Mathematical Model For Dam Safety Monitoring And Dynamic Inverse Model For Power House

The performance characteristics of hydraulic structures are affected by many external factors. The estimate of their real operating states require the evaluation of many parameters such as the elastic modulus, the damping ratio, the friction coefficient, etc, which are extremely difficult to determine with traditional calculation methods or model test. A solution to overcome the difficulty could to be the combination of mathematical model and inverse finite element analysis based on the observed data.As a project financed by Yunfeng and Ming Tomb Hydroelectric Power Station, the study on "Mathematical Model for Dam Safety Monitoring and Dynamic Inverse Model for Underground Power House" was conducted. According to the data observed in-situ, the under-fitting problems in regression models for dam safety monitoring were discussed, and the computational parameters and boundary conditions of underground structures were determined on the base of dynamic inverse mathematical model.No. 15 fault of Yunfeng Dam is a low angle one. Its space structure is very complex. Had not been disposed thoroughly during the period of construction, it has become the control factor in Yunfeng Hydroelectric Project. Based on rigid body method and finite element method, the deep stability against sliding of Yunfeng Dam was analyzed in details. The results show that its safety factor along the fault is on the low side. On account of some parameters in the computation modules were indeterminate, the results obtained could not reflect the real safety situation of Yunfeng Dam. It is necessary to built regression models for dam safety monitoring of Yunfeng Dam to evaluate its operational reliability.Under-fitting problems usually appear in regression models for dam safety monitoring. To overcome those problems, firstly, integral regression temperature factors and periodic time-effect factors were introduced to expand the classical factor set. Then a genetic algorithm (GA) was proposed using a real parameter coding, a ranking selection operator, an arithmetical crossover operator and a uniform mutation operator, and calculated the least-square error of the observed and computed values as its fitness function. The elitist strategy was used to improve the convergence. After that, the modified genetic algorithm was applied to reassess the coefficients of the regression model and a genetic regression modelI?was set up. Based on the two methods above, the observed data of Yunfeng Dam was anglicized. The computational results show that under-fitting problems were solved perfectly. And the periodic orderliness of the horizontal displacement of dam crest indicates that Yunfeng Dam is in a good condition at present situation.The vibration problems in pumped storage hydroplant are outstanding, due to their high-velocity flow, high-speed units, etc. On the base of summarizing the development of inverse analysis methods, this paper proposed a dynamic inverse model for underground hydroplant was built, combining the modal analysis and inverse numerical computation. To optimize the complex structures, the improved genetic algorithm was jointed to the ANSYS Code and an optimal genetic dynamic inverse model was obtained. Make use of these achievement and test data, the numerical computation of Ming Tomb Underground Power Station was carried out, and its average elastic modulus and the elastic resisting force of surrounding rock were obtained.