Application Of The Neural Networks Technology In Oil Refining Enterprises

The developing of oil refining industry has great influence on the national economic development of any country. With the development of science and technology and innovation of the management idea, the oil refining enterprises of the countries all over the world are actively looking for controlling effectively the oil refining craft course and optimizing the course. Some countries have already begun to adopt high-new technological transformation and have quickened the development of oil refining industry. In the last few years, in order to guarantee the security, economy, stability of the oil refining course and meet the demand for sustainable development, a lot of oil refining enterprises are devoting themselves to introduce scientific control of advanced course,optimization and management methods into the course of oil refining, in the hope of obtaining the best economic benefits.Aiming at the deficiency existing in the oil refining craft course of our country's oil enterprise at present, the thesis introduce the neural network technology, and combines it with methods of fuzzy control etc. to carry on the research on the control and optimization of some craft courses in the oil refining course, the main research contents are as follows.First of all, to characteristic of the oil refining course and extant question, point out the deficiency existing in traditional research. Through the exposition of the neural network technology characteristic and the primary application domain, put forward the suitability that the neural network technology uses in oil refining enterprises.Secondly, explain the basic theory of the neural network technology, including structure and classification, multi-layer BP network technology, radial basis function network, genetic algorithm of the artificial neural networks, etc..Thirdly, further investigate the applications of neural network technology in the distillation device under constant decompression, the catalytic cracking device and the oil warehousing and transportation system separately. Utilize neural network technology and some auxiliary technology (such as the fuzzy technology), to model and simulate the above devices, moreover, discuss and analyze some simulation results.Finally, carry on the discussion to the application of neural network technology in the oil refining enterprise's management, including modeling and simulation to forecast of company's stock price and the crude oil appraisal.