Electric vehicles (EVs) with voltage-to-grid (V2G) capability are useful in augmenting grid to handle high energy demand of end users during peak periods. We propose a hybrid state-of-charge (SOC) battery model aggregator optimize charging and maintain stability The proposed SOC leverages the advantages three well-known previously models namely: Shepherd, Unnewehr Nernst models. is combination merits specified empirical Lithium-ion slow charging. This will enhance performance by improving...

Big data analysis has gained popularity over the years as a result of developments in computing and electronics. Several methods have been proposed literature for efficiently mining from dedicated databases wide range electronic sensors. However, volume grows, diversity velocity also grows (sometimes exponentially). Neural networks optimal big mining; however, they suffer problems over-fitting under-fitting. In this paper, an ensemble evolutionary algorithms is proposed, namely: improved...

Demand side management (DSM) has gained a lot of attention in recent years as result increased deployment alternative renewable energy resources electric power grids. This paper presents novel scavenging differential evolution algorithm which reuses unfit population agents previous generations genetic algorithm. The performance the proposed is compared to another popular evolutionary literature: enhanced (EDE). cost minimization model consists parameters describe consumer savings for...

This article investigates the use of optimal reference point placement to improve performance non-dominated sorting genetic algorithm (NSGA). Placement points for many-objective optimization is inspired by wheel and Von Neumann topologies Particle Swarm Optimization (PSO). Results obtained show that pattern determines efficiency NSGA. The better-performing topology (called (wRPGA), compared three other evolutionary algorithms: knee-driven (KnEA), III (NSGAIII) multi-objective based on...

This paper investigates the effect of selected strategies distributed energy resources (DER) on an cost function that optimizes distribution for a mid-rise apartment building. is achieved through comparing parameter optimization results both high-level and low-level optimizer, respectively. The process carried out using following approach: (1) two-objective constructed with one objective similar to optimizer (DER-CAM); (2) evolutionary algorithm (EA) modified selection capability used...

A ceaseless supply of electricity is essential for all developmental sectors and even more important nowadays, as the globe evolves through a call artificially intelligent systems in fourth industrial revolution. Power interruptions lead to unpleasant human situations well great losses within health, education, facilities requiring constant electricity. For this reason, we propose design an automatic transfer switch (ATS) power applications maximize uptime. The ATS ensures between two...

This paper presents an algorithm based on dynamic multiobjective optimization (DMO) which employs a single randomly mutating time-variant archive to balance convergence and diversity in order efficiently select the final, non-dominated Pareto set. The is tested selected benchmark functions, improvement performance of approach validated by improved metrics overall computational time. Overall, proposed single-archive (called DOAEA) generated better faster time for Gee-Tan-Abbas (GTA) test...

This paper examines the role of demand response aggregators in minimizing cost electricity generation by distribution utilities a day-ahead market. In this paper, 2500 standard South African homes are considered as end users. Five clusters (and aggregators) with 500 each cluster. Two cases analysed: (1) Utilization renewable energy sources (RES) is implemented supply operator (DSO), where it meets excess for users during peak hours purchasing from market, and (2) RES alone, assumed that...

Many real-world problems are modeled as multi-objective optimization whose optimal solutions change with time. These commonly termed dynamic (DMOPs). One challenge associated solving such is the fact that Pareto front or set often changes too quickly. This means solution at period t may likely vary from (t+1), and this makes process of optimizing computationally expensive to implement. article proposes use adaptive mutation crossover operators for non-dominated sorting genetic algorithm III...

The role of solar PV in the global energy mix is becoming more significant as cost modules continues to decline. African continent has a good degree irradiation. Therefore, it imperative that benefits alternative generation through are maximized. In this paper, we present an optimization model case study Sierra Leone which can be adapted for Nigerian distribution system. This because both countries located same region (West Africa) and power systems similar. We consider two demand response...

The challenges of many-objective optimization are investigated; and one new algorithm, which is based on the NSGA-II, proposed for multi-objective in this paper. reference points an adaptable crossover rate combined algorithm to improve performance NSGA-II. NSGA optimizing many objective search space examined with without through a constrained two-objective problem up 40 dimensions. Simulation results show that improves selected test generations where non-dominated set not obtained by 39%.

An increasing concentration of ammonia in cooked food is direct proportion to the extent decay. This fact used design an electronic nose (e-nose) based on metal oxide semiconductor odour sensor circuit capable discriminating good and bad food. On basis data produced by e-nose circuit, a feedforward multilayer neural network designed trained recognize varying concentrations Test results prototype system show that it classifying as being or with over 92% average success rate.

Neural networks are an extremely powerful tool for data mining.They especially useful in cases involving classification where it is difficult to establish a pattern the search space.In era when artificial intelligence increasingly being utilised industrial and medical applications throughout world, becoming evident that this emerging trend.This paper explores idea of by employing use feed-forward neural network with two process layers determine concentration ammonia exhaled human breath.The...

Artificial intelligence (AI) is the aspect of computing concerned with programming computers to behave like humans. In spite fact that no artificial system capable fully simulating human behaviour, there are aspects which have been successfully mimicked. One these applications development intelligent systems model sense smell. The neural network one tool makes inferences based on pattern recognition selected parameters in their environment. This paper applies determination food age using...

Electric vehicles (EVs) with vehicle-to-grid (V2G) capability are a promising source of alternative and renewable energy for the existing power grid. However, it is important to prioritize health EV battery since most expensive single component vehicle. In this paper, we present multiobjective function consisting grid parameters which optimized using recently proposed evolutionary algorithm called enhanced dynamic non-dominated sorting genetic III (E-dyNSGA-III). The aggregator model...

Loss of selection pressure in the presence many objectives is one pertinent problems evolutionary optimization. Therefore, it difficult for algorithms to find best-fitting candidate solutions final Pareto optimal front representing a multi-objective optimization problem, particularly when solution space changes with time. In this study, we propose algorithm called enhanced dynamic non-dominated sorting genetic III (E-dyNSGA-III). This an improvement earlier proposed dyNSGA-III, which used...

Softcomputing techniques are fast becoming reliable and efficient means of prediction estimation.This has made their application more wide spread in recent years.With the growing need for intelligent devices systems comes to explore these even further.This paper applies neural networks differential evolution (two most effective softcomputing algorithms) estimation air quality compares accuracy results using mean square error (MSE) method.Air pollution is an ever increasing menace major...