Optimal dispatching method for hydro-wind-solar hybrid power generation system based on wind-solar forecast

Abstract

The present invention discloses a method for optimizing and dispatching a water-wind-solar complementary power generation system based on wind-solar forecasting, which relates to the field of intelligent power dispatching technology. The method comprises: collecting historical meteorological data of a target power plant for meteorological clustering to generate multiple weather scene categories; collecting real-time wind-solar meteorological data and performing periodic forecasting to obtain a target weather scene change sequence; generating an initial power supply strategy based on the sequence; analyzing the correlation between wind-solar environmental indicators and power generation modules, extracting key influencing factors and generating a multi-dimensional power generation evaluation network; optimizing the initial power supply strategy using the evaluation network, forming a power supply optimization strategy, and implementing dynamic power generation scheduling. This method solves the technical problem that the existing water-wind-solar complementary power generation system is difficult to achieve flexible power generation ratio adjustment, resulting in low power generation efficiency and low reliability, and achieves the technical effect of dynamically optimizing the power supply ratio of the water-wind-solar complementary power generation system and improving the power generation efficiency and reliability of the system.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent power dispatching, and in particular to a method for optimizing the dispatching of a water-wind-solar complementary power generation system based on wind-solar prediction. Background Art

[0002] With the transformation of the global energy mix and the rapid development of renewable energy, hydropower, windpower, and photovoltaic (PV) hybrid power generation systems are becoming an important clean energy solution. Combining the advantages of hydropower, wind power, and photovoltaic power generation, they effectively improve energy efficiency. However, the uncertainty of weather changes poses challenges to the stability and economic viability of power generation. Existing technologies lack effective methods for dynamically adjusting the hydropower, wind, and solar power generation ratio based on accurate weather forecasts, resulting in low power generation efficiency and increased costs. Therefore, leveraging historical meteorological data and real-time wind and solar power forecasts to optimize system scheduling has become a key requirement for improving the performance of renewable energy systems.

[0003] At present, there is a technical problem in relevant technologies that the water-wind-solar complementary power generation system is difficult to achieve flexible adjustment of the power generation ratio, resulting in low power generation efficiency and low reliability. Summary of the Invention

[0004] The present application provides a method for optimizing and dispatching a water-wind-solar complementary power generation system based on wind-solar forecasting. The method collects historical meteorological data of a target power plant and uses a probability distribution network for meteorological clustering to generate multiple weather scene categories. Then, real-time wind-solar meteorological data is collected and periodic forecasts are performed, and a target weather scene change sequence is obtained by matching weather scenes. Then, an initial power supply strategy is generated based on the sequence, including multiple power generation ratio switching nodes. Subsequently, the correlation between wind-solar environmental indicators and power generation modules is analyzed, key influencing factors are extracted, and a multidimensional power generation evaluation network is generated. Finally, the evaluation network is used to optimize the initial power supply strategy, form a power supply optimization strategy, and implement dynamic power generation scheduling. This achieves the technical effect of dynamically optimizing the power supply ratio of the water-wind-solar complementary power generation system and improving the power generation efficiency and reliability of the system by combining real-time meteorological data collection with wind-solar forecasting.

[0005] This application provides a method for optimizing the scheduling of a hydro-wind-solar hybrid power generation system based on wind-solar forecasting, including:

[0006] Collect historical meteorological data of the target power plant, use probability distribution network to perform meteorological clustering, and generate multiple weather scene categories; collect real-time wind and solar meteorological data, perform periodic wind and solar forecasts, and traverse the acquired wind and solar forecast data through the multiple weather scene categories to perform weather scene matching to obtain a target weather scene change sequence; based on the target weather scene change sequence, generate an initial power supply strategy for the water-wind-solar complementary power generation system, and the initial power supply strategy includes multiple power generation ratio switching nodes; analyze the correlation between wind and solar environmental indicators and water-wind-solar power generation terminals, extract key influencing factors, and generate a multidimensional power generation evaluation network based on the key influencing factors; based on the multidimensional power generation evaluation network, evaluate and optimize the initial power supply strategy, generate a power supply optimization strategy, and use the power supply optimization strategy for dynamic power generation scheduling.

[0007] The method for optimizing and dispatching a water-wind-solar complementary power generation system based on wind-solar forecast proposed in this application first collects historical meteorological data of the target power plant and uses a probability distribution network to perform meteorological clustering to generate multiple weather scene categories; then, real-time wind-solar meteorological data is collected and periodic forecasts are performed, and a target weather scene change sequence is obtained by matching weather scenes; then, an initial power supply strategy is generated based on the sequence, including multiple power generation ratio switching nodes; subsequently, the correlation between wind-solar environmental indicators and power generation modules is analyzed, key influencing factors are extracted, and a multidimensional power generation evaluation network is generated; finally, the evaluation network is used to optimize the initial power supply strategy, form a power supply optimization strategy, and implement dynamic power generation scheduling, thereby achieving the technical effect of dynamically optimizing the power supply ratio of the water-wind-solar complementary power generation system and improving the power generation efficiency and reliability of the system through the combination of real-time meteorological data collection and wind-solar forecast. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0009] Figure 1 A flow chart of a method for optimizing the scheduling of a hydro-wind-solar hybrid power generation system based on wind-solar forecasting provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of a multi-dimensional power generation evaluation network flow for a method for optimizing scheduling of a water-wind-solar complementary power generation system based on wind-solar prediction provided in an embodiment of the present application. DETAILED DESCRIPTION

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0012] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0014] The embodiment of the present application provides a method for optimizing the scheduling of a water-wind-solar complementary power generation system based on wind-solar prediction, such as Figure 1 As shown, the method includes:

[0015] Step S100: Collect historical meteorological data of the target power plant, perform meteorological clustering using a probability distribution network, and generate multiple weather scene categories. Specifically, in the process of generating weather scene categories based on the probability distribution network, first, extensive historical meteorological data of the target power plant is collected, including long-term observation records of the meteorological department (such as temperature, air pressure, humidity, wind speed, etc., hourly data for many years), satellite meteorological data (such as cloud cover, radiation intensity, etc.), and the power plant's own monitoring data (such as local wind speed, temperature difference, etc.). The collected data is then sorted and preprocessed, the format is unified, and methods such as the 3σ criterion are used to remove outliers. When processing missing values, interpolation methods (such as linear interpolation or spline interpolation) or regression filling methods are used to fill in missing values ​​based on the time series relationship of the data or the correlation between meteorological parameters to ensure data integrity. Next, appropriate probability distribution models are selected for different meteorological parameters, such as Gaussian distribution for temperature and Weibull distribution for wind speed, and the model parameters are estimated. Each meteorological parameter can be described by an appropriate probability distribution model. A probability distribution network is then used to calculate the probability of each data point under different distribution models. These calculated probabilities serve as input features for subsequent clustering, providing contextual information about the probability distribution for each meteorological data point. Next, clustering algorithms (such as density-based clustering) are used to group similar data points together, generating multiple weather scenario categories such as sunny, cloudy, and rainy. Finally, statistical characteristics of each meteorological parameter category are calculated to describe the data, providing a basis for wind and solar forecasting and power generation scheduling.

[0016] Step S200, collect real-time wind and solar meteorological data, perform periodic wind and solar forecasts, and traverse the acquired wind and solar forecast data through the multiple weather scene categories to match the weather scenes, and obtain the target weather scene change sequence. Specifically, first, build a real-time wind and solar meteorological data acquisition system including a variety of sensors and data transmission equipment, install wind speed, light, temperature, humidity and other sensors at the power plant, and the sensors transmit the data to the data acquisition terminal for preliminary processing and storage. The data is collected and timestamp-marked at a determined frequency, and transmitted to the data center for storage via a wired or wireless communication network. Based on the change pattern analysis of the historical meteorological data of the target power plant, a mathematical or empirical model is established considering various factors, and a list of wind and solar forecast periods containing different time scales is generated accordingly. Then, corresponding prediction models such as autoregressive models and machine learning algorithms are used to generate wind and solar forecast data, and the model is continuously optimized and adjusted. Based on the wind and solar power forecast data, a meteorological characteristic curve is generated and smoothed, and a weather scene classifier consisting of a scene segmentation unit and a category matching unit is constructed. The scene segmentation unit cuts the curve according to the threshold table. After fusion analysis, the input is input into the category matching unit for comparison and matching with the weather scene category, thereby obtaining the target weather scene change sequence and providing weather scene information for power generation system scheduling.

[0017] In one possible implementation, real-time wind and solar power meteorological data is collected to perform periodic wind and solar power forecasts. The acquired wind and solar power forecast data is then traversed across multiple weather scene categories for weather scene matching to obtain a target weather scene change sequence. Step S200 further includes step S210, where weather change patterns are analyzed based on historical meteorological data from the target power plant, and a list of wind and solar power forecast cycles is generated based on the analysis results. Specifically, multiple years of historical meteorological data from the target power plant is collected, covering various parameters such as temperature, humidity, air pressure, wind speed, and sunlight. Organize the data and remove errors and missing values. For example, compare the data with the adjacent time periods or use the mean filling method to deal with missing values. Analyze the temporal trend of each meteorological parameter, such as seasonal changes in wind speed and diurnal changes in light intensity. Calculate relevant statistical indicators, such as mean, standard deviation, kurtosis, etc., to understand the data distribution characteristics. Study the correlation between different meteorological parameters, such as the correlation between wind speed and light intensity under certain weather conditions. According to the analysis results, determine different forecast periods. For short-term forecasts, take into account the real-time dispatch needs of power, such as 15 minutes to 1 hour, to quickly respond to the impact of changes in meteorological conditions on power generation. The medium-term forecast period may be several hours to one day, which is used for equipment start-up and shutdown and power trading planning. The long-term forecast period can be up to several days to one week, providing a reference for resource planning and operation strategy formulation. Taking various factors into consideration, generate a list of forecast periods with different time scales to meet the needs of different levels of the power generation system.

[0018] Step S220: Based on the wind / solar forecast cycle list, meteorological data is continuously collected to obtain real-time wind / solar forecast data. Specifically, the collection system is adjusted according to the wind / solar forecast cycle list. For short-term forecasts, the sensor collection frequency is increased, such as once every 5-10 minutes. For medium- and long-term forecasts, the frequency is reduced accordingly while ensuring data continuity. Collection equipment is rationally arranged to avoid interference from factors such as terrain. For example, monitoring points are set up in open areas and near equipment. Equipment is calibrated regularly to ensure measurement accuracy, such as monthly wind speed sensor calibration. The collection system continuously collects meteorological data according to a set period, including real-time wind speed, light intensity, temperature, etc., and transmits the data to the data processing center in a timely manner via wired or wireless communication. Encryption and verification measures are used to ensure data security and integrity. The data processing center performs real-time storage and preliminary quality control on the received data, such as checking the rationality of the data range and marking or correcting abnormal data.

[0019] Step S230, the real-time wind and solar meteorological data is subjected to time series analysis to generate the wind and solar forecast data. Specifically, the real-time data is subjected to a stationary test and seasonal adjustment. If the data is not stationary, differential processing is performed, and a suitable model is selected according to the data characteristics and the forecast period, and automatic adjustment is performed based on model evaluation indicators (such as AIC, BIC, forecast error, etc.). For example, for short-term forecasts (such as wind speed forecasts for the next hour), if the data exhibits strong autoregressive characteristics, the ARMA model is used for forecasting; for medium-term forecasts (such as meteorological data forecasts for the next 7 days), if the data exhibits seasonal fluctuations, the SARIMA model is used to capture seasonal and trend changes; for long-term forecasts or complex situations (such as long-term forecasts of wind and solar power generation), a more flexible machine learning model, such as LSTM (long short-term memory network), is used to handle long-term time dependencies and complex nonlinear relationships. In practice, model selection should be dynamically adjusted based on data characteristics. For example, for short-term forecasts, if the data is highly volatile and lacks a clear seasonal trend, the ARMA model is preferred. In cases of significant seasonality, the SARIMA model is used to capture seasonal variations. If the data exhibits strong nonlinear characteristics or long-term dependence, the LSTM model is selected to improve forecast accuracy. During each model training process, model evaluation indicators such as AIC and BIC are used, combined with forecast errors (such as root mean square error (RMSE) and mean absolute error (MAE)) to evaluate and select the model. If the model's forecast error is large, the system should analyze the cause and consider whether there are sudden meteorological changes, data anomalies, or external influencing factors. The model can be readjusted or more relevant factors can be added for training, such as sudden meteorological changes and historical events, to ensure the model's adaptability and forecast accuracy. Dynamic model selection and adaptive adjustment mechanisms based on data characteristics can effectively avoid the problem of model-data mismatch and improve the accuracy and stability of wind and solar forecasts.

[0020] In one possible implementation, real-time wind and solar meteorological data is collected, periodic wind and solar forecasts are performed, and the acquired wind and solar forecast data is traversed through the multiple weather scene categories to perform weather scene matching to obtain a target weather scene change sequence. Step S200 further includes step S240, generating multiple meteorological characteristic curves based on the wind and solar forecast data. Specifically, according to the wind and light forecast data, each meteorological parameter is processed separately. For example, for wind speed forecast data, a curve showing wind speed changes over time is drawn with time as the horizontal axis and wind speed value as the vertical axis. Corresponding curves are also drawn for parameters such as light intensity, temperature, and humidity. When drawing the curve, ensure that the time sequence of the data is correct and perform appropriate smoothing on the data to remove noise interference. Simple smoothing algorithms such as the moving average method can be used to make the curve more reflective of the basic change trend of the meteorological parameters. For example, for the wind speed curve, the 5-point moving average method is used, that is, the average wind speed of 5 consecutive time points is taken as the new wind speed value to draw the curve. This can reduce the impact of short-term fluctuations in wind speed data on the curve shape and more clearly show the overall change of wind speed. The multiple meteorological characteristic curves generated represent the changes in different meteorological parameters within the forecast time range, providing a basis for subsequent scenario analysis.

[0021] Step S250, constructing a weather scene classifier, the weather scene classifier includes a scene segmentation unit and a category matching unit, wherein the scene segmentation unit includes a meteorological feature threshold table of multiple meteorological features, the meteorological feature threshold table includes multiple level thresholds corresponding to the meteorological features, and the category matching unit is embedded with the multiple weather scene categories. Specifically, a meteorological feature threshold table is formulated for different meteorological features. For example, for wind speed, multiple level thresholds such as low wind speed threshold, medium wind speed threshold and high wind speed threshold are set to divide different wind speed intervals. The threshold is determined based on historical meteorological data and analysis of different weather scenes. By statistics of a large amount of historical data, the typical value range of wind speed under different weather conditions is understood, so as to determine a reasonable threshold. For temperature, a low temperature threshold, a normal temperature threshold and a high temperature threshold are set; for humidity, a low humidity threshold, a medium humidity threshold and a high humidity threshold are set. Each meteorological feature threshold table contains the corresponding meteorological characteristics. Multiple level thresholds of image features are used to divide meteorological data into different intervals for scene segmentation. Among the multiple embedded weather scene categories, each category has its own specific meteorological feature combination definition. The weather scene category is determined by cluster analysis of historical meteorological data and expert experience. For example, different weather scene categories such as sunny, cloudy, overcast, light rain, heavy rain, etc. Each category corresponds to a set of typical meteorological parameter ranges and characteristics. The role of the category matching unit is to match the processed meteorological data with these pre-defined weather scene categories to determine the most suitable weather scene.

[0022] Step S260, based on the scene segmentation unit, the multiple meteorological characteristic curves are scene-cut, and the cut multiple meteorological characteristic curves are fused and analyzed to generate a wind and light forecast curve. Specifically, the meteorological characteristic threshold table in the scene segmentation unit is used to perform scene cutting on the multiple meteorological characteristic curves. For each curve, it is divided into different segments according to the corresponding threshold value. For example, for the wind speed curve, when the wind speed value is higher than the high wind speed threshold, the time period is marked as a high wind speed scene segment; when the wind speed value is between the medium wind speed threshold and the high wind speed threshold, it is marked as a medium wind speed scene segment, etc. Similar scene cutting operations are also performed on the temperature curve, humidity curve, etc. Each meteorological characteristic curve is divided into multiple different scene segments according to the threshold value. These segments represent different meteorological state intervals. The cut multiple meteorological curves are The meteorological characteristic curve is fused and analyzed, and the combination of scene segments of various meteorological parameters at different time points is comprehensively considered, and their mutual relationship and coordinated changes are analyzed. For example, at a certain time point, the wind speed is in the high wind speed scene segment, the temperature is in the high temperature scene segment, and the humidity is in the low humidity scene segment. The comprehensive information is used to determine the overall meteorological scene characteristics at that time point. Through the comprehensive analysis of each time point, a wind and light forecast curve is generated. The curve not only contains the change information of each meteorological parameter, but also integrates the correlation and synergy between different meteorological parameters, and more comprehensively reflects the trend of meteorological scene changes in the future period.

[0023] Step S270: input the wind and light prediction curve into the category matching unit to perform weather scene matching, and generate a target weather scene change sequence according to the scene matching result. Specifically, the generated wind and solar forecast curve is input into the category matching unit. The category matching unit compares and matches the meteorological features in the wind and solar forecast curve with multiple embedded weather scene categories in detail, analyzes the combination of meteorological features at each time point in the curve, and finds the part that best matches the definition of each weather scene category. Based on the matching results, the weather scene category to which each time point belongs is determined. As time goes by, the weather scene matching situation at different time points is recorded to generate a target weather scene change sequence. The sequence shows the dynamic change process of the weather scene in the future, such as gradually changing from sunny to cloudy, and then to overcast, etc. The change sequence is very important for the scheduling decision of the water-wind-solar complementary power generation system, because under different weather scenarios, the power generation efficiency and capacity of hydropower, wind power and photovoltaics will be different. It is necessary to adjust the power generation strategy in time according to the changes in the weather scene to achieve the optimal power generation effect and energy utilization efficiency. The generation of the target weather scene change sequence from wind and solar forecast data is realized, providing key weather scene information support for the optimized operation of the water-wind-solar complementary power generation system.

[0024] Step S300 generates an initial power supply strategy for the hydropower-wind-solar hybrid power generation system based on the target weather scenario change sequence. The strategy includes multiple power generation ratio switching nodes. Specifically, multiple predicted weather scenarios are first extracted from the target weather scenario change sequence, and their durations and scenario switching nodes are determined. Meanwhile, transient scenarios with minimal impact on power generation scheduling are eliminated by setting time thresholds to obtain the weather scenarios to be adjusted. For hydropower generation, the generation characteristics of hydropower are analyzed under different weather scenarios, taking into account the impact of factors such as reservoir water level, water flow velocity, and water temperature on power generation. For wind power generation, the generation characteristics are analyzed based on wind speed, stability, and wind direction variations combined with wind turbine performance curves. For photovoltaic power generation, the generation characteristics under different weather scenarios are analyzed based on light intensity, temperature, and photovoltaic cell performance parameters. Based on the power generation characteristics analysis results, the optimal generation ratio of hydropower, wind power, and photovoltaic power is determined for each weather scenario to be adjusted, and the scenario switching node is used as the generation ratio switching node. Finally, the generation ratios and switching nodes of each scenario are combined to form an initial power supply strategy. This strategy specifies the generation ratios and switching times of each energy source within different time periods, providing preliminary guidance for power generation system operation and enabling subsequent optimization and adjustment.

[0025] In one possible implementation, based on the target weather scene change sequence, an initial power supply strategy for the hydro-wind-solar complementary power generation system is generated, the initial power supply strategy includes multiple power generation ratio switching nodes, and step S300 further includes step S310, based on the target weather scene change sequence, extracting multiple predicted weather scenes, as well as multiple scene durations and scene switching nodes corresponding to the multiple predicted weather scenes. Specifically, the target weather scene change sequence is carefully analyzed. The sequence is obtained by matching wind and solar forecast data with weather scene categories. Different predicted weather scenes, such as sunny, cloudy, overcast, etc., are identified based on the time sequence and changes in meteorological characteristics. By recording the start and end time of each scene, the corresponding scene duration is accurately calculated. The scene switching node is determined based on the transition moment of adjacent scenes. The node marks an important change point in weather conditions and plays a key indicative role in adjusting the power generation strategy.

[0026] Step S320, based on the duration of the multiple scenes, instantaneous scenes are eliminated, multiple weather scenes to be adjusted are generated, and the corresponding scene switching nodes are used as power generation ratio switching nodes. Specifically, the duration of the multiple scenes obtained is compared with the preset time threshold. If the scene duration is less than the threshold, such as some minor scene changes caused by short-term weather fluctuations, it is determined to be an instantaneous scene and eliminated to avoid frequent adjustment of the power generation ratio causing unnecessary interference to the system. The condition for eliminating instantaneous scenes depends only on the time threshold, and there is a risk of ignoring important information. For example, in areas where wind speed changes frequently, even if the wind speed scene duration is short, it may have a greater impact on wind power. It is necessary to introduce an impact assessment mechanism to consider the potential impact of short-term scenes on power generation. In order to avoid ignoring short-term scenes that have a greater impact on power generation efficiency or power, when eliminating instantaneous scenes, the actual impact of the scene can be evaluated in combination with the response characteristics of wind power generation or the changing trends of other key meteorological parameters (such as temperature, humidity, etc.). Specifically, for short-term scenarios, if they have a significant impact on power generation (such as power fluctuations caused by drastic changes in wind speed), the scenario will not be eliminated and will still be processed as a valid weather scenario. The power change threshold or the instantaneous fluctuation amplitude of meteorological parameters will be used to determine whether to retain the scenario. The conditions for eliminating instantaneous scenarios no longer rely solely on the time threshold, but should be combined with the degree of influence of meteorological parameters and power fluctuations. After screening, the remaining scenarios with actual impact are the weather scenarios to be adjusted, and the switching nodes between scenarios are determined as power generation ratio switching nodes, which are used to adjust the power generation ratio in time under different weather conditions. By introducing impact assessment mechanisms and power fluctuation judgments, it is possible to more effectively avoid eliminating short-term scenarios that may affect power generation, thereby ensuring the accuracy of power generation scheduling and the efficient operation of the system.

[0027] Step S330, for multiple weather scenarios to be adjusted, analyze the power generation characteristics, determine the optimal power generation ratio of hydropower, wind power and photovoltaic power according to the power generation characteristics analysis results, and generate the initial power supply strategy. Specifically, for different situations in the weather scenarios to be adjusted, analyze the hydropower generation characteristics. In scenarios with more precipitation, consider the increase in water flow and power generation power caused by the increase in reservoir water level; in drought scenarios, pay attention to the restrictions on power generation caused by the drop in water level, study the changes in turbine efficiency under different water temperature conditions, so as to fully understand the power generation capacity and stability of hydropower in various scenarios, and provide a basis for determining its proportion in the power supply strategy. Analyze the wind speed changes in different weather scenarios. In high wind scenarios, consider the rated power limit and safe operation requirements of the wind turbine; in low wind scenarios, evaluate the reduction in wind power output power, combined with wind The impact of directional changes on wind turbine power generation efficiency and the performance curve of the wind turbine are used to determine the power generation characteristics and output power range of wind power in different weather scenarios to be adjusted, providing a reference for the reasonable allocation of power generation ratio. For photovoltaic power generation, in addition to analyzing the negative impact of temperature increase on photovoltaic cell efficiency in sunny and strong light scenarios; in cloudy and weak light scenarios, the degree of power reduction is analyzed, and the impact of seasonal differences on photovoltaic output needs to be considered. For example, in summer, the light intensity is usually high, but the temperature increase may cause the efficiency of photovoltaic cells to decrease, especially when the temperature exceeds 35°C, the conversion efficiency of photovoltaic cells may decrease. It has obviously attenuated, so it is necessary to correct the attenuation of photovoltaic cell efficiency in high temperature environment. The high temperature effect is mainly reflected in the fact that the increase in the operating temperature of the photovoltaic module will lead to an increase in the internal resistance of the battery, thereby reducing the power generation efficiency. In winter, although the light is weak and the sunshine time is short, low temperature conditions are conducive to improving the efficiency of photovoltaic cells, and usually they can maintain a high conversion efficiency. Especially in sunny weather with strong sunlight in winter, the low temperature environment has a positive effect on photovoltaic power generation. Photovoltaic cells can maintain a high power generation efficiency at low temperatures. Therefore, in winter, the proportion of photovoltaic power generation should be appropriately increased, especially on sunny days. During the day and early morning hours, even when illumination is low, the efficiency of photovoltaic power generation is still high. In order to more accurately arrange the proportion of photovoltaic power generation, it is necessary to comprehensively consider factors such as seasonal differences, light intensity and temperature changes. Specifically, under high temperature conditions in summer, the proportion of photovoltaic power generation should be appropriately reduced to avoid the negative impact of high temperature on the efficiency of photovoltaic cells; while under low temperature conditions in winter, the proportion of photovoltaic power generation can be increased to utilize the potential for efficient power generation under low temperature conditions. Under weak light and cloudy conditions, the efficiency of photovoltaic power generation will drop significantly regardless of the season. At this time, priority should be given to relying on other power generation methods (such as hydropower or wind power) for supplementation.The comprehensive light intensity, temperature and performance parameters of photovoltaic cells are used to determine the power generation efficiency and power changes of photovoltaic power generation in various weather scenarios to be adjusted. On the other hand, adjusting the power generation ratio based solely on weather scenarios cannot fully adapt to the actual needs and load fluctuations of the power grid. The load fluctuations and real-time needs of the power grid are key factors affecting the feasibility and stability of the power supply strategy. During peak load periods, the power grid demand is usually large, and more stable and adjustable power sources (such as hydropower and wind power) are needed to ensure the balance of power supply; during low load periods, especially at night, the contribution of photovoltaic power generation is low, and the regulating role of wind power and hydropower becomes more important. In order to ensure power supply To achieve efficient and stable operation while meeting grid demand, the strategy requires dynamic adjustment of the generation ratio and comprehensive consideration of grid load fluctuations and real-time demand. Load forecasting models and real-time grid dispatch should be combined with weather scenario analysis results to optimize the power supply strategy. For example, when the grid load is high, hydropower and wind power should be prioritized, and their proportions should be appropriately increased to ensure stable grid operation. During low-load periods, photovoltaic power generation can be appropriately increased to reduce reliance on traditional energy sources. When the load is high or there are sudden demands, the generation ratio can be automatically adjusted through an intelligent dispatching system, dynamically adjusted according to the grid load, to ensure power supply stability and optimal utilization of power resources. Combining the actual grid demand and analysis of power generation characteristics, the generated initial power supply strategy will comprehensively consider factors such as weather scenarios, load fluctuations, and power demand. The optimal generation ratio of hydropower, wind power, and photovoltaic power will be determined for each weather scenario to be adjusted, ensuring efficient and stable operation of hydropower, wind power, and photovoltaic power generation under different weather conditions while meeting the real-time needs of the grid.

[0028] Step S400 analyzes the correlation between wind and solar environment indicators and hydropower generation terminals, extracts key influencing factors, and generates a multi-dimensional power generation evaluation network based on the key influencing factors. Specifically, first collect wind and solar environment indicator data related to hydropower generation, including wind speed, wind direction, air density, etc. for wind power generation, light intensity, solar radiation angle, temperature, etc. for photovoltaic power generation, as well as macro-meteorological data. Then, pre-process these data to remove outliers, unify time and format, and fill in missing values. Then, use correlation analysis methods to analyze their correlation with hydropower generation terminals, such as analyzing the correlation between water level, water flow velocity, etc. and power generation power in hydropower generation, the relationship between wind speed, wind direction, etc. and wind turbine output power in wind power generation, and the relationship between light intensity, etc. and power generation efficiency in photovoltaic power generation. Based on the results, determine the key influencing factors and eliminate redundant variables. Then, a multi-dimensional power generation evaluation network structure was designed, including an input layer that receives key factors, an intermediate layer that performs nonlinear transformation and feature extraction, and an output layer that outputs power generation evaluation indicators. The network was trained using historical data, and appropriate optimization algorithms, learning rates, and hyperparameters were selected. The network was evaluated and verified through a test set, and improvements were made to situations with large errors. Finally, the optimized network was integrated into the power generation system scheduling and management platform, which received real-time data to predict power generation output and provide decision support for optimized scheduling, power trading, and energy storage management.

[0029] In one possible implementation, Figure 2 As shown, the relationship between wind and solar environmental indicators and hydropower generation terminals is analyzed, key influencing factors are extracted, and a multidimensional power generation evaluation network is generated based on the key influencing factors. Step S400 further includes step S410. The hydropower-wind-solar complementary power generation system includes a hydropower module, a wind power module, and a photovoltaic power module. Specifically, the hydropower module converts the energy of water flow into electricity and typically involves equipment such as a reservoir and a turbine. The wind power module relies on wind power to rotate the wind turbine blades, which in turn drives the generator to generate electricity. The photovoltaic power module converts solar energy into electricity through photovoltaic cells. Understanding the basic operating principles and characteristics of each module lays the foundation for subsequent analysis of their relationship with wind and solar environmental indicators. For example, the power generation of hydropower is closely related to the reservoir water level and water flow velocity; the output power of wind power is affected by factors such as wind speed and direction; and the efficiency of photovoltaic power generation depends on conditions such as light intensity and temperature.

[0030] Step S420, collect wind and solar environment indicators, and analyze the correlation between the wind and solar environment indicators and the hydropower generation module, wind power generation module and photovoltaic power generation module through correlation analysis. Specifically, collect a variety of wind and solar environment indicators, including but not limited to wind speed, wind direction, light intensity, temperature, humidity, air pressure, etc. The indicators can be obtained through various sensors installed in the power plant, or historical data can be obtained from relevant agencies such as meteorological departments. For example, the wind speed sensor measures the real-time wind speed, and the light intensity sensor records the light conditions to ensure the accuracy and completeness of the data. The data is preliminarily sorted and screened to remove obviously abnormal data points. The correlation analysis method is used to analyze the correlation between wind and solar environment indicators and each power generation module. For the hydropower generation module, the correlation between water level, water flow velocity, etc. and power generation power is analyzed. By calculating statistical indicators such as correlation coefficients, we can determine whether the relationship between them is linear or nonlinear. For example, studies have found that the higher the water level in the reservoir, the greater the water head of the turbine, and the power generation power will usually increase accordingly. There may be a positive correlation between the two. For wind power generation modules, we analyze the relationship between wind speed, wind direction and wind turbine output power. Wind speed is the key factor. Generally, the greater the wind speed, the higher the wind turbine power generation power, but changes in wind direction will also affect the efficiency of the wind turbine. For photovoltaic power generation modules, we analyze the relationship between light intensity, temperature and photovoltaic power generation efficiency. The stronger the light intensity, the greater the photovoltaic power generation, but excessively high temperature may cause the efficiency of photovoltaic cells to decrease.

[0031] Step S430: Based on the results of the correlation analysis, identify the key influencing factors of each power generation module, eliminate redundant variables, and obtain the hydropower influencing factor, wind power operation factor, and photovoltaic influencing factor. Specifically, based on the results of the correlation analysis, identify the key factors that have a significant impact on each power generation module. In hydropower generation, water level and water flow velocity may be identified as key influencing factors; in wind power generation, wind speed and wind direction may be key factors; in photovoltaic power generation, light intensity and temperature may be key influencing factors. Key factors have a significant impact on the output power and efficiency of the power generation module. For example, when wind speed varies within a certain range, the power generated by the wind turbine will fluctuate significantly. Therefore, wind speed can be used as one of the key operating factors of wind power generation. To improve the efficiency and accuracy of subsequent analysis and models, it is necessary to eliminate redundant variables. Redundant variables are generally variables that are highly correlated with other key factors and have relatively small impact on the power generation module. Specifically, redundant variables can be determined by calculating correlation coefficients (such as the Pearson correlation coefficient) between variables to determine which variables have strong correlations. The commonly used standard is that if the correlation coefficient between two variables exceeds a certain threshold (such as 0.8 or 0.9), they can be considered highly correlated variables. For example, temperature and humidity may have a high correlation. If the Pearson correlation coefficient between humidity and temperature is higher than 0.8, humidity can be considered a redundant variable. It can also be judged by analyzing the degree of influence of each variable on key outputs such as power generation power and efficiency. For example, regression analysis, variance analysis or feature importance assessment methods (such as feature importance based on decision trees or random forests) can be used to measure the contribution of each factor to the power generation effect. If the contribution of a variable is less than a set threshold (such as 0.05), it can be considered to have little impact on the power generation output. The elimination of redundant variables avoids unnecessary data interference and improves the accuracy and efficiency of subsequent models. Through dimensionality reduction methods such as principal component analysis, a comprehensive analysis of related variables is conducted to extract the main independent variables as key influencing factors, which can reduce the dimension of the data, reduce the complexity of the model, and improve the predictive ability and generalization ability of the model.

[0032] Step S440, based on the hydropower influencing factors, wind power operating factors and photovoltaic influencing factors, fit and generate the multidimensional power generation evaluation network. Specifically, the data of the hydropower influencing factors, wind power operating factors and photovoltaic influencing factors are sorted out and used as input variables, and the corresponding historical power generation data, such as the actual power generation and power generation efficiency of hydropower, wind power and photovoltaic power generation, are collected as output variables. Select a suitable fitting method, such as a neural network, a multivariate regression model, etc. to construct a multidimensional power generation evaluation network. For example, when using a neural network model, determine the structure of the network, including the number and connection method of neurons in the input layer, hidden layer and output layer, train the model using the prepared data, and adjust the parameters of the model so that the model can accurately fit the relationship between the input factors and the output power generation data. During the training process, use cross-validation and other techniques to prevent overfitting and improve the generalization ability of the model. For example, divide the data into a training set, a validation set and a test set, train the model on the training set, adjust the model parameters through the validation set, and finally evaluate the performance of the model on the test set, and continuously optimize the structure and parameters of the model until the model achieves a better fitting effect and prediction accuracy. For example, adjust the model parameters. The learning rate, number of layers and other parameters of the entire neural network are adjusted to improve the accuracy and stability of the model. After training and optimization, a fitted multi-dimensional power generation evaluation network is obtained. The network can predict the power generation of each power generation module in the water-wind-solar complementary power generation system based on the input hydropower influencing factor, wind power operation factor and photovoltaic influencing factor. The generated network is verified, and new data is used for prediction, which is compared with the actual power generation data to evaluate the accuracy and reliability of the network. If there is a large deviation between the prediction result and the actual situation, the cause is further analyzed, and the model is adjusted and optimized until it meets the requirements of practical applications. Finally, a multi-dimensional power generation evaluation network that can accurately evaluate the operation of the water-wind-solar power generation system is obtained, providing strong support for the optimal scheduling and management of the power generation system.

[0033] Step S500: Based on the multi-dimensional power generation evaluation network, the initial power supply strategy is evaluated and optimized, a power supply optimization strategy is generated, and the power supply optimization strategy is used for dynamic power generation scheduling. Specifically, the multi-dimensional power generation evaluation network is first checked to ensure that it functions normally and its parameters are accurate. At the same time, the real-time data of wind and solar environment indicators and information related to the initial power supply strategy are prepared and pre-processed. Based on this network, multi-objective evaluation indicators such as power generation, power generation efficiency, and power generation cost are set to evaluate the initial power supply strategy. After the relevant data is input into the network, the evaluation results of hydropower, wind power, and photovoltaic power generation are obtained and in-depth analysis is carried out, such as comparing power generation with demand, exploring the reasons for low power generation efficiency, evaluating the power generation cost structure, and paying attention to stability and reliability. The optimization objectives and directions are then determined based on the evaluation results. The generation ratio adjustment unit is used to adjust the generation ratio and generate multiple alternative power supply strategies. A comprehensive objective function is fitted and substituted into the alternative strategies for evaluation. The optimal power supply strategy with the best overall performance is selected based on the function value to ensure stable and efficient operation of the hydropower, wind power, and photovoltaic hybrid power generation system. The multidimensional power generation evaluation network comprehensively analyzes the initial power supply strategy's evaluation results, including detailed data on hydropower, wind power, and photovoltaic power generation, to identify problems and deficiencies, such as an unreasonable generation ratio and high costs. Based on the optimization direction and objectives, the generation ratio adjustment unit is then used to generate multiple alternative power supply strategies. By fitting the comprehensive objective function, the optimal power supply strategy is selected based on the objective function value. In dynamic power generation scheduling, a real-time data monitoring system is established to collect and analyze wind, solar, meteorological, and equipment operation data to determine whether actual conditions align with expectations and promptly activate emergency response mechanisms in the event of anomalies. Based on the power supply optimization strategy, dispatch instructions are issued to each power generation module to adjust the operating parameters of the hydropower, wind, and photovoltaic modules. These parameters are continuously fine-tuned based on real-time data feedback to ensure efficient and stable operation of the power generation system, reduce costs and environmental impact, and improve power supply reliability and sustainability.

[0034] In one possible implementation, based on the multidimensional power generation evaluation network, the initial power supply strategy is evaluated and optimized, a power supply optimization strategy is generated, and the power supply optimization strategy is used for dynamic power generation scheduling. Step S500 further includes step S510, and the multidimensional power generation evaluation network includes a multidimensional power generation evaluation unit and a power generation ratio adjustment unit, wherein the multidimensional power generation evaluation unit includes a hydropower evaluation unit, a wind power evaluation unit and a photovoltaic power generation evaluation unit. Specifically, the multidimensional power generation evaluation network consists of a multidimensional power generation evaluation unit and a power generation ratio adjustment unit. The multidimensional power generation evaluation unit is further subdivided into a hydropower evaluation unit, a wind power evaluation unit and a photovoltaic power generation evaluation unit. The hydropower evaluation unit focuses on evaluating the relevant performance of the hydropower generation module. It considers the impact of factors such as reservoir water level, water flow velocity, and turbine efficiency on hydropower generation. The wind power evaluation unit analyzes factors related to wind power generation such as wind speed, wind direction, air density, and wind turbine performance for the wind power generation module. The photovoltaic power generation evaluation unit mainly focuses on the effects of light intensity, solar radiation angle, temperature, and photovoltaic cell characteristics on photovoltaic power generation. The power generation ratio adjustment unit is the key part for adjusting the power generation ratio in the initial power supply strategy based on the evaluation results.

[0035] Step S520: Perform a multi-objective evaluation of the initial power supply strategy based on the multi-dimensional power generation evaluation unit, including but not limited to evaluating power generation, power generation efficiency, and power generation cost, to obtain hydropower evaluation results, wind power evaluation results, and photovoltaic power generation evaluation results. Specifically, the multi-objective evaluation of the initial power supply strategy is performed based on the multi-dimensional power generation evaluation unit. The hydropower evaluation unit is responsible for evaluating the power generation of hydropower. By analyzing the water inflow of the reservoir, the operating status of the turbine, etc., it predicts the power generation of hydropower under the initial power supply strategy. At the same time, it evaluates the power generation efficiency of hydropower, taking into account factors such as the energy conversion efficiency of the turbine and the operating efficiency of the entire hydropower generation system. The cost of hydropower generation is also assessed, including equipment depreciation, water resource costs, and maintenance costs. The wind power assessment unit performs a similar assessment of wind power, evaluating wind power generation, efficiency, and costs based on wind speed distribution and turbine performance. Costs include turbine equipment investment, maintenance costs, and power transmission losses. The photovoltaic power generation assessment unit evaluates photovoltaic power generation, efficiency, and costs based on lighting conditions and photovoltaic cell performance. Costs cover the purchase and replacement costs and maintenance costs of photovoltaic cells. Through these assessments, hydropower, wind power, and photovoltaic power generation assessment results are obtained, providing a comprehensive understanding of the performance of each power generation module under the initial power supply strategy.

[0036] Step S530 , adjusting the power generation ratio of the initial power supply strategy using the power generation ratio adjustment unit according to the hydropower evaluation result, the wind power evaluation result, and the photovoltaic power generation evaluation result, to generate a power supply optimization strategy. Specifically, based on the evaluation results of hydropower, wind power, and photovoltaic power generation, the advantages and disadvantages of each power generation module are analyzed. If the hydropower evaluation results show that power generation has the potential to increase and the cost is relatively low in certain periods, and the proportion of hydropower generation in the initial power supply strategy does not fully utilize its advantages, then it is necessary to consider increasing the proportion of hydropower generation. Similar analysis is performed for wind power and photovoltaic power generation. Based on these analysis results, the power generation ratio adjustment unit adjusts the power generation ratio in the initial power supply strategy by adjusting the opening of the turbine, the blade angle of the wind turbine, and the working state of the photovoltaic cell to achieve the change in power generation ratio. For example, during periods of high wind speed and high power demand, the power generation power of the wind turbine is increased, and the power allocation of other power generation modules is correspondingly reduced. After multiple adjustments and simulations, a power supply optimization strategy is generated. Based on the comprehensive consideration of multiple objectives such as power generation, power generation efficiency, and power generation cost, this strategy enables the hydro-wind-solar hybrid power generation system to operate more efficiently and economically, improves overall energy utilization efficiency and power supply reliability, and utilizes various units of the multi-dimensional power generation evaluation network to achieve a comprehensive evaluation and optimization of the initial power supply strategy, providing strong support for the smooth operation of the hydro-wind-solar hybrid power generation system.

[0037] In one possible implementation, based on the hydropower evaluation results, wind power evaluation results and photovoltaic power generation evaluation results, the power generation ratio adjustment unit is used to adjust the power generation ratio of the initial power supply strategy to generate a power supply optimization strategy. Step S530 further includes step S531, determining the power supply adjustment direction based on the hydropower evaluation results, wind power evaluation results and photovoltaic power generation evaluation results. Specifically, carefully study the hydropower assessment results to see whether the hydropower generation meets the demand, whether the power generation efficiency is at a reasonable level, and the composition of the power generation cost. If the hydropower assessment results show that the power generation is low in certain periods, it may be due to insufficient reservoir water level or low turbine efficiency. At this time, it is necessary to consider increasing the regulation direction of hydropower generation, such as optimizing reservoir scheduling or maintaining turbine equipment. For power generation costs, if they are too high, you may need to find ways to reduce costs, such as adjusting water resource utilization or optimizing equipment operation mode. Analyze the wind power assessment results and pay attention to the relationship between wind speed and wind power generation. If wind resources are abundant in some areas but wind power generation does not meet expectations, it may be necessary to adjust the operating parameters or layout of the wind turbine to improve wind energy capture efficiency. This is the regulation direction of wind power. At the same time, consider wind power. Factors such as wind turbine maintenance costs and power transmission losses in the cost should be considered to determine whether optimization is needed in these aspects. The photovoltaic power generation evaluation results should be studied to observe the impact of light intensity and temperature on photovoltaic power generation. If the photovoltaic power generation is not ideal during periods of sufficient sunlight, it may be necessary to check the performance or cleanliness of the photovoltaic cells, as well as adjust the angle of the photovoltaic array. This is one of the adjustment directions of photovoltaic power generation. Factors such as the purchase and replacement costs of photovoltaic cells in the photovoltaic power generation cost should be analyzed to consider measures to reduce costs. The evaluation results of hydropower, wind power and photovoltaic power generation should be comprehensively compared to determine the overall power supply adjustment direction. For example, if it is found that the wind power and photovoltaic power generation are both low in a certain period of time, and hydropower has a certain adjustment potential, then the possible adjustment direction is to appropriately increase the proportion of hydropower generation, while looking for ways to improve the efficiency of wind power and photovoltaic power generation.

[0038] In step S532, the power generation ratio adjustment unit is used to adjust the initial power supply strategy according to the adjustment direction and the preset step size, thereby generating multiple alternative power supply strategies. Specifically, the power generation ratio adjustment unit is used to adjust the initial power supply strategy according to the determined adjustment direction and the preset step size. Assuming the preset step size is 5%, if the adjustment direction is to increase the hydropower generation ratio, then starting from the hydropower generation ratio in the initial power supply strategy, the ratio is increased by 5% each time, while the wind power and photovoltaic power generation ratios are correspondingly reduced (keeping the total power generation ratio at 100%), thereby generating a series of different power generation ratio combinations, i.e., multiple alternative power supply strategies. For each alternative power supply strategy, the specific values ​​of the hydropower, wind power, and photovoltaic power generation ratios are recorded. For example, one alternative power supply strategy may have a hydropower generation ratio of 40%, a wind power generation ratio of 30%, and a photovoltaic power generation ratio of 30%; another may have a hydropower generation ratio of 45%, a wind power generation ratio of 27%, and a photovoltaic power generation ratio of 28%, etc., thereby generating multiple alternative power supply strategies with different power generation ratio combinations, thereby providing a basis for subsequent screening and optimization.

[0039] Step S533, fit and generate a comprehensive objective function, screen and optimize the multiple alternative power supply strategies, and obtain the power supply optimization strategy. Specifically, fit and generate a comprehensive objective function, such as Minimize Z = x1·C+c2·E+c3·R (where C is the power generation cost, E is the environmental impact index, R is the power supply reliability, c1, c2, c3 are weight coefficients), and determine the weight coefficients according to the actual situation and decision preferences. For example, when focusing on economic benefits, the weight coefficient c1 of the power generation cost may be set to a larger value; in areas that emphasize environmental protection, the weight coefficient c2 of the environmental impact index may be increased. The generated multiple alternative power supply strategies are substituted into the comprehensive objective function for calculation. For each alternative strategy, according to its power generation ratio and related parameters, the corresponding power generation cost C, environmental impact index E and power supply reliability R values ​​are calculated, and then substituted into the function The objective function value Z is calculated and compared with the objective function values ​​of various alternative power supply strategies. The smaller the objective function value, the better the strategy performs in terms of comprehensive consideration of power generation cost, environmental impact and power supply reliability. The alternative power supply strategy with the smallest objective function value is selected as the power supply optimization strategy. The strategy achieves the optimal balance while meeting multiple objectives. For example, if an alternative power supply strategy has a lower power generation cost and smaller environmental impact while ensuring a certain power supply reliability, and its objective function value is the smallest, then it is determined as the final power supply optimization strategy, which is used to guide the actual operation of the hydro-wind-solar complementary power generation system to improve the overall performance and benefits of the system.

[0040] In one possible implementation, a comprehensive objective function is generated by fitting, and the multiple alternative power supply strategies are screened and optimized to obtain the power supply optimization strategy. Step S533 further includes step S5331, obtaining the optimization target, and the optimization target includes but is not limited to power generation cost, environmental impact index and power supply reliability. Specifically, various factors related to the water-wind-solar complementary power generation system are collected to clarify the optimization target. For power generation cost, various cost components including equipment purchase, installation, maintenance costs, and energy acquisition (such as the cost of utilizing water resources, the equipment operation and maintenance costs of wind and photovoltaic power generation, etc.) are analyzed in detail. For example, the depreciation cost of hydropower equipment, the regular maintenance cost of wind turbines, and the cleaning and replacement cost of photovoltaic panels are calculated, and these costs are comprehensively considered as evaluation factors for power generation cost. Environmental impact indicators are considered, and attention is paid to the various impacts on the environment during the power generation process. In terms of hydropower generation, the changes in the ecological environment caused by reservoir construction and the impact of water flow changes on aquatic organisms are evaluated; for wind power generation, consideration is given to The impact of wind turbine operation on bird migration, noise pollution, etc.; photovoltaic power generation focuses on environmental pollution during the production and waste disposal of photovoltaic panels. Through quantitative analysis of environmental factors, the specific content and evaluation methods of environmental impact indicators are determined, such as the formulation of corresponding ecological impact scoring standards, noise pollution levels, etc., and power supply reliability is analyzed. From the perspective of power supply stability and continuity, the power supply capacity of the power generation system under different weather conditions and load demands is evaluated, including the failure rate of power generation equipment, the availability of backup power supplies, and the reliability of the power transmission network. For example, the number and duration of power outages caused by equipment failures in the past period of time are counted, and the loss and stability of power transmission lines are analyzed to determine quantitative indicators of power supply reliability.

[0041] Step S5332, based on the optimization goal, fit and generate a comprehensive objective function: Minimize Z = c1·C+c2·E+c3·R; where C is the power generation cost, E is the environmental impact index, R is the power supply reliability, and c1, c2, and c3 are weight coefficients. Specifically, based on the determined optimization goal, construct a comprehensive objective function: Minimize Z = c1·C+c2·E+c3·R, with the goal of minimizing the total target value, that is, Minimize Z, where Z is the overall target value after comprehensively considering power generation cost, environmental impact index and power supply reliability. Power generation cost C is an important variable in the function. It is multiplied by the weight coefficient c1 to indicate the relative importance of power generation cost in the overall target. For example, if you want to pay more attention to reducing power generation cost in the optimization process, you can appropriately increase the value of c1. The environmental impact index E is multiplied by the weight coefficient c2, which reflects the weight of environmental factors in the optimization decision. For areas or projects with high environmental protection requirements, the value of c2 may be relatively large to emphasize the importance of reducing environmental impact. For example, if the power generation project is located near an ecological protection area, the environmental impact index will be given a higher weight in the function, prompting more consideration of environmental protection measures in the optimization strategy to reduce the adverse impact on the ecological environment. Power supply reliability R is multiplied by the weight coefficient c3, which reflects the status of power supply stability in the optimization target. In some scenarios with extremely high requirements for power supply stability, such as hospitals and important communication facilities, the value of c3 will be increased accordingly to ensure that power supply reliability is prioritized in the optimization process. When evaluating indicator E, the environmental impact of different power generation methods (hydropower, wind power, and photovoltaic power) is taken into account. The environmental impact calculation will be based on the scoring of various environmental factors. Specifically, the environmental impact of hydropower is mainly reflected in reservoir ecological damage, the impact of flow changes on downstream ecology, and noise pollution. Wind power mainly considers wind turbine noise, bird impacts, and landscape damage. The impact of photovoltaic power generation includes land use, light pollution, and recycling and waste issues. To quantify relevant factors, an ecological impact index can be used to score the impact of water flow changes on downstream ecology and noise pollution (the quantification method includes analyzing the rate of water flow change, noise intensity, and noise impact range), with 1 point indicating almost no impact and 5 points indicating a significant impact. The impact on the surrounding environment is scored by noise pollution (noise intensity and noise impact range), bird impact (bird mortality rate), and landscape damage (changes in tourist flow), with 1 point indicating low impact and 5 points indicating high impact. A similar scoring standard is used to consider factors such as land use (land occupation area and land change) and light pollution (reflected light intensity), with 1 point indicating low impact and 5 points indicating high impact.By reasonably adjusting the three weight coefficients c1, c2, and c3, a comprehensive objective function can be flexibly constructed according to actual needs and focus, achieving comprehensive optimization of power generation costs, environmental impacts, and power supply reliability, providing a scientific basis for decision-making on hydro-wind-solar complementary power generation systems, accurately obtaining optimization targets, and reasonably fitting and generating comprehensive objective functions, providing an effective mathematical model and decision-making tool for subsequent power supply strategy optimization and power generation system management.

[0042] The embodiment of the present application collects historical meteorological data of the target power plant and uses a probability distribution network to perform meteorological clustering to generate multiple weather scene categories; then, real-time wind and solar meteorological data are collected and periodic predictions are performed, and a target weather scene change sequence is obtained by matching weather scenes; then, an initial power supply strategy is generated based on the sequence, including multiple power generation ratio switching nodes; then, the correlation between wind and solar environmental indicators and power generation modules is analyzed, key influencing factors are extracted and a multi-dimensional power generation evaluation network is generated; finally, the evaluation network is used to optimize the initial power supply strategy, form a power supply optimization strategy, and implement dynamic power generation scheduling, thereby achieving the technical effect of dynamically optimizing the power supply ratio of the water-wind-solar complementary power generation system and improving the power generation efficiency and reliability of the system through the combination of real-time meteorological data collection and wind and solar prediction.

[0043] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An optimization scheduling method for a hydro-wind-solar complementary power generation system based on wind-solar forecasting, characterized in that: The method comprises: Collect historical meteorological data of the target power plant, use probability distribution network to perform meteorological clustering, and generate multiple weather scene categories; Collecting real-time wind and solar meteorological data, performing periodic wind and solar forecasting, and traversing the acquired wind and solar forecast data across the multiple weather scene categories to perform weather scene matching to obtain a target weather scene change sequence; Based on the target weather scenario change sequence, generating an initial power supply strategy for the hydro-wind-solar complementary power generation system, the initial power supply strategy including multiple power generation ratio switching nodes; Analyze the correlation between wind and solar environmental indicators and hydropower generation, extract key influencing factors, and generate a multi-dimensional power generation evaluation network based on the key influencing factors; Based on the multi-dimensional power generation evaluation network, the initial power supply strategy is evaluated and optimized to generate a power supply optimization strategy, and the power supply optimization strategy is used to perform dynamic power generation scheduling; Collect real-time wind and solar meteorological data and conduct periodic wind and solar forecasts, including: Analyze meteorological change patterns based on historical meteorological data of the target power plant and generate a list of wind and solar power forecast cycles based on the analysis results; Based on the wind and solar forecast period list, continuously collect meteorological data to obtain real-time wind and solar meteorological data; Performing time series analysis on the real-time wind and solar meteorological data to generate the wind and solar forecast data; The obtained wind and solar forecast data is traversed across the multiple weather scene categories to perform weather scene matching to obtain a target weather scene change sequence, including: generating a plurality of meteorological characteristic curves according to the wind and light forecast data; Constructing a weather scene classifier, the weather scene classifier comprising a scene segmentation unit and a category matching unit, wherein the scene segmentation unit comprises a meteorological feature threshold table for a plurality of meteorological features, the meteorological feature threshold table comprising a plurality of level thresholds corresponding to the meteorological features; and the category matching unit embeds the plurality of weather scene categories; Based on the scene segmentation unit, the multiple meteorological characteristic curves are subjected to scene segmentation, and the segmented multiple meteorological characteristic curves are subjected to fusion analysis to generate a wind and light prediction curve; The wind and light forecast curve is input into the category matching unit for weather scene matching, and a target weather scene change sequence is generated according to the scene matching result.

2. The method for optimizing and dispatching a hydro-wind-solar complementary power generation system based on wind-solar prediction according to claim 1, characterized in that: Based on the target weather scenario change sequence, an initial power supply strategy for the hydro-wind-solar hybrid power generation system is generated, including: Extracting a plurality of predicted weather scenes, and a plurality of scene durations and scene switching nodes corresponding to the plurality of predicted weather scenes based on the target weather scene change sequence; Eliminate instantaneous scenes according to the duration of the multiple scenes, generate multiple weather scenes to be adjusted, and use the corresponding scene switching nodes as power generation ratio switching nodes; For multiple weather scenarios to be adjusted, power generation characteristics analysis is performed, and based on the power generation characteristics analysis results, the optimal power generation ratio of hydropower, wind power and photovoltaic power is determined to generate the initial power supply strategy.

3. The method for optimizing and dispatching a water-wind-solar complementary power generation system based on wind-solar forecasting according to claim 1, characterized in that: Analyze the correlation between wind and solar environmental indicators and hydropower generation, extract key influencing factors, and generate a multi-dimensional power generation evaluation network based on the key influencing factors, including: The water-wind-solar complementary power generation system comprises a hydropower generation module, a wind power generation module and a photovoltaic power generation module; Collecting wind and solar environment indicators, and analyzing the correlation between the wind and solar environment indicators and the hydropower generation module, the wind power generation module, and the photovoltaic power generation module through correlation analysis; Based on the results of the correlation analysis, the key influencing factors of each power generation module are identified, redundant variables are eliminated, and the hydropower influencing factors, wind power operation factors, and photovoltaic influencing factors are obtained; The multi-dimensional power generation evaluation network is generated by fitting based on the hydropower impact factor, wind power operation factor and photovoltaic impact factor.

4. The method for optimizing and dispatching a water-wind-solar complementary power generation system based on wind-solar forecasting according to claim 1, characterized in that: Based on the multi-dimensional power generation evaluation network, the initial power supply strategy is evaluated and optimized, including: The multi-dimensional power generation evaluation network includes a multi-dimensional power generation evaluation unit and a power generation ratio adjustment unit, wherein the multi-dimensional power generation evaluation unit includes a hydropower evaluation unit, a wind power evaluation unit and a photovoltaic power generation evaluation unit; Performing a multi-objective evaluation of the initial power supply strategy based on the multi-dimensional power generation evaluation unit, including but not limited to evaluating power generation, power generation efficiency, and power generation cost, and obtaining a hydropower evaluation result, a wind power evaluation result, and a photovoltaic power generation evaluation result; According to the hydropower evaluation results, wind power evaluation results and photovoltaic power generation evaluation results, the power generation ratio adjustment unit is used to adjust the power generation ratio of the initial power supply strategy to generate a power supply optimization strategy.

5. The method for optimizing and dispatching a hydro-wind-solar complementary power generation system based on wind-solar prediction according to claim 4, characterized in that: The power generation ratio adjustment unit is used to adjust the power generation ratio of the initial power supply strategy to generate a power supply optimization strategy, including: Determining a power supply adjustment direction based on the hydropower assessment results, wind power assessment results, and photovoltaic power generation assessment results; Adopting the power generation ratio adjustment unit, adjusting the initial power supply strategy according to the adjustment direction and the preset step size, and generating a plurality of alternative power supply strategies; A comprehensive objective function is generated by fitting, and the multiple alternative power supply strategies are screened and optimized to obtain the power supply optimization strategy.

6. The method for optimizing and dispatching a hydro-wind-solar complementary power generation system based on wind-solar forecasting according to claim 5, characterized in that: Fitting generates a comprehensive objective function, including: Obtaining optimization objectives, wherein the optimization objectives include but are not limited to power generation cost, environmental impact indicators, and power supply reliability; According to the optimization goal, a comprehensive objective function is generated by fitting: Minimize Z=c1·C+c2·E+c3·R; Among them, C is the power generation cost, E is the environmental impact index, R is the power supply reliability, and c1, c2, and c3 are weight coefficients.

Images