Study On Fault Diagnosis And Fast Service Restoration Of Distribution System In Yangjiaping Power Supply Bureau

Along with the development of the distribution system, the system structure is more and more complicated, the links between the different regions are also becoming closer, this makes the system fault occurred in the system itself also will be the impact of expansion, and along with the constant development of the national economy, users of the power quality requirements of increasingly high. Distribution system links directly to customers whose reliability and supply quality play an important role in national economy and daily life. When faults occurred, rapid fault diagnosis and service restoration is significant to shorten outage time, decrease outage loss, and improve supply reliability and safety.In this paper, according to the distribution system and service restoration, chain storage structure model is introduced to depict the distribution system topology. Object-oriented method is applied to describe the model and electric elements. Based on the established model, the overall and local dynamic connectivity analysis methods are presented. Moreover, branch current based power flow calculation is adopted whose matrix-iterative form is deduced in order to obtain power flow results quickly. Rough set theory is used for fault diagnosis. To apply the method, decision table for fault diagnosis is established firstly. Aiming at the decision table reduction, attribution reduction method based on Tabu search method is proposed, it obtains the reduct of decision table and extracts succinct diagnosis rules promptly. As for the mismatch during fault diagnosis because of lack of important information, rule match method combined Euclid distance and grey theory is used. Rough set based diagnosis method can give satisfactory results even when the key information is lack or incomplete.In order to obtain appropriate restoration schemes, Tabu research is applied for service restoration. The algorithm searches power supply paths by finding effective tie-switches and sectionalizing switches, and eliminates improper switch pair according to power flow calculation. The algorithm takes switch operation number and customer priority into account, and consequently, the initial feasible scheme can restore load as much as possible with minimal switch operation. Considering the variety and flexibility of restoration schemes, rules for generating other candidate restoration schemes are stipulated based on the obtained initial feasible scheme. Both the initial and new generated schemes form a candidate scheme set, and each scheme in it is evaluated by the fuzzy evaluation technique from minimal switch operation number, load transfer, feeder margin and maximal voltage drop. Weighted sum of four indices provides a good guideline for choosing the most optimal scheme for restoration operation.