Hong Wei Ma

Research on an ultrasonic NDT system for complex surface parts

Abstract Aimed at internal quality controlling for complex surface parts, an ultrasonic testing system for complex surface parts has been developed using ultrasonic non-destructive testing (NDT) that has the features of strong penetration, good direction, high sensitivity, low cost, and being harmless to people and materials. The technologies of the computer, numerical control (NC), precision mechanisms, signal analysis and processing were integrated in the testing system. The system includes a PC, a system software, an ultrasonic data acquisition card, a stepper motor drive card and a five-axis precision mechanical device, etc. The software was developed using Windows 98-based VC++. According to CAD data of the parts and interpolation methods, the scanning programs can be programmed. The five-axis scanning system is driven by the CNC (computer numerical control) system to control the attitude of ultrasonic probes. The system’s automatic scanning for complex surface parts, the real-time acquisition of ultrasonic data, and automatic identification of flaw signal have been realized. This system can be used not only for testing complex surface parts, but also, for the random testing of curved parts. With a fast testing speed, high sensitivity, high testing precision and high reliability, the system has wide adaptability.

Remote Monitoring and Control System for Electrical Haulage Shearer

For the requirements and problems in remote monitoring and control system of electrical haulage shearer, this paper research the key technologies of shearers remote monitoring and control system, such as position and attitude testing, control strategy decision, communication real-time and reliability of network. Position and attitude testing method based SINS (Strap down Inertial Navigation System) was designed to measure position, attitude angle, speed and drum height of shearer together. The shearer control strategy based on work mode was proposed to control shearer working in optimal status. Finally, the network control model of shearer was established in order to analyze the time delay that effects on shearer control system.

Dynamic Simulation Analysis of Belt Rupture for Belt Conveyor

To study the dynamic characteristics of belt conveyor during belt rupture, the finite element dynamic model of belt conveyor was built. Dynamic simulation system of DTII type belt conveyor was established with AMESim. The simulation parameters were determined by DTII type belt conveyor design manual and the dynamic model. The speed, displacement of break point and mechanic characteristic carve in break moment were obtained. According to the result, the optimal capture time was confirmed.It provides the theoretical basis for the design of belt rupture equipment.

Studying and Simulation Analysis for Rubber Track of Rescue Robot

Rubber track of Rescue Robot was dispersed into limited number of track blocks by the method of finite segment method used in flexible multibody dynamics. The two neighbor track blocks were connected by springs and dampers, then the moldel was become a multi-rigid-body system with flexible joint. Rubber track was modeled with the help of macro command used in the secondary development of virtual prototype technique software named ADAMS. Flexible connection was realized by the method of adding Bushing, and then a new method was proposed to build rubber track model. The obstacle-surmounting simulation of climbing the barrier of single step was carried out. It intuitively reflected the stress and deformation under the condition of climbing barrier. The method mentioned above laid good foundation for studying obstacle-surmounting abilities of the rubber-tracked robots and dynamic characteristic of the tracks.

Research on Gas Outburst Prediction in Coal Mining in China Based on a New Kind of SVM Algorithm

The research on predicting gas outburst hazard on the basis of the unbalanced problem of gas monitoring samples under usual circumstances is given in this paper. Combined with outburst-preventing monitoring parameters, a new kind of method of predicting gas outburst hazard based on v-SVM algorithm through analyzing features of real-time gas monitoring data and extracting parameters of gas concentration real-time variation trend, parameters of gas variation rates and feature parameters of gas emission is put forward in this paper. And the application shows that this algorithm for the prediction of actual gas emission in coal mines is effective.

Steel Cord Conveyor Belt Defect Signal Noise Reduction Method Based on a Combination of Wavelet Packet and RLS Adaptive Filtering

Electromagnetic signal of steel cord conveyor belt defect is easy to be disturbed by Strong non-stationary noise in coal mine, even some defect signal may be overwhelmed by noise. Therefore, the defect signal noise reduction method based on a combination of wavelet packet and RLS adaptive filtering was put forward. Firstly, defect signal was decomposed by wavelet packet and got multi layer high frequency detail signal for wavelet packet decomposition. Then, the sum of multi layer high frequency detail signal was used as noise reference signal of RLS adaptive filtering. Finally, defect signal was reduced noise by RLS adaptive filtering. The noise reduction method was used to reduce the strong non-stationary noise of steel cord conveyor belt joint electromagnetic signal. The experiment results show that the noise reduction method had good filtering effect to non-stationary noise of steel cord conveyor belt defect signal, which can ensure defect signal to be effectively extracted. It has very important significance for the prevention of broken belt accident.

Production Process Modeling and Realizing Based on Activiti

In a view of the problem of traditional workflow in the manufacturing process management, such as low utilization, complex modeling and difficulty of combing system, modeling based on the commonness of manufacturing enterprise production process, in combination with Activiti workflow engine characteristics, to achieve the goal of integrate process model into system quickly. The whole process, including modeling, packaging, deployment, test validation, is easy to be modified and published quickly. Experimental results show that the Activiti modeling method has the characteristics of flexible, simple and diverse, greatly improvement of the utilization of production process model and the efficiency of modeling.

SVM Classification Model Parameters Optimized by Improved Genetic Algorithm

SVM classification model has been widely applied to mechanical equipment fault diagnosis and material defects classification. It is difficult to choose the optimal value of penalty factor C and kernel function parameter for SVM model. Therefore, an improved genetic algorithm to optimize SVM parameters is put forward, which improves crossover and mutation operators and enhances convergence properties by using the best individual retention strategy. UCI data set is used to verify the algorithm. The testing results show that the algorithm can quickly and effectively select optimal SVM parameters and improve SVM classification accuracy.

Analysis and Design an Electro-Hydraulic Type Belt Capture Device for Belt Conveyor

Aimed at the frequent accidents of conveyor belt and the the lower level of belt fracture protection in coal mine, this paper introduces a novel device equipment, the Eectro-hydraulic type Blt Cpture Dvice (Eh-BCD) to achieve safe and reliable working of steel cord conveyor belt. Firstly, after analysed the typical capture device existing on the market and researched the mechanical properties of the instant belt, an electro-hydraulic control type capture device solution was put forward, its work principle and main modules were introduced also. Secondly, main parts force of capture device calculated using Newtonian mechanics, designed and calculated the mechanical structure of the device, and determined key parameters such as installation space. Finally, after kinematic model built and analysed, device response time parameters the electro-hydraulic controlling capture was estimated.

Defects Classification of Steel Cord Conveyor Belt Based on Rough Set and Multi-Class v-SVM

Because of steel cord conveyor belt with high load operating and complex conditions of coal mine, it is prone to cause conveyor belt horizontal rupture. It will bring tremendous hazards for coal mine production. Twelve time domain features of joints signals, broken wires signals and abrasion signals for steel cord conveyor belt were extracted with weak magnetic detection system. The algorithm of combining rough set based on information entropy with multi-classbased on binary tree was proposed to classify the three categories signals. The experiment results show that rough set reduction algorithm based on information entropy can effectively achieve feature reduction and classification speed of multi-classclassification algorithm based on binary tree can be improved by rough set feature reduction without changing classification accuracy.