Design And Implementation Of Power Line Short-term Load Forecasting System

Power load forecasting is helpful for the electric power department to arrange the electric energy production plan reasonably and to plan the development of power grid.However,in some undeveloped areas,short-term load forecasting basically relies on experience evaluation,with few forecasting methods and unscientific forecasting methods,which makes the prediction results not ideal,causing huge obstacles to the power work and not suitable for the needs of various regional development plans.In order to make up for the shortage of such regional load forecasting methods,based on the actual production environment,this paper collects a large number of historical load data generated in the operation process of power lines.Through the study of a large number of documents,various algorithms of short-term load forecasting of power lines are analyzed and studied,and the functional characteristics of the prediction algorithm and the actual environment are analyzed,and finally the application is selected Elman neural network algorithm with time-series data analysis and processing and delay feedback is used as the basic prediction algorithm in this paper.At the same time,combined with data analysis method,a training data group construction method is developed to overcome the defect of single historical data type and improve the prediction accuracy of algorithm model.Based on the algorithm research,this paper designs the user interface of the forecasting system based on the UI graphical interface design function provided by MATLAB software,and completes the function realization of the prediction system through the software design function of MATLAB.Finally,the design and implementation of the power line load forecasting system is realized by using Microsoft Visual Studio 2010 and MATLAB.In order to verify the applicability of the system,the historical data of different substations are used to verify the functionality and applicability of the system.Finally,it is confirmed that the prediction system meets the requirements of the target environment.