基于大數(shù)據(jù)降維及優(yōu)化神經(jīng)網(wǎng)絡(luò)的負(fù)荷預(yù)測研究

打開文本圖片集

中圖分類號：TP391.92;TM711 文獻(xiàn)標(biāo)志碼：A 文章編號：1001-5922（2025）05-0182-04

Abstract：In order to improve the powerload prediction ability，a setof power loaddata intelligent monitoring system was designed，and itshardwarestructureincluded information processing terminal，network communicationand monitoring model，etc.，and the system could significantlyimprove the operationcapacityof theentire transmission line.In this study，an embedded monitoring system with the main control chip as the TMS32ODM8168 was designed，and the data computing capability was improved by improving the BP neural network model. Based on the XGboost fusion model，abnormal data information monitoring was realized.Through experiments on the 35kV power load data path，it wasfound thatthe number of designed monitoring lineswas 36，the monitoring claritywas ，and the algorithm recognition accuracy was 97.3% ，which greatly improves the monitoring ability. Key words ：power load； monitoring；embedded monitoring；BP neural network model；XGboost fusion model

隨著輸電線路的增加，電力負(fù)荷數(shù)據(jù)的整體規(guī)劃和智能監(jiān)控越發(fā)重要。（剩余5040字）