Research On Theapplication Of Multi-gene Mutation Evolution Strategies In Edss

The city is the most active and most vital part of a country and a region in the people flow,resource flow,information flow,capital flow and technology flow,it is always a regional political, economic, cultural and technology center.However, the city due to its unique high concentration of population and wealth,the complexity of social problems and infrastructure vulnerability, faces the challenges of unexpected events all the time. Developed countries implement advanced management tools and engineering technique in the sudden,catastrophic incidents of the city which leading to significant reduction of property and casualty loss.The urban Emergency Decision Supports System (EDSS) is an effective means which developed countries currently used in urban emergency management.It is used to help managers to make a fast and accurate decision-making of emergency incidents. Compared with developed countries, our country has relatively backward on the EDSS research.Based on the above situation, it is needed to strengthen the urbn Emergency Decision Support System research.Where, the reasoning algorithm is the core component of an urban Emergency Decision Support System, the quality of reasoning algorithm directly determines the efficiency of an EDSS. This article aims to propose an improved Evolution Strategies (ES), compared with the traditional Evolution Strategies which has a better convergence and accuracy, and to research the application of the improved algorithm in EDSS, enabling EDSS with higher efficiency.The paper firstly introduces of EDSS, including system definition, research status, functions and data flow.Then it conducts a study of the Evolution Strategies and proposes a improved Evolution Strategies based on multi-gene mutation, in the chapter 3 gives a detailed algorithm design,algorithm steps and experiments to prove the effectiveness of the improved algorithm.Next, in order to improve the efficiency of the system, the artificial neural network is applied in EDSS to pretreat alarm information of emergency events, which increases the information content of fuzzy knowledge and tacit knowledge and improves upper layer accuracy of reasoning and decision-making. After the paper applies the improved Evolution Strategies which based on multi-gene mutation in the reasoning and decision-making process of emergency incidents, detailed algorithm processes and implementation steps are gived in chapter 4, experiments also are implemented to prove the validity of the method. Finally, by means of a complete simulation experiment, the paper demonstrates the effectiveness of the entire system architecture, processes and functions.