Wang Wan-liang

Soft-sensing method for wastewater treatment based on BP neural network

At present, wastewater treatment quality parameters cannot be detected on-line. In this paper, the soft-sensing method based on artificial neural networks is proposed in order to resolve this problem. The wastewater treatment technique is analyzed systematically. ORP, DO, PH and MLSS which can be detected on-line are taken as the secondary variables. BOD, COD, N and P which cannot be detected on-line are taken as the primary variables. The BP neural network for soft-sensing is proposed and trained using the testing data of the practical treatment process. The simulation results show that the soft-sensing system of wastewater treatment based on the BP neural network can correctly estimate the quality parameters in real time. Thus, the system can be accommodated to the changes of the environment and implement the real time control of wastewater treatment.

Hopfield neural networks approach for job shop scheduling problems

A new method based on Hopfield neural networks for solving job-shop scheduling problems (JSP) is proposed. All constraints of job-shop scheduling problems and its permutation matrix express are developed. A new calculation energy function included all constraints of job-shop scheduling problems is given. A corresponding new Hopfield neural network construction and its weights of job-shop scheduling problems are given. To avoid Hopfield neural network to converge to local minimum volume, and to produce some non-feasible scheduling solutions for JSP, simulated annealing algorithm is applied to Hopfield neural network. Hopfield neural network converging to minimum volume 0, can keep the steady outputs of neural networks as feasible solution for job-shop scheduling problem. This paper improved existing method based on Hopfield neural network for solving job-shop scheduling problems. Compared with the method, modified method can keep the steady outputs of neural networks as feasible solutions for job-shop scheduling problems.

Application of adaptive fuzzy logic system to model for greenhouse climate

In this paper, the greenhouse climate model based on adaptive fuzzy logic system is presented. Greenhouse climate system is a nonlinear system with the various climate factors being coupled. Due to its capability to handle both numerical data and linguistic information, it is feasible to apply adaptive fuzzy logic system to model for greenhouse climate, and then provide prediction for greenhouse climate control.

An improved evaluation method for interactive genetic algorithms and its application in product design

Regarding the fatigue problem in the interactive genetic algorithms (IGA) during the users evaluation process, the paper proposed a Boolean evaluation method, which replaces users grading work with selection. The new method reduces users labor considerably, and makes it possible to construct users image model (by means of user feature curves) through their selections, from which users sensitivity to genes can also be calculated in quantified way. User image helps to estimate individuals sufficiency. User sensitivity parameters help to arrange the searching order while layered optimizing method is adopted, and can also decide which dimensions of the searching space can be neglected according to the preset threshold value. The sufficiency values deduction method from selection information is given, together with the general skills for the recognizing and handling of coupling genes. The algorithms are tested in a product styling design example.

Hopfield neural networks optimization method for VP routing in ATM network

To improve ATM network survivability level, a discrete Hopfield neural network algorithm with transient chaos for VP routing in ATM network was put forward. The computational energy function of VP routing and the motion equation of the discrete Hopfield neural network with transient chaos were also given. Compared with the VP routing algorithms exited, the simulation results show that it is both faster and more effective to achieve VP routing.

Research on unidirectional link routing algorithm for Mobile Ad Hoc Networks

Protocol with QoS guarantee in unidirectional MANET is very important. Many of the classical routing protocols are put forward on the basis of two-way communication. While there exits unidirectional link because of differences in transmission power in each terminals as well as environmental factors such as topography and so on. This paper studies and applies the shortest path method based on graph theory and advances a new routing protocol with QoS guarantee in unidirectional MANET. In order to analyze its performance, adopt MANET to simulate the new routing model and compare it with DSR.

Study on optimal scheduling model and technology based on RPSO for small hydropower sustainability

In order to reduce the local environment impact, the improved methods of optimal scheduling small hydropower are discussed and the sustainability optimal scheduling models which composed of optimal objectives and constraints are presented. The controlled reservoir releases for hydropower generation which based on float coding is used as the decision variable of the model and the reserved minimum flow and reservoir water level as special constraints. The controlled reservoir water level compared with maximal energy output during the period of dispatching is considered as objective. An improved PSO, resilient particle swarm optimization (RPSO), is adopted to the optimization. In RPSO, the velocity of a particle is not dependent on the size of distance between the individual and the optimal particle but only dependent on its direction. An adaptive scheme is adopted to adjust the magnitude of the velocity resiliently. A simulation operation of RPSO Matlab program for a factual small hydropower plant show that the sustainability optimum scheduling models and the technology based RPSO is viable and efficient. RPSO performance was compared with PSO and APSO on the same case, the results show that RPSO performed better.

A new method of networked video surveillance with privacy area protection

Privacy area protection has become more important than before in networked video surveillance. In this paper, we address privacy area protection related issues. First, the characteristics of security camera with pan/tilt rotation mechanism are introduced and the privacy problems of video surveillance in public areas are analyzed. Then, a new method of protecting privacy areas is proposed to automatically adapt the panning rotation and tilting rotation of security camera. This method contains two steps to calculate the masked area matrix from predetermined privacy area matrix and some static/dynamic basic parameters. Experimental results have shown that the proposed method could work well in real-systems and protect the privacy areas efficiently.

Fractal-agent based Web access control of enterprise cooperative manufacturing system

The cooperation mode based on fractal-agent of the networked cooperative manufacturing system is introduced. According to the task executing principle of the fractal-agent, the dynamic task is bound to the corresponding dynamic role. It leads to raise the fractal-agent based Web access control approach, security policy and the relating access control matrix. Furthermore, the approach is carried out to enable the secure access, administration and control of the cooperative manufacturing system. The experimental results show that the research can meet the requirement of the Web access security of the enterprise cooperative manufacturing system.

A heuristic algorithm for minimizing the range of lateness and makespan on non-identical multiprocessors

A generalized multi-objective model is presented for minimizing the range of lateness and makespan on a multiprocessors job scheduling problem. It is proved that this job scheduling problem is an NP complete problem. So a heuristic algorithm is given to solve the scheduling problem. The numerical example illustrates that the heuristic algorithm is designed very well and can find very good approximate optimal solution.