Combining Forecasting Model of Pipelines Corrosion Rate Based on Artificial Immunea Algorithm
The forecast accuracy of combining model can be substantially improved,for it benefits from the individual forecasts on different levels of the input data space.This paper provides a description of forecast combination techniques,and puts forward a combining forecasting model of nonnegative time variant weights.The combination forecasting function is a multi-peak value,and Artificial Immune Algorithm is applied to search multi-modal functions multi-extremism.Then the combining forecasting model of nonnegative time variant weights is applied to forecast the corrosion rate of the certain pipeline,and the error rates obtained from individual models and combining model show that the latter achieves an improvement of forecast quality up to 95 percent.That indicates the best results can be achieved by using combination models,and Artificial Immune Algorithm has validity and accuracy to solve the combination forecast problems.
Corrosion Rate Combining Forecasting Model Forecast Accuracy Artificial Immune Algorithm
Tao Zhang Lizhen Zhou
Engineer, Engineering Faculty, China University of Geosciences; No.388, Lumo Road, Wuhan 430074 Lecturer, Engineering Research Center of Rock-Soil Drilling & Excavation and Protection, Ministry of
国际会议
北京
英文
113-119
2012-10-14（万方平台首次上网日期，不代表论文的发表时间）