Multi-dimensional data model construction method and device for investment benefits of power grid construction project

Abstract

The invention relates to a power grid construction project investment benefit multi-dimensional data model construction method and device, and the method comprises the steps: obtaining the multi-dimensional project data of a power grid construction project, and carrying out the preprocessing of the project data, and obtaining the project processing data; performing feature extraction and conversion processing on the project processing data to obtain feature data; performing iteration processing on the feature data by adopting a Lagrange multiplier method and a convex optimization theory, and constructing a data model; acquiring historical power grid data, and optimizing the Lagrange multiplier of the data model by adopting cross validation and hyper-parameter tuning according to the historical power grid data to obtain an optimized data model. According to the method, multi-dimensional project data of a power grid construction project is obtained, the project data is subjected to preprocessing and feature extraction in sequence, then iteration processing is carried out by adopting a Lagrange multiplier method and a convex optimization theory to obtain a data model, and then model optimization is carried out on the data model to obtain an optimized data model; and outputting the optimal decision variable influencing the evaluation of the power grid construction project.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device and terminal equipment for constructing a multi-dimensional data model of investment benefits of power grid construction projects. Background Art

[0002] With the growth of global electricity demand and the promotion of renewable energy, the investment benefit evaluation of power grid construction projects has become more complex and critical. In the past, the evaluation of power grid construction projects mainly relied on statistical analysis and expert experience, but with the advancement of information technology, the traditional experience model of power grid construction projects can no longer meet the changing market demands and complex environmental conditions.

[0003] With the development of data mining technology. At first, simple regression analysis and basic classification algorithms were used to process investment benefit data in an attempt to understand the key factors that affect the success of construction projects. With the improvement of computing power and data availability, complex machine learning algorithms such as random forests and deep learning have begun to be applied, which can more accurately predict the impact of different investment decisions on benefits. Data mining technology has also gradually been combined with the specific needs of power grid construction projects to develop customized models suitable for practical applications. These models not only take into account economic factors, but also comprehensively consider multi-dimensional factors such as social benefits and environmental impacts, thereby helping power grid companies make more forward-looking and sustainable decisions. The development of data mining technology provides new solutions for power grid construction projects. By mining patterns and associations in large amounts of project data, it can more accurately predict investment benefits and optimize investment decisions.

[0004] Grid construction projects involve a large amount of capital investment, and their investment benefits are directly related to the economic benefits of power companies and the reliability of grid operation. However, the existing technical solutions have largely achieved the improvement of the investment benefits of grid construction projects, but there are still many technical defects, including at least the following prominent aspects: First, although some evaluation models try to integrate multiple factors, most existing evaluation tools still tend to analyze in a single dimension, which directly affects the accuracy and reliability of the evaluation model; second, the currently commonly used evaluation models use static data sets, which often ignore the dynamic changes in time series, which may lead to incomplete or insufficient timeliness of the evaluation results; third, the currently commonly used optimization algorithms such as genetic algorithms and ant colony algorithms, although they perform well in finding the global optimal solution, usually have high computational costs and may be inefficient for large-scale data processing. Summary of the invention

[0005] The present application provides a method, device and terminal equipment for constructing a multi-dimensional data model of the investment benefits of power grid construction projects, which is used to solve the technical problem that the existing power grid construction project evaluation model is constructed by collecting single-dimensional data and the model output results cannot be guaranteed.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] On the one hand, a method for constructing a multi-dimensional data model of investment benefits of power grid construction projects is provided, comprising the following steps:

[0008] Acquire multi-dimensional project data of a power grid construction project, and pre-process the project data to obtain project processing data;

[0009] Performing feature extraction and conversion processing on the project processing data to obtain feature data;

[0010] Iteratively process the characteristic data using Lagrange multiplier method and convex optimization theory to construct a data model for multi-dimensional optimization of investment benefits;

[0011] Historical power grid data is obtained, and Lagrange multipliers of the data model are optimized by cross-validation and hyperparameter tuning according to the historical power grid data to obtain an optimized data model.

[0012] Preferably, the characteristic data is iteratively processed using the Lagrange multiplier method and convex optimization theory to construct a data model for multi-dimensional optimization of investment benefits, including:

[0013] Obtaining the objective function and constraint condition data and necessary and sufficient condition data of the power grid construction project;

[0014] Constructing a Lagrangian function corresponding to the power grid construction project according to the characteristic data, the objective function and the constraint condition data;

[0015] Iterative processing is performed using the interior point method according to the necessary and sufficient condition data and the Lagrangian function to obtain the optimal decision variable and the optimal Lagrangian multiplier that meet the number of iterations;

[0016] A data model for multi-dimensional optimization of investment benefits is constructed based on the optimal Lagrangian multiplier and the Lagrangian function.

[0017] Preferably, the Lagrangian function is:

[0018]

[0019] Where x is the decision variable of the feature data, f(x) is the objective function, L(x, λ, ν) is the Lagrangian function, m is the total number of inequality constraints, p is the total number of equality constraints, g_i(x) is the i-th inequality constraint function, h_j(x) is the j-th inequality constraint function, λ i is the Lagrange multiplier of the ith inequality constraint function, ν j is the Lagrange multiplier of the jth equality constraint function.

[0020] Preferably, the necessary and sufficient condition data include that the gradient is zero, relaxation is complementary, all equality constraint functions are equal to zero, all inequality constraint functions are not greater than zero, and the Lagrange multiplier of the inequality constraint function is not less than zero.

[0021] Preferably, the project processing data is subjected to feature extraction and conversion processing to obtain feature data including:

[0022] Extracting the capital investment, operating cost, expected maintenance cost, and capital benefit of the project from the project processing data based on the timestamp as time characteristics;

[0023] extracting the ratio between the growth rate of electricity demand and the construction cost of the construction project and the interaction coefficient between the market competition degree and the potential customer base as the combined features from the project processing data;

[0024] Extracting topography and population density as geographic features from the project processing data;

[0025] The time feature, the combination feature and the geographic feature are converted into a set of linearly independent feature data by orthogonal transformation using the principal component analysis method.

[0026] Preferably, preprocessing the project data to obtain project processed data includes: preprocessing the project data by data cleaning, data normalization, missing data filling and outlier detection to obtain project processed data.

[0027] In another aspect, a device for constructing a multi-dimensional data model of investment benefits of power grid construction projects is provided, comprising a data acquisition and processing module, a feature extraction module, a model construction module and a model optimization module;

[0028] The data acquisition and processing module is used to acquire multi-dimensional project data of the power grid construction project, pre-process the project data, and obtain project processing data;

[0029] The feature extraction module is used to perform feature extraction and conversion processing on the project processing data to obtain feature data;

[0030] The model building module is used to iteratively process the characteristic data using the Lagrange multiplier method and convex optimization theory to build a data model that obtains multi-dimensional optimization of investment benefits;

[0031] The model optimization module is used to obtain historical power grid data, and optimize the Lagrange multipliers of the data model by cross-validation and hyperparameter tuning according to the historical power grid data to obtain an optimized data model.

[0032] Preferably, the model building module includes a data acquisition submodule, a function building submodule, an iterative calculation submodule and a model building submodule;

[0033] The data acquisition submodule is used to obtain the objective function and constraint condition data and necessary and sufficient condition data of the power grid construction project;

[0034] The function construction submodule is used to construct a Lagrangian function corresponding to the power grid construction project according to the characteristic data, the objective function and the constraint condition data;

[0035] The iterative calculation submodule is used to perform iterative processing using an interior point method according to the necessary and sufficient condition data and the Lagrangian function to obtain an optimal decision variable and an optimal Lagrangian multiplier that meet the number of iterations;

[0036] The model building submodule is used to build a data model for multi-dimensional optimization of investment benefits based on the optimal Lagrangian multiplier and the Lagrangian function.

[0037] Preferably, the Lagrangian function is:

[0038]

[0039] Where x is the decision variable of the feature data, f(x) is the objective function, L(x, λ, ν) is the Lagrangian function, m is the total number of inequality constraints, p is the total number of equality constraints, g_i(x) is the i-th inequality constraint function, h_j(x) is the j-th inequality constraint function, λ i is the Lagrange multiplier of the ith inequality constraint function, ν j is the Lagrange multiplier of the jth equality constraint function;

[0040] The necessary and sufficient condition data include that the gradient is zero, the relaxation is complementary, all equality constraint functions are equal to zero, all inequality constraint functions are not greater than zero, and the Lagrange multiplier of the inequality constraint function is not less than zero.

[0041] In another aspect, a terminal device is provided, comprising a processor and a memory;

[0042] The memory is used to store program codes and transmit the program codes to the processor;

[0043] The processor is used to execute the above-mentioned method for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to the instructions in the program code.

[0044] The invention discloses a method, device and terminal equipment for constructing a multi-dimensional data model of investment benefits of power grid construction projects. The method comprises acquiring multi-dimensional project data of power grid construction projects, pre-processing the project data to obtain project processing data; performing feature extraction and conversion processing on the project processing data to obtain feature data; iteratively processing the feature data using the Lagrange multiplier method and convex optimization theory to construct a data model for multi-dimensional optimization of investment benefits; acquiring historical power grid data, and optimizing the Lagrange multipliers of the data model using cross-validation and hyperparameter tuning according to the historical power grid data to obtain an optimized data model.

[0045] It can be seen from the above technical scheme that the present application has the following advantages: the method for constructing a multi-dimensional data model of investment benefits of power grid construction projects obtains multi-dimensional project data of power grid construction projects, pre-processes and extracts features of the project data in turn, and then uses the Lagrange multiplier method and convex optimization theory to iteratively construct a multi-dimensional optimized data model of investment benefits, and then optimizes the data model to obtain an optimized data model, so that the output of the optimized data model affects the evaluation of the optimal decision variables of the power grid construction project; it solves the technical problem that the existing power grid construction project evaluation model cannot guarantee the output results of the model due to the collection of single-dimensional data.

[0046] The device for constructing a multi-dimensional data model of investment benefit of power grid construction projects cooperates with a data acquisition and processing module, a feature extraction module, a model construction module and a model optimization module to construct a data model for multi-dimensional optimization of investment benefit, and then optimizes the data model to obtain an optimized data model, so that the optimized data model outputs the optimal decision variables that affect the evaluation of power grid construction projects; and solves the technical problem that the existing power grid construction project evaluation model cannot guarantee the output results of the model because it is constructed by collecting single-dimensional data. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0048] Figure 1 A flowchart of the steps of the method for constructing a multi-dimensional data model of investment benefits of power grid construction projects described in an embodiment of the present application;

[0049] Figure 2 A schematic diagram of a framework of a device for constructing a multi-dimensional data model of investment benefits of a power grid construction project according to an embodiment of the present application;

[0050] Figure 3 A schematic diagram of a terminal device described in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, features, and advantages of the invention of this application more obvious and easy to understand, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the embodiments described below are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0052] In the description of the embodiments of the present application, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0053] In the embodiments of the present application, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral one; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0054] The embodiment of the present application provides a method, device and terminal equipment for constructing a multi-dimensional data model of the investment benefit of a power grid construction project, which solves the technical problem that the existing power grid construction project evaluation model is constructed by collecting single-dimensional data, and the output result of the model cannot be guaranteed. In this embodiment, the optimization data model constructed by the method, device and terminal equipment for constructing a multi-dimensional data model of the investment benefit of the power grid construction project integrates multiple data sources by constructing an indicator system covering economic, social, environmental and social impacts, and evaluates the potential benefits and risks of the power grid construction project from multiple dimensions. The multi-objective optimization technology is adopted, and multiple objectives such as cost-effectiveness, environmental impact and social acceptance are considered at the same time, so that the data model finds the optimal solution in multiple possible decision-making schemes. By introducing the analysis of time series data, the performance of the project at different time points can be captured and predicted, providing more dynamic and real-time support for investment decisions. Combining the Lagrange multiplier method and convex optimization theory, when dealing with complex constrained optimization problems, it is possible to ensure that the global optimal solution is found, and high-level algorithms such as interior point methods are used to achieve efficient solution, save computing costs, and improve computing efficiency.

[0055] Embodiment 1:

[0056] Figure 1 This is a flowchart of the steps of the method for constructing a multi-dimensional data model of investment benefits of power grid construction projects described in an embodiment of the present application.

[0057] like Figure 1 As shown, the embodiment of the present application provides a method for constructing a multi-dimensional data model of investment benefits of a power grid construction project, including the following steps:

[0058] S1. Obtain multi-dimensional project data of the power grid construction project, pre-process the project data, and obtain project processing data.

[0059] It should be noted that in step S1, multi-dimensional project data of the power grid construction project is obtained, and the obtained project data is preprocessed. In this embodiment, project-related data of various power grid construction projects are obtained from data including economic indicators, social indicators, environmental indicators, and social impacts as project data. The project data has the potential benefits and risks of evaluating construction projects from multiple dimensions. Among them, the project data includes internal project reports, financial records, market research data, and relevant policies and regulations obtained from power grid-related construction analysis reports. In addition, project data may also include geographic and environmental data, especially for power grid construction projects involving large-scale geographical distribution, so as to obtain sufficient data to support subsequent analysis and model building.

[0060] In an embodiment of the present application, preprocessing the project data to obtain project processed data includes: preprocessing the project data by data cleaning, data normalization, missing data filling and outlier detection to obtain project processed data.

[0061] It should be noted that the acquired project data needs to be preprocessed to ensure the quality and availability of the project processing data, so as to support subsequent analysis and model building. In this embodiment, data cleaning refers to removing or correcting erroneous, incomplete, inconsistently formatted or duplicated project data. Data normalization refers to normalizing the project data to obtain project processing data in a standardized data range, so that the construction of subsequent data models will not be biased due to scale differences in variables. Missing value processing refers to processing missing data of project data by interpolation, using mean or median filling, etc. Outlier detection refers to identifying and processing outliers that may affect the analysis results and removing them.

[0062] S2. Perform feature extraction and conversion on the project processing data to obtain feature data.

[0063] It should be noted that in step S2, feature extraction and conversion are performed on the project processing data obtained in step S1 to obtain feature data. In this embodiment, feature extraction and conversion are performed on the project processing data to facilitate subsequent data mining analysis.

[0064] S3. Use Lagrange multiplier method and convex optimization theory to iteratively process feature data and construct a data model for multi-dimensional optimization of investment benefits.

[0065] It should be noted that in step S3, a model is established based on the characteristic data obtained in step S2 to obtain a data model for multi-dimensional optimization of investment benefits. In this embodiment, in the process of constructing a data model related to the power grid construction project, the problem of power grid construction project investment can usually be transformed into a specific type of optimization problem, that is, minimizing or maximizing an objective function under given constraints. The method based on Lagrange multiplier method and convex optimization theory can be used to solve the problem of optimization algorithm complexity. The Lagrange multiplier method has strong applicability in dealing with optimization problems with constraints, and convex optimization can provide a global optimal solution due to its unique mathematical properties, and can be solved with high efficiency.

[0066] S4. Obtain historical power grid data, and optimize the Lagrange multiplier of the data model using cross-validation and hyperparameter tuning based on the historical power grid data to obtain an optimized data model.

[0067] It should be noted that historical data refers to existing historical power grid data, which includes electricity, capacity, construction cost, etc. The historical power grid data uses cross-validation and hyperparameter tuning to train the Lagrange multipliers of the Lagrange function in the data model to obtain an optimized data model, thereby improving the generalization ability and prediction accuracy of the optimized data model. In this embodiment, the process of optimizing the data model training includes adjusting the data model parameters (such as the parameters involved in the Lagrange function) to minimize the prediction error. In addition, the stability and generalization ability of the data model can be verified by using cross-validation technology. Cross-validation refers to the use of K-fold cross-validation to evaluate the performance of the data model to ensure the performance of the data model on unseen data. Hyperparameter tuning refers to the use of grid search, random search or Bayesian optimization methods to find the optimal data model parameters. Performance evaluation refers to the use of mean square error to evaluate the predictive ability of the data model.

[0068] The present application provides a method for constructing a multi-dimensional data model of investment benefits of power grid construction projects, including obtaining multi-dimensional project data of power grid construction projects, preprocessing the project data to obtain project processing data; extracting features and converting the project processing data to obtain feature data; iteratively processing the feature data using the Lagrange multiplier method and convex optimization theory to construct a data model for multi-dimensional optimization of investment benefits; obtaining historical power grid data, and optimizing the Lagrange multipliers of the data model using cross-validation and hyperparameter tuning based on the historical power grid data to obtain an optimized data model. The method for constructing a multi-dimensional data model of investment benefits of power grid construction projects obtains multi-dimensional project data of power grid construction projects, preprocesses the project data in turn, extracts features, and then iteratively processes the data using the Lagrange multiplier method and convex optimization theory to construct a data model for multi-dimensional optimization of investment benefits, and then optimizes the data model to obtain an optimized data model, so that the optimized data model outputs the optimal decision variables that affect the evaluation of power grid construction projects; it solves the technical problem that the existing power grid construction project evaluation model cannot guarantee the output results of the model because it is constructed by collecting single-dimensional data.

[0069] In one embodiment of the present application, the Lagrange multiplier method and convex optimization theory are used to iteratively process the feature data to construct a data model for multi-dimensional optimization of investment benefits, including:

[0070] Obtain the objective function and constraint data as well as necessary and sufficient condition data of the power grid construction project;

[0071] Construct the Lagrangian function corresponding to the power grid construction project according to the characteristic data, objective function and constraint condition data;

[0072] According to the necessary and sufficient condition data and Lagrangian function, the interior point method is used for iterative processing to obtain the optimal decision variables and optimal Lagrangian multipliers that meet the number of iterations;

[0073] A data model for multi-dimensional optimization of investment benefits is constructed based on the optimal Lagrangian multipliers and Lagrangian functions.

[0074] It should be noted that the objective function f(x) represents the goal of the investment benefit of the power grid construction project, which can also be understood as the maximization of the investment benefit of the power grid construction project. The decision variable x includes influencing factors such as capital investment, operating cost, and expected maintenance cost. The constraint data includes technical constraints, regulatory constraints, and financial constraints. Technical constraints include capacity limitations of power grid transmission lines. Regulatory constraints include environmental protection regulations. Financial constraints include budget limitations of power grid construction projects, etc.

[0075] In an embodiment of the present application, the Lagrangian function is:

[0076]

[0077] Where x is the decision variable of the feature data, f(x) is the objective function, L(x, λ, ν) is the Lagrangian function, m is the total number of inequality constraints, p is the total number of equality constraints, g_i(x) is the i-th inequality constraint function, h_j(x) is the j-th inequality constraint function, λ i is the Lagrange multiplier of the ith inequality constraint function, ν j is the Lagrange multiplier of the jth equality constraint function.

[0078] It should be noted that λ i It also measures the degree of influence of the inequality constraint function on the result when maximizing the objective function. j It also measures the degree of influence of the equality constraint function on the result when maximizing the objective function.

[0079] In an embodiment of the present application, the necessary and sufficient condition data include gradient zero, relaxation complementarity, all equality constraint functions are equal to zero, all inequality constraint functions are not greater than zero, and the Lagrange multiplier of the inequality constraint function is not less than zero.

[0080] It should be noted that a gradient of zero means It means finding the point where the first-order derivative of the Lagrangian function in the objective function and the constraint is zero. represents the gradient of the objective function L with respect to the decision variable x, represents the Lagrangian function, represents the optimal solution of the decision variables, is the Lagrange multiplier corresponding to the inequality constraint function, is the Lagrange multiplier corresponding to the equality constraint function. Relaxed complementarity can be understood as: ; For all i, the inequality constraint function g_i(x) and the corresponding Lagrange multiplier The product of is equal to zero; for each inequality constraint, either the constraint is strictly satisfied and the corresponding multiplier is zero, or the multiplier is non-zero and the constraint becomes active. Original feasibility: ; At the optimal solution, all inequality constraint functions are satisfied (less than or equal to zero), and the equality constraint functions are also satisfied. Dual feasibility: ; The Lagrange multiplier must be non-negative to ensure that the constraints of the dual problem are met. In the process of constructing the data model for multi-dimensional optimization of investment benefits, it is ensured that all solutions are within the feasible domain of necessary and sufficient condition data, and all multipliers are legal.

[0081] In the embodiment of the present application, the interior point method is used to perform iterative processing according to the necessary and sufficient condition data and the Lagrangian function, and the optimal decision variables and the optimal Lagrangian multipliers that meet the number of iterations are obtained, including:

[0082] First, the parameters of the Lagrangian function are initialized: starting from an initial point that meets all the constraint data;

[0083] Secondly, at each step, the decision variable x of the characteristic function is adjusted so that the Lagrange multipliers λ and ν approach the optimal solution while staying within all necessary and sufficient condition data;

[0084] Finally, the iterative calculation is stopped when the number of iterations reaches the preset number of iterations, and the optimal decision variable is output. The value of , and the optimal Lagrange multiplier and , explain the influence of constraints on the final solution. According to the optimal Lagrange multiplier and Input the Lagrangian function to construct a data model for multi-dimensional optimization of investment benefits.

[0085] In one embodiment of the present application, feature extraction and conversion processing are performed on the project processing data to obtain feature data including:

[0086] The project processing data is extracted based on the timestamp, including the project's capital investment, operating cost, expected maintenance cost, and capital benefit as time features;

[0087] The ratio between the growth rate of electricity demand and the construction cost of the construction project and the interaction coefficient between the market competition degree and the potential customer base are extracted from the project processing data as combined features;

[0088] Extract terrain and population density as geographic features from project processing data;

[0089] The orthogonal transformation of principal component analysis method is used to transform the time features, combination features and geographical features into a set of linearly independent feature data.

[0090] It should be noted that the method for constructing a multi-dimensional data model of the investment benefit of the power grid construction project first selects the most influential features related to the power grid construction project from the project processing data. Secondly, the most useful features for data model prediction are selected from the project processing data to improve the performance of the data model and reduce the computational burden. In the investment benefit evaluation of power grid construction projects, there may be a large number of variables, such as project cost, timeliness, geographical location, market environment, etc., which have different degrees of influence on the final return on investment. Feature data includes but is not limited to the following features: First, time features: If the project processing data contains a timestamp, information such as year, month, day, and season can be extracted; for power grid projects, power demand in a specific time period may affect the cost and benefit of the project. Second, combined features: Create new features, such as the ratio between the growth rate of power demand and the cost of new facility construction, or the interaction coefficient between the degree of market competition and the base of potential customers. Third, geographical features: According to the geographical location of the power grid construction project, features related to terrain, population density, or regional economic activity intensity can be constructed. Finally, the extracted data is reduced in dimensionality: Dimensionality reduction includes using principal component analysis to convert the original features such as time features, combination features and geographical features into a set of linearly independent features through orthogonal transformation. Usually, the first few principal components with higher cumulative contribution rates (such as load, electricity, construction cost) are selected; data dimensions are reduced through techniques such as linear discriminant analysis to simplify the construction of data models and reduce computational complexity.

[0091] Embodiment 2:

[0092] Figure 2 A schematic diagram of the framework of a device for constructing a multi-dimensional data model of investment benefits of power grid construction projects as described in an embodiment of the present application.

[0093] like Figure 2 As shown, the embodiment of the present application provides a multi-dimensional data model construction device for investment benefits of power grid construction projects, including a data acquisition and processing module 10, a feature extraction module 20, a model construction module 30 and a model optimization module 40;

[0094] The data acquisition and processing module 10 is used to acquire multi-dimensional project data of the power grid construction project, pre-process the project data, and obtain project processing data;

[0095] The feature extraction module 20 is used to extract and convert the project processing data to obtain feature data;

[0096] A model building module 30 is used to iteratively process the characteristic data using the Lagrange multiplier method and convex optimization theory to build a data model for multi-dimensional optimization of investment benefits;

[0097] The model optimization module 40 is used to obtain historical power grid data, and optimize the Lagrange multipliers of the data model by cross-validation and hyperparameter tuning according to the historical power grid data to obtain an optimized data model.

[0098] It should be noted that the contents of the modules in the device for constructing a multidimensional data model of the investment benefits of power grid construction projects in Example 2 correspond to the contents of the steps in the method for constructing a multidimensional data model of the investment benefits of power grid construction projects. The contents of the steps in the method for constructing a multidimensional data model of the investment benefits of power grid construction projects have been described in Example 1, and the contents of the modules in the device for constructing a multidimensional data model of the investment benefits of power grid construction projects will not be described in detail in this embodiment. The device for constructing a multidimensional data model of the investment benefits of power grid construction projects constructs a multidimensional optimized data model of investment benefits through the cooperation of a data acquisition and processing module, a feature extraction module, a model construction module, and a model optimization module, and then optimizes the data model to obtain an optimized data model, so that the output of the optimized data model affects the optimal decision variables for evaluating power grid construction projects; this solves the technical problem that the existing power grid construction project evaluation model is constructed due to the collection of single-dimensional data, and the output results of the model cannot be guaranteed.

[0099] In the embodiment of the present application, the model construction module 30 includes a data acquisition submodule, a function construction submodule, an iterative calculation submodule and a model construction submodule;

[0100] The data acquisition submodule is used to obtain the objective function and constraint condition data and necessary and sufficient condition data of the power grid construction project;

[0101] Function construction submodule, used to construct the Lagrangian function corresponding to the power grid construction project according to the characteristic data, objective function and constraint condition data;

[0102] Iterative calculation submodule, used to perform iterative processing using the interior point method according to necessary and sufficient condition data and Lagrangian function, to obtain the optimal decision variables and optimal Lagrangian multipliers that meet the number of iterations;

[0103] The model building submodule is used to construct a data model for multi-dimensional optimization of investment benefits based on the optimal Lagrangian multipliers and Lagrangian functions.

[0104] In the embodiment of the present application, the Lagrangian function is:

[0105]

[0106] Where x is the decision variable of the feature data, f(x) is the objective function, L(x, λ, ν) is the Lagrangian function, m is the total number of inequality constraints, p is the total number of equality constraints, g_i(x) is the i-th inequality constraint function, h_j(x) is the j-th inequality constraint function, λ i is the Lagrange multiplier of the ith inequality constraint function, ν j is the Lagrange multiplier of the jth equality constraint function;

[0107] The necessary and sufficient condition data include that the gradient is zero, the relaxations are complementary, all equality constraint functions are equal to zero, all inequality constraint functions are not greater than zero, and the Lagrange multiplier of the inequality constraint function is not less than zero.

[0108] Embodiment three:

[0109] Figure 3 A schematic diagram of a terminal device described in an embodiment of the present application.

[0110] like Figure 3 As shown, an embodiment of the present application provides a terminal device, including a processor and a memory;

[0111] A memory, used for storing program codes and transmitting the program codes to a processor;

[0112] The processor is used to execute the above-mentioned method for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to the instructions in the program code.

[0113] It should be noted that the processor is used to execute the steps in the above-mentioned embodiment of a method for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to the instructions in the program code. Alternatively, the processor implements the functions of each module / unit in the above-mentioned system / device embodiments when executing the computer program.

[0114] Exemplarily, the computer program may be divided into one or more modules / units, one or more modules / units are stored in a memory and executed by a processor to complete the present application. One or more modules / units may be a series of computer program instruction segments capable of completing a specific function, and the instruction segments are used to describe the execution process of the computer program in a terminal device.

[0115] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that this does not constitute a limitation on the terminal device, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.

[0116] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0117] The memory may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. The memory may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device. Furthermore, the memory may include both an internal storage unit of the terminal device and an external storage device. The memory is used to store computer programs and other programs and data required by the terminal device. The memory may also be used to temporarily store data that has been output or is to be output.

[0118] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0119] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0120] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0121] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0122] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0123] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing a multi-dimensional data model of investment benefits of power grid construction projects, characterized in that: The following steps are involved: Acquire multi-dimensional project data of a power grid construction project, and pre-process the project data to obtain project processing data; Performing feature extraction and conversion processing on the project processing data to obtain feature data; Iteratively process the characteristic data using Lagrange multiplier method and convex optimization theory to construct a data model for multi-dimensional optimization of investment benefits; Historical power grid data is obtained, and Lagrange multipliers of the data model are optimized by cross-validation and hyperparameter tuning according to the historical power grid data to obtain an optimized data model.

2. The method for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to claim 1, characterized in that: The characteristic data are iteratively processed using the Lagrange multiplier method and convex optimization theory to construct a data model for multi-dimensional optimization of investment benefits, including: Obtaining the objective function and constraint condition data and necessary and sufficient condition data of the power grid construction project; Constructing a Lagrangian function corresponding to the power grid construction project according to the characteristic data, the objective function and the constraint condition data; Iterative processing is performed using the interior point method according to the necessary and sufficient condition data and the Lagrangian function to obtain the optimal decision variable and the optimal Lagrangian multiplier that meet the number of iterations; A data model for multi-dimensional optimization of investment benefits is constructed based on the optimal Lagrangian multiplier and the Lagrangian function.

3. The method for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to claim 2, characterized in that: The Lagrangian function is: Where x is the decision variable of the feature data, f(x) is the objective function, L(x, λ, ν) is the Lagrangian function, m is the total number of inequality constraints, p is the total number of equality constraints, g_i(x) is the i-th inequality constraint function, h_j(x) is the j-th inequality constraint function, λ i is the Lagrange multiplier of the ith inequality constraint function, ν j is the Lagrange multiplier of the jth equality constraint function.

4. The method for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to claim 2, characterized in that: The necessary and sufficient condition data include that the gradient is zero, the relaxation is complementary, all equality constraint functions are equal to zero, all inequality constraint functions are not greater than zero, and the Lagrange multiplier of the inequality constraint function is not less than zero.

5. The method for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to claim 1, characterized in that: The project processing data is subjected to feature extraction and conversion processing, and the feature data obtained includes: Extracting the capital investment, operating cost, expected maintenance cost, and capital benefit of the project from the project processing data based on the timestamp as time characteristics; extracting the ratio between the growth rate of electricity demand and the construction cost of the construction project and the interaction coefficient between the market competition degree and the potential customer base as the combined features from the project processing data; Extracting topography and population density as geographic features from the project processing data; The time feature, the combination feature and the geographic feature are converted into a set of linearly independent feature data by orthogonal transformation using the principal component analysis method.

6. The method for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to claim 1, characterized in that: Preprocessing the project data to obtain project processed data includes: preprocessing the project data by data cleaning, data normalization, missing data filling and outlier detection to obtain project processed data.

7. A device for constructing a multi-dimensional data model of investment benefits of power grid construction projects, characterized in that: It includes data acquisition and processing module, feature extraction module, model building module and model optimization module; The data acquisition and processing module is used to acquire multi-dimensional project data of the power grid construction project, pre-process the project data, and obtain project processing data; The feature extraction module is used to perform feature extraction and conversion processing on the project processing data to obtain feature data; The model building module is used to iteratively process the characteristic data using the Lagrange multiplier method and convex optimization theory to build a data model that achieves multi-dimensional optimization of investment benefits; The model optimization module is used to obtain historical power grid data, and optimize the Lagrange multipliers of the data model by cross-validation and hyperparameter tuning according to the historical power grid data to obtain an optimized data model.

8. The device for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to claim 7, characterized in that: The model building module includes a data acquisition submodule, a function building submodule, an iterative calculation submodule and a model building submodule; The data acquisition submodule is used to obtain the objective function and constraint condition data and necessary and sufficient condition data of the power grid construction project; The function construction submodule is used to construct a Lagrangian function corresponding to the power grid construction project according to the characteristic data, the objective function and the constraint condition data; The iterative calculation submodule is used to perform iterative processing using an interior point method according to the necessary and sufficient condition data and the Lagrangian function to obtain an optimal decision variable and an optimal Lagrangian multiplier that meet the number of iterations; The model building submodule is used to build a data model for multi-dimensional optimization of investment benefits based on the optimal Lagrangian multiplier and the Lagrangian function.

9. The device for constructing a multi-dimensional data model of investment benefits of power grid construction projects according to claim 8, characterized in that: The Lagrangian function is: Where x is the decision variable of the feature data, f(x) is the objective function, L(x, λ, ν) is the Lagrangian function, m is the total number of inequality constraints, p is the total number of equality constraints, g_i(x) is the i-th inequality constraint function, h_j(x) is the j-th inequality constraint function, λ i is the Lagrange multiplier of the ith inequality constraint function, ν j is the Lagrange multiplier of the jth equality constraint function; The necessary and sufficient condition data include that the gradient is zero, the relaxation is complementary, all equality constraint functions are equal to zero, all inequality constraint functions are not greater than zero, and the Lagrange multiplier of the inequality constraint function is not less than zero.

10. A terminal device, characterized in that: including a processor and a memory; The memory is used to store program codes and transmit the program codes to the processor; The processor is used to execute the method for constructing a multi-dimensional data model of investment benefits of power grid construction projects as described in any one of claims 1-6 according to the instructions in the program code.

Images