An Intelligent E-Learning Recommender System Using Hybrid Feature Selection and Deep Learning

Estimation of transformer parameters from nameplate data by imperialist competitive and gravitational search algorithms

Network lifetime play a vital role in setting up an energy efficient wireless sensor network. The lifetime of the network can be improved by efficient deployment and scheduling of the sensor nodes inside the network. In this paper, on the basis of mathematically calculated upper bound lifetime of the network, the sensor nodes are efficiently deployed by using Simulated Annealing and Particle Swarm Optimization with pre-specified sensing range of the sensor nodes and on the second fold the sensor nodes are efficiently scheduled by applying the Simulated Annealing and Dempster Shafer Theory. The overall objective of this paper is to find the optimized location and schedule for sensor nodes. The comparative study shows that the simulated annealing and particle swarm optimization algorithms performs better for sensors deployment. Simulated annealing and Dempster Shafer theory achieves the goal to provide the different schedules with high efficiency.

In striving to remain competitive in the global market, the concept of optimization of manufacturing processes has been extensively employed to meet the diverse production requirements. Optimization analysis of machining processes is usually based on either minimizing or maximizing certain objective functions. Recently, various non-traditional optimization techniques have evolved to optimize the process parameters of machining processes. The objective of this work is to study the effectiveness of the most commonly used non-traditional optimization methods as applied to a particular machining optimization problem. In this work, surface grinding processes are optimized using i) Particle Swarm Optimization (PSO) ii) Adaptive Genetic Algorithm (AGA) iii) Simulated Annealing (SA) and iv) Memetic algorithm (MA). Memetic algorithm used here has two variations as MA-1 and MA-2, each having the combination of PSO and SA and AGA and SA respectively. The mathematical model of surface grinding operations was adopted from a literature. A computer program was written in Visual C++ for the optimization computations. The computation results of various optimization methods are compared and it is observed that the results of PSO method have outperformed the results of other methods in terms of the combined objective function (COF).

The integration of renewable sources and energy storage in residential microgrids offers energy efficiency and emission reduction potential. Effective energy management is vital for optimizing resources and lowering costs. In this paper, we propose a novel approach, combining the imperialist competitive algorithm (ICA) with particle swarm optimization (PSO) as ICA-PSO to enhance energy management. The proposed energy management system operates in an offline mode, anticipating data for the upcoming 24 h, including consumption predictions, tariff rates, and meteorological data. This anticipatory approach facilitates optimal power distribution among the various connected sources within the microgrid. The performance of the proposed hybrid ICA-PSO algorithm is evaluated by comparing it with three selected benchmark algorithms, namely the genetic algorithm (GA), ICA, and PSO. This comparison aims to assess the effectiveness of the ICA-PSO algorithm in optimizing energy management in multi-source residential microgrids. The simulation results, obtained using Matlab 2023a, provide clear evidence of the effectiveness of the hybrid ICA-PSO algorithm in achieving optimal power flows and delivering substantial cost savings. The hybrid algorithm outperforms the benchmark algorithms with cost reductions of 4.47%, 14.93%, and 26% compared to ICA, PSO, and GA, respectively. Furthermore, it achieves a remarkable participation rate of 50.6% for renewable resources in the energy mix, surpassing the participation levels of the ICA (42.88%), PSO (40.51%), and GA (38.95%). This research contributes to the advancement of power flow management techniques in the context of multi-source residential microgrids, paving the way for further research and development in this field.

Production scheduling optimization in foundry using hybrid Particle Swarm Optimization algorithm

This study compares the performance of four different metaheuristics for solving a constraint satisfaction scheduling problem of the outfitting process of shipbuilding. The ship outfitting process is often unorganised and chaotic due to the complex interactions between the stakeholders and the overall lack of sufficiently detailed planning. The examined methods are genetic algorithms (GA), simulated annealing (SA), genetic simulated annealing (GSA) and discrete particle swarm optimisation (PSO). Each of these methods relies on a list scheduling heuristic to transform the solution space into feasible schedules. Although the SA had the best performance for a medium-sized superstructure section, the GSA created the best schedules for engine room double-bottom sections, the most complex sections in terms of outfitting. The GA provided the best scalability in terms of computational time while only marginally sacrificing solution quality. The solution quality of the PSO was very poor in comparison with the other methods. All methods generated schedules with sufficiently high resource utilisation, approximately 95%. The findings from this work will be incorporated into a larger project with the aim of creating a tool which can automatically generate an outfitting planning for a vessel.

Non-traditional algorithms are the realistic models that are used for solving many complex engineering optimization problems. Machining is one of the most important and widely used manufacturing processes which rely on optimization. Optimization problems may be either constrained or unconstrained in nature. The complication factor with constrained optimization is that there is a possible existence of one or more complex constraints. These constraints must be incorporated into the problem otherwise solution is unacceptable from a practical standpoint. Non-traditional algorithms are the best way for solving the constrained optimization problems. The advantages of non-traditional techniques are its tolerant of imprecision, uncertainty, and meta-heuristics. In this work, three different non-traditional algorithms such as genetic algorithm (GA), simulated annealing (SA), and particle swarm optimization (PSO) are used in multi-pass turning machining operation to identify the set of optimal parameter values for minimizing unit production cost. The mathematical model is taken from the literature (Chen and Tsai, Int J Prod Res 34:2803–2825, 1996 and it is used for the evaluation of algorithmic parameters. The main aim of this work is to analyze the effect of algorithmic parameters of non-traditional technique in turning operation. The analyses performed are (i) sensitivity analysis in GA, PSO, and SA, (ii) analysis to find the best reproduction method in GA (iii) analysis to analyze the performance of GA, PSO, and SA by varying algorithmic parameters, and (iv) analysis to compare the results of the algorithms with the literature, and the best one is proposed.

Optimal management of hybrid PV/fuel cell/battery power system: A comparison of optimal hybrid approaches

In view of the problem of premature convergence of traditional particle swarm optimization (PSO) algorithm in path planning, which is easy to converge to local optimal solution and poor path quality, some theory about the corresponding PSO algorithm of simulated annealing optimization is studied in this paper. While planning the moving path of the robot, it analyzes the effect of initial temperature and cooling coefficient on path length and iteration times from the major contributing factors of simulated annealing algorithm. Thus deduce the law of its change and seek the optimal parameter matching. Simulated annealing algorithm can not only move the updated particle position on the basis of the particle swarm optimization formula, but also select the updated position with a certain probability. The method is used to avoid the particle converging into the local optimal solution in the whole iterative process. The capabilities of the global optimization is strengthened. Compared with the traditional PSO algorithm, the simulated annealing PSO in complex environment has better optimization ability, shorter path and fewer iterations in the simulation results.KeywordsParticle swarmPath planningSimulated annealingLinear inertia weight

Chapter 10 - Constraint-based heuristic algorithms for software test generation

The mathematical model of a multi-product multi-period multi-echelon closed-loop supply chain network design under uncertainty is designed in this paper. The designed network consists of raw material suppliers, plants, warehouses, distribution centers, and customer zones in forward chain and collection centers, repair centers, recovery/decomposition center, and disposal center in the reverse chain. The goal of the model is to determine the quantities of products and raw material transported between the supply chain entities in each period by considering different transportation mode, the number and locations of the potential facilities, the shortage of products in each period, and the inventory of products in warehouses and plants with considering discount and uncertainty parameters. The robust possibilistic optimization approach was used to control the uncertainty parameter. At the end to solve the proposed model, five meta-heuristic algorithms include genetic algorithm, bee colony algorithm, simulated annealing, imperial competitive algorithm, and particle swarm optimization are utilized. Finally, some numerical illustrations are provided to compare the proposed algorithms. The results show the genetic algorithm is an efficient algorithm for solving the designed model in this paper.

<p><strong>Context and relevance.</strong>The traveling salesman problem (TSP) is one of the key NP-hard combinatorial optimization problems with wide applications in logistics, transportation planning, and other fields. Exact methods for solving TSP become inefficient as the problem dimension increases, making the use of metaheuristic methods such as simulated annealing (SA), ant colony optimization (ACO), and particle swarm optimization (PSO) relevant. <strong>Objective. </strong>To conduct a comparative analysis of the effectiveness of SA, ACO and PSO for solving the traveling salesman problem with 100 points, to evaluate their convergence and to identify the optimal parameters. <strong>Hypothesis. </strong>The ant algorithm, due to the pheromone trail mechanism, will provide higher accuracy of solutions compared to SA and PSO, but will require more computational resources. <strong>Methods and materials. </strong>The study conducted computational experiments with varying parameters of each algorithm. For SA, the initial temperature and cooling coefficient were analyzed, for ACO, the influence of pheromones and distances, for PSO, the cognitive and social coefficients. <strong>Results. </strong>ACO showed the best results, finding a route with a length of 9.23, demonstrating high efficiency for medium-sized problems. PSO took an intermediate position with a route length of 32.64, making it suitable for moderately complex problems. SA was the least effective, with a route length of 45.2, which limits its use for complex routing networks. <strong>Conclusions. </strong>The ant algorithm demonstrated the greatest efficiency due to the balance between exploring the solution space and exploiting the found routes, which is especially important for maritime Arctic logistics, where sustainable route optimization is required under high uncertainty. PSO showed good convergence, but it is inferior to ACO in terms of accuracy. SA, despite its simplicity, is less effective for high-dimensional problems. The obtained results are of particular practical value for route optimization in the Arctic region, where traditional methods often prove to be ineffective.</p>

The evaluation and precise prediction of safety factor (SF) of slopes can be useful in designing/analyzing these important structures. In this study, an attempt has been made to evaluate/predict SF of many homogenous slopes in static and dynamic conditions through applying various hybrid intelligent systems namely imperialist competitive algorithm (ICA)-artificial neural network (ANN), genetic algorithm (GA)-ANN, particle swarm optimization (PSO)-ANN and artificial bee colony (ABC)-ANN. In fact, ICA, PSO, GA and ABC were used to adjust weights and biases of ANN model. In order to achieve the aim of this study, a database composed of 699 datasets with 5 model inputs including slope gradient, slope height, friction angle of soil, soil cohesion and peak ground acceleration and one output (SF) was established. Several parametric investigations were conducted in order to determine the most effective factors of GA, ICA, ABC and PSO algorithms. The obtained results of hybrid models were check considering two performance indices, i.e., root-mean-square error and coefficient of determination $$(R^{2})$$ . To evaluate capability of all hybrid models, a new system of ranking, i.e., the color intensity rating, was developed. As a result, although all predictive models are able to approximate slope SF values, PSO-ANN predictive model can perform better compared to others. Based on $$R^{2}$$ , values of (0.969, 0.957, 0.980 and 0.920) were found for testing of ICA-ANN, ABC-ANN, PSO-ANN and GA-ANN predictive models, respectively, which show higher efficiency of the PSO-ANN model in predicting slope SF values.

As blasting has some environmental constraints, ripping has got more prevalent as a method of breaking ground in both civil and mining engineering applications. A rippability model of a higher applicability is, therefore, needed to appropriately predict the ripping production (Q) prior to carrying out such tests. For the purpose of predicting the results of ripping production that were attained in three sites located in Johor state, Malaysia, the present study applied three hybrid intelligent techniques, i.e., neuro-bee, neuro-imperialism, and neuro-swarm, to predict Q. In fact, the effects of artificial bee colony (ABC), imperialism competitive algorithm (ICA), and particle swarm optimization (PSO) on weights and biases of neural networks were examined in the current research to receive better prediction/evaluation of ripping production. To do this, totally, 74 ripping tests were taken into consideration in the investigated regions and their influential parameters were assessed. Many parametric studies on ABC, ICA, and PSO parameters were conducted, and then, a comparison was done on the obtained results of the predictive hybrid models using several performance indices. As confirmed by the comparative results, the neuro-bee model proposed in this study estimated Q with a higher accuracy than the other hybrid models. The root-mean-square error (RMSE) values of 0.060, 0.076, and 0.094 were obtained for testing the data sets of neuro-bee, neuro-imperialism, and neuro-swarm techniques, respectively. This clearly shows that the newly developed hybrid model was superior to its rival in terms of predicting the ripping production. Furthermore, results of sensitivity analysis showed that weathering zone is the most influential factor on ripping production as compared to other inputs.

In this article, we present an application of metaheuristics optimization approaches to improve medical classifier performance. Genetic Algorithm (GA), Simulated Annealing (SA) and Particle Swarm Optimization (PSO) have been applied in conjunction with Least Square Support Vector Machine (LS-SVM) approach to optimize the total misclassification error in term of False Positive and False Negative rates. We validate our experimental results, based on five well known unbalanced medical datasets. Presented results show that the SA achieved the best results. Both SA and GA outperform PSO metaheuristic.