This paper develops a state of charge (SOC) estimation model for a lithium-ion battery using an improved extreme learning machine (ELM) algorithm. ELM is suitable for SOC estimation since ELM algorithm has fast estimation speed, good generalization performance, and high accuracy. However, the performance of ELM is highly dependent on training accuracy and number of neurons in a hidden layer. Hence, gravitational search algorithm (GSA) is applied to improve the ELM computational intelligence by searching for the optimal value hidden layer neurons. The optimal ELM based GSA model does not require internal battery knowledge and mathematical model for SOC estimation. The model robustness is validated at different temperatures using different electric vehicle (EV) drive cycles. The performance of ELM-GSA model is verified with two popular neural network methods; back-propagation neural network (BPNN) and radial basis function neural network (RBFNN). The results are evaluated using different error rates and computation cost. The results demonstrate that the ELM based GSA model offers high accuracy and low SOC error rate than BPNN based GSA and RBFNN based GSA models. Furthermore, a detailed comparative study between the proposed model and existing SOC strategies is conducted which also demonstrates the superiority of the proposed model.

In recent years, a trend of shifting from traditional power grids to modern smart grids has emerged the formation of microgrids (MGs), connecting low-voltage distributed generation (DG) units, loads and local storage apparatus to the medium voltage distribution system. This revolution in power distribution systems has pledged ample advantages for customers including reliability, quality, and efficiency of generated power. Furthermore, it also eliminates the necessity to construct long transmission lines resulting in cost saving and reduced power losses. In contrary to the benefits provided by MGs, protection of these entities is an enormously perplexing procedure predominantly due to dynamic behavior of MGs, bidirectional power flow, and high penetration of inverter-interfaced DG sources that interferes the conventional operation and coordination of protection relays. This article presents an analytical appraisal on state-of-the-art protection techniques to address problems associated with the MG protection. Advantages and disadvantages of each protection technique, as well as proper selection of protective relays suitable for each protection zone are discussed. Recommendations on protection procedures and effective techniques be employed to resolve the MG protection issues are also presented.

This paper presents a novel binary backtracking search algorithm (BBSA) for an optimal scheduling controller applied to the IEEE 14-bus test system for controlling distributed generators (DGs) in microgrids (MGs) in the form of virtual power plant (VPP) toward sustainable renewable energy sources integration. The VPP and MGs models are simulated and tested based on real parameters and loads data recorded in Perlis, Malaysia, employed on each bus of the system for 24 h. BBSA optimization algorithm provides the best binary fitness function, i.e., global minimum fitness for finding the best cell to generate the optimal schedule. The fitness function is generated based on real conditions such as solar irradiation and wind speed and preparation of battery charge/discharges, fuel states and demand of the specific hour. The obtained results show that the BBSA algorithm provides the best schedule to control DGs ON and OFF based on controller decision. Results obtained from the BBSA are compared with binary particle swarm optimization in terms of objective function and power saving to validate the developed controller. The developed BBSA optimization algorithm minimizes the power generation cost, reduces power losses, delivers reliable and high-quality power to the loads, and integrates priority-based sustainable MGs into the grid. Thus, VPP can enable efficient integration of DGs and MGs into the grid by balancing their variability.

This article presents the fuzzy-based charging-discharging control technique of lithium-ion battery storage in microgrid application. Considering available power, load demand, and battery state-of-charge (SOC), the proposed fuzzy-based scheme enables the storage to charge or discharge within the safe operating region. Various controlling techniques have been implemented to evaluate and control the battery performance, which has the limitation of controlling overcharging or overdischarging, complexity in control, and longer charging time. Besides, a fuzzy controller is less complex and faster, as it obviates the extra sensing components, requires no additional deep discharging and overcharging protection, and easy to implement due to the absence of mathematical calculation. The numerical simulations with the load demand and the generations demonstrate the effectiveness of the proposed charging-discharging controller strategy. The efficacy of the proposed controller is tested under certain load variations for real-time application. The obtained experimental result shows that the developed model can control the battery charging-discharging efficiently. Moreover, it is also seen from the output that the battery SOC does not go beyond the limit of the respective safe battery operating region (20%-80%). Thus, the main contribution of this research is to develop an improved fuzzy model and, thus, implement the system for real-time application to control the charging-discharging of the battery.

The main objective of this study is to develop a quantum-behaved lightening search algorithm (QLSA) to improve the indirect field-oriented fuzzy-proportional-integral (PI) controller technique to control a three-phase induction motor (TIM) drive. The generated adaptive PI current control parameters and fuzzy membership functions are carried to design induction motor drive speed controller to minimize the fitness function formulated by QLSA. An optimal QLSA-based indirect field-oriented control (QLSA-IFOC) fitness function is used to reduce the mean absolute error of the rotor speed to improve the performance of the TIM with varying speed and mechanical load. Results obtained from the QLSA-IFOC are compared with those obtained through lightening search algorithm, gravitational search algorithm, backtracking search algorithm, and particle swarm optimization to validate the developed controller. The optimization results of objective functions in terms of box plots and iterations show that the QLSA algorithm outperforms the other optimization algorithms. Moreover, the QLSA-IFOC controller performed well in all tests in terms of transient response. The developed controller also minimizes overshoot, increases damping capability, and reduces the root-mean-square error, as well as standard deviation under sudden change of speed and mechanical loads. A comparative analysis is performed between simulation and experimental results to justify the efficiency of the developed controller.