At present, the back-propagation (BP) network algorithm widely used in the short-term output prediction of photovoltaic power stations has the disadvantage of ignoring meteorological factors and weather conditions in the input. The existing traditional BP prediction model lacks a variety of numerical optimization algorithms, such that the prediction error is large. The back-propagation (BP) neural network is easy to fall into local optimization thus reducing the prediction accuracy in photovoltaic power prediction. In order to solve this problem, an improved grey wolf optimization (GWO) algorithm is proposed to optimize the photovoltaic power prediction model of the BP neural network. So, an improved grey wolf optimization algorithm optimized BP neural network for a photovoltaic (PV) power prediction model is proposed. Dynamic weight strategy, tent mapping and particle swarm optimization (PSO) are introduced in the standard grey wolf optimization (GWO) to construct the PSO–GWO model. The relative error of the PSO–GWO–BP model predicted data is less than that of the BP model predicted data. The average relative error of PSO–GWO–BP and GWO–BP models is smaller, the average relative error of PSO–GWO–BP model is the smallest, and the prediction stability of the PSO–GWO–BP model is the best. The model stability and prediction accuracy of PSO–GWO–BP are better than those of GWO–BP and BP.

The demand for energy on a global scale increases day by day. Unlike renewable energy sources, fossil fuels have limited reserves and meet most of the world’s energy needs despite their adverse environmental effects. This study presents a new forecast strategy, including an optimization-based S-curve approach for coal consumption in Turkey. For this approach, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), Grey Wolf Optimization (GWO), and Whale Optimization Algorithm (WOA) are among the meta-heuristic optimization techniques used to determine the optimum parameters of the S-curve. In addition, these algorithms and Artificial Neural Network (ANN) have also been used to estimate coal consumption. In evaluating coal consumption with ANN, energy and economic parameters such as installed capacity, gross generation, net electric consumption, import, export, and population energy are used for input parameters. In ANN modeling, the Feed Forward Multilayer Perceptron Network structure was used, and Levenberg-Marquardt Back Propagation has used to perform network training. S-curves have been calculated using optimization, and their performance in predicting coal consumption has been evaluated statistically. The findings reveal that the optimization-based S-curve approach gives higher accuracy than ANN in solving the presented problem. The statistical results calculated by the GWO have higher accuracy than the PSO, WOA, and GA with R ² = 0.9881, RE = 0.011, RMSE = 1.079, MAE = 1.3584, and STD = 1.5187. The novelty of this study, the presented methodology does not need more input parameters for analysis. Therefore, it can be easily used with high accuracy to estimate coal consumption within other countries with an increasing trend in coal consumption, such as Turkey.

Multimedia networks utilize low-power scalar nodes to modify wakeup cycles of high-performance multimedia nodes, which assists in optimizing the power-toperformance ratios. A wide variety of machine learning models are proposed by researchers to perform this task, and most of them are either highly complex, or showcase low-levels of efficiency when applied to large-scale networks. To overcome these issues, this text proposes design of a Q-learning based iterative sleep-scheduling and fuses these schedules with an efficient hybrid bioinspired multipath routing model for largescale multimedia network sets. The proposed model initially uses an iterative Q-Learning technique that analyzes energy consumption patterns of nodes, and incrementally modifies their sleep schedules. These sleep schedules are used by scalar nodes to efficiently wakeup multimedia nodes during adhoc communication requests. These communication requests are processed by a combination of Grey Wolf Optimizer (GWO) & Genetic Algorithm (GA) models, which assist in the identification of optimal paths. These paths are estimated via combined analysis of temporal throughput & packet delivery performance, with node-to-node distance & residual energy metrics. The GWO Model uses instantaneous node & network parameters, while the GA Model analyzes temporal metrics in order to identify optimal routing paths. Both these path sets are fused together via the Q-Learning mechanism, which assists in Iterative Adhoc Path Correction (IAPC), thereby improving the energy efficiency, while reducing communication delay via multipath analysis. Due to a fusion of these models, the proposed Q-Learning based Iterative sleep-scheduling & hybrid Bioinspired Multipath Routing model for Multimedia Networks (QIBMRMN) is able to reduce communication delay by 2.6%, reduce energy consumed during these communications by 14.0%, while improving throughput by 19.6% & packet delivery performance by 8.3% when compared with standard multimedia routing techniques.

In global path planning (GPP), an autonomous underwater vehicle (AUV) tracks a predefined path. The main objective of GPP is to generate a collision free sub-optimal path with minimum path cost. The path is defined as a set of segments, passing through selected nodes known as waypoints. For smooth planar motion, the path cost is a function of the path length, the threat cost and the cost of diving. Path length is the total distance travelled from start to end point, threat cost is the penalty of collision with the obstacle and cost of diving is the energy expanse for diving deeper in ocean. This paper addresses the GPP problem for multiple AUVs in formation. Here, Grey Wolf Optimization (GWO) algorithm is used to find the suboptimal path for multiple AUVs in formation. The results obtained are compared to the results of applying Genetic Algorithm (GA) to the same problem. GA concept is simple to understand, easy to implement and supports multi-objective optimization. It is robust to local minima and have wide applications in various fields of science, engineering and commerce. Hence, GA is used for this comparative study. The performance analysis is based on computational time, length of the path generated and the total path cost. The resultant path obtained using GWO is found to be better than GA in terms of path cost and processing time. Thus, GWO is used as the GPP algorithm for three AUVs in formation. The formation follows leader-follower topography. A sliding mode controller (SMC) is developed to minimize the tracking error based on local information while maintaining formation, as mild communication exists. The stability of the sliding surface is verified by Lyapunov stability analysis. With proper path planning, the path cost can be minimized as AUVs can reach their target in less time with less energy expanses. Thus, lower path cost leads to less expensive underwater missions.