Residential Electricity Load Forecasting Based on Fuzzy Cluster Analysis and LSSVM with Optimization by the Fireworks Algorithm

Abstract

As the construction of the energy internet progresses, the proportion of residential electricity consumption in end-use energy consumption is increasing, the peak load on the grid is growing year on year, and seasonal and regional peak power supply tensions, mainly for residential electricity consumption, have become common problems across the country. Accurate residential load forecasting can provide strong data support for the operation of electricity demand response and the incentive setting of the response. For the accuracy and stability of residential electricity load forecasting, a forecasting model is presented in this paper based on fuzzy cluster analysis (FC), least-squares support vector machine (LSSVM), and a fireworks algorithm (FWA). First of all, to reduce the redundancy of input data, it is necessary to reduce the dimension of data features. Then, FWA is used to optimize the arguments γ and σ2 of LSSVM, where γ is the penalty factor and σ2 denotes the kernel width. Finally, a load forecasting method of FC–FWA–LSSVM is developed. Relevant data from Beijing, China, are selected for training tests to demonstrate the effectiveness of the proposed model. The results show that the FC–FWA–LSSVM hybrid model proposed in this paper has high accuracy in residential power load forecasting, and the model has good stability and versatility.

Short-term Electricity Load Forecasting Based on TPE-optimized CNN-BIGRU-Attention

“…compared with other hyperparameter optimization algorithms, the coefficient of determination2 R of the model reached 98.8% after the hyperparameters of the CNN-BiGRU-Attention prediction model were optimized by TPE, verifying the effectiveness of TPE hyperparameter search.…”

“…The author Jin [1] points out that BiGRU can effectively learn the bi-directional temporal characteristics of data. The author Zhao [2] points out that the attention mechanism can effectively highlight important features that are more effective for prediction models, which can further improve prediction accuracy. The author Zou [3] proposes a CNN-BiGRU model for power load forecasting, which effectively extracts high-dimensional features and bidirectional timeseries features of load data, but the selection of model hyperparameters is decided by the author's personal experience, and the selected hyperparameters may not be optimal, so the prediction accuracy can be further improved by seeking optimal hyperparameters.…”

Research on short-term power load forecasting method based on multi-factor feature analysis and LSTM

“…LSTM overcomes the problem of gradient disappearance or gradient explosion during the training of Recurrent Neural Network (RNN) [17][18]. Compared with RNN, LSTM is characterized by the introduction of gating mechanism, which can mine the data information of longer time series.…”

Day-Ahead Spot Market Price Forecast Based on a Hybrid Extreme Learning Machine Technique: A Case Study in China

“…Market demand includes the market main body, the different time scales of market demand, market demand prediction deviation in the province and the provincial requirements, etc., Thermal power output includes the history and actual output of thermal power, scheduling, and modulation of and participation in peak shaving, etc. [47,48]. New energy output includes the historical output of New energy, forecast deviation of New energy output, and proportion of New energy output in market demand, etc.…”