Electric power engineering project optimization method and system based on full life cycle engineering cost

Abstract

The invention relates to the technical field of electric power engineering construction, and discloses an electric power engineering project optimization method and system based on full life cycle engineering cost. Training a project cost prediction model by analyzing the relationship among the cost, progress and quality in the historical project configuration data corresponding to the historical electric power project and the full life cycle project cost data, and further predicting the target prediction cost corresponding to the target electric power project and the target project configuration data; optimizing the target project configuration data through a genetic algorithm to generate a plurality of target project alternative configuration schemes according to an optimal configuration result; simulating each target project alternative configuration scheme by using a virtual reality technology to obtain a target simulation cost, comparing the target simulation cost with a target prediction cost, and selecting the corresponding target project alternative configuration scheme as a project optimization result of the target electric power project according to a comparison result to output; by optimizing the project configuration, the utilization rate of project resources is effectively improved, and the project management efficiency is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power engineering construction, and in particular to an electric power engineering project optimization method and system based on full life cycle engineering cost. Background Art

[0002] The construction of power engineering projects currently relies mainly on field visits, calculations and evaluations by professionals. However, human experience is easily affected by subjectivity and will ignore changes in the current situation and the input of new information, leading to wrong decisions. This makes resource allocation in power engineering construction unreasonable, such as waste of materials, equipment, manpower and other resources, thereby increasing project costs.

[0003] It can be seen that how to improve the resource utilization rate of existing power engineering construction projects has become a technical problem that technical personnel in this field need to solve urgently. Summary of the invention

[0004] The present invention provides a method and system for optimizing electric power engineering projects based on full life cycle engineering cost, which solves the problem of how to improve resource utilization of existing electric power engineering construction projects.

[0005] In order to solve the above technical problems, the first aspect of the present invention provides a method for optimizing a power engineering project based on the full life cycle engineering cost, comprising:

[0006] Obtain historical power projects and their corresponding historical project configuration data and full life cycle project cost data;

[0007] Mining and analyzing the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data, and using the analysis results as training samples to train the neural network model to obtain a project cost prediction model;

[0008] Inputting the target power project and its corresponding target project configuration data into the project cost prediction model for processing to obtain the target predicted cost;

[0009] Based on the target predicted cost, the target project configuration data is optimized by a genetic algorithm to obtain an optimized configuration result, and a plurality of target project candidate configuration data are generated according to the optimized configuration result;

[0010] The alternative configuration data of each target project is simulated by virtual reality technology to obtain the target simulation cost, and according to the relationship between the target predicted cost and each target simulation cost, the corresponding target project alternative configuration data is selected as the project optimization result output of the target power engineering project.

[0011] Furthermore, the acquisition of historical power projects and their corresponding historical project configuration data and full life cycle project cost data includes:

[0012] Extract several historical power projects and their corresponding historical project configuration data and full life cycle project cost data from the pre-built power project database through the AES encryption algorithm;

[0013] The historical project configuration data and the full life cycle engineering cost data are subjected to data cleaning and format conversion processing to obtain the historical project configuration data and the full life cycle engineering cost data in a unified data format.

[0014] Furthermore, the mining and analysis of the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data includes:

[0015] The historical project configuration data and the full life cycle engineering cost data are constructed into historical project time series data, and the historical project time series data are analyzed by a time series analysis algorithm to obtain historical project cost trend data;

[0016] Integrate the historical project cost trend data, the historical project configuration data and the full life cycle engineering cost data, and perform standardization on the integrated data to obtain normalized historical data;

[0017] The normalized historical data is mined and analyzed by an association rule mining algorithm to obtain a first correlation between the project cost and the project progress, and a second correlation between the project cost and the project quality, and the first correlation and the second correlation are output as the analysis results.

[0018] Furthermore, the analysis results are used as training samples to train the neural network model to obtain a project cost prediction model, including:

[0019] Performing feature screening on the first association relationship and the second association relationship by using a feature selection algorithm to obtain a first key feature and a second key feature;

[0020] Add corresponding cost tags to the first key feature and the second key feature respectively to construct training samples, and divide the training samples into a training set and a test set;

[0021] The first neural network model and the second neural network model are trained respectively by the training set, and the trained first neural network model and the trained second neural network model are used to predict the test set respectively to obtain a first prediction result and a second prediction result;

[0022] The evaluation index values ​​of the first prediction result and the second prediction result on the test set are quantified respectively, so as to output the neural network model whose evaluation index value is greater than a preset evaluation threshold as the project cost prediction model.

[0023] Furthermore, the feature selection algorithm is used to perform feature screening on the first association relationship and the second association relationship to obtain a first key feature and a second key feature, including:

[0024] Preliminarily screening the first association relationship and the second association relationship based on a filtering feature selection algorithm to remove features that are irrelevant to the first association relationship and the second association relationship;

[0025] A decision tree model is constructed by using an embedded feature selection algorithm, and the feature importance of the first association relationship that has been preliminarily screened and the feature importance of the second association relationship that has been preliminarily screened are respectively evaluated by using the decision tree model to obtain a first feature importance score and a second feature importance score;

[0026] According to the first feature importance score and the second feature importance score, the first association relationship that has been initially screened and the second association relationship that has been initially screened are respectively secondary screened to obtain a first key feature and a second key feature.

[0027] Furthermore, based on the target predicted cost, the target project configuration data is optimized by a genetic algorithm to obtain an optimized configuration result, including:

[0028] Taking the minimum target prediction cost as the goal, encoding the target project configuration data into individuals in the genetic algorithm, and initializing the population and algorithm parameters; wherein each individual represents a project configuration scheme;

[0029] Calculate the fitness value of each individual, perform selection, crossover and mutation operations on each individual according to the fitness value to construct a new population, and iteratively execute the fitness calculation step and the new population construction step through the new population until a preset number of iterations is reached, and output the current best individual as the optimization configuration result.

[0030] Furthermore, the generating of a plurality of target project candidate configuration data according to the optimization configuration result includes:

[0031] Analyze the optimization configuration results to obtain key influencing factors whose influence on the target predicted cost reaches a preset influence standard;

[0032] According to the constraint conditions corresponding to the target power project, the key influencing factors are adjusted based on the optimization configuration result to obtain first optimization configuration data;

[0033] Based on the constraint conditions, the key influencing factors are combined and random parameters are added to obtain combined influencing factors, and the combined influencing factors are adjusted according to the optimization configuration result to obtain second optimization configuration data;

[0034] The first optimized configuration data and the second optimized configuration data are combined to obtain a plurality of candidate configuration data for the target item.

[0035] Further, the selecting corresponding target project candidate configuration data as the project optimization result output of the target power project according to the relationship between the target predicted cost and each of the target simulation costs includes:

[0036] Quantifying the cost deviation between the target prediction cost and each of the target simulation costs, preliminarily screening the candidate configuration data of each of the target items based on the cost deviation, and using the preliminarily screened candidate configuration data of the target items as preliminary alternative solutions;

[0037] A comprehensive evaluation is performed on the preliminary alternative plans, and the preliminary alternative plans whose comprehensive evaluation results meet the preset evaluation standards are output as the project optimization results of the target power project.

[0038] Furthermore, the comprehensive evaluation of the preliminary alternative solutions and outputting the preliminary alternative solutions whose comprehensive evaluation results meet the preset evaluation criteria as the project optimization results of the target power project include:

[0039] Conduct sensitivity analysis on each of the preliminary alternatives to evaluate the degree of dependence of the key influencing factors on each of the preliminary alternatives, obtain multiple sensitivity scores, and use the preliminary alternatives whose sensitivity scores reach a preset sensitivity threshold as secondary alternatives;

[0040] The economic indicators of each of the secondary alternative plans are quantified to obtain a plurality of economic quantification results, and the secondary alternative plans whose economic quantification results reach a preset economic threshold are output as project optimization results of the target power project.

[0041] A second aspect of the present invention provides a power engineering project optimization system based on full life cycle engineering cost, comprising:

[0042] Data acquisition module, used to acquire historical power projects and their corresponding historical project configuration data and full life cycle project cost data;

[0043] A model building module is used to mine and analyze the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data, and use the analysis results as training samples to train the neural network model to obtain a project cost prediction model;

[0044] A cost prediction module, used for inputting the target power project and its corresponding target project configuration data into the project cost prediction model for processing to obtain a target predicted cost;

[0045] A solution generation module is used to optimize the target project configuration data through a genetic algorithm based on the target predicted cost to obtain an optimized configuration result, and generate a plurality of target project candidate configuration data according to the optimized configuration result;

[0046] The scheme optimization module is used to simulate the alternative configuration data of each target project through virtual reality technology to obtain the target simulation cost, and select the corresponding target project alternative configuration data as the project optimization result output of the target power project according to the relationship between the target predicted cost and each target simulation cost.

[0047] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0048] (1) By mining the correlation between historical project configuration data and full life cycle engineering cost data corresponding to historical power projects, we can more accurately understand the utilization efficiency of project resources at different stages and configurations;

[0049] (2) Based on the target prediction cost, the target project configuration data is optimized through genetic algorithms, ensuring the rational allocation and efficient use of resources during project execution and avoiding waste and redundancy of resources;

[0050] (3) The simulation function of virtual reality technology helps to identify potential resource waste scenarios, so that adjustments and optimizations can be made in advance, thereby reducing dependence on and consumption of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 It is a flow chart of a method for optimizing a power engineering project based on the full life cycle engineering cost provided by a certain embodiment of the present invention;

[0053] Figure 2 It is a structural diagram of a power engineering project optimization system based on full life cycle engineering cost provided by a certain embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. 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.

[0055] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0056] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two components. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the system or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0057] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0058] In one embodiment, if Figure 1As shown, the first aspect of the present invention provides a method for optimizing a power engineering project based on the full life cycle engineering cost, comprising:

[0059] S1. Obtain historical power projects and their corresponding historical project configuration data and full life cycle project cost data;

[0060] In one embodiment, step S1 includes:

[0061] Extract several historical power projects and their corresponding historical project configuration data and full life cycle project cost data from the pre-built power project database through the AES encryption algorithm;

[0062] The historical project configuration data and the full life cycle engineering cost data are subjected to data cleaning and format conversion processing to obtain the historical project configuration data and the full life cycle engineering cost data in a unified data format.

[0063] Specifically, the pre-built power engineering project database contains several historical power projects and is numbered by time. The historical project configuration data of each historical power project includes project length, material type, geographical conditions, environmental requirements, historical construction plans, equipment selection, construction schedule, construction phases, key construction nodes, material consumption and quality data, project risk factors, construction difficulties, quality requirements, etc. The full life cycle engineering cost data of each historical power project includes the engineering cost data and construction-related data of the project in various stages such as the project pre-stage, design stage, construction stage, completion stage, and operation and maintenance stage.

[0064] The early stage of a project is the starting point of a power engineering project, which mainly includes project approval, investment estimation, fund raising and bidding management. In this stage, it is necessary to prepare a project proposal and a feasibility study report, and then use the built-in estimation model to make an investment estimate to provide a basis for project approval.

[0065] The design phase focuses on cost control and quality management of project design. The power engineering project database supports online preparation, review and version control of design documents to ensure the orderly progress of design work.

[0066] The construction phase includes monitoring of costs, progress and quality during the construction process. The database also supports the preparation and adjustment of construction plans. By comparing the actual progress with the planned progress, progress deviations can be discovered and resolved in a timely manner. At the same time, various expenses during the construction process are recorded in real time and compared with the budget to help managers control costs through cost analysis reports. In addition, the database also provides a quality management system to ensure that the construction quality meets the standards through quality inspection and acceptance record functions;

[0067] The completion stage mainly includes project acceptance and final settlement. The database stores the acceptance data uploaded by users, and can also automatically generate an acceptance report based on it to ensure that the project quality meets the standards. In the final settlement stage, all expenses during the construction process are summarized and a final settlement report is automatically generated to provide a basis for the payment of the project funds.

[0068] The operation and maintenance stage is mainly about the cost and efficiency of power project operation and maintenance. The database stores equipment management, maintenance plans, and fault reporting records to help users achieve refined management of operation and maintenance activities, thereby reducing operation and maintenance costs. At the same time, the database also supports the statistics and analysis of operation and maintenance costs, facilitating the full life cycle cost management of subsequent projects and the retrieval and processing of subsequent data.

[0069] In addition, the data in the database can be collected from project management systems, financial systems, construction management systems, etc., or recorded in real time, or scanned and processed paper documents and uploaded. The source and acquisition method are not specifically limited here. When storing data in the database, it is done by establishing data tables, indexes and views to ensure fast access and efficient management of data; the data storage process is represented as: create data tables → define data fields → store data records; data management includes regular data backup, data compression and data security measures to ensure data security and availability.

[0070] To retrieve historical data from the database, you first need to send a data request signal to the power engineering project database. The signal generally contains the year and type of the required data to facilitate data retrieval. After the database responds to the request, it will process the required data through the AES encryption algorithm, including:

[0071] Initialize the extended key: Assume that the 256-bit key K256 has been used to generate an extended key sequence through the key expansion algorithm;

[0072] Byte substitution enhancement: Use S' box to replace each byte of the plaintext block; assuming the first byte of the plaintext block is P0, after S' box replacement, S'0 is obtained;

[0073] Row shift optimization: Determine the row shift pattern based on a part of the key, such as certain bits of K256[0]; if the first bit of the key is 1, the first row is shifted to the left, otherwise it is shifted to the right;

[0074] Column confusion improvement: Use the new confusion matrix M' to confuse the state after row shifting; if the first column is C0, C4, C8, C12, then after M' confusion, new columns C'0, C'4, C'8, C'12 are obtained;

[0075] Round key plus extension: XOR the obfuscated state with the dynamically generated round key; the first round round key K'1 is dynamically generated based on K256 and the plaintext block;

[0076] Additional round key addition: After each round, an additional round key addition operation may be performed;

[0077] Final ciphertext generation: After all rounds, the final ciphertext block is obtained;

[0078] The encrypted ciphertext block is sent over. After receiving the ciphertext block sent by the database, it is processed by the AES encryption algorithm. The specific processing process includes:

[0079] Initialize the extended key: Use the same key K256 and key expansion algorithm as in the encryption process to generate an extended key sequence;

[0080] Reverse shift optimization: Assume that the last row of the ciphertext block C is shifted to the left by X bits during encryption. Then, during decryption, shift the row to the right by X bits.

[0081] Reverse byte substitution enhancement: Use the inverse mapping of the S' box to replace each byte of the ciphertext block; if the first byte of the ciphertext block is C0, then S'0 is obtained through the inverse mapping of the S' box;

[0082] Improvement of reverse column confusion: Use the inverse matrix of the confusion matrix M' to perform reverse column confusion on the state after row shift; if the first column of the ciphertext block is C'0, C'4, C'8, C'12, then after reverse column confusion, new columns C0, C4, C8, C12 are obtained;

[0083] Reverse round key plus extension: XOR the state after reverse column obfuscation with the inverse sequence of the last round key used in the encryption process; if the round key used in the last round is K'10, then the inverse of K'10 is used for XOR operation during decryption;

[0084] Additional reverse round key addition: If there are additional round key addition operations during encryption, these operations need to be performed in reverse order during decryption;

[0085] Final plaintext recovery: After all reverse rounds, the original plaintext block is finally restored, which is the required historical data.

[0086] Then, the acquired historical project configuration data and full life cycle engineering cost data will be cleaned, such as removing duplicate data, correcting erroneous data, filling in missing data, etc., and then using automated scripts to perform preliminary screening and format unification on them, so as to achieve pre-processing operations on these historical data, so as to improve the integrity and accuracy of the data, improve the data quality, and provide a reliable data foundation for subsequent data mining and decision support.

[0087] S2, mining and analyzing the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data, and using the analysis results as training samples to train the neural network model to obtain a project cost prediction model;

[0088] In one embodiment, the mining and analysis of the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data includes:

[0089] The historical project configuration data and the full life cycle engineering cost data are constructed into historical project time series data, and the historical project time series data are analyzed by a time series analysis algorithm to obtain historical project cost trend data;

[0090] Integrate the historical project cost trend data, the historical project configuration data and the full life cycle engineering cost data, and perform standardization on the integrated data to obtain normalized historical data;

[0091] The normalized historical data is mined and analyzed by an association rule mining algorithm to obtain a first correlation between the project cost and the project progress, and a second correlation between the project cost and the project quality, and the first correlation and the second correlation are output as the analysis results.

[0092] Specifically, the present invention adds timestamps to the pre-processed historical project configuration data and the full life cycle engineering cost data and arranges them in chronological order to form time series data to fuse them and obtain the historical project time series data; then the ARIMA model is used to perform trend analysis on the historical project time series data to obtain the historical project cost trend data, which includes the cost change trend and fluctuation over time, etc., and these analysis results are interpreted to reveal the law and reason of cost change; then the historical project cost trend data, historical project configuration data and full life cycle engineering cost data are integrated to form a complete data set. Note that during the integration process, care should be taken to preserve Maintain the consistency and integrity of the data, and standardize the integrated data, such as minimum-maximum scaling, z-score standardization or root mean square standardization. Standardization helps to eliminate dimensional differences between data and improve data comparability. Finally, the Apriori algorithm is applied to the standardized data for association rule mining. By setting algorithm parameters, such as support and confidence, the quantity and quality of the mining results are controlled. The output includes the first correlation between project cost and project progress, and the second correlation between project cost and project quality. The association rule mining results are output to reveal the correlation and influence mechanism between historical project costs and other factors.

[0093] The present invention deeply mines the correlation and regularity between historical data through time series analysis algorithm and association rule mining algorithm, which helps to improve the accuracy and reliability of data analysis and provide more powerful support for decision-making; by analyzing the correlation between project cost and project progress and project quality, the key factors affecting project cost can be identified, which helps to optimize project management and resource allocation, reduce project cost and improve the utilization rate of project resources.

[0094] In one embodiment, the step of using the analysis results as training samples to train a neural network model to obtain a project cost prediction model includes:

[0095] Performing feature screening on the first association relationship and the second association relationship by using a feature selection algorithm to obtain a first key feature and a second key feature;

[0096] Add corresponding cost tags to the first key feature and the second key feature respectively to construct training samples, and divide the training samples into a training set and a test set;

[0097] The first neural network model and the second neural network model are trained respectively by the training set, and the trained first neural network model and the trained second neural network model are used to predict the test set respectively to obtain a first prediction result and a second prediction result;

[0098] The evaluation index values ​​of the first prediction result and the second prediction result on the test set are quantified respectively, so as to output the neural network model whose evaluation index value is greater than a preset evaluation threshold as the project cost prediction model.

[0099] Specifically, the present invention uses a feature selection algorithm to extract key features from the association relationship, and adds corresponding cost tags to the first key feature and the second key feature respectively, that is, the actual project cost data for supervised learning, and combines the key features and the cost tags into training samples, so that each training sample should contain a set of key features and their corresponding cost tags; then randomly divide them into training sets and test sets, and use the training sets to train the first and second neural network models; wherein the first and second neural network models can be multi-layer perceptrons (MLP), convolutional neural networks (CNN), recurrent neural networks (RNN), etc., and the specific training process is: initialize the weights and biases of the neural network, set the hyperparameters of the model, such as learning rate, The number of iterations, batch size, etc., use the training set to train the first and second neural network models respectively, calculate the output through forward propagation, update the weights and biases through back propagation, and repeat the training process until the preset number of iterations is reached or the model performance no longer improves; then use the trained first and second neural network models to predict the test set to obtain the first and second prediction results, and quantify the evaluation index values ​​of the first and second prediction results on the test set, and select the neural network model with an evaluation index value greater than the preset evaluation threshold as the output of the project cost prediction model; among them, commonly used evaluation indicators include mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), accuracy, recall rate, F1 score, etc.

[0100] By training the neural network model, the present invention can learn the complex relationship between key features and costs, which helps to improve the accuracy of project cost prediction; dividing the training samples into a training set and a test set, and using the test set to evaluate the model performance can enhance the generalization ability of the model, which helps the model to better adapt to new data in practical applications; comparing the evaluation index values ​​of the two models, it is possible to select a model with better performance as the project cost prediction model, which helps to ensure the effectiveness and reliability of the model in practical applications.

[0101] The step of performing feature screening on the first association relationship and the second association relationship by using a feature selection algorithm to obtain a first key feature and a second key feature includes:

[0102] Preliminarily screening the first association relationship and the second association relationship based on a filtering feature selection algorithm to remove features that are irrelevant to the first association relationship and the second association relationship;

[0103] A decision tree model is constructed by using an embedded feature selection algorithm, and the feature importance of the first association relationship that has been preliminarily screened and the feature importance of the second association relationship that has been preliminarily screened are respectively evaluated by using the decision tree model to obtain a first feature importance score and a second feature importance score;

[0104] According to the first feature importance score and the second feature importance score, the first association relationship that has been initially screened and the second association relationship that has been initially screened are respectively secondary screened to obtain a first key feature and a second key feature.

[0105] Specifically, the present invention constructs the first association relationship and the second association relationship into a first data set including a feature variable (i.e., project progress) and a target variable (i.e., project cost) and a second data set including a feature variable (i.e., project quality) and a target variable (i.e., project cost); and uses a filtering feature selection algorithm, such as a chi-square test, a correlation coefficient, mutual information, etc., to sort the features in the first and second data sets respectively, and removes the features that are irrelevant or weakly correlated with the target variable according to the sorting results, so that the first and second data sets that have been preliminarily screened can be output; then, using the preliminarily screened data sets, respectively construct decision tree models for the first association relationship and the second association relationship, so that they should be able to reflect the relationship between the feature variable and the target variable, and obtain the first and second decision tree models to evaluate the feature importance of the first and second data sets that have been preliminarily screened, and obtain the first and second feature importance scores; wherein, the feature importance can usually be measured by indicators such as the frequency of use of the feature in the model, information gain, and Gini index; finally, according to the feature importance scores, the first and second data sets that have been preliminarily screened are secondary screened, and the data whose importance scores in the two data sets reach the preset importance threshold are selected as key features, and the first key feature and the second key feature are obtained.

[0106] The present invention can remove irrelevant or redundant features through the combination of filtered feature selection and embedded feature selection, improve the performance and accuracy of the model, and help to build a more concise and effective decision tree model; the key features obtained by multiple screenings can provide more direct and effective information support for project decisions, help improve decision-making efficiency, and thus improve the utilization of project resources.

[0107] S3. Input the target power project and its corresponding target project configuration data into the project cost prediction model for processing to obtain the target predicted cost; wherein, the target power project and its corresponding target project configuration data can be solution data obtained by professionals based on experience, or can be obtained by other means. The specific source and construction process are not the focus of this application and will not be described in detail here. However, it should be noted that before inputting into the project cost prediction model, it is necessary to perform preprocessing like historical data before cost prediction can be performed.

[0108] S4. Based on the target predicted cost, the target project configuration data is optimized by a genetic algorithm to obtain an optimized configuration result, and a plurality of target project candidate configuration data are generated according to the optimized configuration result;

[0109] In one embodiment, the target project configuration data is optimized by a genetic algorithm based on the target predicted cost to obtain an optimized configuration result, including:

[0110] Taking the minimum target prediction cost as the goal, encoding the target project configuration data into individuals in the genetic algorithm, and initializing the population and algorithm parameters; wherein each individual represents a project configuration scheme;

[0111] Calculate the fitness value of each individual, perform selection, crossover and mutation operations on each individual according to the fitness value to construct a new population, and iteratively execute the fitness calculation step and the new population construction step through the new population until a preset number of iterations is reached, and output the current best individual as the optimization configuration result.

[0112] Specifically, the present invention encodes the target project configuration data into individuals in a genetic algorithm, and each individual represents a project configuration scheme, randomly generates a group of individuals as an initial population, and sets parameters for the genetic algorithm, such as population size, crossover probability, mutation probability, preset number of iterations, etc.; then calculates the fitness value of each individual, and the fitness function comprehensively considers factors such as cost, progress, quality, etc., and performs weighted calculation on them; then selects, crosses, and mutates the individuals according to the fitness value; among them, commonly used selection methods include roulette selection, tournament selection, etc., to ensure that excellent individuals have a greater probability of being selected; and crossover operations on selected individuals can generate new offspring individuals to exchange some genes of the individuals, thereby introducing new configuration schemes; mutation operations on offspring individuals can introduce new gene information, increase the diversity of the population, and randomly change some gene values ​​of the individuals; finally, the new population is used to iteratively execute the fitness calculation step and the new population construction step (selection, crossover, mutation), and the iterative process is repeated until the preset number of iterations is reached, and the final population and its corresponding optimal individual, that is, the optimal solution obtained by the genetic algorithm, are output as the optimization configuration result. The present invention utilizes the global search capability of the genetic algorithm to quickly find a better solution space area to improve the optimization efficiency.

[0113] In one embodiment, the generating of a plurality of target project candidate configuration data according to the optimization configuration result includes:

[0114] Analyze the optimization configuration results to obtain key influencing factors whose influence on the target prediction cost reaches a preset influence standard;

[0115] According to the constraint conditions corresponding to the target power project, the key influencing factors are adjusted based on the optimization configuration result to obtain first optimization configuration data;

[0116] Based on the constraint conditions, the key influencing factors are combined and random parameters are added to obtain combined influencing factors, and the combined influencing factors are adjusted according to the optimization configuration result to obtain second optimization configuration data;

[0117] The first optimized configuration data and the second optimized configuration data are combined to obtain a plurality of candidate configuration data for the target item.

[0118] Specifically, the present invention uses sensitivity analysis, correlation analysis and other methods to conduct in-depth analysis of the optimization configuration results, identify factors that have a significant impact on the target predicted cost, quantify the degree of influence of each factor on the target predicted cost, and then screen out the key influencing factors that have an impact on the target predicted cost to reach a preset impact standard. These key factors may be certain parameters, resource allocation, and technology selection in the project configuration; then sort out the constraints corresponding to the target power project, including reducing the total cost of the project, the project duration, and the project quality must meet national and industry standards, etc., and based on these constraints, fine-tune the identified key influencing factors. The control variable method can be used to further optimize the target predicted cost under the premise of meeting the constraints, and the adjusted key influencing factors are substituted into the optimization configuration result model to recalculate and obtain the first optimization configuration data.

[0119] The identified key influencing factors are then combined, and random parameters are added to simulate the uncertainty in actual situations to obtain combined influencing factors. The combined influencing factors are then adjusted based on the constraints to ensure that the adjusted scheme still meets the constraints, and are substituted into the optimization configuration result model to recalculate the second optimization configuration data. In this process, different key factors are combined to generate multiple alternative schemes. Each scheme should, while ensuring feasibility, try to maintain a certain difference from the optimal configuration. In order to increase the diversity of the scheme, some parameters can be randomly selected for change to generate some additional alternative schemes. However, it should be noted that random exploration should be carried out within a certain range to avoid generating completely unrealistic schemes.

[0120] Finally, the first optimized configuration data and the second optimized configuration data are combined to generate several target project alternative configuration data. The solutions formed by each alternative configuration data are adjusted based on different key influencing factors and combined influencing factors, but they all meet the constraints, so as to make full use of resources while saving costs. The present invention can further improve the precision and accuracy of optimization by deeply analyzing the optimization configuration results, identifying key influencing factors, and making targeted adjustments; by combining influencing factors and adding random parameters, it can simulate the uncertainty in actual situations, and the generated alternative solutions have stronger robustness and adaptability.

[0121] S5. Simulate the target project alternative configuration data by virtual reality technology to obtain the target simulation cost, and select the corresponding target project alternative configuration data as the project optimization result output of the target power project according to the relationship between the target predicted cost and each target simulation cost;

[0122] Specifically, the present invention simulates the solutions corresponding to the alternative configuration data of each target project through virtual reality technology to intuitively show the impact of different decisions on project cost and progress, and obtain the target simulation cost. The simulation process is:

[0123] The data mapping tool in the data integration interface provides mapping templates for the alternative configuration data of each target project to achieve standardized data import, and uses a data validator to check the integrity and accuracy of the imported data to ensure that no missing or erroneous data enters the simulation environment; the format converter of the data conversion engine is used to convert the format of the input data, such as converting DWG to FBX or OBJ format for use in the VR environment, and the converted data is compressed with a data compressor to optimize loading time and operating efficiency;

[0124] The parametric modeling tools in the 3D model builder are used to quickly build input parameters into standardized power engineering components, such as transmission towers and substations. Custom modeling is also provided, that is, advanced modeling tools are provided to allow users to create non-standard or specially designed components. Different levels of detail are set for different scenes and perspectives through the LOD setter to balance visual effects and performance requirements, and model details are dynamically adjusted according to user perspective and hardware performance to maintain a smooth experience. The interactive logic programmer is then used to write logic scripts for interactive elements, define their behavior and response in the VR environment, and configure the physical properties of mass and friction coefficient for interactive elements to achieve real physical interaction.

[0125] Use the terrain editor to edit and generate the corresponding target terrain to match the topography of the actual project site, and use the scene object placer to place buildings, equipment, vegetation objects, etc. in the scene; use the physical engine integrated block to define the gravity, wind force, collision physics rules, etc. in the simulated environment; then simulate the movement of mechanical equipment and the stress of the structure, such as crane hoisting operations; and simulate environmental dynamics, such as: simulating different weather conditions such as sunny days, rainy days, and snowy days, and their impact on construction; perform time control: simulate the alternation of day and night, affect light and shadows, and control the work efficiency of workers to simulate and build the overall environment;

[0126] Based on the alternative configuration data of each target project, the decision tab provides a series of corresponding and adjustable decision options, such as material selection, construction method, and staffing. The real-time feedback system immediately displays the impact of the decision on the project after the decision is made, and uses the cost calculator to calculate the direct and indirect costs of the project in real time according to the decision changes. The sum of the two is the total cost, which is the target simulation cost. The schedule planner simulates the impact of the decision on the project schedule and provides adjustment suggestions. Finally, the decision log records the time, content, and results of all simulated decisions, and provides decision version management and backtracking to facilitate comparison of different decision plans.

[0127] In addition, VR controller mapping maps the buttons, joysticks, and touchpad actions of the VR controller to specific interactive behaviors, and force feedback settings provide force feedback for specific operations to enhance immersion. The user interface design includes a menu system: design easy-to-navigate VR menus that provide tool selection and parameter adjustment functions; data panel: real-time display of key data and indicators such as cost, progress, and resource usage; voice command parser: recognizes user voice commands and performs corresponding operations for decision makers to use;

[0128] The simulation effect of the alternative configuration data of each target project, i.e., the project quality, cost, etc., can be visualized and analyzed through charts and color coding for intuitive display, and presented through reports. When the simulation results exceed the preset threshold, an early warning prompt is issued; and the simulation data is uploaded to the cloud server to achieve cross-platform sharing. In addition, annotation and discussion functions are provided to support team collaboration for decision makers. It can be seen that VR simulation not only provides a powerful tool for power engineering cost management, but also ensures that users can make efficient project decisions and risk management in a highly realistic and interactive environment, so as to facilitate subsequent solution optimization and save resources.

[0129] In one embodiment, selecting corresponding target project candidate configuration data as the project optimization result output of the target power project according to the relationship between the target predicted cost and each of the target simulation costs includes:

[0130] Quantifying the cost deviation between the target prediction cost and each of the target simulation costs, preliminarily screening the candidate configuration data of each of the target items based on the cost deviation, and using the preliminarily screened candidate configuration data of the target items as preliminary alternative solutions;

[0131] A comprehensive evaluation is performed on the preliminary alternative plans, and the preliminary alternative plans whose comprehensive evaluation results meet the preset evaluation standards are output as the project optimization results of the target power project.

[0132] Specifically, the present invention compares the target predicted cost with each target simulation cost, calculates the cost deviation, that is, the difference between the actual simulation cost and the target predicted cost, and sets a reasonable screening standard according to the size of the cost deviation, such as if the cost deviation rate does not exceed a certain percentage, performs a preliminary screening of the target project alternative configuration data, eliminates configurations with excessively large cost deviations, and retains configurations with smaller cost deviations as preliminary alternatives; determines comprehensive evaluation indicators according to the characteristics of the power engineering project, and sets preset evaluation standards according to the goals and requirements of the power engineering project, such as if the comprehensive evaluation score is not less than a certain score, and outputs the preliminary alternative plan whose comprehensive evaluation result meets the preset evaluation standard as the project optimization result of the target power engineering.

[0133] By quantifying cost deviations, the present invention can more accurately understand the cost situation of each alternative configuration data, thereby screening out a configuration plan with a more reasonable cost; the comprehensive evaluation process takes into account multiple factors, making project decisions more comprehensive and scientific; through preliminary screening and comprehensive evaluation, it can be ensured that the final output project optimization results have excellent performance in terms of cost control, market competitiveness, resource utilization, etc., thereby achieving optimal resource allocation.

[0134] In one embodiment, the comprehensive evaluation of the preliminary alternative solutions and outputting the preliminary alternative solutions whose comprehensive evaluation results meet the preset evaluation criteria as the project optimization results of the target power project include:

[0135] Conduct sensitivity analysis on each of the preliminary alternatives to evaluate the degree of dependence of the key influencing factors on each of the preliminary alternatives, obtain multiple sensitivity scores, and use the preliminary alternatives whose sensitivity scores reach a preset sensitivity threshold as secondary alternatives;

[0136] The economic indicators of each of the secondary alternative plans are quantified to obtain a plurality of economic quantification results, and the secondary alternative plans whose economic quantification results reach a preset economic threshold are output as project optimization results of the target power project.

[0137] Specifically, the present invention constructs a sensitivity analysis model for each preliminary alternative plan. The model takes key influencing factors as input variables and the target predicted cost of the preliminary alternative plan as output. By changing the value of the key influencing factors, the change in the target predicted cost of the preliminary alternative plan is observed, and then the model is used to calculate the sensitivity score of each key influencing factor to the preliminary alternative plan. The score can reflect the degree of influence of the change of the influencing factor on the project performance, and the preliminary alternative plan whose sensitivity score reaches the preset sensitivity threshold is used as the secondary alternative plan. Since these plans are highly dependent on the key influencing factors, they deserve special attention in the subsequent quantification of economic indicators.

[0138] Then, according to the project characteristics and needs, determine the economic indicators that need to be quantified, such as total cost, net present value, internal rate of return, etc., and for each secondary alternative plan, calculate the corresponding economic indicators according to its configuration data and the values ​​of key influencing factors, and obtain multiple economic quantification results, each result corresponds to the economic performance of a secondary alternative plan; the secondary alternative plan whose economic quantification results reach the preset economic threshold is used as the project configuration optimization plan for the target power project. These plans have excellent economic performance and meet the economic goals and requirements of the project. It should be noted that the preset values ​​involved in the plan can be determined according to actual needs, and no specific numerical restrictions are made.

[0139] Through sensitivity analysis, the present invention can clarify the impact of key influencing factors on project configuration alternatives, which is helpful to identify and avoid potential risks in advance, and provide strong support for the subsequent quantification of economic indicators. The quantification results of economic indicators can intuitively reflect the economic performance of each plan, reduce the economic risks in the project implementation process, and provide a scientific basis for decision makers; through screening and optimization, excellent performance plans can be selected, thereby achieving optimal allocation and efficient utilization of resources.

[0140] In the embodiment of the present application, based on the problem of how to improve the resource utilization rate of existing power engineering construction projects, a power engineering project optimization method based on the full life cycle engineering cost is designed, which realizes the acquisition of historical power engineering and its corresponding historical project configuration data and full life cycle engineering cost data; the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data is mined and analyzed, and the analysis results are used as training samples to train the neural network model to obtain a project cost prediction model; the target power engineering and its corresponding target project configuration data are input into the project cost prediction model for processing to obtain the target predicted cost; based on the target predicted cost, the target project configuration data is optimized by a genetic algorithm to obtain an optimized configuration result, and a number of target project alternative configuration data are generated according to the optimized configuration result; the target project alternative configuration data of each target project is simulated by virtual reality technology to obtain a target simulation cost, and according to the relationship between the target prediction cost and each target simulation cost, the corresponding target project alternative configuration data is selected as the technical solution for outputting the project optimization result of the target power engineering; by optimizing the project configuration, the utilization rate of the project resources is effectively improved, thereby improving the project management efficiency.

[0141] It should be noted that although the steps in the above flowchart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders.

[0142] In another embodiment, if Figure 2 As shown, the second aspect of the present invention provides a power engineering project optimization system based on the full life cycle engineering cost, comprising:

[0143] The data acquisition module 10 is used to acquire historical power engineering and its corresponding historical project configuration data and full life cycle engineering cost data;

[0144] A model building module 20 is used to mine and analyze the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data, and use the analysis results as training samples to train the neural network model to obtain a project cost prediction model;

[0145] The cost prediction module 30 is used to input the target power project and its corresponding target project configuration data into the project cost prediction model for processing to obtain a target predicted cost;

[0146] A solution generation module 40 is used to optimize the target project configuration data by a genetic algorithm based on the target predicted cost to obtain an optimized configuration result, and generate a plurality of target project candidate configuration data according to the optimized configuration result;

[0147] The scheme optimization module 50 is used to simulate the alternative configuration data of each target project through virtual reality technology to obtain the target simulation cost, and select the corresponding target project alternative configuration data as the project optimization result output of the target power project according to the relationship between the target predicted cost and each target simulation cost.

[0148] It should be noted that each module in the above-mentioned power engineering project optimization system based on the full life cycle engineering cost can be fully or partially implemented by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. For the specific definition of a power engineering project optimization system based on the full life cycle engineering cost, please refer to the definition of a power engineering project optimization method based on the full life cycle engineering cost above. The two have the same functions and effects, which will not be repeated here.

[0149] In summary, the present invention relates to the field of electric power engineering construction technology, and discloses an electric power engineering project optimization method and system based on full life cycle engineering cost. The method trains a project cost prediction model by analyzing the relationship between cost, progress and quality in historical project configuration data corresponding to historical electric power engineering and full life cycle engineering cost data, and then predicts the target predicted cost corresponding to the target electric power engineering and target project configuration data; optimizes the target project configuration data by a genetic algorithm to generate multiple target project alternative configuration schemes according to the optimized configuration results; simulates each target project alternative configuration scheme by using virtual reality technology to obtain a target simulation cost for comparison with the target predicted cost, and selects the corresponding target project alternative configuration scheme as the project optimization result output of the target electric power engineering according to the comparison result; and effectively improves the utilization rate of project resources by optimizing the project configuration, thereby improving the project management efficiency.

[0150] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0151] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. A method for optimizing power engineering projects based on the full life cycle engineering cost, characterized in that: include: Obtain historical power projects and their corresponding historical project configuration data and full life cycle project cost data; Mining and analyzing the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data, and using the analysis results as training samples to train the neural network model to obtain a project cost prediction model; Inputting the target power project and its corresponding target project configuration data into the project cost prediction model for processing to obtain the target predicted cost; Based on the target predicted cost, the target project configuration data is optimized by a genetic algorithm to obtain an optimized configuration result, and a plurality of target project candidate configuration data are generated according to the optimized configuration result; The alternative configuration data of each target project is simulated by virtual reality technology to obtain the target simulation cost, and according to the relationship between the target predicted cost and each target simulation cost, the corresponding target project alternative configuration data is selected as the project optimization result output of the target power engineering project.

2. The method for optimizing electric power engineering projects based on full life cycle engineering cost according to claim 1, characterized in that: The acquisition of historical power projects and their corresponding historical project configuration data and full life cycle project cost data includes: Extract several historical power projects and their corresponding historical project configuration data and full life cycle project cost data from the pre-built power project database through the AES encryption algorithm; The historical project configuration data and the full life cycle engineering cost data are subjected to data cleaning and format conversion processing to obtain the historical project configuration data and the full life cycle engineering cost data in a unified data format.

3. The method for optimizing electric power engineering projects based on full life cycle engineering cost according to claim 1 is characterized in that: The mining and analysis of the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data includes: The historical project configuration data and the full life cycle engineering cost data are constructed into historical project time series data, and the historical project time series data are analyzed by a time series analysis algorithm to obtain historical project cost trend data; Integrate the historical project cost trend data, the historical project configuration data and the full life cycle engineering cost data, and perform standardization on the integrated data to obtain normalized historical data; The normalized historical data is mined and analyzed by an association rule mining algorithm to obtain a first correlation between the project cost and the project progress, and a second correlation between the project cost and the project quality, and the first correlation and the second correlation are output as the analysis results.

4. The method for optimizing electric power engineering projects based on full life cycle engineering cost according to claim 3 is characterized in that: The analysis results are used as training samples to train the neural network model to obtain a project cost prediction model, including: Performing feature screening on the first association relationship and the second association relationship by using a feature selection algorithm to obtain a first key feature and a second key feature; Add corresponding cost tags to the first key feature and the second key feature respectively to construct training samples, and divide the training samples into a training set and a test set; The first neural network model and the second neural network model are trained respectively by the training set, and the trained first neural network model and the trained second neural network model are used to predict the test set respectively to obtain a first prediction result and a second prediction result; The evaluation index values ​​of the first prediction result and the second prediction result on the test set are quantified respectively, so as to output the neural network model whose evaluation index value is greater than a preset evaluation threshold as the project cost prediction model.

5. The method for optimizing electric power engineering projects based on full life cycle engineering cost according to claim 4 is characterized in that: The feature selection algorithm is used to perform feature screening on the first association relationship and the second association relationship to obtain the first key feature and the second key feature, including: Preliminarily screening the first association relationship and the second association relationship based on a filtering feature selection algorithm to remove features that are irrelevant to the first association relationship and the second association relationship; A decision tree model is constructed by using an embedded feature selection algorithm, and the feature importance of the first association relationship that has been preliminarily screened and the feature importance of the second association relationship that has been preliminarily screened are respectively evaluated by using the decision tree model to obtain a first feature importance score and a second feature importance score; According to the first feature importance score and the second feature importance score, the first association relationship that has been initially screened and the second association relationship that has been initially screened are respectively secondary screened to obtain a first key feature and a second key feature.

6. The method for optimizing electric power engineering projects based on full life cycle engineering cost according to claim 1, characterized in that: The step of optimizing the target project configuration data based on the target predicted cost by using a genetic algorithm to obtain an optimized configuration result includes: Taking the minimum target prediction cost as the goal, encoding the target project configuration data into individuals in the genetic algorithm, and initializing the population and algorithm parameters; wherein each individual represents a project configuration scheme; Calculate the fitness value of each individual, perform selection, crossover and mutation operations on each individual according to the fitness value to construct a new population, and iteratively execute the fitness calculation step and the new population construction step through the new population until a preset number of iterations is reached, and output the current best individual as the optimization configuration result.

7. The method for optimizing electric power engineering projects based on full life cycle engineering cost according to claim 1, characterized in that: The step of generating a plurality of target project candidate configuration data according to the optimization configuration result includes: Analyze the optimization configuration results to obtain key influencing factors whose influence on the target predicted cost reaches a preset influence standard; According to the constraint conditions corresponding to the target power project, the key influencing factors are adjusted based on the optimization configuration result to obtain first optimization configuration data; Based on the constraint conditions, the key influencing factors are combined and random parameters are added to obtain combined influencing factors, and the combined influencing factors are adjusted according to the optimization configuration result to obtain second optimization configuration data; The first optimized configuration data and the second optimized configuration data are combined to obtain a plurality of candidate configuration data for the target item.

8. The method for optimizing electric power engineering projects based on full life cycle engineering cost according to claim 7 is characterized in that: The step of selecting corresponding target project candidate configuration data as the project optimization result output of the target power project according to the relationship between the target predicted cost and each target simulation cost includes: Quantifying the cost deviation between the target prediction cost and each of the target simulation costs, preliminarily screening the candidate configuration data of each of the target items based on the cost deviation, and using the preliminarily screened candidate configuration data of the target items as preliminary alternative solutions; A comprehensive evaluation is performed on the preliminary alternative plans, and the preliminary alternative plans whose comprehensive evaluation results meet the preset evaluation standards are output as the project optimization results of the target power project.

9. A method for optimizing electric power engineering projects based on full life cycle engineering cost according to claim 8, characterized in that: The comprehensive evaluation of the preliminary alternatives and outputting the preliminary alternatives that meet the preset evaluation criteria as the project optimization results of the target power project include: Conduct sensitivity analysis on each of the preliminary alternatives to evaluate the degree of dependence of the key influencing factors on each of the preliminary alternatives, obtain multiple sensitivity scores, and use the preliminary alternatives whose sensitivity scores reach a preset sensitivity threshold as secondary alternatives; The economic indicators of each of the secondary alternative plans are quantified to obtain a plurality of economic quantification results, and the secondary alternative plans whose economic quantification results reach a preset economic threshold are output as project optimization results of the target power project.

10. A power engineering project optimization system based on the full life cycle engineering cost, characterized in that: include: Data acquisition module, used to acquire historical power projects and their corresponding historical project configuration data and full life cycle project cost data; A model building module is used to mine and analyze the correlation between the project cost, project progress and project quality in the historical project configuration data and the full life cycle engineering cost data, and use the analysis results as training samples to train the neural network model to obtain a project cost prediction model; A cost prediction module, used for inputting the target power project and its corresponding target project configuration data into the project cost prediction model for processing to obtain a target predicted cost; A solution generation module is used to optimize the target project configuration data through a genetic algorithm based on the target predicted cost to obtain an optimized configuration result, and generate a plurality of target project candidate configuration data according to the optimized configuration result; The scheme optimization module is used to simulate the alternative configuration data of each target project through virtual reality technology to obtain the target simulation cost, and select the corresponding target project alternative configuration data as the project optimization result output of the target power project according to the relationship between the target predicted cost and each target simulation cost.

Images