Study On Safety Evaluation Model Of Oil And Gas Drilling Based On Rough Set And Neural Network

Drilling operation is one of areas where accidents often occur in oil and natural gas exploration and development activities. The number of drilling site hazards and personnel violations is often very high which can easily induce safety incidents. Once the safety accident occurred, it will cause casualties, equipment damage and environmental pollution, and have a huge impact on the economic efficiency and the social efficiency. How to ensure the safety of drilling operation and prevent the occurrence of accidents are always the problem which drilling industry needs to focus on.Thus, it is necessary to have a comprehensive identification and analysis of the hazards in oil and gas drilling operations and understand the security status of the drilling site. To establish a safety assessment model of oil and gas drilling is the problem which needs to be solved immediately. This paper is intended to provide an effective evaluation method for drilling operations and a real-time, objective decision-making basis for safety regulators of drilling operations. This study is a new exploration to make drilling safety management more scientific. It also have a great significance to improve the safety management level of the drilling company.Oil and gas drilling is a complicated system, its biggest feature is dynamic, stochastic and fuzzy. Many factors which have mutual restraint affect the safety of drilling operations. Drilling safety assessment is a nonlinear problem. Considering the nonlinear mapping ability of BP neural network and the powerful analysis ability on incomplete and uncertain information of rough set, this paper use rough set and neural network to build the drilling operation safety assessment model. The contents of this paper are as follows.(1) To understand the researching actuality of safety assessment and drilling operations safety assessment at home and abroad.(2) Based on the comprehensive identification on hazards in drilling operations, to analyze it and establish drilling safety assessment index system.(3) To establish a drilling qualitative safety assessment model based on rough set and neural network and a drilling quantitative safety assessment model based neural network. In the process of constructing the qualitative safety evaluation model, first of all, this paper use rough set to do attribute reduction of the sample data. Then, based on the minimum condition attribute set, this paper select training samples and test samples of neural networks. Finally, training samples are used to build and train neural network and testing samples are used to test the network. After that, this paper construct a quantitative safety assessment model using neural network, and use the training samples and testing samples to train and test the network.(4) To design the drilling safety assessment models, it mainly includes:the overall design of the system, the program design of rough set and neural network.