Computational Methods in Engineering

---

Soft Switcing techniques have recently been applied in the design of dc-ac converters, in order to achive better performance, higher efficiency, and power density. One of the soft switching techniques uesd in inverters is resonant dc links. These topologies have some disadvantages such as irregular current peaks, large voltage peaks, uncotrollble pulse width, etc. Another soft switching method in inverters is using Quasi –resonant links, which have PWM modulation capability. Inverters with series or parallel Quasi-resonant dc links use several quasi-resonant current or voltage pulses, respectively, to produce PWM modualation. In this paper an inverter with a novel Quasi-resonant series dc link is introduced. This topology enables current source inverters to have characteristics such as resonant pulse peak limition and pulse width controllability. This circuit provides the inverter with two to three ranges of PWM control capability which increases the switching time control in a larger range. Various operational modes of this novel Quasi-resonant dc link is analyesed and then the circuit losses is calculated. Finally, simulation results by PSPICE software is presented to justify the circuit operation. Keyword: Inverter, Soft switching, Novel quasi-series resonant link, increasing control areas, Losses

---

Introducing distributed generation into a power system can lead to numerous benefits including technical, economic, environmental, etc. To attain these benefits, distributed generators with proper rating should be installed at suitable locations. Given the similar effects of distributed generators and capacitor banks on operation indices of a distribution system, simultaneous assignment of best locations and sizes to both will not only lead to greatest benefits from distributed generators but also to lower reactive power capacity requirements. In this paper, a new combined planning problem involving distributed generation and Volt/VAr control means planning is formulated and solved in which the quantity of distributed generators and reactive power sources are simultaneously assigned to buses in a distribution system. Also tap positions of voltage regulators are computed such that with a given distributed generation under peak load conditions, power losses and the reactive power capacity required are minimized. Like many other problems in power network planning, the problem formulated here is a nonlinear combinatorial one. Hence, we employ the tabu search algorithm to solve the optimization problem. The results from applying the algorithm to distribution networks with 6, 19, and 33 buses are presented and compared with those obtained from employing the second order method.