Power system scheduling method and system based on wind power consumption capability evaluation
The wind power consumption capacity is evaluated through the hidden Markov model and the covariance matrix adaptive evolution algorithm, and the problems of large delay and high error of power system scheduling caused by wind power uncertainty are solved, and fast, automatic and accurate power system scheduling is achieved. Improve wind power consumption capacity and resource utilization rate.
Patent Information
- Application Number
- CN202410980007.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-22
AI Technical Summary
The uncertainty of wind power leads to large delays in the scheduling of power systems and high errors, resulting in wind and power waste, and reduced resource waste and profits.
The hidden Markov model is used to evaluate the matching degree of wind power and load scenarios, establish a wind power consumption capacity assessment model, and optimize the power system scheduling through the covariance matrix adaptive evolution algorithm to achieve fast, automatic and accurate power system scheduling.
It improves wind power consumption capacity and resource utilization rate, reduces wind and power waste, and improves the flexibility and response speed of the power system.
Smart Images

Figure QLYQS_1 
Figure QLYQS_2 
Figure QLYQS_3
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of wind power consumption, and in particular to a method and system for dispatching a power system based on wind power consumption capacity assessment. Background Art
[0002] As an important support for the transformation of energy structure, wind power grid connection can effectively reduce the use of thermal power technology and greatly reduce carbon dioxide emissions. However, wind power has significant uncertainty. Its large-scale grid connection has led to a sharp increase in the demand for power system flexibility, and the absorption problem has brought new challenges to the regulation and support capabilities of the power system. In the current environment of imperfect power system scheduling, the mismatch between source and load in time and space has led to widespread abandonment of wind power, resulting in waste of wind resources and reduced profits.
[0003] Currently, deep peak regulation of coal-fired power units is usually adopted to promote the consumption of new energy. Traditional solutions mostly use random optimization and robust optimization methods to dispatch power systems. Traditional solutions usually have the problem of one-sided conservatism, resulting in large delays, high errors and insufficient accuracy in power system dispatching. Summary of the invention
[0004] Purpose of the invention: The purpose of the present invention is to provide a method and system for dispatching a power system based on wind power absorption capacity assessment to solve the problems raised in the background technology.
[0005] Technical solution: The power system dispatching method based on wind power absorption capacity evaluation described in the present invention comprises the following steps:
[0006] (1) Use hidden Markov models to create wind power and load scenarios and calculate matching;
[0007] (2) Establish a wind power absorption capacity assessment model;
[0008] (3) The covariance matrix adaptive evolutionary algorithm is used to solve the optimal solution for power system dispatch based on wind power absorption capacity assessment; and a power dispatch system based on wind power absorption capacity assessment is constructed.
[0009] Furthermore, step (1) is specifically as follows: first input wind power and load historical data; calculate the target association matrix; randomly generate scenarios; use the hidden Markov model to calculate the target association matrix and the target normalized moment; check whether the result error is less than the threshold, if not, repeat the iteration of the previous step until the error converges; transform the normalized scenario, and output the result and the target moment.
[0010] Furthermore, in step (1), the wind power and load matching calculation formula is:
[0011]
[0012] Among them, ψ is the matching degree of wind power and load; γ is the slope correlation of the daily average load curve; P′ v is the peak-to-valley difference rate of the generalized daily average power load; f is the load fluctuation rate.
[0013] Furthermore, in step (2), the wind power absorption capacity is mainly measured from three aspects, namely, wind farm power generation, wind power abandonment and wind energy utilization rate, as evaluation indicators of wind power absorption capacity, and the calculation formula is:
[0014]
[0015] Among them, E t 、E quit and χ W are wind farm power generation, wind power abandonment and wind energy utilization rate respectively; m is the wind farm unit number; o is the sampling period; P m,o is the power of unit m in period o; E com Indicates the grid-connected wind power;
[0016] Combining wind power and load matching, the following wind power consumption assessment model is constructed:
[0017]
[0018] Among them, F(t) is the wind power absorption capacity evaluation function; It is the expansion factor of wind power installation; is the maximum value of wind power output; P W,t is the wind power output value corresponding to time t.
[0019] Furthermore, step (3) is specifically as follows: first, a series of initial solutions are randomly generated using normal distribution as the initial population, and then the fitness value of each individual in the population is calculated, and the first n individuals with the best fitness values are selected as the parents of the next generation. A new population is generated according to the normal distribution characteristics based on the individual position information, and the optimization is stopped by iterating until a satisfactory solution is found or the stopping condition is reached.
[0020] The present invention provides a power system dispatching system based on wind power absorption capacity assessment, comprising:
[0021] Creation module: used to create wind power,load scenarios and calculate matching degree using hidden Markov models;
[0022] Evaluation module: used to establish a wind power absorption capacity evaluation model;
[0023] Solution module: used to use the covariance matrix adaptive evolutionary algorithm to solve the optimal solution for power system dispatch based on wind power absorption capacity evaluation; to build a power dispatch system based on wind power absorption capacity evaluation.
[0024] Furthermore, in the creation module, the details are as follows: first input wind power and load historical data; calculate the target association matrix; randomly generate scenarios; use the hidden Markov model to calculate and obtain the target association matrix and target normalized moment; check whether the result error is less than the threshold, if not, repeat the iteration of the previous step until the error converges; transform the normalized scenario, and output the result and target moment;
[0025] Furthermore, in the creation module, the calculation formula for wind power and load matching is:
[0026]
[0027] Among them, ψ is the matching degree of wind power and load; γ is the slope correlation of the daily average load curve; P′ v is the peak-to-valley difference rate of the generalized daily average power load; f is the load fluctuation rate.
[0028] Furthermore, in the evaluation module, the wind power absorption capacity is mainly measured from three aspects, namely, wind farm power generation, wind power abandonment and wind energy utilization rate. As the evaluation index of wind power absorption capacity, the calculation formula is:
[0029]
[0030] Among them, E t 、E quit and χ W are wind farm power generation, wind power abandonment and wind energy utilization rate respectively; m is the wind farm unit number; o is the sampling period; P m,o is the power of unit m in period i; E com Indicates the grid-connected wind power;
[0031] Combining wind power and load matching, the following wind power consumption assessment model is constructed:
[0032]
[0033] Among them, F(t) is the wind power absorption capacity evaluation function; is the expansion factor of wind power installations; is the maximum value of wind power output; P W,t is the wind power output value corresponding to time t.
[0034] Furthermore, in the solution module, the details are as follows: first, a series of initial solutions are randomly generated using normal distribution as the initial population, and then the fitness value of each individual in the population is calculated, and the first n individuals with the best fitness values are selected as the parents of the next generation. A new population is generated according to the normal distribution characteristics based on the individual position information, and the optimization is stopped by iterating until a satisfactory solution is found or the stopping condition is reached.
[0035] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: a power system dispatching method based on the covariance matrix adaptive evolutionary algorithm to evaluate wind power absorption capacity, taking into account the matching degree of wind power and load, wind power output characteristics, and using an intelligent optimization algorithm to improve system calculation and response speed. The main advantage is to achieve fast, automatic and accurate power system dispatching, and improve wind power absorption capacity and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 A schematic diagram designed for the present invention;
[0037] Figure 2 A comparison chart of the speed and accuracy of the method of the present invention and the traditional method;
[0038] Figure 3 This is the electric power system dispatching curve diagram of the present invention. DETAILED DESCRIPTION
[0039] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.
[0040] like Figure 1-3 As shown, an embodiment of the present invention provides a power system dispatching method based on wind power absorption capacity evaluation, comprising the following steps:
[0041] (1) Use the hidden Markov model to create wind power and load scenarios and calculate the matching degree; the details are as follows: first input the historical data of wind power and load; calculate the target association matrix; randomly generate scenarios; use the hidden Markov model to calculate the target association matrix and target normalized moment; check whether the result error is less than the threshold, if not, repeat the iteration of the previous step until the error converges; transform the normalized scenario, and output the result and target moment;
[0042] The calculation formula for wind power and load matching is:
[0043]
[0044] Among them, ψ is the matching degree of wind power and load; γ is the slope correlation of the daily average load curve; P′ v is the peak-to-valley difference rate of the generalized daily average power load; f is the load fluctuation rate.
[0045] (2) Establish a wind power absorption capacity assessment model; wind power absorption capacity is mainly measured from three aspects, namely wind farm power generation, wind power abandonment and wind energy utilization rate, which are used as wind power absorption capacity assessment indicators. The calculation formula is:
[0046]
[0047] Among them, E t 、E quit and χ W are wind farm power generation, wind power abandonment and wind energy utilization rate respectively; m is the wind farm unit number; o is the sampling period; P m,o is the power of unit m in period o; E com Indicates the grid-connected wind power;
[0048] Combining wind power and load matching, the following wind power consumption assessment model is constructed:
[0049]
[0050] Among them, F(t) is the wind power absorption capacity evaluation function; is the expansion factor of wind power installations; is the maximum value of wind power output; P W,t is the wind power output value corresponding to time t.
[0051] (3) The covariance matrix adaptive evolutionary algorithm is used to solve the optimal solution for power system dispatch based on wind power absorption capacity evaluation; and a power dispatch system based on wind power absorption capacity evaluation is constructed. The details are as follows: First, a series of initial schemes are randomly generated using normal distribution as the initial population, and then the fitness value of each individual in the population is calculated, and the first n individuals with the best fitness value are selected as the parents of the next generation. A new population is generated according to the normal distribution characteristics based on the individual position information, and the optimization is stopped by iterating until a satisfactory solution is found or the stopping condition is reached.
[0052] The embodiment of the present invention further provides a power system dispatching system based on wind power absorption capacity evaluation, including:
[0053] Creation module: used to create wind power and load scenarios and calculate matching degree using hidden Markov model; specifically as follows: first input wind power and load historical data; calculate target association matrix; randomly generate scenarios; use hidden Markov model to calculate target association matrix and target normalized moment; check whether the result error is less than the threshold, if not, repeat the previous iteration until the error converges; transform the normalized scenario, output the result and target moment;
[0054] The calculation formula for wind power and load matching is:
[0055]
[0056] Among them, Ψ is the matching degree of wind power and load; γ is the slope correlation of the daily average load curve; P′ v is the peak-to-valley difference rate of the generalized daily average power load; F is the load fluctuation rate.
[0057] Evaluation module: used to establish a wind power absorption capacity evaluation model; wind power absorption capacity is mainly measured from three aspects, namely wind farm power generation, wind power abandonment and wind energy utilization rate, as wind power absorption capacity evaluation indicators, the calculation formula is:
[0058]
[0059] Among them, E t 、E quit and χ W are wind farm power generation, wind power abandonment and wind energy utilization rate respectively; M is the wind farm unit number; o is the sampling period; P m,o is the power of unit m in period o; E com Indicates the grid-connected wind power;
[0060] Combining wind power and load matching, the following wind power consumption assessment model is constructed:
[0061]
[0062] Among them, F(t) is the wind power absorption capacity evaluation function; is the expansion factor of wind power installations; is the maximum value of wind power output; P W,t is the wind power output value corresponding to time t.
[0063] Solution module: used to solve the optimal solution of power system dispatch based on wind power absorption capacity assessment by using covariance matrix adaptive evolutionary algorithm; to build a power dispatch system based on wind power absorption capacity assessment. The details are as follows: first, a series of initial schemes are randomly generated as the initial population using normal distribution, and then the fitness value of each individual in the population is calculated, and the first n individuals with the best fitness value are selected as the parents of the next generation. A new population is generated according to the normal distribution characteristics based on the individual position information, and the optimization is stopped by iteration until a satisfactory solution is found or the stopping condition is reached. .
Claims
1. A power system dispatching method based on wind power absorption capacity assessment, It is characterized in that The following steps are involved: (1) Use hidden Markov models to create wind power and load scenarios and calculate matching; (2) Establish a wind power absorption capacity assessment model; (3) The covariance matrix adaptive evolutionary algorithm is used to solve the optimal solution for power system dispatch based on wind power absorption capacity assessment; and a power dispatch system based on wind power absorption capacity assessment is constructed.
2. According to claim 1, a power system dispatching method based on wind power absorption capacity evaluation, It is characterized in that Step (1) is as follows: first input wind power and load historical data; calculate the target association matrix; randomly generate scenarios; use the hidden Markov model to calculate the target association matrix and target normalized moment; check whether the result error is less than the threshold, if not, repeat the iteration of the previous step until the error converges; Transform the normalized scene and output the result and target moments.
3. According to the power system dispatching method based on wind power absorption capacity evaluation according to claim 1, It is characterized in that In step (1), the wind power and load matching calculation formula is: Among them, ψ is the matching degree of wind power and load; γ is the slope correlation of the daily average load curve; P′ v is the peak-to-valley difference rate of the generalized daily average power load; f is the load fluctuation rate.
4. A method for dispatching a power system based on wind power absorption capacity evaluation according to claim 1, It is characterized in that In step (2), the wind power absorption capacity is mainly measured from three aspects, namely, wind farm power generation, wind power abandonment and wind energy utilization rate, which are used as evaluation indicators of wind power absorption capacity. The calculation formula is: Among them, E t 、E qui t and χ W are wind farm power generation, wind power abandonment and wind energy utilization rate respectively; m is the wind farm unit number; o is the sampling period; P m,o is the power of unit m in period o; E com Indicates the grid-connected wind power; Combining wind power and load matching, the following wind power consumption assessment model is constructed: Among them, F(t) is the wind power absorption capacity evaluation function; It is the expansion factor of wind power installation; is the maximum value of wind power output; P W,t is the wind power output value corresponding to time t.
5. A method for dispatching a power system based on wind power absorption capacity evaluation according to claim 1, It is characterized in that Step (3) is as follows: first, a series of initial solutions are randomly generated using normal distribution as the initial population, then the fitness value of each individual in the population is calculated, and the first n individuals with the best fitness values are selected as the parents of the next generation. A new population is generated according to the normal distribution characteristics based on the individual position information, and the optimization is stopped by iterating until a satisfactory solution is found or the stopping condition is reached.
6. A power system dispatching system based on wind power absorption capacity assessment, It is characterized in that include: Creation module: used to create wind power,load scenarios and calculate matching degree using hidden Markov models; Evaluation module: used to establish a wind power absorption capacity evaluation model; Solution module: used to use the covariance matrix adaptive evolutionary algorithm to solve the optimal solution for power system dispatch based on wind power absorption capacity evaluation; to build a power dispatch system based on wind power absorption capacity evaluation.
7. The power system dispatching system based on wind power absorption capacity evaluation according to claim 1, It is characterized in that In the creation module, the details are as follows: first input wind power and load historical data; calculate the target association matrix; randomly generate scenarios; use the hidden Markov model to calculate the target association matrix and target normalized moment; check whether the result error is less than the threshold, if not, repeat the previous iteration until the error converges; Transform the normalized scene and output the result and target moments.
8. The power system dispatching system based on wind power absorption capacity evaluation according to claim 1, It is characterized in that In the creation module, the calculation formula for wind power and load matching is: Among them, ψ is the matching degree of wind power and load; γ is the slope correlation of the daily average load curve; P′ v is the peak-to-valley difference rate of the generalized daily average power load; f is the load fluctuation rate.
9. The power system dispatching system based on wind power absorption capacity evaluation according to claim 1, It is characterized in that In the evaluation module, wind power absorption capacity is mainly measured from three aspects, namely wind farm power generation, wind power abandonment and wind energy utilization rate. As the evaluation index of wind power absorption capacity, the calculation formula is: Among them, E t 、E quit and χ W are wind farm power generation, wind power abandonment and wind energy utilization rate respectively; m is the wind farm unit number; o is the sampling period; P m,o is the power of unit m in period o; E com Indicates the grid-connected wind power; Combining wind power and load matching, the following wind power consumption assessment model is constructed: Among them, F(t) is the wind power absorption capacity evaluation function; It is the expansion factor of wind power installation; is the maximum value of wind power output; P W,t is the wind power output value corresponding to time t.
10. The power system dispatching system based on wind power absorption capacity evaluation according to claim 1, It is characterized in that In the solution module, the details are as follows: first, a series of initial solutions are randomly generated using normal distribution as the initial population, then the fitness value of each individual in the population is calculated, and the first n individuals with the best fitness values are selected as the parents of the next generation. A new population is generated according to the normal distribution characteristics based on the individual position information, and the optimization is stopped by iterating until a satisfactory solution is found or the stopping condition is reached.
