Cost data management system based on electric power construction

Abstract

The invention discloses a cost data management system based on electric power construction, and relates to the field of data management, and the key points of the technical scheme are that a data acquisition and preprocessing module is used for acquiring comprehensive cost data in a cycle time, and preprocessing the comprehensive cost data by using a rule map to obtain standard cost data; the data combination and reconstruction module is used for carrying out combination and reconstruction on the standard cost data to obtain a time-class data packet; the curve fitting module uses a curve fitting model to carry out curve expression on the time class data packet, a standard curve set is obtained, trend analysis is carried out on the standard curve set, and comprehensive evaluation data is obtained based on the comprehensive set; the cost analysis and decision-making module constructs a cost function based on the comprehensive evaluation data, and determines distribution funds of each data category; and the visual output module carries out visual display based on the distribution fund of each data category, so that high-quality management of the cost data in the electric power construction process is realized.

Description

Technical Field

[0001] The present invention relates to the field of data management, and more specifically, to a cost data management system based on electric power construction. Background Art

[0002] In power construction, if you want to use cost data to effectively control costs, you will encounter many problems and difficulties in management. These problems and difficulties mainly focus on the accuracy and timeliness of data.

[0003] The problem of data accuracy is mainly caused by market price fluctuations. Specifically, the prices of materials and equipment are affected by market supply and demand, policies and regulations, and human factors, and fluctuate greatly, resulting in differences between the cost data in the budget stage and the actual purchase price. After changes in market prices and policies and regulations, cost data cannot be updated in a timely manner, resulting in the failure of cost control basis, which in turn deteriorates the timeliness of data. This is mainly due to the lag in data collection, and the inability to timely feedback the on-site construction progress and actual cost information to the cost management department, resulting in the inability to implement cost control measures in a timely manner.

[0004] Therefore, based on the above problems, it is necessary to design cost data management based on power construction. Summary of the invention

[0005] In view of the problems existing in the prior art, the purpose of the present invention is to provide a cost data management system based on electric power construction to achieve high-quality management of cost data during the electric power construction process.

[0006] To achieve the above object, the present invention provides the following technical solution: The cost data management system based on electric power construction includes: Data collection and preprocessing module: used to collect comprehensive cost data within the cycle time, preprocess the comprehensive cost data using rule maps, and obtain standard cost data; Data combination and reconstruction module: used to combine and reconstruct standard cost data to obtain time-class data packets; Curve fitting module: Use the curve fitting model to express the time-class data packets with curves, obtain the standard curve set, perform trend analysis on the standard curve set, obtain the comprehensive group set of each data category, and obtain comprehensive evaluation data based on the comprehensive group set; Cost analysis and decision-making module: constructs a cost function based on comprehensive evaluation data and determines the allocated funds for each data category; Visualization output module: Visualization of allocated funds based on each data category.

[0007] Preferably, the comprehensive cost data includes direct engineering cost data, indirect engineering cost data and other engineering cost data; Direct engineering cost data include material costs, equipment costs, labor costs, machinery usage fees and measures costs; Indirect engineering cost data include regulatory fees and enterprise management fees; Other project cost data include tax data, construction land fees, bidding agency fees and contingency fees.

[0008] Preferably, the method of preprocessing the comprehensive cost data using the rule graph to obtain standard cost data includes: Construct a rule graph, which includes a structure data table and preprocessing rules. Each preprocessing rule is simulated into a node. The relationship between nodes is defined as being arranged in order according to priority and represented by directed edges, so that the nodes are connected by directed edges to form a rule network. Import the structure data table into each node of the rule expansion network in sequence, and output the pre-processed standard cost data; Preprocessing rules include input rules, type format rules, cleaning rules, standard rules and output rules. Each node includes rule name, rule description, triggering condition, processing method and priority. Among them, the rule description of the input rule is used for data input, the rule description of the type format rule is used to determine whether the format of the data conforms to the preset data standard format, the rule description of the cleaning rule is used to determine whether the data is normal, the rule description of the standard rule is used to standardize the data, and the rule description of the output rule is used for data output; The trigger condition is defined as the condition that triggers the execution of the rule, the processing method is defined as the processing method when the data meets the rule or does not meet the rule, and the priority is defined as the order in which the nodes process the data.

[0009] Preferably, the method for constructing the structure data table includes: Extract the comprehensive cost data within the cycle time, discretize the cycle time into O_p time points, and correspond the time points to the timestamps of the comprehensive cost data; Create a blank table with three horizontal spaces, which are used to fill in direct engineering cost data, indirect engineering cost data and other engineering cost data respectively. Each space is set with dimensions according to the categories of the filled data. Set O_p spaces horizontally in the table, and fill in O_p time points in chronological order; Extract each value of the comprehensive cost data at each time point, fill the value into the corresponding position of the table, and form a structural data table within the cycle time.

[0010] Preferably, the method for combining and reconstructing standard cost data includes: In the cycle time, take O_p discrete time points as time set groups, take this time set group as the basis, generate new time set groups by shifting, repeat the shifting in sequence, obtain d_b time set groups, and each time set group is different; For any time set group, extract the time points corresponding to each data category in the standard cost data, and arrange and combine the data values ​​of the same data category in time order to form a category data group; Collecting the category data groups of all data categories forms a time-class data set; Collect the corresponding time class data sets on all time set groups to form a time class data packet.

[0011] Preferably, the method of using a curve fitting model to express the time-class data packet by curves to obtain a standard curve set includes: Step A1: construct a curve fitting model based on a neural network, set the input of the model as a time class data set, and the output as a class curve set, and the class curve set includes curve expressions of each data class in the time class data set; The constraints of the model are set as ,in, Indicates The category data group corresponding to each data category is expressed by the curve output by the model. Indicates The actual curve expression of the category data group corresponding to the data category, is the limit threshold, represents the integral over time; Step A2: Set a sample set, which includes S_H samples. Each sample includes a time class data set and a corresponding class curve set. Use the sample set to train the curve fitting model. When the constraint conditions are met, stop and obtain the trained model. Step A3: Import the time class data package into the trained model to obtain a standard curve set; The standard curve set includes d_b curve expressions for each data category on d_b time set groups.

[0012] Preferably, the method of performing trend analysis on the standard curve set to obtain comprehensive sets of various data categories and obtaining comprehensive evaluation data based on the comprehensive sets includes: Step E1: using a clustering algorithm to perform cluster analysis on the standard curve set to obtain a comprehensive group set; Step E2: Extract the individual sets of each data category in the comprehensive set and calculate the standard features of the individual sets And typical trend ; in, Indicates The number of curves expressed in each group, Indicates the total number of curve expressions, Indicates The standard deviation of the curve in the group, Indicates The average curve expression in each group; Step E3: Take the standard features and typical trends of a data category as a feature evaluation, and collect the feature evaluations of all data categories to form comprehensive evaluation data.

[0013] Preferably, the method of using a clustering algorithm to perform cluster analysis on a set of standard curves to obtain a comprehensive set includes: Step D1: For d_b curve expressions of each data category in the standard curve set, randomly select K_a curve expressions as cluster centers; Step D2: Calculate the distance between each curve expression and the cluster center, and assign it to the cluster center with the nearest distance, dividing the curve expressions into K_a groups; Step D3: Identify the number of curve expressions in each group. If the number of curve expressions is less than the limit threshold, release the curve expressions in the group, delete the cluster center corresponding to the group, recalculate the distance between the released curve expressions and the remaining cluster centers, and redistribute them to obtain the groups K_b into which the updated curve expressions are divided; Step D4: Update the cluster center of each group ,in, Represents the curve expression within the cycle time, is the number of curve expressions in the group; Step D5: Repeat steps D2 to D4 until all cluster centers no longer change, obtain the curve expressions divided into K_c groups, and collect the curve expressions of K_c groups to form a single group set; Step D6: Collect individual sets of all data categories to form a comprehensive set.

[0014] Preferably, the cost function is constructed based on the comprehensive evaluation data to determine the allocated funds for each data category, and the specific method includes: Step F1: For the comprehensive evaluation data, use the prediction model to obtain the price data of each data category in the next cycle; Step F2: Calculate the funding allocation weight for each data category, and Fund allocation weights for each data category ; in, Indicates The predicted price of each data category in the next cycle time, Indicates The actual average price of each data category in the current cycle time, Indicates Standard features for each data category, is the index of the data category; Step F3: Set the number of seconds allocated to the first Funds in data categories ; in, Indicates Total budget for each data category; Step F4: Set the objective function to , the constraints are .

[0015] Preferably, the visualization method includes using visualization tools and visualization algorithms to intuitively display data.

[0016] Compared with the prior art, the present invention has the following beneficial effects: Through the data collection and preprocessing module, the system can collect comprehensive cost data from multiple sources and preprocess it through rule maps to ensure the standardization and accuracy of the data. Each rule node processes the data in an orderly manner, eliminating the problems of inconsistent data formats and abnormal data, improving the consistency and reliability of data processing, and deleting or correcting abnormal data through cleaning rules to ensure the accuracy and completeness of the data.

[0017] The data combination and reconstruction module reconstructs the standard cost data within the cycle time to generate multiple time-class data packets, providing a variety of data combination methods for subsequent analysis. By generating time set groups by dislocation and extracting corresponding category data groups, the system can flexibly reconstruct data and improve the diversity and depth of data analysis. The data reconstruction method is flexible and diverse, which can adapt to different analysis needs and provide a variety of data combination methods, providing a rich data foundation for subsequent trend analysis and decision support. Using time set groups to combine and reconstruct data can better capture the time characteristics of data and improve the accuracy of data analysis.

[0018] The curve fitting module uses a neural network model to express the time-class data packets in a curve, obtain a set of standard curves, and then perform trend analysis, providing efficient curve fitting and prediction capabilities. The clustering analysis method can effectively classify the standard curve set, extract the standard features and typical trends of each data category, and form comprehensive evaluation data. Efficient fitting: The curve fitting model based on the neural network has efficient learning and prediction capabilities and can accurately fit complex time series data. The clustering analysis method can effectively identify the potential patterns and trends of the data and provide strong support for decision-making.

[0019] The cost analysis and decision-making module constructs a cost function based on comprehensive evaluation data, determines the funding allocation for each data category, and provides scientific cost analysis and decision-making support. Through the prediction model and optimization objective function, the system can accurately calculate the funding allocation weight and allocation amount for each data category to ensure the optimal use of funds. Based on the comprehensive evaluation data and prediction model, it provides a scientific funding allocation plan to support decision makers in accurate cost management. Through objective function optimization, it ensures the reasonable allocation and effective use of funds and improves the economic benefits of project management.

[0020] The visualization output module uses visualization tools and algorithms to intuitively display the allocation of funds and trend analysis results for each data category, making it easier to understand and analyze. Through charts, dashboards, and other forms, decision makers can intuitively view the allocation of funds and project cost trends, improving decision-making efficiency.

[0021] In summary, the system realizes the comprehensive management and optimization of power construction project cost data through standardized data processing, flexible reconstruction and combination, efficient curve fitting and trend analysis, scientific cost analysis and decision support, and intuitive visual display, thereby improving the overall efficiency and economic benefits of project management. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The present invention proposes a structural schematic diagram of a cost data management system based on electric power construction; Figure 2 It is a schematic diagram of the method of the present invention applied to the cost data management system based on electric power construction. DETAILED DESCRIPTION

[0023] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here. Embodiment 1

[0024] Reference Figure 1 and Figure 2 Embodiment 1 further illustrates a cost data management system based on electric power construction proposed by the present invention.

[0025] In the process of power construction, cost data refers to all data and information related to the cost of engineering construction. It covers all expenses incurred during the entire life cycle of the project from project initiation to completion acceptance, and is an important basis for project investment control, cost management, and benefit evaluation.

[0026] In power construction, if you want to use cost data to effectively control costs, you will encounter many problems and difficulties in management. These problems and difficulties mainly focus on the accuracy and timeliness of data.

[0027] The problem of data accuracy is mainly caused by market price fluctuations. Specifically, the prices of materials and equipment are affected by market supply and demand, policies and regulations, and human factors, and fluctuate greatly, resulting in differences between the cost data in the budget stage and the actual purchase price. After changes in market prices and policies and regulations, cost data cannot be updated in a timely manner, resulting in the failure of cost control basis, which in turn deteriorates the timeliness of data. This is mainly due to the lag in data collection, and the inability to timely feedback the on-site construction progress and actual cost information to the cost management department, resulting in the inability to implement cost control measures in a timely manner.

[0028] Data collection and preprocessing module: used to collect comprehensive cost data within the cycle time, preprocess the comprehensive cost data using rule maps, and obtain standard cost data; Comprehensive cost data includes direct engineering cost data, indirect engineering cost data and other engineering cost data; Direct engineering cost data include material costs, equipment costs, labor costs, machinery usage fees and measures costs; Material costs refer to the costs of various raw materials, auxiliary materials, components, semi-finished products, etc. that constitute the engineering entity, including the original price of materials, transportation fees, procurement and storage fees, etc. For example: the price and quantity of materials such as steel, cement, cables, transformers, insulators, etc.

[0029] Equipment costs refer to the costs of purchasing or self-making various production equipment, power equipment, transmission equipment, transportation equipment, testing equipment, etc. that meet the fixed asset standards, including the original price of the equipment, transportation and miscellaneous expenses, installation fees, etc. For example: the price and quantity of equipment such as generator sets, transformers, circuit breakers, control systems, etc.

[0030] Labor costs refer to the wages, bonuses, allowances, subsidies and other expenses of production workers directly engaged in project construction.

[0031] Machinery usage fees refer to the depreciation, maintenance, fuel and power costs, labor costs, etc. incurred when using various construction machinery during the construction process.

[0032] Measures fees refer to the costs incurred for various technical measures, safety measures, civilized construction measures, etc. taken to ensure the smooth progress of the project, such as scaffolding erection, temporary facilities construction, safety protection and other costs.

[0033] Indirect engineering cost data include regulatory fees and enterprise management fees; Regulatory fees refer to fees that must be paid as stipulated by the government and relevant departments, such as engineering sewage fees, social security fees, housing provident funds, etc.

[0034] Enterprise management fees refer to the various expenses incurred by construction companies in organizing and managing construction production activities, such as management staff salaries, office expenses, travel expenses, insurance premiums, and survey and design fees.

[0035] Other project cost data include tax data, construction land fees, bidding agency fees and contingency fees.

[0036] Tax data refers to the various taxes and fees payable as stipulated by the state, such as value-added tax, urban maintenance and construction tax, and education surcharge.

[0037] Construction land fees refer to the fees paid for obtaining the right to use construction land.

[0038] The bidding agency fee refers to the fees paid by entrusting a bidding agency to conduct bidding agency work.

[0039] The reserve fund refers to the reserve for unforeseen expenses that may occur during the construction process.

[0040] The above cost data can be obtained through the following methods, such as market inquiry (obtaining price information of materials and equipment by inquiring material suppliers, equipment manufacturers, etc.), quota pricing (pricing based on the engineering cost quotas issued by the state or industry to obtain cost information such as labor costs and machinery usage fees), historical data (referring to the cost data of similar projects in the past for estimation and analysis), professional databases (using professional engineering cost databases to query relevant cost information) and network tool crawling (using information crawling tools to crawl and update the latest price information or policy information, such as Internet of Things (IoT) device integration: installing IoT devices such as sensors and RFID tags at the construction site to automatically collect data such as material consumption, equipment usage, and man-hours. Electronic procurement platform docking: docking with the electronic procurement platform to automatically obtain the price, purchase quantity, supplier information, etc. of materials and equipment. Mobile application data entry: providing mobile applications for on-site personnel to facilitate them to enter on-site data in real time, such as progress, changes, acceptance, etc. API interface integration: integrating data with other systems (such as ERP, project management system) through API interfaces to achieve automatic data synchronization.).

[0041] Methods for preprocessing comprehensive cost data using rule graphs to obtain standard cost data include: Construct a rule graph, which includes a structure data table and preprocessing rules. Each preprocessing rule is simulated into a node. The relationship between nodes is defined as being arranged in order according to priority and represented by directed edges, so that the nodes are connected by directed edges to form a rule network. Import the structure data table into each node of the rule expansion network in sequence, and output the pre-processed standard cost data; Preprocessing rules include input rules, type format rules, cleaning rules, standard rules and output rules. Each node includes rule name, rule description, triggering condition, processing method and priority. The rule name of each node is the corresponding input rule, type format rule, cleaning rule, standard rule and output rule; Among them, the rule description of the input rule is used for data input, the rule description of the type format rule is used to determine whether the format of the data conforms to the preset data standard format, the rule description of the cleaning rule is used to determine whether the data is normal, the rule description of the standard rule is used to standardize the data, and the rule description of the output rule is used for data output; The trigger condition is defined as the condition that triggers the execution of the rule, for example, it can be set as data input or import. The processing method is defined as the processing method when the data meets the rule or does not meet the rule. For example, when the data does not meet the rule, it can be processed by correction, deletion, filling, etc. and then output. When the data meets the rule, it is output normally. The priority is defined as the order in which the nodes process the data. The preset data standard format refers to a data category expression format that is pre-set according to the data category. For example, it stipulates that the "equipment number" field must be "letters + numbers", and that the "material unit price" must be "value + unit".

[0042] The cleaning rule is to determine whether the data meets the pre-set standards. Assume that the pre-set standards are to limit the values ​​of each data type to a reasonable range. For example, the price of steel materials is pre-set within the range of 1000-5000 based on historical prices and predicted prices. If a value of 120000 suddenly appears, it means that this value is beyond the normal range. The value bytes of each data type can also be superimposed and set. If the byte is 0, it means that the value corresponding to the data type is missing. If the byte exceeds the standard, it means that the value corresponding to the data type is repeated, etc. Cleaning rules are superimposed and set according to the required data processing functions.

[0043] Data can be standardized by unit unification, or by data standardization and normalization, so that the data is under a unified standard.

[0044] The rule map checks and filters the data according to pre-defined rules to determine whether the data complies with these rules, thereby achieving data cleaning and data quality control.

[0045] The construction method of the structure data table includes: Extract the comprehensive cost data within the cycle time, discretize the cycle time into O_p time points, and correspond the time points to the timestamps of the comprehensive cost data; Create a blank table with three horizontal spaces set up to fill in direct engineering cost data, indirect engineering cost data and other engineering cost data respectively. Set the dimension of each space according to the category of the filled-in data; for example, direct engineering cost data includes material costs, equipment costs, labor costs, machinery usage fees and measures fees, then the dimension of the space corresponding to the direct engineering cost data is set to 5.

[0046] Set O_p spaces horizontally in the table, and fill in O_p time points in chronological order; Extract each value of the comprehensive cost data at each time point, fill the value into the corresponding position of the table, and form a structural data table within the cycle time.

[0047] Data combination and reconstruction module: used to combine and reconstruct standard cost data to obtain time-class data packets; The methods for combining and reconstructing standard cost data include: In the cycle time, take O_p discrete time points as time set groups, take this time set group as the basis, generate new time set groups by shifting, repeat the shifting in sequence, obtain d_b time set groups, and each time set group is different; For any time set group, extract the time points corresponding to each data category in the standard cost data, and arrange and combine the data values ​​of the same data category in time order to form a category data group; Collecting the category data groups of all data categories forms a time-class data set; Collect the corresponding time class data sets on all time set groups to form a time class data packet.

[0048] For example, within the cycle time 1-3, there are 5 discrete time points, namely 1, 1.5, 2, 2.5, and 3, forming a time set group (1, 1.5, 2, 2.5, 3). Taking (1, 1.5, 2, 2.5, 3) as the benchmark, a new time set group (1.1, 1.6, 2.1, 2.6, 2.9) is generated by shifting, and the shifting is repeated in sequence. Assuming that 6 time sets are obtained, then for the data category construction land fee, there will be 6 time sets within the cycle time, that is, there will be 6 curve expressions.

[0049] Curve fitting module: Use the curve fitting model to express the time-class data packets with curves, obtain the standard curve set, perform trend analysis on the standard curve set, obtain the comprehensive group set of each data category, and obtain comprehensive evaluation data based on the comprehensive group set; The method of using the curve fitting model to express the standard cost data by curve and obtaining the standard curve set includes: Step A1: construct a curve fitting model based on a neural network, set the input of the model as a time class data set, and the output as a class curve set, and the class curve set includes curve expressions of each data class in the time class data set; The constraints of the model are set as ,in, Indicates The category data group corresponding to each data category is expressed by the curve output by the model. Indicates The actual curve expression of the category data group corresponding to the data category, is the limit threshold, Represents the integral over time; the limit threshold can be pre-set based on historical data analysis, experimental data analysis or work experience.

[0050] Step A2: Set a sample set, which includes S_H samples. Each sample includes a time class data set and a corresponding class curve set. Use the sample set to train the curve fitting model. When the constraint conditions are met, stop and obtain the trained model. Step A3: Import the time class data package into the trained model to obtain a standard curve set; The standard curve set includes d_b curve expressions for each data category on d_b time set groups.

[0051] Perform trend analysis on the standard curve set to obtain comprehensive sets of various data categories. The method for obtaining comprehensive evaluation data based on the comprehensive sets includes: Step E1: using a clustering algorithm to perform cluster analysis on the standard curve set to obtain a comprehensive group set; Step E2: Extract the individual sets of each data category in the comprehensive set and calculate the standard features of the individual sets And typical trend ; in, Indicates The number of curves expressed in each group, Indicates the total number of curve expressions, Indicates The standard deviation of the curve in the group, Indicates The average curve expression in each group; By calculating the standard deviation of each cluster group, the volatility of the price within the group can be evaluated. The larger the standard deviation, the more drastic the price fluctuation; the smaller the standard deviation, the more stable the price fluctuation. The "typical trend" is used as a representative trend curve of the data category for prediction and decision-making reference.

[0052] Step E3: Take the standard features and typical trends of a data category as a feature evaluation, and collect the feature evaluations of all data categories to form comprehensive evaluation data.

[0053] The method of using a clustering algorithm to perform cluster analysis on a set of standard curves to obtain a comprehensive set of groups includes: Step D1: For d_b curve expressions of each data category in the standard curve set, randomly select K_a curve expressions as cluster centers; Step D2: Calculate the distance between each curve expression and the cluster center, and assign it to the cluster center with the nearest distance, dividing the curve expressions into K_a groups; The distance here can be the absolute difference between the two curves expressed in cycle time, or it can be the closed area of ​​the encircled city expressed by the two curves in cycle time.

[0054] Step D3: Identify the number of curve expressions in each group. If the number of curve expressions is less than the limit threshold, release the curve expressions in the group, delete the cluster center corresponding to the group, recalculate the distance between the released curve expressions and the remaining cluster centers, and redistribute them to obtain the groups K_b into which the updated curve expressions are divided; Step D4: Update the cluster center of each group ,in, Represents the curve expression within the cycle time, is the number of curve expressions in the group; Step D5: Repeat steps D2 to D4 until all cluster centers no longer change, obtain the curve expressions divided into K_c groups, and collect the curve expressions of K_c groups to form a single group set; Step D6: Collect individual sets of all data categories to form a comprehensive set.

[0055] Cost analysis and decision-making module: constructs a cost function based on comprehensive evaluation data and determines the allocated funds for each data category; In order to decide the purchase or fund allocation of each data category in the next cycle, a cost function needs to be constructed. If a data category is likely to increase in the future, the purchase volume of this category should be increased first; if the price is likely to drop, the current purchase volume can be reduced and purchase can be made after the price drops.

[0056] Based on the comprehensive evaluation data, a cost function is constructed to determine the allocated funds for each data category. The specific methods include: Step F1: For the comprehensive evaluation data, use the prediction model to obtain the price data of each data category in the next cycle; Prediction models (such as the time series model ARIMA or the deep learning model LSTM) are used to obtain future data (prices) for each data category within the next cycle.

[0057] For the ARIMA model: determine the parameters of the ARIMA model, including the number of autoregressive terms, the order of differencing, and the number of moving average terms; use grid search and cross-validation to find the best parameter combination; and use mean squared error (MSE) to evaluate model performance.

[0058] For the LSTM model: Design the LSTM network structure, including determining the size and number of hidden layers; convert the comprehensive evaluation data into a format suitable for LSTM input, usually a three-dimensional array; use mean square error (MSE) as the loss function, and perform backpropagation and parameter optimization.

[0059] Use the trained ARIMA or LSTM model to evaluate the features of each data category and make price predictions at future time points.

[0060] Step F2: Calculate the funding allocation weight for each data category, and Fund allocation weights for each data category ; in, Indicates The predicted price of each data category in the next cycle time, Indicates The actual average price of each data category in the current cycle time, Indicates Standard features for each data category, is the index of the data category; Step F3: Set the number of seconds allocated to the first Funds in data categories ; in, Indicates Total budget for each data category; Step F4: Set the objective function to , the constraints are ; Based on the clustering results of data categories and predicted price trends, funds can be allocated reasonably to achieve a balance between risk and return.

[0061] Visualization output module: Visualization of allocated funds based on each data category; Visual display methods include using visualization tools and visualization algorithms to display data intuitively.

[0062] Visualization tools such as Neo4j, Gephi, etc. use data visualization algorithms (such as scatter plots, line charts, bar charts, heat maps, etc.) to convert data into intuitive graphics, presenting complex cost data in an intuitive way, making it easier for users to analyze and make decisions.

[0063] You can also design various dashboards to display key indicators, such as budget execution, cost deviation, material price trends, etc. You can also generate various reports, such as cost analysis reports, budget execution reports, material consumption reports, etc.

[0064] The rule nodes and relationships are visualized to form a clear and intuitive rule map. Visualization can help us better understand the relationship between rules and facilitate the maintenance and management of rules.

[0065] This method is also applicable to the following scenarios: Analyze the price trends of different energy types (such as photovoltaic, wind energy, power grid, etc.), optimize procurement or allocation strategies, reduce costs and improve efficiency in energy transactions.

[0066] Analyze and predict price changes of different raw materials or commodities in the supply chain to optimize procurement plans and reduce inventory costs.

[0067] In order to ensure the security and confidentiality of cost data, different permissions are assigned according to user roles, user access and operations to data are controlled, and sensitive data is encrypted for storage and transmission to prevent data leakage.

[0068] Through the above methods, the problems and difficulties encountered in cost control management of cost data can be effectively solved, the cost control level of power construction projects can be improved, and the investment benefits can be maximized.

[0069] Embodiment 2 Reference Figure 1 and Figure 2 , Example 2 further illustrates a cost data management system based on electric power construction proposed by the present invention.

[0070] The cost data management method based on electric power construction comprises the following steps: Step S1: continuously collect comprehensive cost data within a period of time, and use a rule map to pre-process the comprehensive cost data to obtain standard cost data; Step S2: Combine and reconstruct the standard cost data to obtain a time-based data packet; Step S3: construct and train a curve fitting model, input the time class data packet into the trained curve fitting model, and obtain a standard curve set; Step S4: Calculate the standard characteristics and typical trends of each data category in the comprehensive group set to form comprehensive evaluation data; Step S5: By constructing a cost function, the fund allocation weight of each data category is calculated to determine the allocated funds for each data category; Step S6: The visualization interface displays the allocated funds for each data category to form an interface interaction.

[0071] The cost data management system based on electric power construction is applied to a cost data management method based on electric power construction and includes: Data collection and preprocessing module: used to collect comprehensive cost data within the cycle time, preprocess the comprehensive cost data using rule maps, and obtain standard cost data; Data combination and reconstruction module: used to combine and reconstruct standard cost data to obtain time-class data packets; Curve fitting module: Use the curve fitting model to express the time-class data packets with curves, obtain the standard curve set, perform trend analysis on the standard curve set, obtain the comprehensive group set of each data category, and obtain comprehensive evaluation data based on the comprehensive group set; Cost analysis and decision-making module: constructs a cost function based on comprehensive evaluation data and determines the allocated funds for each data category; Visualization output module: Visualization of allocated funds based on each data category; Summarize: Using rule graphs to preprocess data can effectively clean and standardize data and improve data quality.

[0072] Using time set groups to combine and reconstruct data can better capture the temporal characteristics of the data and improve the accuracy of data analysis.

[0073] Using neural networks to build curve fitting models can better express data trends and improve the accuracy of data expression.

[0074] Using clustering algorithms to perform cluster analysis on a set of standard curves can identify potential patterns in the data and improve the accuracy of trend analysis.

[0075] Using the prediction model and the objective function to construct a cost function can scientifically determine the allocation of funds for each data category and improve the efficiency of fund use.

[0076] The cost data management system solution based on power construction can effectively improve project cost control efficiency, optimize fund allocation, and enhance project management level by combining data collection, preprocessing, analysis, visualization and other technologies. Its technical solution is innovative and practical, and can provide strong data support for power construction projects.

[0077] The modules are connected to each other via wired and / or wireless means.

[0078] In addition, according to the implementation of the present application, the process described in the attached figure of a cost data management system based on electric power construction can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application. Of course, the architecture shown in the attached figure of a cost data management system based on electric power construction is only exemplary. When implementing different devices, adaptive selection or adjustment can be performed according to actual needs.

[0079] 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 and thresholds in the formula are set by technicians in this field according to actual conditions.

[0080] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A cost data management system based on electric power construction, characterized in that: The cost data management system based on electric power construction includes: Data collection and preprocessing module: used to collect comprehensive cost data within the cycle time, preprocess the comprehensive cost data using rule maps, and obtain standard cost data; Data combination and reconstruction module: used to combine and reconstruct standard cost data to obtain time-class data packets; Curve fitting module: Use the curve fitting model to express the time-class data packets with curves, obtain the standard curve set, perform trend analysis on the standard curve set, obtain the comprehensive group set of each data category, and obtain comprehensive evaluation data based on the comprehensive group set; Cost analysis and decision-making module: constructs a cost function based on comprehensive evaluation data and determines the allocated funds for each data category; Visualization output module: Visualization of allocated funds based on each data category.

2. The cost data management system based on electric power construction according to claim 1 is characterized in that: The comprehensive cost data includes direct engineering cost data, indirect engineering cost data and other engineering cost data; Direct engineering cost data include material costs, equipment costs, labor costs, machinery usage fees and measures costs; Indirect engineering cost data include regulatory fees and enterprise management fees; Other project cost data include tax data, construction land fees, bidding agency fees and contingency fees.

3. The cost data management system based on electric power construction according to claim 2 is characterized in that: The method of preprocessing the comprehensive cost data using the rule graph to obtain standard cost data includes: Construct a rule graph, which includes a structure data table and preprocessing rules. Each preprocessing rule is simulated into a node. The relationship between nodes is defined as being arranged in order according to priority and represented by directed edges, so that the nodes are connected by directed edges to form a rule network. Import the structure data table into each node of the rule expansion network in sequence, and output the pre-processed standard cost data; Preprocessing rules include input rules, type format rules, cleaning rules, standard rules and output rules. Each node includes rule name, rule description, triggering condition, processing method and priority. Among them, the rule description of the input rule is used for data input, the rule description of the type format rule is used to determine whether the format of the data conforms to the preset data standard format, the rule description of the cleaning rule is used to determine whether the data is normal, the rule description of the standard rule is used to standardize the data, and the rule description of the output rule is used for data output; The trigger condition is defined as the condition that triggers the execution of the rule, the processing method is defined as the processing method when the data meets the rule or does not meet the rule, and the priority is defined as the order in which the nodes process the data.

4. The cost data management system based on electric power construction according to claim 3 is characterized in that: The method for constructing the structure data table includes: Extract the comprehensive cost data within the cycle time, discretize the cycle time into O_p time points, and correspond the time points to the timestamps of the comprehensive cost data; Create a blank table with three horizontal spaces, which are used to fill in direct engineering cost data, indirect engineering cost data and other engineering cost data respectively. Each space is set with dimensions according to the categories of the filled data. Set O_p spaces horizontally in the table, and fill in O_p time points in chronological order; Extract each value of the comprehensive cost data at each time point, fill the value into the corresponding position of the table, and form a structural data table within the cycle time.

5. The cost data management system based on electric power construction according to claim 4 is characterized in that: The method for combining and reconstructing standard cost data includes: In the cycle time, take O_p discrete time points as time set groups, take this time set group as the basis, generate new time set groups by shifting, repeat the shifting in sequence, obtain d_b time set groups, and each time set group is different; For any time set group, extract the time points corresponding to each data category in the standard cost data, and arrange and combine the data values ​​of the same data category in time order to form a category data group; Collecting the category data groups of all data categories forms a time-class data set; Collect the corresponding time class data sets on all time set groups to form a time class data packet.

6. The cost data management system based on electric power construction according to claim 5 is characterized in that: The method of using the curve fitting model to express the time class data packet by curve to obtain the standard curve set includes: Step A1: construct a curve fitting model based on a neural network, set the input of the model as a time class data set, and the output as a class curve set, and the class curve set includes curve expressions of each data class in the time class data set; The constraints of the model are set as ,in, Indicates The category data group corresponding to each data category is expressed by the curve output by the model. Indicates The actual curve expression of the category data group corresponding to the data category, is the limit threshold, represents the integral over time; Step A2: Set a sample set, which includes S_H samples. Each sample includes a time class data set and a corresponding class curve set. Use the sample set to train the curve fitting model. When the constraint conditions are met, stop and obtain the trained model. Step A3: Import the time class data package into the trained model to obtain a standard curve set; The standard curve set includes d_b curve expressions for each data category on d_b time set groups.

7. The cost data management system based on electric power construction according to claim 6 is characterized in that: The method of performing trend analysis on the standard curve set to obtain comprehensive sets of various data categories and obtaining comprehensive evaluation data based on the comprehensive sets includes: Step E1: using a clustering algorithm to perform cluster analysis on the standard curve set to obtain a comprehensive group set; Step E2: Extract the individual sets of each data category in the comprehensive set and calculate the standard features of the individual sets And typical trend ; in, Indicates The number of curves expressed in each group, Indicates the total number of curve expressions, Indicates The standard deviation of the curve in the group, Indicates The average curve expression in each group; Step E3: Take the standard features and typical trends of a data category as a feature evaluation, and collect the feature evaluations of all data categories to form comprehensive evaluation data.

8. The cost data management system based on electric power construction according to claim 7 is characterized in that: The method of using a clustering algorithm to perform cluster analysis on a set of standard curves to obtain a comprehensive set includes: Step D1: For d_b curve expressions of each data category in the standard curve set, randomly select K_a curve expressions as cluster centers; Step D2: Calculate the distance between each curve expression and the cluster center, and assign it to the cluster center with the nearest distance, dividing the curve expressions into K_a groups; Step D3: Identify the number of curve expressions in each group. If the number of curve expressions is less than the limit threshold, release the curve expressions in the group, delete the cluster center corresponding to the group, recalculate the distance between the released curve expressions and the remaining cluster centers, and redistribute them to obtain the groups K_b into which the updated curve expressions are divided; Step D4: Update the cluster center of each group ,in, Represents the curve expression within the cycle time, is the number of curve expressions in the group; Step D5: Repeat steps D2 to D4 until all cluster centers no longer change, obtain the curve expressions divided into K_c groups, and collect the curve expressions of K_c groups to form a single group set; Step D6: Collect individual sets of all data categories to form a comprehensive set.

9. The cost data management system based on electric power construction according to claim 8 is characterized in that: The cost function is constructed based on the comprehensive evaluation data to determine the allocated funds for each data category. The specific method includes: Step F1: For the comprehensive evaluation data, use the prediction model to obtain the price data of each data category in the next cycle; Step F2: Calculate the funding allocation weight for each data category, and Fund allocation weights for each data category ; in, Indicates The predicted price of each data category in the next cycle time, Indicates The actual average price of each data category in the current cycle time, Indicates Standard features for each data category, is the index of the data category; Step F3: Set the number of seconds allocated to the first Funds in data categories ; in, Indicates Total budget for each data category; Step F4: Set the objective function to , the constraints are .

10. A cost data management system based on electric power construction, characterized in that: The visualization method includes using visualization tools and visualization algorithms to intuitively display data.

Images