Applied Mechanics and Materials Vol. 389

Abstract: Aiming at the control of a class of bounded disturbance uncertain nonlinear systems,we applied robust control and backstepping control,introduced the concept of virtual control quantity,chose Lyapunov function through stepwise recursion,offered the adaptive law of parameters estimate,designed an adaptive controller with state feedback in the premise of unknown uncertain parameters in the systems, and analyzed its stability. Compared with the conventional PID control in simulation results, the designed controller has the better robustness of the system parameter uncertainty and bounded disturbances, and can ensure the entire closed-loop system is globally asymptotically stable, thus verifying the effectiveness of the proposed control strategy.

Abstract: Aiming at the control problem of dual-motor driving servo system with backlash nonlinearity, we proposed the model of system linear part. We used describing function methods to analyze backlash nonlinearity characteristic. The simulation experiments of nonlinear system with backlash were carried out. The results of simulation show that system represents residual self-oscillation with step response, nonstationarity with low speed tracking and error abrupt change produced by sinusoidal tracking reversing. In the end, we used compensation control strategy to weaken the impact of backlash nonlinearity on dual-motor driving servo system. Finally the simulation illustrates the effectiveness of the proposed compensation control strategy.

Abstract: Aiming at the control problem of dual-motor driving servo system with dead-zone nonlinearity,we proposed the model of system linear part.We used describing function methods to analyze dead-zone nonlinearity characteristic.The simulation experiments of nonlinear system with dead-zone were carried out.The results of simulation show that system represents the phenomenon of zero deflection with step response,error curve exists peak at commutation time with sine response.In the end,we used compensation control strategy to weaken the impact of dead-zone nonlinearity on dual-motor driving servo system.Finally the simulation illustrates the effectiveness of the proposed compensation control strategy.

Abstract: The objective of this work is to define assembly body of processing center by using the contact and node coupling combined method. The information of node displacement, stress and strain can be got by static analysis. Besides, we can check the strength of key parts through the strength theory. Modal analysis results indicate that resonance would be caused in machine tools when external drive frequency is 36.477 Hz, 41.436 Hz and the 44.796 Hz, which are serious accidents. Harmonic response analysis indicates that the certain frequency of 41Hz should be avoided in practice.

Abstract: In this paper, introduce the power supply for full-bridge inverter high frequency high voltage electron beam welder, applying PWM-Buck integrated circuit and zero-voltage switching technology and the control system. Getting continuously adjustable, stable over current and over voltage protection, and the regulator the accuracy requirements of output voltage from high voltage power supply, and improve the energy conversion efficiency of the system.

Abstract: A new method is proposed for the delay-dependent stability control of fuzzy systems with time-varying delay. A new fuzzy Lyapunov-Krasovskii functional (LKF) is introduced to establish a delay-dependent stability criterion. Based on parallel distributed compensation (PDC) scheme, a stabilization condition is derived and the corresponding controller can be obtained by solving a set of linear matrix inequalities (LMIs).

Abstract: A microelectronic packaging position detection method based on machine vision is proposed. Analyze the color and shape characteristics of the FC-D30 handle circuit boards, extract the R component. Complex background is removed through using image segmentation technology, this research utilizes image smoothing technology to eliminate the noise after segmentation. And use Sobel operator to do edge detection. The package position is detected by using image feature extraction techniques, and finally the position is located accurately. The experimental results show that, the average detection time of this algorithm is 180 milliseconds, the correct positioning is 99%. Microelectronic packaging position can be located fast and accurately by using machine vision technology, this mothod has a good theoretical and practical value.

Abstract: A precision temperature control system that uses ATmega128 as the main control microcomputer is built. The temperature is sensed by a Pt100 thermistor, and then converted into the voltage signal by a temperature transmitter to the microcomputer. Two kinds of digital PID (Proportional-Integral-Differential) algorithms are adopted to control the output power of the thermo-electric cooler (TEC) automatically. The test results of the improved temperature control system show that the set-up time of this system is about 340s and the temperature control accuracy is up to±0.1°C.

Abstract: In order to detect the weak signal deeply buried in the noise, a weak signal detection system based on lock-in amplifier is proposed. The system includes the preamplifier circuit, active low pass filter circuit, AC amplifying circuit and phase sensitive demodulation circuit. Test results show that it can greatly increase the signal-to-noise ratio (SNR) up to 12.7db.

Abstract: In order to improve the decoding performance of MIMO-OFDM system in the case of the channel state information is not accurate enough, a new algorithm is proposed, which combines SAGE algorithm, DFT-LS channel estimation and maximum likelihood detection algorithm. The algorithm utilizes joint iterative technology to achieve channel estimation and decoding effect, thereby enhances the reliability of the system. Theoretical study and simulation results show that the proposed algorithm can track the channel change correctly without increasing the system overhead, and the convergence speed is accelerated. Besides, the performance is superior to the commonly used joint detection algorithm. Moreover, comparing with the ideal channel estimation under the maximum likelihood detection algorithm, the new proposed algorithm only has a loss of 0.5dB with the same bit error rate.