STUDY ON DEFORMATION PARALLEL COMBINATION PREDICTION
UNDER DIFFERENT RESTRICTION CRITERION
1)College of Geoscience and Surveying Engineering，China University of Mining and Technology，Beijing 100083
2）Geomatics & City Spatial Information School，Henan University of Urban Construction，Pingdingshan 467036
Abstract：The parallel combination deformation prediction models were studied.According to the weighted constraint criterion，five kinds of deformation parallel combination prediction models were established，including suboptimal weight combination，optimal weigh combination，synthesis relational grade of greyweight，entropyweight and artificial neural networkweight.Based on the characteristics of the models，the effects of parallel combination prediction under different constraint criterion are compared.The results of analysis with real data show that the nonlinear combination prediction based on neural network is the best one.
Gao Caiyun,Cui Ximin,Gao Ning. STUDY ON DEFORMATION PARALLEL COMBINATION PREDICTION
UNDER DIFFERENT RESTRICTION CRITERION[J]. jgg, 2014, 34(3): 91-94.
