Research On Construction Control System For Long-span Continuous Beam Bridge On Multi-Agent And Its Key Technologies

Chinese government would like to develop the transportation infrastructure construction today, and the bridge project construction is an arduous task. The mechanical behavior and seismic behavior of beam bridge are good. The beam bridge can also be built with many different modeling. So most of the bridge designers would like to choose prestressed concrete continuous beam bridge design scheme.The cantilever casting construction method of prestressed concrete continuous beam bridge is much more mature. There are many advantages of cantilever casting construction method, such as low cost, fast construction speed, less equipment, simple process, take up less bridge space and so on. With the heavy task of beam bridge construction, the disadvantages of beam bridge cantilever casting construction have to be paid much attention by the bridge construction masters. Today the span of beam bridge is increasing, the pier is much higher, and there is much more requirement of bridge ride comfort. So in the course of beam bridge construction, the structure stress state, the construction error and the construction safety should be strengthen control.A long-span continuous beam bridge multi-Agent construction control system is built in this paper. The construction control system takes full advantage of the intelligent of multi-Agent system. As a result of that, the construction control system can do positive control to stress、safety and linear in the beam bridge construction. With the distributed collaborative construction control pattern of the construction control system, the beam bridge construction control efficiency is greatly improved. The main work and research result of the paper is as follow:(1) First, the long-span continuous beam bridge construction control system is analysed. The content of analysis is the goodness and methods of construction system optimization. Based on the basic conception of multi-Agent, the long-span continuous beam bridge multi-Agent construction control system is built. Compared with the traditional construction control system, the multi-Agent construction control system is obvious intelligent. The self-adaptive construction control can be really achieved in the multi-Agent construction control system. The system also includes engineering data base, knowledge base and construction state analysis module. There is a new construction control method in the system, which is the combination of the self-adaptive control method and the predictive control method. The new construction control method can improve the reliability of construction control decision by analyzing all of the influencing factors of beam bridge construction.(2) Based on the long-span continuous beam bridge multi-Agent construction control system and real-time control theory, the long-span continuous beam bridge multi-Agent construction control real-time control flow is proposed. There are four subsystems in the control flow, which are construction field monitoring subsystem, construction simulated analysis subsystem, construction state analysis subsystem and parameter identification construction prediction subsystem. In the multi-Agent control flow, the Agent tasks of four subsystems are presented. The Agent internal structure and the cooperative relationship of construction tasks are also presented in the control flow. With the multi-Agent control flow, the beam bridge construction control is easy to implement. Besides, the beam bridge construction control safeguard measure is proposed to ensure construction control system operation.(3) The construction state analysis method based on error correction is studied. Firstly, the construction field monitoring data is analyzed preliminarily, and judges the construction early warning need or not. In the further analysis of beam bridge construction state, the error correction analysis method is temperature effect correction calculation and shear lag effect correction calculation. With the two sides’correction calculation, the construction field monitoring data of structure deformation and structure stress will be more significant. Compared with construction simulation model calculated value and the construction field monitoring data which has been modified, the linear state and stress state of beam bridge can be judged rapidly. The error correction construction state analysis method is more reasonable than the traditional method. The error correction analysis method can also reduce the temperature demand of construction monitoring, which is beneficial to construction of beam bridge.(4) The common methods of parameter identification and construction prediction are analyzed at first. Then many new methods of parameter identification and construction prediction are proposed, such as parameters canonical correlation analysis, GRNN, grey neural network, genetic neural network and combination forecasting method. The parameters canonical correlation analysis method can identify the parameters which have a significant influence on structure linear and stress at the same time, this method can improve the efficiency of parameter identification calculation. Compared with BP neural work, GRNN and grey neural network and genetic neural network have better nonlinear mapping ability, fault tolerance and robustness. Construction combination forecasting model can make use of the superiority of the above three neural network methods, and the higher precision of the combination model has been confirmed in the application of engineering.(5) On the basis of object-oriented technology, the long-span continuous beam bridge multi-Agent construction control system has been realized preliminarily in the construction control of Xijun Bridge. The intelligence of the control system is embodied in practical application. On the combination of the system realization with parameter identification and construction prediction methods which are proposed in this paper, the construction control precision of Xijun Bridge is guaranteed, and the closures of Xijun Bridge are all finished naturally.