Research On Cost Forecast Of Transmission And Transformation Project Based On Artificial Neural Network

The cost of grid engineering is a multivariable,highly nonlinear problem.In the past,the prediction of the cost of power transmission and transformation project mainly depends on the practical analysis and operation of the technical personnel with many years of practical experience in this field.However,the cost of grid engineering is a multivariable,highly non-linear problem,especially when the engineering situation is complex and difficult,it is difficult to obtain empirical results through a single project to ensure reliable results to guide the project cost control.The unit capacity cost in the substation project and the unit cost in the transmission project are the main technical and economic indicators which are the most concerned parties of the two investors and the construction unit.It is also the core index of the cost management and control of the power transmission project.Therefore,investors and construction units urgently need an ideal forecasting method to make use of the historical cost data of the built engineering to quickly predict the main technical and economic indicators of the new power engineering in order to make the construction plan reasonably and strive for the active time for the power engineering construction,To improve the efficiency of project funding and project review the quality of the project to guide the cost of new power construction projects.In this paper,the data preprocessing technology of power transmission and transformation project is studied.Based on the specific characteristics of engineering cost history data,the data preprocessing method of power transmission and transformation project including data cleaning,data conversion and data reduction is proposed.Power transmission project as a case simulation,verify the validity of the data preprocessing method.Then,based on the data of power transmission and transformation project data,an easy-to-operate,fast and effective model of power transmission and transformation project is proposed,namely MEA-BP cost forecasting model.Among them,artificial neural network algorithm in the small sample learning field performance is very advantageous,but also suitable for a limited number of projects,a considerable number of factors affecting the cost of power transmission project to learn.Thinking evolutionary algorithm as a powerful parameter optimization algorithm,in the model of BP artificial neural network parameters to optimize.Finally,an example is used to simulate the simulation.It is proved that the prediction model has a large improvement in accuracy compared with the traditional cost measurement method.The simulation of power transmission and substation project shows that the system can achieve the project cost management effectively and effectively.Smooth implementation of the provision of technical support.