Water conservancy project cost early-stage evaluation method

Abstract

The invention relates to the field of hydraulic engineering cost evaluation, in particular to a hydraulic engineering cost early-stage evaluation method, which comprises the steps of data acquisition and processing: collecting data related to hydraulic engineering cost from multiple channels by using a big data technology, including historical project data, market information and project quota, the data quality is improved by means of data cleaning and preprocessing; through index screening, the novel evaluation method can make full use of big data and artificial intelligence technologies, the acquisition and processing capability of evaluation data is improved, and thus the evaluation accuracy is improved; according to the novel evaluation method, the evaluation process is simplified, the evaluation efficiency is improved, and the evaluation period is shortened; the novel evaluation method can perform adaptive adjustment for hydraulic engineering projects of different types and different scales to meet actual requirements; the proposal and application of the novel evaluation method are helpful for promoting the innovative development of the water conservancy project cost evaluation industry.

Description

Technical Field

[0001] The invention relates to the technical field of water conservancy project cost assessment, in particular to a water conservancy project cost pre-assessment method. Background Art

[0002] With the rapid development of my country's economy and society, water conservancy project construction, as an important part of the national infrastructure, has seen its investment scale continue to expand, which has put forward higher requirements for the assessment of project costs. In order to ensure the smooth implementation of water conservancy projects, reasonably control investment costs, and improve investment benefits, it is crucial to conduct accurate cost assessment of water conservancy projects.

[0003] At present, the objects of water conservancy project cost assessment are wide, the assessment methods are diverse, the assessment basis is complex and there are many factors affecting the assessment results. These all affect the preliminary cost of water conservancy projects. At the same time, there are problems with traditional assessment methods. Traditional assessment methods rely on a large amount of historical data, but in actual projects, there is often a lack of complete and reliable data sources, resulting in distorted assessment results; traditional assessment methods have poor adaptability when dealing with water conservancy projects of different types and scales, and it is difficult to meet actual needs; traditional assessment methods require the collection, organization and analysis of a large amount of data, and the assessment cycle is long, which affects the efficiency of project decision-making; due to problems with data sources and method adaptability, traditional assessment methods have low accuracy, which may lead to investment decision-making errors. Therefore, it is necessary to propose a preliminary assessment method for water conservancy project cost. Summary of the invention

[0004] In view of the problems in the prior art, the present invention provides a method for pre-evaluation of the cost of a water conservancy project.

[0005] The technical solution adopted by the present invention to solve its technical problem is: a method for preliminary evaluation of water conservancy project cost, including data collection and processing: using big data technology to collect data related to water conservancy project cost from multiple channels, including historical project data, market information, and engineering quotas, and improving data quality through data cleaning and preprocessing means; Constructing the indicator system: completing the core indicator system through indicator screening, indicator classification and indicator weight determination. The indicator system is the core method of the new evaluation, which determines the comprehensiveness and scientificity of the evaluation; Feature extraction and selection: Through machine learning algorithms, features that have a significant impact on the evaluation results are extracted from the original data, and features with higher predictive power are selected for modeling; Model construction and training: Build a theoretical model based on deep learning technology and train the model with a large amount of historical data to enable it to accurately predict project costs; Evaluation and optimization: Apply the trained model to new engineering projects to conduct cost evaluation, and optimize the model based on the evaluation results to improve the accuracy of the prediction.

[0006] The data collection and processing include data collection and data processing; The data collection collects data related to the cost of water conservancy projects through various channels, including but not limited to: Cost data of historical engineering projects; engineering design drawings and related technical parameters; market price information, such as material prices and labor rates; policies, regulations and industry standards; various information sources on the geographical environment and socio-economic conditions of engineering projects; The data processing performs preprocessing on the collected data, including data cleaning, deduplication, filling missing values, and standardization to ensure the quality and availability of the data.

[0007] The model construction and training theory model is constructed through model selection, model structure design, model training, model verification and testing, and model deployment and application; The model selection is based on the project characteristics and evaluation requirements, and selects appropriate machine learning algorithms, such as decision trees, random forests, support vector machines, and neural networks; The model structure design designs the network structure of the model, including the input layer, hidden layer and output layer, as well as the number of neurons in each layer and the connection method; The model training uses historical data to train the model and adjusts the model parameters through an optimization algorithm so that the model can accurately predict the project cost; The model validation and testing is performed by cross-validation and test set validation to evaluate the performance and generalization ability of the model; The model deployment and application deploys the verified model to the actual project, conducts cost evaluation, and guides project decision-making based on the evaluation results.

[0008] Indicator selection in the construction of the index system is based on the characteristics of the water conservancy project and the purpose of evaluation, and key indicators that can reflect the project cost are selected, such as project scale, project structure, regional economic level, and technical difficulty; The indicator classification divides the selected indicators into direct impact indicators and indirect impact indicators, as well as quantitative indicators and qualitative indicators, to construct a clearly hierarchical indicator system; The indicator weights are determined by expert consultation and analytic hierarchy process to reflect the impact of each indicator on the project cost.

[0009] Said evaluation and optimization is performed through data mining and analysis, application of artificial intelligence algorithms, and optimization of model parameters; The data mining and analysis are the basis of a new evaluation method, and the data mining and analysis include data mining, data analysis and data visualization.

[0010] The application of artificial intelligence algorithm is the core of the new evaluation method, which can handle complex data relationships and improve the intelligence level of evaluation. The application of artificial intelligence algorithm includes machine learning algorithm, deep learning algorithm and reinforcement learning algorithm.

[0011] The model parameter optimization is the key to improving the performance of the new evaluation method. It involves adjusting and optimizing the model parameters to achieve more accurate evaluation results. The model parameter optimization includes parameter optimization, model parameter adjustment, model integration and hyperparameter optimization.

[0012] A method for using a preliminary evaluation method for water conservancy project cost includes the following steps: Step 1: The preparation stage is the basis for the implementation of the new evaluation method. The purpose of this stage is to ensure the smooth progress of the evaluation process; clarify the evaluation objectives, scope and specific requirements, analyze the project characteristics, and determine the key points and difficulties of the evaluation; according to project requirements, form an evaluation team composed of data analysts, engineers, and economists; collect and organize the data required for the evaluation, including historical project data, market information, technical parameters, and pre-process the data; select appropriate data mining and artificial intelligence algorithms to build a preliminary evaluation model framework; Step 2: The implementation phase is the core of the new evaluation method. The tasks of this phase are to build and run the evaluation model; use data mining technology to conduct in-depth analysis of the collected data and extract useful information and features; build an evaluation model based on the results of data analysis and technology selection, and use historical data to train the model; verify the model through cross-validation and test sets to evaluate the performance and generalization ability of the model; prepare a detailed evaluation report based on the model evaluation results, including the evaluation process, result analysis, and suggestions; Step 3: The purpose of the verification and adjustment phase is to ensure the accuracy and reliability of the evaluation model and make necessary adjustments to the model; apply the model to new engineering projects, compare the actual results with the model prediction results, and verify the accuracy of the model; collect feedback from users and experts on the evaluation model and evaluation report; adjust and optimize the model parameters based on the feedback and actual operation conditions; deploy the verified and adjusted model to actual projects to guide investment decisions and cost control of engineering projects; continuously monitor the operation of the model, and regularly evaluate and update it to adapt to new changes and needs.

[0013] Beneficial effects of the present invention: The present invention discloses a method for preliminary assessment of the cost of a water conservancy project. The new assessment method can make full use of big data and artificial intelligence technology to improve the collection and processing capabilities of assessment data, thereby improving the accuracy of the assessment; the new assessment method simplifies the assessment process, improves the assessment efficiency, and helps to shorten the assessment cycle; the new assessment method can be adaptively adjusted for water conservancy projects of different types and sizes to meet actual needs; the proposal and application of the new assessment method will help promote the innovative development of the water conservancy project cost assessment industry. DETAILED DESCRIPTION

[0014] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0015] The present invention discloses a method for preliminary evaluation of the cost of a water conservancy project, comprising data collection and processing: using big data technology to collect data related to the cost of a water conservancy project from multiple channels, including historical project data, market information, and engineering quotas, and improving data quality through data cleaning and preprocessing; constructing an indicator system: completing a core indicator system through indicator screening, indicator classification, and indicator weight determination. The indicator system is the core method of a new type of evaluation, which determines the comprehensiveness and scientificity of the evaluation; feature extraction and selection: extracting features that have an important impact on the evaluation results from the original data through a machine learning algorithm, and selecting features with higher predictive capabilities for modeling; model construction and training: constructing a theoretical model based on deep learning technology, and training the model through a large amount of historical data so that it can accurately predict the project cost; evaluation and optimization: applying the trained model to a new project to perform cost evaluation, and optimizing the model based on the evaluation results to improve the accuracy of the prediction.

[0016] The data collection and processing include data collection and data processing; the data collection collects data related to the cost of water conservancy projects through multiple channels, including but not limited to: cost data of historical engineering projects; engineering design drawings and related technical parameters; market price information, such as material prices, labor rates; policies, regulations and industry standards; various information sources on the geographical environment and socio-economic conditions of engineering projects; the data processing pre-processes the collected data, including data cleaning, deduplication, filling in missing values, and standardization to ensure the quality and availability of the data.

[0017] The model construction and training theory model is constructed through model selection, model structure design, model training, model verification and testing, and model deployment and application; the model selection selects appropriate machine learning algorithms such as decision trees, random forests, support vector machines, and neural networks according to project characteristics and evaluation requirements; the model structure design designs the network structure of the model, including the input layer, hidden layer, and output layer, as well as the number of neurons in each layer and the connection method; The model training uses historical data to train the model, and adjusts the model parameters through optimization algorithms so that the model can accurately predict the project cost; the model verification and testing verifies the model through cross-validation and test sets to evaluate the performance and generalization ability of the model; the model deployment and application deploys the verified model to the actual project, conducts cost evaluation, and guides project decisions based on the evaluation results.

[0018] In the construction of the index system, the index screening is based on the characteristics of the water conservancy project and the evaluation purpose, and the key indicators that can reflect the project cost, such as project scale, project structure, regional economic level, and technical difficulty, are screened out; the index classification divides the screened indicators into direct impact indicators and indirect impact indicators, as well as quantitative indicators and qualitative indicators, to construct a hierarchical index system; the indicator weight determination determines the weight of each indicator through expert consultation and hierarchical analysis method, reflecting the degree of influence of each indicator on the project cost.

[0019] Said evaluation and optimization is performed through data mining and analysis, application of artificial intelligence algorithms, and optimization of model parameters; The data mining and analysis is the basis of the new evaluation method, and the data mining and analysis includes data mining, data analysis and data visualization; The data mining uses data mining techniques to analyze the collected historical engineering data, including association rule mining, cluster analysis, classification analysis and predictive analysis, to discover potential relationships and patterns between data; The data analysis uses statistical analysis, principal component analysis, and factor analysis methods to conduct in-depth analysis of the data set and extract key factors and features that affect the project cost; The data visualization uses data visualization tools to display the analysis results in the form of graphics or charts to help decision makers intuitively understand the data and analysis results.

[0020] The artificial intelligence algorithm application is the core of the new evaluation method, which can handle complex data relationships and improve the intelligence level of evaluation. The artificial intelligence algorithm application includes machine learning algorithm, deep learning algorithm and reinforcement learning algorithm; The machine learning algorithm applies machine learning algorithms, linear regression, decision tree, random forest, and support vector machine to predict the project cost; The deep learning algorithm uses deep learning algorithms, such as convolutional neural networks, recurrent neural networks, and long short-term memory networks, to process unstructured data and extract high-dimensional features; The reinforcement learning algorithm uses the reinforcement learning algorithm to train the model in a specific evaluation environment, so that the model can optimize the evaluation results by self-learning and adjusting strategies.

[0021] The model parameter optimization is the key to improving the performance of the new evaluation method. It involves adjusting and optimizing the model parameters to achieve more accurate evaluation results. The model parameter optimization includes parameter optimization, model parameter adjustment, model integration and hyperparameter optimization. The parameter optimization searches for the optimal parameter combination of the model through grid search, random search and gradient descent method. The model parameter adjustment is based on the performance of the model on the validation set, and the model adjusts parameters, such as learning rate, number of hidden layer nodes, and regularization term, to improve the prediction performance of the model; The model integration combines the results of multiple models through integrated learning methods, such as Bagging and Boosting, to improve the robustness and generalization ability of the model; The hyperparameter optimization: uses Bayesian optimization and genetic algorithm advanced optimization techniques to optimize the hyperparameters of the model to obtain the best model performance.

[0022] Through the comprehensive application of the above key technologies, the new evaluation method can achieve accurate evaluation of water conservancy projects and provide a scientific basis for investment decisions and cost control.

[0023] Specific use: Preliminary preparation Clarify the evaluation objectives: Determine the evaluation objectives according to project requirements, such as investment estimation, cost control, benefit analysis, etc.; Form an evaluation team: a professional team consisting of data analysts, engineers, economists, etc. is formed to ensure the professionalism and comprehensiveness of the evaluation; Data collection and processing: Collect historical data, market information, technical parameters, etc. related to the project, and perform data cleaning and preprocessing; Indicator system construction: According to the project characteristics and evaluation objectives, an evaluation system including quantitative and qualitative indicators is constructed; Evaluation model selection: Select an appropriate evaluation model, such as regression analysis, neural network, etc., based on data characteristics and evaluation requirements.

[0024] Implementation steps Model building: Using the collected data, the selected evaluation model is trained, including parameter optimization and model adjustment; Evaluation and prediction: Apply the model to the project to be evaluated to predict its construction cost; Result analysis: Analyze the evaluation results, evaluate the difference between the predicted value and the actual value, and analyze the reasons; Evaluation report preparation: Prepare a detailed evaluation report based on the evaluation results, including the evaluation process, result analysis, recommendations, etc.; Model validation: Apply the model to new engineering projects to verify its predictive ability and adaptability.

[0025] Post-maintenance Model update: Regularly update the evaluation model based on the actual project situation and new data to improve the accuracy of the model; Result feedback: Collect user feedback on evaluation results, analyze existing problems, and provide a basis for model optimization; Technical support: Provide users with necessary technical support to help solve problems encountered during use; Continuous improvement: Combine user feedback and technological development to continuously optimize evaluation methods and models and improve evaluation results.

[0026] Operation method: Collect various data related to the project to be evaluated, including historical project data, market information, technical parameters, etc.; use historical data to train the evaluation model, adjust model parameters, and improve prediction accuracy; apply the trained model to the project to be evaluated to predict its construction cost; compare the predicted value with the actual value, analyze the cause of the error, and optimize the model; based on the evaluation results, prepare a detailed evaluation report to provide a basis for project decision-making.

[0027] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for early stage evaluation of water conservancy project cost, characterized in that: include, S1. Data collection and processing: Using big data technology, collect data related to water conservancy project costs from multiple channels, including historical project data, market information, and engineering quotas, and improve data quality through data cleaning and preprocessing; S2. Construct the indicator system: Complete the core indicator system through indicator screening, indicator classification and indicator weight determination. The indicator system is the core method of the new evaluation, which determines the comprehensiveness and scientificity of the evaluation; S3. Feature extraction and selection: Through machine learning algorithms, features that have a significant impact on the evaluation results are extracted from the original data, and features with higher predictive power are selected for modeling; S4. Model construction and training: Build a theoretical model based on deep learning technology and train the model with a large amount of historical data to enable it to accurately predict project costs; S5. Evaluation and optimization: Apply the trained model to new engineering projects to conduct cost evaluation, and optimize the model based on the evaluation results to improve the accuracy of the prediction.

2. A method for early stage evaluation of water conservancy project cost according to claim 1, characterized in that: The data collection and processing include data collection and data processing; The data collection collects data related to the cost of water conservancy projects through various channels, including but not limited to: Cost data of historical engineering projects; engineering design drawings and related technical parameters; market price information, such as material prices and labor rates; policies, regulations and industry standards; Various sources of information on the geographical environment and socio-economic conditions of the project; The data processing performs preprocessing on the collected data, including data cleaning, deduplication, filling missing values, and standardization to ensure the quality and availability of the data.

3. The method for early stage evaluation of water conservancy project cost according to claim 1 is characterized in that: The model construction and training theory model is constructed through model selection, model structure design, model training, model verification and testing, and model deployment and application; The model selection is based on the project characteristics and evaluation requirements, and selects appropriate machine learning algorithms, such as decision trees, random forests, support vector machines, and neural networks; The model structure design designs the network structure of the model, including the input layer, hidden layer and output layer, as well as the number of neurons in each layer and the connection method; The model training uses historical data to train the model and adjusts the model parameters through an optimization algorithm so that the model can accurately predict the project cost; The model validation and testing is performed by cross-validation and test set validation to evaluate the performance and generalization ability of the model; The model deployment and application deploys the verified model to the actual project, conducts cost evaluation, and guides project decision-making based on the evaluation results.

4. The method for early stage evaluation of water conservancy project cost according to claim 1 is characterized in that: Indicator selection in the construction of the index system is based on the characteristics of the water conservancy project and the purpose of evaluation, and key indicators that can reflect the project cost are selected, such as project scale, project structure, regional economic level, and technical difficulty; The indicator classification divides the selected indicators into direct impact indicators and indirect impact indicators, as well as quantitative indicators and qualitative indicators, to construct a clearly hierarchical indicator system; The indicator weights are determined by expert consultation and analytic hierarchy process to reflect the impact of each indicator on the project cost.

5. The method for early stage evaluation of water conservancy project cost according to claim 1 is characterized by: Said evaluation and optimization is performed through data mining and analysis, application of artificial intelligence algorithms, and optimization of model parameters; The data mining and analysis are the basis of a new evaluation method, and the data mining and analysis include data mining, data analysis and data visualization.

6. A method for early stage evaluation of water conservancy project cost according to claim 5, characterized in that: The application of artificial intelligence algorithm is the core of the new evaluation method, which can handle complex data relationships and improve the intelligence level of evaluation. The application of artificial intelligence algorithm includes machine learning algorithm, deep learning algorithm and reinforcement learning algorithm.

7. A method for early stage evaluation of water conservancy project cost according to claim 5, characterized in that: The model parameter optimization is the key to improving the performance of the new evaluation method. It involves adjusting and optimizing the model parameters to achieve more accurate evaluation results. The model parameter optimization includes parameter optimization, model parameter adjustment, model integration and hyperparameter optimization.

8. A method for using the method for preliminary evaluation of water conservancy project cost according to any one of claims 1 to 7, characterized in that: The following steps are involved: Step 1: Preparation The preparation stage is the basis for the implementation of the new evaluation method. The purpose of this stage is to ensure the smooth progress of the evaluation process; clarify the evaluation objectives, scope and specific requirements, analyze the project characteristics, and determine the key points and difficulties of the evaluation; according to project requirements, form an evaluation team composed of data analysts, engineers, and economists; collect and organize the data required for the evaluation, including historical project data, market information, technical parameters, and pre-process the data; select appropriate data mining and artificial intelligence algorithms to build a preliminary evaluation model framework; Step 2: Implementation The implementation phase is the core of the new evaluation method. The tasks of this phase are to build and run the evaluation model; use data mining technology to conduct in-depth analysis of the collected data and extract useful information and features; build an evaluation model based on the results of data analysis and technology selection, and use historical data to train the model; verify the model through cross-validation and test sets, and evaluate the performance and generalization ability of the model; based on the model evaluation results, prepare a detailed evaluation report, including the evaluation process, result analysis, and suggestions; Step 3: Verification and Adjustment Phase The purpose of the verification and adjustment phase is to ensure the accuracy and reliability of the evaluation model and make necessary adjustments to the model; apply the model to new engineering projects, compare the actual results with the model prediction results, and verify the accuracy of the model; collect feedback from users and experts on the evaluation model and evaluation report; adjust and optimize the model parameters based on the feedback and actual operation conditions; deploy the verified and adjusted model to actual projects to guide investment decisions and cost control of engineering projects; continuously monitor the operation of the model, and regularly evaluate and update it to adapt to new changes and needs.