Non-dominated Multi-objective Genetic Algorithm Research Articles

Multi-objective optimization of elliptical tube fin heat exchangers based on neural networks and genetic algorithm

The application of machine learning based on neural networks (NNs) and genetic algorithm (GA) in multi-objective optimization of heat exchangers is studied. Taking the tube fin heat exchanger (TFHE) as the research object, the inlet air velocity and the ellipticity of tubes are taken as the optimization variables. In order to obtain the optimal heat transfer performance and pressure drop performance, Computational Fluid Dynamics (CFD) simulation is carried out for different Reynolds based on the hydraulic diameter numbers (150–750) and tube ellipticity (0.2–1). Then use simulation data to train the Back-Propagation neural networks and establish the prediction model of heat transfer coefficient and pressure drop. The non-dominated multi-objective genetic algorithm with elitist retention strategy (NSGA-II) is used to optimize two prediction results of NNs. Finally, the optimal heat transfer coefficient and pressure drop are given in the form of Pareto front. The optimization results show that when the Reynolds number is 541 and the ellipticity is 0.34, the pressure drop of the TFHE decreases 20%, and the heat transfer coefficient is basically unchanged, whose j/f is 1.28 times as much as that of the original heat exchanger.

Design of a Parallel Gripper Based on Topology Synthesis and Evolutionary Optimization

Abstract A compliant gripper with nearly parallel gripping motion is developed by a topology synthesis and a dimensional synthesis approach. The topology synthesis process can generate linkage-type compliant mechanisms. Suitable boundary conditions of the topology synthesis process are selected to achieve the desired functions of the device. The dimensional synthesis is based on an evolutionary optimal design process. To meet various design goals, a nondominated multi-objective genetic algorithm is selected for the optimal design process. A kinetostaic model based on the chained beam constraint model is developed for force–displacement analysis of the designs. Efficiency and accuracy of the design approach are proved by experiments. Appropriate linkage types of compliant mechanisms may be discovered by the topology optimization process before moving on to the dimensional synthesis to obtain final designs.

Efficient Non-Dominated Multi-Objective Genetic Algorithm (NDMGA) and network security policy enforcement for Policy Space Analysis (PSA)

Computer network security is the first line of protection for guaranteeing information reliability. A computer network can be threatened when it does not have a well-designed or properly implemented network security policy. The major issue is that network operators cannot always support their network security strategy. Network operators often depend on security facilities to defend their Information Technology (IT) organizations. Different forms of security policies concerned with the network are distinct and circulated among various security middleboxes (MBs) organized into networks. Conversely, due to the inherent sophistication of security policies, it is not appropriate to use recent path-wise implementation algorithms. The present study solves the problem of policy implementation with MBs relatively faster compared to existing methods. In particular, this research models security policy implementation as a Weighted K Set Covering Problem (WKSCP) and employs a computational-geometry-based Policy Space Analysis (PSA) tool for a set of procedures as the security mechanism. Using the PSA tool leads to a Non-Dominated Multi-Objective Genetic Algorithm (NDMGA), which reveals the inherent complexities involved in a security policy and informs implementation of the policy enforcement algorithm. The resolution to this issue is to determine a collection of enforcement network nodes for an MB to execute on-datapath examination with globally distinct security policies. Therefore, the chosen policy enforcement nodes should be initially positioned on paths that are related to security strategies. Compared to the scope-wise strategy enforcement mechanism, where the procedures are executed over the objects of the network path, the scheme in this present work is implemented over the objects pertaining to the network nodes. A secure node may contain multiple paths, thus storing a large possible flow space that preserves the distribution and circumvents the ineffective usage of network bandwidth due to random traffic steering. Compared to current methodologies, the suggested NDMGA-based PSA tool yielded a 95% decrease in policy implemented nodes and a 93% decline in implemented rules.

Contribution au dialogue entre la psychanalyse, la psychiatrie et les neurosciences dans la recherche clinique actuelle individuelle et sociale, par l’étude des phénomènes d’automatisme

Computer network security is the first line of protection for guaranteeing information reliability. A computer network can be threatened when it does not have a well-designed or properly implemented network security policy. The major issue is that network operators cannot always support their network security strategy. Network operators often depend on security facilities to defend their Information Technology (IT) organizations. Different forms of security policies concerned with the network are distinct and circulated among various security middleboxes (MBs) organized into networks. Conversely, due to the inherent sophistication of security policies, it is not appropriate to use recent path-wise implementation algorithms. The present study solves the problem of policy implementation with MBs relatively faster compared to existing methods. In particular, this research models security policy implementation as a Weighted K Set Covering Problem (WKSCP) and employs a computational-geometry-based Policy Space Analysis (PSA) tool for a set of procedures as the security mechanism. Using the PSA tool leads to a Non-Dominated Multi-Objective Genetic Algorithm (NDMGA), which reveals the inherent complexities involved in a security policy and informs implementation of the policy enforcement algorithm. The resolution to this issue is to determine a collection of enforcement network nodes for an MB to execute on-datapath examination with globally distinct security policies. Therefore, the chosen policy enforcement nodes should be initially positioned on paths that are related to security strategies. Compared to the scope-wise strategy enforcement mechanism, where the procedures are executed over the objects of the network path, the scheme in this present work is implemented over the objects pertaining to the network nodes. A secure node may contain multiple paths, thus storing a large possible flow space that preserves the distribution and circumvents the ineffective usage of network bandwidth due to random traffic steering. Compared to current methodologies, the suggested NDMGA-based PSA tool yielded a 95% decrease in policy implemented nodes and a 93% decline in implemented rules.

Intelligent Residential Energy Management in Smart Grid

Objectives: Providing efficient energy management for smart home through appropriate scheduling of household appliances is addressed in this paper, with two objectives namely, electricity peak demand minimization and electricity cost minimization. Methods/Statistical Analysis: The residential load scheduling problem requires the prior knowledge about the residential electricity demand and electricity price information, for scheduling the appliances. Two different algorithms namely, Discrete Non-dominated Multi-objective Particle Swarm Optimization (DNMPSO) algorithm and Manhattan distance based Non-dominated Multi-objective Genetic Algorithm (MNMGA) are proposed to solve the problem in this paper. Findings: Both the algorithms were simulated in order to evaluate their performance. The Peak-To-Average Ratio value is used as the measure to assessing the peak load in electricity demand. Based on the results, it is observed that the DNMPSO algorithm obtains significant cost reduction with the acceptable Peak-To-Average Ratio value than the MNMGA algorithm. In addition to that, the DNMPSO algorithm provides better diversity than the MNMGA algorithm. Application/Improvements: The proposed algorithm can be used to enhance the energy management for residential load in smart grid. Further, the cost efficiency can be improved by incorporating renewable energy resource.

Fuzzy Control of Base-Isolation System Using Multi-Objective Genetic Algorithm

Abstract: In this study, a friction pendulum system (FPS) and a magnetorheological (MR) damper are employed as the isolator and supplemental damping device, respectively, of a smart base-isolation system. Neuro-fuzzy models are used to represent dynamic behavior of the MR damper and FPS. A fuzzy logic controller (FLC) is used to modulate the MR damper so as to minimize structural acceleration while maintaining acceptable base displacement levels. To this end, a multi-objective optimization scheme that uses a nondominated multi-objective genetic algorithm (NSGA-II) is used to optimize parameters of membership functions and find appropriate fuzzy rules. To demonstrate the effectiveness of the proposed multi-objective genetic algorithm for FLC, a numerical study of a smart base-isolation system is conducted using several historical earthquakes. It is shown that the proposed method can find optimal fuzzy rules and that the NSGA-II-optimized FLC outperforms not only a passive control strategy but also a human-designed FLC and a conventional semiactive control algorithm.