摘要
通过对城市天然气中长期历史消费情况的分析研究,建立了模拟城市天然气消费的趋势外推预测模型——龚珀兹负指数函数模型,并采用最小二乘法和人工神经网络对该模型进行求解。实例计算结果表明:趋势外推模型能够模拟城市天然气中长期消费的变化趋势,在根据最小二乘法、BP算法、微粒群算法(PSO算法)训练神经网络求解得到的3个预测模型中,采用PSO算法得到的模型精度高、耗时少,能较准确地反映城市天然气中长期消费情况。
According to historical data of the city natural gas medium-long term consumption, a trend exploration prediction model based on simulating city natural gas consumption, named Gompertz negative exponential function model, has been established. Then, least square method and artificial neural network are used to solve this model. The calculated results of a practical example show that the established trend extrapolation model can simulate the medium-long term natural gas consumption changes. Among the three model which gained by using least square method, BP algorithm method, PSO algorithm method, PSO algorithm method can more accurately reflect the medium-long term city natural gas consumption trend with high accuracy, less time-consuming.
出处
《天然气技术》
2007年第6期77-79,共3页
NATURAL GAS TECHNOLOGY
基金
四川省高校重点学科建设资助项目(SZD0416)
关键词
天然气消费
预测
数学模型
算法
Natural gas consumption Prediction Mathematical model Algorithm
