Method for predicting liquid holdup of moisture pipeline based on GA-BP neural network

Abstract

The invention discloses a method for predicting the liquid holdup of a moisture pipeline based on a GA-BP neural network, and the method comprises the steps: collecting and analyzing a large amount of moisture pipeline experiment data, and carrying out the large amount of machine learning as cloud data; constructing a BP neural network optimized by a genetic algorithm, and mainly performing the following work: determining an initial structure of the BP neural network; determining the population size, the coding length and the fitness in the genetic algorithm; selecting a genetic operator; calculating an optimal weight and an optimal threshold; checking the model accuracy; and calculating the liquid holdup parameter of the moisture pipeline system. In order to overcome the defects of the BP neural network, the genetic algorithm is adopted to optimize the BP neural network, the principle of survival of the fittest is followed, good global search performance is achieved, the defect of local optimum of the BP algorithm is well overcome, meanwhile, the initial weight and threshold value of the BP neural network can be optimized, and the calculation precision of the BP neural network is further improved. The application range is wide, and the accuracy rate is up to 95% or above. Besides a moisture pipeline, the method is also suitable for predicting the liquid holdup flow parameter of the gas-liquid two-phase flow in the industrial fields of power engineering, nuclear energy utilization, chemical engineering and the like.

Description

technical field [0001] The invention relates to a gas-liquid flow parameter prediction method based on GA (genetic algorithm)-BP neural network, which is used for online prediction of liquid holdup of wet gas pipelines, and belongs to the field of multiphase flow prediction. Background technique [0002] As one of the main strategic resources in the world today, oil has an unshakable position, and countries have caused wars because of the competition for oil energy. However, in the process of oil and gas field development, oil field output mainly includes oil, natural gas and water, among which oil and water are mixed. In order to reduce production costs, they are usually transported in the form of mixture. In the process of wet gas pipeline transportation, gas-liquid mixed transportation is usually used for transportation, which not only reduces the economic cost of single-phase transportation after gas-liquid separation, but also saves the construction period and improves ...

AI Technical Summary

Problems solved by technology

However, the traditional BP neural algorithm has great randomness in the selection of initial weights and thresholds, which is accompanied by problems such as poor prediction accuracy and unstable prediction.

Images