| With the acceleration development of the scientific outlook on development,our country also puts forward the new request to the development of construction projects,"energy saving,water saving,electricity saving,section,material saving,protect the environment,reduce pollution" has become the new trend of the development of the construction project in China.Some scholars found that the factors which restrict the development of the green building such as green building cost,the demand of the policy system of green building,green building environment,etc.,among them,the green building cost is the first factor of restricting the green building large implementation.The cost control of the project runs through the whole life cycle of the project,including the project proposal stage,feasibility study report stage,drawing design stage,bidding stage and construction stage.In order to control the construction cost better,the cost of the project should be estimated first.There have been many methods of predicting project cost in our country,mainly including the unit production capacity estimation,the production ability index method,the langge coefficient method and the proportion estimation method.The project cost,grey prediction model and artificial neural network method are predicted by fuzzy mathematics.Among them,based on artificial neural network prediction method is a hotspot of current,mainly including standard neural network,improve the neural network and the combination of neural network,improved neural network and the combination forecasting method of neural network is used the most.The key of neural network method is the selection of sample engineering and selection of engineering feature vectors.It’s worth noting that the BP neural network and RBF neural network,BP neural network is a feedforward neural network,its work procedures including signal is transmitted and error backward propagation,by constantly adjust the network weights so as to achieve target error of the method,but this method is slow convergence speed,and the disadvantage of easy to fall into local minimum.In order to avoid the disadvantages existing in the BP neural network,RBF neural network has been found out,RBF neural network convergence speed,high precision,has the advantages of good push universality,this paper is to use RBF neural network for green building engineering cost estimation.In this paper,the RBF neural network model is constructed by using the MATLAB neural network toolbox based on the extensive reading of domestic and foreign literatures and related experts.In the process of building model,you first need to accurately select engineering feature vector,this paper use the analytic hierarchy process(ahp)to select engineering feature vector,and then the selected 36 group similar already built green building engineering sample data for network training and simulation,using the actual data validation and the feasibility and the advanced nature of the RBF neural network to confirm the method in the application prospect of green building engineering cost estimation.It is hoped that the research results of this paper can be applied in the estimation of actual green construction project cost,so as to play an active role in promoting the green building in China. |
