Power transmission and transformation project cost review system based on big data
Through big data technology, combined with gray correlation analysis and clustering algorithm, a cluster-level cost prediction model is constructed, which solves the problem of large cost prediction deviation in traditional methods, and achieves more accurate and flexible cost prediction for power transmission and transformation projects.
Patent Information
- Application Number
- CN202411453954.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-17
- Publication Date
- 2024-12-31
AI Technical Summary
The traditional cost prediction method for power transmission and transformation projects assumes that there is a linear relationship between characteristic factors and cost, and it is impossible to effectively analyze the nonlinear and interactive effects, resulting in large prediction deviations and the incomplete analysis of the relationship between multiple characteristic factors, resulting in insufficient prediction results. precise.
Using a cost review system for power transmission and transformation projects based on big data, the key characteristic factors are identified through gray correlation analysis and k-means clustering algorithm, and a cluster hierarchical cost prediction model is constructed, and the dimension reduction is reduced by combining principal component analysis and self-attention mechanism. Feature extraction, and build an efficient cost prediction model.
It improves the accuracy and robustness of cost prediction, reduces the complexity of model and computational overhead, is suitable for complex and large-scale transmission and transformation engineering projects, dynamically deals with multi-dimensional factors, and enhances the ability to capture nonlinear relationships.
Smart Images
Figure CN119228316A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of cost review, and more specifically, to a cost review system for power transmission and transformation engineering projects based on big data. Background Art
[0002] Power transmission and transformation projects are an important part of the power system, responsible for the transmission and transformation of electric energy. Their construction costs are often high and involve a variety of complex equipment and materials. Therefore, the cost review of power transmission and transformation projects is particularly critical in power transmission and transformation projects. With the continuous growth of electricity demand, the scale and complexity of power transmission and transformation projects have also increased, and cost management and control face new challenges.
[0003] At present, traditional power transmission and transformation project cost prediction methods usually assume that there is a linear relationship between characteristic factors and cost, while the complex factors such as equipment, materials, and construction environment involved in modern power transmission and transformation projects often have nonlinear and interactive effects, resulting in large prediction deviations; secondly, traditional methods cannot comprehensively analyze the relationship between multiple characteristic factors, and are prone to omitting key characteristic factors or over-focusing on certain minor factors, resulting in inaccurate prediction results. 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, an embodiment of the present invention provides a power transmission and transformation project cost review system based on big data, which solves the problem of low cost prediction accuracy in power transmission and transformation station projects.
[0005] To achieve the above object, the present invention provides the following technical solutions: The cost review system for power transmission and transformation project based on big data includes a data acquisition module, a feature engineering module, a model construction module and a cost prediction review module, and there are connections between the modules: the data acquisition module is used to acquire the engineering data and the characteristic factor data of the power transmission and transformation project and perform preprocessing; the feature engineering module is used to calculate the grey correlation between the characteristic factor data of the power transmission and transformation project and the cost of the power transmission and transformation project by using the grey correlation analysis method, and divide the characteristic factor data of the power transmission and transformation project into k power transmission and transformation project characteristic clusters according to the corresponding grey correlation based on the k-means clustering algorithm; the model construction module is used to perform dimensionality reduction processing on the k power transmission and transformation project characteristic clusters by using the principal component analysis method, and construct a cluster-level power transmission and transformation project cost prediction model; the cost prediction review module inputs the engineering data of the power transmission and transformation project into the cluster-level power transmission and transformation project cost prediction model, obtains the power transmission and transformation project cost prediction value, and compares the power transmission and transformation project cost prediction value with a preset cost threshold value to perform cost review warning. In a preferred embodiment, the grey correlation analysis method is used to calculate the grey correlation between the characteristic factor data of the power transmission and transformation project and the cost of the power transmission and transformation project, specifically: the characteristic factor data of the power transmission and transformation project is standardized by the Min-Max normalization method to obtain the standardized characteristic factor data of the power transmission and transformation project; the grey correlation between the characteristic factor data of the power transmission and transformation project and the cost of the power transmission and transformation project is calculated by the gray correlation analysis formula.
[0006] In a preferred embodiment, the k-means clustering algorithm is used to divide the power transmission and transformation project characteristic factor data into k power transmission and transformation project characteristic clusters according to the corresponding grey correlation degrees, specifically: setting an initial k value, and selecting k grey correlation degrees from the grey correlation degrees corresponding to the characteristic factor data of each power transmission and transformation project as the centers of the initial power transmission and transformation project characteristic clusters; calculating the Euclidean distance between the grey correlation degree corresponding to each power transmission and transformation project characteristic factor data and the center of the initial power transmission and transformation project characteristic cluster; based on the Euclidean distance, allocating power transmission and transformation project characteristic clusters and updating the centers of the power transmission and transformation project characteristic clusters; repeating the allocation of power transmission and transformation project characteristic clusters and updating the centers of the power transmission and transformation project characteristic clusters to obtain k power transmission and transformation project characteristic clusters.
[0007] In a preferred embodiment, the k power transmission and transformation engineering feature clusters are subjected to dimensionality reduction processing using a principal component analysis method, specifically: the intra-cluster characteristic factors in the k power transmission and transformation engineering feature clusters are converted into m power transmission and transformation engineering feature matrices; based on the m power transmission and transformation engineering feature matrices, a principal component analysis method is used to perform dimensionality reduction processing.
[0008] In a preferred embodiment, the principal component analysis method is specifically as follows: calculating the covariance matrix of the k power transmission and transformation project feature clusters and the eigenvalues and eigenvectors of the covariance matrix; based on the eigenvalues, sorting the eigenvectors in descending order, selecting the first n eigenvectors to convert the k power transmission and transformation project feature clusters into principal component space, performing dimensionality reduction, and obtaining n power transmission and transformation project feature clusters.
[0009] In a preferred embodiment, the construction of the cluster hierarchical power transmission and transformation project cost prediction model is specifically as follows: based on the n power transmission and transformation project feature clusters, n independent sub-neural network models are constructed; the feature factor data in the n power transmission and transformation project feature clusters are input into the k independent sub-neural network models for feature extraction to obtain n cluster power transmission and transformation project feature matrices; a self-attention mechanism is introduced to assign corresponding cluster weights to the n cluster power transmission and transformation project feature matrices to obtain n cluster power transmission and transformation project feature matrices after the corresponding weights are assigned; the n cluster power transmission and transformation project feature matrices after the corresponding weights are assigned are weightedly fused to obtain a fused cluster power transmission and transformation project feature matrix; The fused cluster power transmission and transformation project feature matrix is input into a preset LSTM prediction model to perform power transmission and transformation project cost prediction.
[0010] In a preferred embodiment, the self-attention mechanism is introduced to assign corresponding cluster weights to the n-cluster power transmission and transformation engineering feature matrices to obtain the n-cluster power transmission and transformation engineering feature matrices after the corresponding weights are assigned, specifically: calculating the attention scores between the n-cluster power transmission and transformation engineering feature matrices; normalizing the attention scores to obtain the cluster weights corresponding to the n-cluster power transmission and transformation engineering feature matrices.
[0011] The technical effects and advantages of the power transmission and transformation project cost review system based on big data of the present invention are as follows: 1. The present invention adopts the grey correlation analysis method to calculate the grey correlation between the characteristic factor data of the power transmission and transformation project and the cost of the power transmission and transformation project, and divides the various characteristic factors into k power transmission and transformation project characteristic clusters according to the corresponding grey correlation based on the k-means clustering algorithm. First, the grey correlation analysis can effectively identify and quantify the influence of characteristic factors on the cost, which is particularly suitable for situations with incomplete data or small samples, and helps to find key influencing factors; secondly, by dividing the features with similar correlation into clusters through k-means clustering, the data structure can be simplified, redundant information can be reduced, and the factors in different feature clusters can be more relevant, which is conducive to the refined processing of the model. Finally, this combined method helps to improve the efficiency of feature extraction and the accuracy of cost prediction in complex power transmission and transformation projects, and reduce the complexity of the model, while retaining the core features that have a greater impact on the cost. 2. The present invention constructs a cluster-level power transmission and transformation project cost prediction model. First, by clustering the features, the features with strong correlation are classified into the same cluster, so that the model can process different feature clusters separately, thereby enhancing the ability to capture complex feature relationships. Secondly, hierarchical modeling can better handle the nonlinear relationship between features, avoid feature interference, and improve the accuracy and robustness of the prediction. Each cluster uses an independent sub-model to extract features and then fuses them, which can make the model more flexible and able to dynamically respond to multi-dimensional factors in complex projects. Finally, the cluster hierarchical structure reduces the complexity and computational overhead of the model, helps to improve the efficiency and adaptability of the model, and is suitable for large-scale and complex power transmission and transformation project cost prediction tasks. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 It is a schematic diagram of the structure of the power transmission and transformation project cost review system based on big data 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 present invention provides a power transmission and transformation project cost review system based on big data. It includes data acquisition module, cost prediction module and cost review module. There are connections between the modules: A data acquisition module is used to acquire engineering data of power transmission and transformation engineering projects and data of power transmission and transformation engineering characteristic factors and perform preprocessing; In this example, the engineering data of the power transmission and transformation project and the power transmission and transformation project characteristic factor data are specifically: The engineering data of power transmission and transformation projects include basic project data, project progress data, project cost data, project quality data and external factors of the project. The basic project data mainly include project name, project type, project scale, voltage level, etc.; the project progress data mainly include the project start time and end time; the project cost data mainly include the initial budget material cost, labor cost, equipment cost and operation and maintenance cost; the project quality data mainly include various inspection reports and equipment failure rate during the construction process; the external factors of the project mainly include geographical location and environmental data.
[0015] The data of characteristic factors of power transmission and transformation projects include technical characteristic factor data, project scale data, and geographical and environmental factor data. The technical characteristic factor data mainly include voltage level, line type, equipment type, and substation type; the project scale data mainly include line length, substation capacity, and the number of line tower bases; the geographical and environmental factor data mainly include terrain conditions, climate conditions, and soil conditions.
[0016] It should be noted that in the technical characteristic factor data, voltage level: different voltage levels (such as 35kV, 110kV, 220kV, 500kV, 750kV) have a greater impact on the project cost. Generally, the higher the voltage level, the higher the cost; line type: overhead lines, underground cable lines, submarine cable lines, etc., the cost of overhead lines is generally lower, and the cost of underground and submarine cable lines is higher; equipment type: the cost of equipment from different manufacturers and models is different. For example, the prices of transformers, switchgear, lightning arresters, etc. of different brands will vary; substation type: GIS (gas insulated substation), AIS (air insulated substation). GIS substation is suitable for cities because of its compact equipment, and its cost is generally higher than AIS.
[0017] In the project scale data, line length: the length of the transmission line directly affects the material and construction costs. The longer the line, the higher the cost; substation capacity: the total capacity of the substation, usually in MVA. The larger the capacity, the higher the cost; number of line tower bases: the number of transmission towers affects the material and construction costs, especially for high-voltage lines. The height of the tower base and the complexity of the structure also have a great impact on the cost.
[0018] A feature engineering module, for calculating the grey correlation between the characteristic factor data of the power transmission and transformation project and the cost of the power transmission and transformation project by using a grey correlation analysis method, and dividing the characteristic factors into k power transmission and transformation project characteristic clusters according to the corresponding grey correlation based on a k-means clustering algorithm; In this example, the grey correlation analysis method is used to calculate the grey correlation between the characteristic factor data of the power transmission and transformation project and the cost of the power transmission and transformation project, specifically: The power transmission and transformation engineering characteristic factor data is standardized by using a Min-Max normalization method to obtain standardized power transmission and transformation engineering characteristic factor data; The standardized power transmission and transformation project characteristic factor data is used to calculate the grey correlation degree between the power transmission and transformation project characteristic factor data and the power transmission and transformation project cost by using a grey correlation analysis formula.
[0019] In this example, the standardized power transmission and transformation engineering characteristic factor data is used to calculate the gray correlation degree of each feature using the gray correlation analysis formula, specifically:
[0020]
[0021] in, is the grey correlation coefficient of the jth feature in the i-th power transmission and transformation project, is the normalized value of the target variable, is the normalized value of the jth power transmission and transformation project feature, is the resolution coefficient, m is the total number of power transmission and transformation projects, is the grey relational degree of the jth power transmission and transformation project feature.
[0022] It should be noted that the gray correlation analysis formula is used to calculate the gray correlation degree of each characteristic of the standardized power transmission and transformation project characteristic factor data, mainly to accurately identify and quantify the impact of each characteristic factor on the project cost. Power transmission and transformation projects involve a variety of complex factors, such as technical parameters, geographical environment, construction methods, etc., and their contribution to the project cost is nonlinear and uncertain. Grey correlation analysis is an analysis method suitable for small samples and incomplete information systems. It can reveal the relative importance of different characteristics to the cost by calculating the correlation between characteristic factors and cost.
[0023] The purpose of standardizing data is to eliminate the differences in the dimensions and magnitudes of each feature, so that they can be compared on the same scale. The calculation of grey correlation can comprehensively consider the trend and magnitude of feature changes, thereby identifying the key factors that have a greater impact on the cost. In this way, grey correlation analysis provides a strong quantitative basis for project cost prediction, which helps to improve the prediction accuracy of the model and reduce cost risks.
[0024] In this example, the k-means clustering algorithm is used to divide the power transmission and transformation project characteristic factor data into k power transmission and transformation project characteristic clusters according to the corresponding grey correlation degree, specifically: Set the initial k value, and select k grey correlation degrees from the grey correlation degrees corresponding to the characteristic factor data of each power transmission and transformation project as the initial power transmission and transformation project characteristic cluster center; Calculate the Euclidean distance between the grey correlation degree corresponding to the characteristic factor data of each power transmission and transformation project and the center of the initial power transmission and transformation project characteristic cluster; Based on the Euclidean distance, the power transmission and transformation project characteristic clusters are allocated and the centers of the power transmission and transformation project characteristic clusters are updated; Repeat the steps of allocating power transmission and transformation engineering feature clusters and updating the power transmission and transformation engineering feature cluster centers to obtain k power transmission and transformation engineering feature clusters.
[0025] In this example, the initial k value is set, and k grey correlation degrees are selected from the grey correlation degrees corresponding to the characteristic factor data of each power transmission and transformation project as the initial power transmission and transformation project characteristic cluster center, specifically: The performance and results of the K-means clustering algorithm are highly dependent on the selection of the initial cluster center. The initial position of the cluster center affects the formation and convergence results of each cluster. If the initial cluster center is not selected properly, the algorithm may fall into a local optimal solution and produce unreasonable clustering results. By using features with physical or practical significance as the initial cluster center, the clustering process can be better guided and converged to a more appropriate cluster division.
[0026] Grey correlation is an indicator to measure the degree of correlation between each characteristic factor and the project cost. The larger the value, the greater the impact of the characteristic on the target. By selecting characteristic factors with higher grey correlation as the initial cluster center, it can ensure that those characteristics that have a greater impact on cost prediction are given priority, thereby improving the rationality of the clustering results.
[0027] In this example, the calculation of the Euclidean distance between the grey correlation degree corresponding to each power transmission and transformation project characteristic factor data and the center of the initial power transmission and transformation project characteristic cluster is specifically as follows:
[0028] in, is the Euclidean distance between the grey correlation degree corresponding to each characteristic factor and the center of the initial power transmission and transformation engineering characteristic cluster, is the characteristic factor of the jth power transmission and transformation project, The center of the ith cluster.
[0029] A model building module, used to perform dimensionality reduction processing on the k power transmission and transformation project feature clusters using a principal component analysis method, and to build a cluster-level power transmission and transformation project cost prediction model; In this example, the k power transmission and transformation engineering feature clusters are subjected to dimensionality reduction processing using the principal component analysis method, specifically: Converting the intra-cluster characteristic factors in the k power transmission and transformation engineering characteristic clusters into m power transmission and transformation engineering characteristic matrices; Based on the m power transmission and transformation engineering characteristic matrices, a principal component analysis method is used to perform dimensionality reduction processing.
[0030] In this example, the principal component analysis method is specifically: Calculating the covariance matrix of the k power transmission and transformation engineering feature clusters and the eigenvalues and eigenvectors of the covariance matrix; Based on the eigenvalues, the eigenvectors are sorted in descending order, and the first n eigenvectors are selected to convert the k power transmission and transformation engineering feature clusters into principal component space, and perform dimensionality reduction to obtain n power transmission and transformation engineering feature clusters.
[0031] In this example, the covariance matrix of the k power transmission and transformation engineering feature clusters and the eigenvalues and eigenvectors of the covariance matrix are calculated, specifically:
[0032] in is the power transmission and transformation engineering feature matrix of the i-th power transmission and transformation engineering feature cluster, is the covariance matrix of the power transmission and transformation engineering feature cluster, with a dimension of mxm, where m is the number of characteristic factors in the power transmission and transformation engineering feature cluster, is the eigenvector corresponding to the covariance matrix of the power transmission and transformation engineering feature cluster, and n is the number of power transmission and transformation engineering feature clusters.
[0033] It should be noted that the principal component analysis (PCA) method is used to reduce the dimension of k power transmission and transformation project feature clusters, mainly to reduce redundant information, simplify the model structure, and improve the model's operating efficiency and prediction accuracy. In the power transmission and transformation project feature cluster, there may be high correlation or collinearity between features, resulting in information redundancy. These redundant features not only unnecessarily increase the complexity of the model, but may also generate noise and interference in the cost prediction.
[0034] By reducing the dimensionality of feature clusters through PCA, multiple related features can be converted into a few independent principal components with the largest explained variance, retaining the main information in the data. PCA compresses high-dimensional data into low-dimensional data, while retaining as much important information in the original features that explains the cost changes as possible. Doing so can not only reduce the amount of calculation and improve the efficiency of the model, but also reduce the risk of overfitting the model.
[0035] In addition, PCA can help identify and focus on the core features that contribute most to cost prediction, improve the model's ability to explain actual costs, and further improve the accuracy and reliability of cost prediction. Therefore, using PCA for dimensionality reduction is an effective optimization strategy.
[0036] In this example, the construction of a cluster-level power transmission and transformation project cost prediction model is specifically as follows: Based on the n power transmission and transformation engineering feature clusters, constructing n independent sub-neural network models; Inputting the characteristic factor data in the n power transmission and transformation engineering characteristic clusters into the k independent sub-neural network models for feature extraction to obtain the n cluster power transmission and transformation engineering characteristic matrix; A self-attention mechanism is introduced to assign corresponding cluster weights to the n-cluster power transmission and transformation engineering feature matrices, so as to obtain the n-cluster power transmission and transformation engineering feature matrices after the corresponding weights are assigned; Performing weighted fusion on the n cluster power transmission and transformation engineering feature matrices after the corresponding weights are assigned to obtain a fused cluster power transmission and transformation engineering feature matrix; The fused cluster power transmission and transformation project feature matrix is input into a preset LSTM prediction model to perform power transmission and transformation project cost prediction.
[0037] In this example, the self-attention mechanism is introduced to assign corresponding cluster weights to the n-cluster power transmission and transformation engineering feature matrices, and the n-cluster power transmission and transformation engineering feature matrices after the corresponding weights are assigned are obtained, specifically: Calculating the attention scores between the n cluster power transmission and transformation engineering feature matrices; The attention scores are normalized to obtain cluster weights corresponding to the n cluster power transmission and transformation engineering feature matrices.
[0038] In this example, based on the n power transmission and transformation engineering feature clusters, n independent sub-neural network models are constructed, specifically: Based on n power transmission and transformation project feature clusters, n independent sub-neural network models are constructed in order to perform more refined feature extraction and processing based on the characteristics of different feature clusters. Each feature cluster contains similar characteristic factors. Through independent sub-neural network processing, the unique impact of the features in each cluster on the power transmission and transformation project cost can be fully explored to avoid interference and redundancy between features. In addition, the sub-neural network can learn different patterns for the features of each cluster, so that the model can better capture complex nonlinear relationships, thereby improving the accuracy and robustness of the overall cost prediction. This hierarchical processing method can optimize the neural network structure and improve prediction efficiency.
[0039] In this example, the attention scores between the n cluster power transmission and transformation engineering feature matrices are calculated, specifically:
[0040] in, For power transmission and transformation engineering cluster Power Transmission and Transformation Engineering Cluster The attention weight, For power transmission and transformation engineering cluster The query vector is For power transmission and transformation engineering cluster The key value of is the scaling factor.
[0041] It should be noted that the self-attention mechanism is introduced to assign weights to the n-cluster power transmission and transformation project feature matrix in order to dynamically capture the importance of different feature clusters to cost prediction. The features in different clusters have an uneven impact on the cost of power transmission and transformation projects. By calculating the importance weight of each feature cluster, the attention mechanism can automatically focus on clusters that contribute more to the prediction results, while weakening unimportant cluster information. Compared with averaging all clusters, the attention mechanism gives the model flexibility, allowing it to perform weighted combinations based on the importance of different feature clusters, significantly improving the accuracy and generalization of predictions. In addition, it can dynamically adjust the weight distribution between feature clusters to adapt to the cost structure of different engineering projects and enhance the applicability and robustness of the model.
[0042] The cost forecast review module inputs the engineering data of the power transmission and transformation project into the cluster-level power transmission and transformation project cost forecast model to obtain the power transmission and transformation project cost forecast value, and compares the power transmission and transformation project cost forecast value with a preset cost threshold to conduct a cost review warning. In this example, the engineering data of the power transmission and transformation project is input into the cluster-level power transmission and transformation project cost prediction model to obtain the power transmission and transformation project cost prediction value, and the power transmission and transformation project cost prediction value is compared with the preset cost threshold value to conduct cost review and early warning, specifically: Dividing the engineering data of the power transmission and transformation project into a training set and a test set according to a preset ratio, and inputting the training set into the cluster-level power transmission and transformation project cost prediction model to train the model; The trained model is tested using the test set to obtain a trained cluster-level power transmission and transformation project cost prediction model; The trained cluster hierarchical power transmission and transformation project cost prediction model is used to predict the power transmission and transformation project cost.
[0043] When the cost forecast value of the power transmission and transformation project exceeds the preset cost threshold, an overspending warning is triggered; When the predicted cost of a power transmission and transformation project is less than the preset cost threshold, an underestimation warning is triggered.
[0044] It should be noted that the engineering data of the power transmission and transformation project is input into the cluster hierarchical cost prediction model, and the predicted value is compared with the preset cost threshold to conduct cost review and early warning. The main purpose is to effectively control the engineering cost and reduce the risk of project over-budget. The cluster hierarchical model can fully explore the inherent structure of the feature data through cluster analysis and principal component dimensionality reduction processing, thereby improving the prediction accuracy. After the engineering data is input into the model, a more accurate cost prediction value can be obtained.
[0045] By comparing the predicted value with the preset cost threshold (such as the budget cap or the safety range set by historical cost data), possible cost overruns can be discovered in a timely manner. Once the predicted value exceeds the threshold, the system will trigger an early warning to remind managers to review and adjust the project costs to avoid major financial risks caused by cost out of control in the later stage. This process can help project managers take corrective measures in advance, optimize engineering plans, achieve more scientific cost management and risk control, and improve the economic benefits of the project.
[0046] 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.
[0047] 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.
[0048] 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.
[0049] 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.
[0050] 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.
[0051] 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. The power transmission and transformation project cost review system based on big data is characterized by: It includes data acquisition module, cost prediction module and cost review module, and there are connections between the modules: A data acquisition module is used to acquire engineering data of power transmission and transformation engineering projects and data of power transmission and transformation engineering characteristic factors and perform preprocessing; A feature engineering module, for calculating the grey correlation between the characteristic factor data of the power transmission and transformation project and the cost of the power transmission and transformation project by using a grey correlation analysis method, and dividing the characteristic factor data of the power transmission and transformation project into k power transmission and transformation project feature clusters according to the corresponding grey correlation based on a k-means clustering algorithm; A model building module, used to perform dimensionality reduction processing on the k power transmission and transformation project feature clusters using a principal component analysis method, and to build a cluster-level power transmission and transformation project cost prediction model; The cost forecast review module inputs the engineering data of the power transmission and transformation project into the cluster-level power transmission and transformation project cost forecast model to obtain the power transmission and transformation project cost forecast value, and compares the power transmission and transformation project cost forecast value with a preset cost threshold to conduct a cost review warning.
2. The power transmission and transformation project cost review system based on big data according to claim 1 is characterized in that: The grey correlation analysis method is used to calculate the grey correlation between the characteristic factor data of the power transmission and transformation project and the cost of the power transmission and transformation project, specifically: The power transmission and transformation engineering characteristic factor data is standardized by using a Min-Max normalization method to obtain standardized power transmission and transformation engineering characteristic factor data; The standardized power transmission and transformation project characteristic factor data is used to calculate the grey correlation degree between the power transmission and transformation project characteristic factor data and the power transmission and transformation project cost by using a grey correlation analysis formula.
3. The power transmission and transformation project cost review system based on big data according to claim 2 is characterized in that: The k-means clustering algorithm is used to divide the power transmission and transformation project characteristic factor data into k power transmission and transformation project characteristic clusters according to the corresponding grey correlation degree, specifically: Set the initial k value, and select k grey correlation degrees from the grey correlation degrees corresponding to the characteristic factor data of each power transmission and transformation project as the initial power transmission and transformation project characteristic cluster center; Calculate the Euclidean distance between the grey correlation degree corresponding to the characteristic factor data of each power transmission and transformation project and the center of the initial power transmission and transformation project characteristic cluster; Based on the Euclidean distance, allocating power transmission and transformation engineering feature clusters and updating power transmission and transformation engineering feature cluster centers; Repeat the steps of allocating power transmission and transformation engineering feature clusters and updating the power transmission and transformation engineering feature cluster centers to obtain k power transmission and transformation engineering feature clusters.
4. The power transmission and transformation project cost review system based on big data according to claim 3 is characterized in that: The k power transmission and transformation engineering feature clusters are subjected to dimensionality reduction processing using the principal component analysis method, specifically: Converting the intra-cluster characteristic factors in the k power transmission and transformation engineering characteristic clusters into m power transmission and transformation engineering characteristic matrices; Based on the m power transmission and transformation engineering characteristic matrices, a principal component analysis method is used to perform dimensionality reduction processing.
5. The power transmission and transformation project cost review system based on big data according to claim 4 is characterized in that: The principal component analysis method is specifically: Calculating the covariance matrix of the k power transmission and transformation engineering feature clusters and the eigenvalues and eigenvectors of the covariance matrix; Based on the eigenvalues, the eigenvectors are sorted in descending order, and the first n eigenvectors are selected to convert the k power transmission and transformation engineering feature clusters into principal component space, and perform dimensionality reduction to obtain n power transmission and transformation engineering feature clusters.
6. The power transmission and transformation project cost review system based on big data according to claim 5 is characterized in that: The construction of the cluster-level power transmission and transformation project cost prediction model is specifically as follows: Based on the n power transmission and transformation engineering feature clusters, constructing n independent sub-neural network models; Inputting the characteristic factor data in the n power transmission and transformation engineering characteristic clusters into the k independent sub-neural network models for feature extraction to obtain the n cluster power transmission and transformation engineering characteristic matrix; A self-attention mechanism is introduced to assign corresponding cluster weights to the n-cluster power transmission and transformation engineering feature matrices, so as to obtain the n-cluster power transmission and transformation engineering feature matrices after the corresponding weights are assigned; Performing weighted fusion on the n cluster power transmission and transformation engineering feature matrices after the corresponding weights are assigned to obtain a fused cluster power transmission and transformation engineering feature matrix; The fused cluster power transmission and transformation project feature matrix is input into a preset LSTM prediction model to perform power transmission and transformation project cost prediction.
7. The power transmission and transformation project cost review system based on big data according to claim 6 is characterized in that: The self-attention mechanism is introduced to assign corresponding cluster weights to the n-cluster power transmission and transformation engineering feature matrices, and the n-cluster power transmission and transformation engineering feature matrices after the corresponding weights are assigned are obtained, specifically: Calculating the attention scores between the n cluster power transmission and transformation engineering feature matrices; The attention scores are normalized to obtain cluster weights corresponding to the n cluster power transmission and transformation engineering feature matrices.
8. The power transmission and transformation project cost review system based on big data according to claim 2 is characterized in that: The standardized power transmission and transformation engineering characteristic factor data is used to calculate the grey correlation degree of each feature using the grey correlation analysis formula, specifically: in, is the grey correlation coefficient of the jth feature in the i-th power transmission and transformation project, is the normalized value of the target variable, is the normalized value of the jth power transmission and transformation project feature, is the resolution coefficient, m is the total number of power transmission and transformation projects, is the grey relational degree of the jth power transmission and transformation project feature.
9. The power transmission and transformation project cost review system based on big data according to claim 5 is characterized in that: Calculate the covariance matrix of the k power transmission and transformation engineering feature clusters and the eigenvalues and eigenvectors of the covariance matrix, specifically: in is the power transmission and transformation engineering feature matrix of the i-th power transmission and transformation engineering feature cluster, is the covariance matrix of the power transmission and transformation engineering feature cluster, with a dimension of mxm, where m is the number of characteristic factors in the power transmission and transformation engineering feature cluster, is the eigenvector corresponding to the covariance matrix of the power transmission and transformation engineering feature cluster, and n is the number of power transmission and transformation engineering feature clusters.
10. The power transmission and transformation project cost review system based on big data according to claim 7 is characterized in that: Calculate the attention scores between the n cluster power transmission and transformation engineering feature matrices, specifically: in, For power transmission and transformation engineering cluster Power Transmission and Transformation Engineering Cluster The attention weight, For power transmission and transformation engineering cluster The query vector is For power transmission and transformation engineering cluster The key value of is the scaling factor.
Citation Information
Cited By
Construction cost data query method and system for constructional engineering
CN120013490A