Fengqiang Zhao

A human-computer cooperative particle swarm optimization based immune algorithm for layout design

Packing and layout problems have wide applications in engineering practice. However, they belong to NP (non-deterministic polynomial)-complete problems. In this paper, we introduce human intelligence into the computational intelligent algorithms, namely particle swarm optimization (PSO) and immune algorithms (IA). A novel human-computer cooperative PSO-based immune algorithm (HCPSO-IA) is proposed, in which the initial population consists of the initial artificial individuals supplied by human and the initial algorithm individuals are generated by a chaotic strategy. Some new artificial individuals are introduced to replace the inferior individuals of the population. HCPSO-IA benefits by giving free rein to the talents of designers and computers, and contributes to solving complex layout design problems. The experimental results illustrate that the proposed algorithm is feasible and effective.

Swarm-based intelligent optimization approach for layout problem

Layout problem is a kind of NP-Complete problem. It is concerned more and more in recent years and arises in a variety of application fields such as the layout design of spacecraft modules, plant equipment, platforms of marine drilling well, shipping, vehicle and robots. The algorithms based on swarm intelligence are considered powerful tools for solving this kind of problems. While usually swarm intelligence algorithms also have several disadvantages, including premature and slow convergence. Aiming at solving engineering complex layout problems satisfactorily, a new improved swarm-based intelligent optimization algorithm is presented on the basis of parallel genetic algorithms. In proposed approach, chaos initialization and multi-subpopulation evolution strategy based on improved adaptive crossover and mutation are adopted. The proposed interpolating rank-based selection with pressure is adaptive with evolution process. That is to say, it can avoid early premature as well as benefit speeding up convergence of later period effectively. And more importantly, proposed PSO update operators based on different versions PSO are introduced into presented algorithm. It can take full advantage of the outstanding convergence characteristic of particle swarm optimization (PSO) and improve the global performance of the proposed algorithm. An example originated from layout of printed circuit boards (PCB) and plant equipment shows the feasibility and effectiveness of presented algorithm.

Parallel Adaptive Artificial Fish Swarm Algorithm Based on Differential Evolution

Artificial fish swarm algorithm (AFSA) is a newly proposed swarm intelligent optimization algorithm. It is proved to be a promising approach to complex engineering problems, yet still there exist some defects of this algorithm. To solve the problem that AFSA has a low rate of convenience, low optimization precision, premature convergence and poor ability of balancing exploitation and exploration, an improved artificial fish swarm algorithm (PAAFSA-DE) is proposed. This algorithm divides the population into two sub groups with the same size, and different adaptive strategies are applied to the two groups respectively to make one group focus on global search and the other on local search. The two sub populations evolve independently and individual migration are conducted regularly to achieve information communication, increase the population diversity and improve convergence rate of algorithm. When the information on the bulletin board does not change for a certain times, the differential evolution strategy will be introduced to make the algorithm escape from local extreme. The comparing simulation results on the benchmark function optimization problems demonstrate that the improved algorithm is feasible and effective. It performs better than basic AFSA, the balance ability of exploitation and exploration is enhanced, and convergence efficiency and optimization precision are improved greatly as well as the stability is strengthened.

A novel genetic algorithm based on immunity and its application

In this paper, a novel genetic algorithm based on immunity (GABI) on the basis of parallel genetic algorithms (PGA) is proposed in order to overcome some defects of them, such as premature and slow convergence rate. The global performance of the algorithm is improved by introducing immunity theory into PGA. This is revealed in the following two aspects. One is that the immune selection based on proposed adjustable geometric-progression rank-based selection can prevent the algorithm from premature. The other is that convergence rate can be accelerate by individual migration strategy between subpopulations based on immune memory mechanism. In this algorithm, the idea of multiple subpopulations evolution based on improved adaptive crossover and mutation is adopted. To be hybridized with the Powell method can further improve local searching performance of the algorithm. An example of layout design shows that GABI is feasible and effective.

A novel improved hybrid particle swarm optimisation based genetic algorithm for the solution to layout problems

Layout problems belong to NP(non-deterministic polynomial)-Complete problems theoretically. They are paid more and more attention in recent years and arise in a variety of application fields such as the layout design of spacecraft modules, shipping, vehicle and robots, plant equipments, platforms of marine drilling well. The algorithms based on swarm intelligence are relatively effective to solve this kind of problems. But usually there still exist two main defects of them, i.e. premature convergence and slow convergence rate. To overcome them, a novel improved hybrid PSO-based genetic algorithm (HPSO-GA) is proposed on the basis of parallel genetic algorithms (PGA). In this algorithm, chaos initialization and multi-subpopulation evolution based on improved adaptive crossover and mutation are adopted. And more importantly, in accordance with characteristics of different classes of subpopulations, different modes of PSO update operator are introduced. It aims at making full use of the fast convergence property of particle swarm optimization (PSO). The proposed adjustable arithmetic-progression rank-based selection can prevent the algorithm from premature in the early stage and benefit accelerating convergence in the late stage as well. An example of layout problems shows that HPSO-GA is feasible and effective.

A New Hybrid PSO-based Genetic Algorithm and its Application to LayoutProblems

Layout problems belong to NP-Complete problems theoretically. They are concerned more and more in recent years and arise in a variety of application fields such as the layout design of spacecraft modules, plant equipments, plat- forms of marine drilling well, shipping, vehicle and robots. The algorithms based on swarm intelligence are relatively ef- fective to solve these kind of problems. But usually there still exist two main defects, i.e. premature convergence and slow convergence rate. To overcome them, a new improved hybrid PSO-based genetic algorithm (HPSO-GA) is proposed on the basis of parallel genetic algorithms (PGA). In this algorithm, chaos initialization, hybrid strategy and multi- subpopulation evolution based on improved adaptive crossover and mutation are adopted. The proposed interpolating rank-based selection with pressure can prevent the algorithm from premature in the early stage and benefit accelerating convergence in the late stage as well. And more importantly, in accordance with characteristics of different classes of sub- populations, different modes of PSO update operator are introduced. It aims at making full use of the fast convergence property of particle swarm optimization (PSO). An example of layout problems shows that HPSO-GA is feasible and ef- fective.

Research on Diesel Engine Piston Wear Fault Diagnosis Method Based on the Local Wave Time-frequency Analysis

A large number of non-stationary dynamic signals are generated in the working machinery and equipment. Es- pecially, diesel engines often encounter non-stationary transient and time-varying modulation signals, such as the impulse response signals caused by cylinder piston wear. These kinds of signals generated from diesel engine are analyzed by the method of the Local Wave time-frequency proposed in this paper, and then according to the analysis to diagnose the working state of the diesel engine. It proved that the proposed method is feasible and effective. Moreover, it provides an effective way for the diesel engine condition monitor and fault diagnosis.

An improved parallel genetic algorithm based on particle swarm optimization and its application to packing layout problems

Packing layout problems belong to NP-Complete problems theoretically. They are concerned more and more in recent years and arise in a variety of application fields such as the layout design of spacecraft modules, plant equipments, platforms of marine drilling well, shipping, vehicle and robots. The algorithms based on swarm intelligence are relatively effective to solve this kind of problems. But usually there still exist two main defects of them, i.e. premature convergence and slow convergence rate. To overcome them, an improved parallel genetic algorithm based on particle swarm optimization (PSO-PGA) is proposed on the basis of traditional parallel genetic algorithms (PGA). In this algorithm, parallel evolution of multiple subpopulations based on improved adaptive crossover and mutation is adopted. And more importantly, in accordance with characteristics of different classes of subpopulations, different modes of PSO update operators are introduced. It aims at making full use of the fast convergence property of particle swarm optimization (PSO). The proposed arithmetic-progression rank-based selection with pressure can prevent the algorithm from premature in the early stage and benefit accelerating convergence in the late stage as well. An example of packing layout problems shows the proposed PSO-PGA is feasible and effective.

A parallel hybrid immune genetic algorithm and its application

We propose a parallel hybrid immune genetic algorithm (PHIGA) based on parallel genetic algorithms (PGA) in order to overcome some defects of them, such as premature and slow convergence rate. The global performance of the algorithm is improved by introducing immunity theory into PGA. This is revealed in the following two aspects. One is that immune selection can prevent the algorithm from premature. The other is that convergence rate can be accelerate by individual migration strategy between subpopulations based on immune memory mechanism. In this algorithm, chaos initialization and multiple subpopulations evolution based on improved adaptive crossover and mutation are adopted. To be hybridized with the complex method can further improve local searching performance of the algorithm. An example of layout design shows that PHIGA is feasible and effective.