Jianchao Zeng

Self-Organization particle swarm optimization based on information feedback

The paper develops a self-organization particle swarm optimization (SOPSO) with the aim to alleviate the premature convergence. SOPSO emphasizes the information interactions between the particle-lever and the swarm-lever, and introduce feedback to simulate the function. Through the feedback information, the particles can perceive the swarm-lever state and adopt favorable behavior model to modify their behavior, which not only can modify the exploitation and the exploration of the algorithm adaptively, but also can vary the diversity of the swarm and contribute to a global optimum output in the swarm. Relative experiments have been done; the results show SOPSO performs very well on benchmark problems, and outperforms the basic PSO in search ability.

Adaptive particle swarm optimization with feedback control of diversity

Swarm-diversity is an important factor influencing the global convergence of particle swarm optimization (PSO). In order to overcome the premature convergence, the paper introduced a negative feedback mechanism into particle swarm optimization and developed an adaptive PSO. The improved method takes advantage of the swarm-diversity to control the tuning of the inertia weight (PSO-DCIW), which in turn can adjust the swarm-diversity adaptively and contribute to a successful global search. The proposed PSO-DCIW was applied to some well-known benchmarks and compared with the other notable improved methods for PSO. The relative experimental results show PSO-DCIW is a robust global optimization method for the complex multimodal functions, which can improve the performance of the standard PSO and alleviate the premature convergence validly.

Particle Swarm Optimization with Diversity-Controlled Acceleration Coefficients

In order to overcome the premature convergence, the paper introduced a negative feedback mechanism into particle swarm optimization and developed an adaptive PSO. The improved method takes advantage of the swarm-diversity to control the tuning of the acceleration coefficients (PSO-DCAC). Through the feedback control of the diversity, PSO-DCAC can manipulate the weight of the cognitive part and the social part to fluctuate with the search state, which in turn can adjust the exploration and exploitation adaptively and contributes to a successful global search. The proposed PSO-DCAC was applied to some well-known benchmarks and compared with the other notable improved PSO. Experimental results show diversity-controlled acceleration coefficients is a feasible technique to improve the global performance of PSO and performs very well on the complex optimization problems.

A differential evolutionary particle swarm optimization with controller

To improve the computational efficiency,a new uniform model of particle swarm optimization (PSO) and corresponding algorithm, differential evolutionary PSO (DEPSO), are described, and the convergence is analyzed with transfer function. To enhance the diversity of swarm, PID controller is used to control dynamic evolutionary behavior of DEPSO. Simulation results have proved the algorithms efficiency.

Dynamic circle nonlinear genetic algorithm

Through mechanism analysis of simple genetic algorithm (SGA), every genetic operator is considered as a linear transform. So some disadvantages of SGA is solved if genetic operators are modified to nonlinear transform. According to the above method, the nonlinear genetic algorithm is introduced, and different nonlinear genetic operators with some probability are designed and applied to numerical optimization problems. The optimization computing of some examples is made to show that the new genetic algorithm is useful and simple.

A constraint multi-objective artificial physics optimization algorithm

The use of evolutionary algorithms to solve unconstraint multi-objective problems (MOPs) has attracted much attention recently. However, research on constraint multi-objective algorithms is relatively less. The authors introduce a novel evolutionary paradigm of artificial physics optimization (APO) into constraint multi-objective optimization domain and modify the original mass function and virtual force rules in order to fit constraint multi-objective optimization problems. Moreover the authors present a method of virtual force decreasing to improve the efficiency. Finally, simulation tests show that the algorithm is effective.

A new organizational nonlinear genetic algorithm for numerical optimization

Based on the concept of organization in economics, a novel genetic algorithm, organizational nonlinear genetic algorithm (ONGA), is proposed to solve global numerical optimization problems with continuous variables. In ONGA, genetic operators do not act on individuals directly, but on organizations, and four genetic operators,organization establish, organization classify, multi-parent crossover, and multi-parent mutation operators, are designed for organizations. Simulation results indicate that ONGA performs much better than the real-coded genetic algorithm both in the quality of solution and in the computational complexity.

Distributed control system in water plant based on ControlNet

The paper discusses a kind of multilevel distributed control system in a water plant, which is based on ControlNet, and describes the control system of the filter pool in detail. This control system has been implemented successfully in a water plant.

Practical Stability Analysis for Exponential Type Stochastic Swarms

A novel Lagrangian individual-based isotropic continuous time exponential type stochastic swarming model in an n-dimensional Euclidean space with a family of attrac- tion/repulsion function is proposed in this article. The stability of aggregating behavior of the swarms system are verified by practical stability theoretical analysis and numerical simulation. Practical stability analysis and numerical simulations results further indicate that the individual members living in group during the course of coordinative motion can realize the mutual aggregating behavior, the motion of each individual member is a combination of the inter-individual interactions, meanwhile, which are also presented to demonstrate the effectiveness of our model. The attraction/repulsion function is odd, so the attractive force and repulsion force taking effect in opposite direction that leads to aggregation behavior.

Design of the agent-based genetic algorithm

In the standard GA, the individual has no intelligence and must act upon some rules established by a programmer in advance, such as various genetic operator. The result is to make the evolutionary process to be trapped into the local optimization of the objective function. In order to solve this problem, through studying the structure of an agent and selection operator, the paper designs a new genetic algorithm based on agent, called AGA (Agent-based Genetic Algorithm). At the premise of giving the definition of the outer environment where an agent lives and of an agents belief, this paper gives some rules on how an agent selects one agent to cross their genes and some rules on how to solve competition. In addition, a communication method based on blackboard is presented to solve the communication among the agent society. Finally, the paper gives the structure of AGA and the simulation result for a multi-peak function, which demonstrates the validity of the AGA.