DBO-LSTM-based Demand-side Resource Load Forecasting Method
Authors:
- 26 January 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 53-57
Abstract
A combined forecasting method based on Dung Beetle Optimization (DBO) and Long Short-Term Memory (LSTM) network is proposed to address the problems of strong nonlinearity of demand-side user load data and the difficulty of forecasting. Firstly, under the consideration of the weather influence factor, the DBO optimizes the parameters of LSTM network, and explores the feature correlation and time sequence correlation between the data, and finally obtains the load prediction result of a typical demand response day. The experiment adopts the actual electricity load data of a region in China, and after comparing and analyzing with the typical LSTM prediction model, it is verified that the proposed model has higher prediction accuracy and greatly reduces the relative error of load prediction.Keywords
This publication has 7 references indexed in Scilit:
- Weather Sensitive Residential Load Forecasting Using Neural Networks2023
- User-Adjustable Capability Mining Technology Based on Improved PSO and LSTM Model2022
- Analysis on Adjustable Potential of Air-Conditioning in Aggregated Load2022
- Typical Commercial User Adjustable Potential Assessment Method2022
- Review on Analysis of Steel Production Load Characteristics and Adjustable Potential2021
- Data Driven Load Forecasting Method Considering Demand Response2020
- Industrial power demand response analysis for one-part real-time pricingIEEE Transactions on Power Systems, 1998