New energy power prediction monitoring platform and method for power grid dispatching optimization

Abstract

The invention discloses a new energy power prediction monitoring platform and method for power grid dispatching optimization, and relates to the technical field of power systems, and the method comprises the steps: determining a distribution diagram of a target new energy power station, collecting the historical data and weather forecast data of the power station, and obtaining a plurality of weather forecasts through parallel calculation. And based on weather forecast and historical data, establishing a weather power prediction model. And generating a plurality of power prediction results through the prediction model according to user requirements, performing rolling prediction according to the new energy type, and continuously optimizing the model based on the prediction results. And adjusting a scheduling plan and a power transaction strategy of the power grid through a plurality of power prediction results so as to optimize the utilization of new energy and the scheduling of the power grid. The technical problems of inaccurate new energy power prediction and untimely response in the prior art are solved, and the technical effect of improving the new energy power prediction precision and real-time performance is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a new energy power prediction and monitoring platform and method for power grid dispatch optimization. Background Art

[0002] As the global demand for renewable energy continues to increase, the proportion of new energy power stations such as wind and solar energy in electricity supply is increasing. However, due to the instability of renewable energy power generation and the influence of meteorological factors, power grid dispatching faces many challenges. Traditional power dispatching methods mainly rely on the controllability of conventional energy generation and cannot effectively deal with the volatility and uncertainty of renewable energy generation. Therefore, how to make full use of renewable energy generation under the premise of ensuring the safe and stable operation of the power grid has become a key research direction in the current power dispatching field. Through accurate power prediction and reasonable grid dispatch optimization, the utilization rate of renewable energy can be maximized, the dependence on traditional fossil energy can be reduced, and the operation cost and environmental pollution of the power grid can be reduced. In this context, power prediction technology based on meteorological forecasts and historical data has gradually become an important means to improve the accuracy and efficiency of renewable energy generation dispatching.

[0003] At the current stage, relevant technologies have technical problems such as inaccurate prediction of new energy power and untimely response. Summary of the invention

[0004] This application provides a new energy power prediction and monitoring platform and method for power grid dispatch optimization, adopts a distribution map of the target new energy power station, collects historical data and weather forecast data of the power station, and obtains multiple weather forecasts through parallel computing. Then, based on these weather forecasts and historical data, a meteorological power prediction model is established. According to user needs, multiple power prediction results are generated through the prediction model, and rolling predictions are performed according to the type of new energy, and the model is continuously optimized based on the prediction results. Finally, through multiple power prediction results, the dispatch plan and power trading strategy of the power grid are adjusted to optimize the utilization of new energy and the dispatch of the power grid, achieving the technical effect of improving the accuracy and real-time performance of new energy power prediction.

[0005] This application provides a new energy power prediction and monitoring platform for power grid dispatch optimization, including:

[0006] A distribution map determination module, the distribution map determination module is used to determine the distribution map of the target new energy power station, and collect the historical data set and meteorological forecast data set of the target new energy power station; a parallel calculation module, the parallel calculation module is used to parallelly calculate the meteorological forecast data set to obtain multiple meteorological forecasts; a meteorological power prediction model establishment module, the meteorological power prediction model establishment module is used to establish a meteorological power prediction model based on the multiple meteorological forecasts and the historical data set; a power prediction result acquisition module, the power prediction result acquisition module is used to obtain multiple power prediction results through the meteorological power prediction model according to user needs; a feedback optimization module, the feedback optimization module is used to perform a rolling prediction of new energy power according to the new energy type, and based on the prediction results, feedback optimize the meteorological power prediction model; a scheduling plan adjustment module, the scheduling plan adjustment module is used to adjust the power grid scheduling plan and power trading strategy according to the multiple power prediction results.

[0007] The present application provides a new energy power prediction and monitoring method for power grid dispatch optimization, including:

[0008] Determine the distribution map of the target new energy power station, collect the historical data set and meteorological forecast data set of the target new energy power station; calculate the meteorological forecast data set in parallel to obtain multiple meteorological forecasts; establish a meteorological power prediction model based on the multiple meteorological forecasts and the historical data set; according to user needs, obtain multiple power prediction results through the meteorological power prediction model; according to the type of new energy, perform a rolling prediction of new energy power, and based on the prediction results, feedback and optimize the meteorological power prediction model; according to the multiple power prediction results, adjust the power grid dispatch plan and power trading strategy.

[0009] The new energy power prediction and monitoring platform and method for power grid dispatch optimization proposed in this application first determine the distribution map of the target new energy power station, collect historical data and weather forecast data of the power station, and obtain multiple weather forecasts through parallel computing. Then, based on these weather forecasts and historical data, a meteorological power prediction model is established. According to user needs, multiple power prediction results are generated through the prediction model, and rolling predictions are performed according to the type of new energy, and the model is continuously optimized based on the prediction results. Finally, through multiple power prediction results, the dispatch plan and power trading strategy of the power grid are adjusted to optimize the utilization of new energy and the dispatch of the power grid, achieving the technical effect of improving the accuracy and real-time performance of new energy power prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in this application to illustrate the operations performed by the platform according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0011] Figure 1 A schematic diagram of the structure of a new energy power prediction and monitoring platform for power grid dispatch optimization provided in an embodiment of the present application;

[0012] Figure 2 A flow chart of a new energy power prediction and monitoring method for power grid dispatch optimization provided in an embodiment of the present application.

[0013] Explanation of the reference numerals: distribution diagram determination module 10 , parallel calculation module 20 , meteorological power prediction model establishment module 30 , power prediction result acquisition module 40 , feedback optimization module 50 , scheduling plan adjustment module 60 . DETAILED DESCRIPTION

[0014] 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.

[0015] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0016] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is 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 to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, platform, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are 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 those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0017] The present application embodiment provides a new energy power prediction and monitoring platform for power grid dispatch optimization, such as Figure 1 As shown, the method includes:

[0018] The distribution map determination module 10 is used to determine the distribution map of the target new energy power station and collect the historical data set and weather forecast data set of the target new energy power station. Specifically, in the new energy power prediction and monitoring platform for power grid dispatch optimization, the distribution map determination module 10 uses geographic information system technology combined with a variety of information resources to determine the distribution map of the target new energy power station, and then connects with the power station monitoring system to interactively collect historical data sets, including information such as the actual power generation power of the power station in the past and the operating status of the equipment. At the same time, it cooperates with the meteorological department or data service provider to collect weather forecast data sets, including future wind speed, light intensity and other meteorological parameters, to provide basic data support for the subsequent links of the platform.

[0019] A parallel computing module 20 is used to parallelly compute the weather forecast data set to obtain multiple weather forecasts. Specifically, in a new energy power prediction and monitoring platform for power grid dispatch optimization, after the parallel computing module 20 receives the weather forecast data set, the sub-area division unit divides the target new energy power station distribution map into multiple sub-areas and allocates computing nodes, the time period division unit divides the forecast time range into multiple time periods and also allocates computing nodes, the independent computing task formulation unit formulates tasks for computing nodes, the exchange boundary determination subunit of the data consistency processing unit determines the data exchange boundary of adjacent sub-areas, the data exchange frequency determination subunit determines the exchange frequency, the data exchange subunit performs data exchange, the abnormal data warning subunit sets a gradient change threshold to check the exchanged data and warns of abnormalities, and finally the weather forecast acquisition unit statistically analyzes the results of multiple independent computing tasks to obtain multiple weather forecasts, providing input data for subsequent model establishment.

[0020] In a possible implementation, the parallel computing module 20 further includes: a sub-region division unit, the sub-region division unit is used to analyze the distribution map of the target new energy power station, divide it into multiple sub-regions, and allocate computing nodes based on the spatial characteristics of the sub-regions. Specifically, the sub-region division unit first conducts an in-depth analysis of the distribution map of the target new energy power station, and divides the entire target area into multiple sub-regions by studying the geographical location distribution, topographical characteristics and other relevant spatial factors of the power station. In order to decompose the complex weather forecast calculation task into multiple relatively small and easier to manage parts, the computing nodes are allocated based on the spatial characteristics of the divided sub-regions, such as the size of the area of ​​the sub-region, the complexity of the terrain, etc. For sub-regions with larger areas or more complex terrains, more computing resources can be allocated to ensure that the weather forecast calculation of the sub-region can be carried out efficiently. The allocation method can fully take into account the computing needs of different sub-regions and improve the overall computing efficiency.

[0021] A time period division unit, the time period division unit is used to divide the forecast time range of the target new energy power station into multiple time periods, and allocate computing nodes based on the time characteristics of the time periods. Specifically, the time period division unit divides the forecast time range of the target new energy power station into multiple time periods, which are determined according to factors such as the law of meteorological changes, the time requirements of power grid dispatching, and the availability of computing resources. For example, a day can be divided into several hourly segments, or a week can be divided into different time periods of different days. Based on the time characteristics of the time periods, such as the severity of meteorological changes in different time periods, the importance to power grid dispatching, etc., computing nodes are allocated. For time periods with more drastic meteorological changes or those that are critical to power grid dispatching, more computing resources can be allocated to ensure that meteorological forecast calculations can be accurately performed in these time periods.

[0022] An independent computing task formulation unit, the independent computing task formulation unit is used to formulate independent computing tasks for multiple computing nodes based on the spatial characteristics of the sub-region or the time characteristics of the time period. Specifically, the independent computing task formulation unit formulates independent computing tasks for multiple computing nodes according to the spatial characteristics of the sub-region or the time characteristics of the time period. If the task is formulated based on the spatial characteristics of the sub-region, the specific meteorological conditions and terrain influences of each sub-region will be considered, and the corresponding meteorological forecast computing tasks will be assigned to the computing nodes responsible for the sub-region, such as calculating the change trend of meteorological parameters such as wind speed and light intensity in a specific sub-region. If the task is formulated based on the time characteristics of the time period, the task will be assigned to the corresponding computing node according to the meteorological change law and power grid dispatching requirements in different time periods. For example, in a certain time period, the focus may be on the change of wind speed, while in another time period, the focus may be on the change of light intensity. When formulating independent computing tasks, the scheduler will reasonably allocate tasks according to the performance and load of the computing nodes. If a computing node has strong performance and light load, more complex computing tasks can be assigned; while for computing nodes with weak performance or heavy load, relatively simple tasks are assigned to ensure full utilization of computing resources and efficient execution of tasks.

[0023] A data consistency processing unit, the data consistency processing unit is used to periodically perform data consistency processing on multiple independent computing tasks. Specifically, the data consistency processing unit periodically performs data consistency processing on multiple independent computing tasks. In the parallel computing process, different computing nodes may produce inconsistent data results, affecting the accuracy of the final weather forecast. First, the exchange boundary determination subunit analyzes the distribution map of the target new energy power station and determines the data exchange boundaries of adjacent sub-areas. It can be clear which sub-areas need to exchange data to ensure the continuity of the meteorological field. For example, the meteorological parameters at the boundary of two adjacent sub-areas should change continuously. By determining the exchange boundary, it can be ensured that the data at these boundaries can be correctly processed during the calculation process. The data exchange frequency determination subunit determines the frequency of data exchange according to the time characteristics of the time period. If the time period is short and the meteorological changes are fast, a higher exchange frequency may be required to ensure. To ensure the real-time and accuracy of the data; if the time period is long or the weather changes slowly, the exchange frequency can be appropriately reduced to reduce the consumption of computing resources. The data exchange subunit exchanges data based on the determined data exchange boundary and exchange frequency. The computing nodes of adjacent sub-areas will exchange meteorological data according to predetermined rules, such as wind speed, temperature and other parameters at the boundary, to ensure that the meteorological field of the entire computing area is consistent. The abnormal data warning subunit analyzes the spatial characteristics of adjacent sub-areas and sets the gradient change threshold. During the data exchange process, the exchanged data is checked. If the difference of the exchanged data is found to be greater than the gradient change threshold, it will be marked as consistent abnormal data and a warning will be issued. Abnormal situations in the calculation process can be discovered in time so that appropriate measures can be taken to deal with them and ensure the accuracy and reliability of the data.

[0024] A weather forecast acquisition unit, the weather forecast acquisition unit is used to perform statistical analysis on the calculation results of multiple independent calculation tasks to obtain multiple weather forecasts. Specifically, the weather forecast acquisition unit performs statistical analysis on the calculation results of multiple independent calculation tasks, collects the weather forecast results of the sub-regions or time periods completed by each calculation node, comprehensively considers the weather changes in different sub-regions and time periods, and integrates and processes the results by using statistical analysis methods such as averaging, weighted averaging, interpolation, etc., to obtain multiple more accurate and comprehensive weather forecasts. The weather forecast will provide an important basis for the subsequent establishment of a weather power prediction model and optimization of power grid dispatching.

[0025] In a possible implementation, the data consistency processing unit further includes: an exchange boundary determination subunit, which is used to analyze the distribution map of the target new energy power station and determine the data exchange boundary of the adjacent sub-areas. Specifically, the exchange boundary determination subunit first analyzes the distribution map of the target new energy power station in detail, and determines the data exchange boundary between adjacent sub-areas by studying the geographical distribution of the power station and the division of sub-areas. It is necessary to take into account the shape, size and relative position relationship of each sub-area. For example, if the target new energy power station is distributed in a wide geographical area, the sub-area division may be based on factors such as terrain, climate zone or power grid distribution. In this case, the exchange boundary determination subunit will carefully analyze the edge parts of these sub-areas to determine which places are the boundaries where data exchange is required between adjacent sub-areas. The purpose is to ensure that in the subsequent data exchange process, the meteorological information interaction between adjacent sub-areas can be accurately processed to ensure the continuity of the meteorological field in the entire calculation area.

[0026] The data exchange frequency determination subunit is used to determine the data exchange frequency according to the time characteristics of the time period. Specifically, the data exchange frequency determination subunit determines the frequency of data exchange according to the time characteristics of the time period. If the time period is short and the meteorological changes are rapid, a higher data exchange frequency may be required to ensure that the meteorological information between adjacent sub-areas can be synchronized in time. For example, when severe convective weather is approaching, the meteorological conditions change very quickly. At this time, a higher data exchange frequency is required so that the computing nodes of each sub-area can obtain the latest meteorological information in time and perform accurate meteorological forecast calculations. On the contrary, if the time period is long and the meteorological changes are relatively slow, the data exchange frequency can be appropriately reduced to reduce communication overhead and improve computing efficiency. For example, under stable weather conditions, the meteorological changes are not obvious. At this time, the data exchange frequency can be reduced to avoid unnecessary waste of communication resources. It should be noted that the data exchange boundary and the data exchange frequency need to be updated at each time step, because the meteorological conditions are dynamically changing. As time goes by, the demand for meteorological information interaction between adjacent sub-areas will also change. If the update frequency is too high, it may cause excessive communication overhead and affect computing efficiency; while if the update frequency is too low, it may cause discontinuity or error accumulation at the boundary of the meteorological field.

[0027] A data exchange subunit, the data exchange subunit is used to perform data exchange based on the data exchange boundary and the data exchange frequency. Specifically, the data exchange subunit performs data exchange based on the determined data exchange boundary and data exchange frequency. After determining the data exchange boundary between adjacent sub-areas, the data exchange subunit will exchange meteorological data at the boundary according to predetermined rules and frequencies. For example, for two adjacent sub-areas, the data exchange subunit may exchange data on meteorological parameters such as wind speed, temperature, and humidity at their boundaries to ensure that the meteorological field between adjacent sub-areas is continuous and to avoid discontinuity or mutation of meteorological information. The data exchange subunit also needs to consider the accuracy and reliability of the data to ensure that the exchanged data is verified and processed and does not introduce erroneous information.

[0028] The abnormal data early warning subunit is used to analyze the spatial characteristics of the adjacent sub-areas, set the gradient change threshold, check the exchanged data, determine whether consistent abnormal data appears, and issue an early warning for the consistent abnormal data. Specifically, the abnormal data warning subunit first analyzes the spatial characteristics of adjacent sub-areas, including factors such as the terrain differences, climate differences, and power station distribution of adjacent sub-areas. According to the spatial characteristics, a gradient change threshold is set. The threshold is used to determine whether the exchanged data is abnormal. During the data exchange process, the abnormal data warning subunit will check the exchanged data. If it is found that the difference in the exchanged data is greater than the set gradient change threshold, it is considered that consistent abnormal data has occurred. For example, if the temperature difference between two adjacent sub-areas at the boundary suddenly increases and exceeds the preset threshold, it means that an abnormal situation may have occurred, such as a failure of the meteorological monitoring equipment in a sub-area or an abnormal change in meteorological conditions. Once it is determined that consistent abnormal data has occurred, the abnormal data warning subunit will issue an early warning, which is achieved by sending an alarm message, recording an abnormal data log, or triggering a specific processing procedure. The purpose of the early warning is to promptly notify relevant personnel or systems so that appropriate measures can be taken to handle the problem, ensure the accuracy and reliability of the data, and avoid adverse effects on weather forecasts and power grid dispatch optimization.

[0029] In a possible implementation, the abnormal data warning subunit further includes: an exchange data judgment microunit, the exchange data judgment microunit is used to judge whether the difference of the exchange data is greater than the gradient change threshold. If the difference of the exchange data is greater than the gradient change threshold, the exchange data is marked as consistent abnormal data. If the difference of the exchange data is less than the gradient change threshold, the exchange data is marked as normal exchange data. Specifically, the exchange data judgment microunit first obtains the exchange data that needs to be judged. The exchange data comes from the data exchange process between adjacent sub-areas and usually includes various meteorological parameters, such as wind speed, temperature, humidity, etc. The microunit calculates the difference of the exchange data and determines the difference of the exchange data by comparing the numerical differences of the corresponding meteorological parameters of adjacent sub-areas. The difference reflects the degree of change of adjacent sub-areas in specific meteorological parameters. The microunit compares the difference with the preset gradient change threshold. The gradient change threshold is a critical value set according to factors such as the spatial characteristics of adjacent sub-regions and the law of meteorological changes. It is used to determine whether the exchange data is abnormal. If the difference of the exchange data is greater than the gradient change threshold, it means that the meteorological parameters between adjacent sub-regions change more drastically and exceed the normal range of change. The micro-unit marks the exchange data as consistent abnormal data. The mark can trigger the subsequent early warning mechanism and abnormal handling process so as to promptly notify relevant personnel to check and repair. For example, if the temperature difference between two adjacent sub-regions at the boundary suddenly increases and exceeds the threshold, it may mean that the temperature monitoring equipment in one of the sub-regions fails or an abnormal meteorological event occurs. On the contrary, if the difference of the exchange data is less than the gradient change threshold, it means that the meteorological parameters between adjacent sub-regions are within the normal range. The unit marks the exchange data as normal exchange data. Normal exchange data will continue to participate in the subsequent meteorological forecast calculation and data processing process, providing a reliable basis for accurately predicting the power generation of new energy power stations and optimizing power grid scheduling. The exchange data judgment micro-unit effectively screens out abnormal data and normal data by comparing the difference of the exchange data with the gradient change threshold, providing an important guarantee for the stable operation and accurate prediction of the platform.

[0030] A meteorological power prediction model establishment module 30 is used to establish a meteorological power prediction model based on the multiple meteorological forecasts and the historical data set. Specifically, in the new energy power prediction and monitoring platform for power grid dispatch optimization, the meteorological power prediction model establishment module 30 receives multiple meteorological forecasts of the parallel computing module 20 and the historical data set of the distribution map determination module 10. The key data extraction unit extracts key data from the meteorological forecast and stores it according to the time period sequence and sub-region location. The meteorological power generation model construction unit combines the historical power generation data and key data in the historical data set, and uses data analysis and modeling algorithms to establish a meteorological power generation model, considering the impact of various factors on the power generation of new energy power stations, and establishing a mapping relationship between meteorological conditions and power generation. The power generation prediction value is output to provide a decision-making basis for the adjustment of the power grid dispatch plan.

[0031] In a possible implementation, the meteorological power prediction model establishment module 30 further includes: a key data extraction unit, the key data extraction unit is used to extract the meteorological forecast key data of the meteorological forecast, and store the meteorological forecast key data according to the time period sequence and sub-region position. Specifically, the key data extraction unit first conducts an in-depth analysis of the meteorological forecast, which contains a lot of meteorological parameter information, but not all data have a direct and important impact on the power generation of the new energy power station, and extracts the meteorological forecast key data, which are meteorological parameters closely related to the power generation of the new energy power station, such as wind speed, wind direction, etc. for wind power stations, and light intensity, temperature, etc. for solar power stations. After extracting the meteorological forecast key data, store it according to the time period sequence and sub-region position. The storage method helps to quickly and accurately obtain the key data of the corresponding time and space position when establishing the meteorological power generation model in the future. For example, when it is necessary to predict the power generation of a specific sub-region in a certain time period in the future, the corresponding meteorological information can be directly extracted from the stored key data to improve the efficiency and accuracy of modeling.

[0032] A meteorological power generation model building unit is used to obtain historical power generation data from the historical data set, and to establish a meteorological power generation model in combination with the key meteorological forecast data. Specifically, according to the type (such as wind power, photovoltaic, etc.) and characteristics of the new energy power station, a suitable power prediction model is selected. If it is a wind power station, considering that its power generation principle mainly relies on air flow to drive the rotation of the wind turbine blades, a physical model based on aerodynamics will be selected. The model predicts power output by considering the impact of meteorological conditions on power generation equipment. For example, meteorological elements such as wind speed and wind direction are used, combined with equipment parameters such as the hub height and blade length of the wind turbine, and the relationship between wind speed and wind turbine output power is calculated according to the principles of aerodynamics. For some new energy power stations with richer data and more complex relationships, the statistical relationship between historical meteorological data and power data is used to establish a linear equation between meteorological parameters and power generation for prediction. Time series analysis can consider the trend of power generation over time, use past power data to predict future power values, and determine the input variables of the model, which usually include meteorological forecast data and other related factors. For wind power prediction, the input variables can include meteorological elements such as wind speed, wind direction, air pressure, temperature, and equipment parameters such as the hub height and blade length of the wind turbine. The input variables can fully reflect the various factors affecting wind power generation. At the same time, time information can also be used as one of the input variables. For example, the power generation in different seasons and different time periods may be different. The output variable of the model, that is, the power generation prediction result of the new energy power station, is determined. The output variable can be the average power, peak power or total power generation in the future period, etc., which is determined according to actual needs. The model is trained and the parameters are adjusted according to historical data. The historical meteorological data and the corresponding power generation data are input into the model. By continuously adjusting the parameters of the model, the prediction results of the model are made as close as possible to the actual power generation to improve the prediction accuracy of the model. Using the cross-validation method, the historical data is divided into multiple subsets, which are used for training and validating the model respectively to avoid overfitting. Methods such as grid search can also be used to optimize the parameters of the model. By traversing different parameter combinations, the parameter settings with the best performance are found, and different models are compared and evaluated. Different types of models are trained and tested using the same historical data, and then the model with the best performance is selected as the final power prediction model based on the evaluation indicators. The evaluation indicators can include root mean square error, mean absolute error, correlation coefficient, etc.The smaller the root mean square error and the mean absolute error, the smaller the deviation between the model's prediction results and the actual values; the closer the correlation coefficient is to 1, the higher the correlation between the model's prediction results and the actual values. When new weather forecast data is received, it is input into the established power prediction model. Before inputting the data, it is necessary to ensure that the format and unit of the input data are consistent with the model requirements. For example, if the model requires the input wind speed unit to be meters per second, and the wind speed unit in the received weather forecast data is kilometers per hour, it is necessary to perform unit conversion and run the power prediction model. According to the input weather forecast data and model parameters, the power generation prediction results of the new energy power station are calculated. The prediction results can provide important decision-making basis for the power grid dispatch plan adjustment module 60, help power grid managers to reasonably arrange the power generation plan of the new energy power station, and optimize the operation efficiency and stability of the power grid.

[0033] The power forecast result acquisition module 40 further includes: a power forecast result composition unit, and the power forecast result composition unit is used for the power forecast result to include an interval prediction result and a density prediction result. Specifically, in a new energy power forecasting and monitoring platform for power grid dispatch optimization, the power forecast result composition unit of the power forecast result acquisition module 40 generates a power forecast result including an interval prediction result and a density prediction result. Among them, the confidence determination subunit determines the confidence according to the operation demand characteristics and the new energy use scenario, and the interval prediction result acquisition subunit calculates the confidence interval based on the prediction result of the confidence statistics meteorological power generation power model to obtain the interval prediction result; the density prediction result acquisition subunit uses the prediction result of the meteorological power generation power model to generate a probability distribution curve to obtain the density prediction result, which provides support for the adjustment of the power grid dispatch plan and the formulation of power trading strategies.

[0034] In a possible implementation, the power prediction result composition unit further includes: a confidence determination subunit, which is used to determine the confidence according to the operation demand characteristics and the new energy use scenario. Specifically, the confidence determination subunit first analyzes the operation demand characteristics, which cover the specific requirements for power prediction accuracy. For example, in some key power grid dispatching scenarios, very high prediction accuracy is required to ensure the stable operation of the power grid, including expectations for the reliability of the prediction results. For example, when making major power trading decisions, a higher confidence is required to reduce risks. At the same time, the subunit will also consider the new energy use scenario. Different types of new energy power stations have different power generation characteristics and uncertainties. For example, wind power generation is greatly affected by wind speed changes, while photovoltaic power generation is restricted by light intensity and weather conditions. Different grid access locations and load requirements will also affect the requirements for confidence. By comprehensively considering the operation demand characteristics and the new energy use scenario, the subunit determines a suitable confidence level, which will be used as an important parameter for the subsequent calculation of the interval prediction results.

[0035] The interval prediction result acquisition subunit is used to count the prediction results of the meteorological power generation model, and calculate the confidence interval in combination with the confidence level to obtain the interval prediction result. Specifically, the interval prediction result acquisition subunit first counts the prediction results of the meteorological power generation model. The prediction result is the predicted value of the power generation of the new energy power station calculated based on the meteorological forecast data and the meteorological power generation model. In combination with the confidence level determined by the confidence level determination subunit, the subunit uses statistical methods to calculate the confidence interval. For example, if a confidence level of 95% is determined, the upper and lower limits of the confidence interval can usually be calculated based on the normal distribution assumption and the mean and standard deviation of the prediction results. Through calculation, the subunit obtains the interval prediction result. The interval prediction result represents the possible value range of the power generation of the new energy power station under a given confidence level. The interval prediction result provides a prediction method that takes uncertainty into account for power grid dispatching and power trading decisions.

[0036] The density prediction result acquisition subunit is used to use the prediction results of the meteorological power generation model to generate a probability distribution curve and obtain a density prediction result. Specifically, the density prediction result acquisition subunit uses the prediction results of the meteorological power generation model as basic data. The prediction results include the possible power generation values ​​of the new energy power station under different meteorological conditions. The subunit generates a probability distribution curve by analyzing and processing these prediction results. It can use methods such as kernel density estimation to construct a continuous probability density function based on a large number of prediction results to reflect the probability of occurrence of different power generation values. The subunit obtains the density prediction result. The result provides the probability distribution information of the power generation of the new energy power station, so that the power grid manager can have a more comprehensive understanding of the uncertainty of the power generation. For example, the probability of different power intervals can be seen from the probability distribution curve, which is helpful to formulate more reasonable power grid dispatching plans and risk management strategies.

[0037] Feedback optimization module 50, the feedback optimization module 50 is used to perform rolling prediction of new energy power according to the type of new energy, and feedback optimize the meteorological power prediction model based on the prediction results. Specifically, in the new energy power prediction and monitoring platform for power grid dispatch optimization, the feedback optimization module 50 performs rolling power prediction according to different new energy types. In actual operation, new meteorological and actual power generation data are continuously collected, and real-time or periodic power forecasts are performed based on the new data. Compare the prediction results with the actual values, analyze the reasons for the deviation, if the meteorological forecast data is inaccurate, cooperate with more accurate meteorological data sources or adopt data fusion technology; if the model parameters are not suitable, adjust the parameters or introduce new feature variables. Through continuous rolling prediction and feedback optimization, the meteorological power prediction model can adapt to changes, improve accuracy and reliability, and provide precise support for power grid dispatch and power trading.

[0038] In one possible implementation, the feedback optimization module 50 further includes: a grid load change prediction data determination unit, which is used to analyze seasonal characteristics and new energy type characteristics to determine grid load change prediction data. Specifically, seasonal characteristics and new energy type characteristics are analyzed. Different seasons have a significant impact on grid load. For example, in summer, the temperature is high, and the use of refrigeration equipment such as air conditioners increases significantly, causing a sharp increase in grid load; in winter, the use of heating equipment will also increase grid load. In addition, meteorological conditions such as sunshine time and wind speed in different seasons will also affect the power generation capacity of new energy. Historical grid load data is collected and sorted by season. The changing trend of grid load in different seasons, the time of peak occurrence, and the amplitude of load fluctuations are analyzed. For example, through the analysis of summer grid load data in the past few years, it was found that from 2 pm to 5 pm every day, the peak value of the load is 1.5 pm, and the peak value of the load is 2.5 pm. 8 o'clock is usually the peak load period, and as the temperature rises, the load peak will also increase accordingly. According to seasonal characteristics, the impact of seasonal factors on the grid load is considered. For example, when predicting the grid load in summer, the weight of the air-conditioning load is appropriately increased; when predicting in winter, the load demand of heating equipment is considered. Different types of new energy have different power generation characteristics. Solar power generation mainly depends on sunshine, and its power generation is higher during the day and zero at night; wind power generation depends on wind speed and has greater randomness and volatility; hydropower generation is affected by water flow and water level difference, which is relatively stable but also has seasonal changes. Research on the power generation laws of different types of new energy and their relationship with the grid For example, solar power generation can meet part of the grid load demand during the day, but it cannot supply power at night, and other energy sources are needed to supplement it. The volatility of wind power generation may lead to instability in the grid load, and corresponding adjustment measures need to be taken to consider the impact of factors such as installed capacity, power generation efficiency, and reliability of new energy types on the grid load. For example, if a region has a large installed capacity of wind power generation, but its power generation efficiency is low and unstable, then when predicting the grid load, it is necessary to fully consider its contribution to the grid and the possible risks it may bring, determine the grid load change prediction data, and comprehensively analyze the seasonal characteristics and new energy type characteristics to establish a grid load forecast. The load change prediction model adopts statistical analysis methods, machine learning algorithms or physical models, for example, a multivariate linear regression model is used, with seasonal factors, new energy power generation, traditional energy power generation, etc. as independent variables, and power grid load as the dependent variable. The parameters of the model are determined by training historical data. The established prediction model is used in combination with current seasonal information, new energy power generation forecast data, and traditional energy supply conditions to predict the power grid load change trend in the future. For example, based on the current season being summer, the forecast power of solar and wind power generation, and the supply plan of traditional energy, the power grid load at different times of day in the next week is predicted.

[0039] A prediction frequency determination unit, the prediction frequency determination unit is used to determine the prediction frequency of the rolling prediction of the new energy power based on the grid load change prediction data. Specifically, based on the grid load change prediction data, analyze the characteristics and change rules of the grid load change prediction data, for example, observe the fluctuation frequency, amplitude and relationship with time of the grid load. If it is found that the grid load changes more drastically in certain time periods, then it is necessary to increase the frequency of the rolling prediction of the new energy power to better adapt to the load changes, consider the correlation between the grid load changes and the new energy power generation, if the new energy power generation accounts for a large proportion of the grid load, and the change of its power generation has an important impact on the grid load, then it is necessary to timely adjust the frequency of the rolling prediction of the new energy power according to the changes in the grid load. For example, when the proportion of wind power generation in the grid load is high and the wind speed changes frequently, it is necessary to increase the prediction frequency of wind power generation in order to timely adjust the grid dispatch plan. According to the characteristics and needs of the grid load changes, determine the appropriate rolling prediction frequency of the new energy power. If the grid load changes rapidly, a higher prediction frequency is required in order to timely adjust the new energy The output power of renewable energy generation and the grid dispatch plan are determined by the forecast frequency. For example, during peak hours of electricity consumption, the grid load changes rapidly, and it may be necessary to conduct a rolling forecast of renewable energy power every hour or even shorter. Considering the limitations of forecast costs and computing resources, a higher forecast frequency may require more computing resources and time, and will also increase the forecast cost. Therefore, it is necessary to balance forecast accuracy and cost to determine the optimal forecast frequency. For example, if the forecast frequency is too high, resulting in tight computing resources, the forecast frequency can be appropriately reduced, but it is necessary to ensure that it can be adjusted in time when the grid load changes greatly. A dynamic adjustment mechanism is established to continuously adjust the frequency of renewable energy power rolling forecasts as the grid load and renewable energy power generation change. For example, some trigger conditions can be set. When the grid load changes by more than a certain amount or the renewable energy power generation fluctuates greatly, the forecast frequency is automatically increased; conversely, when the load and power generation are relatively stable, the forecast frequency is reduced to save computing resources.

[0040] In a possible implementation, the predicted frequency determination unit further includes: a difference value comparison subunit, and the difference value comparison subunit is used to compare the difference value between the actual demand of the power grid dispatching and the predicted result. Specifically, the actual demand of the power grid dispatching is determined. The actual demand of the power grid dispatching refers to the power supply required to meet the power demand of power users and ensure the safe and stable operation of the power grid in a specific time period. The demand is affected by many factors, including but not limited to the user's power consumption behavior, industrial production activities, seasonal changes, weather conditions, etc. The power grid operation parameters such as voltage, current, power, etc., and the demand information of the power market, such as power consumption forecasts in different time periods and users' power contracts, are obtained in real time through the power grid monitoring system to determine the actual demand of the power grid dispatching. For example, in the hot summer weather, the power consumption of refrigeration equipment such as air conditioners increases significantly, and the actual demand of the power grid dispatching will also increase accordingly. The prediction result is obtained. The prediction result is the predicted value of the new energy power generation power obtained by the new energy power rolling prediction model. The prediction model combines meteorological data, historical operation data of new energy power stations and other relevant factors to predict the future. The power generation of new energy in a certain period of time is predicted. For example, for wind power generation, the prediction model will predict the wind power generation in the next few hours based on meteorological parameters such as wind speed, wind direction, air pressure, as well as the performance parameters and historical operation data of the wind turbine. For solar power generation, the prediction model will consider factors such as sunshine intensity, cloud coverage, seasonal changes, etc., and predict the solar power generation in the next day. The difference value is compared, and the actual demand of the power grid dispatch is compared with the prediction result. The difference between the two is calculated. The difference value can be expressed as an absolute difference or a relative difference. For example, if the actual demand of the power grid dispatch is 1000 MW and the prediction result is 900 MW, then the absolute difference value is 100 MW; the relative difference value can be calculated as (1000-900) / 1000=0.1, that is, 10%. By comparing the difference values, the degree of deviation between the prediction result and the actual demand can be understood, providing a basis for subsequent adjustment of the prediction frequency.

[0041] The prediction frequency adjustment subunit is used to adjust the prediction frequency of the rolling prediction of the new energy power according to the difference value. Specifically, a difference value threshold is set. In order to determine when the prediction frequency needs to be adjusted, it is first necessary to set the difference value threshold. The threshold can be determined based on factors such as the operation requirements of the power grid, the stability of new energy power generation, and the accuracy of the prediction model. For example, it can be set that when the absolute difference value exceeds a certain number of megawatts (such as 50 megawatts) or the relative difference value exceeds a certain percentage (such as 5%), it is considered that the difference is large and the prediction frequency needs to be adjusted. The direction and amplitude of the prediction frequency are adjusted. If the difference value exceeds the threshold, it is necessary to adjust the prediction frequency of the rolling prediction of the new energy power according to the size and direction of the difference value. If the prediction result is lower than the actual demand of the power grid dispatch, it means that the new energy power generation may not be able to meet the needs of the power grid. At this time, the prediction frequency needs to be increased in order to more timely adjust the output power of the new energy power generation or take other measures to meet the needs of the power grid. For example, if the difference value is large and continues to increase, the prediction frequency can be increased from every The prediction frequency can be increased from once an hour to once every half an hour so as to update the prediction results more frequently and provide more timely decision-making basis for power grid dispatching. On the contrary, if the prediction result is higher than the actual demand of power grid dispatching, it means that there may be excess new energy power generation. At this time, the prediction frequency can be appropriately reduced to save computing resources, but at the same time, it is also necessary to pay close attention to changes in power grid load to prevent sudden increases in load from causing insufficient new energy power generation and dynamically adjust the prediction frequency. As the power grid operation and new energy power generation change, the difference value will continue to change. Therefore, it is necessary to establish a mechanism for dynamically adjusting the prediction frequency and continuously adjust the prediction frequency according to the real-time difference value. For example, a timer can be set to recalculate the difference value every certain period of time (such as 15 minutes) and adjust the prediction frequency according to the new difference value. This can ensure that the prediction frequency is always in line with the actual situation of power grid demand and new energy power generation, and improve the efficiency and accuracy of power grid dispatching.

[0042] The scheduling plan adjustment module 60 is used to adjust the power grid scheduling plan and power trading strategy according to the multiple power prediction results. Specifically, in the new energy power prediction and monitoring platform for power grid scheduling optimization, the scheduling plan adjustment module 60 receives multiple power prediction results from the power prediction result acquisition module 60, including interval prediction results and density prediction results. After analyzing these results, the power grid scheduling plan is optimized, such as adjusting the traditional energy generation plan and arranging spare capacity according to the power of new energy generation, adjusting the power trading strategy, and formulating the power purchase and sales strategy in advance according to the prediction results. It also continuously monitors the actual situation and compares it with the prediction results, and makes timely adjustments when deviations occur to ensure the safe and stable operation of the power grid and maximize economic benefits.

[0043] In the above, refer to Figure 1The new energy power prediction and monitoring platform for power grid dispatch optimization according to an embodiment of the present invention is described in detail. Figure 2 A new energy power prediction and monitoring method for power grid dispatch optimization according to an embodiment of the present invention is described.

[0044] New energy power prediction and monitoring methods for power grid dispatch optimization, such as Figure 2 As shown, the method includes:

[0045] Determine the distribution map of the target new energy power station, collect the historical data set and meteorological forecast data set of the target new energy power station; calculate the meteorological forecast data set in parallel to obtain multiple meteorological forecasts; establish a meteorological power prediction model based on the multiple meteorological forecasts and the historical data set; according to user needs, obtain multiple power prediction results through the meteorological power prediction model; according to the type of new energy, perform a rolling prediction of new energy power, and based on the prediction results, feedback and optimize the meteorological power prediction model; according to the multiple power prediction results, adjust the power grid dispatch plan and power trading strategy.

[0046] In a possible implementation, the new energy power prediction and monitoring method for power grid dispatch optimization also includes: analyzing the distribution map of the target new energy power station and dividing it into multiple sub-areas, and allocating computing nodes based on the spatial characteristics of the sub-areas; dividing the forecast time range of the target new energy power station into multiple time periods, and allocating computing nodes based on the time characteristics of the time periods; formulating independent computing tasks for multiple computing nodes based on the spatial characteristics of the sub-areas or the time characteristics of the time periods; regularly performing data consistency processing on multiple independent computing tasks; and performing statistical analysis on the calculation results of multiple independent computing tasks to obtain multiple weather forecasts.

[0047] In a possible implementation, the new energy power prediction and monitoring method for power grid dispatch optimization also includes: analyzing the distribution map of the target new energy power station to determine the data exchange boundaries of adjacent sub-areas; determining the data exchange frequency according to the time characteristics of the time period; performing data exchange based on the data exchange boundary and the data exchange frequency; analyzing the spatial characteristics of adjacent sub-areas, setting a gradient change threshold, checking the exchanged data, determining whether consistent abnormal data appears, and issuing an early warning for the consistent abnormal data.

[0048] In one possible implementation, the new energy power prediction and monitoring method for power grid dispatch optimization also includes: determining whether the difference of the exchange data is greater than the gradient change threshold; if the difference of the exchange data is greater than the gradient change threshold, marking the exchange data as consistent abnormal data; if the difference of the exchange data is less than the gradient change threshold, marking the exchange data as normal exchange data.

[0049] In one possible implementation, the new energy power prediction and monitoring method for power grid dispatch optimization also includes: extracting key meteorological forecast data of the meteorological forecast, and storing the key meteorological forecast data according to time period sequence and sub-region location; obtaining historical power generation data from the historical data set, and establishing a meteorological power generation model in combination with the key meteorological forecast data.

[0050] In a possible implementation, the new energy power prediction and monitoring method for power grid dispatch optimization further includes: the power prediction result includes an interval prediction result and a density prediction result.

[0051] In one possible implementation, the new energy power prediction and monitoring method for power grid dispatch optimization also includes: determining the confidence level based on the operation demand characteristics and new energy usage scenarios; statistically analyzing the prediction results of the meteorological power generation model, combining the confidence level, calculating the confidence interval, and obtaining the interval prediction results; using the prediction results of the meteorological power generation model to generate a probability distribution curve and obtain a density prediction result.

[0052] In one possible implementation, the new energy power prediction and monitoring method for power grid dispatch optimization also includes: analyzing seasonal characteristics and new energy type characteristics to determine power grid load change prediction data; and determining the prediction frequency of the new energy power rolling prediction based on the power grid load change prediction data.

[0053] In a possible implementation, the new energy power prediction and monitoring method for power grid dispatch optimization also includes: comparing the difference between the actual power grid dispatch demand and the prediction result; and adjusting the prediction frequency of the new energy power rolling prediction based on the difference.

[0054] The embodiment of the present application adopts a distribution map of the target new energy power station, collects historical data and weather forecast data of the power station, and obtains multiple weather forecasts through parallel computing. Then, based on these weather forecasts and historical data, a meteorological power prediction model is established. According to user needs, multiple power prediction results are generated through the prediction model, and rolling predictions are performed according to the type of new energy, and the model is continuously optimized based on the prediction results. Finally, through multiple power prediction results, the dispatch plan and power trading strategy of the power grid are adjusted to optimize the utilization of new energy and the dispatch of the power grid, achieving the technical effect of improving the accuracy and real-time performance of new energy power prediction.

[0055] The new energy power prediction and monitoring platform for power grid dispatch optimization provided by the embodiment of the present invention can execute the new energy power prediction and monitoring method for power grid dispatch optimization provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0056] Although the present application makes various references to certain modules in the platform according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0057] The above specific implementation manner does not constitute a limitation to the protection scope of the present 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 substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does 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. A new energy power prediction and monitoring platform for power grid dispatch optimization, characterized in that: include: A distribution map determination module, the distribution map determination module is used to determine the distribution map of the target new energy power station, and collect the historical data set and weather forecast data set of the target new energy power station; A parallel computing module, the parallel computing module is used to parallelly compute the weather forecast data set to obtain multiple weather forecasts; A meteorological power prediction model establishment module, the meteorological power prediction model establishment module is used to establish a meteorological power prediction model based on the multiple meteorological forecasts and the historical data set; A power prediction result acquisition module, wherein the power prediction result acquisition module is used to obtain a plurality of power prediction results through the meteorological power prediction model according to user needs; A feedback optimization module, wherein the feedback optimization module is used to perform rolling prediction of new energy power according to the new energy type, and to feedback optimize the meteorological power prediction model based on the prediction results; A scheduling plan adjustment module is used to adjust the power grid scheduling plan and power trading strategy according to the multiple power prediction results.

2. The new energy power prediction and monitoring platform for power grid dispatch optimization according to claim 1, characterized in that: The parallel computing module comprises: A sub-region division unit, the sub-region division unit is used to analyze the distribution map of the target new energy power station, divide it into a plurality of sub-regions, and allocate computing nodes based on the spatial characteristics of the sub-regions; A time period division unit, the time period division unit is used to divide the forecast time range of the target new energy power station into a plurality of time periods, and allocate computing nodes based on the time characteristics of the time periods; An independent computing task formulation unit, the independent computing task formulation unit is used to formulate independent computing tasks for the plurality of computing nodes based on the spatial characteristics of the sub-region or the temporal characteristics of the time period; A data consistency processing unit, the data consistency processing unit is used to periodically perform data consistency processing on a plurality of the independent computing tasks; A weather forecast acquisition unit is used to perform statistical analysis on the calculation results of multiple independent computing tasks to obtain multiple weather forecasts.

3. The new energy power prediction and monitoring platform for power grid dispatch optimization according to claim 2, characterized in that: The data consistency processing unit includes: An exchange boundary determination subunit, the exchange boundary determination subunit is used to analyze the distribution map of the target new energy power station and determine the data exchange boundary of the adjacent sub-areas; a data exchange frequency determination subunit, the data exchange frequency determination subunit being used to determine the data exchange frequency according to the time characteristics of the time period; a data exchange subunit, the data exchange subunit being used to perform data exchange based on the data exchange boundary and the data exchange frequency; The abnormal data early warning subunit is used to analyze the spatial characteristics of the adjacent sub-areas, set the gradient change threshold, check the exchanged data, determine whether consistent abnormal data appears, and issue an early warning for the consistent abnormal data.

4. The new energy power prediction and monitoring platform for power grid dispatch optimization according to claim 3, characterized in that: The exchange data judgment micro-unit is used to judge whether the difference of the exchange data is greater than the gradient change threshold. If the difference of the exchange data is greater than the gradient change threshold, the exchange data is marked as consistent abnormal data; if the difference of the exchange data is less than the gradient change threshold, the exchange data is marked as normal exchange data.

5. The new energy power prediction and monitoring platform for power grid dispatch optimization according to claim 1, characterized in that: The meteorological power prediction model establishment module includes: A key data extraction unit, the key data extraction unit is used to extract the weather forecast key data of the weather forecast, and store the weather forecast key data according to the time period sequence and the sub-region position; A meteorological power generation model building unit is used to obtain historical power generation data from the historical data set, and to establish a meteorological power generation model in combination with the key meteorological forecast data.

6. The new energy power prediction and monitoring platform for power grid dispatch optimization according to claim 1, characterized in that: A power prediction result composition unit, wherein the power prediction result composition unit is used for the power prediction result to include an interval prediction result and a density prediction result.

7. The new energy power prediction and monitoring platform for power grid dispatch optimization according to claim 6, characterized in that: The power prediction result composition unit comprises: A confidence determination subunit, the confidence determination subunit is used to determine the confidence according to the operation demand characteristics and the new energy use scenario; An interval prediction result acquisition subunit, the interval prediction result acquisition subunit is used to collect statistics on the prediction results of the meteorological power generation model, calculate the confidence interval in combination with the confidence level, and obtain the interval prediction result; The density prediction result acquisition subunit is used to generate a probability distribution curve using the prediction result of the meteorological power generation model to obtain a density prediction result.

8. The new energy power prediction and monitoring platform for power grid dispatch optimization according to claim 1, characterized in that: The feedback optimization module comprises: A power grid load change prediction data determination unit, the power grid load change prediction data determination unit is used to analyze seasonal characteristics and new energy type characteristics to determine power grid load change prediction data; A prediction frequency determination unit, wherein the prediction frequency determination unit is used to determine the prediction frequency of the new energy power rolling prediction based on the power grid load change prediction data.

9. The new energy power prediction and monitoring platform for power grid dispatch optimization according to claim 8, characterized in that: The predicted frequency determination unit comprises: A difference value comparison subunit, the difference value comparison subunit is used to compare the difference value between the actual demand of power grid dispatching and the predicted result; A prediction frequency adjustment subunit, wherein the prediction frequency adjustment subunit is used to adjust the prediction frequency of the new energy power rolling prediction according to the difference value.

10. A new energy power prediction and monitoring method for power grid dispatch optimization, characterized in that: The method is applied to the new energy power prediction and monitoring platform for power grid dispatch optimization according to any one of claims 1 to 9, and the method comprises: Determine a distribution map of a target new energy power station, and collect a historical data set and a weather forecast data set of the target new energy power station; Parallel computing of the weather forecast data set to obtain multiple weather forecasts; Establishing a meteorological power prediction model based on the multiple meteorological forecasts and the historical data set; According to user needs, multiple power prediction results are obtained through the meteorological power prediction model; Carry out rolling forecast of new energy power according to the type of new energy, and provide feedback to optimize the meteorological power forecast model based on the forecast results; According to the multiple power prediction results, the power grid dispatch plan and power trading strategy are adjusted.

Images