A method and system for predicting carbon emissions of power transmission line projects based on improved Beluga algorithm

Abstract

The present invention discloses a method and system for predicting carbon emissions of power transmission line projects based on an improved Beluga algorithm. The method comprises obtaining data on carbon emissions from the construction of power transmission line projects and the main influencing factors of carbon emissions from the design scheme of power transmission line projects, and constructing a data set; preprocessing the data in the data set; constructing a support vector machine prediction model; using the improved Beluga algorithm to find the optimal parameters of the support vector machine prediction model; bringing the optimal parameters into the support vector machine prediction model, and inputting the preprocessed data set into the support vector machine prediction model to obtain carbon emission prediction data. The solution provided by the present invention uses an improved Beluga optimization algorithm to optimize the core parameters of the support vector machine, which effectively improves the accuracy of the model's carbon emission prediction. By predicting the carbon emissions of the design scheme of the power transmission line project, it can provide a decision-making reference for power grid companies to plan and design low-carbon cost power transmission lines.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission assessment for new power system infrastructure, and in particular to a method and system for predicting carbon emissions of power transmission line projects based on an improved Beluga algorithm. Background Art

[0002] The traditional form of power generation, which is mainly based on thermal power, is limited by the stock of fossil fuel resources and greenhouse gas emissions, and it is difficult to independently and comprehensively support the needs of sustainable development. Therefore, vigorously developing renewable energy has become the key to the transformation of the energy structure. However, the distribution of renewable energy resources such as wind power, solar energy and hydropower is uneven with electricity consumption, which has also led to long-distance, large-capacity transmission line projects becoming a key link in helping to supply and absorb new energy. As the cornerstone of the new power system, the transmission line project has been expanded on a large scale, accompanied by a large amount of resource and energy input, resulting in a large amount of carbon emissions. The design scheme of the transmission line project directly affects the overall resource input of the transmission line project. In this context, how to scientifically and reasonably predict the carbon emissions of the transmission line project design scheme in order to formulate a transmission line project design scheme with low carbon emission costs has become a major test for achieving low-carbonization of cross-regional transmission infrastructure.

[0003] The existing assessment of carbon emissions of transmission line infrastructure mainly uses a hybrid life cycle assessment method, combined with a standard cost list for transmission line projects, to calculate the carbon emissions of engineering construction, and propose management opinions on pollution reduction and carbon reduction of transmission infrastructure based on the calculation results. However, when carbon emissions are calculated based on the budget list form of the project under construction or the cost list form after the project is completed and accepted, the management opinions proposed are mostly focused on the selection of engineering materials, and lack specific guidance on the design and planning of transmission line projects. Therefore, studying new carbon emission prediction methods is of great significance for the construction of low-carbon transmission lines. Summary of the invention

[0004] In order to solve the technical problem that the existing carbon emission assessment method cannot effectively optimize the design planning of transmission line projects, the embodiment of the present invention provides a method and system for predicting carbon emissions of transmission line projects based on an improved Beluga algorithm.

[0005] The technical solution of the embodiment of the present invention is achieved as follows:

[0006] The embodiment of the present invention provides a method for predicting carbon emissions of a power transmission line project based on an improved Beluga algorithm, the method comprising:

[0007] Obtain data on carbon emissions from power transmission line construction and major influencing factors of carbon emissions from power transmission line engineering design solutions, and construct a data set; preprocess the data in the data set; construct a support vector machine prediction model; use an improved Beluga algorithm to find the optimal parameters of the support vector machine prediction model; bring the optimal parameters into the support vector machine prediction model, and input the preprocessed data set into the support vector machine prediction model to obtain carbon emission prediction data;

[0008] The improved Beluga algorithm is a Beluga optimization algorithm that introduces an elite reverse learning strategy and a dynamic refraction reverse learning strategy. The elite reverse learning strategy is used to retain the elite individual information in the search process of the Beluga optimization algorithm; the dynamic refraction reverse learning strategy is used to dynamically adjust parameters through the number of iterations while retaining the elite individual information, so that the search range decreases as the number of iterations increases;

[0009] The implementation process of the elite reverse learning strategy includes:

[0010]

[0011]

[0012]

[0013] in, Initialize the position for the i-th individual; and is a random number in the range of (0,1); is the lower bound of the parameter; is the upper bound of the parameter; is the reverse solution corresponding to the i-th individual; For elite individual positions; Solve the function for fitness;

[0014] The implementation process of the dynamic refraction reverse learning strategy includes:

[0015]

[0016]

[0017] in, is the dynamic refraction parameter; is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm; is the lower bound of the parameter; is the upper bound of the parameter; For elite individual positions; is the reverse solution of the i-th elite individual.

[0018] In one embodiment, data on major influencing factors of carbon emissions of a transmission line engineering design scheme are obtained, including: obtaining all carbon emission influencing factors of the transmission line engineering design scheme; taking all carbon emission influencing factors as a total sample, using a resampling technique, selecting m samples from the total sample to determine as in-bag data, and using the in-bag data as a training sample set; determining the remaining samples after the selection as out-of-bag data, and using the out-of-bag data as a test set sample; using the training sample set to generate a decision tree and each node, repeatedly sampling the total sample with replacement, repeating n times, and generating n decision trees; training a random forest model based on the generated n decision trees; using the trained random forest model to predict the test set samples, and using a voting method to obtain prediction results; calculating the importance of each carbon emission influencing factor through the error rate of the prediction result; setting an influencing factor importance threshold, and using the carbon emission influencing factors whose importance is greater than or equal to the influencing factor importance threshold as the main influencing factors.

[0019] In one embodiment, calculating the importance of each carbon emission influencing factor by using the error rate of the prediction result includes: calculating the importance of each carbon emission influencing factor by using the following calculation formula by using the error rate of the prediction result:

[0020]

[0021] in, Factors affecting carbon emissions The importance of is the number of decision trees, Factors affecting carbon emissions The importance of the i-th decision tree is calculated as:

[0022]

[0023] in, is the number of observation columns of out-of-bag data of the i-th tree; is an indicator function, which takes 1 when two values ​​are equal and 0 when they are not equal; is the true result of the pth observation; is the prediction result of the p-th observation value of the out-of-bag data by the i-th decision tree before random permutation; is the prediction result of the i-th decision tree for the p-th observation of the out-of-bag data after random permutation.

[0024] In one embodiment, constructing a support vector machine prediction model includes:

[0025] The problem form of the support vector machine is defined as follows:

[0026]

[0027] in, is the direction of the hyperplane; C is the regularization constant; m is the number of samples; , are the horizontal and vertical coordinates of the samples respectively; is an insensitive loss function;

[0028] Introduce slack variables to transform the problem form, and the transformed problem form is:

[0029]

[0030] in, is the direction of the hyperplane; C is the regularization constant; m is the number of samples; and are two slack variables;

[0031] Based on the transformed problem form, Lagrange multipliers and kernel functions are introduced, and the support vector machine prediction model obtained is:

[0032]

[0033] Among them, m is the number of samples, and is the Lagrange multiplier, is the bias term, is the kernel function, and its calculation formula is as follows:

[0034]

[0035] in, and are two data points in the input space, is the bandwidth of the radial basis kernel function.

[0036] In one embodiment, the method of finding the optimal parameters of the support vector machine prediction model using the improved Beluga algorithm includes:

[0037] Set the initial parameters of the White Whale optimization algorithm; calculate and compare the balance factor and the White Whale drop probability to determine the stage of the algorithm; obtain the optimal values ​​of the regularization constant and radial basis kernel function bandwidth of the support vector machine prediction model.

[0038] In one embodiment, the balance factor is calculated as:

[0039]

[0040] in, is the balance factor, is a value that changes randomly from (0,1) during each iteration. is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm;

[0041] The calculation formula for the drop probability of White Whale is:

[0042]

[0043] in, The drop rate of White Whale. is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm;

[0044] The calculation compares the balance factor and the White Whale drop probability to determine the stage of the algorithm, including:

[0045] Balance Factor When , the algorithm is judged to be in the exploration stage; When , the algorithm is judged to be in the development stage; When , the algorithm is judged to be in the whale fall stage.

[0046] In one embodiment, the position update formula in the exploration phase is:

[0047]

[0048] in, is the updated position of the i-th beluga whale in the j-th dimension; For the i-th beluga whale P j The location of the dimension; P j is a randomly selected integer from the beluga population d; and is a random operator in the range (0,1); The current location of a random beluga whale; , To mirror the orientation of the beluga whale's fin relative to the water surface;

[0049] The position update formula in the development phase is:

[0050]

[0051] in, The current iteration position of the i-th white whale; are beluga whales randomly selected from the population; Updated position for the i-th beluga whale; The best position for white whales; and is a random operator in the range (0,1); The jump strength for Levy's flight; is the Levy flight function; and is a random number that conforms to the normal distribution; is the gamma function; is a constant of 1.5; is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm;

[0052] The position update formula in the whale fall phase is:

[0053]

[0054] in, is the updated position of the i-th beluga whale, is the current iteration position of the i-th white whale, is a beluga whale randomly selected from the population, , and is a random operator in the range (0,1); for the step length of the whale's fall; is the step factor related to the beluga population size and drop probability; is the beluga population size, is the lower bound of the parameter; is the upper bound of the parameter; is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm; The drop rate of White Whale.

[0055] In one embodiment, a whirlwind foraging strategy is introduced in the algorithm development stage, so that the beluga population approaches the optimal solution in a spiral motion; the position update formula in the post-introduction development stage is:

[0056]

[0057]

[0058] in, is the updated position of the i-th beluga whale, The best position for the white whale. is the current iteration position of the i-th white whale, is the weight coefficient and changes with the number of iterations; The current iteration position of the i-1th white whale; , is a random operator in the range (0,1), is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm.

[0059] An embodiment of the present invention also provides a carbon emission prediction system for a transmission line project based on an improved Beluga algorithm, comprising: a processor and a memory for storing a computer program that can be run on the processor; wherein the processor executes the steps of the above-mentioned method when running the computer program.

[0060] The method and system for predicting carbon emissions of power transmission line projects based on the improved Beluga algorithm provided in the embodiments of the present invention can predict carbon emissions of power transmission line project design schemes in the early stage of power transmission line construction by screening the factors affecting carbon emissions of power transmission line project design schemes. This can provide a decision-making reference for power grid companies in planning and designing power transmission lines with low carbon emission costs.

[0061] At the same time, by introducing the elite reverse learning strategy, dynamic refraction reverse learning strategy and whirlwind foraging strategy to improve the White Whale algorithm, the shortcomings of slow convergence speed and unstable optimization results of the algorithm are compensated, and the global search and local search capabilities of the White Whale optimization algorithm are balanced. The support vector machine prediction model suitable for small samples is adopted, and the parameters of the support vector machine model are optimized using the improved White Whale algorithm, which can effectively improve the prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flow chart of a method for predicting carbon emissions of a power transmission line project based on an improved Beluga algorithm according to an embodiment of the present invention;

[0063] Figure 2 It is a flow chart of the application process of the method for predicting carbon emissions of power transmission line projects based on the improved Beluga algorithm according to an embodiment of the present invention;

[0064] Figure 3 1 is a diagram showing the internal structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0065] The present invention will be described in further detail below with reference to the accompanying drawings and embodiments.

[0066] The embodiment of the present invention provides a method for predicting carbon emissions of power transmission line projects based on an improved Beluga algorithm. Figure 1 As shown, the method includes:

[0067] Step 101: Acquire data on carbon emissions from construction of a transmission line project and major influencing factors of carbon emissions from a design of a transmission line project, and construct a data set; pre-process the data in the data set;

[0068] Step 102: construct a support vector machine prediction model; use the improved Beluga algorithm to find the optimal parameters of the support vector machine prediction model;

[0069] Step 103: Bring the optimal parameters into the support vector machine prediction model, and input the preprocessed data set into the support vector machine prediction model to obtain carbon emission prediction data.

[0070] Specifically, see Figure 2 This embodiment is based on the improved Beluga algorithm to optimize the support vector machine transmission line engineering design scheme carbon emission prediction method, specifically including the following steps:

[0071] S1: Build the dataset.

[0072] Since there are few studies on factors affecting carbon emissions of power transmission line engineering design solutions, this embodiment selects the most representative influencing factors and uses a random forest algorithm to screen important influencing factors, including the following sub-steps:

[0073] S11: Construct a set of factors affecting carbon emissions of transmission line engineering design schemes;

[0074] S12, using random forest algorithm to screen the main influencing factors;

[0075] S13, construct the final dataset.

[0076] Among them, the influencing factor set in S11 includes the unit investment of the transmission line, the size of the wind speed zone where the transmission line is located, the icing conditions, and the length of the terrain such as flat land, swamps, river networks, hills, mountains, high mountains, steep mountains and deserts that the transmission line passes through.

[0077] In S12, the random forest algorithm is used to screen the main influencing factors, including:

[0078] Using resampling technology, m samples (in-bag data) are selected from the initial influencing factor set as the training sample set; a decision tree and each node are generated using the sample set, and the samples are sampled repeatedly with replacement, repeated n times, to generate n decision trees; the trained random forest model is used to predict the test set samples (out-of-bag data), and the voting method is used to obtain the prediction results; the importance of each influencing factor is calculated by the out-of-bag error rate (Out-of-Bag Error); the importance threshold of the influencing factor is set , the importance is greater than or equal to The influencing factors are taken as the main influencing factors.

[0079] The importance of each influencing factor is obtained through the out-of-bag data error rate It is the feature selection method provided by the random forest algorithm, and the calculation formula is:

[0080]

[0081]

[0082] in, Influencing factors The importance of the i-th decision tree; is the number of decision trees, is the number of observation columns of out-of-bag data of the i-th tree; is an indicator function, which takes 1 when two values ​​are equal and 0 when they are not equal; is the true result of the pth observation; is the prediction result of the p-th observation value of the out-of-bag data by the i-th decision tree before random permutation; is the prediction result of the i-th decision tree for the p-th observation of the out-of-bag data after random permutation.

[0083] The final data set constructed in S13 includes the carbon emissions from transmission line project construction and the main influencing factors screened in S12.

[0084] S2: Preprocess the data.

[0085] Data normalization. Since the data value range of the carbon emission influencing factors of the transmission line engineering design scheme collected is large and the dimensions are different, it is necessary to normalize the data to eliminate the impact of different data dimensions and magnitudes on the prediction results. This embodiment uses the range normalization method to normalize the data:

[0086]

[0087] in, , They are the maximum and minimum values ​​in the same set of data respectively; The data for input model; is the normalized data.

[0088] S3: Build a support vector machine prediction model.

[0089] In this embodiment, a support vector machine regression prediction model suitable for small sample data is adopted, which includes the following steps:

[0090] S31: Define the problem form of support vector machine;

[0091] S32, to make the problem easier to solve, slack variables are introduced to transform the problem form;

[0092] S33, introduce Lagrange multipliers and kernel functions to obtain the final support vector machine prediction model.

[0093] Among them, the support vector machine problem form in S31 is:

[0094]

[0095] in, is the direction of the hyperplane; C is the regularization constant; m is the number of samples; , are the horizontal and vertical coordinates of the samples respectively; is an insensitive loss function.

[0096] After the slack variables are introduced in S32, the problem form is transformed into:

[0097]

[0098] in, and are two slack variables.

[0099] After introducing Lagrange multipliers and kernel functions in S33 for solution, the final support vector machine prediction model is obtained as follows:

[0100]

[0101] in, is the kernel function, and is the Lagrange multiplier, is the bias term.

[0102] The kernel function of this embodiment uses the radial basis kernel function (RBF) with fewer parameters and stronger generalization ability, and the calculation formula is as follows:

[0103]

[0104] in, is the bandwidth of the radial basis kernel function, and are two data points in the input space.

[0105] S4: Find the optimal parameters of support vector machine based on the improved Beluga optimization algorithm.

[0106] The Beluga optimization algorithm used in this embodiment is inspired by the three behaviors of beluga whales: swimming, hunting, and whale falls in the sea. It is a new meta-Qi method optimization algorithm proposed in 2022. The Beluga optimization algorithm has been improved by introducing the elite reverse learning strategy, dynamic refraction reverse learning strategy, and cyclone foraging strategy to effectively improve the convergence speed of the algorithm while avoiding falling into the local optimal solution, and at the same time balance the global search and local search capabilities of the Beluga optimization algorithm.

[0107] The search for optimal parameters of support vector machine based on improved Beluga optimization algorithm mainly focuses on finding the regularization constant C and radial basis kernel function bandwidth in the support vector machine prediction model. The optimal value of includes the following sub-steps:

[0108] S41: Initial parameter setting of White Whale optimization algorithm;

[0109] S42: Calculate the comparison balance factor Drop rate with White Whale Size, to determine the stage of the algorithm;

[0110] S43: Obtain the regularization constant C and radial basis kernel function bandwidth of the support vector machine prediction model The optimal value of .

[0111] Among them, the initial parameter settings of the S41 White Whale optimization algorithm include population initialization position setting, population fitness function setting, parameter upper limit setting, parameter lower limit setting, optimization parameter number setting, search quantity setting and algorithm maximum iteration number setting.

[0112] Although the White Whale optimization algorithm in S41 has a strong global search capability, the population initialization often uses random initialization, which is prone to falling into local optimal solutions and limited adaptability. Based on this defect, this embodiment introduces an elite reverse learning strategy to retain elite individual information when setting the population initialization position, so that the algorithm maintains a high performance level during the search process, and further introduces a dynamic refraction reverse learning strategy. While retaining elite individuals, the parameters are dynamically adjusted through the number of iterations, so that the search range decreases as the number of iterations increases. This is conducive to improving the algorithm convergence speed and avoiding the algorithm from falling into a local optimal solution.

[0113] The implementation process of the elite reverse learning strategy includes:

[0114]

[0115]

[0116]

[0117] in, Initialize the position for the i-th individual; and is a random number in the range of (0,1); is the lower bound of the parameter; is the upper bound of the parameter; is the reverse solution corresponding to the i-th individual; For elite individual positions; Solve the function for fitness.

[0118] The implementation process of the dynamic refraction reverse learning strategy includes:

[0119]

[0120]

[0121] in, is the dynamic refraction parameter; is the reverse solution of the i-th elite individual, For elite individual positions.

[0122] S42 Balance Factor Drop rate with White Whale The calculation formula is:

[0123]

[0124]

[0125] in, It is a value that changes randomly from (0,1) during each iteration; is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm. The algorithm is balanced factor During the exploration phase, At the development stage, It is in the whale fall stage.

[0126] The position update formula of the exploration phase of the White Whale optimization algorithm is:

[0127]

[0128] in, is the updated position of the i-th beluga whale in the j-th dimension; is the position of the ith beluga whale in the Pj dimension; Pj is a randomly selected integer from the beluga population d; and is a random operator in the range (0,1); The current location of a random beluga whale; , Mirrors the orientation of the beluga whale's fin relative to the water surface.

[0129] The position update formula in the development phase of the White Whale optimization algorithm is:

[0130]

[0131] in, The current iteration position of the i-th white whale; are beluga whales randomly selected from the population; Updated position for the i-th beluga whale; The best position for white whales; and is a random operator in the range (0,1); The jump strength for Levy's flight; is the Levy flight function; and is a random number that conforms to the normal distribution; is the gamma function; is a constant of 1.5.

[0132] In order to balance the global search capability and local search capability of the Beluga optimization algorithm, this embodiment introduces a whirlwind foraging strategy in the algorithm development stage, so that the Beluga population approaches the optimal solution in a spiral motion. The specific position update formula is:

[0133]

[0134]

[0135] in, is the weight coefficient and changes with the number of iterations; , is a random operator in the range (0,1), The current iteration position of the i-1th white whale.

[0136] The position update formula in the whale fall phase is:

[0137]

[0138] in, , and is a random operator in the range (0,1); for the step length of the whale's fall; is the step factor related to the beluga population size and drop probability; The size of the beluga whale population.

[0139] S43 improves the White Whale optimization algorithm, effectively avoiding the algorithm from falling into the local optimal solution, and balances the algorithm's global search ability and local search ability. The obtained support vector machine prediction model regularization constant C and radial basis kernel function bandwidth The optimal value is obtained faster.

[0140] S5: Bring the optimal parameters into the support vector machine and input the processed data set into the model to achieve carbon emission prediction.

[0141] In this example, the optimal parameters obtained by the improved Beluga optimization algorithm are brought into the support vector machine prediction model. The carbon emission prediction of the transmission line engineering design scheme is realized by inputting the processed data set into the model, and the final carbon emission prediction value is obtained by denormalizing the prediction results.

[0142] To summarize, the present embodiment provides a method for predicting carbon emissions from transmission line engineering design schemes based on an improved Beluga algorithm to optimize a support vector machine. The method optimizes the initial position of the Beluga population by combining an elite reverse learning strategy with a dynamic refraction reverse learning strategy, thereby effectively avoiding the Beluga optimization algorithm from falling into a local optimal solution while accelerating the convergence of the algorithm. The whirlwind foraging strategy is introduced to optimize the development phase of the Beluga optimization algorithm, thereby balancing the global search capability and local search capability of the algorithm and ensuring the stability of obtaining the optimal parameters. The improved Beluga optimization algorithm is used to optimize the core parameters of the support vector machine, thereby effectively improving the accuracy of the model's carbon emission prediction. By predicting the carbon emissions from the transmission line engineering design scheme, it can provide a decision-making reference for power grid companies to plan and design low-carbon transmission lines.

[0143] This embodiment introduces an elite reverse learning strategy, a dynamic refraction reverse learning strategy, and a whirlwind foraging strategy to effectively avoid the White Whale optimization algorithm from falling into a local optimal solution while accelerating the convergence of the algorithm, balancing the global search capability and local search capability of the algorithm, and ensuring the stability of obtaining the optimal parameters. The improved White Whale optimization algorithm is used to optimize the core parameters of the support vector machine, which effectively improves the accuracy of the model's carbon emission prediction. By predicting the carbon emissions of the transmission line engineering design scheme, it can provide a decision-making reference for power grid companies to plan and design low-carbon transmission lines.

[0144] In order to implement the method of an embodiment of the present invention, an embodiment of the present invention also provides a carbon emission prediction system for a transmission line project based on an improved Beluga algorithm, comprising: a processor and a memory for storing a computer program that can be run on the processor; wherein, when the processor is used to run the computer program, the steps of the above-mentioned method are executed.

[0145] The above-mentioned system provided in this embodiment belongs to the same concept as the above-mentioned method embodiment. The specific implementation process thereof is detailed in the method embodiment and will not be repeated here.

[0146] In order to implement the method of the embodiment of the present invention, the embodiment of the present invention also provides a computer program product, which includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of the above method.

[0147] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiment of the present invention, the embodiment of the present invention further provides an electronic device (computer device). Specifically, in one embodiment, the computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05 and a memory (not shown in the figure) connected through a system bus. Among them, the processor A01 of the computer device is used to provide computing and control capabilities. The memory of the computer device includes an internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A06. The network interface A02 of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor A01, the method of any one of the above embodiments is implemented. The display screen A04 of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device A05 of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0148] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0149] The device provided by an embodiment of the present invention includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the method of any one of the above embodiments is implemented.

[0150] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0151] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0152] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0153] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0154] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0155] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0156] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0157] It can be understood that the memory of the embodiment of the present invention can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAMbus random access memory (DRRAM, Direct Rambus Random Access Memory).The memories described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memories.

[0158] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0159] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for predicting carbon emissions of power transmission line projects based on an improved Beluga algorithm, characterized in that: The method comprises: Acquire data on carbon emissions from power transmission line project construction and major influencing factors of carbon emissions from power transmission line project design solutions, and construct a data set; pre-process the data in the data set; Constructing a support vector machine prediction model; using an improved Beluga algorithm to find the optimal parameters of the support vector machine prediction model; Bringing the optimal parameters into the support vector machine prediction model, and inputting the preprocessed data set into the support vector machine prediction model to obtain carbon emission prediction data; The improved Beluga algorithm is a Beluga optimization algorithm that introduces an elite reverse learning strategy and a dynamic refraction reverse learning strategy. The elite reverse learning strategy is used to retain the elite individual information in the search process of the Beluga optimization algorithm; the dynamic refraction reverse learning strategy is used to dynamically adjust parameters through the number of iterations while retaining the elite individual information, so that the search range decreases as the number of iterations increases; The implementation process of the elite reverse learning strategy includes: in, Initialize the position for the i-th individual; and is a random number in the range of (0,1); is the lower bound of the parameter; is the upper bound of the parameter; is the reverse solution corresponding to the i-th individual; For elite individual positions; Solve the function for fitness; The implementation process of the dynamic refraction reverse learning strategy includes: in, is the dynamic refraction parameter; is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm; is the lower bound of the parameter; is the upper bound of the parameter; For elite individual positions; is the reverse solution of the i-th elite individual; Among them, the data on the main factors affecting carbon emissions of transmission line engineering design solutions are obtained, including: Obtain carbon emission influencing factors of all transmission line engineering design schemes; All carbon emission influencing factors are taken as the total sample, and the resampling technology is used to select m samples from the total sample to determine them as in-bag data, and the in-bag data is used as the training sample set; the remaining samples after the selection are determined as out-of-bag data, and the out-of-bag data is used as the test set samples; Generate a decision tree and each node using the training sample set, repeatedly sample the total sample with replacement, repeat n times, and generate n decision trees; train a random forest model based on the generated n decision trees; The trained random forest model is used to predict the test set samples, and the prediction results are obtained by using a voting method; The importance of each carbon emission influencing factor is calculated by the error rate of the prediction results; An importance threshold of influencing factors is set, and carbon emission influencing factors whose importance is greater than or equal to the importance threshold of influencing factors are taken as main influencing factors.

2. The method for predicting carbon emissions of power transmission line projects based on the improved Beluga algorithm according to claim 1 is characterized in that: The importance of each carbon emission influencing factor is calculated by the error rate of the prediction results, including: The importance of each carbon emission influencing factor is calculated by the error rate of the prediction results using the following calculation formula: in, Factors affecting carbon emissions The importance of is the number of decision trees, Factors affecting carbon emissions The importance of the i-th decision tree is calculated as: in, is the number of observation columns of out-of-bag data of the i-th tree; is an indicator function, which takes 1 when two values ​​are equal and 0 when they are not equal; is the true result of the pth observation; is the prediction result of the p-th observation value of the out-of-bag data by the i-th decision tree before random permutation; is the prediction result of the p-th observation of the out-of-bag data by the i-th decision tree after random permutation.

3. The method for predicting carbon emissions of power transmission line projects based on the improved Beluga algorithm according to claim 1 is characterized in that: Build a support vector machine prediction model, including: The problem form of the support vector machine is defined as follows: in, is the direction of the hyperplane; C is the regularization constant; m is the number of samples; , are the horizontal and vertical coordinates of the samples respectively; is an insensitive loss function; Introduce slack variables to transform the problem form, and the transformed problem form is: in, is the direction of the hyperplane; C is the regularization constant; m is the number of samples; and are two slack variables; Based on the transformed problem form, after introducing Lagrange multipliers and kernel functions for solving, the support vector machine prediction model obtained is: Among them, m is the number of samples, and is the Lagrange multiplier, is the bias term, is the kernel function, and its calculation formula is as follows: in, and are two data points in the input space, is the bandwidth of the radial basis kernel function.

4. The method for predicting carbon emissions of power transmission line projects based on the improved Beluga algorithm according to claim 1, characterized in that: The method of using the improved Beluga algorithm to find the optimal parameters of the support vector machine prediction model includes: Set the initial parameters of the White Whale optimization algorithm; Calculate and compare the balance factor and the probability of White Whale dropping to determine the stage of the algorithm; Obtain the optimal values ​​of the regularization constant and the bandwidth of the radial basis kernel function of the support vector machine prediction model.

5. The method for predicting carbon emissions of power transmission line projects based on the improved Beluga algorithm according to claim 4 is characterized in that: The balance factor is calculated as: in, is the balance factor, is a value that changes randomly from (0,1) during each iteration. is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm; The calculation formula for the drop probability of White Whale is: in, is the drop rate of White Whale. is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm; The calculation compares the balance factor and the White Whale drop probability to determine the stage of the algorithm, including: Balance Factor When , the algorithm is judged to be in the exploration stage; When , the algorithm is judged to be in the development stage; When , the algorithm is judged to be in the whale fall stage.

6. The method for predicting carbon emissions of power transmission line projects based on the improved Beluga algorithm according to claim 5, characterized in that: The position update formula in the exploration phase is: in, is the updated position of the i-th beluga whale in the j-th dimension; For the i-th beluga whale P j The location of the dimension; P j is a randomly selected integer from the beluga population d; and is a random operator in the range (0,1); The current location of a random beluga whale; , To mirror the orientation of the beluga whale's fin relative to the water surface; The position update formula in the development phase is: in, The current iteration position of the i-th white whale; are beluga whales randomly selected from the population; Updated position for the i-th beluga whale; The best position for the white whale; and is a random operator in the range (0,1); The jump strength for Levy's flight; is the Levy flight function; and is a random number that conforms to the normal distribution; is the gamma function; is a constant of 1.5; is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm; The position update formula in the whale fall phase is: in, is the updated position of the i-th beluga whale, is the current iteration position of the i-th beluga whale, is a beluga whale randomly selected from the population, , and is a random operator in the range (0,1); The step length for the whale to fall; is the step factor related to the beluga population size and drop probability; is the beluga population size, is the lower bound of the parameter; is the upper bound of the parameter; is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm; The drop rate of White Whale.

7. The method for predicting carbon emissions of power transmission line projects based on the improved Beluga algorithm according to claim 6, characterized in that: The whirlwind foraging strategy was introduced in the algorithm development stage, so that the beluga population approaches the optimal solution in a spiral motion; the position update formula in the post-introduction development stage is: in, is the updated position of the i-th beluga whale, The best position for the white whale. is the current iteration position of the i-th beluga whale, is the weight coefficient and changes with the number of iterations; The current iteration position of the i-1th white whale; , is a random operator in the range (0,1), is the current iteration number of the algorithm; is the maximum number of iterations of the algorithm.

8. A carbon emission prediction system for power transmission line projects based on an improved Beluga algorithm, characterized in that: include: A processor and a memory for storing a computer program that can be run on the processor; wherein, when the processor is used to run the computer program, the steps of the method according to any one of claims 1 to 7 are executed.

Images