Cost information base upgrading method based on technical maintenance
By constructing a cost -cost predictive model based on linear regression and convolutional neural networks, the problem of actual cost and predictive costs caused by market changes is solved, and the accuracy of cost prediction and the transparency of management are improved.
Patent Information
- Application Number
- CN202411461962.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-18
- Publication Date
- 2025-01-28
AI Technical Summary
The existing technology is difficult to effectively deal with the problem of excessive gap between actual cost and predicted cost caused by market changes, resulting in inaccurate cost management and forecasting.
By obtaining the cost cost data of the substation during the preset period cycle, combined with market factors characteristic data, the linear regression model and convolutional neural network are used to build cost cost prediction models, and the optimal cost cost prediction model is obtained through parameter optimization.
It has improved the accuracy of future maintenance costs, optimized budget and expense management, provided more accurate cost information and maintenance records, and enhanced the transparency of cost management and the scientificity of decision -making.
Smart Images
Figure CN119379331A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of cost management, and more specifically, to a cost information base upgrading method based on technical maintenance. Background Art
[0002] The cost information database based on technical maintenance is a database specially used to record, manage and analyze various cost data related to technical maintenance. This information database usually covers all costs of equipment maintenance, repair and upgrading, including but not limited to labor costs, material costs, equipment depreciation, maintenance time costs, and outsourcing service costs; this type of information database has been widely used in many industries to effectively manage and control the total cost of technical maintenance and repair; the cost information database based on technical maintenance plays a vital role in modern enterprises. By providing comprehensive, accurate and real-time cost data, it supports enterprises in scientific cost management and decision-making. This information database not only improves the accuracy of cost control, but also promotes the optimal allocation of resources and the improvement of business processes.
[0003] Deficiencies of existing technologies: Market conditions and cost structures are changing dynamically. Raw material prices, labor costs, equipment costs, etc. will fluctuate with changes in market supply and demand. Upgrading the cost information database can ensure that the data used is the latest, thereby helping companies to formulate more accurate budgets and cost control strategies. Therefore, the present invention provides a cost information database upgrade method based on technical maintenance, which takes into account the impact of market fluctuations on predicted costs, solves the problem of actual costs being changed due to market changes, resulting in a large difference between actual costs and predicted costs, and makes the prediction of cost information more accurate.
[0004] In view of the above problems, the present invention proposes a solution. Summary of the invention
[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for upgrading a cost information database based on technical maintenance, which solves the problem of a large difference between the actual cost and the predicted cost due to market changes raised in the above-mentioned background technology by analyzing the cost information.
[0006] To achieve the above object, the present invention provides the following technical solutions: A method for upgrading a cost information database based on technical maintenance includes the following steps: obtaining cost data of a substation within a preset time period, wherein the cost data includes historical cost data and actual cost data; obtaining market factors, performing feature analysis on the market factors, extracting factor feature data that can be quantitatively analyzed, and obtaining market factor feature data based on a linear regression model according to the factor feature data; constructing a cost prediction model based on a convolutional neural network according to the market factor feature data and the historical cost data to obtain predicted cost data; and optimizing parameters of the cost prediction model according to the actual cost data and the predicted cost data to obtain an optimal cost prediction model.
[0007] In a preferred embodiment, the historical cost data includes initial construction cost, operation cost and maintenance cost; the actual cost data includes various actual costs in the construction and operation of the substation.
[0008] In a preferred embodiment, the specific method for acquiring the market factor characteristic data is as follows: market factors include raw material prices, labor costs, inflation rates, and market demand; market factor data is cleaned, outliers are removed, and characteristic analysis is performed to extract factor characteristic data that can be quantitatively analyzed; a linear regression model is created, and the factor characteristic data that can be quantitatively analyzed extracted from the market factors is substituted into the linear regression model to obtain the market factor characteristic data.
[0009] In a preferred embodiment, the specific method for obtaining the cost prediction model is as follows: The construction cost of the substation is used as the fixed step size of the convolution kernel, and the input substation feature information is convolved to generate the corresponding weight coefficient of the feature information; The sensitivity error coefficient and entropy loss function are introduced to calculate the historical cost correction coefficient; Based on the weight coefficient of historical cost characteristic information, combined with the historical cost correction coefficient and market factor characteristic data, a cost prediction model is established.
[0010] In a preferred embodiment, the specific method for obtaining the optimal cost prediction model is as follows: compare and analyze the actual cost data in the cost data with the predicted cost data, and calculate the absolute error between the actual cost data and the predicted cost data; when the absolute error is greater than a preset error threshold, adjust the cost prediction model parameters and feature information, and test and optimize the cost prediction model again.
[0011] The technical effects and advantages of the cost information database upgrade method based on technical maintenance of the present invention are as follows: 1. The present invention ensures that the data in the cost information database accurately reflects the current maintenance cost through real-time collection and analysis of technical maintenance data, improves the cost prediction model, and improves the prediction accuracy of future maintenance costs; dynamically adjusts cost information according to actual maintenance data, optimizes budget and expense management, provides detailed cost information and maintenance records, and enhances the transparency of cost management; uses the updated information database to provide more accurate data support for decision-making, improves the scientific nature of decision-making, and the automated update process reduces the workload of manually maintaining the information database, improves work efficiency, and can quickly respond to cost changes and adapt to dynamic changes in the market and technology.
[0012] 2. The present invention reduces unnecessary expenses and waste through accurate cost information and optimized budget management, can adapt to the dynamic changes in the market and technology, and improves the economic return on investment through accurate cost control and forecasting; provides more accurate cost data to support decision-making, helps to formulate more effective maintenance and procurement strategies, better identify and manage financial risks related to maintenance, and reduce decision-making risks; improves cost control and maintenance management processes and overall management efficiency through the upgrade of the information database, and the continuously updated information database supports the continuous improvement of cost management and maintenance strategies. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 The present invention is a schematic structural diagram of a method for upgrading a cost information database based on technical maintenance. DETAILED DESCRIPTION
[0014] 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.
[0015] Embodiment 1, Figure 1 The invention provides a cost information base upgrading method based on technical maintenance.
[0016] S10, obtaining cost data of the substation within a preset time period, wherein the cost data includes historical cost data and actual cost data; The cost data set within the preset period refers to the cost data of the substation every year within N years, the historical cost data refers to the cost data of each year within the previous N-1 years, and the actual cost data refers to the cost data remaining in the cost data set except the historical cost data; The specific method for obtaining the historical cost data is as follows: Contact the company that builds and operates the substation, check the company's project files and annual financial reports, obtain the initial construction cost, operating cost and maintenance cost data of the substation, extract the construction area, construction efficiency and equipment status of the substation as feature information; perform data cleaning on the historical cost data based on the feature information of the substation, process the missing values, outliers and duplicate values that may exist in the data, and obtain a set of historical cost data; The specific method for obtaining the actual cost data is as follows: Obtain annual reports, quarterly reports and financial details from the company's financial department, obtain various expenses in the substation construction and operation project, and integrate various records to obtain actual expenditures, that is, actual cost data; Obtaining the cost data of the substation within a preset time period and establishing it as a historical cost data set is an important basis for long-term cost management and budget forecasting. This process plays a significant role in the upgrade and maintenance of the cost information database, which is specifically reflected in the following aspects: Improve data accuracy and reliability: The upgraded information base can provide more accurate historical data to help analyze cost trends and changes. Based on accurate historical data, better future cost forecasting and budgeting can be carried out; Support decision-making: By analyzing historical cost data, more reasonable budgets and investment plans can be formulated, and the causes of cost growth can be identified and controlled, thereby achieving cost savings and resource optimization; Enhance data analysis capabilities: Identify cost trends and patterns to help develop long-term strategies, compare historical data with current costs, and evaluate the cost-effectiveness of projects; Promote standardization and regularization: Establish unified data collection and reporting standards, improve data consistency, ensure compliance with industry standards, and improve the standardization of project management; Improve information sharing and communication: All departments can communicate and collaborate effectively based on a unified data set, providing transparent data support, which helps improve the transparency and accountability of project management; Support technological innovation and improvement: By analyzing historical cost data, evaluate the impact of technological improvements on costs, and apply new technologies or methods to reduce costs and improve efficiency; S20, obtaining market factors, performing feature analysis on the market factors, extracting factor feature data that can be quantitatively analyzed, and obtaining market factor feature data based on the factor feature data based on a linear regression model; The market factors include raw material prices, labor costs, inflation rates, and market demand. After data cleaning of the market factors and removal of outliers, feature analysis is performed to extract factor feature data that can be quantitatively analyzed. Create a linear regression model, substitute the quantifiable analysis factor characteristic data extracted from the market factors into the linear regression model, and obtain the market factor characteristic data; The linear regression model can obtain the total market factor characteristic data by performing linear calculation on the quantitatively analyzable factor characteristic data and the preset linear regression coefficient; The specific calculation formula of the market factor characteristic data is as follows:
[0017] In the formula, is the market factor characteristic data, are the preset linear regression coefficients, It is the characteristic data of factors that can be quantitatively analyzed. is the number of characteristic data of factors that can be quantitatively analyzed; S30, obtaining market factors, performing feature analysis on the market factors, extracting factor feature data that can be quantitatively analyzed, and obtaining market factor feature data based on the factor feature data based on a linear regression model; The specific method for obtaining the cost prediction model is as follows: The construction cost of the substation is used as the fixed step size of the convolution kernel, and the input substation feature information is convolved to generate the corresponding feature information weight coefficient. The specific calculation formula of the weight coefficient of the substation characteristic information is as follows:
[0018] In the formula, It is Year The weight coefficient of the eigenvalue information is is a nonlinear activation function; For the The feature subset corresponding to the feature value information is is the substation construction cost, For the Year Specific data of characteristic values, After convolution Year eigenvalue information bias term, is a positive integer; When calculating the predicted cost data, numerical calculation errors may occur and the model response speed is not sensitive. Therefore, the sensitivity error coefficient is introduced to reduce the error problem caused by insufficient model sensitivity. The specific calculation formula of the sensitivity error coefficient is as follows:
[0019] In the formula, is the sensitivity error coefficient; When performing neural convolution, convolution entropy loss may occur, which will affect the output data results. Therefore, the entropy loss function is introduced. The specific calculation formula of the entropy loss function is as follows:
[0020] In the formula, is the entropy loss function, For the Year Specific data of characteristic values, It is a positive integer, indicating the number of historical cost data obtained; Combining the entropy loss function and the sensitivity error coefficient, the historical cost correction coefficient is calculated. The specific calculation formula of the correction coefficient is:
[0021] In the formula, is the historical cost correction factor; Based on the weight coefficient of the historical cost characteristic information, combined with the historical cost correction coefficient and the market factor characteristic data, a cost prediction model is established. The specific calculation formula of the cost prediction model is as follows:
[0022] In the formula, It is the forecast cost data. It is Year The weight coefficient of the eigenvalue information is is the market factor characteristic data, is the historical cost correction factor; Combining market factors with cost data and using convolutional neural networks to build a cost forecasting model can significantly improve the accuracy and practicality of forecasts. The upgrade of the cost information database has the following effects in this process: Provide comprehensive market factor data: The upgraded information database can contain the latest market factor characteristic data, such as economic indicators, industry trends, supply chain changes, etc. This information is crucial for the input of the model and helps to capture subtle changes in the market, thereby improving the accuracy of the model's cost forecast; Improved data quality and diversity: The updated information base contains more diverse and high-quality data, which helps to extract valuable features and patterns during the training process. In this way, the model can better learn the relationship between complex cost and market factors and improve forecasting accuracy; Adapt to market changes and trends: The market environment and cost structure change over time. The upgrade of the information database can incorporate the latest market dynamics and trends into the model. In this way, the convolutional neural network model can better adapt to these changes and provide more accurate cost forecasts. In short, the upgrade of the information database ensures that the model can utilize the latest market factors and high-quality data, improve the accuracy and reliability of the forecast results, and enable the model to maintain efficient performance in an ever-changing market environment.
[0023] S40, optimizing the parameters of the cost prediction model according to the actual cost data and the predicted cost data to obtain the optimal cost prediction model; When analyzing cost data, first compare the actual cost data with the predicted cost data; Calculate the absolute error between each pair of data points, that is, the absolute value of the difference between the actual cost and the predicted cost; The prediction accuracy of the model is evaluated through these absolute errors. If some error values are found to exceed the preset threshold, it means that the prediction accuracy does not meet the standard, and the cost prediction model needs to be adjusted. Adjusting the cost prediction model includes optimizing the model's parameter settings and adjusting feature information, such as adding new features or removing irrelevant features; After the adjustment is completed, the model is retrained and its prediction performance is verified on the test set to ensure that the prediction error is reduced to an acceptable range, thereby optimizing and improving the model.
[0024] Optimizing the parameters of the cost prediction model to obtain the optimal model plays an important role in upgrading the cost information database: Providing more accurate input data to make the model closer to reality: Upgrading the cost information database can introduce the latest market data and cost information, such as recent material prices, labor costs and other related costs. These data updates help reduce model prediction deviations caused by the use of outdated or inaccurate data. By providing current actual cost information, the model's input data becomes more accurate, making the prediction results more in line with reality and improving the overall prediction accuracy. Enhance the model's ability to adapt to cost fluctuations and market changes: Market conditions and cost factors are constantly changing, such as material price fluctuations, changes in the labor market, etc. Upgrading the information database can timely incorporate these dynamic changes into the model, allowing the model to reflect and adapt to these fluctuations in real time. By integrating the latest data, the model can better capture and understand the trend of market changes, thereby adjusting the forecasting strategy and enhancing its foresight and adaptability to future cost fluctuations; Help discover and capture new cost patterns and trends and optimize forecast results: Over time, new cost patterns and trends may emerge, such as new building materials, construction technologies, or changes in the economic environment. By updating the cost information base, the model can access these new data points to identify and capture emerging cost trends and patterns. In this way, the model can conduct a more comprehensive analysis and optimize the forecast results to reflect new market realities and trends; In short, the upgrade of the information base ensures that the model can remain efficient and accurate in an ever-changing environment. By providing the latest and relevant data, the model can improve the reliability and accuracy of predictions.
[0025] 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.
[0026] 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.
[0027] 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.
[0028] 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.
[0029] 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.
[0030] 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. A method for upgrading a cost information database based on technical maintenance, characterized in that: The following steps are involved: Obtaining cost data of the substation within a preset time period, wherein the cost data includes historical cost data and actual cost data; Obtain market factors, perform feature analysis on market factors, extract factor feature data that can be quantitatively analyzed, and obtain market factor feature data based on the linear regression model according to the factor feature data; According to the market factor characteristic data and historical cost data, a cost prediction model is constructed based on a convolutional neural network to obtain predicted cost data; The parameters of the cost prediction model are optimized according to the actual cost data and the predicted cost data to obtain the optimal cost prediction model.
2. A method for upgrading a cost information database based on technical maintenance according to claim 1, characterized in that: The historical cost data include initial construction costs, operating costs and maintenance costs; the actual cost data include various actual costs in the substation construction and operation projects.
3. The method for upgrading the cost information database based on technical maintenance according to claim 1 is characterized in that: The specific method for obtaining the market factor characteristic data is as follows: Market factors include raw material prices, labor costs, inflation rates, and market demand. Market factors are cleaned and outliers are removed before feature analysis to extract factor feature data that can be quantitatively analyzed. A linear regression model is created, and the factor characteristic data that can be quantitatively analyzed extracted from market factors is substituted into the linear regression model to obtain the market factor characteristic data.
4. The method for upgrading the cost information database based on technical maintenance according to claim 1 is characterized in that: The specific method for obtaining the cost prediction model is as follows: The construction cost of the substation is used as the fixed step size of the convolution kernel, and the input substation feature information is convolved to generate the corresponding weight coefficient of the feature information; The sensitivity error coefficient and entropy loss function are introduced to calculate the historical cost correction coefficient; Based on the weight coefficient of historical cost characteristic information, combined with the historical cost correction coefficient and market factor characteristic data, a cost prediction model is established.
5. The method for upgrading the cost information database based on technical maintenance according to claim 1 is characterized in that: The specific method for obtaining the optimal cost prediction model is as follows: Compare and analyze the actual cost data and the predicted cost data in the cost data, and calculate the absolute error between the actual cost data and the predicted cost data; When the absolute error is greater than the preset error threshold, the cost prediction model parameters and feature information are adjusted, and the cost prediction model is tested again to optimize the cost prediction model.
6. The method for upgrading the cost information database based on technical maintenance according to claim 1 is characterized in that: The specific calculation formula for the market factor characteristic data is as follows: In the formula, is the market factor characteristic data, are the preset linear regression coefficients, It is the characteristic data of factors that can be quantitatively analyzed. It is a positive integer, indicating the number of factor characteristic data that can be quantitatively analyzed.
7. A method for upgrading a cost information database based on technical maintenance according to claim 5, characterized in that: The specific calculation formula of the weight coefficient of the substation characteristic information is as follows: In the formula, It is Year The weight coefficient of the eigenvalue information is is a nonlinear activation function; For the The feature subset corresponding to the feature value information is is the substation construction cost, For the Year Specific data of characteristic values, After convolution Year eigenvalue information bias term, It is a positive integer, indicating the number of historical cost data obtained.
8. The method for upgrading the cost information database based on technical maintenance according to claim 1 is characterized in that: The specific calculation formula of the correction coefficient is: In the formula, is the historical cost correction factor, is the sensitivity error coefficient, is the entropy loss function.
9. The method for upgrading the cost information database based on technical maintenance according to claim 1 is characterized in that: The specific calculation formula of the sensitivity error coefficient is as follows: In the formula, is the sensitivity error coefficient, It is Year The weight coefficient of the eigenvalue information is is the substation construction cost, After convolution Year eigenvalue information bias term, It is a positive integer, indicating the number of historical cost data obtained; The specific calculation formula of the entropy loss function is as follows: In the formula, is the entropy loss function, For the Year feature data, It is a positive integer, indicating the number of historical cost data obtained.
10. The method for upgrading the cost information database based on technical maintenance according to claim 1, characterized in that: The specific calculation formula of the cost prediction model is as follows: In the formula, It is the forecast cost data. It is Year The weight coefficient of the eigenvalue information is is the market factor characteristic data, It is the historical cost correction factor.