A power supply and demand prediction system and method based on multi-source meteorological wind power prediction
By adding power prediction, evaluation and scheduling modules in the power supply and demand prediction system, and using multi-source meteorological data to predict wind power farm wind power and power supply and demand scheduling, the problem that existing systems cannot effectively predict wind power in the wind power farm is solved, and the applicability and prediction accuracy of the system are improved.
Patent Information
- Application Number
- CN202411690829.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-25
- Publication Date
- 2025-04-04
AI Technical Summary
The existing power supply and demand prediction system based on multi-source meteorological wind power prediction cannot effectively predict the wind power of the wind farm, resulting in the inability to scientifically evaluate the wind power and power supply and demand situation of the wind farm, affecting the applicability of the power supply and demand prediction system.
By adding power prediction modules, evaluation modules and scheduling modules, a power supply and demand prediction system based on multi-source meteorological wind power prediction is built. The power prediction module uses historical data to establish statistical or learning models for prediction, the evaluation module evaluates the prediction results through root mean square error and average absolute error, and the scheduling module performs reasonable power demand scheduling.
It realizes scientific evaluation of wind power in the wind farm and accurate judgment of power supply and demand conditions, and improves the applicability and prediction accuracy of the power supply and demand prediction system.
Smart Images
Figure CN119765267A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of wind power prediction, and in particular to an electricity supply and demand prediction system and method based on multi-source meteorological wind power prediction. Background Art
[0002] With the continuous advancement of "dual carbon" development and energy transformation, the traditional power system is evolving into a new power system that is clean, low-carbon, safe, controllable, flexible, efficient, intelligent and friendly. The core feature of the new power system is the widespread access of a high proportion of new energy. New energy occupies a dominant position and accelerates the replacement of fossil energy to become the main form of energy. Wind, photovoltaic and other energy sources have become an important part of the new power system. The new power system with new energy as the main body is a complex nonlinear dynamic system with real-time balance. The randomness and intermittency of new energy wind power have brought new challenges to power balance. Improving the prediction ability of new energy wind power is an effective means to meet the challenges.
[0003] The existing electricity supply and demand forecasting system based on multi-source meteorological wind power forecasting has difficulties in predicting the wind power of wind farms. There is no error calculation, and it is impossible to determine the physical solution with the best forecasting effect, and it is impossible to scientifically evaluate the wind power of wind farms. The wind power of wind farms and the supply and demand forecasting system of electricity supply and demand are unclear, and reasonable scheduling cannot be carried out, which affects the applicability of the electricity supply and demand forecasting system.
[0004] In view of the above, this case proposes a power supply and demand forecasting system and method based on multi-source meteorological wind power forecasting, which solves the above technical problems by adding a power prediction module, an evaluation module and a scheduling module. Summary of the invention
[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a power supply and demand forecasting system and method based on multi-source meteorological wind power forecasting, which solves the above technical problems by adding a power forecasting module, an evaluation module and a scheduling module.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] An electricity supply and demand prediction system based on multi-source meteorological wind power prediction, comprising a system control platform, a database, a collection module, a transmission module, a statistical module, a power prediction module, an evaluation module and a scheduling module;
[0008] The system control platform is used to regulate the operation of the power supply and demand prediction system based on multi-source meteorological wind power prediction;
[0009] The database stores power prediction results calculated by the prediction model in real time;
[0010] The acquisition module is used to collect and store real-time data required by the system;
[0011] The transmission module is used for generating and reporting forecast files and downloading weather forecast data;
[0012] The statistical module is used to calculate the prediction accuracy of the statistical model and the reporting success rate of the files that need to be uploaded in the system, and to periodically read the power prediction data and real-time power data stored in the database;
[0013] The power prediction module is used to predict the power of the wind farm;
[0014] The evaluation module is used to evaluate and compare the predicted power of the wind farm;
[0015] The dispatching module is used to dispatch the power demand of the wind farm.
[0016] Furthermore, the power prediction module is implemented by establishing a wind power prediction model, which is mainly based on the historical data of the wind farm. The collected historical data is used as the input of the model, and prediction is performed by establishing a statistical model or a learning model to obtain a power prediction value.
[0017] Furthermore, the power prediction value indicator is the real-time daily accuracy, and the calculation formula is:
[0018]
[0019] Where: P MI is the actual power of the wind farm at time i, P PI is the wind power forecast value of the wind farm at time i, Cap is the total installed capacity of the wind farm, and n is the number of samples for calculating the accuracy.
[0020] Furthermore, the evaluation module obtains the prediction results through the transmission module, and evaluates and compares the prediction results.
[0021] Furthermore, the criteria for evaluating the prediction results are root mean square error and mean absolute error, and the definitions of root mean square error and mean absolute error are respectively:
[0022]
[0023]
[0024] Where: P' i represents the true value, P i is the predicted value.
[0025] Furthermore, the power prediction module establishes a prediction curve according to the prediction result, and the specific steps are as follows:
[0026] Step 1: Divide the wind speed obtained from the wind tower into many small intervals with an interval length of 0.5m / s;
[0027] Step 2: Randomly select a small interval (V i1 ,V i2 ), the power value with the highest probability of occurrence in the power corresponding to the wind speed in this interval is determined and recorded as P imax , forming a wind speed-power data pair (V i2 ,P imax )
[0028] Step 3: Take the wind speed-power data pair as input, and use the least squares fitting method to obtain a curve connecting all the points to obtain the wind speed-power curve of the wind farm.
[0029] Furthermore, the scheduling module is implemented by establishing a load scheduling model;
[0030] The relationship between power consumption and variables such as temperature, energy efficiency, and time can be expressed using a parameter model as follows:
[0031]
[0032] Where: θ in(t) and θ out(t) are the indoor and outdoor temperatures of the wind farm at time t, C is the equivalent heat capacity of the wind farm, R is the equivalent thermal resistance of the wind farm, μ is the energy efficiency ratio of the wind turbine, S(t is the operating state of the wind turbine, P(t) is the operating power of the wind turbine at time t, θ + and θ - It is the upper and lower thresholds for switching the load start and stop status;
[0033] Get the power p of a single wind turbine in the on state at time t on,i ;
[0034]
[0035] According to the law of large numbers, the approximate aggregate power P of the wind turbine load is obtained. agg (t);
[0036]
[0037] According to the inequality transformation, we can get p on,i Transformation interval;
[0038]
[0039] Estimating wind turbine aggregate power range
[0040]
[0041] Where: P agg (t) is the aggregate power of the wind turbine at time t, θ set Set the temperature for the wind farm, P i is the power consumption of the i-th wind turbine.
[0042] Furthermore, the scheduling module power supply and demand estimated total response cost minimum objective function is as follows:
[0043]
[0044] Where: C i is the unit capacity corresponding cost of resource i, μ i The response capacity of resource i, is the weight factor of the aggregate r.
[0045] Furthermore, the acquisition module performs data normalization and feature selection on the acquired data to prevent different data magnitudes of various parameters from affecting feature selection.
[0046] A method for using an electricity supply and demand prediction system based on multi-source meteorological wind power prediction, the prediction method comprising the following steps:
[0047] S1. Real-time collection and storage of real-time data of wind power generators, and data normalization and feature selection processing of the collected data;
[0048] S2, the transmission module realizes the generation and reporting of forecast files and the downloading of weather forecast data, and establishes a connection with the database;
[0049] S3. Based on the historical data of the wind farm, the collected historical data is used as the input of the model, and a statistical model or a learning model is established to perform prediction, obtain the power prediction value, obtain the prediction result through the transmission module, and evaluate and compare the prediction results according to the root mean square error and the mean absolute error;
[0050] S4. Determine the supply and demand situation of the wind power and electricity supply and demand forecasting system of the wind farm, and make reasonable dispatch.
[0051] Compared with the prior art, the present invention has the following beneficial effects:
[0052] 1. In the present invention, by setting up a power prediction module and an evaluation module, using the real-time data collected from the wind farm as the input of the model, the simplest and most effective continuous prediction method in the statistical model is used to obtain the predicted power, and by calculating the error, the physical scheme with the best prediction effect is compared and selected to scientifically evaluate the wind power of the wind farm;
[0053] 2. In the present invention, by setting up a scheduling module, the wind power of the wind farm and the supply and demand situation of the electricity supply and demand forecasting system are judged, and reasonable scheduling is performed, thereby improving the applicability of the electricity supply and demand forecasting system. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 The present invention is a block diagram of a system and method for predicting power supply and demand based on multi-source meteorological wind power forecasting. DETAILED DESCRIPTION
[0055] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0056] like Figure 1 As shown, a power supply and demand forecasting system based on multi-source meteorological wind power forecasting includes a system control platform, a database, a collection module, a transmission module, a statistical module, a power forecasting module, an evaluation module and a scheduling module;
[0057] The system control platform is used to regulate the operation of the power supply and demand forecasting system based on multi-source meteorological wind power forecasting;
[0058] The database stores the power prediction results calculated by the prediction model in real time;
[0059] The acquisition module is used to collect and store the real-time data required by the system, and store all the acquired wind field data with time tags in the database;
[0060] The transmission module is used for forecast file generation and reporting and downloading weather forecast data. When file transmission fails, an alarm message is automatically generated.
[0061] The statistical module is used to calculate the prediction accuracy of the statistical model and the reporting success rate of the files that need to be uploaded in the system. It reads the power prediction data and real-time power data stored in the database regularly, and calculates the prediction accuracy of yesterday according to the prediction accuracy calculation formula.
[0062] The power prediction module is used to predict the power of the wind farm, and the prediction result is stored in the database;
[0063] The evaluation module is used to evaluate and compare the predicted power of the wind farm;
[0064] The dispatching module is used to dispatch the electricity demand of the wind farm.
[0065] It should be noted that the criteria for evaluating the prediction results are the root mean square error and the mean absolute error, which are implemented in the dispatch module by establishing a load dispatch model;
[0066] The power prediction module is implemented by establishing a wind power prediction model, which is mainly based on the historical data of the wind farm. The collected historical data is used as the input of the model, and prediction is performed by establishing a statistical model or a learning model to obtain the power prediction value.
[0067] It should be noted that since wind power is closely related to wind speed, when the wind speed remains unchanged, it can be known that the value of wind power will not change. Therefore, in the continuous prediction model, it can be assumed that the predicted value of wind energy at the wind farm at the current moment is equal to the actual observed value of wind energy at the previous moment.
[0068] The indicator of power prediction value is real-time daily accuracy, and the calculation formula is:
[0069]
[0070] Where: P MI is the actual power of the wind farm at time i, P PI is the wind power forecast value of the wind farm at time i, Cap is the total installed capacity of the wind farm, and n is the number of samples for calculating the accuracy.
[0071] It should be noted that the forecast uses a time scale of 15 minutes and uses the continuous forecasting method to predict wind power for the next 4 hours. The calculation process of the continuous forecasting model is simple and effective, and the performance of the model is good in the ultra-short term. Experiments have shown that in the ultra-short-term wind power forecast, the real-time accuracy of the wind power forecast value calculated using this forecasting method is 90%, which meets the requirement that the power forecast accuracy must not be less than 85%.
[0072] The criteria for evaluating the prediction results are the root mean square error and the mean absolute error. The definitions of the root mean square error and the mean absolute error are:
[0073]
[0074]
[0075] Where: P' i represents the true value, P i is the predicted value.
[0076] It should be noted that the root mean square error of the wind speed prediction results calculated is below 3.6 m / s. The prediction results of the model are good and the prediction accuracy is improved.
[0077] The power prediction module establishes a prediction curve based on the prediction results. The specific steps are as follows:
[0078] Step 1: Divide the wind speed obtained from the wind tower into many small intervals with an interval length of 0.5m / s;
[0079] Step 2: Randomly select a small interval (V i1 ,V i2 ), the power value with the highest probability of occurrence in the power corresponding to the wind speed in this interval is determined and recorded as P imax , forming a wind speed-power data pair (V i2 ,P imax )
[0080] Step 3: Take the wind speed-power data pair as input, and use the least squares fitting method to obtain a curve connecting all the points to obtain the wind speed-power curve of the wind farm.
[0081] It should be noted that the accuracy of the predicted power value obtained by the wind speed-power curve of the wind farm is 83.5%.
[0082] The scheduling module is implemented by establishing a load scheduling model;
[0083] The relationship between power consumption and variables such as temperature, energy efficiency, and time can be expressed using a parameter model as follows:
[0084]
[0085] Where: θ in(t) and θ out(t) are the indoor and outdoor temperatures of the wind farm at time t, C is the equivalent heat capacity of the wind farm, R is the equivalent thermal resistance of the wind farm, μ is the energy efficiency ratio of the wind turbine, S(t) is the operating state of the wind turbine, P(t) is the operating power of the wind turbine at time t, θ + and θ- are the upper and lower thresholds for switching the load start and stop states;
[0086] Get the power p of a single wind turbine in the on state at time t on，i ;
[0087]
[0088] According to the law of large numbers, the approximate aggregate power P of the wind turbine load is obtained. agg (t);
[0089]
[0090] According to the inequality transformation, we can get p on，i Transformation interval;
[0091]
[0092] Estimating wind turbine aggregate power range
[0093]
[0094] Where: P agg (t) is the aggregate power of the wind turbine at time t, θ set Set the temperature for the wind farm, P i is the power consumption of the i-th wind turbine.
[0095] It should be noted that the influence of electric power consumption on variables such as temperature, energy efficiency, and time is quantitatively analyzed to obtain the adjustable margin, and the adjustable range of the aggregated power and the maximum load reduction potential are solved in combination with the controllability of the wind farm and the approximate aggregation model.
[0096] The objective function for minimizing the total response cost of the power supply and demand of the scheduling module is as follows:
[0097]
[0098] Where: C i is the unit capacity corresponding cost of resource i, μ i The response capacity of resource i, is the weight factor of the aggregate r.
[0099] It should be noted that the response strategy is optimized based on the regulation cost, regulation appropriateness, etc., and the results are fed back to the supply and demand forecasting system.
[0100] The acquisition module performs data normalization and feature selection on the collected data to prevent the different data levels of each parameter from affecting the feature selection.
[0101] It should be noted that the linear function normalization method is used to normalize the data to eliminate the dimensional differences between data sets, thereby eliminating the adverse effects of dimensional differences on prediction performance. The classification and regression tree method is then used to rank the importance of the candidate input variables, and parameters with higher importance are selected to reduce the dimension of the input variables.
[0102] A prediction method for an electricity supply and demand prediction system based on multi-source meteorological wind power prediction, the prediction method comprising the following steps:
[0103] S1. Real-time collection and storage of real-time data of wind power generators, and data normalization and feature selection processing of the collected data;
[0104] S2, the transmission module realizes the generation and reporting of forecast files and the downloading of weather forecast data, and establishes a connection with the database;
[0105] S3. Based on the historical data of the wind farm, the collected historical data is used as the input of the model, and a statistical model or a learning model is established to perform prediction, obtain the power prediction value, obtain the prediction result through the transmission module, and evaluate and compare the prediction results according to the root mean square error and the mean absolute error;
[0106] S4. Determine the supply and demand situation of the wind power and electricity supply and demand forecasting system of the wind farm, and make reasonable dispatch.
[0107] In the embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation; the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the method of this embodiment.
[0108] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A power supply and demand forecasting system based on multi-source meteorological wind power forecasting, characterized by: It includes system control platform, database, acquisition module, transmission module, statistics module, power prediction module, evaluation module and scheduling module; The system control platform is used to regulate the operation of the power supply and demand prediction system based on multi-source meteorological wind power prediction; The database stores power prediction results calculated by the prediction model in real time; The acquisition module is used to collect and store real-time data required by the system; The transmission module is used for generating and reporting forecast files and downloading weather forecast data; The statistical module is used to calculate the prediction accuracy of the statistical model and the reporting success rate of the files that need to be uploaded in the system, and to periodically read the power prediction data and real-time power data stored in the database; The power prediction module is used to predict the power of the wind farm; The evaluation module is used to evaluate and compare the predicted power of the wind farm; The dispatching module is used to dispatch the power demand of the wind farm.
2. The power supply and demand forecasting system based on multi-source meteorological wind power forecasting according to claim 1 is characterized in that: The power prediction module is implemented by establishing a wind power prediction model, which is mainly based on the historical data of the wind farm. The collected historical data is used as the input of the model, and prediction is performed by establishing a statistical model or a learning model to obtain a power prediction value.
3. The power supply and demand forecasting system based on multi-source meteorological wind power forecasting according to claim 2 is characterized in that: The indicator of the power prediction value is the real-time daily accuracy, and the calculation formula is: Where: P MI is the actual power of the wind farm at time i, P PI is the wind power forecast value of the wind farm at time i, Cap is the total installed capacity of the wind farm, and n is the number of samples for calculating the accuracy.
4. The power supply and demand forecasting system based on multi-source meteorological wind power forecasting according to claim 1 is characterized in that: The evaluation module obtains the prediction results through the transmission module, and evaluates and compares the prediction results.
5. The power supply and demand forecasting system based on multi-source meteorological wind power forecasting according to claim 4 is characterized in that: The evaluation criteria for the prediction results are root mean square error and mean absolute error, and the definitions of root mean square error and mean absolute error are: Where: P' i represents the true value, P i is the predicted value.
6. The power supply and demand forecasting system based on multi-source meteorological wind power forecasting according to claim 4 is characterized in that: The power prediction module establishes a prediction curve according to the prediction results, and the specific steps are as follows: Step 1: Divide the wind speed obtained from the wind tower into many small intervals with an interval length of 0.5m / s; Step 2: Randomly select a small interval (V i1 ,V i2 ), the power value with the highest probability of occurrence in the power corresponding to the wind speed in this interval is determined and recorded as P imax , forming a wind speed-power data pair (V i2 ,P imax ) Step 3: Take the wind speed-power data pair as input, and use the least squares fitting method to obtain a curve connecting all the points to obtain the wind speed-power curve of the wind farm.
7. The power supply and demand forecasting system based on multi-source meteorological wind power forecasting according to claim 1 is characterized by: The scheduling module is implemented by establishing a load scheduling model; The relationship between power consumption and variables such as temperature, energy efficiency, and time can be expressed using a parameter model as follows: Where: θ in(t) and θ out(t) are the indoor and outdoor temperatures of the wind farm at time t, C is the equivalent heat capacity of the wind farm, R is the equivalent thermal resistance of the wind farm, μ is the energy efficiency ratio of the wind turbine, S(t) is the operating state of the wind turbine, P(t) is the operating power of the wind turbine at time t, θ + and θ - The upper and lower thresholds for switching the load start and stop states; Get the power p of a single wind turbine in the on state at time t on,i ; According to the law of large numbers, the approximate aggregate power P of the wind turbine load is obtained. agg (t); According to the inequality transformation, we can get p on,i Transformation interval; Estimating wind turbine aggregate power range Where: P agg (t) is the aggregate power of the wind turbine at time t, θ set Set the temperature for the wind farm, P i is the power consumption of the i-th wind turbine.
8. The power supply and demand forecasting system based on multi-source meteorological wind power forecasting according to claim 7 is characterized by: The scheduling module power supply and demand estimated total response cost minimum objective function is as follows: Where: C i is the unit capacity corresponding cost of resource i, μ i The response capacity of resource i, is the weight factor of the aggregate r.
9. The power supply and demand forecasting system based on multi-source meteorological wind power forecasting according to claim 1, characterized in that: The acquisition module performs data normalization and feature selection on the collected data to prevent different data magnitudes of various parameters from affecting feature selection.
10. A prediction method for an electricity supply and demand prediction system based on multi-source meteorological wind power prediction according to any one of claims 1 to 9, characterized in that: The prediction method includes the following steps: S1. Real-time collection and storage of real-time data of wind power generators, and data normalization and feature selection processing of the collected data; S2, the transmission module realizes the generation and reporting of forecast files and the downloading of weather forecast data, and establishes a connection with the database; S3. Based on the historical data of the wind farm, the collected historical data is used as the input of the model, and a statistical model or a learning model is established to perform prediction, obtain the power prediction value, obtain the prediction result through the transmission module, and evaluate and compare the prediction results according to the root mean square error and the mean absolute error; S4. Determine the supply and demand situation of the wind power and electricity supply and demand forecasting system of the wind farm, and make reasonable dispatch.