Advanced Materials Research Vols. 383-390

Abstract: High-frequency digital inverter power supply has the advantages of small, no noise, high efficiency, small ripp- le, and perfect protection, therefore, is gradually replacing the traditional electron beam welder which uses the 400Hz frequency generator. However, with more high frequency of the power supply, the design of the transformer used by the high-frequency high-voltage electron beam welding becomes difficult. Therefore, this paper put forward a diversified transformer mode and a voltage doubling rectifier circuit, which were used to reduce the transformer turns ratio, thus reducing the topology of the transformer distribution para- meters. Accordingly, a special high-frequency high-voltage electron beam welding tank is designed. Tests show that the set of performance indicators of the fuel tank meets the design goals to meet the actual needs of welding process.

Abstract: To overcome the shortage of standard Particle Swarm Optimization(SPSO) on premature convergence, Quantum-behaved Particle Swarm Optimization (QPSO) is presented to solve engineering constrained optimization problem. QPSO algorithm is a novel PSO algorithm model in terms of quantum mechanics. The model is based on Delta potential, and we think the particle has the behavior of quanta. Because the particle doesn’t have a certain trajectory, it has more randomicity than the particle which has fixed path in PSO, thus the QPSO more easily escapes from local optima, and has more capability to seek the global optimal solution. In the period of iterative optimization, outside point method is used to deal with those particles that violate the constraints. Furthermore, compared with other intelligent algorithms, the QPSO is verified by two instances of engineering constrained optimization, experimental results indicate that the algorithm performs better in terms of accuracy and robustness.

Abstract: In this paper, a method of multistep imprint lithography process is described. Through comparing among the loading process factors, a multistep loading locus, which includes a pre-cure release of the pressing force, is proposed for the high-conformity transfer of nano-patterns from the template to the wafer. A series of imprint experiments show that the new multistep loading process can meet the needs for different pressing areas, feature sizes and repetitious imprints. This loading process can effectively reduce the residual resist thickness while maintaining a uniform residual resist and non-distorted transfer of nano-patterns to the resist-coated wafer. And a high-conformity of 100 nm feature can be achieved.

Abstract: Electrical conductivity of silicon nanomembranes (SiNMs) was measured by van der Pauw method under two surface modifications: hydrofluoric acid (HF) treatment and vacuum-hydrogenated(VH) treatment, which create hydrogen-terminated surface; and one interface modification: forming gas (5% H2 in N2) anneal, which causes hydrogen passivated interfaces. The results show that thinner SiNMs are more sensitive to the surface modifications, and HF treatment can cause larger drop of sheet resistance than that caused by VH treatment probably because of Fluorine (F). Forming gas anneal can also improve the conductivity depending on the interface trap density.

Abstract: A HEMP Simulator was constructed based on coaxial peaking technology. The experimental results show that: a HEMP field matching the MIL-STD-464A standard is generated under the bounded wave transmission line when the parameters are selected reasonably; the variation coefficient of output wave parameters such as peaking filed strength, rise time and half time show that the repeatability of the HEMP simulator are very good; there is linear relation between the peaking field strength and the charging voltage of the Marx generator.

Abstract: A kind of multiple bi-directional DC/DC converter used in supercapacitor energy storage unit of the wind power flow optimization and control system is analyzed and studied. Based on DSPs, the converter uses a double closed-loop control strategy to stabilize voltage and uses a digital parallel current sharing strategy to eliminate circulation current among each converter unit. In the paper, firstly, the working principle and the working process of the multiple bi-directional converter are analyzed in detail. Then, on the basis of the small signal model, the way to voltage stabilization and current balance are presented. Finally, a low power converter prototype is designed and made to carry on the experiment. The experimental results show that the proposed control strategies are feasible and effective. So the multiple bi-directional DC/DC converter designed in the paper has comparatively high practicable value in the wind power flow optimization and control system.

Abstract: Analyzing the structural property of model vehicle, a multi-body dynamic model of a full independent 4-wheel-drive electric vehicle(4WD EV) is established by using automatic dynamic analysis of mechanical systems (ADAMS) software, and an electric motor and drive-control system is established under Matlab/Simulink environment. Co-simulation based on the two soft wares is under control of the equality power driving force distribution strategy controller. The results of co-simulation show that the co-simulation is effective and the model’s rationality is validated. It could be applied in the following experiments in complex circumvents, and give theoretic supports.

Abstract: Based on the analysis of the current long-distance pipeline network running conditions, an economic optimal mathematical model of the gas transmission network including compressor station is used. The natural gas pipeline network is divided into different parts, and adopting the cooperation co-evolutionary genetic algorithm, the subpopulations are created. The fitness function is established by taking advantage of the punish function. The results of the simulation show that this approach has better convergence. It is an effective method to solve the optimization problem.

Abstract: Permanent magnet synchronous motor (PMSM) is a multi-variables, non-linear and strong coupling system. A model reference adaptive controller (MRAC) for PMSM based on back propagation (BP) neural network (NN) is proposed to solve the shortcoming of traditional proportion integration (PI) control method, which is widely used in linear system. According to the proposed method, the simulation model is established and simulated with Simulink. The adaptive control of motor speed is achieved with the training of BP neural network. Simulation results show that the system has long response time, small overshoot and high static performance.

Abstract: In this paper, based on linear quadratic optimal control design the controller of single inverted pendulum system, using the current epidemic method of Co-simulation to play each of the strengths of two software for simulation, Through two methods of the static and dynamic to observe and analyze the quality of feedback controller the based on linear quadratic optimal control.