Murata Junichi

Simultaneous determination of optimal sizes and locations of Distributed Generation units by Differential Evolution

With the fuel depletion problem, fluctuation of fuel prices and environment crisis, the introduction of renewable energy based Distributed Generation (DG) into distribution networks is a valuable solution. However, a proper installation plan and operation of DGs are required, otherwise the network can suffer from degradation of power quality, stability and reliability. Many studies on the determination of optimal DG sizes and locations were conducted previously either with separated or simultaneous optimization with a restricted number of DGs leading to non-optimal solutions. Furthermore, most of the studies treated the load as constant, considering only the peak load. This paper proposes a planning to determine the sizes and the locations of DGs simultaneously that attains maximum fuel cost saving while satisfying the technical constraints. The number of DGs are unrestricted and the payback period is included to ensure the DG investment is worth. Moreover, the load variations are considered to make this proposed planning applicable in practical situations. Possible DG locations are considered based on geographical features and the optimal locations are then determined with the corresponding optimal sizes using Differential Evolution (DE). This paper treats a mixed combinatorial and continuous problem as a continuous problem. To show its applicability, the proposed method is tested on a distribution network model.

Control of nonlinear mechatronics systems by using universal learning networks

Nonlinear elements such as friction, dead zone, backlash are mixed in most mechatronics systems, and these are factors that make control accuracy of systems decrease and cause an oscillation. In this paper, a design method for nonlinear mechatronics control systems is proposed, in which both of the control object and its controller are represented by using a universal learning network, and the network parameters are trained using a random search algorithm called RasID. In this approach, knowledge and experience about models and controllers are easily incorporated into the network including the nonlinear elements and their compensation elements expressed by non-differentiable functions. Some simulations of a nonlinear crane control system with dead-zone characteristic were carried out. The effectiveness of the proposed design method is illustrated via simulations.