Xiao Long Wang

Fault Diagnosis of PT Fuel System Based on Particle Swarm Optimization-Support Vector Machine

In order to overcome the difficulty in selecting parameters of support vector machine (SVM) when modeling the PT fuel system fault diagnosis, SVM optimized by particle swarm optimization (PSO) algorithm was proposed. The PSO-SVM model was established and the fault multi-classifiers of the SVM were got. The pressure signal of the PT fuel inlet and outlet at different rotational speed and conditions was collected. The algorithm of PSO-SVM was used to train and recognize the pressure signal. The result of experiment confirms the validity of this method through comparison of the BP-NN, SVM and the PSO-SVM.

Tribological properties of laser surface cladding Ni-base alloy coatings under dry friction and seawater conditions

Ni-base alloy coatings were prepared on aluminium alloy 7005 by laser surface cladding (LSC). The tribological properties of the coatings under dry friction and seawater conditions were investigated. Results showed that the coatings contained the phases of NiAl, NiAl3, Al3Ni2 and a few Cr23C6 and CrB. Compared with atmosphere plasma spraying (APS) Ni-base alloy coatings, LSC Ni-base alloy coatings exhibited higher microhardness, interfacial adhesion and cohesion force, and smaller friction coefficients and wear losses. The wear mechanisms of LSC Ni-base alloy coatings changed from micro-cutting wear into multi-plastic deformation wear, adhesive wear and oxidative wear as loads increased in dry friction. In seawater environment, LSC Ni-base coatings experienced micro-cutting wear and pitting corrosion wear at the loads of 3∼6 N, and multi-plastic deformation wear and local fracture wear at the loads higher than 9 N.

Research of Equipment Precise Support System Based on Internet of Things

In this paper, conception, structure, key technologies of Internet of Things are described. According to the characteristics and trend of equipment support in informative war, equipment precise support system based on Internet of Things is put forward. Through layering research of precise support system, overall structure of the system is designed, main functions of the system is described in detail, a theoretical basis is laid for development of system.

An Improved Extension Method of EMD Based on SVRM

The Support Vector Regression Machines (SVRM) is considered to be an effective method for time series prediction, while it is not suitable for long-term prediction. In order to improve the SVRM application in eliminating the end effect of EMD, we apply the SVRM to achieve an accurate short-term extension in the end of the original signal before EMD, and then extend two extreme with the Self-adaptive Triangular Waveform Matching extension method (STWM) in the every sifting process of EMD. The results show that the new extension method can effectively contain end effect of EMD.

Tribological Behavior and Mechanisms of the Deloro60 Alloy Coatings by Electro Spark Deposition under Dry Sliding Friction Conditions

The Ni-base alloy coatings of Derolo60 were prepared on the surface of carbon steel by electro spark deposition. The tribological properties of the coatings were investigated in a tribometer under dry sliding friction conditions. The results show that the coatings exhibit excellent properties of wear resistance because of their unique microstructure with a rational combination of hard phases and tough matrix. With the increase of the normal loads and sliding speeds, the friction coefficients of the coatings decrease, while the wear losses increase. The main wear mechanisms of the coatings are micro-cutting wear and multi-plastic deformation wear at low speed and light load conditions, and then gradually change into micro-cutting wear and adhesive wear as well as fatigue fracture accompanied by some oxidation wear at high speed and heavy load conditions.

The Improved Fuzzy Analytic Hierarchy Process

In FAHP, discontinuous scale, such as 0.1-0.9 scale, is usually used to get fuzzy judgment matrix. But when evaluating system is complex or index systems are too many, we may get stuck in a comparison dilemma by using 0.1-0.9 scale, resulting in a fuzzy judgment matrix of poor consistency. This paper improved 0.1-0.9 scale and established 0-1 continuous scale. We first apply 0-1 three scales to get a comparison matrix, and then converted it into a fuzzy judgment matrix according to 0-1 continuous scale.

Research on Parameters of SVRM in Time Series Prediction

Though the Support Vector Regression Machines (SVRM) is considered to be an effective method for time series prediction, its performance is greatly influenced by its parameters. In order to improve the rationality of parameter setting, the influences of the parameters (the number of support vectors NS and the prediction length NE) and signal characteristics on the SVRM performance were discussed. The results proved that the existence of confliction between prediction accuracy and prediction efficiency, and SVRM may inappropriate to long-term prediction, and NS should be greater than a threshold which depends on the signal characteristics for an accurate prediction result. The research results may provide a theoretical basis for the improvments of SVRM algorithm.

Microstructure and Tribological Properties of Stellite21 Coating by Electro-Spark Deposition

Setellite21 cobalt-based alloy coating was deposited on 45 steel by electro-spark deposition. Microstructure and phase composition of the coating were analyzed. Wear resistance and wear mechanism of the coating were researched. The results indicate that the coating with compact structure is mainly composed of Co, Co6W6C, CoCx and CoCr. Average microhardness of the coating is 445.34 HV0.5, which is about 2 times to that of the substrate. The coating presents excellent wear resistance with no obvious peelings and scratches. Wear resistance of the coating is about 2.3~2.7 times to that of the substrate. Wear mechanism of the coating mainly contains abrasive wear and fatigue wear, and along with oxidization wear.

Design of State Monitoring System of Internal Combustion Engine Based on Serial Bus

This paper introduces a type of control and test system for internal combustion engine based on Serial bus and virtual instrument technique. It consists of host and intelligent node, with parameter signals communication through RS485 serial bus to form a distributed high-precision status parameters controlling testing system. The host computer implements the user interaction, data storage, status monitoring, power measurement and computation as well as management. The client unit reads the commands of host, controls the rotation speed of engine and excitation current, detects all the parameters of system and transmits the data. The hardware and software structure of the distributed testing system (client) with ATmega128 micro single chip computer and communication module are implemented. And the application software was developed with graphical virtual instrument language LabVIEW. The solution may have some reference value to the design of other internal combustion monitoring system with other control bus and the development of general single chip microcomputer and industrial field monitoring and controlling system.

Distributed High-Precision Field Testing System of Construction Machinery Based on CAN Bus

The main objective of this work is development of a type of status parameter testing system based on CAN bus for construction machinery. Various difficulties due to over-abundant parameters of testing and controlling, many different types of transducer, changes and anomalies in the operation temperature and harsh EMI environment are considered in the accomplishment. The design of intelligent field testing node is demonstrated in detail. The intelligent node consists of embedded microcontroller AT89C52, high-precision signal conditioning circuits, CAN interface module and other peripheral circuits. The intelligent node uses software auto-zero correction and full-scale correction as well as the method of second-degree parabola interpolation to deal with the non-linear characteristic of the measured data. Hence, it implements the high-precision detection and measurement. In addition, the intelligent connects to the host computer via CAN bus to form a distributed high-precision status parameters testing system. The hardware circuit of operation status parameters measurement and communication interface circuit are developed and described, and the software is also accomplished with C language and LabVIEW. The system has the advantages of simple structure, reliable performance, significant capability of real time processing, powerful anti-jamming performance and so forth. Meanwhile, the system has great application prospect.