Cai Guoqiang

Multi-objective fuzzy modeling using NSGA-II

An approach to construct multiple Pareto-optimal fuzzy systems based on NSGA-II is proposed in this paper. First, in order to obtain a good initial fuzzy system, a modified fuzzy clustering algorithm is used to identify the antecedents of fuzzy system, while the consequents are designed separately to reduce computational burden. Second, a Pareto multi-objective genetic algorithm based on NSGA-II and the interpretability-driven simplification techniques are used to evolve the initial fuzzy system iteratively with three objectives: the precision performance, the number of fuzzy rules and the number of fuzzy sets. Resultantly, multiple Pareto- optimal fuzzy systems are obtained. The proposed approach is applied to two benchmark problems, and the results show its validity.

Monte-Carlo simulation based on FTA in reliability analysis of door system

the paper proposes a reliability analysis method based on Monte-Carlo simulation and fault-tree analysis method (FTA), with which the reliability of door system of metro vehicles in the urban rail is analysis. Given the states transfer rules, the simulation model of the door system based on FTA is built. With the model we make the reliability simulation based on Monte-Carlo method. The calculated results contain the failure rate of the top event and importance degrees of basic events. Then the dynamic simulation of the system reliability is made and the distribution of reliability is got. The simulation method proposed in the paper is more modular and convenient for application.

Approach and Application of Transportation State Analysis

Urban traffic state analysis plays an important role in the solution of traffic congestion problem. To identify traffic state effectively is a foundational work for improving traffic condition and preventing traffic congestion. In this paper, an index system of traffic state was established, it through analysis of the multi-source and heterogeneous data from the traffic network, and for the demand of complex structure of large-scale road networks. The proposed index system includes micro-layer, which describes flow tendency on links or nodes, and meso-layer, which represents link-to-link or given region flow condition and diversification, and macro-layer, which indicates more large-scale than meso-layer region flow status and variety. A new method of estimating level of service (LOS) of link, node and network is developed to identify traffic state. A numerical example is studied in Stabilized-door bridge to Bell-tower bridge urban traffic. The proposed index system is employed in evaluating traffic level of service to get traffic flow condition and variety. The evaluation results show that level of service can truly reflect the traffic state. Accordingly, the truly traffic status testified the effective for the proposed evaluation method in traffic level of service.

Modeling and Application of Urban Dynamic Region Traffic Model Based on Information Fusion

Existing urban transportation network models or travel choice models focus on network application. While, region traffic situation modeling are becoming crucial problem for urban traffic management and traffic flow control. A novel dynamic region traffic model is proposed in this paper. The region traffic mathematical model is classified as macroscopic model, mesoscopic model and microscopic model. Macroscopic model represents region traffic state, which is appropriate for large area analysis. Mesoscopic model is capable for the corresponding transportation network problem. Microscopic model is suit for vehicle cluster behavior. Considering probability and trends, the information fusion techniques are used to process historical data. The dynamic region traffic model can be fit for the long term and short term traffic situation estimate. Application results show these contributions.

A Novel Evaluate Method of Urban Region Transportation Level of Service

Congestion in urban traffic is critical points in the all problems of traffic. Foundational work for improving network traffic condition and preventing traffic congestion is to identify traffic network states effectively. In this paper, a novel evaluate method of urban region transportation level of service is proposed based on the attribute mathematics. The approach proposed in this paper is convenient and simple for application. The illustrative example shows the effectiveness of the proposed approach. Especially, this approach has achieved quite successful application to the practical problem of estimation the level of service.

Research on Computer Aided System Design Method

Computer aided design technologies is the hot topic applied in engineering design in order to create better complex products. Information fusion modeling forms the basis of manufacturing and functional design because integrated information can solve the constraints problem comes from different kinds of people with different knowledge into a design team. This paper propose a architecture model based on information fusion whose essence is to turn the isolated activities, design and manufacture, design and assembly to an integrated collaborate process managing product life cycle information.

Adaptive control of response in a rotor with loose foundation

Research on the loosing rotor system, a nonlinear dynamic equation of a rotor with loosing foundation is discussed. Then, we studied the problem of controlling the multi-cycles response of the nonlinear rotor system by the adaptive control theory. Numerical simulations on the rotor system are performed with a multi-parameter adaptive algorithm. The numerical results show that the parametric adaptive control method is effective to suppress a multi-cycles system response once happened into a required steady State periodic response. This control strategy is helpful to rotor system design and maintenance.Research on the loosing rotor system, a nonlinear dynamic equation of a rotor with loosing foundation is discussed. Then, we studied the problem of controlling the multi-cycles response of the nonlinear rotor system by the adaptive control theory. Numerical simulations on the rotor system are performed with a multi-parameter adaptive algorithm. The numerical results show that the parametric adaptive control method is effective to suppress a multi-cycles system response once happened into a required steady State periodic response. This control strategy is helpful to rotor system design and maintenance.

Quantity Modeling and Application of Multivariable Correlation Analysis

Abstract-This study focuses on quantitative correlation problem of four Railway Parcel traffic parameters: Number of Initial trains (NIT), GDP of cities, Number of Parcel Traffic Agencies (NPTA) and Number of Parcel traffic Nodes (NPTN). It can be seen as a multivariable systems that called Multiple-Input Single-Output(MISO). Then ANN is used in to resolve the multivariable Correlation Analysis problems in China Railway Parcel forecast. Based on Artificial Neural Networks (ANN), the prediction of China Railway Parcel Traffic Volume is modeling. The model can effectively solve the variable multiple correlation problem. Good performance is demonstrated when Application proves the accuracy of the model and its contribution.