Search

Capacitors are widely used in power electronic converters to buffer the pulsation power, filter the harmonics and support voltage for stable operation. For these applications where single capacitor can not fulfill the voltage rating or capacitance requirements, capacitor bank is always used as the energy buffer by connecting several capacitors in parallel for larger capacitance, or in series for higher voltage capability. The existing design considerations for the capacitor bank are in terms of voltage ripple, electromagnetic-interference, power loss, weight and volume. With more stringent constrains on volume for high power density applications, reliability as well as temperature of capacitor banks should be considered into the design phase. In order to estimate the temperature of capacitor bank with sufficient accuracy and computational efficiency, this paper proposes a lumped thermal model for a capacitor bank by taking into account the boundary conditions (e.g., ambient temperature, power loss for each cell) and the thermal coupling effect among the capacitor cells. Considering the variable boundary with different loading conditions in FEM simulation, fast thermal resistance extraction method is investigated to obtain the self-heating and coupling thermal impedance. Critical uneven temperature distribution among capacitors in a bank may be observed. A case study of an electrolytic capacitor bank is presented to verify the accuracy of the proposed method.

Shuffled frog leaping algorithm (SFLA) is one of promising optimistic methods which are based on swarm intelligence. SFLA combines the advantages of memetic algorithm and particle swarm optimization. It has been widely used to solve complex optimization problems in engineering fields. To enhance the convergence and effectiveness of local search process in SFLA, a novel individual update method is presented and the dynamic behaviors of local search are analyzed in detail. On the basis of the proposed approach, the modified SFLA is presented. Simulation experimental results show that the modified SFLA achieves more efficiency than the classical shuffled frog leaping algorithm.

Shuffled frog leaping algorithm (SFLA) is meta-heuristic for solving complex optimization problems. It is one of promising optimistic methods which are based on swarm intelligence. SFLA combines the advantages of memetic algorithm and particle swarm optimization and has been widely used in engineering fields. In order to overcome the shortcomings of local search in the classic SFLA, a novel update method with convergence property is presented in this paper. On the basis of the proposed approach, the modified SFLA is presented afterwards. Experimental results show that the efficiency and convergence of the modified SFLA can be enhanced significantly.

The turning ability of heavy transportation vehicle is investigated and improved using new control method in this paper. The investigation is carried out on the model building of the whole turning system includes electromagnetic valve, hydraulic cylinder and turning mechanical part. Based on the model, the factors of blocking the turning ability of heavy transportation vehicle are studied. Then a new control algorithm which combined Bang-Bang control and PID control is presented, the simulation is carried out to verify the algorithm's validity. According to the actual application demand, the turning control algorithm is realized by using CAN fieldbus and PLC controllers. The ability of the turning system of heavy transportation vehicle has been improvement greatly and the effect shows the algorithm meets the requirement.

Circuit breaker switching operation will generate transient voltage and inrush current. The synchronous switching control strategies research can effectively reduce the transient effect and improve power quality. The transient operation process of vacuum circuit breaker is analyzed by deriving the system voltage and current expressions when vacuum circuit breaker switching reactive compensation capacitor bank. The theory switching phases without transient transition process are obtained by referring to the voltage and current expression. The optimum switching capacitor banks phase and new synchronous switching control strategies of different connections capacitor bank are given by synthetically considering closing action time dispersion, pre-strike characteristic and opening arcing time of vacuum circuit breaker. The voltage and current waveforms when vacuum circuit breaker randomly, according to traditional and new synchronous switching strategies switching ungrounded Y capacitor banks connection are simulated in the article. Simulation results show that the inrush current under the new synchronous switching control strategies can be reduced 1.2p.u.and closing voltage has no oscillation transition process, that is, directly accesses to steady-state. So the theoretical analysis and synchronous switching control strategies in the article are verified correctly.

Digital image coding includes vector coding and scalar coding. Vector transform is an important content to realize image vector coding. According to the unequal dimension between the input matrix and output matrix in the classic transform, a fast vector transform with equal dimension is deduced. The computational efficiency and the transform performance are analyzed, and the basic block of the fast vector transform implementation, the circuit diagram of a skew-rotation unit, is presented. The research can provide the theoretical and technical foundation for the vector transform.

Modern equipment is large structure and complex function with redundant components within the system. As a result, the complexity is increased when fault diagnosis and analysis model is set up. In order to reduce diagnosis model complexity, a method of directed fault propagation model and optimization is proposed. Fault relationship among the components is analyzed based on the probability and influence of fault propagation, the multiple fault modes and fault propagation relationship is also studied. The directed fault propagation graph model is optimized according to the fault influence factor on the basis of quantitative assessment, and redundant component within fault diagnosis model is cut, so the model is simplified, the computational process is reduced and the efficiency for fault diagnosis and analysis is improved. A practical model shows the validation of the model and optimization method.