Energy generation control method driven by multi-source power coupling model

Abstract

The present invention relates to the field of energy generation control technology, and specifically includes an energy generation control method driven by a multi-source power coupling model, including: constructing and fitting wind power and photovoltaic power models; predicting future weather and inputting the model to generate power generation fitting results; combining power supply demand, performing economic-stable balance optimization, and controlling energy generation according to the optimization results, so as to solve the technical problems of lack of comprehensive consideration of the power coupling characteristics of multiple energy generation, the risk of imbalance between power supply and demand, and the inability to ensure the overall stability and safety of the power system. The actual power generation situation is restored by constructing a power model, and at the same time, power generation fitting is performed in combination with real-time weather data to improve the accuracy and timeliness of the prediction. On this basis, economic-stable balance optimization is performed in combination with power supply demand, and energy generation is controlled according to the optimization results, so as to realize the refined and intelligent management of the power generation power of multiple energy sources, and effectively ensure the overall stability and reliability of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field related to energy generation control, and in particular to an energy generation control method driven by a multi-source power coupling model. Background Art

[0002] Wind and solar energy are two major renewable energy sources. Their generation methods have the inherent characteristics of intermittency and uncertainty, which makes it difficult to accurately predict their output power. This not only increases the difficulty of power system scheduling and operation, but may also lead to an imbalance between supply and demand, thereby affecting the overall stability and security of the power system. This is especially true in areas where wind and solar energy account for a high proportion, where this uncertainty and intermittency have a more significant impact on the power grid.

[0003] In addition, it is difficult to retain the autocorrelation of wind power output power and photovoltaic output power during the conventional control simulation generation process, which may cause a large deviation between the simulated time series and the actual series, further limiting the effective use of wind and solar energy by the power system. For example, when the actual values ​​of wind power and photovoltaic output power are higher than the predicted values, it may cause grid overload or energy waste; and when the actual values ​​are lower than the predicted values, it may lead to insufficient power supply and affect the normal electricity demand of users. This deviation between prediction and reality increases the complexity of power system scheduling and has a negative impact on the reliability of the power system.

[0004] In summary, the existing technology lacks comprehensive consideration of the power coupling characteristics of multiple energy sources, poses a risk of imbalance in power supply and demand, and cannot ensure the overall stability and security of the power system. Summary of the invention

[0005] This application provides an energy generation control method driven by a multi-source power coupling model, aiming to solve the technical problems in the prior art of lack of comprehensive consideration of the power coupling characteristics of multiple energy generation, the risk of imbalance in power supply and demand, and the inability to ensure the overall stability and safety of the power system.

[0006] In view of the above problems, the technical solution to implement this application is:

[0007] The present application provides a multi-source power coupling model driven energy generation control method, wherein the method comprises: establishing a sample database, wherein the sample database is constructed by reading power generation data after connecting to a power generation control system, wherein the sample data in the sample database comprises weather data and power generation data; constructing an initial wind power model and an initial photovoltaic power model respectively, and performing model fitting of the initial wind power model and the initial photovoltaic power model after data preprocessing of the data in the sample database, and establishing a wind power model and a photovoltaic power model; acquiring real-time weather data, and performing weather forecasting based on the real-time weather data, and establishing a weather forecast result, wherein In the method, the weather forecast result is provided with a forecast trust mark, and the forecast trust mark indicates the reliability of the weather forecast result; the weather forecast result is synchronized to the wind power model and the photovoltaic power model to perform power generation fitting, and establish a time-series power generation fitting result; the power supply demand is read, and the economic-stability balance optimization under the forecast trust mark is performed based on the power supply demand and the time-series power generation fitting result, and the optimization result is established; energy power generation control is performed according to the optimization result; the economic-stability balance optimization under the forecast trust mark is performed based on the power supply demand and the time-series power generation fitting result, and the optimization result is established, which also includes: establishing a power balance constraint according to the power supply demand, as follows: ;in, Representation in time power supply needs, Representation in time The power generation capacity, Characterizes the charging and discharging power of energy storage equipment, where: Negative values ​​represent charging, and positive values ​​represent discharging; according to the predicted confidence mark, the importance of the flexibility margin is corrected, and the importance correction result is established; according to the importance correction result and the power balance constraint, the economic-stability balance is optimized, and the optimization result is established; according to the importance correction result and the power balance constraint, the economic-stability balance is optimized, and the optimization result is established, which also includes: establishing an objective function, as follows: L= ; where L represents the objective function, The optimized total time window established based on power supply demand, Characterizes the cost of electricity generation, Characterizing the cost of power fluctuations, , Characterize the total number of samples, Characterization time point The power generation capacity, Characterizes the average power generation value of the total optimization time window, Characterize the flexibility margin, Characterize the prediction confidence mark, , Characterize historical actual power generation data, Characterize the fitted power generation based on weather forecasts, , are the weight coefficients of volatility cost and flexibility margin respectively.

[0008] In summary, one or more technical solutions provided in this application restore the actual power generation situation by constructing a power model. At the same time, power generation fitting is performed in combination with real-time weather data, which further improves the accuracy and timeliness of the prediction. On this basis, economic-stability balance optimization is performed in combination with power supply demand, and energy power generation is controlled according to the optimization results, thereby realizing refined and intelligent management of power generation power of multiple energy sources, and effectively ensuring the technical effect of the overall stability and reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A flow chart of an energy generation control method driven by a multi-source power coupling model is provided for this application;

[0010] Figure 2 A flow chart of fault cost compensation in an energy generation control method driven by a multi-source power coupling model is provided for this application. DETAILED DESCRIPTION Example

[0011] The present application is described in detail below with reference to the accompanying drawings. Figure 1 As shown, the present application provides a multi-source power coupling model driven energy generation control method, wherein the method includes:

[0012] S1: Establish a sample database, which is constructed by reading power generation data after connecting to the power generation control system. The sample data in the sample database includes weather data and power generation data; S2: Construct an initial wind power model and an initial photovoltaic power model respectively, and after preprocessing the data of the sample database, perform model fitting of the initial wind power model and the initial photovoltaic power model to establish a wind power model and a photovoltaic power model.

[0013] Specifically, the sample database refers to a collection of historical data stored in a storage medium, which is used to train and verify the power generation prediction model. It contains weather data and power generation data and is the basis for model learning. The wind power model is used to describe how wind speed and other meteorological conditions affect the power generation of wind turbines. The photovoltaic power model is used to describe how conditions such as light intensity and temperature affect the power generation of solar photovoltaic panels.

[0014] Execution steps: Connect the power generation control system (such as the monitoring system of wind farms and photovoltaic power stations) with the database system, which usually involves the setting of hardware interfaces and the integration of software to ensure that data can be transmitted from the power generation equipment to the database in real time; once the system connection is completed, you can start reading data from the power generation control system. These data include weather data such as wind speed, temperature, humidity, and light intensity, as well as power generation data such as the speed of wind turbines, power generation, and output power of photovoltaic panels.

[0015] The collected data is stored in the storage medium corresponding to the sample database. The sample database is the basis for subsequent model training and verification, and can efficiently store and retrieve large amounts of data. Using machine learning technology, initial wind power and photovoltaic power models are constructed. The initial wind power model and the initial photovoltaic power model preliminarily describe the relationship between power generation and weather conditions.

[0016] Before model fitting, the data in the sample database is preprocessed. This may include steps such as removing outliers, filling missing values, and data normalization to improve the quality of the data. For example, if there are abnormally high readings in the wind speed data, these may be caused by sensor errors and need to be identified and removed; then, the model is fitted, and further, the sample database is used to adjust the model parameters so that the model can best reflect the relationships and patterns in the data.

[0017] Using the preprocessed data, statistical analysis or machine learning methods (such as linear regression, neural network, decision tree, etc.) are used to fit wind power and photovoltaic power models. Through model fitting, wind power models and photovoltaic power models are established. Wind power models and photovoltaic power models will be used for subsequent power generation control and optimization. For example, the initial wind power model established with a linear regression model as the framework, however, the relationship between wind speed and power generation is not linear. A polynomial regression model is selected as the framework to better fit the relationship between wind speed and power generation, providing a model basis for subsequent analysis.

[0018] S3: Acquire real-time weather data, and make weather forecasts based on the real-time weather data to establish weather forecast results, wherein the weather forecast results carry a forecast trust mark, and the forecast trust mark represents the reliability of the weather forecast results; S4: Synchronize the weather forecast results to the wind power model and the photovoltaic power model to perform power generation fitting, and establish a time-series power generation fitting result.

[0019] Specifically, real-time weather data refers to the weather conditions at the current moment, such as temperature, humidity, wind speed, wind direction, light intensity, etc. These data are crucial for predicting future weather changes; the prediction confidence level indicator is used to indicate the reliability of weather forecast results. High confidence means that the forecast results are more reliable, while low confidence means that the forecast results are more uncertain; the time series power generation fitting result refers to the power generation prediction result generated based on the time series, which reflects the change of power generation over time.

[0020] Execution steps: Collect current real-time weather data through meteorological monitoring equipment such as weather stations, satellites, and radars. This data can be updated in real time by meteorological service providers or directly obtained through sensors installed on-site at wind farms and photovoltaic power stations; use meteorological principles and mathematical models to predict weather conditions in the future. Furthermore, use meteorological models such as the numerical weather prediction (NWP) model. The NWP model analyzes the collected real-time weather data to predict weather changes in the future.

[0021] Based on the output of the NWP model, weather forecast results for future weather conditions are established, including predicted values ​​of temperature, humidity, wind speed, wind direction, light intensity, etc. in the future time period; for each forecast result, a forecast confidence flag is calculated, and the forecast confidence flag is determined based on factors such as the accuracy of historical forecasts, the stability of the model, and the integrity of the data. For example, if the weather forecast model has a high accuracy rate in past forecasts, then the weight corresponding to the corresponding forecast confidence flag is close to 1.

[0022] The weather forecast results with prediction confidence marks are input into the wind power model and photovoltaic power model, and data integration tools and middleware are used for real-time data transmission and processing. The wind power model and photovoltaic power model are used in combination with weather forecast results to perform fitting calculations of power generation. Specifically, the predicted wind speed and predicted temperature are input into the wind power model to calculate the expected power generation.

[0023] According to the output of wind power model and photovoltaic power model, a series of power generation prediction results that change with time are generated, that is, the time-series power generation fitting results, which are used for bidding in the power market, dispatching of power grids and energy management. Through the above steps, using real-time weather data and prediction technology, combined with wind power and photovoltaic power models, the power generation is generated, taking into account the impact of weather changes and the reliability of the prediction results, providing support for energy generation control.

[0024] S5: Read the power supply demand, and optimize the economic-stability balance under the prediction trust mark based on the power supply demand and the time-series power generation fitting results, and establish the optimization result; S6: Control energy power generation according to the optimization result.

[0025] Specifically, power demand refers to the total demand for electricity in a power grid or power market within a specific period of time, usually in megawatts (MW); the prediction confidence level is used to indicate the reliability of the power generation prediction results. The weight corresponding to high confidence is close to 1, which means that the prediction result is relatively reliable; the weight corresponding to low confidence is close to 0, which means that the uncertainty of the prediction result is high.

[0026] Execution steps: Obtain power demand data for the upcoming time period from the grid operator or the power market, including the predicted power load curve, which shows the power demand in different time periods; Use the time-series power generation fitting results of wind power generation and photovoltaic power generation, and further, for each time-series power generation fitting result, consider its prediction confidence mark, which helps to give more reliable predictions higher weights in the optimization process, thereby improving the reliability of the overall power generation plan; Use optimization algorithms, such as linear programming, dynamic programming or genetic algorithms, to combine power demand and time-series power generation fitting results to find the optimal power generation plan. In this process, it is necessary to consider multiple factors such as power generation cost, grid stability, and prediction confidence. For example, if the prediction confidence of wind power is high and the prediction confidence of photovoltaic power is low, it will tend to increase wind power generation during the period when wind power prediction is reliable, and reduce photovoltaic power generation during the period when photovoltaic prediction is uncertain.

[0027] Output the optimization results, which include power generation plans, indicating how much electricity wind and photovoltaic power should generate in each time period to meet power supply needs while achieving the lowest cost and the most stable system; apply the optimization results to actual energy generation control, adjust the speed of wind turbines, control the tilt angle of photovoltaic panels, or schedule the charging and discharging of energy storage systems. Specifically, if the optimization results show that wind power generation needs to be increased within a certain period of time, the control center will send instructions to the wind farm to adjust the operating parameters of the wind turbines to increase power generation.

[0028] Through the above steps, it is possible to ensure that the power supply demand is met while minimizing the cost of power generation and maximizing the stability of the grid, taking into account the uncertainty of power generation forecasts. This approach improves the efficiency and reliability of energy generation, which is essential for the stable operation of the grid and the healthy development of the power market.

[0029] Furthermore, the economic-stability balance optimization under the prediction trust level mark is performed based on the power supply demand and the time-series power generation fitting results, and the optimization result is established. The method of this application also includes:

[0030] Establish power balance constraints according to power supply requirements, as follows: ;in, Representation in time power supply needs, Representation in time The power generation capacity, Characterizes the charging and discharging power of energy storage equipment, where: Negative values ​​represent charging, and positive values ​​represent discharging; the importance of the flexibility margin is corrected according to the predicted confidence mark, and the importance correction result is established; the economic-stability balance is optimized according to the importance correction result and the power balance constraint, and the optimization result is established.

[0031] Specifically, in order to ensure the stability and continuity of power supply, the power supply demand and the generated power must be balanced. The power balance constraint is established according to the power supply demand. ; Flexibility margin refers to the additional power generation or storage capacity reserved in the power system to cope with forecast errors and uncertainties; The importance correction result refers to adjusting the flexibility margin based on the forecast confidence, for example, increasing the margin when uncertainty is high; reducing the margin when the forecast is more accurate.

[0032] Execution steps: According to the power supply demand, establish the power balance constraint equation to ensure that at any given time point, the power supply demand Equal to power generation Plus the charging and discharging power of the energy storage device For example, if at a certain hour, the power demand is 100 MW and the total predicted power generation of wind power and photovoltaic power is 80 MW, then the energy storage equipment needs to provide an additional 20 MW to meet the power balance.

[0033] Charging and discharging power of energy storage equipment It can be a positive value (discharging) or a negative value (charging). This value needs to be determined based on the current grid status and the predicted power generation. For example, if photovoltaic power generation is predicted to decrease at night, the energy storage device may need to be charged during the day (negative value) so that it can be discharged at night (positive value) to meet the power supply demand.

[0034] The flexibility margin is adjusted according to the prediction confidence at each time point. If the prediction confidence is low, the flexibility margin needs to be increased to cope with greater uncertainty. For example, if the prediction confidence of wind power is low, more energy storage capacity needs to be reserved to cope with the power generation fluctuations caused by wind speed changes. The adjusted flexibility margin is applied to the power balance constraint, and the optimal power generation and energy storage scheduling plan is found by combining the revised power balance constraint and flexibility margin. For example, the optimization algorithm may consider multiple factors such as power generation cost, grid stability, prediction confidence, etc. to determine how much electricity should be generated at each time point and how to schedule the charging and discharging of energy storage equipment.

[0035] Output the optimization result, which is the power generation and energy storage scheduling plan, indicating the optimal operating status of various power generation resources and energy storage equipment in each time period. For example, the optimization result may show that wind power generation is increased during periods with higher wind speed forecasts, and the discharge of energy storage equipment is increased during periods with lower forecast confidence to ensure the stability of power supply. Through the above steps, it is ensured that while meeting the power supply demand, the power generation cost is minimized and the grid stability is maximized, while taking into account the uncertainty of power generation forecasts, the overall efficiency and reliability of the power system are improved.

[0036] Furthermore, the economic-stability balance optimization is performed according to the importance correction result and the power balance constraint, and the optimization result is established. The method of the present application also includes:

[0037] The objective function is established as follows: ; where L represents the objective function, The optimized total time window established based on power supply demand, Characterizes the cost of electricity generation, Characterizing the cost of power fluctuations, , Characterize the total number of samples, Characterization time point The power generation capacity, Characterizes the average power generation value of the total optimization time window, Characterize the flexibility margin, Characterize the prediction confidence mark, , Characterize the historical actual power generation data, Characterize the fitted power generation based on weather forecasts, , are the weight coefficients of volatility cost and flexibility margin respectively.

[0038] Specifically, the total optimization time window T refers to the time period corresponding to the economic-stability balance optimization, such as a working day, which is used to calculate the cost and stability in the entire time period; the power generation cost It refers to the cost of generating electricity at time point t, including fuel costs, etc. Power fluctuation cost Refers to the additional costs caused by fluctuations in generated power, usually related to power generation stability.

[0039] The total number of samples N refers to the number of samples used to calculate the mean and variance within the total optimization time window; flexibility margin Used to characterize the additional power generation or storage capacity reserved to deal with uncertainties and emergencies; prediction confidence mark A quantitative indicator that represents the difference between the predicted power generation and the historical actual power generation. The lower the confidence level (e.g. →0), the optimization model focuses on the flexibility requirement (i.e. The higher the historical actual power generation data Refers to the actual power generation data at time point t in the past; the fitted power generation is the power generation at time point t predicted by the weather forecast model; the weight coefficient , They represent the parameters used to adjust the importance of volatility cost and flexibility margin in the objective function, respectively.

[0040] Execution steps: define the total optimization time window and determine the total time period T corresponding to the economic-stable balance optimization. For example, if the power dispatch within a day is considered, T is 24 hours; construct the objective function , the objective function aims to minimize the generation cost and power volatility cost, while considering the flexibility margin and forecast confidence, where, , Characterization time point The power generation capacity, Characterize the mean power generation power of the total optimized time window; calculate the power volatility cost at each time point and the standard deviation of the power generation power based on historical data and forecast data.

[0041] According to the prediction confidence Adjusting flexibility margin The importance of prediction is that if the prediction confidence is high, the flexibility margin is reduced accordingly; if the prediction confidence is low, the flexibility margin is increased to cope with uncertainty; the weight coefficient is used , To balance the importance of power generation cost, power volatility cost and flexibility margin in the objective function; solve the objective function to find the optimal power generation plan and energy storage scheduling strategy; output the optimization result, which includes the optimal power generation power and energy storage scheduling strategy at each time point, ensuring power supply stability while meeting power supply demand.

[0042] Furthermore, the establishment of the wind power model and the photovoltaic power model, the method of the present application also includes:

[0043] The wind power model is constructed as follows: ;in, Characterizes the wind turbine power generation, Characterizes the wind speed at the start of the fan. Characterizes the wind speed at which the fan stops. Characterizes the rated wind speed, Characterizes the rated power of the fan, is the adjustment factor.

[0044] Specifically, the wind turbine power generation Refers to the actual output power of a wind turbine at a specific wind speed; wind speed at which the wind turbine starts It refers to the minimum wind speed required for the wind turbine to start generating electricity; the wind speed at which the wind turbine stops It refers to the maximum wind speed at which the wind turbine needs to stop operating in order to ensure safety; rated wind speed It refers to the wind speed when the wind turbine reaches its rated power; the rated power of the wind turbine It refers to the maximum electric power that a wind turbine can output at rated wind speed; the adjustment factor k is used to adjust the proportional coefficient of the output power of the wind turbine at different wind speeds.

[0045] Implementation steps: Determine the technical parameters of each wind turbine, including the starting wind speed , Shutdown wind speed , Rated wind speed and rated power ; Construct a wind power model, which describes the relationship between wind speed and wind turbine power generation, including: ; ; ,The wind power model considers that the wind turbine does not generate electricity when the wind speed is lower than the starting wind speed and higher than the shutdown wind speed, the power generated increases with the wind speed when the wind speed is between the starting wind speed and the rated wind speed, and the rated power is reached when the wind speed reaches the rated wind speed.

[0046] The adjustment factor k is an empirical parameter determined based on actual wind turbine performance data, which affects the growth rate of wind turbine power generation between the starting wind speed and the rated wind speed. After the wind power model is built, the wind turbine power generation is calculated using real-time or predicted wind speed data. This can be achieved through programming, such as using Python language and Pandas library to process data and perform calculations. Specifically, historical wind speeds and corresponding actual power generation data are used to verify the accuracy of the model. In this way, the wind power model can predict the wind turbine power generation based on the wind speed, thereby performing effective power scheduling and optimization.

[0047] Furthermore, the present application method also includes:

[0048] The photovoltaic power model is constructed as follows: ;in, Characterize photovoltaic power generation, Characterize the conversion efficiency of photovoltaic panels, Characterizes the effective light-receiving area of ​​the photovoltaic panel, Characterizes light intensity, Characterize the temperature coefficient, Characterizes the operating temperature of photovoltaic panels, Characterizes the reference calibration temperature of photovoltaic panels, is the dynamic adjustment factor of ambient temperature, Characterizes the ambient temperature.

[0049] Specifically, photovoltaic power generation Refers to the actual electrical power output of photovoltaic cells under specific conditions; the conversion efficiency of photovoltaic panels Refers to the efficiency of photovoltaic cells in converting sunlight into electrical energy; the effective light-receiving area of ​​the photovoltaic panel Refers to the area of ​​​​the photovoltaic panel that receives sunlight; light intensity Refers to the intensity of sunlight received by the photovoltaic panel, usually in W / m²; temperature coefficient Refers to the rate at which the efficiency of photovoltaic panels changes with temperature; the operating temperature of photovoltaic panels Refers to the actual temperature of the photovoltaic panel when it is working; the reference calibration temperature of the photovoltaic panel Refers to the calibrated temperature of the photovoltaic panel under standard test conditions, usually 25°C; the ambient temperature dynamic adjustment factor It refers to the coefficient that adjusts the photovoltaic power generation efficiency according to the change of ambient temperature; ambient temperature Refers to the temperature of the environment where the photovoltaic panels are located.

[0050] Implementation steps: Determine the technical parameters of each photovoltaic panel, including conversion efficiency , Effective light receiving area , Temperature coefficient and reference calibration temperature ; The photovoltaic power model describes the influence of light intensity, temperature and photovoltaic panel characteristic parameters on photovoltaic power generation, thereby calculating the photovoltaic power generation , comprehensively considering the direct impact of light intensity on power generation, the impact of temperature on power generation efficiency, and the dynamic adjustment of ambient temperature on power generation efficiency; further, the ambient temperature dynamic adjustment factor It is an empirical parameter that affects the power generation efficiency of photovoltaic panels under different ambient temperatures. The power generation power of photovoltaic panels is calculated using real-time or predicted light intensity and temperature data. This can be achieved through programming, such as using Python language and Pandas library to process data and perform real-time calculations.

[0051] Use historical light intensity, temperature and corresponding actual power generation data to verify the accuracy of the model. If the power generation predicted by the model is different from the actual power generation, the ambient temperature dynamic adjustment factor needs to be adjusted. In this way, the power generation of photovoltaic panels is predicted based on light intensity and temperature, allowing for efficient power scheduling and optimization.

[0052] Furthermore, the method of establishing the optimization result also includes:

[0053] The failure rate model is established as follows: ;in, Characterize the probability of failure, Characterizes the baseline failure rate under rated load, Characterizes the load sensitivity factor, Characterization time The actual power generated by the equipment, is the rated power of the equipment, Characterize the wear coefficient of equipment, Characterization time The power generation capacity of the equipment, is the complex nonlinear exponent of power, Characterize the thermal stress failure rate; perform failure cost compensation under economic-stability balance optimization based on the failure probability model, and reconstruct the optimization results according to the failure cost compensation results.

[0054] Specifically, the failure probability Refers to the probability of equipment failure at a specific point in time; baseline failure rate Refers to the inherent failure rate of the equipment under rated load; load sensitivity coefficient It refers to the sensitivity of the equipment failure rate to load changes; the actual power generation of the equipment Refers to the actual power output of the equipment at a specific point in time; rated power generation It refers to the maximum output power that the equipment can achieve in design; Equipment wear coefficient It refers to the coefficient of increased failure rate caused by gradual wear of equipment due to long-term use; power complex nonlinear index Index used to describe the nonlinear relationship between a device's power and its failure rate; thermal stress failure rate Refers to the failure rate caused by thermal stress due to temperature changes.

[0055] Execution steps: Establish a failure rate model, which describes the relationship between equipment failure rate and load, equipment wear, historical power output and thermal stress, and failure probability , considering the change in failure rate of the equipment under different loads, the impact of equipment wear on the failure rate, the cumulative impact of historical power output on the failure rate, and the impact of thermal stress on the failure rate; the baseline failure rate , load sensitivity factor , Equipment wear coefficient , power complex nonlinear exponential , thermal stress failure rate It is necessary to estimate the thermal stress failure rate through historical failure data and equipment performance data. ( ), specifically, No unit, represents a probability value, It refers to the operating temperature of the equipment at time t (unit: °C); Characterization reference temperature (unit: °C, the calibration temperature for normal operation of the equipment); Characterize the critical temperature (unit: °C, if the critical temperature is exceeded, the equipment will enter a high-risk state); substitute the time Actual power generation of the equipment , calculation time The probability of failure .

[0056] When optimizing for economic-stability balance, the impact of failure probability on power generation plan is considered, and the power generation plan is adjusted to reduce power generation during high failure probability periods, or to increase spare capacity to cope with possible equipment failures; the optimization results are reconstructed to obtain an optimized power generation plan to ensure that the cost and risk caused by equipment failure are minimized while meeting power supply demand. In this way, the failure probability of wind turbines under specific conditions is predicted, and the power generation plan is adjusted accordingly to reduce the risk of failure.

[0057] Furthermore, if Figure 2 As shown, before the failure cost compensation under the economic-stability balance optimization based on the failure probability model is performed, the method of the present application also includes:

[0058] Perform equipment health self-inspection on wind power equipment and photovoltaic equipment, and establish equipment health self-inspection results; initialize the fault probability model based on the equipment health self-inspection results, and perform fault cost compensation under economic-stability balance optimization based on the initialized fault probability model.

[0059] Specifically, equipment health self-check refers to regular or real-time inspections of wind power and photovoltaic equipment to assess their operating status and potential failure risks; equipment health self-check results refer to detailed reports on the equipment status obtained after self-check, including whether the equipment is operating normally and whether there are signs of failure; model initialization refers to setting the initial parameters of the model based on the latest data and conditions before starting optimization.

[0060] Implementation steps: Perform health checks on wind power and photovoltaic equipment regularly or under specific conditions. This may include physical inspections, performance tests, and sensor data analysis. For example, for wind turbines, the inspection may include blade damage inspections, electrical tests of the generator, and vibration analysis; establish equipment health self-inspection results, and further organize the data collected during the self-inspection process into a report that clearly indicates the health status of the equipment and any signs of failure that need attention. For example, the self-inspection results may show that the bearing temperature of a wind turbine is abnormal, indicating that there may be a risk of failure.

[0061] The equipment health self-check results are used to adjust and initialize the parameters of the fault probability model to ensure that the model can reflect the actual operating status of the equipment. During the optimization process, the impact of equipment failure probability on the power generation plan is considered, and possible failure costs are compensated. For example, if the failure probability model predicts that a certain wind power equipment has a higher probability of failure under high wind speed conditions, the optimization algorithm may reduce the power generation of the equipment, or increase the backup capacity to reduce the risk of failure.

[0062] According to the updated fault probability model, the economic-stability balance optimization is re-performed to determine the optimal power generation plan. Specifically, the fault probability model shows that the failure risk of a photovoltaic power station increases under high temperature conditions. The optimization result may suggest reducing the power generation during high temperature periods to avoid potential failures. In the above steps, after the model is initialized, the economic-stability balance optimization is performed using the updated fault probability model. The optimization result may suggest reducing the power generation of the wind turbine when the wind speed is high, or arranging more maintenance work on days with higher wind speed forecasts to reduce the risk of failures, thereby ensuring the stability of power supply and reducing the impact of failures.

[0063] Furthermore, the present application method also includes:

[0064] Supply defect prediction is performed based on power supply demand and time-series power generation fitting results, and supply defect prediction results are established; energy storage demand is established based on the supply defect prediction results, and energy storage strategy is matched based on the energy storage demand, and energy generation control is performed with the energy storage strategy.

[0065] Specifically, supply defect prediction refers to the prediction of the gap between power supply capacity and power supply demand due to various reasons (such as equipment failure, power generation prediction error, etc.) within a specific time period; the supply defect prediction result is based on the power supply demand and the time-series power generation fitting result, and predicts the specific time period and degree of insufficient power supply capacity; the energy storage demand is to determine how much energy storage resources are needed to make up for the shortfall in power supply capacity based on the supply defect prediction result; the energy storage strategy is used to use energy storage resources to meet power supply demand within a specific time period, including the timing and power of charging and discharging.

[0066] Implementation steps: Collect and analyze power demand data and the time-series power generation fitting results obtained through wind power and photovoltaic power models; use statistical analysis or machine learning methods to compare the predicted power generation and power demand, and identify time periods when power supply capacity may be insufficient. For example, use time series analysis tools (such as ARIMA models) to predict power demand in the future and compare it with the power generation fitting results to identify potential power supply gaps.

[0067] Based on the supply defect prediction, a detailed prediction result is established, including the specific time point and size of the power supply gap. For example, the prediction result shows that between 2:00 p.m. and 4:00 p.m. on the Zth day, there will be a 100-megawatt power supply gap due to the decrease in photovoltaic power generation. Based on the supply defect prediction results, the amount of energy storage resources needed to fill these power supply gaps is calculated. Combined with the above example, the prediction result shows that an additional 100 megawatts of electricity is needed to meet demand. Correspondingly, it is necessary to determine the energy storage system needs to provide 100 megawatts of electricity.

[0068] According to the energy storage demand, formulate a specific energy storage strategy, including when to charge, when to discharge, and the power of discharge. Combined with the above example, if it is predicted that there will be a power supply gap in the afternoon of the Zth day, the energy storage strategy may include charging during the peak period of photovoltaic power generation in the morning of the Zth day, and discharging in the afternoon of the Zth day to meet the demand; apply the energy storage strategy to the actual energy generation control, adjust the operating parameters of wind power, photovoltaic and energy storage equipment, and use the energy management system (EMS) software to automatically control the charging and discharging operations of the energy storage equipment to ensure that the required power is provided during the predicted power supply gap period. In this way, energy storage resources are effectively used to make up for the lack of power supply capacity and ensure the stable operation of the microgrid.

[0069] In summary, the beneficial effects of the embodiments of the present application are:

[0070] 1. By establishing wind power model and photovoltaic power model, and combining real-time weather data for power generation fitting, the accuracy of power generation forecast is improved, adapting to different environmental conditions, including different weather and temperature changes, and improving the adaptability and flexibility of the power system.

[0071] 2. By optimizing the economic-stability balance under the prediction trust level, the stability of the power grid is improved, the supply defect prediction and energy storage demand are matched, the impact of the volatility of renewable energy generation is reduced, the energy utilization efficiency is improved, and the optimal match between power supply demand and power generation is achieved.

[0072] 3. By considering the prediction trust, the reliability of the control strategy is enhanced, and the failure rate model and equipment health self-check are established to reduce the failure rate of equipment and the risk of grid instability caused by prediction errors.

[0073] 4. Due to the adoption of the failure rate model, as follows: ;in, Characterize the probability of failure, Characterizes the baseline failure rate under rated load, Characterizes the load sensitivity factor, Characterization time The actual power generated by the equipment, is the rated power of the equipment, Characterize the wear coefficient of equipment, Characterization time The power generation capacity of the equipment, is the complex nonlinear exponent of power, Characterize the thermal stress failure rate; perform failure cost compensation under economic-stability balance optimization based on the failure probability model, and reconstruct the optimization result according to the failure cost compensation result. In this way, the failure probability of wind turbines under specific conditions is predicted, and the power generation plan is adjusted accordingly to reduce the risk of failure.

[0074] In summary, any step can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor, and no unnecessary restrictions are made here.

[0075] Furthermore, the above-mentioned technical scheme only reflects the preferred technical scheme of the technical scheme of the embodiment of the present application. Some changes that may be made to some parts thereof by technicians in this technical field all reflect the novel principles of the embodiment of the present application. Obviously, technicians in this field can make various changes and modifications to the present application without departing from the scope of the present application.

Claims

1. An energy generation control method driven by a multi-source power coupling model, characterized in that: The method comprises: Establishing a sample database, wherein the sample database is constructed by reading power generation data after connecting to a power generation control system, and the sample data in the sample database includes weather data and power generation data; Respectively constructing an initial wind power model and an initial photovoltaic power model, and after preprocessing the data of the sample database, performing model fitting of the initial wind power model and the initial photovoltaic power model to establish a wind power model and a photovoltaic power model; Acquire real-time weather data, and perform weather forecast based on the real-time weather data to establish a weather forecast result, wherein the weather forecast result carries a forecast confidence mark, and the forecast confidence mark represents the reliability of the weather forecast result; Synchronize the weather forecast results to the wind power model and the photovoltaic power model to perform power generation fitting, and establish a time-series power generation fitting result; Read the power supply demand, and conduct economic-stability balance optimization under the prediction trust mark based on the power supply demand and time-series power generation fitting results, and establish the optimization results; Control energy generation based on the optimization results; Based on the power supply demand and the time-series power generation fitting results, the economic-stability balance optimization under the prediction trust mark is carried out, and the optimization results are established, which also includes: Establish power balance constraints according to power supply requirements, as follows: ； in, Representation in time power supply needs, Representation in time The power generation capacity, Characterizes the charging and discharging power of energy storage equipment, where: Negative values ​​represent charging, positive values ​​represent discharging; According to the prediction confidence mark, the importance of the flexibility margin is corrected and the importance correction result is established; Perform economic-stability balance optimization based on importance correction results and power balance constraints, and establish optimization results; According to the importance correction results and power balance constraints, the economic-stability balance is optimized and the optimization results are established, which also include: The objective function is established as follows: L= ； Among them, L represents the objective function, The optimized total time window established based on power supply demand, Characterizes the cost of electricity generation, Characterizing the cost of power fluctuations, , Characterize the total number of samples, Characterization time point The power generation capacity, Characterizes the average power generation value of the total optimization time window, Characterize the flexibility margin, Characterize the prediction confidence mark, , Characterize historical actual power generation data, Characterize the fitted power generation based on weather forecasts, , are the weight coefficients of volatility cost and flexibility margin respectively.

2. The energy generation control method driven by a multi-source power coupling model according to claim 1, characterized in that: The establishing of the wind power model and the photovoltaic power model also includes: The wind power model is constructed as follows: ； in, Characterizes the wind turbine power generation, Characterizes the wind speed at the start of the fan. Characterizes the wind speed at which the fan stops. Characterizes the rated wind speed, Characterizes the rated power of the fan, is the adjustment factor.

3. The energy generation control method driven by a multi-source power coupling model as claimed in claim 2, characterized in that: The method further comprises: The photovoltaic power model is constructed as follows: ； in, Characterize photovoltaic power generation, Characterize the conversion efficiency of photovoltaic panels, Characterizes the effective light-receiving area of ​​the photovoltaic panel, Characterizes light intensity, Characterize the temperature coefficient, Characterizes the operating temperature of photovoltaic panels, Characterizes the reference calibration temperature of photovoltaic panels, is the dynamic adjustment factor of ambient temperature, Characterizes the ambient temperature.

4. The energy generation control method driven by a multi-source power coupling model as claimed in claim 1, characterized in that: The establishing of the optimization result also includes: The failure rate model is established as follows: ； in, Characterize the probability of failure, Characterizes the baseline failure rate under rated load, Characterizes the load sensitivity factor, Characterization time The actual power generated by the equipment, is the rated power of the equipment, Characterize the wear coefficient of equipment, Characterization time The power generation capacity of the equipment, is the complex nonlinear exponent of power, Characterize thermal stress failure rate; Based on the failure probability model, failure cost compensation is carried out under economic-stability balance optimization, and the optimization results are reconstructed according to the failure cost compensation results.

5. The energy generation control method driven by a multi-source power coupling model as claimed in claim 4, characterized in that: Before the failure cost compensation under the economic-stability balance optimization based on the failure probability model is performed, it also includes: Conduct equipment health self-inspection on wind power equipment and photovoltaic equipment, and establish equipment health self-inspection results; The fault probability model is initialized based on the equipment health self-check results, and the fault cost compensation is performed under the economic-stability balance optimization according to the initialized fault probability model.

6. The energy generation control method driven by a multi-source power coupling model according to claim 1, characterized in that: The method further comprises: Supply defect prediction is performed based on power supply demand and time-series power generation fitting results, and supply defect prediction results are established; Establish energy storage demand based on supply deficiency prediction results, match energy storage strategy based on energy storage demand, and use energy storage strategy to control energy generation.

Images