基于粒子群优化RBF神经网络原油含水率预测

原油含水率预测对于确定油井水、油层位以及估计原油产量有着非常重要意义。BP神经网络是最近常用的原油含水率预测方法,然而,由于BP神经网络存在容易陷入局部极小值、收敛速度慢等问题,影响了其预测的实用性和准确性,对此,提出基于粒子群优化RBF神经网络(PSO-RBFNN)的原油含水率预测方法,粒子群优化算法用于RBF神经网络参数优化。在分析原油含水率预测的影响因素基础上,建立粒子群优化RBF神经网络的原油含水率预测模型。实验结果表明,在原油含水率预测中,基于粒子群优化RBF神经网络比BP神经网络有着更高的预测精度。

Prediction of Crude Oil Moisture Based on RBF Neural Network Optimized by PSO

Prediction of crude oil moisture is very significant to determine the water and oil level of oil well and estimate the production of crude oil.Recently BP neural network is common prediction method of crude oil moisture.However,the practicability and accuracy are affected due to its drawbacks of falling into local optimization and low convergence rate.Thus,RBF neural network optimized by particle swarm optimization algorithm(PSO-RBFNN) is proposed to predict crude oil moisture in the paper,where particle sw...