Research On The Healthy Monitoring Model Of Large Aqueduct In The Middle Route Of South-to-North Water Transfer Project

To compensate for the lack water resources in north areas, the large water-diversion project between different drainage basin—South-to-North Water Transfer Project was carried out. The project is characterized by large scale, high investment, numerous building and difficult technique requirements. It has thousands of canal structure and aqueduct is a very important part. Open-channel free water collection was adopted in general main canal of the middle route of The South-to-North Water Transfer Project. The section size of the channel and structure is large. Because of the complicated working condition and load, it is very difficult to analysis and confirms its boundary condition, loads and the relationship between structure influencing quantity and external condition. A lot of indeterminate factors exist at the construction and operation stage, so there are many differences of aqueduct between operation period and the design condition. The safe operation of aqueduct will be influenced by the aging, fatigue and the abrupt earthquake etc. So the healthy monitoring, precise analysis and evaluation are needed during diagnose the healthy condition of the aqueduct timely. The safety construction and operation period need to be insured and significant accident, huge economic loss and bad society influences need to be prevented.In this paper, the purpose of the aqueduct healthy monitoring, research situation and main content of this paper are introduced briefly in chapter 1. The monitoring system, contents, data analysis and several often-used healthy monitoring models are introduced in chapter 2. In chapter 3, large finite element computer software ANSYS and temperature loads effect to the bulk concrete hydraulic structure are introduced briefly; the general situation of the Ming river aqueduct engineering is introduced and its ANSYS finite element model is constructed. In the fourth part of this paper, representative operation performances are chosen by the method of homogeneous design. Structural analysis to the aqueduct is carried out through compile APDL script language of ANSYS and the monitoring result of crucial monitoring point is extracted as sample data to build determinacy model. In the fifth part of this paper, traditional multiple linear regression method is introduced to build the determinacy healthy monitoring model of the aqueduct, and genetic algorithm method is used to optimize the model primarily. In chapter 6, Radial-basis Function (RBF) neuron network is also used to optimize the healthy monitoring model, so the precision of prediction of model is raised and nonlinear relationship between aqueduct displacement, stress and aqueduct water level is explained, temperature load etc. Summary and a long view of the research are made at the end of the paper.