Research And Implementation Of Intelligent Decision Support System For Underground Pipeline Incidents

Urban public facilities are the fundamental for maintaining the normal function of a city, which shows the economic strength of the city at the same time and also embodies the quality of life of the general public, so the normal operation of public facilities of the city is of particular importance.Underground pipelines are an integral part of urban public facilities, but in recent years, pipeline accidents have occurred frequently, which will not only bring great inconvenience to people's life, but also bring the loss of economic and even life, so that the underground pipeline incidents can be dealt with promptly and effectively plays an important role in raising the level of livable cities.Therefore, disposing of Beijing public facilities accident timely and effectively is very important for maintaining the stability of Beijing's economy and society. However, currently there is no intelligent system to help building and upgrading the experience of experts. Currently, the accident disposal process is mainly depending on the individual experience, including experts, managers and front-line staffs in different fields. It's very hard to comprehensive and promotes the experience from various sources.This paper analysis current situation and problems of underground pipelines incidents in Beijing and associates with real demand, completed the following work using expert system relative knowledge:1. Research theoretical knowledge and key technologies of Expert System, Decision Support System and Intelligent Decision Support System for further study and design.2. Apply rule-based reasoning technology to expert system, that is extracting rules from incident case base from the past, using production rule representation and positive reasoning technology to construct rule-based reasoning expert system.3. Implement the shortest path between nodes in topology using MapX control and Dijkstra algorithm, and displayed in form of drawing lines on MapX control panel.4. Research principle of case-based reasoning, adopt combination of features to describe problems of cases and search the most similar case compared with the target case from case base using nearest neighbor algorithm.