Intelligent And Informationization Research On The Key Technology For Soft Ground Improvement Of Highway

The soft ground improvement of highway is a complicated system engineering, in the course of ground improvement, the key issues are as follows:correct understanding of the characteristics of soft soil, select the appropriate treatment scheme, monitor the settlement and stability of embank. Today, the decision for soft ground improvement is mainly rely on the knowledge and experience of people, the science and reliability is not enough. In order to better solve the key issues and enhance the science and reliability for soft ground improvement, combined with the feature of soft ground improvement knowledge, and based on the theory of artificial intelligence, knowledge engineering, mathematical statistics, fuzzy mathematics, neural net, grey theory, information technology etc, the intelligent and informationization technique for key issues in highway soft ground improvement is studied. The main content of this paper including the followings six aspects:1.Based on analyzing the change confines, average value, variability, relationship, probabilistic distribution characteristics for physical and strength indexes of Tianjin soft soil, the district empirical relationship formulas and probabilistic distribution models of soil parameters are estabilshed for Tianjin soft soil. At the same time, the correlation of shear strength results by straight shear test, triaxial test, vane shear test are studied, and a method for using random factor to evaluate the uncertainty involved in different test methods of soft soil parameter is proposed.2.Aimed at existing various influential factors, massive indeterminacy knowledge in the highway soft ground improvement, the fuzzy reasoning model about scheme decision of soft ground improvement is confirmed. The comment membership of different grades and the weight values of various influencing factors are determined. Moreover, uses the average weighting operator and the multi-layer fuzzy synthetic decision model to optimize the model, all of these have realized effectively reasoning and evaluation for schemes decision of highway soft ground improvement, which make up the weak point about experience decision, and improve the reliability level of decision.3.The paper established BP neural network model for the scheme decision of highway soft ground. Based on the design information, the parameters, algorithm, weitht of network are confirmed by training with massive data. Because the network model input parameter’s indexes were divided into multiple levels, so the reasoning ground improvement scheme is specific. The result shows the model is reliable, can meet the needs of design and decision for soft soil ground improvement.4.Aimed at existing massive indeterminacy and unknown in the improvement of highway soft ground, the model based on grey theory for the scheme decision of highway soft ground improvement is proposed, uses grey analysis method, by calculating the ratio of the correlation between the alternative scheme and the ideal scheme, the best soft soil ground improvement scheme can be obtained, and the scheme optimization of soft ground improvement under the multiple factors is solved also.5.In this paper, a long distance monitor and manage system for embank settlement is developed, which does not interfere with construction and traffic. At the same time, the distinguishing rule with monitoring index, monitoring prediction model, monitoring curve for the noise and real exceptional monitored data of the settlement is confirme. All of these have provided better informationization and intelligent technology of settlement and stablity monitoring for soft soil ground embank.6.Based on the embank settlement dataes, the influence of expression fitting parameters of hyperbola method, Xingye method, Asaoka method to settlement prediction results are discussed, by adding the time factor index to modify verhulst curve model, and has trained reliable neural network model of settlement prediction for soft ground improvement embank. All the results have solved the settlement prediction question in soft ground improvement embank.