Dissertation > Environmental science, safety science > Safety Science > Safety Engineering > Petroleum, chemical industry security

Research and Application of Safety Assessment Method Based on Neural Network

Author HuYingYing
Tutor LiuWeiZuo
School Shandong University of Science and Technology
Course Safety Technology and Engineering
Keywords safety assessment Preliminary Hazard Analysis Neural Network theory Six Stage Evaluation MATLAB software design
CLC X937
Type Master's thesis
Year 2011
Downloads 55
Quotes 0
Download Dissertation

Safety problem is becoming one of the severe prolems faced by petrochemical industry, which is the main barrier to harmonious society construction. Safety science is the basis and security of industry safety and society safety, which supports national economy to develop sustainably and healthily. Applying safety science to petrochemical enterprise becomes an important theme. Hazard identification is the fundamentation of the accident prevention, and safety assessment is the premise. Enterprise safety management aims to control hazard accident and guarantee safety. Adopting better assessment methods to evaluate enterprises’ actual situation can make safety management have a definite object in view as one who shoots the arrow at the target.This dissertation analyses the main problems existing in the safety assessment system and improves the method based on the theory understanding. To solve the problems, a Six Stage Evaluation method based on Neural Network theory is proposed and improved on the basis of the existing method. Through hazard identification and assessment unit divided of Xinhai petrolchemical enterprize, Preliminary Hazard Analysis assessment method and the new improved Six Stage Evaluation method based on Neural Network are applied to Xinhai petrolchemical enterprize programme, which proves the feasibility of the improved method. By the analysis of the improved assessment system, the software of Six Stage Assessment base on Neural Network is created by MATLAB and Visual C++. The research achieves good effect in the example applications.

Related Dissertations
More Dissertations