Papers by Author: Jian Chao Zeng

Abstract: BP neural network for failure pattern recognition has been used in hydraulic system fault diagnosis.However, its convergence rate is relatively small and always trapped at the local minima. So a new modified PSO-BP hydraulic system fault diagnosis method was proposed,which combined the respective advantages of particle swarm algorithm and BP algorithm. Firstly, the inertia weight and learning factor of the standard particle swarm algorithm was improved, then BP neural network’s weights and thresholds were optimized by modified PSO algorithm. BP network performance was ameliorated. The simulation results showed that this method improved the convergence rate of the BP network, and it could reduce the diagnostic errors.

Abstract: For the complex environment and multi influencing factor in coalface, Wireless Sensor Networks (WSN) is a effective solutions for the coal face of environmental monitoring and miner positioning. Based on the environmental features of the coalface, a WSN deployment paradigm is proposed in this paper, and the DV-HOP localization algorithm is determined to be the miner location solution. Due to the low accuracy of DV-HOP algorithm on condition of the uneven structure of the coalface WSN and the big distance gap between adjacent node, a RSSI-based weight DV-HOP algorithm is proposed. In this algorithm the hops of adjacent node is weighted by the RSS of flooding packets transmitted between adjacent nodes, which associated the hops with the distance of adjacent nodes. Simulation results show that under the same experimental the accuracy of our algorithm is better than the DV-HOP algorithm, and can meet the requirements of localization on the coalface environment.

Abstract: Detecting and characterizing the community structure of complex network is fundamental. We compare the classical optimization indexes of modularity and modularity density, which are quality indexes for a partition of a network into communities. Based on this, we propose a quantitative function for community partition, named communitarity or C value. We demonstrate that the quantitative is superior to modularity Q and modularity density D. Both theoretical and numerical results show that optimizing the new index not only can resolve small modules, but also can correctly identify the number of communities.

Abstract: The fast location is especially important, in the license plate recognition. In this paper, we received the candidate regions by the mathematical morphological operation, and then scanned the middle line of the binarized candidate areas. Determine the plate region using the black-white variations of the pixel in this middle line. This method is based on single line scanning which can reduce the computation time and improve the speed of license plate location.

Abstract: This paper studied the problem of deadlock performance analysis for collaborative design process model based on extended objective Petri Net (OPN). Firstly, the definition of CSCD extended OPN model based on decision rules has been proposed by analyzing the collectivity requirement for the machinery products collaborative design system to be built. Besides, models for the OPN model of process controlling nets and design unit objective class. Then OPN invariant analysis theory has been adopted to analyze the dynamic performance of deadlock to the part design unit by building its object communication net. The dynamic performance of the whole model could be obtained by analyzing the performance of all the design unit object class.

Abstract: The routing protocol which can meet the requirement of underground wireless sensor network plays a decisive role in improving the monitoring quality and the survival time of network. Because of the different energy supplement method of anchor nodes and unknown nodes, these two types of nodes are different energy levels, which determines that the task of the data forwarding should be completed by the blind node as far as possible. By combining the position information and cluster routing technology, a position-based cluster routing (PBCR) protocol is proposed. In the protocol the tunnel is divided into a number of regions and the nodes within the region form a cluster. The protocol chooses the blind node nearest to the base station in the region as cluster head and builds up the chain between the cluster heads as the network data transmission backbone which can help achieve the high-speed data transmission. Simulation results show that the PBCR protocol can ensure a balanced use of node energy and prolong the life cycle of the network.

Abstract: Crank block steering mechanism optimization, which is a nonlinear constrained optimization problem, is important for forklift truck to get preferable steering performance. Vector particle swarm optimization (VPSO) is a swarm intelligent algorithm proposed to solve constrained optimization problems with simple concept and easy implementation. Therefore, VPSO was proposed to optimize crank block steering mechanism of forklift truck in this paper, which will be intended for minimizing the maximal error of outer wheel steering angle, maximizing the minimal transmission angle and making the transmission capability of the force better. The experimental results obtained by VPSO are compared with the results obtained by enumeration algorithm which are reliable for optimization problems. The comparison results showed that vector particle swarm optimization can get same optimal results as enumeration algorithm in much less calculation times.

Abstract: Complex manufacture collaborative design process is a dynamic process with uncertainty, which is a process with multi-task executing in a parallel manner. It has obviously time constraint, which has been few studied up to now. Temporary inference and simulation problem for complex manufacture design process are mainly studied in this paper. Firstly, a Petri net model for collaborative design process is proposed based on time Petri net theory and four temporary inference algorithms of ordering, paralleling, selecting and circling structures in collaborative design process. Secondly, the authors design and develop a visually modeling and simulating circumstance for collaborative design process temporary inference based on time Petri net. Temporary inference for the whole collaborative design process and optimization to design task can be implemented by this simulation software. Finally, collaborative design process for driver of chain-transportation-driven equipment is undertaken to shown the efficiency of the proposed method. Research in this paper affords a determinant principle for time arrangement of collaborative design task and it also affords schema scheduling.