Day-Ahead Prediction of Microgrid Electricity Demand Using a Hybrid Artificial Intelligence Model

Abstract

Improved-performance day-ahead electricity demand forecast is important to deliver necessary information for right decision of energy management of microgrids. It supports microgrid operators and stakeholders to have better decisions on microgrid flexibility, stability and control. The available conventional forecasting methods for electricity demand at national or regional level are not effective for electricity demand forecasting in microgrids. This is due to the fact that the electricity consumption in microgrids is many times less than the regional or national demands and it is highly volatile. In this paper, an integrated Artificial Intelligence (AI) based approach consisting of Wavelet Transform (WT), Simulated Annealing (SA) and Feedforward Artificial Neural Network (FFANN) is devised for day-ahead prediction of electric power consumption in microgrids. The FFANN is the basic forecasting engine of the proposed model. The WT is utilized to extract relevant features of the target variable (electric load data series) to obtain a cluster of enhanced-feature subseries. The extracted subseries of the past values of the electric load demand data are employed as the target variables to model the FFANN. The SA optimization technique is employed to obtain the optimal values of the FFANN weight parameters during the training process. Historical information of actual electricity consumption, meteorological variables, daily variations, weekly variations, and working/non-working day indicators have been employed to develop the forecasting tool of the devised integrated AI based approach. The approach is validated using electricity demand data of an operational microgrid in Beijing, China. The prediction results are presented for future testing days with one-hour time interval. The validation results demonstrated that the devised approach is capable to forecast the microgrid electricity demand with acceptably small error and reasonably short computation time. Moreover, the prediction performance of the devised approach has been evaluated relative to other four approaches and resulted in better prediction accuracy.

Performance Analysis of Machine Learning Algorithms for Energy Demand–Supply Prediction in Smart Grids

“…Power consumption prediction algorithms for the day ahead are either ML or AI. In [22], a hybrid AI model (a combination of feed-forward artificial neural network (FFANN), wavelet transform (WT), and simulated annealing (SA)) was used to predict power demand for a day ahead. The hybrid model was shown to be more efficient as compared to using just one method, as in [23], which implemented the neural network technique for similar day-ahead prediction conditions.…”

Systematic Review of Electricity Demand Forecast Using ANN-Based Machine Learning Algorithms

“…Despite the fact that this structure is common to most of the reviewed papers, the authors typically introduce innovations in either the methodology or the algorithms and techniques used in the different stages of the process. The parts where more innovative approaches can be found are: The use of clustering and other categorization techniques with load input data in order to tune the model depending on the patterns found therein [ 31 , 35 , 49 , 51 ]; The ML algorithm proposed to build the forecasting model. The proposed models are typically compared with commonly used ML models.…”

“…The use of clustering and other categorization techniques with load input data in order to tune the model depending on the patterns found therein [ 31 , 35 , 49 , 51 ];…”

5G Wireless Networks in the Future Renewable Energy Systems

“…Artificial intelligence (AI), in conjunction with advanced energy generation and energy storage technologies, has proven its potential in managing energy production and consumption, as well as energy supply and demand (Alreshidi, 2019;Pinto et al, 2019;Ji et al, 2020;Shi et al, 2020). This will improve the ability to understand the real conditions of the energy market, such as demand and supply, and to make very accurate predictions and simulations (Ma and Zhai, 2019;Márquez et al, 2020;Ngarambe et al, 2020). Such circumstances which can obviously have cascading effects on the entire regional network and pose a difficult task for utilities, could be avoided.…”