Applied Mechanics and Materials Vols. 494-495

Abstract: Establishing a test system model is necessary guarantee to realize the state monitoring, and also the premise of optimizing test points. The modeling method, principle and steps based on multi-signal flow model were analyzed in this paper. And a test point optimization method based on this model was studied. Through analysis and comparison, the proposed method can reduce complexity of monitoring system, avoid blindness of the diagnosis and reasoning process, and improve efficiency of fault diagnosis.

Abstract: Real-time reliability estimation is very significant for system operation safety, particularly in railway transportation system. This paper proposes a method, integrating sensor-driven prognostic model with modular approach, to accurate estimate system real-time reliability. Modular approach is utilized to divide system fault tree into independent subtrees (modulars), and solves system reliability using Binary Decision Diagrams and Bayesian Networks according to their characteristics. Sensor-driven prognostic models use in situ sensor data from system components to compute their failure density functions or reliability functions, and continuously update system reliability. The method for system reliability assessment presented in this paper, integrating sensor-driven prognostic model with modular approach can overcome static characteristics of reliability analysis, and better correspond to practical applications.

Abstract: Eddy current is presented in electric coil when it approaches metallic conductor. Different metals have different conductivity and permeability. Therefore, the equivalent impedances of the electric coil are also different. Based on the principle, a Steel material sorting instrument was made to detect the steel permeability level, through the permeability level, combined with the central processing circuit steel hardness values reflect iron parts, materials and the carbon content of steel, mixing nondestructive sorting.

Abstract: In the process of the piezoresistive sensor design, the placement of varistor has a great influence on the sensitivity of the sensor.In this paper, three theoretical models of piezoresistive sensor are established.The stress simulation analysis of the different varistor path way placed on the cantilever beam is conducted in Ansys software. The sensitivities of the axis and transverse direction are calcuated and compared by Matlab compiler. The results show that the placement of varistor which is parallel to the cantilever beam is the most optimal choice.

Abstract: In this paper, two robots in different positions with two different paths are planned based on RRT algorithm. And two robots search the ball according to the planned paths coordinated with each other under the circumstance of existing static obstacles. Simulation result showed that the correctness of the planned path. RRT algorithm is suitable for path planning of high dimensional robot under complex environment, which avoids complicated modeling of robot and obstacles in C space. And RRT algorithm uses tree structure, which tracks between adjacent node accord with requirements of robots kinematics, dynamics and avoiding collision. Therefore, it is suitable for practical application of robots.

Abstract: In the past, the semi-active control study generally is calculated based on the MATLAB numerical procedure, which cant achieve fine simulation. In this paper, to solve this problem, the finite element software ABAQUS is taken to be secondary development, a numerical simulation method of studying semi-active control is proposed. And the seismic response of a steel column is taken for the research. the numerical simulation analysis of the semi-active control is carried out. The results show that the control effect of the control method is significant, and the simulation result is clearly visible.

Abstract: . An automatic cleaning device used for cleaning cellular evaporator in sugar factory is designed, which uses high-pressure water jet to clean fouls. The device includes supporting frame, linear guides, sending and closing pipes mechanism, hydraulic system and high-pressure water jet generating system, etc. Actions of sending and closing pipes as well as other cleaning operations are controlled by a numerical control system. Experiments show this device can replace manual cleaning for achieving automatic cleaning to a cellular evaporator, and it significantly increases cleaning efficiencies.

Abstract: We designed a color recognition algorithm based on HSI color space, through front-facing camera to identify the scene in front of the robot, then video stream was cut into frames and image processing was conducted. The presented algorithm does well on the robot experiment platform, we can carry out the color recognition, and also can identify the size of the red area through the parameter setting, so as to choose the preset area for tracking. Simulation results show that the optimization function has successfully filtered distractors, the system basically meets the requirements of real-time, providing effective support for the robot tracking.

Abstract: An intelligent control model, called fuzzy self-organization model, is proposed based on the requirements of electromagnetic bionics system design principle and fuzzy algorithm. What the intelligent control model is and how it can be used in projects and the validation are elaborated. An intelligent control system - a heating system as selected, is developed according to the proposed model, in which the intelligent control model is constructed and coded into a DSP chip to organize the heating circuits/branches and control the heating procedure. The test results of the system show that the model can be carried out successfully, and the heating procedure is optimized by the settings of accurate heating temperature and the heating speed.

Abstract: According to the phase delaying question of estimated rotor position angle in traditional sliding mode observer, using fractional order calculus theory and singular perturbation two-time-scale decomposition theory, making up the parameter designing rules of fractional order PD sliding mode observer, proposed position sensorless control strategy for permanent magnet synchronous motor based on fractional order PD sliding mode observer. On one side, taking fast time-varying subsystem of singular perturbation two-time-scale decomposition theory to establish fractional order sliding mode observer; on the other side, taking slow time-varying subsystem of singular perturbation two-time-scale decomposition theory to obtain parameter of fractional order sliding mode observer. The results show that the method could not only making estimated rotor position angle follow actual rotor angle, but also establishing a global compensating control strategy of estimated rotor position angle effectively.