论文部分内容阅读
针对加热输油管道仿真中管道内油流温度和压力变化,本文基于神经网络算法,利用管道数据采集与监控(SCADA)系统获取的500组历史运行数据,建立了管道沿线温度和压力的预测模型.提出的预测模型采用混沌粒子群改进的RBF(CPSO-RBF)神经网络算法.对RBF神经网络的参数(中心和宽度)、连接权重进行优化,通过与其他方法对比可知提出的CPSO-RBF预测模型具有精度高、收敛快等特点.在日照-仪征热油管道实际运行方案中验证了提出的CPSO-RBF预测模型的可行性.
Aiming at the change of temperature and pressure of oil flow in pipeline during heating oil pipeline simulation, based on the neural network algorithm, this paper establishes a forecast model of temperature and pressure along the pipeline by using 500 historical operating data acquired by pipeline data acquisition and monitoring (SCADA) system. The proposed prediction model uses the improved RBF (CPSO-RBF) neural network algorithm based on chaos particle swarm optimization to optimize the parameters (center and width) and the connection weights of the RBF neural network. By comparing with other methods, we can see that the CPSO-RBF prediction model It has the characteristics of high accuracy and fast convergence.It verifies the feasibility of the proposed CPSO-RBF prediction model in the practical operation plan of Rizhao-Yizheng thermal oil pipeline.