The Exchange Algorithm And Its Application For Determining The Optimal Position Of Switching Devices

This thesis was supported in part by the National Natural Science Foundation of China ("Study on the optimal switching model and algorithm for electrical distribution networks using immune-algorithm", No.50307015). This paper discusses the exchange algorithm of switch positions (EASP) for determining the optimal position of switching devices(OPSD) in electrical distribution networks (EDN), the improved EASP and the sensitivity analysis of the OPSD schemes varying with the reliability and economic parameters.The literatures show that 80% of the total customer interruptions results from the faults occurred in an EDN. The measures used to improve the reliability of the EDN considering several factors, such as costs and node voltages, currently became a popular research field. The OPSD is a significant measure to improve the system reliability, decrease the outage loss and save the investment of an EDN. This thesis presents the EASP for determining the OPSD in EDN using the cost-benefit principle and the reliability theories for the EDN. The main contents are as follows.â‘ This thesis presents a model and its algorithm based on the EASP for OPSD in the EDN. The calculation model is based on the equivalent annual value method. The objective function of the model is consisted of the outage costs, the annual investment and maintenance costs of switches. The power flow constraints, the reliability performance of the EDN and the cost of components are also considered in this model. Only one switch is moved and the others stay primary position every time when the algorithm is applied for OPSD, namely, each switch is moved in sequence to the around for improving the reliability of END, then the objective function is calculated for gradually implementing the optimal location of switching devices. The reliability evaluation algorithm based on the section-algorithm method regards a section not a component as a calculation unit during fault enumeration processes. A few sections is changed during each switch position exchange, therefore, in order to obtain the reliability indices, outage costs and objective function value of the EDN after a switch moved, it is only needed to calculate the reliability indices of the sections whose structure had been changed. The test of the RBTS-BUS6 and some practical END show that EASP is accurate and suitable to be used for the OPSD.â‘¡This thesis presents an improved EASP integrated with Genetic Algorithm(GA). The EASP relies strongly on the choices of the initial values, so how to confirm the initial value is a very important problem. In order to improve EASP, EASP is integrated with GA for OPSD, which the results of GA is regarded as the initial input data instead of the engineering experience data. The hybrid algorithm based on the global search feature of GA and local search feature of EASP is applied to find the global optimal solution. The proposed method is applied for switching devices optimization in a practical 10kV EDN. The results show that it can be used to provide new schemes with decreased outage costs for the original OPSD schemes using GA. This indicates that the method usually finds a better result than the one from GA. Practical EDN examples show that the method has significant impacts on the OPSD resultss with an increasing number of allocating switches.â‘¢The sensitivity analysis of the OPSD schemes varying with the reliability and economic parameters are presented in this thesis. The OPSD scheme for an EDN is depended on a wide range of factors including the reliability and economic parameters of components. Many models are used to prepose the OPSD scheme for an EDN using the constant reliability and economic parameters, which are usually changed with the development of the economy and society. A 10kV practical EDN is used to perform the sensitivity analysis of the reliability and economic parameters.â‘£The EASP for determining the OPSD was implemented in VC++6.0 in WINDOWS operation system. The practical cases show that the proposed algorithms are effective and practical.