Medium-Term Load Forecasting Using Support Vector Regression, Feature Selection, and Symbiotic Organism Search Optimization

Abstract

An accurate load forecasting has always been one of the main indispensable parts in the operation and planning of power systems. Among different time horizons of forecasting, while short-term load forecasting (STLF) and long-term load forecasting (LTLF) have respectively got benefits of accurate predictors and probabilistic forecasting, medium-term load forecasting (MTLF) requires more attentions due to its vital role in power system operation and planning such as optimal scheduling of generation units, robust planning programs for customer service, and economic supply. In this study, a hybrid method, composed of Support Vector Regression (SVR) and Symbiotic Organism Search Optimization (SOSO) method, is proposed for MTLF. In the proposed forecasting model, SVR is the main part of the forecasting algorithm while SOSO is embedded into it to optimize the parameters of SVR. In addition, a minimum redundancy-maximum relevance feature selection algorithm is applied in the preprocessing of input data. The proposed method is tested on EUNITE competition dataset and compared with some previous works to demonstrate its high performance.

Medium-Term Load Forecasting with Power Market Survey: GEPCO Case Study

“…In the proposed study, the authors also compared the model's computational time. A hybrid model consisting of support vector regression (SVR) and symbiotic organism search optimization (SOSO) is proposed in [13] for MTLF. The SOSO is used to tune the hyper-parameter of SVR.…”

Research on renewable energy power demand forecasting method based on IWOA-SA-BILSTM modeling

“…Artificial intelligence forecasting methods include Extreme Learning Machine (ELM) (He et al, 2021), Support Vector Machine (SVM) (Shi et al, 2019;MuSAA et al, 2021), and various neural network forecasting models (Machado et al, 2021;Rajbhandari et al, 2021;Hu et al, 2023). For example, Shi et al (Shi et al, 2012) utilized SVM to forecast the amount of photovoltaic (PV) load generation and claimed that the results were good; Zare-Noghabi et al (Zare-Noghabi et al, 2019) demonstrated the effectiveness of Support Vector Regression (SVR) in forecasting power system load demand using actual data; Guo et al (Guo et al, 2021) developed a load forecasting model using LSTM, considering demand response, and demonstrated its practicality through experiments; Wen et al (Wen et al, 2022) proposed a short-term load demand forecasting model based on Bi-directional Long Short-Term Memory(BILSTM) considering the uncertainty of short-term load demand and claimed that the model was superior to the traditional forecasting methods; Su Chang et al (Su et al, 2023) utilized LSTM and combined it with multi-feature fusion coding to forecast the power load demand, which improved the accuracy of the power load forecasting; Zhang Suning et al (Zhang et al, 2022) proposed a cross-region power demand forecasting model based on XGBoost for different forms of power demand in multiple regions and claimed that the method can provide fast and accurate forecasting of power demand; Shu Zhang et al (Zhang Shu et al, 2021) proposed a neural network forecasting model based on feature analysis of the LSTM, which improves the prediction accuracy of short-term power demand. Hybrid forecasting methods (Qinghe et al, 2022;He et al, 2023;Sekhar and Dahiya, 2023) combine various effective forecasting methods to enhance the accuracy of electricity demand forecasting.…”

Medium-term impact of daily heat pump load profiles to forecast congestion of distribution network assets

“…Medium-term studies are less covered in the literature, but they provide a solution to the limitations of the short-term and long-term studies discussed. The considered time horizon and time resolution of medium-term studies in the literature vary from a month to a few years ahead [8]. For this mediumterm study, a time horizon of a year ahead with an hourly time resolution is considered.…”