Application Of Neural Network With Chaos Genetic Algorithm In The Oil Pipeline

In today’s pipeline transportation industry, along with the rapid development the attendant is more environments was leakage of oil pollution, and it can cause energy waste and inestimable economic losses. In recent years, the pipeline in most areas of varying degrees of aging and corrosion phenomenon, combined with some criminals stealing oil incident, leading to frequent leakage accident. The traditional leakage detection has a certain error and efficiency is also relatively low. So a set of effective automatic pipeline detection and location system is very important. The pipeline leakage detection is a topic involving extensive, sensor technology, signal detection and digital signal processing discipline.This paper systematically summarizes the domestic and international oil pipeline leak detection and the research status, more detailed expounds the pipeline leak positioning principle, difficulties and solutions. Wavelet packet algorithm and the neural network technology, the wavelet packet transforms the pressure signal denoising processing. At the same time, extract feature vectors as the input of neural network.In order to improve the accuracy of pipeline leakage detection, a new algorithm for BP neural network based on chaos genetic algorithm is proposed. BP neural network is used widely but is prone to converge to minimum, and the convergence speed is relatively slow. The improved algorithm is based on chaos genetic algorithm to optimize the initial weights and thresholds of BP neural network. The chaos’ s ergodicity and the advantages of genetic algorithm inversion have been combined. Chaotic variables are added to the genetic algorithm, improved genetic algorithm’s global search ability and convergence speed. The improved CGA-BP algorithm is used for pipeline leakage prediction. The result shows that the classification of the pipeline leakage is better than that of BP neural network.Nowadays, the single leak detection are often not able to meet our needs. In practical application, a variety of methods are combined to get more accurate proposed by the system software and hardware and the overall design. The system collects the information from the pressure sensor at both ends of the pipe, and carries out the function of the data transmission and timely alarming and locating.