Study On Ontology-Based Evolutionary Reasoning Mechanism In Emergency Decision System

With the popularization of information technology development in intelligent control, business, finance, science experiments, information services in the field of application of a new series of complex intelligent decision problems, they have some characteristics such as huge data, including random factors, requiring environmental adaptability, automatic decision-making, real-time decision-making and high reliability etc. These problems of traditional modeling methods and technology for intelligent decision have new requirements. The fundamental ideology of EDS (Emergency Decision System) is to study the theory and technology of intelligent decision support system, to improve the efficiency of Emergency treatment and the effectiveness of decision-making. Due to complexity of emergency incidents decision, uncertainty of solutions and urgency of time, EDSS (Emergency Decision Support System) requires better performance than other EDSs, which needs characteristics including faster response, simple interaction and reliable operation etc.Ontology performs effectively as a semantic model and the concept of hierarchical structure, in response to the decision-making system to provide consistent understanding of the public domain knowledge. Inquiries can be used by decision-makers to search for relevant and accurate information for decision-making inquiries and provide guarantees. Evolutionary computation is a mechanism from biological evolutionary course for solving optimization problems, which have a great deal of applications in the field of machine learning. Although success has been achieved a lot, its potential, especially for complex application problems, doesn't release fully. We can use evolutionary computation when universal quantifiable models of decision-making rules can be constructed so that the emergency decision problems are attributed to find the best optimizational parameter of decision-making rules.Based on the background of decision-making process for fire emergency, this article tried to introduce evolutionary computation technology into emergency decision-making process, to build the model of ontology-based evolutionary reasoning, and to analyze its feasibility and universality.Firstly, on the basis of careful analysis of firefighting field's background knowledge, this article researched on ontology knowledge representation of emergency decision system, established ontology-based case and knowledge library on fire emergency ontology knowledge in the field of decision-making process, made use of characteristics of the definite description and formalization to field knowledge to improve the efficiency and quality of obtaining problem-solving solutions.Secondly, on the basis of analyzing related technology of ontology reasoning, a kind of Ontology-based evolutionary decision technology was put forward against the question of slow decision-making process and single reasoning-technology. A fire emergency decision-making mathematical model of modern city was constructed which is based on time and space distribution of actual fire in the decision-making process. Time and space were two aspects to the prerequisite for the deployment of resources, established a two-level optimization mathematical model of minimum response time and losses. A dual-gene mixed-evolution strategy is established based on (μ+λ+κ) - ES and mixed (μ+λ)-ES and improved the speed and accuracy of the algorithm under the premise of maintaining the diversity of groups. The algorithm will be applied to evolutionary strategies to improve the process of solving mathematical model; it had researched ontology data technology and designed interface module of the algorithm access ontology data, implement the operation of algorithm module accessing ontology data.Finally, this article designed and implemented the Fire Emergency Decision prototype system with research results on the basis of evolutionary reasoning mechanism. The system which has fire emergency field knowledge as ontology knowledge and case Warehouse gave the ultimate decision Proposal with the requirements of the proposed policy through the Handling of ontology matching module, Decision knowledge module, the evolutionary reasoning module, the ultimate decision-makers presented with the requirements of the proposed policy. The operating results show that it can implement Real-time system with the correctness of the decision-making reasoning and the rationality of solutions approximately. It also approved that using evolutionary algorithms auxiliary in emergency decision-making process is feasible and valuable. In the end, this article generalized contents and prospected further development in the direction of the problem.