Study On Back Analysis And Slab’s Thermal Stress Of Face Rockfill Dam

Nowadays, along with the emphasis of hydropower construction being shifted to the West China gradually, Concrete Faced Rock-fill Dam develops rapidly in recent years due to its some advantages such as using local materials, short project durationt, low cost and good security etc. Especially with the progress of dam construction technology and improvement of construction machinery, it has made remarkable achievements at home and abroad. With the development to300m grade of CFRD, how to control the deformation of rock body effectively in order to avoiding or reducing the panel crack has become one of the key problems of its development. In view of this, under the background of Shuibuya Concrete Faced Rock-fill Dam Project, this thesis has systematically analyzed and studied key problems as well as analysis methods of Concrete Face Rock-fill Dam intelligent back analysis and slab’s thermal stress. The principal research was as follows:(1) Research on Rock-fill Dam displacement back analysis methods. Based on the basic theory and method of inverse analysis, this paper gives intelligent displacement back analysis which is the more commonly used at present a more comprehensive paper, as well as summarizes many methods to solve the problems that results are affected deeply by initial value and its results easy to fall into local extremum when multi-parameter are being optimized existed in traditional back analysis method, and studies the advantages and disadvantages of the displacement back analysis method based on intelligent optimization algorithm-neural network,which utilizes not only nonlinear mapping, network reasoning and prediction function of neural network but also evolution characteristics of Intelligent optimization algorithm. It has a high application value in the treatment of complex project problems that there is no obvious mathematical expression between variables and the objective function value.(2) Intelligent optimization algorithm in inverse displacement analysis. BP neural networks, genetic algorithms, genetic neural network, the basic theory and implementation process of particle swarm optimization and the issues concerned by algorithm are analyzed comprehensively. This paper briefly expounds some experimental design methods constructing the training sample parameter group, as well as solves problems exiting in rock-fill dam back analysis such as iteration convergence slowly, poor stability of the solution, falling into the local minimum value easily, finite element calculation consuming time and labor etc by building inversion analysis model based on particle swarm-genetic algorithm of neural network combined with the rock-fill dam deformation and stress analysis principle. It effectively improves the accuracy and efficiency of rock-fill dam back analysis and has remarkable effect to control the deformation of the dam.(3) The study of concrete-faced temperature stress of Rock-fill Dam. This paper combined with temperature field and numerical calculation theory of temperature stress, considering temperature, placing temperature, temperature rise of cement hydration heat, temperature variation, concrete hardening process and creep while concrete-faced were poured, analyzing temperature field and temperature stress, obtained distribution law of temperature stress in the temperature field, providing technical reference for comprehensive analyzing force condition of concrete-faced and guiding design and construction of concrete-faced.(4) Application of displacement back analysis and temperature stress calculation of panel in Shuibuya CFRD. In view of the above questions research, based on the particle swarm-genetic algorithm of neural network inversion analysis model, this paper carry out Inverse Analysis to Duncan E-B model parameter of Shuibuya CFRD in construction, meanwhile, more excellent parameter group obtained by Inverse Analysis carried out finite element analysis, verified reliability of the model; Meantime, in the situation such as in early January, in late April, no heat preservation plate and heat preservation plate during pouring period of the Concrete-Faced, temperature field and temperature stress which panel came across cooling are carried out calculation and analysis, and analytical results is referential to avoid and decrease panel fracture.