Papers by Author: Zheng Yao

Abstract: Fault diagnosis has been the research hotspot in the industry fields, but, with the gradual complication in modern industry equipments and systems, it is more hard to quickly diagnose complicated or exceptional faults. For overcoming the diagnosis weakness of traditional fault diagnosis methods in the rotating machinery, this paper presents a hybrid method that combines the wavelet with neural networks theory. Both the blindness of framework designs for BP neural networks and the problem of nonlinear optimizations were solved and this method was used in rotating machinery fault diagnosis. The research shows that this method is feasible and effective and can be applied to the other rotating machinery fault diagnosis.

Abstract: Energy consumption is a critical problem in operation of wireless sensor networks. For the sake of avoiding the data abundance and balancing the energy consumption in wireless sensor networks, this paper makes a research on network nodes Optimization in wireless sensor network based on ant colony algorithm and WIA-PA protocol stack. The novel design improved on hardware and software to control consumption of the energy and used transition probability of ant colony algorithm from one node to the other to calculate and determine the optimal path of network node in traversal of these locations. The results of the examples show that this method has lower energy consumption, computational briefness and higher positioning accuracy; it can not easily run into the local optimum, and also be applied to other tracking of complex network systems.

Abstract: In order to more rapidly and accurately diagnose the fault, the CAN bus was applied to system. In this article, an intelligent fault diagnosis method is presented for the diversity, uncertainty and complexity of device faults. The system can directly access self-test information and on-line test information, and information fusion technology, stator current and rotor vibration signals as a diagnostic characteristics input signal were introduced into the motor fault diagnosis. In accordance with the appropriate information, monitoring signs separately adopt a diagnosis sub-network to complete different aspects of fault diagnosis; diagnostic reasoning and drawing conclusions of fault diagnosis. Experiments show that this method is simple and effective. It can also be applied to other fault diagnosis of complex systems and has certain portability.

Abstract: The complexity of modern equipment is higher than before, thus it cannot meet the need of speedy fault diagnosis to use single diagnosis pattern. According to these troubles, a fault tree was introduced to analyze fault information of gear. Fault tree analysis is a logical and diagrammatic method to evaluate the probability of an accident resulting from sequences and combinations of faults and failure events. In conventional fault tree analysis, probabilities and consequences are treated as exact values. In many engineering applications, however, it is difficult to evaluate the probabilities and consequences from past experiences, so a fuzzy set defined in probability is proposed. The principle of fuzzy diagnosis is taken as basics and symptom sets and faults sets are extracted through fault tree analysis. Experiments show this method is effective for fault diagnosis of gear.

Abstract: The on-line fault diagnostics technology for machines is fast emerging for the detection of incipient faults as to avoid the unexpected failure. On the basis of fault diagnosis theory and method, this paper presents a applications of techniques for fault detection and classification in rotating machinery based on fuzzy theory and neural network theory, the basic structure and working principle of the fault intelligent diagnosis system are introduced, the knowledge stored in the neuron-fuzzy system has been extracted by a fuzzy rule set with an acceptable degree of interpretability, the model of fuzzy fault diagnosis and the self-study principle are described. The practice proves that this is an effective method of large-scale and complicated electronic equipment, and it can also be applied to other fault diagnosis of complex systems and has certain portability.

Abstract: Profibus and CAN bus are two kinds of widely used field bus in the industrial control field. However, this two bus protocol conversion devices are not introduced on the market, it can not meet the needs of users. Therefore, the design of a Profibus and CAN protocol translation ngateway is very important. By comparing the OSI reference model, the same points of Profibus and CAN bus can be found, their communication model structure is streamlined by the OSI model, they are only three layers, which makes it possible to interconnect them. This paper designs an embedded protocol conversion gateway, S3C2410 processor is used as main controller, SPC3 and SJA1000 are selected as Profibus and CAN protocol chips. The design has realized Profibus and CAN bus protocol conversion, using this gateway in industrial field can bring certain economic benefits for the enterprise.