Lean management system and management method for operation and inspection cost type projects

Abstract

The invention discloses a lean management system and management method for operation and maintenance cost projects, and relates to the technical field of lean management, and the method comprises the following steps: collecting historical settlement data of operation and maintenance projects, and carrying out the preprocessing of the historical settlement data; dividing the preprocessed historical settlement data into a training set and a verification set, and performing variational mode decomposition on the training set to obtain a plurality of orders of settlement mode components; performing time serialization on the settlement mode components of the plurality of orders, and constructing an operation and maintenance project cost evaluation model based on a machine learning algorithm in combination with historical settlement data; performing structure optimization and parameter adjustment on the operation and maintenance project cost evaluation model to obtain an optimal operation and maintenance project cost evaluation model; and obtaining actual settlement data, obtaining an evaluation result through the optimal operation and maintenance project cost evaluation model, and judging the feasibility of the project according to a preset cost-benefit ratio. And project feasibility judgment is performed in combination with the preset cost-benefit ratio, so that the system can screen out potential projects, and the project management capability is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of project lean management, and more specifically, to a lean management system and management method for operation and inspection cost projects. Background Art

[0002] With the continuous reform of the power market in recent years, the expansion of the scope of influence, the acceleration of urbanization and industrialization, many old city renovation projects have also begun to be carefully deployed, smart meters have been promoted and used, power capacity has been expanded in industrial areas, and line inspections, power transmission and other live work operations have become increasingly frequent. This not only brings Development has brought new opportunities, but also new challenges. The power industry needs to have high performance and reliability in the construction of transmission and distribution grids and various facilities. Therefore, promoting technological innovation in the power industry and building stable, clean, safe, efficient and reliable 10KV and above transmission and distribution projects will have a strong driving effect on the development of the power grid itself and the sustainable development of national energy. At present, the requirements for power system safety and power supply reliability have been greatly improved. The operation and maintenance of power grid equipment, as an important part of ensuring the safe and stable operation of the power system, has also increased year by year. The operation and maintenance of the power system is usually arduous and complex. The operation and maintenance business carried out every year is collectively referred to as the operation and inspection cost project. Its main work content mainly includes patrol monitoring, operation and maintenance, hidden danger defect investigation and treatment, line "six preventions", detection services, sporadic maintenance and special treatment of the power transmission and transformation project itself. The number of operation and inspection cost projects carried out by power supply companies and their directly affiliated units in various cities and counties is large, the scope is wide, and the types are diverse, basically covering the three major business types of transmission projects, substation projects and distribution projects.

[0003] The above disclosed technical solutions have at least the following technical problems: in project management, data collection may not be comprehensive, resulting in an inability to accurately evaluate the cost-effectiveness of the project; some projects lack a sound cost supervision mechanism, resulting in the inability to effectively guarantee the effect of cost management. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a lean management system and management method for operation and maintenance cost projects, and solve the problem of inaccurate original power grid cost assessment by constructing a lean management model for operation and maintenance projects.

[0005] To achieve the above object, the present invention provides the following technical solutions: The lean management system and management method for operation and maintenance cost projects include the following steps: collecting historical settlement data of operation and maintenance projects and preprocessing them; dividing the preprocessed historical settlement data into a training set and a validation set, and decomposing the training set based on variational mode to obtain settlement modal components of several orders; time-seriesizing the settlement modal components of several orders, and building an operation and maintenance project cost evaluation model based on a machine learning algorithm in combination with the historical settlement data; optimizing the structure and adjusting the parameters of the operation and maintenance project cost evaluation model based on a loss function and a validation set to obtain an optimal operation and maintenance project cost evaluation model; obtaining actual settlement data, obtaining an evaluation result through the optimal operation and maintenance project cost evaluation model, and judging the feasibility of the project based on a preset cost-effectiveness ratio.

[0006] In a preferred embodiment, the historical settlement data includes fixed demand data and project characteristic data; the fixed demand data includes a fixed cost coefficient and an indirect cost coefficient, and the project characteristic data includes a project scale coefficient; the specific method for obtaining the fixed cost coefficient is: by reviewing the financial statements of the power grid, the basic data of the fixed cost coefficient is obtained, and the average value of historical input costs and employee wages is calculated based on the basic data, and this is used as the parameter of the exponential function to operate the exponential function; the product of the historical input cost and employee wages is logarithmically operated, and multiplied by the result of the exponential function to obtain the fixed cost coefficient.

[0007] In a preferred embodiment, the specific method for obtaining the indirect cost coefficient is: collect specific material cost data, downtime data and risk data, record detailed information of each maintenance according to the maintenance plan and market price, including maintenance time, maintenance content, materials and parts used, and personnel costs, calculate the cost of materials and parts required in the maintenance process, and use logarithmic functions and square root functions for processing to obtain the indirect cost coefficient.

[0008] In a preferred embodiment, the specific method for obtaining the project scale coefficient is as follows: according to the length, area, and volume parameters provided in the bill of quantities, multiply by the construction efficiency to obtain the actual project quantity of each work; according to the historical project data, the average project quantity of each work is calculated by the median statistical method; according to the project's bill of quantities, the actual project quantity of each line is calculated in combination with a preset line length scale calculation formula; the actual project quantity of the support structure is calculated in combination with a preset support structure calculation formula; the actual project quantity of the equipment installation is calculated in combination with a preset installation equipment quantity calculation formula; the actual project quantity is compared with the average project quantity and added to obtain the project scale coefficient; In a preferred embodiment, the training set is decomposed based on variational mode to obtain settlement modal components of several orders, specifically: the engineering settlement data of the training set is represented as a one-dimensional time series , where n is the length of the time series; set the number of modal components according to the periodicity and frequency of the engineering cost forecasting requirements, and map the regularization parameters; analyze the periodicity and frequency characteristics of the engineering cost data. According to the forecasting requirements and data characteristics, obtain the main frequency resolution; initialize the frequency core function and center frequency of each modal component; perform alternating optimization steps, iteratively optimize the frequency core function and center frequency until convergence or reach the preset number of iterations, and obtain the final modal component.

[0009] In a preferred embodiment, the cost evaluation model of the cost operation and inspection project is structurally optimized and parameter adjusted based on the loss function, specifically: defining a loss function, the loss function includes a mean square error index and a mean absolute error index; dividing the historical settlement data into a training set and a test set; extracting features from the training set based on a convolutional neural network to obtain several feature maps, and establishing an initial training model; extracting features from the training set based on a convolutional neural network to obtain multiple feature maps, and establishing an initial training model; establishing a linear weighting function based on several feature maps, taking the actual cost data as the output of the weighting function, setting a preset first loss function as a constraint, and setting initial weights, and obtaining an initial operation and inspection project cost evaluation model in combination with the initial training model; inputting the training set into the initial operation and inspection project cost evaluation model to determine whether the first loss function converges, and if not, performing gradient descent iterative optimization on the weights of the linear weighting function; until the first loss function converges, stopping the iterative optimization of the weights to obtain the optimal operation and inspection project cost evaluation model.

[0010] In a preferred embodiment, it includes a data collection module, a data processing module, a model building module, a model optimization module, and an evaluation module: Data collection module: used to collect and preprocess the historical settlement data of the operation and inspection project; Data processing module: used to divide the preprocessed historical settlement data into a training set and a validation set, and decompose the training set based on variational mode to obtain settlement modal components of several orders; Model building module: used to time-series the settlement modal components of several orders, and build an operation and inspection project cost evaluation model based on a machine learning algorithm in combination with the historical settlement data; Model optimization module: used to perform structural optimization and parameter adjustment on the operation and inspection project cost evaluation model based on the loss function and the validation set to obtain the optimal operation and inspection project cost evaluation model; Evaluation module: used to obtain actual settlement data, obtain evaluation results through the optimal operation and inspection project cost evaluation model, and judge the feasibility of the project based on a preset cost-effectiveness ratio. The technical effects and advantages of the lean management system and management method of the operation and inspection cost projects of the present invention are as follows: 1. The present invention obtains the optimal cost evaluation model for operation and inspection projects by analyzing historical data. Through the optimal cost evaluation model for operation and inspection projects, the evaluation results are obtained. Through accurate cost evaluation, the system can help enterprises optimize resource allocation, avoid unnecessary waste, and improve resource utilization efficiency. Combined with the preset cost-benefit ratio to judge the feasibility of the project, the system can screen out potential projects and improve the management ability of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a structural schematic diagram of the lean management method for operation and inspection cost items of the present invention.

[0012] Figure 2 It is a schematic diagram of the structure of the lean management system for operation and inspection cost items of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] Embodiment 1, Figure 1 The invention provides a lean management method for operation and inspection cost items, comprising the following steps: S1, collect historical settlement data of operation and inspection projects and perform preprocessing; The historical settlement data includes fixed demand data and project characteristic data; The fixed demand data include fixed cost coefficients and indirect cost coefficients, and the project characteristic data include project scale coefficients; The preprocessing includes data cleaning, classification and statistics; Both automatic and manual collection methods are used to ensure the comprehensiveness and accuracy of the data.

[0015] Clean the collected historical settlement data, specifically: Import the collected historical settlement data into the data processing tool; Identify and process missing values ​​in the O&M cost data using filling, deletion, and interpolation methods; Check and delete duplicate records to ensure the uniqueness and accuracy of operation and inspection cost data; Identify and correct errors and inconsistencies in O&I cost data; Unify the data format, such as date format, currency unit, etc., to ensure data consistency; Identify and handle outliers and correct these data points; Convert data into a format suitable for analysis, such as converting categorical data into numerical data.

[0016] The fixed cost coefficient of the operation and inspection project mainly includes equipment maintenance and management costs, energy loss, staff wages, equipment replacement costs, and system monitoring and management expenses. Costs may also increase due to factors such as network expansion, technology upgrades, and responding to emergencies. In order to reduce the fixed cost coefficient, the operation and inspection project should strengthen cost control, do a good job of cost budgeting and forecasting, track cost changes in a timely manner, and take effective measures to adjust and optimize.

[0017] The specific method for obtaining the fixed cost coefficient is: By reviewing the financial statements of the power grid, we obtain the basic data of the fixed cost coefficient. Based on the basic data, we calculate the average value of historical input costs and employee wages, and use this as the parameter of the exponential function to operate the exponential function. We perform logarithmic operations on the product of historical input costs and employee wages, and multiply it with the result of the exponential function to obtain the fixed cost coefficient.

[0018]

[0019] Where GD is the fixed cost coefficient, is the historical input cost, It's employee wages.

[0020] Fixed cost coefficients have the following benefits for lean management of operation and inspection cost items: Budgeting: By calculating the fixed cost coefficient, companies can more accurately formulate annual or quarterly budgets to ensure that all expenditures are within controllable range.

[0021] Pricing strategy: By understanding the production cost of a product, companies can formulate more competitive pricing strategies while ensuring profit margins.

[0022] Investment decision: When deciding whether to introduce new equipment, launch new projects or expand the market, fixed cost coefficient data provides important economic basis for decision-making.

[0023] Resource allocation: Based on the fixed cost coefficient data, enterprises can optimize resource allocation and ensure the rational allocation of resources among departments and projects.

[0024] Efficiency improvement: By identifying inefficiencies in operations, companies can take steps to improve them, thereby increasing overall operational efficiency.

[0025] Cost fluctuation warning: Regular calculation of fixed cost coefficients can help companies detect cost fluctuations in a timely manner and take corresponding countermeasures to reduce potential risks.

[0026] Adjustment of cost control strategies: In the face of changes in the market environment or internal operational adjustments, companies can adjust their cost control strategies based on fixed cost coefficient data to maintain operational stability.

[0027] Improved competitiveness: Cost leadership strategy: By continuously reducing the fixed cost coefficient, companies can implement a cost leadership strategy in the market and attract customers with low prices.

[0028] Product and service innovation: On the basis of effective control of fixed cost coefficients, enterprises have more resources to invest in product and service innovation to meet the ever-changing needs of customers.

[0029] In summary, calculating the fixed cost coefficient is crucial for enterprises and organizations. It not only helps to achieve accurate financial control and effective decision support, but also promotes resource optimization, efficiency improvement, risk management and competitiveness improvement. Therefore, enterprises should attach great importance to the calculation and analysis of fixed cost coefficients.

[0030] Indirect cost factors include: Material cost: When equipment parts are damaged or reach the end of their service life, they need to be replaced with new parts. The purchase cost of these parts is an important part of the indirect cost coefficient. During the maintenance process, some consumables such as lubricants and detergents are also consumed, which also need to be included in the indirect cost coefficient.

[0031] Downtime cost: The cost of production loss caused by downtime of operation and inspection projects due to maintenance. Although this cost is not directly reflected in financial expenditure, it has an important impact on the operation of the enterprise, so it should also be considered in the scope of indirect cost coefficient.

[0032] Risk cost: Risk cost caused by equipment failure and accidents due to untimely or incomplete maintenance. This cost is often difficult to quantify accurately, but it poses a huge potential threat to enterprises.

[0033] The specific method for obtaining the indirect cost coefficient is: Collect specific material cost data, downtime data and risk data. According to the maintenance plan and market price, record the detailed information of each maintenance, including maintenance time, maintenance content, materials and parts used, and personnel costs. Calculate the cost of materials and parts required in the maintenance process, and use logarithmic functions and square root functions for processing to obtain the indirect cost coefficient.

[0034] The specific calculation formula of the indirect cost coefficient is:

[0035] Where WX is the indirect cost coefficient, is the material cost data, It is the downtime data. is the risk data, n is the number of downtime days, is the preset scaling factor.

[0036] The indirect cost coefficient has the following benefits for the lean management of operation and inspection cost items: Precise budget control: By accurately calculating the indirect cost coefficient, companies can more accurately formulate and monitor maintenance budgets. This helps avoid budget overruns, ensure the effective use of funds, and make timely adjustments when necessary.

[0037] Improve decision-making quality: Detailed data on indirect cost factors provides management with valuable decision-making information. Whether selecting suppliers, purchasing spare parts or evaluating equipment upgrade options, this cost information can help decision makers make more informed choices.

[0038] Enhanced cost control capabilities: By continuously tracking and analyzing indirect cost coefficients, companies can identify trends and causes of cost increases and take appropriate measures to control them. This helps reduce overall fixed cost coefficients and improve profitability.

[0039] Improve operational efficiency: Effective indirect cost factor management can promote the improvement of operational efficiency. For example, by properly arranging maintenance plans and optimizing maintenance processes, equipment downtime can be reduced, and production efficiency and capacity utilization can be improved.

[0040] Risk management: Understanding the indirect cost coefficient helps companies better assess potential risks. For example, predicting and analyzing the indirect cost coefficient of key equipment can help companies prepare countermeasures in advance to avoid production interruptions and losses caused by equipment failure.

[0041] Support continuous improvement: Indirect cost coefficient data is an important input for continuous improvement. By regularly reviewing and analyzing indirect cost coefficients, companies can identify opportunities for improvement, such as improving maintenance technology, optimizing spare parts management, or improving employee skills, thereby continuously improving maintenance efficiency and quality.

[0042] Improve customer satisfaction: For companies that provide maintenance services, accurate calculation of indirect cost factors helps to set reasonable service prices while ensuring service quality. This helps to improve customer satisfaction and loyalty, thereby promoting the long-term development of the company.

[0043] The project scale factor is a quantitative indicator used to measure the size of a project. In project management, the scale factor is usually used to determine key factors such as resource allocation, budget preparation, and construction schedule. By calculating the project scale factor, the complexity and risk of the project can be more accurately assessed, thereby formulating a more scientific and reasonable project management strategy.

[0044] The specific method for obtaining the project scale coefficient is as follows: According to the length, area and volume parameters provided in the bill of quantities, multiply them by the construction efficiency to get the actual quantity of each work; Based on historical project data, the average engineering quantity of each work is calculated by the median statistical method; According to the project's bill of quantities, the actual engineering quantity of each line is calculated in combination with the preset line length scale calculation formula; the actual engineering quantity of the support structure is calculated in combination with the preset support structure calculation formula; the actual engineering quantity of the equipment installation is calculated in combination with the preset installation equipment quantity calculation formula; Compare and add the actual project quantity with the average project quantity to obtain the project scale coefficient; The specific calculation formula for the actual engineering quantity of each line is as follows:

[0045] The specific calculation formula for the actual engineering quantity of the support structure is as follows:

[0046] The specific calculation formula for the actual engineering quantity of the equipment installation is as follows:

[0047] The specific calculation formula of the project scale coefficient is as follows:

[0048] In the formula, is the actual engineering quantity of each line, u is the number of sections of the line, is the laying length of the line, is the line angle, is the number of cable axes between lines, is the actual engineering quantity of the supporting structure, M is the total number of transmission towers, is the width of the tower foundation pit, is the height of the tower foundation pit, is the volume of the tower foundation pit, T is the number of tower floors, is the actual engineering quantity of equipment installation, N is the total number of equipment, is the number of equipment types, is the required material quantity for the equipment, is the number of installations, GM is the project scale factor, is the average engineering quantity of each line, is the average engineering quantity of the supporting structure, The average engineering workload for equipment installation.

[0049] The project scale factor has the following benefits for the lean management of operation and inspection cost projects: Reflecting scale effects: The scale factor can reflect the differences in costs between projects of different sizes, thereby more accurately estimating the costs of power grid projects. For large power grid projects, the scale factor can help identify and consider scale-related cost factors, such as equipment procurement, construction difficulty, and operation and maintenance costs.

[0050] Optimizing resource allocation: By taking scale factors into account, grid companies can allocate resources more efficiently, ensuring that large projects receive adequate funding while preventing small projects from being hobbled by insufficient funding.

[0051] Cost forecasting: During the planning stage of a power grid project, the scale factor can be used to predict the cost of the project at different scales, providing a scientific basis for cost control.

[0052] Cost comparison: By comparing the costs of projects of different sizes, power grid companies can identify weak links in cost control and take targeted measures to improve them.

[0053] Assisted decision-making: The scale factor can serve as an important reference for power grid project decision-making, helping decision makers to more comprehensively evaluate the cost-effectiveness of the project and make more scientific and reasonable decisions.

[0054] Optimizing investment portfolio: Across multiple grid projects, the scale factor can be used to evaluate the cost-effectiveness of each project and optimize the investment portfolio to ensure that funds are invested in the projects with the greatest potential.

[0055] Planning optimization: Incorporating the scale factor into grid planning can optimize the grid structure and improve the reliability and economy of the grid.

[0056] Sustainable development: By properly controlling the cost of power grid projects, power grid companies can ensure the sustainability of projects and lay a solid foundation for future power grid development.

[0057] S2, the preprocessed historical settlement data is divided into a training set and a validation set, and the training set is decomposed based on variational mode to obtain settlement modal components of several orders; The training set is decomposed based on variational mode to obtain several orders of settlement modal components, specifically: Represent the engineering settlement data of the training set as a one-dimensional time series , where n is the length of the time series; The number of modal components is set according to the period and frequency required for project cost forecasting, and is mapped into the regularization parameter; Analyze the periodicity and frequency characteristics of construction cost data.

[0058] According to the prediction requirements and data characteristics, the main frequency resolution is obtained; Initialize the frequency kernel function and center frequency of each modal component; Perform alternating optimization steps to iteratively optimize the frequency core function and center frequency until convergence or the preset number of iterations is reached to obtain the final modal components.

[0059] S3, time series the settlement modal components of several orders, and build a cost assessment model for the operation and inspection project based on the machine learning algorithm in combination with the historical settlement data; Time series processing: The modal component data within each time window are arranged in chronological order to form time series data.

[0060] Time series processing can simply arrange the data in each time window into a vector or matrix in chronological order to facilitate subsequent time series modeling or deep learning model input.

[0061] We collect and integrate historical settlement data to build the basic data set for the cost assessment model of the operation and maintenance project. Then, based on these data sets, we use technical means such as statistical analysis methods and machine learning algorithms to establish the mathematical model of the cost assessment model of the operation and maintenance project.

[0062] The cost assessment model of the operation and inspection project is specifically as follows:

[0063] In the formula, is the preset weight factor of the fixed cost coefficient, is the preset weight factor of the fixed cost coefficient, is the preset weight factor of the project scale coefficient, GD is the fixed cost coefficient, JJ is the fixed cost coefficient, and GM is the project scale coefficient.

[0064] S4, based on the loss function, the cost evaluation model of the cost operation and inspection project is optimized and the parameters are adjusted to obtain the optimal cost evaluation model of the operation and inspection project; The structure optimization and parameter adjustment of the cost evaluation model of the cost operation and inspection project based on the loss function are specifically as follows: Define the loss function, which includes the mean square error index and the mean absolute error index; Divide historical settlement data into training set and test set; Extract features from the training set based on a convolutional neural network, obtain several feature maps, and establish an initial training model; Extract features from the training set based on a convolutional neural network to obtain multi-feature mapping and establish an initial training model; A linear weighting function is established based on several feature maps, the actual cost data is used as the output of the weighting function, a preset first loss function is set as a constraint condition, and an initial weight is set, and an initial operation and inspection project cost evaluation model is obtained in combination with the initial training model; The training set is input into the initial operation and inspection project cost assessment model to determine whether the first loss function converges. If not, the weights of the linear weighting function are iteratively optimized by gradient descent. Until the first loss function converges, stop iterative optimization of weights and obtain the optimal cost evaluation model for operation and inspection projects; The specific calculation formula of the mean square error index is:

[0065] The specific calculation formula of the mean absolute error index is:

[0066] in, is the actual value, is the predicted value, n is the number of samples; S5, obtain the actual settlement data, obtain the evaluation results through the optimal operation and inspection project cost evaluation model, and judge the feasibility of the project based on the preset cost-benefit ratio; Collect and organize the latest settlement data of the operation and maintenance projects, and input these actual data into the optimal operation and maintenance project cost evaluation model for evaluation. The model will evaluate the cost of the operation and maintenance project based on the input data and optimized parameters, using the set algorithm and model structure.

[0067] If the evaluation result is greater than the preset cost-benefit ratio, the project is not feasible; conversely, if the evaluation result is less than the preset cost-benefit ratio, the project is feasible; Embodiment 2, Figure 2 The invention provides a lean management system for operation and inspection cost projects, including a data collection module, a data processing module, a model building module, a model optimization module, and an evaluation module: Data collection module: used to collect historical settlement data of operation and inspection projects and perform preprocessing; Data processing module: used to divide the pre-processed historical settlement data into a training set and a validation set, and decompose the training set based on variational mode to obtain settlement modal components of several orders; Model building module: used to time-series the settlement modal components of several orders, and build a cost assessment model for the operation and inspection project based on the machine learning algorithm in combination with historical settlement data; Model optimization module: used to optimize the structure and adjust the parameters of the operation and inspection project cost evaluation model based on the loss function and the validation set to obtain the optimal operation and inspection project cost evaluation model; Evaluation module: used to obtain actual settlement data, obtain evaluation results through the optimal operation and maintenance project cost evaluation model, and judge the feasibility of the project based on the preset cost-benefit ratio.

[0068] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0069] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0070] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0071] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0072] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0073] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Lean management method for operation and inspection cost projects, characterized by: The steps include: Collect historical settlement data of operation and inspection projects and perform pre-processing; The preprocessed historical settlement data is divided into a training set and a validation set, and the training set is decomposed based on variational mode to obtain settlement modal components of several orders; The settlement modal components of several orders are time-seriesized, and the cost assessment model of the operation and inspection project is constructed based on the machine learning algorithm in combination with the historical settlement data; Based on the loss function and validation set, the structure of the operation and inspection project cost evaluation model is optimized and the parameters are adjusted to obtain the optimal operation and inspection project cost evaluation model; Obtain actual settlement data, obtain the evaluation results through the optimal operation and maintenance project cost evaluation model, and judge the feasibility of the project based on the preset cost-benefit ratio.

2. The lean management method for operation and inspection cost items according to claim 1 is characterized in that: The historical settlement data includes fixed demand data and project characteristic data; The fixed demand data include fixed cost coefficients and indirect cost coefficients, and the project characteristic data include project scale coefficients; The specific method for obtaining the fixed cost coefficient is: By reviewing the financial statements of the power grid, we obtain the basic data of the fixed cost coefficient. Based on the basic data, we calculate the average value of historical input costs and employee wages, and use this as the parameter of the exponential function to operate the exponential function. We perform logarithmic operations on the product of historical input costs and employee wages, and multiply it with the result of the exponential function to obtain the fixed cost coefficient.

3. The lean management method for operation and inspection cost items according to claim 2 is characterized in that: The specific method for obtaining the indirect cost coefficient is: Collect specific material cost data, downtime data and risk data. According to the maintenance plan and market price, record the detailed information of each maintenance, including maintenance time, maintenance content, materials and parts used, and personnel costs. Calculate the cost of materials and parts required in the maintenance process, and use logarithmic functions and square root functions for processing to obtain the indirect cost coefficient.

4. The lean management method for operation and inspection cost items according to claim 3 is characterized in that: The specific method for obtaining the project scale coefficient is as follows: According to the length, area and volume parameters provided in the bill of quantities, multiply them by the construction efficiency to get the actual quantity of each work; Based on historical project data, the average engineering quantity of each work is calculated by the median statistical method; According to the project's bill of quantities, the actual engineering quantity of each line is calculated in combination with the preset line length scale calculation formula; Calculate the actual engineering quantity of the supporting structure in combination with the preset supporting structure calculation formula; Calculate the actual engineering quantity of equipment installation by combining the preset calculation formula for the number of installed equipment; Compare the actual project quantity with the average project quantity and add them together to obtain the project scale coefficient.

5. The lean management method for operation and inspection cost items according to claim 4 is characterized in that: The training set is decomposed based on variational mode to obtain several orders of settlement modal components, specifically: Represent the engineering settlement data of the training set as a one-dimensional time series , where n is the length of the time series; The number of modal components is set according to the period and frequency required for project cost forecasting, and is mapped into the regularization parameter; Analyze the periodicity and frequency characteristics of engineering cost data; According to the prediction requirements and data characteristics, the main frequency resolution is obtained; Initialize the frequency kernel function and center frequency of each modal component; Perform alternating optimization steps to iteratively optimize the frequency core function and center frequency until convergence or the preset number of iterations is reached to obtain the final modal components.

6. The lean management method for operation and inspection cost items according to claim 5 is characterized in that: The structure optimization and parameter adjustment of the cost evaluation model of the cost operation and inspection project based on the loss function are specifically as follows: Define the loss function, which includes the mean square error index and the mean absolute error index; Divide historical settlement data into training set and test set; Extract features from the training set based on a convolutional neural network, obtain several feature maps, and establish an initial training model; Extract features from the training set based on a convolutional neural network to obtain multi-feature mapping and establish an initial training model; A linear weighting function is established based on several feature maps, the actual cost data is used as the output of the weighting function, a preset first loss function is set as a constraint condition, and an initial weight is set, and an initial operation and inspection project cost evaluation model is obtained in combination with the initial training model; The training set is input into the initial operation and inspection project cost assessment model to determine whether the first loss function converges. If not, the weights of the linear weighting function are iteratively optimized by gradient descent. Until the first loss function converges, stop the iterative optimization of the weights and obtain the optimal operation and inspection project cost evaluation model.

7. The lean management method for operation and inspection cost items according to claim 6 is characterized in that: The cost assessment model of the operation and inspection project is specifically as follows: In the formula, is the preset weight factor of the fixed cost coefficient, is the preset weight factor of the indirect cost coefficient, is the preset weight factor of the project scale coefficient, GD is the fixed cost coefficient, JJ is the indirect cost coefficient, and GM is the project scale coefficient.

8. The lean management method for operation and inspection cost items according to claim 7 is characterized in that: The specific calculation formula of the project scale coefficient is as follows: In the formula, is the actual engineering quantity of each line, is the actual engineering quantity of the supporting structure, is the actual engineering quantity of equipment installation, GM is the engineering scale factor, is the average engineering quantity of each line, is the average engineering quantity of the supporting structure, The average engineering workload for equipment installation.

9. The lean management method for operation and inspection cost items according to claim 8 is characterized in that: The specific calculation formula of the indirect cost coefficient is: In the formula, is the initial cost, is an additional cost, is the surplus value, is the preset residual value coefficient, N is the installed life, and M is the preset estimated service life.

10. A lean management system for operation and inspection cost items using any one of claims 1 to 9, characterized in that: Including data collection module, data processing module, model building module, model optimization module, and evaluation module: Data collection module: used to collect historical settlement data of operation and inspection projects and perform preprocessing; Data processing module: used to divide the pre-processed historical settlement data into a training set and a validation set, and decompose the training set based on variational mode to obtain settlement modal components of several orders; Model building module: used to time-series the settlement modal components of several orders, and build a cost assessment model for the operation and inspection project based on the machine learning algorithm in combination with historical settlement data; Model optimization module: used to optimize the structure and adjust the parameters of the operation and inspection project cost evaluation model based on the loss function and the validation set to obtain the optimal operation and inspection project cost evaluation model; Evaluation module: used to obtain actual settlement data, obtain evaluation results through the optimal operation and maintenance project cost evaluation model, and judge the feasibility of the project based on the preset cost-benefit ratio.

Images