Application And Improvement Of BP Neural Network In Dam Deformation Analysis

As an indispensable water conservancy project in today's society,the reservoir has been playing a huge engineering benefit.At the same time as the core of the project scheduling of flood control,to ensure peace and made a great contribution.Hydroelectric power for economic and social development and industrial production to provide the necessary energy for people's daily drinking water to provide a stable source of power and.The dam is the core of building construction and operation of the reservoir,the dam needs to be considered in the design structure and height,while the height of dam structure and is based on the flood submerging into consideration,sometimes also need to consider the downstream cascade hydropower station.The complexity of the construction environment makes the dam usually bear huge load,prone to various types of migration and deformation,so safety monitoring is very important for the dam.Based on regular monitoring,but also requires a combination of deformation of the dam to analyze and forecast data,so that you can do to prevent early in danger before.Dam deformation is related to many factors,such as the downstream reservoir water level difference,pressure,temperature,time,and because most of these factors has strong randomness,the relationship between the factors is complex,deformation of the dam so that the impact of these factors cannot be described by the exact quantitative relation.At present,many methods of dam deformation monitoring data analysis of diversity,and most of them are long-term deformation monitoring data and analysis to determine the safety status of the dam with the use of statistical models,experiments show that statistical models of defects and deficiencies in dealing with similar dam deformation the nonlinear fuzzy system,analysis of the results of prediction errors can not meet the requirements of reliability and stability.Because the BP network model can deal with the nonlinear mapping problem,it has obvious advantages in dealing with the problem.In this paper,BP neural network model is used to analyze the early warning of dam deformation.At the same time,defects of BP network model for the existence of the traditional genetic algorithm is improved by adding and improved particle swarm optimization algorithm based on traditional BP network model,the use of these two kinds of optimization algorithm to achieve the optimization of the traditional BP network model,and its application in engineering.The results show that the numerical analysis method of genetic algorithm and the improved BP model of particle swarm optimization algorithm compared with the traditional single BP model and analysis method with high accuracy based on the advantages of fault tolerance and generalization ability.