; is the power of conventional units; is the load power; It is the credible capacity value after the specified energy storage strategy acts on wind power; (2) In the process of smoothing the output of the wind farm, the energy storage system mainly considers the charging and discharging constraints and the power constraints of the energy storage system. In order to protect the battery, the minimum power stored in the battery shall not be less than 5% of the battery capacity, and the maximum power shall not be higher than 95% of the battery capacity. In addition, considering the constraints of the system on the wind power acceptance capacity, the constraints of the above optimization model are as follows: In the formula, For the i Power for a time period; is the capacity of the battery; is the charge-discharge conversion efficiency; (3) The constraints of energy storage are processed using the exterior point penalty function method. The objective function of the battery is expressed as follows: The corresponding penalty function constructed by the exterior point method is: First, the credible capacity value calculated by wind power alone is used as , bring it into the system calculation to optimize the specified objective function, and solve the energy storage charging and discharging strategy for each point; apply this strategy to the wind power system, obtain a new power curve on the power supply side, calculate the credible capacity of the wind power curve after energy storage optimization, and use this credible capacity to add it to the objective function of the next step, and iterate the calculation until the credible capacity value of the wind power energy storage system that meets the accuracy requirements is obtained, and the energy storage strategy that maximizes the credible capacity is obtained; In step 4, the improved chaotic particle swarm optimization algorithm is used to solve the model to obtain the specific steps of energy storage charging and discharging strategies under different scenarios: (1) Assume that the particle swarm i Particles in D The position and velocity in the solution space can be expressed as and ; In the optimization process, particles keep tracking their own individual optimal position and the global optimal position of the population To update its own state, through the personalized learning of particles and the collaborative interaction between each other, the entire group is constantly approaching the optimal solution to the problem; i Its velocity and position will be updated according to the following formula: In the formula, i =1, 2, 3, …, M ; d =1, 2, 3, …, D ; k for evolutionary algebra; w is the inertia weight factor; c 1 and c 2 is the acceleration factor, which is usually taken as 2.0; r 1 and r 2 is a random number between [0,1]. In addition, in order to improve the problem that the standard PSO algorithm is prone to fall into the local optimal value and thus premature convergence, the idea of ​​chaos optimization is introduced into the standard PSO algorithm, and a chaotic particle swarm optimization algorithm (CPSO) is proposed. The chaotic mutation processing of the optimal position of the group is carried out by using the characteristics of chaotic motion ergodicity and randomness to increase the diversity of the population and improve the global convergence performance of the algorithm. (2) In the actual optimization process, the inertia weight w It is mainly used to balance the global search and local search capabilities in the CPSO algorithm. This paper will adopt a strategy of adaptively adjusting the inertia weight based on the group's premature convergence degree and individual fitness value. The calculation formula is: In the formula, and yes w The upper and lower limits of the value of ; For particles Adaptability; is the mean fitness of all particles; The fitness of the particle swarm is better than The average fitness value of the particles; Used to evaluate the degree of premature convergence in the particle evolution process; generally ; In order to further improve the optimization efficiency of the algorithm, the convergence factor is introduced δ The particle position equation is modified as follows: in, (3) Compared with Logistic mapping, Tent mapping has better traversal uniformity and higher optimization efficiency. Therefore, Tent chaotic mutation processing is introduced to improve the problem that the algorithm is prone to fall into local optimality. The dynamic equation of Tent chaotic mapping is: (4) In order to avoid the problem that the iterative sequence falls into the small period point (0.2, 0.4, 0.6, 0.8) or the fixed point (0, 0.25, 0.5, 0.75), the perturbation equation is introduced to correct equation (10): (5) When the global optimal fitness value of the particle swarm stops updating for several consecutive times, the optimal particle position of the swarm Perform chaos mutation processing: In the formula, It is called the shrinkage factor, which determines the variation space of the particle position.

2. According to the method for optimizing the reliable capacity of a wind-storage system based on chaotic particle swarm optimization in claim 1, it is characterized in that: In the step 1, the actual active power data of the wind farm over the past years is mainly derived from the regional power grid dispatching and operation management system according to the conditions such as the wind farm name, grid-connected voltage level, and location.

3. A method for optimizing the reliable capacity of a wind-storage system based on chaotic particle swarm optimization according to claim 1 or 2, characterized in that: In step 5, the sensitivity change curve of wind power credible capacity under different energy storage capacity configurations is studied by taking an actual case as an example, specifically including the basic situation of the power system of the example, the analysis of wind power output characteristics and the comparison with the change trend of the regional load curve. By configuring energy storage capacity of different proportions, the ICPSO optimization algorithm is used to solve the model one by one to obtain the corresponding wind power credible capacity under each energy storage ratio.

Images