Abstract
Prediction of soft soil settlement is an important research topic in the field of civil engineering, and the least square support vector machine is one of the commonly used prediction methods at present. Nonetheless, the existing LSSVM models have problems of low search efficiency in the search process and lack of global optimal solution in the search results. In order to solve this problem, based on the leave-one-out cross-validation method, the homotopy continuation method was used to optimize the LSSVM model parameters, and then the HC-LSSVM model was constructed with the goal of minimizing the sum of squares of the prediction error of the full sample retention one. Finally, the rationality and correctness of the model are verified by engineering application. The results show that the HC-LSSVM model constructed in this study can accurately predict the settlement of soft ground, which is superior to the common LSSVM model and solves the problem that the parameters of LSSVM model cannot be solved optimally. The research results provide a new method for prediction of soft soil settlement.
Highlights
More and more buildings, roads and railways are built on soft soil, and their construction schemes and use functions require more accurate determination of their settlement in the construction period and operation period [1]
In order to make full use of the advantages and overcome the disadvantages of the two methods of Least Squares Support Vector Machines (LSSVM) model parameter solving, this study proposes the HCLSSVM model: based on the leave-one-out cross-validation method, the optimization problem of LSSVM model parameters is transformed into the solution problem of nonlinear equations with the goal of minimizing the sum of squares of the prediction error of full sample retention one, and considering that the homotopy continuation method is an effective method to solve nonlinear equations in a large range of search [20,21], the homotopy continuation method is adopted to solve the nonlinear equations, and the results of solution are taken as the optimal parameters of LSSVM model
The HC-LSSVM model of soft soil settlement based on homotopy continuation method is mainly expressed as follows: based on the observation data of soft soil settlement, the training sample set and test sample set are established, and on the basis of leave-oneout cross-validation method, the optimization problem of LSSVM model parameters is transformed into solving nonlinear equations problem with the goal of minimizing the sum of squares of prediction error for the whole sample
Summary
Roads and railways are built on soft soil, and their construction schemes and use functions require more accurate determination of their settlement in the construction period and operation period [1]. Chapelle et al used a leave-one-out cross-validation method and support vector counting to optimize SVM parameters: the leave-one-out cross-validation method divided the sample set into a training sample set and a test sample set, and the minimum statistical index of test error rate of SVM for many times was used as the criterion of optimization parameters; the support vector counting method takes the minimum ratio of the number of support vectors to the total number of samples as the criterion of SVM parameter optimization [19] Both methods have their advantages and disadvantages in solving model parameters: the first method solves parameters by intelligent method or optimization method, which can comprehensively search the optimal solution of model parameters. This study tested the model through the measured data of soft soil settlement, and the test results proved that the LSSVM model had a good optimization result and the LSSVM model is a stable model with a good prediction result in the selection of hyperparameters (Figure 1)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
