In this paper, we propose mathematical optimization models of household energy units to optimally control the major residential energy loads while preserving the user preferences. User comfort is modelled in a simple way, which considers appliance class, user preferences and weather conditions. The wind-driven optimization (WDO) algorithm with the objective function of comfort maximization along with minimum electricity cost is defined and implemented. On the other hand, for maximum electricity bill and peak reduction, min-max regret-based knapsack problem (K-WDO) algorithm is used. To validate the effectiveness of the proposed algorithms, extensive simulations are conducted for several scenarios. The simulations show that the proposed algorithms provide with the best optimal results with a fast convergence rate, as compared to the existing techniques.

With the rapid increase in the world’s population, the global electricity demand has increased drastically. Therefore, it is required to adopt efficient energy management mechanisms. Since the energy consumption trends are rather dynamic. Therefore, precise energy demand estimation and short and/or long-term forecasting results with higher accuracy are required to develop the optimization and control mechanism. Consequently, the machine learning (ML) techniques along with distributed demand response programs are being adopted to predict the future energy demand requirement with satisfactory results. In this paper, different state-of-the-art ML algorithms such as logistic regression (LR), support vector machines (SVM), naive Bayes (NB), decision tree classifier (DTC), K-nearest neighbor (KNN), and neural networks (NNs), have been implemented to analyze their performance. The main objective of this paper is to present a comparative analysis of ML algorithms for short-term load forecasting (STLF) regarding accuracy and forecast error. Based on the implementation and analysis, we have identified that, among other algorithms, the DTC provides comparatively better results. Therefore, we devised the enhanced DTC (EDTC) by integrating fitting function, loss function, and gradient boosting in DTC mathematical model for fine-tuning the control variables. The implementation results show that the proposed EDTC algorithm provides better forecast results (i.e., 99.9 % recall, 100% F1, 100% precision, 99.21 % training accuracy, and 99.70% testing accuracy.)

Despite the universal importance of price based demand response (DR) for managing electric vehicle (EV) charging load, the academic literature has explored various mechanisms to its implementation. The prequel to this work has demonstrated that implementation of load management schemes on the basis of price based DR programs leads to costlier scheduling for low or constant energy consumers. In this regard, the proposed work has considered and expanded the same idea from analytical as well as implementation point of view to multiple EV charging regions and respective loads. We present a novel mechanism to calculate EV charging prices using individualized energy consumption patterns of EVs in each region. In this regard, all EV regions/stations receive a dynamic price signal which is non-discriminatory in nature. The dynamic price signals are specifically designed to mitigate the impact of discriminatory prices on end user’s cost. Furthermore, the other objectives of these non-discriminatory prices are to lower energy cost and rebound peaks without affecting utility objective (i.e., net revenue). Initially, a new mathematical model is presented to calculate charging prices based on real time load demand and market dynamics. Then relatively a well behaved functional form of the optimization problem is formulated and the cost minimization objective function is solved by using genetic algorithm (GA). The optimization program successfully converges to give global optimum solution validating the effectiveness of proposed mechanism. Finally, the analytical and simulation results are conducted to show the achievements of our proposed work in terms of fair cost distribution with high user satisfaction. It is also proved that in both mechanisms, the utility’s revenue remains unaffected.

This paper presents real time information based energy management algorithms to reduce electricity cost and peak to average ratio (PAR) while preserving user comfort in a smart home. We categorize household appliances into thermostatically controlled (tc), user aware (ua), elastic (el), inelastic (iel) and regular (r) appliances/loads. An optimization problem is formulated to reduce electricity cost by determining the optimal use of household appliances. The operational schedules of these appliances are optimized in response to the electricity price signals and customer preferences to maximize electricity cost saving and user comfort while minimizing curtailed energy. Mathematical optimization models of tc appliances, i.e., air-conditioner and refrigerator, are proposed which are solved by using intelligent programmable communication thermostat ( iPCT). We add extra intelligence to conventional programmable communication thermostat (CPCT) by using genetic algorithm (GA) to control tc appliances under comfort constraints. The optimization models for ua, el, and iel appliances are solved subject to electricity cost minimization and PAR reduction. Considering user comfort, el appliances are considered where users can adjust appliance waiting time to increase or decrease their comfort level. Furthermore, energy demand of r appliances is fulfilled via local supply where the major objective is to reduce the fuel cost of various generators by proper scheduling. Simulation results show that the proposed algorithms efficiently schedule the energy demand of all types of appliances by considering identified constraints (i.e., PAR, variable prices, temperature, capacity limit and waiting time).

This paper proposes a new hybrid framework for short-term load forecasting (STLF) by combining the Feature Engineering (FE) and Bayesian Optimization (BO) algorithms with a Bayesian Neural Network (BNN). The FE module comprises feature selection and extraction phases. Firstly, by merging the Random Forest (RaF) and Relief-F (ReF) algorithms, we developed a hybrid feature selector based on grey correlation analysis (GCA) to eliminate feature redundancy. Secondly, a radial basis Kernel function and principal component analysis (KPCA) are integrated into the feature-extraction module for dimensional reduction. Thirdly, the Bayesian Optimization (BO) algorithm is used to fine-tune the control parameters of a BNN and provides more accurate results by avoiding the optimal local trapping. The proposed FE-BNN-BO framework works in such a way to ensure stability, convergence, and accuracy. The proposed FE-BNN-BO model is tested on the hourly load data obtained from the PJM, USA, electricity market. In addition, the simulation results are also compared with other benchmark models such as Bi-Level, long short-term memory (LSTM), an accurate and fast convergence-based ANN (ANN-AFC), and a mutual-information-based ANN (ANN-MI). The results show that the proposed model has significantly improved the accuracy with a fast convergence rate and reduced the mean absolute percent error (MAPE).