Profile Design Of 100-meter Grade Cementitious Sand Gravel Dam Based On Artificial Bee Colony Algorithm And Genetic Algorithm

As a new kind of dam,the cementite gravel dam has many advantages such as saving materials,fast construction progress,high economic efficiency and adapting to different types of foundations.It has good prospects in the application of water conservancy projects in our country.The structural characteristics and operation mechanism of cementified gravel dam have been studied at home and abroad,but there are few researches on the profile design and stability and safety prediction of the 100-meter cementified gravel dam,and there is no actual project of the 100-meter cementified gravel dam in China.Therefore,it is more important to study the stability and safety of the 100-meter cementified gravel dam with different profile forms and cementified materials.In this paper,the finite element analysis and response surface method are used to create a large number of data sets,and the regression equation is established for the data.The artificial bee colony algorithm and genetic algorithm are used to obtain the maximum principal stress,vertical displacement,riverside displacement and overload safety factor under the same dam height,different dam profile forms and different cementing materials,and the optimization results of the two algorithms are compared.On this basis,a design scheme with the same optimization results of the two algorithms is selected to simulate the operation mechanism of the 100-meter cementitious gravel dam under the homogeneous foundation by using finite element and structural model test.Meanwhile,the failure process and safety degree of the 100-meter cementitious gravel dam are obtained by using the overload method,and the accuracy and reliability of the regression equation of the prediction model are verified.The specific research contents are as follows:(1)Using finite element analysis and response surface method to create a large number of data sets and establish the regression equation,establish the finite element model of different dam height profile forms,extract the maximum principal stress,vertical displacement and horizontal displacement under normal operation,and use the overload method to obtain the overload safety factor,use the response surface method to create the regression equation of the prediction model.(2)On the basis of the established regression equation,it is taken as the objective function.MATLAB artificial colony algorithm and genetic algorithm are used to optimize and compare the dam profile and cementing material.Both algorithms generate initial population and optimize it according to geometric constraints.In the artificial colony,the lead bees will search for a new nectar source and replace the original nectar source with a better nectar source.In the genetic algorithm,the initial population is selected,crossed and mutated to form a new population.In the aspect of data presentation,it is found that the artificial bee colony algorithm is easier to fall into the local optimal solution than the genetic algorithm,but at the same time,the genetic algorithm is more complicated than the artificial bee colony algorithm with more control parameters.A group with the same optimization results is selected from the results of the two algorithms for the 100 m dam height optimization profile form as the subsequent research object.(3)The finite element software is used to apply the load,boundary constraint and the indoor parameters of materials in the actual project to simulate the operation of the dam body in the actual project.The overload method is used to obtain the maximum principal stress,minimum principal stress,riverside displacement,vertical displacement and the changing trend of the dam body strain.The overload safety factor of the dam body is Kp=5.0.This is verified with the overload safety factor Kp=5.08 calculated by the optimization algorithm,indicating that the regression equation and optimization results are reliable.(4)The overload method is used to carry out structural model test on the model.According to the displacement and strain data of each measuring point measured by the measuring equipment,the failure mechanism,failure form and overload safety factor of the cement-gravel dam are obtained.The final results show that they are consistent with the displacement,strain change trend and failure form obtained in the finite element test.Verify the rationality of the prediction model and the optimization of the profile form of the cementitious sand gravel dam.