A Hybrid Short-Term Load Forecasting Model Based on Improved Fuzzy C-Means Clustering, Random Forest and Deep Neural Networks

Abstract

Short-term load forecasting (STLF) plays an important role in the secure and reliable operation of the electric power system. Grouping similar load profiles by a clustering algorithm is a common method to reduce the uncertainty of electric consumption data. However, due to the uneven distribution of different date types in a historical data set, the tradition fuzzy c-means clustering (FCM) algorithm cannot identify typical load consumption patterns accurately. To solve this problem, a novel STLF model based on the improved FCM (IFCM) algorithm, random forest (RF) and deep neural networks (DNN) is proposed in this paper. First, IFCM is used to partition the load consumption profiles into several groups, and each group represents a typical load consumption pattern. The optimal number of clusters is determined by a recent clustering validity index. Then, a RF model is trained by the meteorological and calendar features of the historical data set. Finally, a DNN model is established for each group, and is trained using the features of the days that are partition into this group by IFCM. The experimental results on two daily load consumption data sets have showed that the proposed STLF model achieves better prediction performance as compared to other methods. In addition, the load consumption pattern of holidays was extracted from the historical data sets by utilizing IFCM, and the prediction performance of holidays in the testing set therefore has been significantly improved. INDEX TERMS clustering, fuzzy c-mean algorithm, load forecasting, random forest, deep neural network List of Abbreviations FCM, fuzzy c-means clustering STLF, short-term load forecasting RF, random forest DNN, deep neural network CNN, convolutional neural network RNN, recurrent neural network DBN, deep brief network LSTM, long short-term memory GRNN, generalized regression neural network ELM, extreme learning machine MAPE, mean absolute percentage error MASE, mean absolute scaled error RMSE, root mean square error SVM, support vector machine

Hierarchical Clustering and Deep Learning for Short-Term Load Forecasting with Influenced Factors

“…The best architectural modeling value is determined by calculating the Maximum Absolute Percentage Error (MaxAPE) value from each test and accumulated by the average value approach. Forecasting is done using limited training data, namely four days or 4 x 48 data, to predict the next one day or 1 x 48 data, which is sufficient according to the author's previous research [6], [25] The results of the LSTM architectural modeling test are presented in Table 2. on these results, it was found that the best LSTM architecture model lies in model number 9, with the smallest MaxAPE average value of 7,061.…”

“…In general, this method is used to predict Sequential Values or forecasting. This method is applied in many fields such as cellular network [16], power generation [24], bearing conditions [15], Load Forecasting [6], [25], etc. LSTM was developed to solve the RNN problem [26], namely the loss of important information at the beginning of knowledge; if the sequence processed is long enough (forward propagation) and the gradient vanishing problem (backward propagation), the gradient value is very small, so it does not contribute to changes in weight.…”

A New Medium and Long-Term Power Load Forecasting Method Considering Policy Factors

“…In the same category, the load growth rate of known years is used as the load growth rate of predicted years, so as to realize load forecasting. Fuzzy cluster analysis method can consider many related factors at the same time, which is consistent with the characteristics of policy influencing many related factors [32]. However, fuzzy clustering cannot carry quantification weights, and the influence of different factors on load is not considered, so it is improved to weighted fuzzy clustering on the basis of fuzzy clustering.…”

A Novel Short-Term Residential Electric Load Forecasting Method Based on Adaptive Load Aggregation and Deep Learning Algorithms

“…Accurate short-term load forecasting results can be used to form a more reasonable home energy scheduling plan [7][8][9]. Since dispatching management is only applied in the large-scale regions in the traditional power grid, existing electric load forecasting methods are mainly used to obtain regional power load for generation scheduling, transaction scheduling, and network dispatching [10,11]. These load forecasting methods can be mainly grouped into three categories, including similar day or similar time interval-based forecasting methods, frequency component-based forecasting methods, and meteorological factor-based load forecasting methods.…”