A modern rural power grid planning method and system

Abstract

The present invention relates to the technical field of power grid planning, and particularly to a modern rural power grid planning method and system. The method specifically includes: determining the carbon emission data of various power generation technologies within the power grid planning area; obtaining the historical power demand data of different types of agricultural industries within the power grid planning area, and determining the power demand change trend of each type of agricultural industry according to the historical power demand data; establishing a power grid planning optimization model by using a multi-objective evolutionary algorithm based on the prediction results of the unit efficiency prediction model, the evaluation results of the unit status evaluation model, the carbon emission data, and the power demand change trend of each type of agricultural industry. The power grid planning optimization model is used to determine the optimal power grid planning scheme. The present invention improves the scientificity and rationality of power grid planning, enhances the environmental protection performance of the power grid planning scheme, and strengthens the compatibility between the power grid planning scheme and the actual needs of agricultural production.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid planning, and in particular, to a method and system for modern rural power grid planning. Background Art

[0002] With the rapid development of modern agriculture, as an important infrastructure to support agricultural production and rural life, the planning and construction of rural power grids are particularly important. Traditional rural power grid planning methods often focus on meeting basic power demands, while ignoring various factors such as the efficiency of generating units, the status of units, the carbon emissions of power generation technologies, and the changing trends of power demands in the agricultural industry, resulting in low comprehensive benefits of power grid planning schemes. Specifically, the traditional rural power grid planning methods have the following problems:

[0003] 1. Insufficient data utilization: The potential value of data from generating units fails to be fully explored and utilized. Due to the lack of effective data analysis and application, it is impossible to accurately predict the efficiency and status of units, and this defect further affects the scientificity and rationality of power grid planning, making the planning scheme possibly deviate from the actual operation conditions.

[0004] 2. Lack of environmental protection consideration: In traditional rural power grid planning, the carbon emission data of various power generation technologies are often not comprehensively considered, which greatly reduces the environmental performance of power grid planning schemes and cannot effectively meet the urgent social needs for green and low-carbon development.

[0005] 3. Insufficient demand analysis: The changing trends of power demands in the agricultural industry fail to be deeply analyzed and grasped, resulting in a serious disconnection between the power grid planning scheme and the actual demands of agricultural production. Agricultural production has seasonal and cyclical characteristics, and power demands also fluctuate accordingly, but traditional planning methods often ignore this important factor. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and system for modern rural power grid planning to solve at least one of the above-mentioned existing technical problems.

[0007] In a first aspect, the present invention provides a method for modern rural power grid planning, which specifically includes:

[0008] Obtain the operation data and historical maintenance data of generating units within the power grid planning area, establish a unit efficiency prediction model based on the operation data, and establish a unit status evaluation model based on the historical maintenance data;

[0009] Determine the carbon emission data of various power generation technologies within the power grid planning area, where the carbon emission data includes the carbon emission data of power generation technologies of various types of renewable energy and the carbon emission data of traditional energy power generation technologies;

[0010] Obtain historical power demand data of different types of agricultural industries within the power grid planning area, and determine the power demand change trend of each type of agricultural industry according to the historical power demand data;

[0011] According to the prediction result of the unit efficiency prediction model, the evaluation result of the unit status evaluation model, the carbon emission data, and the power demand change trend of each type of agricultural industry, establish a power grid planning optimization model by using a multi-objective evolutionary algorithm, and the power grid planning optimization model is used to determine the optimal power grid planning scheme.

[0012] In a second aspect, the present invention provides a modern rural power grid planning system, which specifically includes:

[0013] A first planning module, configured to obtain the operation data and historical maintenance data of the generator sets within the power grid planning area, establish a unit efficiency prediction model according to the operation data, and establish a unit status evaluation model according to the historical maintenance data;

[0014] A second planning module, configured to determine the carbon emission data of various power generation technologies within the power grid planning area, and the carbon emission data includes the carbon emission data of power generation technologies of various renewable energy sources and the carbon emission data of traditional energy power generation technologies;

[0015] A third planning module, configured to obtain the historical power demand data of different types of agricultural industries within the power grid planning area, and determine the power demand change trend of each type of agricultural industry according to the historical power demand data;

[0016] A fourth planning module, configured to establish a power grid planning optimization model by using a multi-objective evolutionary algorithm according to the prediction result of the unit efficiency prediction model, the evaluation result of the unit status evaluation model, the carbon emission data, and the power demand change trend of each type of agricultural industry, and the power grid planning optimization model is used to determine the optimal power grid planning scheme.

[0017] In a third aspect, the present invention provides a computer device, including: a memory, a processor, and a computer program stored on the memory. When the computer program is executed on the processor, it implements the modern rural power grid planning method described in any one of the above methods.

[0018] In a fourth aspect, the present invention provides a computer-readable storage medium, characterized in that a computer program is stored thereon, and when the computer program is run by a processor, it implements the modern rural power grid planning method described in any one of the above methods.

[0019] Compared with the prior art, the present invention has at least one of the following technical effects:

[0020] 1. It improves the scientificity and rationality of power grid planning. By establishing a unit efficiency prediction model and a unit status evaluation model, accurate prediction and evaluation of the efficiency and status of generating units are achieved.

[0021] 2. It enhances the environmental protection performance of power grid planning schemes. By comprehensively considering the carbon emission data of various power generation technologies, optimization of power grid planning schemes in terms of environmental protection is realized.

[0022] 3. It strengthens the compatibility between power grid planning schemes and the actual needs of agricultural production. By deeply analyzing the changing trend of power demand in the agricultural industry, the power grid planning scheme becomes more in line with the actual needs of agricultural production, thereby improving the comprehensive benefits of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0024] Figure 1 is a schematic flowchart of a modern rural power grid planning method provided by an embodiment of the present invention;

[0025] Figure 2 is a schematic structural diagram of a modern rural power grid planning system provided by an embodiment of the present invention;

[0026] Figure 3 is a schematic structural diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0028] It should be understood that when used in the specification and appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0029] It should also be understood that the term "and / or" as used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0030] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "once" or "in response to determining" or "in response to detecting" depending on the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined" or "in response to determining" or "once [the described condition or event] is detected" or "in response to detecting [the described condition or event]" depending on the context.

[0031] In addition, in the description of the specification and appended claims of this application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0032] Reference to "one embodiment" or "some embodiments" etc. described in the specification of this application means that a specific feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in another way. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in another way.

[0033] In the embodiments of this application, the execution subject of the process includes a terminal device. The terminal device includes, but is not limited to: devices such as servers, computers, smartphones, and tablet computers that can execute the methods disclosed in this application. Figure 1 The flowchart of a modern rural power grid planning method disclosed in an embodiment of the present invention is shown and described in detail as follows:

[0034] S101, Obtain the operation data and historical maintenance data of the generating units in the power grid planning area, establish a unit efficiency prediction model according to the operation data, and establish a unit status evaluation model according to the historical maintenance data.

[0035] In this embodiment, in order to specifically implement the power grid planning, first, the operation data and historical maintenance data of all generating units in the power grid planning area are obtained, including operation parameters such as the power output, operation time, load rate, and fuel consumption of the generating units, as well as historical maintenance information such as the maintenance records, fault types, and maintenance time of the units.

[0036] Then, using these data, a unit efficiency prediction model and a unit status evaluation model were established respectively. For the unit efficiency prediction model, we adopted the Support Vector Regression (SVR) algorithm, with the running time and load rate as independent variables and the unit efficiency as the dependent variable for modeling. By preprocessing and feature selection of the operation data, the characteristic variables that mainly affect the unit efficiency were extracted, and the SVR model was trained to predict the unit efficiency under different operating conditions.

[0037] For the unit status evaluation model, the decision tree algorithm was used for modeling. First, the historical maintenance data was mined to analyze the correlation between the fault symptoms and fault causes of the generator set. Then, each performance index of the generator set was selected as the characteristic variable, and the unit status rating was used as the target variable to train the decision tree model to evaluate the unit status.

[0038] In this embodiment, by establishing the unit efficiency prediction model and the unit status evaluation model, the accurate prediction and evaluation of the efficiency and status of the generator sets in the power grid planning area were realized. Specifically, the unit efficiency prediction model can predict the efficiency value of the unit according to the input parameters under different operating conditions, providing an important reference basis for power grid planning. The unit status evaluation model can evaluate the status level of the unit according to the performance indexes of the generator set, helping power grid planners understand the health status of the unit, so as to formulate a more scientific and reasonable power grid planning scheme.

[0039] In some embodiments, in the above step S101, the establishment of the unit efficiency prediction model according to the operation data specifically includes:

[0040] The principal component analysis method is used to perform dimensionality reduction processing on the operation data to obtain the main characteristic variables;

[0041] According to the type of each unit, the main characteristic variables are divided into different subsets by the K-means clustering algorithm;

[0042] Taking the running time and load rate as independent variables and the unit efficiency as the dependent variable, a support vector regression model is used to establish a corresponding unit efficiency prediction model for each subset respectively.

[0043] In this embodiment, for the operation data, the principal component analysis method is used for dimensionality reduction processing to extract the main characteristic variables from the high-dimensional data, realizing data dimensionality reduction. By preprocessing the operation data, noise and outliers are removed to improve the data quality, preparing for the principal component analysis. According to the results of the principal component analysis, the first k principal components with the cumulative contribution rate reaching the preset threshold are selected as the main characteristic variables to achieve data dimensionality reduction. The K-means clustering algorithm is used to perform clustering analysis on the selected main characteristic variables, and the optimal number of clusters K is determined through iterative optimization; according to the clustering results, the characteristic variables of each unit type are divided into different subsets, and each subset reflects different unit operation modes.

[0044] Historical operation data is obtained according to the operation time and load rate and used as the training data set of the support vector regression model; for each subset, the corresponding historical operation data is divided into a training set and a test set. The training set is used for model training, and the test set is used for model evaluation; the grid search and cross-validation methods are used to optimize the hyperparameters of the support vector regression model to obtain the optimal model parameter configuration; the optimized support vector regression model is used to train each subset to establish a unit efficiency prediction model; the operation time and load rate are used as independent variables, and the unit efficiency is used as the dependent variable and input into the trained prediction model; the unit efficiency prediction values of each subset at the given operation time and load rate are calculated through the prediction model; the efficiency prediction results of each subset are summarized to obtain the efficiency prediction situation of the overall unit.

[0045] In some embodiments, in step S101 above, the establishing a unit status evaluation model according to the historical maintenance data specifically includes:

[0046] The association rule mining algorithm is used to analyze the historical maintenance data to determine the association relationship between the fault symptoms and fault causes of the generator set;

[0047] Taking the performance indicators of the generator set as characteristic variables and the unit status rating of the generator set as the target variable, according to the association relationship between the fault symptoms and fault causes of the generator set, a decision tree algorithm is used to establish a unit status evaluation model.

[0048] In this embodiment, based on the historical maintenance data of the generator set, data preprocessing technology is used to clean, integrate, transform and reduce the data to improve data quality. Through feature engineering, relevant features such as fault symptoms and fault causes are extracted from the preprocessed historical maintenance data to construct a feature set. An association rule mining algorithm, such as the Apriori algorithm or the FP-Growth algorithm, is used to mine the association rules between fault symptoms and fault causes. According to the mined association rules, the fault diagnosis rules are determined to form a rule base. If a certain fault symptom occurs, the possible cause of the fault is determined according to the association rules. For the new generator set fault data, the fault symptom features are extracted, and the cause of the fault is determined by matching the fault diagnosis rule base. If the cause of the fault cannot be diagnosed according to the existing rules, the fault data is added to the historical maintenance data set, the association rules are re-mined, and the fault diagnosis rule base is updated.

[0049] The historical operation data of the generator set, including the performance index data and the corresponding unit status rating data, are obtained as the training data set. The historical operation data obtained are preprocessed, abnormal data are eliminated, missing data are filled, and the data are normalized. According to the correlation between the fault symptoms and fault causes of the generator set, the performance indicators related to the fault are selected from the preprocessed data as feature variables. The unit status rating of the generator set is used as the target variable, and the preprocessed data is divided into a training set and a test set. The training set data is trained using the decision tree algorithm, and the decision tree model is constructed by recursively dividing the feature variables until the preset stop condition is met. The constructed decision tree model is used to predict the test set data to obtain the predicted unit status rating results, which are compared with the actual unit status rating, and the accuracy of the model and other evaluation indicators are calculated. If the accuracy and other evaluation indicators of the model meet the preset thresholds, the decision tree model is used as the final unit status evaluation model; otherwise, the parameters of the decision tree algorithm are adjusted, and the training and testing are re-performed until an evaluation model that meets the requirements is obtained.

[0050] S102, determining carbon emission data of various power generation technologies in the power grid planning area, wherein the carbon emission data includes carbon emission data of various types of renewable energy power generation technologies and carbon emission data of traditional energy power generation technologies.

[0051] In this embodiment, in order to comprehensively consider environmental factors during the grid planning process, the carbon emission data of various power generation technologies in the grid planning area are determined. This process first involves the classification of all power generation facilities in the area, including renewable energy power generation facilities such as wind power, solar power, and hydropower, as well as traditional energy power generation facilities such as coal, oil, and natural gas.

[0052] Specifically, first, through the energy statistical data released by the government, the annual reports of power generation enterprises, the databases of authoritative institutions such as the International Energy Agency, and on-site investigations, etc., detailed operation data of various power generation technologies and carbon emission-related information are collected.

[0053] Then, according to different types of power generation technologies, corresponding carbon emission calculation methods are adopted. For renewable energy power generation technologies, such as wind power and photovoltaic power, since they do not produce direct carbon dioxide emissions during power generation, the carbon emissions mainly come from life cycle links such as equipment manufacturing, transportation, installation, and maintenance. We use the life cycle assessment (LCA) method to calculate their carbon emissions. For traditional energy power generation technologies, such as coal power generation and gas power generation, the emission factor method is adopted, and the carbon dioxide emissions are calculated according to the fuel consumption and the corresponding emission factors.

[0054] Integrate the carbon emission data of various power generation technologies collected to form a complete carbon emission database. Subsequently, we use data analysis tools to deeply mine the database, analyze the carbon emission characteristics, change trends, and influencing factors of various power generation technologies, and provide a scientific basis for power grid planning.

[0055] In this embodiment, by determining the carbon emission data of various power generation technologies within the power grid planning area, the environmental protection performance of various power generation technologies can be understood more clearly, so as to give priority to low-carbon and environmentally friendly power generation methods in power grid planning and reduce the negative impact of power grid operation on the environment. The carbon emission data provides important decision-making support for power grid planning. Based on these data, the environmental costs of different power generation schemes can be evaluated, and the best scheme that meets both power demand and environmental protection requirements can be selected.

[0056] In some embodiments, in the above step S102, the determining the carbon emission data of various power generation technologies within the power grid planning area specifically includes:

[0057] Obtain the meteorological data within the power grid planning area and the energy resource data of different types of renewable energy, and establish an association relationship between the energy resource data of each type of renewable energy and the meteorological data respectively to form a renewable energy production prediction model;

[0058] Establish a power generation model for each type of renewable energy, and determine the power generation data of each type of renewable energy according to the power generation model and the renewable energy production prediction model;

[0059] Obtain the traditional energy consumption data within the power grid planning area, and establish an energy network diagram according to the traditional energy consumption data. The energy network diagram includes the association relationship between electricity and other different forms of traditional energy;

[0060] Determine the carbon emission factors of various power generation technologies in each energy supply chain link, and determine the carbon emission data of various power generation technologies according to the carbon emission factors, the power generation data, and the energy network diagram.

[0061] In this embodiment, meteorological data within the power grid planning area is obtained, including temperature, humidity, wind speed, radiation intensity, etc., and at the same time, historical output data of renewable energy sources such as wind energy and solar energy within this area is obtained. The obtained meteorological data is preprocessed to remove outliers and missing values, and data standardization is performed to unify data of different dimensions to the same scale. The correlation analysis method, such as Pearson correlation coefficient or mutual information, is used to analyze the correlation between meteorological conditions and the output of renewable energy sources, and the key meteorological factors affecting the output of renewable energy sources are determined. According to the results of the correlation analysis, an appropriate machine learning algorithm, such as support vector machine, random forest, or neural network, is selected to construct a renewable energy production prediction model. The preprocessed meteorological data and historical output data of renewable energy sources are divided into a training set and a test set. The training set data is used to train the prediction model, and the test set data is used to evaluate the prediction performance of the model. The trained prediction model is optimized by adjusting model parameters, feature selection, and other methods to improve the prediction accuracy and stability of the model.

[0062] Obtain the historical power generation data and influencing factor data of various types of renewable energy sources, and preprocess the data using data cleaning and feature engineering methods to remove outliers and redundant information, and extract key feature parameters. According to the characteristics of different types of renewable energy sources, a suitable machine learning algorithm, such as support vector machine, random forest, or neural network, is selected to construct a power generation prediction model for each type of renewable energy source. During the training process of the power generation prediction model, methods such as cross-validation are used to optimize the model, and the prediction accuracy and generalization ability of the model are improved by adjusting model hyperparameters, feature selection, and other means. The power generation prediction model and the production prediction model are fused, and factors such as power generation efficiency and energy loss are comprehensively considered, and the model weights are dynamically adjusted to generate more accurate and comprehensive estimated power generation data of renewable energy sources. According to the estimated power generation data and the actual power generation data, the prediction model is continuously evaluated and optimized, and the prediction performance and adaptability of the model are continuously improved through methods such as incremental learning and online learning.

[0063] By docking with the power grid planning department, historical consumption data of various traditional energy sources in the region are obtained, including consumption volume and consumption time series data of energy sources such as electricity, coal, oil, and natural gas. The obtained traditional energy consumption data are preprocessed to clean outliers and missing values, and the data are normalized to a unified time granularity and numerical range for subsequent correlation analysis and network construction. Correlation analysis algorithms such as Pearson correlation coefficient or mutual information are used to calculate the correlation between different energy consumption time series, and the association strength between electric energy and other traditional energy sources is mined. According to the association strength between energy sources, a weighted undirected graph is constructed, where nodes represent different energy forms and the weights of edges represent the association strength between energy sources. Network embedding algorithms such as DeepWalk are used to learn the low-dimensional vector representation of nodes to capture the high-order association relationships between energy sources. Community discovery is carried out on the energy network graph, and the modularity-optimized Louvain algorithm is used to divide closely associated energy nodes into the same community to mine the internal structure and laws of energy consumption. The energy network graph is visually displayed, and the size and color of nodes are set according to the importance of nodes and community membership, and the thickness of edges is set according to the weights of edges to intuitively present the intricate association relationships between electric energy and other energy sources.

[0064] According to the pre-established energy network graph, carbon emission factor data of various power generation technologies in different links of the energy supply chain are obtained; the power generation volume data of various power generation technologies over a period of time are obtained, and the corresponding supply chain links of various power generation technologies are determined according to the energy network graph; for each type of power generation technology, its power generation volume data are multiplied by the carbon emission factor of the corresponding supply chain link to obtain the carbon emissions of this power generation technology in this link; the carbon emissions of each type of power generation technology in all supply chain links are added up to obtain the total carbon emissions data of this power generation technology; according to the total carbon emissions data of various power generation technologies, clustering algorithms are used to classify power generation technologies to obtain power generation technology categories with high, medium, and low carbon emissions; for the power generation technology category with high carbon emissions, the relationship between its carbon emissions, power generation volume, and energy supply chain links is determined through regression analysis to judge the main influencing factors of carbon emissions; according to the main influencing factors of carbon emissions, optimization algorithms are used to determine the strategies for adjusting the power generation structure and optimizing the supply chain to obtain a power generation technology combination and energy supply method that can effectively reduce carbon emissions.

[0065] Furthermore, establishing an association relationship between the energy resource data of each type of renewable energy and the terrain data to form a renewable energy production capacity prediction model specifically includes:

[0066] Using the principal component analysis method to perform feature processing on the meteorological data to extract key meteorological factors related to the production capacity of renewable energy;

[0067] Analyze the time series data of the meteorological factors through an autoregressive integrated moving average model to obtain the dynamic change characteristics of the meteorological factors;

[0068] According to the dynamic change characteristics and the energy resource data of each type of renewable energy, use a long short-term memory network model to predict the renewable energy production capacity, and obtain the renewable energy production capacity prediction models for each type of renewable energy.

[0069] In this embodiment, obtain the original meteorological parameters related to the renewable energy production capacity according to the meteorological data, perform data cleaning and preprocessing operations on the original meteorological parameters to obtain a preprocessed meteorological parameter data set. Use the principal component analysis method to perform feature analysis on the preprocessed meteorological parameter data set. By calculating the correlation coefficient matrix between each meteorological parameter and performing eigenvalue decomposition on the correlation coefficient matrix, obtain the principal component eigenvectors of the meteorological parameters. According to the variance contribution rate of each component in the principal component eigenvector, select the top N principal components with larger variance contribution rates as the key meteorological factors, where N is a preset positive integer threshold. Determine whether the correlation between the key meteorological factors and the renewable energy production capacity is greater than a preset correlation threshold. If so, determine the key meteorological factor as the key meteorological factor related to the renewable energy production capacity.

[0070] For the time series data of the meteorological factors, capture the time correlation of the data through an autoregressive model, use an integration model to eliminate the non-stationarity of the data, and combine a moving average model to smooth the random fluctuations of the data, and comprehensively establish a time series model of the meteorological factor changes. Use the established autoregressive integrated moving average model to fit and predict the time series data of the meteorological factors to obtain the dynamic change characteristics of the meteorological factors.

[0071] According to the renewable energy types, obtain the historical production capacity data of each type of energy and the data of the key factors affecting the production capacity; preprocess the obtained historical data, extract the dynamic change characteristics of the energy production capacity to form time series data; for the preprocessed time series data, use a long short-term memory network model for training to establish a production capacity prediction model; during the training process of the production capacity prediction model, optimize the prediction performance of the model by setting model hyperparameters and evaluation indicators; use the trained long short-term memory network model, combined with the real-time collected energy data, to predict the renewable energy production capacity in the future for a period of time; summarize the production capacity prediction results of each type of renewable energy to form an overall renewable energy production capacity prediction report; according to the production capacity prediction results, optimize the scheduling and energy storage strategies of renewable energy to improve the utilization efficiency of renewable energy.

[0072] Exemplarily, first, perform principal component analysis on the meteorological data within the power grid planning area to extract 10 key meteorological factors closely related to renewable energy production capacity, such as temperature, humidity, wind speed, sunshine hours, etc. Then, use the autoregressive integrated moving average model to analyze the time series data of these meteorological factors over the past 5 years to obtain their dynamic change characteristics. For example, the temperature shows an obvious seasonal change trend, and the wind speed change is relatively random, etc. Next, match the dynamic change characteristic data of the meteorological factors with the historical data of 5 types of renewable energy such as wind energy and solar energy in this area to construct a training data set. Finally, use the long short-term memory network model to learn the training data, set the number of hidden layers to 3 layers, the number of neurons in each layer to 128, 256, and 128 respectively, the learning rate to 0.1, and obtain the production prediction models of various types of renewable energy after training for 1000 epochs. After evaluation, this model can relatively accurately predict the output of various types of renewable energy in the future for a period of time, and control the average error rate within 8%, providing a reliable basis for the optimal dispatching of the power grid.

[0073] Further, establishing the energy network diagram according to the traditional energy consumption data specifically includes:

[0074] Determine the node sets of the energy production unit, energy conversion unit, and energy consumption unit according to the traditional energy consumption data;

[0075] Construct an initial energy network diagram according to the node sets, and the edges of the initial energy network diagram represent the energy flow paths;

[0076] Obtain the association strength between different nodes by analyzing the energy flow paths;

[0077] Optimize the initial energy network diagram using the minimum spanning tree algorithm and calculate the weights of each edge;

[0078] Adjust the initial energy network diagram to the target energy network diagram according to the association strength between different nodes and the weights of each edge.

[0079] In this embodiment, according to the traditional energy consumption data, the energy consumption of each node in the energy system is obtained and stored in the energy consumption database. For the data in the energy consumption database, a clustering algorithm is used to cluster each node to obtain an energy production unit node set, an energy conversion unit node set, and an energy consumption unit node set. According to the energy production unit node set, the actual energy production equipment corresponding to each production unit node is determined, and the energy production data of each production equipment is obtained. According to the energy conversion unit node set, the actual energy conversion equipment corresponding to each conversion unit node is determined, and the energy conversion efficiency data of each conversion equipment is obtained. According to the energy consumption unit node set, the actual energy consumption equipment corresponding to each consumption unit node is determined, and the energy demand data of each consumption equipment is obtained.

[0080] Obtain a given node set, determine the attribute information of each node, including node type, node position coordinates, etc. According to the node attribute information, use graph theory algorithms to construct the topological structure of the initial energy network diagram and generate the edges between nodes. For each edge, associate the attributes of energy flow, including parameters such as energy type, flow direction, and flow rate. Through the edge weight calculation of the network diagram, judge the feasible paths of energy flow between nodes and filter out the optimized transmission paths. Based on the network flow algorithm, calculate the optimal energy scheduling scheme under different scenarios, balance the load distribution, and improve the energy utilization efficiency. According to the topological structure and node information of the initial energy network diagram, establish a mathematical model of the network, and quantify the attributes of nodes and edges as weight parameters.

[0081] Use the Prim algorithm or Kruskal algorithm to implement the construction of the minimum spanning tree, and gradually generate the minimum cost tree by selecting the edge with the smallest weight through a greedy strategy. During the construction of the minimum spanning tree, dynamically update the connectivity status of nodes, judge whether the selected edge will form a loop, and avoid the failure of the spanning tree. Obtain the constructed minimum spanning tree, extract the topological structure and node connection relationship of the tree, and use it as the optimized energy network topology. According to the edge weight information of the minimum spanning tree, calculate the total transmission cost of the network, and evaluate the effect and benefit of network optimization. Feed back the structure and cost information of the minimum spanning tree to the scheduling and control system of the energy network, and realize the efficient distribution and balance of energy according to the optimized topology. Continuously monitor the load changes and node status of the energy network. When the network topology changes, re-trigger the minimum spanning tree algorithm to adaptively optimize the network and maintain the stable and efficient operation of the energy system.

[0082] According to the association strength between nodes in the initial energy network diagram, obtain the initial association degree weight matrix between nodes; according to the weights of the edges in the initial energy network diagram, obtain the initial weight matrix of the edges in the network; through a deep learning algorithm, train the initial association degree weight matrix and the initial weight matrix to obtain an adjusted node association degree weight matrix and edge weight matrix; according to the adjusted node association degree weight matrix, update the association strength between nodes; according to the adjusted edge weight matrix, update the weights of the edges in the network; according to the updated node association strength and edge weights, reconstruct the energy network topology diagram; determine whether the reconstructed energy network topology diagram meets the requirements of the target energy network diagram. If it meets the requirements, output the topology diagram as the target energy network diagram. If it does not meet the requirements, return to the third step to continue training and adjustment.

[0083] Exemplarily, first, according to the traditional energy consumption data, use the clustering analysis algorithm to divide the energy production units, energy conversion units, and energy consumption units into different node sets. Among them, the energy production units include thermal power plants, hydropower plants, etc., the energy conversion units include substations, distribution stations, etc., and the energy consumption units include industrial parks, commercial areas, residential areas, etc. Then, construct an initial energy network diagram based on the node sets. The edges in the diagram represent the energy flow paths between different nodes. By analyzing factors such as the energy transmission volume and distance on the energy flow path, calculate the association strength between different nodes. For example, the association strength between a certain thermal power plant and a nearby substation is 8, while the association strength with a far - away distribution station is 4. Next, use the minimum spanning tree algorithm to optimize the initial energy network diagram and calculate the weight of each edge. This algorithm constructs a minimum spanning tree by continuously selecting the edge with the smallest weight, and finally obtains an optimized energy network diagram. During the optimization process, factors such as the distance between nodes and the energy transmission efficiency need to be comprehensively considered to reasonably set the weights of the edges. For example, the weight of an edge with a relatively short distance and high transmission efficiency can be set to 9, while the weight of an edge with a relatively long distance and low transmission efficiency can be set to 2. Finally, according to the association strength between different nodes and the optimized edge weights, adjust the initial energy network diagram to form a target energy network diagram. During the adjustment process, it is necessary to take into account the energy supply - demand balance and network stability, appropriately delete the edges with relatively weak association strength or small weights, and retain or appropriately increase the edges with relatively strong association strength or large weights. For example, edges with an association strength lower than 3 and a weight lower than 4 can be deleted, and edges with an association strength higher than 7 and a weight higher than 8 can be retained or increased. Through the above - mentioned analysis and optimization process, a target energy network diagram with a reasonable structure and high efficiency is finally formed, providing an important reference for the optimal operation of the energy system.

[0084] S103, obtain the historical electricity demand data of different types of agricultural industries in the power grid planning area, and determine the electricity demand change trend of each type of agricultural industry according to the historical electricity demand data.

[0085] In this embodiment, historical power demand data of different types of agricultural industries (such as planting, animal husbandry, fishery, agricultural product processing, etc.) within the power grid planning area are obtained, which usually include monthly, quarterly or annual power consumption, covering multiple time periods for trend analysis. Then, time series analysis methods, such as trend line fitting, seasonal decomposition, etc., are used to perform trend analysis on the power demand data of various agricultural industries. By calculating indicators such as the growth rate and periodic fluctuations of power demand, the long-term trends and seasonal characteristics of the power demand of different agricultural industries are identified. To visually display the analysis results, the change trends of the power demand of various agricultural industries can be presented in the form of charts (such as line charts, bar charts, trend lines, etc.), which can not only display the absolute value changes of power demand, but also clearly show the growth rate and fluctuations.

[0086] In this embodiment, by determining the change trends of the power demand of different types of agricultural industries within the power grid planning area, it can provide strong decision-making support for power grid planning, power facility construction, and the development of agricultural industries. According to the change trends of power demand, future power demand can be reasonably predicted to ensure the supply capacity and stability of the power grid.

[0087] In some embodiments, in the above step S103, the determining the change trends of the power demand of each type of agricultural industry according to the historical power demand data specifically includes:

[0088] Using an autoregressive integrated moving average model to perform data preprocessing and trend analysis on the historical energy demand data to obtain the time change characteristics of the energy demand of each type of agricultural industry;

[0089] Obtaining weather forecast data, crop growth cycle data, and livestock growth stage data to form multi-source heterogeneous data;

[0090] Using the principal component analysis method to perform dimensionality reduction processing on the multi-source heterogeneous data to obtain the dimensionality-reduced multi-source heterogeneous data;

[0091] Using the Kalman filter algorithm to fuse the dimensionality-reduced multi-source heterogeneous data to obtain the target multi-source heterogeneous data;

[0092] Taking the time change characteristics of the energy demand of each type of agricultural industry and the target multi-source heterogeneous data as input variables, and taking the energy demand of each type of agricultural industry as the output variable, and using a support vector machine regression model to construct an energy demand feature dataset;

[0093] Based on the above-mentioned energy demand characteristic dataset, the long short-term memory neural network algorithm is adopted to model and predict the energy demand of each type of agricultural industry, and the changing trend of the electricity demand of each type of agricultural industry is obtained.

[0094] In this embodiment, the historical energy demand data of each agricultural industry is obtained, and data cleaning and correction are performed on the missing values and outliers to obtain complete and accurate time series data. An autoregressive integrated moving average model is used to model the processed historical energy demand data, and the autoregressive coefficient, integration order, and moving average term coefficient of the energy demand of each agricultural industry are obtained through model parameter estimation. According to the estimated model parameters, trend analysis is performed on the energy demand data of each agricultural industry to determine whether there are obvious growth trends, periodic fluctuations, or random disturbances, and the corresponding time-varying characteristics are extracted.

[0095] Weather forecast data, crop growth cycle data, and livestock and poultry growth stage data are obtained to form an initial multi-source heterogeneous dataset; preprocessing is performed on the initial multi-source heterogeneous dataset, including operations such as data cleaning, missing value processing, and data standardization, to obtain a preprocessed multi-source heterogeneous dataset; principal component analysis is used to perform feature extraction and dimensionality reduction on the preprocessed multi-source heterogeneous dataset to obtain a dimensionality-reduced multi-source heterogeneous dataset; according to the crop growth cycle data and livestock and poultry growth stage data, a state transition matrix and an observation matrix are constructed, and the initial state and covariance matrix are set; the dimensionality-reduced multi-source heterogeneous dataset is input into the Kalman filter model, and through recursive calculation and state update, the fused multi-source heterogeneous data is obtained; post-processing is performed on the fused multi-source heterogeneous data, including operations such as data inverse standardization and result visualization, to obtain the final target multi-source heterogeneous dataset.

[0096] Data fusion technology is adopted to fuse the time series characteristics of agricultural industry energy consumption with multi-source heterogeneous data to construct an energy demand characteristic dataset. The energy demand characteristic dataset is divided into a training set and a test set. The training set is used to train the support vector machine regression model, and the test set is used to evaluate the model performance. Methods such as grid search and cross-validation are used to optimize the hyperparameters of the support vector machine regression model to improve the prediction accuracy of the model. The trained support vector machine regression model is used to predict the test set to obtain the energy demand prediction results of various agricultural industries and evaluate the generalization ability of the model.

[0097] According to the energy demand characteristic dataset, preprocess the data, including data cleaning, missing value handling, outlier handling, etc., to ensure data quality. For the preprocessed dataset, extract the key attributes reflecting the energy demand characteristics of the agricultural industry and construct an energy demand feature vector. Adopt the long short-term memory neural network algorithm to design a network structure including an input layer, an LSTM layer, and an output layer, where the LSTM layer can capture the long-term dependencies of the energy demand time series data. Input the constructed energy demand feature vector into the long short-term memory neural network model, and by setting appropriate network parameters, train the model to fit the changing trend of the energy demand of the agricultural industry. During the model training process, adopt strategies such as early stopping to avoid overfitting problems and improve the generalization ability of the model. At the same time, optimize the model hyperparameters through methods such as grid search to improve the prediction accuracy. Use the trained long short-term memory neural network model to predict the electricity demand of different types of agricultural industries and obtain the changing trend of the electricity demand of each type of agricultural industry in the future for a period of time.

[0098] S104. According to the prediction results of the unit efficiency prediction model, the evaluation results of the unit status evaluation model, the carbon emission data, and the changing trends of the electricity demands of various types of agricultural industries, establish a grid planning optimization model using a multi-objective evolutionary algorithm. The grid planning optimization model is used to determine the optimal grid planning scheme.

[0099] In this embodiment, in order to comprehensively consider multiple factors such as unit efficiency, unit status, carbon emissions, and the changing trends of agricultural industry electricity demands, a grid planning optimization model is established using a multi-objective evolutionary algorithm. Specifically, first, according to the aforementioned prediction results and evaluation data, the objective functions of the grid planning optimization model are defined. These objective functions include, but are not limited to: maximizing grid efficiency: based on the prediction results of the unit efficiency prediction model, by optimizing the grid structure and operation strategies, make the overall operation efficiency of the grid reach the maximum; minimizing carbon emissions: according to the carbon emission data of various power generation technologies, optimize the power generation portfolio and grid layout to reduce the carbon emissions during the operation of the grid; meeting the electricity demands of the agricultural industry: according to the changing trends of the electricity demands of various types of agricultural industries, ensure that the grid planning scheme can meet the electricity demands for future agricultural production; improving the stability and reliability of the grid: combining the evaluation results of the unit status evaluation model, optimize the grid redundancy and emergency response capabilities to improve the stability and reliability of the grid.

[0100] To ensure the feasibility and rationality of the power grid planning scheme, corresponding constraint conditions are further set, such as power grid capacity limitation, capital budget limitation, technical feasibility limitation, etc. A multi-objective evolutionary algorithm (such as NSGA-II, SPEA2, etc.) is used to solve the power grid planning problem. This algorithm searches for the optimal solution set (Pareto optimal solution set) of multiple objective functions in the solution space by simulating the natural evolution process. During the operation of the algorithm, the solution set is evaluated and screened according to the objective functions and constraint conditions, gradually approaching the optimal solution. After multiple rounds of iteration, the algorithm will output a set of Pareto optimal solution sets. A decision analysis tool (such as a decision matrix, radar chart, etc.) is used to comprehensively analyze these solution sets, considering the trade-off relationship between various objectives, and select the optimal solution that best meets the power grid planning requirements and objectives.

[0101] In this embodiment, by using a multi-objective evolutionary algorithm to establish a power grid planning optimization model, multiple factors such as unit efficiency, unit status, carbon emissions, and power demand of the agricultural industry can be comprehensively considered, and the comprehensive benefit of the power grid planning scheme can be maximized. This not only helps to improve the economy and environmental protection of power grid operation, but also better meets the power demand of agricultural production.

[0102] In some embodiments, in the above step S104, according to the prediction result of the unit efficiency prediction model, the evaluation result of the unit status evaluation model, the carbon emission data, and the power demand change trend of each type of agricultural industry, using a multi-objective evolutionary algorithm to establish a power grid planning optimization model specifically includes:

[0103] Taking minimizing the total cost, maximizing the energy utilization rate, and minimizing the carbon emissions as the objective functions, and taking the prediction result of the unit efficiency prediction model, the evaluation result of the unit status evaluation model, the carbon emission data, and the power demand change trend of each type of agricultural industry as input parameters, to establish a power grid planning optimization model;

[0104] Using a multi-objective evolutionary algorithm based on Pareto sorting to solve the multi-objective optimization model to obtain a Pareto optimal solution set;

[0105] For each solution in the Pareto optimal solution set, using conditional value at risk as a risk measurement index to quantify the uncertainty of renewable energy and obtain the risk value of each solution;

[0106] According to the risk value, determine the optimal solution from the Pareto optimal solution set, and the optimal solution is the optimal power grid planning scheme.

[0107] In this embodiment, historical power demand data of various types of agricultural industries are obtained, and the changing trends of power demand for various types of agricultural industries in a future period are predicted through time series analysis algorithms. Real-time operation data of generator sets are collected, including power generation and fuel consumption, etc., and the power generation efficiency of each unit under different operating conditions is predicted according to the unit efficiency prediction model. Status parameters of each generator set are obtained, such as the operating duration and failure rate of the unit, and the health status and reliability of each unit are evaluated through the unit status assessment model. Carbon emission data of each generator set are collected, and the carbon emission intensity per unit of electricity is calculated, which is used as one of the optimization objectives. The power demand prediction results, unit efficiency prediction results, unit status assessment results, and carbon emission data of various types of agricultural industries are used as inputs to construct a multi-objective optimization model, and the objective functions include minimizing the total cost, maximizing the energy utilization rate, and minimizing the carbon emissions.

[0108] The multi-objective evolutionary algorithm based on Pareto ranking (such as NSGA-II) is used to solve the model. The algorithm searches for the optimal solution set of multiple objective functions in the solution space by simulating the natural evolution process. After multiple rounds of iteration, the algorithm outputs a set of Pareto optimal solution sets, and these solutions achieve an optimal trade-off among the objective functions. For each solution in the Pareto optimal solution set, the conditional value at risk (CVaR) is used as a risk measurement index to quantify the uncertainty of the power generation of renewable energy sources (such as wind energy and solar energy). CVaR measures the potential loss caused by the insufficient power generation of renewable energy sources at a certain confidence level.

[0109] Calculate the risk value of each solution to evaluate its robustness in the face of the uncertainty of renewable energy. According to the trade-off between the risk value and the original objective function, the optimal solution is selected from the Pareto optimal solution set. This optimal solution not only performs well in each objective function but also can better cope with the uncertainty of renewable energy. Specifically, calculate the risk value of each solution in the Pareto optimal solution group; sort the Pareto optimal solution group in ascending order of the risk value; obtain the solution with the smallest risk value after sorting and determine it as the optimal solution; generate a detailed power grid planning scheme according to the power grid topology structure, equipment parameters, etc. included in the optimal solution; use the Monte Carlo simulation algorithm to conduct a multi-scenario risk assessment of the power grid planning scheme and determine whether the risk value exceeds the threshold; if the risk value exceeds the threshold, return to select the solution with the second smallest risk value and regenerate the planning scheme; if the risk value does not exceed the threshold, determine the current power grid planning scheme as the final optimal planning scheme.

[0110] Exemplarily, in power grid planning and optimization, the objective functions are to minimize the total cost, maximize the energy utilization rate, and minimize the carbon emissions. First, the unit efficiency prediction model is used to predict the efficiency of each generating unit in the future for a period of time. For example, the average efficiency of a thermal power unit in the next year is predicted to be 35%. Then, the unit status evaluation model is adopted to evaluate the operating status of each unit and obtain their health indexes. For example, the health index of a hydropower unit is 85. Meanwhile, the carbon emission data of each unit is collected, and the annual average carbon emission reduction of a wind farm is calculated to be 20,000 tons by the wind0 model. In addition, according to the prediction result of the electricity demand of the agricultural industry, the electricity consumption of the fishery will increase at an annual rate of 4% in the next five years. The above data is input into the power grid planning and optimization model, and the multi-objective evolutionary algorithm NSGA-III is used to solve it. When the population size is 100 and the number of evolutionary generations is 500, 100 Pareto optimal solutions are obtained. Then, the conditional value at risk CVaR is selected as the risk measurement index. When the confidence level is 95%, the CVaR value of each optimal solution is calculated, which represents the economic loss risk brought by the uncertainty of the renewable energy power generation. Finally, the solution with the minimum CVaR value is selected from the Pareto optimal solution set as the optimal planning scheme. This scheme considers multiple objectives of power grid planning and quantifies the uncertainty risk of renewable energy, and can guide the optimal construction and operation of the power grid.

[0111] Referring to Figure 2 , an embodiment of the present invention provides a modern rural power grid planning system 2, and the system 2 specifically includes:

[0112] The first planning module 201 is configured to obtain the operation data and historical maintenance data of the generating units in the power grid planning area, establish a unit efficiency prediction model according to the operation data, and establish a unit status evaluation model according to the historical maintenance data;

[0113] The second planning module 202 is configured to determine the carbon emission data of various power generation technologies in the power grid planning area, and the carbon emission data includes the carbon emission data of the power generation technologies of various types of renewable energy and the carbon emission data of traditional energy power generation technologies;

[0114] The third planning module 203 is configured to obtain the historical electricity demand data of different types of agricultural industries in the power grid planning area, and determine the electricity demand change trend of each type of agricultural industry according to the historical electricity demand data;

[0115] The fourth planning module 204 is configured to establish a grid planning optimization model by using a multi-objective evolutionary algorithm according to the prediction result of the unit efficiency prediction model, the evaluation result of the unit status evaluation model, the carbon emission data, and the power demand change trend of each type of agricultural industry. The grid planning optimization model is used to determine an optimal grid planning scheme.

[0116] It can be understood that the content in the embodiment of the modern rural power grid planning method as Figure 1 shown is applicable to the embodiment of the modern rural power grid planning system. The functions specifically implemented in the embodiment of the modern rural power grid planning system are the same as those in the embodiment of the modern rural power grid planning method as Figure 1 shown, and the beneficial effects achieved are also the same as those in the embodiment of the modern rural power grid planning method as Figure 1 shown.

[0117] It should be noted that the information interaction, execution process, etc. between the above systems, due to being based on the same concept as the method embodiment of the present invention, for the specific functions and the technical effects brought, reference can be specifically made to the method embodiment part, and details are not described herein again.

[0118] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the system is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working process of the units and modules in the above system can refer to the corresponding process in the foregoing method embodiment, and details are not described herein again.

[0119] Referring to Figure 3 , an embodiment of the present invention further provides a computer device 3, including: a memory 302, a processor 301, and a computer program 303 stored on the memory 302. When the computer program 303 is executed on the processor 301, the modern rural power grid planning method as described in any one of the above methods is implemented.

[0120] The computer device 3 may be a computing device such as a desktop computer, a notebook, a handheld computer, and a cloud server. The computer device 3 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art can understand that Figure 3 merely examples of the computer device 3, which do not constitute a limitation on the computer device 3, may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0121] The so-called processor 301 may be a central processing unit (CPU), and the processor 301 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0122] The memory 302 may be an internal storage unit of the computer device 3 in some embodiments, such as the hard disk or memory of the computer device 3. The memory 302 may also be an external storage device of the computer device 3 in other embodiments, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device 3. Further, the memory 302 may also include both the internal storage unit and the external storage device of the computer device 3. The memory 302 is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program, etc. The memory 302 may also be used to temporarily store data that has been output or will be output.

[0123] The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it implements the modern rural power grid planning method as described in any one of the above methods.

[0124] In this embodiment, 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, to implement all or part of the processes in the method of the above embodiment in this application, a computer program can be used to instruct relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the photographing device / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be an electrical carrier signal and a telecommunication signal.

[0125] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0126] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0127] In the embodiments disclosed in this application, it should be understood that the disclosed device / terminal device and method can be implemented in other ways. For example, the device / terminal device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, 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 displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical or other form.

[0128] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed over multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

Claims

1. A modern rural power grid planning method, characterized in that: The method specifically comprises: Acquire the operating data and historical maintenance data of the generator sets in the power grid planning area, establish a unit efficiency prediction model based on the operating data, and establish a unit status assessment model based on the historical maintenance data; Determine the carbon emission data of various power generation technologies in the power grid planning area, wherein the carbon emission data includes carbon emission data of various types of renewable energy power generation technologies and carbon emission data of traditional energy power generation technologies; Obtain historical power demand data of different types of agricultural industries in the power grid planning area, and determine the power demand change trend of each type of agricultural industry based on the historical power demand data; A multi-objective evolutionary algorithm is used to establish a power grid planning optimization model based on the prediction results of the unit efficiency prediction model, the evaluation results of the unit status evaluation model, the carbon emission data and the power demand change trend of various types of agricultural industries. The power grid planning optimization model is used to determine the optimal power grid planning scheme; Among them, according to the prediction results of the unit efficiency prediction model, the evaluation results of the unit status evaluation model, the carbon emission data and the change trend of electricity demand of various types of agricultural industries, a multi-objective evolutionary algorithm is used to establish a power grid planning optimization model, which specifically includes: Taking minimizing total cost, maximizing energy utilization and minimizing carbon emissions as objective functions, and taking the prediction results of the unit efficiency prediction model, the evaluation results of the unit status evaluation model, the carbon emission data and the power demand change trends of various types of agricultural industries as input parameters, a power grid planning optimization model is established; A multi-objective evolutionary algorithm based on Pareto sorting is used to solve the multi-objective optimization model and obtain the Pareto optimal solution group; For each solution in the Pareto optimal solution group, conditional risk value is used as a risk measurement indicator to quantify the uncertainty of renewable energy and obtain the risk value of each solution; According to the risk value, an optimal solution is determined from the Pareto optimal solution group, and the optimal solution is an optimal power grid planning scheme.

2. The method according to claim 1, characterized in that The step of establishing a unit efficiency prediction model according to the operation data specifically includes: The operating data is processed by principal component analysis to reduce the dimension and obtain the main characteristic variables; According to the type of each unit, the main characteristic variables are divided into different subsets by using a K-means clustering algorithm; With the operating time and load rate as the independent variable and the unit efficiency as the dependent variable, the support vector regression model is used to establish the corresponding unit efficiency prediction model for each subset.

3. The method according to claim 2, characterized in that The step of establishing a unit status assessment model based on the historical maintenance data specifically includes: Using an association rule mining algorithm to analyze the historical maintenance data, and determine the association between the fault symptoms and the fault causes of the generator set; Taking the performance indicators of the generator set as characteristic variables and the unit status rating of the generator set as the target variable, a unit status assessment model is established using the decision tree algorithm according to the correlation between the fault symptoms and the fault causes of the generator set.

4. The method according to claim 1, characterized in that: The carbon emission data of various power generation technologies in the grid planning area are specifically determined as follows: Obtain meteorological data and energy resource data of different types of renewable energy in the power grid planning area, establish a correlation between the energy resource data of each type of renewable energy and the meteorological data, and form a renewable energy capacity prediction model; Establishing a power generation model for each type of renewable energy, and determining power generation data of each type of renewable energy according to the power generation model and the renewable energy capacity prediction model; Acquire traditional energy consumption data in the power grid planning area, and establish an energy network diagram based on the traditional energy consumption data, wherein the energy network diagram includes the association relationship between electricity and other different traditional energy forms; Determine the carbon emission factors of various power generation technologies at various energy supply chain links, and determine the carbon emission data of various power generation technologies based on the carbon emission factors, the power generation data and the energy network diagram.

5. The method according to claim 4, characterized in that The step of establishing a correlation between the energy resource data of each type of renewable energy and the terrain data to form a renewable energy capacity prediction model specifically includes: Using principal component analysis to perform feature processing on the meteorological data, and extracting key meteorological factors related to renewable energy production capacity; The time series data of the meteorological factors are analyzed by an autoregressive integrated moving average model to obtain the dynamic change characteristics of the meteorological factors; According to the dynamic change characteristics and the energy resource data of each type of renewable energy, a long short-term memory network model is used to predict the renewable energy capacity, and a renewable energy capacity prediction model for each type of renewable energy is obtained.

6. The method according to claim 4, characterized in that The step of establishing an energy network diagram according to the traditional energy consumption data specifically includes: Determine a node set of an energy production unit, an energy conversion unit, and an energy consumption unit according to the traditional energy consumption data; Constructing an initial energy network graph according to the node set, wherein the edges of the initial energy network graph represent energy flow paths; By analyzing the energy flow path, the correlation strength between different nodes is obtained; The initial energy network diagram is optimized using a minimum spanning tree algorithm to calculate the weight of each edge; According to the association strength between different nodes and the weight of each edge, the initial energy network diagram is adjusted to a target energy network diagram.

7. The method according to claim 1, characterized in that Determining the power demand change trend of each type of agricultural industry according to the historical power demand data specifically includes: The autoregressive integrated moving average model is used to preprocess and analyze the historical energy demand data to obtain the time variation characteristics of energy demand for various types of agricultural industries; Obtain weather forecast data, crop growth cycle data, and livestock and poultry growth stage data to form multi-source heterogeneous data; Using principal component analysis to perform dimensionality reduction processing on the multi-source heterogeneous data to obtain multi-source heterogeneous data after dimensionality reduction; The multi-source heterogeneous data after dimensionality reduction are fused through the Kalman filter algorithm to obtain the target multi-source heterogeneous data; Taking the time variation characteristics of energy demand of each type of agricultural industry and the target multi-source heterogeneous data as input variables, taking the energy demand of each type of agricultural industry as output variables, and using a support vector machine regression model to construct an energy demand characteristic data set; Based on the energy demand characteristic data set, a long short-term memory neural network algorithm is used to model and predict the energy demand of various types of agricultural industries, and obtain the changing trends of electricity demand of various types of agricultural industries.

8. A modern rural power grid planning system, characterized in that: The system specifically comprises: The first planning module is used to obtain the operating data and historical maintenance data of the generator sets in the power grid planning area, establish a unit efficiency prediction model based on the operating data, and establish a unit status assessment model based on the historical maintenance data; The second planning module is used to determine the carbon emission data of various power generation technologies in the power grid planning area, wherein the carbon emission data includes the carbon emission data of various types of renewable energy power generation technologies and the carbon emission data of traditional energy power generation technologies; The third planning module is used to obtain historical power demand data of different types of agricultural industries in the power grid planning area, and determine the power demand change trend of each type of agricultural industry according to the historical power demand data; The fourth planning module is used to establish a power grid planning optimization model using a multi-objective evolutionary algorithm according to the prediction results of the unit efficiency prediction model, the evaluation results of the unit status evaluation model, the carbon emission data and the power demand change trend of various types of agricultural industries, and the power grid planning optimization model is used to determine the optimal power grid planning scheme; Among them, according to the prediction results of the unit efficiency prediction model, the evaluation results of the unit status evaluation model, the carbon emission data and the change trend of electricity demand of various types of agricultural industries, a multi-objective evolutionary algorithm is used to establish a power grid planning optimization model, which specifically includes: Taking minimizing total cost, maximizing energy utilization and minimizing carbon emissions as objective functions, and taking the prediction results of the unit efficiency prediction model, the evaluation results of the unit status evaluation model, the carbon emission data and the power demand change trends of various types of agricultural industries as input parameters, a power grid planning optimization model is established; A multi-objective evolutionary algorithm based on Pareto sorting is used to solve the multi-objective optimization model and obtain the Pareto optimal solution group; For each solution in the Pareto optimal solution group, conditional risk value is used as a risk measurement indicator to quantify the uncertainty of renewable energy and obtain the risk value of each solution; According to the risk value, an optimal solution is determined from the Pareto optimal solution group, and the optimal solution is an optimal power grid planning scheme.

Images