Scene generation and reduction-based wind and light energy random output prediction method and system

Abstract

The invention discloses a scene generation and reduction-based wind and light energy random output prediction method and system, and the method comprises the steps: obtaining wind energy and solar power generation historical data of a target region, and selecting Weibull distribution and Beta distribution to construct a wind and light probability density function; based on a wind and light output historical data set, using a Latin hypercube sampling method to extract sample points from the probability density function to represent different wind power and solar power generation scenes; and based on the generated scene, performing scene reduction by using a K-means clustering method, and finally generating a wind and light output curve. The system part comprises a data acquisition module, a scene generation module, a scene reduction module and an output curve generation module, and by adopting an advanced scene generation and reduction technology, the fluctuation of new energy power generation can be managed and reduced more effectively, so that the stability and reliability of a power grid are improved, and the power generation efficiency is improved. And the problem of wind and light uncertainty under large-scale access of new energy is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power data processing, and in particular to a method and system for predicting random output of wind and solar energy based on scenario generation and reduction. Background Art

[0002] Due to the inherent randomness, intermittency and volatility of renewable energy output, while renewable energy is facing unprecedented development opportunities, it also brings huge challenges to the planning and operation of power systems. Traditional power system planning and operation optimization methods based on deterministic theory can no longer effectively cope with today's power systems that are connected to a large number of renewable energy sources. How to safely and efficiently use renewable energy under highly uncertain conditions has become the focus of future development of power systems. Dealing with power system planning and operation issues under uncertainty can lay a theoretical foundation for further improving the penetration rate of renewable energy, increasing the power generation of renewable energy and the proportion of renewable energy in total power generation, and promoting the low-carbon, efficient and safe development of power systems.

[0003] Traditional power dispatching plans, especially for the generation plan formulation of the day before, usually do not take into account the participation of renewable energy, but regard it as a load that can be reduced. However, when the proportion of renewable energy in the power grid increases significantly, the situation becomes complicated. Before formulating the generation plan, the power generation of renewable energy must be predicted and the prediction results must be incorporated into the generation plan. In order to cope with the uncertainty of renewable energy generation, some additional backup power sources need to be arranged to compensate for the impact of forecast errors and power generation fluctuations. A major problem with this approach is that it usually formulates plans based on specific renewable energy forecasts, which may lead to large deviations between actual operations and plans. In other words, this traditional planning method may not be able to adapt to the variability and forecast errors of renewable energy generation, and thus cannot meet the needs of power generation plans under different power generation levels and forecast deviations. In short, the traditional method may not be able to flexibly adjust to the actual power generation situation when the proportion of renewable energy generation increases. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for predicting the random output of wind and solar energy based on scenario generation and reduction, so as to overcome the problem of wind and solar energy uncertainty under the large-scale access of new energy through scenario generation and reduction.

[0005] To achieve the above object, the present invention provides the following technical solutions: The method for predicting the random output of wind and solar energy based on scenario generation and reduction includes the following steps: S1: Obtain historical data of wind and solar power generation in the target area; S2: Based on historical data, Weibull distribution and Beta distribution are selected to construct wind and light probability density functions; S3: Based on the historical data set of wind and solar power output, Latin hypercube sampling method is used to extract sample points from the probability density function. The sample points are evenly distributed in the multidimensional parameter space to represent different wind and solar power generation scenarios; S4: Based on the generated scenes, the K-means clustering method is used to reduce the scenes, and the K value is determined by the elbow rule and silhouette coefficient; S5: Generate wind and solar power output curve.

[0006] Furthermore, the Weibull distribution is used in S2 to describe the uncertainty of wind speed v, and its probability density function is: in, is the wind speed, is the shape parameter, is a scale parameter. Furthermore, wind power output With wind speed With wind speed The relationship between is expressed as a piecewise function:

[0007] Where: is the cut-in speed of wind power; is the rated speed of wind power; is the cut-out speed of wind power; is the rated active power of the fan.

[0008] Furthermore, the Beta distribution is used in S2 to construct the photovoltaic power generation probability density function as follows:

[0009] Where: is the maximum power output of photovoltaic power generation; Г is the gamma function; is the Beta distribution parameter, Its expression is:

[0010] Where: is the average light intensity over a period of time; is the variance of light intensity.

[0011] The wind and solar energy random output prediction system based on scenario generation and reduction includes: Data acquisition module, used to obtain historical data of wind and solar power generation in the target area; The scenario generation module first collects comprehensive historical data on wind and solar power output from the target area, including but not limited to meteorological data on wind speed, solar radiation intensity, temperature, and socioeconomic indicators related to electricity demand. The collected data is cleaned, including removing outliers, filling missing values, and standardizing and normalizing the data. The historical data and selected features are used to generate wind and solar power output scenarios through Latin hypercube sampling. The scenario reduction module is used to optimize and streamline the large number of renewable energy output scenarios generated by the scenario generation module; The curve generation module is a tool for simulating and analyzing the output characteristics of wind and solar power generation systems. It generates output curves that reflect the volatility and seasonal changes of wind and solar power generation by integrating meteorological data, historical power generation data and genetic algorithms.

[0012] Furthermore, the scene reduction module uses a K-means clustering algorithm to group similar scenes into several representative clusters, so as to identify and retain the most representative and diverse scenes, while removing scenes that are not statistically significant or too similar.

[0013] Furthermore, it also includes a computer device, which is provided with a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the above method.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The random output prediction method and system of wind and solar energy based on scenario generation and reduction of the present invention are highly applicable. By comprehensively considering the meteorological conditions, historical power generation data and socio-economic indicators of the target area, a highly representative wind and solar output scenario is generated. The method does not rely on specific geographical or climatic conditions, so it has good universality and can be applied to different regions and environments. In addition, the probability density function and clustering algorithm adopted by the method can flexibly adapt to wind and solar power generation facilities of different scales and types. Whether it is a large-scale wind farm or a distributed solar power generation system, accurate output prediction and effective scheduling support can be obtained. This high applicability enables the present invention to provide strong technical support for the planning and operation of various power systems.

[0015] 2. The random output prediction method and system of wind and solar energy based on scenario generation and reduction of the present invention are highly accurate. By using Weibull distribution and Beta distribution to approximate the probability density function of wind power and photovoltaic power generation, it can more accurately describe the probability distribution characteristics of wind and solar power generation output, thereby improving the accuracy of the prediction.

[0016] 3. The random output prediction method and system of wind and solar energy based on scenario generation and reduction of the present invention are easy to integrate. The proposed wind and solar output prediction method and system design take into account the compatibility with the existing power system management and dispatching platform, ensuring a high degree of integration. The system adopts a modular design, and each component such as data acquisition, scenario generation, scenario reduction, and output curve generation are designed to operate independently while being seamlessly integrated into a wider power system architecture. This design allows power companies to flexibly select and integrate the required modules according to their own needs, whether it is adding new functions to the existing power dispatching system or deploying a complete solution in a brand new system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 A curve diagram initially generated for wind and solar power output of the present invention; Figure 3 It is a typical output diagram after clustering of the present invention; Figure 4 A schematic diagram of the structure of a system in an embodiment of the present invention; Figure 5 1 is a diagram showing the internal structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] like Figure 1 As shown: The method for predicting the random output of wind and solar energy based on scenario generation and reduction provided in an embodiment of the present invention comprises the following steps: S1: Obtain historical data on wind and solar power generation in the target area. These data cover daily power generation within a preset period of time. By analyzing these historical data, we can understand the volatility and predictability of wind and solar power generation, and provide data support for the dispatch and management of the power grid. S2: Based on historical data, Weibull distribution and Beta distribution are selected to construct wind and light probability density functions; specifically, wind power output is affected by wind speed and presents relatively large randomness and volatility. In order to accurately describe the probability distribution characteristics of wind power output, probability density functions are usually used for modeling and analysis. The present invention uses Weibull distribution to describe the uncertainty of wind speed v, and its probability density function is:

[0020] in, is the wind speed, is the shape parameter, As the scale parameter, the Weibull distribution can well describe the wind power output characteristics under different wind speed conditions; Wind power output With wind speed The relationship between can be approximately expressed as a piecewise function:

[0021] Where: is the cut-in speed of wind power; is the rated speed of wind power; is the cut-out speed of wind power; is the rated active power of the fan.

[0022] The output power of photovoltaic power generation is mainly affected by factors such as solar irradiance, temperature, and weather conditions. It has significant randomness and intermittency. Photovoltaic power generation is similar to wind power generation in that it is greatly affected by natural climate and weather conditions. The intensity of solar light over a period of time can be approximately regarded as a Beta distribution:

[0023] Where: is the maximum power output of photovoltaic power generation; Г is the gamma function; is the Beta distribution parameter, and its expression is:

[0024] Where: is the average light intensity over a period of time; is the variance of light intensity.

[0025] S3: Based on the historical data set of wind and solar power output, Latin hypercube sampling method is used to extract sample points from the probability density function. The sample points are evenly distributed in the multidimensional parameter space to represent different wind and solar power generation scenarios; S4: Based on the generated scenes, the K-means clustering method is used to reduce the scenes, and the K value is determined by the elbow rule and silhouette coefficient; S5: Generate wind and solar output curve, such as Figure 2 shown.

[0026] like Figure 4 To further better explain the above method, an embodiment of the present invention further provides a prediction system for implementing the above method, which includes: The data acquisition module is used to obtain historical data on wind and solar power generation in the target area. These data cover daily power generation within a preset period of time. By analyzing these historical data, the volatility and predictability of wind and solar power generation can be understood, thereby providing data support for the dispatch and management of the power grid. The scenario generation module first collects comprehensive historical data on wind and solar power output from the target area, including but not limited to meteorological data such as wind speed, solar radiation intensity, temperature, and socioeconomic indicators related to electricity demand; cleans the collected data, including removing outliers, filling missing values, and standardizing and normalizing the data to ensure data quality and provide accurate input for subsequent analysis. It uses historical data and selected features to generate a series of possible wind and solar power output scenarios through Latin hypercube sampling, which cover different weather conditions, seasonal changes, and load demand patterns; The Scenario Reduction Module is an efficient data processing tool designed to optimize and streamline the large number of renewable energy output scenarios generated by the Scenario Generation Module. This module uses the K-means clustering algorithm, an unsupervised learning method that can group similar scenarios into several representative clusters. In this way, the Scenario Reduction Module can identify and retain the most representative and diverse scenarios while removing those that are not statistically significant or too similar; The curve generation module is a tool for simulating and analyzing the output characteristics of wind and solar power generation systems. By integrating meteorological data, historical power generation data and genetic algorithms, this module can generate output curves that reflect the volatility and seasonal changes of wind and solar power generation.

[0027] like Figure 5 As shown, the wind and solar energy random output prediction system based on scenario generation and reduction provided in the above embodiment also includes a computer device, which is provided with a memory, a processor and a computer program stored in the memory and can be run on the processor, wherein the processor is used to execute the computer program, thereby implementing the steps of the above method. In order to further verify the effectiveness of the model built in the above method, this embodiment uses a calculation example for calculation and analysis, and the system wind power output prediction curve is as follows Figure 3 As shown, the wind and solar power output data within 24 hours is taken as the original data, and the data is shown in Table 1: Table 1: Original data of wind and solar power output from 0 to 24 hours Wind power output Photovoltaic output 341 0 290 0 450 0 471 0 515 0 540 0 527 99 641 138 634 156 525 180 400 188 640 193 595 176 540 170 512 140 505 101 201 77 84 0 81 0 80 0 85 0 110 0 353 0 523 0 Matlab was used to verify the operation. 1000 scenarios were randomly generated based on the data within 24 hours. The final scenario generation diagram is shown in the figure below. Figure 2As shown in the figure, it can be concluded that the random scene curve obeys a certain distribution function; then the 1000 scenes are iteratively reduced until 20 typical scenes are reduced, and finally the typical scene wind and solar output curve is obtained as follows Figure 3 shown.

[0028] It can be seen from the above that the advantage of the present invention is that it does not rely on specific geographical or climatic conditions, is suitable for different regions and environments, can flexibly adapt to wind and solar power generation facilities of different sizes and types, and adopts Weibull distribution and Beta distribution to approximate the probability density function of wind power and photovoltaic power generation to improve the accuracy of prediction. At the same time, the design takes into account the compatibility with the existing power system management and dispatching platform, adopts modular design, and is easy to integrate into a wider power system architecture. By constructing multiple possible new energy output scenarios, it comprehensively considers and evaluates various possible power generation situations, and provides accurate and flexible decision support for power grid dispatching. Through the scenario reduction mechanism, the most representative and most likely scenarios are intelligently screened out, unnecessary calculations and analysis work are reduced, and the efficiency of power generation plan formulation is improved. It provides a more scientific and reasonable power generation plan compilation method for power systems under large-scale new energy access, which helps to improve the adaptability of the power grid to new energy fluctuations and provide strong guarantees for the safety, economy and sustainable development of the power system.

[0029] To sum up: the random output prediction method and system of wind and solar energy based on scenario generation and reduction provided by the present invention are used to deal with the output uncertainty problem faced when new energy sources such as wind energy and solar energy are connected to the power grid on a large scale. By adopting advanced scenario generation and reduction technologies, it is possible to more effectively manage and reduce the volatility of new energy power generation, thereby improving the stability and reliability of the power grid; by constructing multiple possible new energy output scenarios, various possible power generation situations are comprehensively considered and evaluated, which can not only predict the power generation of new energy, but also evaluate the power generation fluctuations under different situations, providing more accurate and flexible decision support for power grid dispatching.

[0030] In addition, the present invention also includes a scenario reduction mechanism, which selects the most representative and most likely scenarios through computer programs, thereby reducing unnecessary calculations and analysis work and improving the efficiency of power generation plan formulation. This method helps to optimize the configuration of power generation resources and ensure that reasonable power generation plans can be formulated under various new energy output levels and forecast deviation conditions, providing a more scientific and reasonable power generation plan formulation method for power systems under large-scale new energy access. This can not only improve the adaptability of the power grid to new energy fluctuations, but also provide strong guarantees for the safety, economy and sustainable development of the power system.

[0031] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A random output prediction method for wind and solar energy based on scenario generation and reduction, characterized in that: The following steps are involved: S1: Obtain historical data of wind and solar power generation in the target area; S2: Based on historical data, Weibull distribution and Beta distribution are selected to construct wind and light probability density functions; S3: Based on the historical data set of wind and solar power output, Latin hypercube sampling method is used to extract sample points from the probability density function. The sample points are evenly distributed in the multidimensional parameter space to represent different wind and solar power generation scenarios; S4: Based on the generated scenes, the K-means clustering method is used to reduce the scenes, and the K value is determined by the elbow rule and silhouette coefficient; S5: Generate wind and solar power output curve.

2. The method for predicting random output of wind and solar energy based on scenario generation and reduction as claimed in claim 1, characterized in that: In S2, Weibull distribution is used to describe the uncertainty of wind speed v, and its probability density function is: in, is the wind speed, is the shape parameter, is the scale parameter.

3. The method for predicting random output of wind and solar energy based on scenario generation and reduction as claimed in claim 2, characterized in that: Wind power output With wind speed The relationship between is expressed as a piecewise function: Where: is the cut-in speed of wind power; is the rated speed of wind power; is the cut-out speed of wind power; is the rated active power of the fan.

4. The method for predicting random output of wind and solar energy based on scenario generation and reduction as claimed in claim 1, characterized in that: The Beta distribution is used in S2 to construct the photovoltaic power generation probability density function: Where: is the maximum power output of photovoltaic power generation; Г is the gamma function; is the Beta distribution parameter, and its expression is: Where: is the average light intensity over a period of time; is the variance of light intensity.

5. A wind and solar energy random output prediction system based on scenario generation and reduction, characterized by: include: Data acquisition module, used to obtain historical data of wind and solar power generation in the target area; The scenario generation module first collects comprehensive historical data on wind and solar power output from the target area, including but not limited to meteorological data on wind speed, solar radiation intensity, temperature, and socioeconomic indicators related to electricity demand. The collected data is cleaned, including removing outliers, filling missing values, and standardizing and normalizing the data. The historical data and selected features are used to generate wind and solar power output scenarios through Latin hypercube sampling. The scenario reduction module is used to optimize and streamline the large number of renewable energy output scenarios generated by the scenario generation module; The curve generation module is a tool for simulating and analyzing the output characteristics of wind and solar power generation systems. It generates output curves that reflect the volatility and seasonal changes of wind and solar power generation by integrating meteorological data, historical power generation data and genetic algorithms.

6. The wind and solar energy random output prediction system based on scenario generation and reduction as claimed in claim 5, characterized in that: The scene reduction module uses a K-means clustering algorithm to group similar scenes into several representative clusters to identify and retain the most representative and diverse scenes, while removing scenes that are not statistically significant or too similar.

7. The wind and solar energy random output prediction system based on scenario generation and reduction as claimed in claim 5, characterized in that: It also includes a computer device, which is provided with a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the above method.

Images