Research And Development Of Safety Assessment System For Dangerous Chemical Hazard

Safety system engineering, as a new research field and an important component of safety science, implements system engineering approach to every section of industry production to analyze the safety status, and gives a comprehensive evaluation and hazard control programs to achieve the best safety. At present, the standardization and information technology are highly emphasized in production safety field, in the background of which, it is of great theoretical and practical instructional effect to penetrate the safety system engineering further to the enterprises'safety management.Consisting with the management process of safety analysis, safety assessment and accident prevention on the basis of safety system engineering theory and methods, this thesis mainly focuses on the dangerous chemical hazards, and the work includes research of safety assessment methods and information system development. According to the Bayesian network characteristics of uncertainty knowledge representation and reasoning as well as the conclusion of accident-causing theory, a three-layered Bayesian network topological model is proposed, meanwhile the linked relationships between the layers, the construction and application of the new model are expounded respectively. Compared to the current Bayesian based fault diagnosis method, this model extends the causes of accidents and carries out systematic evaluation of the hidden weaknesses, and also takes the harmful effects into consideration. Moreover, in combination with the existing evaluation methods, a safety assessment system under C/S LAN environment is developed in this thesis, including authority verification, hazard identification and inquiry, evaluation, results inquiry and output modules. To obtain objects'safety situation, the system is designed to perform safety assessment though three aspects: safety ratings, accident consequences and comprehensive Bayesian network evaluation. And it can also quantitatively express the possible property losses, damage to human health and environment within the affected areas when the fire, explosion or leakage accident occur in order to provide decision support for emergency decision system.