| Water conservancy projects are important people’s livelihood projects with long construction periods and large investment scales.They are easily affected by natural conditions such as topography,geology,and hydrology.At the same time,they are also restricted by local economic development and other market conditions.Improving the quality of construction cost management of water conservancy projects is an important way to ensure that the project achieves its pre-determined economic goals and gives full play to its economic benefits.Due to the strong public welfare nature of water conservancy projects,under the current cost management model,the phenomenon of "final accounts exceeding budget,budget exceeding estimates,and estimates exceeding cost estimation" occurs from time to time,and sometimes even costs are out of control.Therefore,strengthening the estimation and control of construction costs and scientifically judging the cost status of the project is an important part of the cost management of the projects,and it is also a scientific method to improve the level and quality of project cost management.Aiming at the small sample characteristics of the construction cost estimation problem of water conservancy projects,this paper adopted the Least Square Support Vector Machine algorithm(LS-SVM)as the core algorithm of the cost estimation model to obtain the complex nonlinear relationship between the cost and its influencing factors.The key cost influencing factors obtained by using the factor analysis and sensitivity analysis methods comprehensively were used as the input vectors of the model.The Levy flight algorithm was also used to improve the generation way of the new food resource at the scout stage in the basic Artificial Bee Colony(ABC)algorithm,and the values of the core parameters of the model were determined by using the improved IABC algorithm.The validity and scientificity of the model were verified after training and testing the cost estimation model by the actual collected engineering case data.At the same time,on the basis of the cost estimation model constructed in this paper,the cost control interval was constructed based on the acceptable project cost change percentage value η.As a result,the cost control model was constructed.Meanwhile,the idea of realizing the dynamic cost control of project life cycle was realized based on the cost estimation model and control model constructed in this paper.The verification results of engineering examples showed that the Mean Absolute Percentage Error(MAPE)of the cost estimation model constructed in this paper was less than10%,and the modeling time of the model was about 17.0s,indicating that the cost estimation model constructed in this paper can make full use of the advantages of various algorithms to achieve the improvement of the estimation accuracy and modeling efficiency.At the same time,the constructed cost control model can effectively monitor the changes in project construction costs,and the cost status and management level of the projects can be determined by judging the relative position relationship between the project completion cost and the cost control interval.In this paper,machine learning algorithm and bionic intelligent algorithm are introduced into the problem of the cost estimation and control of the water conservancy projects.The improvement of the accuracy of project construction cost estimation has a certain driving effect on the improvement of the level of project cost management.It also provides a more convenient way and a more scientific reference for achieving the high cost management quality of water conservancy projects.At the same time,the proposal of the dynamic control method framework for construction costs of the projects based on the entire life cycle of the project can provide designers,construction organization staff and project cost management personnel with a powerful tool to help them obtain more valuable cost management information. |
PDF Full Text Request