Rural smart energy area three-dimensional data acquisition method

Abstract

The invention provides a rural smart energy area three-dimensional data acquisition method, which comprises the following steps: constructing a data acquisition sample set which covers energy facility samples and infrastructure samples; constructing a grid region set based on the position of the energy facility; iteratively processing the sample set by using an optimized data processing algorithm to generate a high-precision three-dimensional point cloud model, adjusting the model weight and the sample weight according to a fitting error during processing, and updating the weight and the label of the infrastructure sample set; obtaining a final three-dimensional data model through weighted combination of the three-dimensional point cloud models; and analyzing a to-be-analyzed data set by using the model, and mapping data to a gridding region set, thereby analyzing the spatial positions and running states of the energy facility and the infrastructure. According to the invention, the precision and management efficiency of rural energy area data acquisition can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data collection and analysis of rural smart energy areas, and more specifically, to a method for collecting three-dimensional data of rural smart energy areas. Background Art

[0002] In the current rural energy sector, the traditional energy development model faces problems such as decentralized energy consumption, low energy density, lack of energy facilities and poor information interconnection. Energy development in rural areas is difficult, but the energy Internet provides a new solution, which is centered on electricity, relies on a strong smart grid, combines distributed energy equipment, modern information and communication technology and a complete application platform, supports the large-scale development and utilization of renewable energy, and improves comprehensive energy efficiency. The development of rural energy Internet aims to achieve clean production, supply and consumption of energy, and achieve multi-energy complementarity by integrating multiple renewable resources such as wind, light, and biomass, and promote the upgrading and transformation of rural energy systems.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: First, the investment demand for energy infrastructure in rural areas is large and the payback period is long, which leads to limited motivation for sustainable development; second, the improvement of the level of universal energy services in rural areas needs to be strengthened, and the existing service model relies on government subsidies and the support of state-owned enterprises, lacking a mechanism for market-oriented operation and multi-party participation; third, rural environmental monitoring and protection methods have not yet fully utilized digital technology, resulting in inaccurate and inefficient environmental governance; fourth, the level of intelligence in rural production management is not high, resulting in the need to improve agricultural production efficiency and food safety. Summary of the invention

[0004] The present invention provides a method for collecting three-dimensional data of rural smart energy areas, comprising:

[0005] Based on the same rural area, a data collection sample set is constructed, wherein the data collection sample set includes an energy facility sample set and an infrastructure sample set in the rural smart energy area;

[0006] Construct a gridded regional collection based on the location of energy facilities in the rural smart energy area;

[0007] Based on the optimized data processing algorithm, the data collection sample set is iteratively processed to obtain a high-precision three-dimensional point cloud model;

[0008] Each time the processing is performed, the weight of the current three-dimensional point cloud model and the sample weight of the energy facility sample set are set according to the fitting error of the current three-dimensional point cloud model to the energy facility sample set, and the sample weight and original label of the infrastructure sample set are updated in combination with the spatial position information of the samples in the infrastructure sample set, and the next three-dimensional point cloud model is processed according to the updated data acquisition sample set;

[0009] The final 3D data model is obtained by weighted combination of the 3D point cloud models;

[0010] The energy facilities and infrastructure data of the rural smart energy area are obtained again as the data set to be analyzed and passed into the final three-dimensional data model. The data to be analyzed are mapped to the gridded area set to analyze the spatial location and operation status of the energy facilities and infrastructure in the rural smart energy area.

[0011] Furthermore, each sample in the data collection sample set includes an energy facility feature vector and an original label, wherein the energy facility feature vector includes: facility type, facility size, facility location, facility operating status, and facility connection relationship; the original labels of the energy facility sample set include: distribution network, distributed photovoltaic, and wind power generation facilities; the original labels of the infrastructure sample set include: transportation facilities, communication facilities, and water conservancy facilities.

[0012] Furthermore, the fitting error of the current three-dimensional point cloud model to the energy facility sample set is the sum of the weights of samples in the energy facility sample set whose classification labels are inconsistent with the original labels.

[0013] Furthermore, the weight of the current three-dimensional point cloud model is set by the following formula:

[0014]

[0015] Among them, w k is the weight of the kth 3D point cloud model, e k is the fitting error of the kth 3D point cloud model for the energy facility sample set.

[0016] Furthermore, the sample weight of the energy facility sample set is updated by the following formula:

[0017]

[0018] in, is the sample weight of the i-th sample in the k-th 3D point cloud model in the target sample set, w i is the original weight of the i-th sample in the target sample set, e k is the fitting error of the kth 3D point cloud model for the energy facility sample set.

[0019] Furthermore, the updating of the sample weights and original labels of the infrastructure sample set in combination with the spatial location information of the samples in the infrastructure sample set includes:

[0020] Update sample weights for samples in the infrastructure sample set that are classified as non-energy facilities;

[0021] The samples classified as energy facilities in the infrastructure sample set are taken as samples to be processed, the sample set in the leaf node to which each sample to be processed is classified by the current three-dimensional point cloud model is obtained, and the energy facility confidence probability of each sample to be processed is calculated; after sorting the samples to be processed from large to small according to the energy facility confidence probability, the original labels of the samples to be processed before the energy facility confidence probability are updated to energy facilities, and the sample weights of all samples to be processed are updated, where Z∈1,2,...,Zmax-1, and Zmax is the number of samples to be processed.

[0022] Furthermore, the sample weights of the samples classified as non-energy facilities in the infrastructure sample set are updated by the following formula:

[0023]

[0024] in, is the sample weight of the jnth sample in the auxiliary sample set classified as a non-energy facility in the kth 3D point cloud model, w jn is the original weight of the jnth sample in the auxiliary sample set classified as a non-energy facility, d jn is the spatial position error of the jnth sample classified as a non-energy facility in the auxiliary sample set.

[0025] Furthermore, the energy facility confidence probability of each sample to be processed is the ratio of the number of samples whose original label is energy facility in the sample set where each sample to be processed is located.

[0026] Furthermore, the sample weights of all samples to be processed are updated by the following formula:

[0027]

[0028] in, is the sample weight of the juth sample to be processed in the kth 3D point cloud model in the auxiliary sample set, w ju is the original weight of the juth sample to be processed in the auxiliary sample set, p ju It is the proportion of the number of samples whose original label is energy facilities in the sample set of each sample to be processed.

[0029] Furthermore, each piece of data in the data set to be analyzed includes the latitude and longitude coordinates of energy facilities and infrastructure, energy facility feature vectors and original labels; the data to be analyzed is mapped to the gridded area set to analyze the spatial location and operation status of energy facilities and infrastructure in the rural smart energy area, including:

[0030] For the data to be analyzed that are classified as energy facilities, the grid area where each energy facility is located is obtained according to the longitude and latitude coordinates of each piece of data to be analyzed; the connection relationship between the energy facilities corresponding to the grid area is incremented by 1;

[0031] The connection relationship between any two points in the energy facilities is summarized to analyze the spatial location and operation status of the energy facilities and infrastructure in the rural smart energy area.

[0032] The above-mentioned embodiments according to the present invention have at least the following beneficial effects: the rural smart energy area three-dimensional data acquisition method of the present invention can improve the analysis accuracy of the spatial position and operating status of energy facilities and infrastructure. By comprehensively utilizing the energy facility sample set and the infrastructure sample set, combined with the construction of a gridded area set, this method can generate a more accurate three-dimensional point cloud model. During the data processing process, the dynamic adjustment of the model weights and sample weights further optimizes the fitting effect, so that the final three-dimensional data model can more realistically reflect the actual situation of the rural smart energy area. In addition, the method can enhance the robustness of the model by weighted combination of multiple three-dimensional point cloud models, providing strong data support for the planning and management of rural smart energy areas.

[0033] In addition, this method can also effectively improve the efficiency and accuracy of data processing. By adjusting the weight according to the fitting error in each iterative process, it can converge to the optimal solution faster and reduce the consumption of computing resources. At the same time, the sample weights and labels are updated in combination with the spatial location information of the infrastructure sample set, so that the model can more accurately identify and distinguish different types of facilities, which can improve the degree of automation of data processing. This method is particularly suitable for areas where the distribution of rural energy facilities and infrastructure is more complex and dispersed. Through this refined data collection and management, it can promote the rational development and utilization of rural energy, and then promote the upgrading and transformation of rural energy structure, and provide technical guarantees for achieving rural sustainable development and environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0035] Figure 1 A schematic flow chart of a method for collecting three-dimensional data of rural smart energy areas provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0036] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0037] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, apparatus, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0038] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0039] Reference below Figure 1 , Figure 1 The following is a flow chart of a method for collecting three-dimensional data of rural smart energy areas provided by an embodiment of the present invention. Figure 1 As shown, a rural smart energy area three-dimensional data collection method 100 includes:

[0040] Step 101, based on the same rural area, construct a data collection sample set, wherein the data collection sample set includes an energy facility sample set and an infrastructure sample set of the rural smart energy area;

[0041] Step 102, constructing a gridded area set according to the locations of energy facilities in the rural smart energy area;

[0042] Step 103, iteratively processing the data acquisition sample set based on the optimized data processing algorithm to obtain a high-precision three-dimensional point cloud model;

[0043] Step 104, wherein each time the processing is performed, the weight of the current three-dimensional point cloud model and the sample weight of the energy facility sample set are set according to the fitting error of the current three-dimensional point cloud model to the energy facility sample set, and the sample weight and original label of the infrastructure sample set are updated in combination with the spatial position information of the samples in the infrastructure sample set, and the next three-dimensional point cloud model is processed according to the updated data acquisition sample set;

[0044] Step 105, weighted combination of the three-dimensional point cloud models to obtain a final three-dimensional data model;

[0045] Step 106, again obtain the energy facilities and infrastructure data of the rural smart energy area as the data set to be analyzed, pass it into the final three-dimensional data model, map the data to be analyzed to the grid area set, and analyze the spatial location and operation status of the energy facilities and infrastructure in the rural smart energy area.

[0046] It should be noted that constructing a data collection sample set is the first step of this method, which involves collecting energy facility sample sets and infrastructure sample sets in rural smart energy areas. The data collection sample set refers to the basic data set used for subsequent data processing and model building. The energy facility sample set includes energy-related data such as distribution networks, distributed photovoltaics, and wind power generation facilities, while the infrastructure sample set covers non-energy-related data such as transportation facilities, communication facilities, and water conservancy facilities.

[0047] Specifically, these sample sets can be obtained through field measurements, remote sensing technology, historical data collection, etc. For example, the energy facility sample set can be constructed by recording feature vectors such as facility type, size, location, operating status, and connection relationship, while the infrastructure sample set can be formed by collecting spatial location information of transportation, communication, water conservancy, etc. Each sample data is accompanied by an original label to identify its corresponding category.

[0048] More specifically, during the construction of the sample set, machine learning algorithms can be used to pre-process the collected data to improve the quality and availability of the data. For example, for the energy facility sample set, different weight parameters can be set to reflect the importance of various types of facilities, such as the distribution network may have a higher weight than distributed photovoltaics.

[0049] Furthermore, for the infrastructure sample set, the sample weights can be adjusted according to the frequency of use and coverage of the facilities to ensure that the model can more accurately reflect the actual infrastructure distribution. In addition, outliers or inaccurate data can be filtered out by setting thresholds to improve the accuracy and robustness of the model.

[0050] In some embodiments, each sample in the data collection sample set includes an energy facility feature vector and an original label, wherein the energy facility feature vector includes: facility type, facility size, facility location, facility operating status, and facility connection relationship; the original labels of the energy facility sample set include: distribution network, distributed photovoltaic, and wind power generation facilities; the original labels of the infrastructure sample set include: transportation facilities, communication facilities, and water conservancy facilities.

[0051] It should be noted that each sample in the energy facility sample set and infrastructure sample set in this embodiment includes a feature vector and an original label. The feature vector refers to a series of values ​​that can characterize the characteristics of the sample, such as the type, size, location, operating status and connection relationship of the energy facility, while the original label refers to the label when the sample is initially classified, which is used for subsequent data processing and model training.

[0052] Specifically, the facility type in the energy facility feature vector can be a distribution network, distributed photovoltaic or wind power generation facility, etc., and the facility size can be the physical size of the facility, such as length, width and height. The location of the facility can be determined by the longitude and latitude coordinates, and the operating status can be the current working status of the facility, such as normal operation, maintenance or failure. The connection relationship of the facility describes the interconnection between the facilities. For example, the distribution network may be connected to multiple distributed photovoltaic facilities. The original labels of the infrastructure sample set can include transportation facilities, communication facilities, and water conservancy facilities.

[0053] Preferably, each sample in the energy facility sample set can be further refined into multiple sub-features. For example, the operating status of the facility can be subdivided into multiple levels, such as efficient operation, inefficient operation, and shutdown for repair.

[0054] More specifically, the sample weights of the infrastructure sample set can be adjusted according to the importance and frequency of use of the facilities to ensure that more critical infrastructures receive higher attention during model training. In addition, timestamps can be introduced to record the collection time of sample data so that the impact of time factors can be considered in model training and analysis.

[0055] In some embodiments, the fitting error of the current three-dimensional point cloud model to the energy facility sample set is the sum of the weights of samples in the energy facility sample set whose classification labels are inconsistent with the original labels.

[0056] It should be noted that the fitting error of the current 3D point cloud model for the energy facility sample set is determined by calculating the sum of the weights of the samples in the energy facility sample set whose classification labels are inconsistent with the original labels. The fitting error here refers to the difference between the model prediction result and the actual result, and the classification label refers to the label given by the model after classifying the sample. This process ensures that the model can accurately reflect the actual distribution of the sample set.

[0057] Specifically, the fitting error can be calculated by comparing the classification label predicted by the model with the original label of the sample. If the classification label predicted by the model is inconsistent with the original label of the sample, it is considered that the sample has a fitting error. At this time, it is necessary to calculate the weight of the sample and add the weights of all samples with fitting errors to obtain the fitting error of the entire sample set. This weight can be set based on the importance of the sample in the data set or its influence on the model prediction results.

[0058] Preferably, in order to calculate the fitting error more accurately, a threshold can be introduced to judge the consistency between the predicted label and the original label. For example, if the predicted label and the original label are considered consistent in a certain degree of similarity, the fitting error will not be counted.

[0059] Furthermore, we can also consider dynamic adjustment of sample weights, that is, adjusting sample weights according to the performance of the model in different iterations so that the model can converge to a better solution faster. This method helps to improve the generalization ability of the model and enable it to better adapt to new and unseen data.

[0060] In some embodiments, the weight of the current three-dimensional point cloud model is set by the following formula:

[0061]

[0062] Among them, w k is the weight of the kth 3D point cloud model, e k is the fitting error of the kth 3D point cloud model for the energy facility sample set.

[0063] It should be noted that the weight setting of the 3D point cloud model is determined based on the model's fitting error for the energy facility sample set. The weight mentioned here refers to the influence or importance of the model in subsequent data processing and model optimization. By considering the fitting error, the weight of the model can be adjusted to give it an appropriate proportion in the weighted combination of the final data model.

[0064] Specifically, the model weight can be calculated by the formula

[0065]

[0066] To achieve this, w k represents the weight of the kth 3D point cloud model, e k represents the fitting error of the model for the energy facility sample set. In this formula, the fitting error e k The larger the model weight w k The smaller the value, the smaller the value, and vice versa. This setting can ensure that the models with good fitting effect have a greater weight in the final data model.k It can be the sum of sample weights based on inconsistent classification labels, or other metrics that quantify fitting errors.

[0067] Preferably, the calculation of model weights may also take into account other factors, such as the complexity of the model or the stability during training. For example, an adjustment factor may be introduced to balance the accuracy and complexity of the model to prevent overfitting.

[0068] Furthermore, cross-validation methods can be used to evaluate the generalization ability of the model and adjust the model weights accordingly. These methods help improve the accuracy and robustness of the final data model, so that it better reflects the actual situation of rural smart energy areas.

[0069] In some embodiments, the sample weights of the energy facility sample set are updated by the following formula:

[0070]

[0071] in, is the sample weight of the i-th sample in the k-th 3D point cloud model in the target sample set, w i is the original weight of the i-th sample in the target sample set, e k is the fitting error of the kth 3D point cloud model for the energy facility sample set.

[0072] It should be noted that the sample weight update process of the energy facility sample set is based on the fitting error of the 3D point cloud model to the sample set. The sample weight update here refers to adjusting the sample weight according to the model fitting error after each iteration so as to optimize the model in the next iteration. The fitting error refers to the difference between the model prediction result and the actual result, while the sample weight represents the influence of the sample in the model training.

[0073] Specifically, the update of sample weights can be achieved by the following formula:

[0074]

[0075] in represents the new weight of the i-th sample after the k-th iteration, w i,k is the original weight, e k is the fitting error of the kth 3D point cloud model. In this process, the fitting error e k The larger the value, the greater the decrease in sample weight, thereby reducing the influence of samples that cause large fitting errors in subsequent model training.

[0076] Preferably, the update of sample weights can also be fine-tuned in combination with the importance of the samples and other relevant factors. For example, a higher base weight can be set for samples that are closely related to the safety and efficiency of energy facilities, so that they can maintain a certain influence even if their fitting errors are large.

[0077] Furthermore, a weight decay factor can be introduced to ensure that the weight update process does not fluctuate drastically due to the fitting error of a single iteration, thereby improving the stability of model training.

[0078] In some embodiments, the updating of the sample weights and original labels of the infrastructure sample set in combination with the spatial location information of the samples in the infrastructure sample set includes:

[0079] Update sample weights for samples in the infrastructure sample set that are classified as non-energy facilities;

[0080] The samples classified as energy facilities in the infrastructure sample set are taken as samples to be processed, the sample set in the leaf node to which each sample to be processed is classified by the current three-dimensional point cloud model is obtained, and the energy facility confidence probability of each sample to be processed is calculated; after sorting the samples to be processed from large to small according to the energy facility confidence probability, the original labels of the samples to be processed before the energy facility confidence probability are updated to energy facilities, and the sample weights of all samples to be processed are updated, where Z∈1,2,...,Zmax-1, and Zmax is the number of samples to be processed.

[0081] It should be noted that updating sample weights and original labels in combination with the spatial location information of the infrastructure sample set means that in the process of processing the three-dimensional point cloud model, not only the fitting of the energy facility samples is considered, but also the spatial distribution characteristics of the infrastructure samples are used to optimize the model. Here, the spatial location information refers to the coordinate data of the infrastructure samples in three-dimensional space, which can be used to assist in judging and updating the weights and labels of the samples.

[0082] Specifically, the updating process can be performed by analyzing the spatial distribution pattern of the infrastructure samples. For example, if some infrastructure samples are close to known energy facility samples in space, it can be considered that the possibility of these infrastructure samples being associated with the energy facilities is high.

[0083] Furthermore, in this case, the weights of these infrastructure samples can be increased, and their original labels can be considered to be updated to energy facilities. In addition, for samples with classification labels as non-energy facilities, their weights can be adjusted according to their spatial distance from energy facility samples. The closer the distance, the greater the magnitude of the weight adjustment.

[0084] Preferably, the process of updating the sample weights and original labels of the infrastructure sample set can be further refined. For example, a spatial position error threshold can be set, and only when the spatial distance between the infrastructure sample and the energy facility sample is less than the threshold, their weights and labels are considered to be updated.

[0085] Furthermore, a confidence scoring mechanism can be introduced to calculate the confidence score based on the spatial relationship, similarity and other relevant factors between the infrastructure samples and the energy facility samples, and decide whether to update the weight and label of the sample based on the scoring results.

[0086] In some embodiments, the sample weights of samples in the infrastructure sample set classified as non-energy facilities are updated by the following formula:

[0087]

[0088] in, is the sample weight of the jnth sample in the auxiliary sample set classified as a non-energy facility in the kth 3D point cloud model, w jn is the original weight of the jnth sample in the auxiliary sample set classified as a non-energy facility, d jn is the spatial position error of the jnth sample classified as a non-energy facility in the auxiliary sample set.

[0089] It should be noted that the weights of samples classified as non-energy facilities in the infrastructure sample set are updated through the formula. Non-energy facilities refer to infrastructure that is not directly involved in energy production or distribution, such as transportation, communications, and water conservancy facilities. This process ensures that during model training, the weights of non-energy facility samples can be dynamically adjusted according to their performance in the model, thereby optimizing the overall model effect.

[0090] Specifically, the update of sample weights can be achieved by the following formula:

[0091]

[0092] in represents the new weight of the jth non-energy facility sample after the kth iteration, w j,k is the original weight, e j is the spatial position error of the sample. The spatial position error can be obtained by calculating the distance between the sample and its nearest energy facility sample. The original weight of the sample can be set according to its importance in the data set to ensure that the key samples maintain a certain influence during the update process.

[0093] Preferably, the process of updating the non-energy facility sample weights can be further refined. For example, a weighting factor of the spatial position error can be introduced so that the update amplitude of the sample weights is larger when the spatial position error is smaller.

[0094] Furthermore, the historical performance of the samples can also be considered. For example, if a non-energy facility sample maintains a consistently high fitting error over multiple iterations, a threshold can be set to decide whether to reduce its weight to a minimum value to avoid negative impact on the model.

[0095] In some embodiments, the energy facility confidence probability of each sample to be processed is the ratio of the number of samples whose original label is energy facility in the sample set where each sample to be processed belongs.

[0096] It should be noted that the energy facility confidence probability of the sample to be processed is calculated based on the proportion of the number of samples related to energy facilities in the sample set. The energy facility confidence probability refers to the degree of confidence that a sample is classified as an energy facility, and this probability can be used to evaluate the relevance of the sample to the energy facility. In this context, the sample set refers to the set of infrastructure samples collected in a specific area, and the number of samples originally labeled as energy facilities refers to the number of samples in this set that have been labeled as energy facilities.

[0097] Specifically, the calculation of the confidence probability of energy facilities can be achieved by the following formula:

[0098]

[0099] Where P ju is the confidence probability of the energy facility of the juth sample to be processed, N energy is the number of samples whose original label is energy facilities in the sample set where the sample to be processed is located, N total is the total number of all samples in the sample set. This calculation method is simple and intuitive, and can reflect the universality of energy facility samples in a sample set.

[0100] Preferably, the process of calculating the confidence probability of energy facilities can be further refined in combination with other characteristic information. For example, the spatial distribution characteristics of the samples can be considered. If a sample to be processed is not only similar to the energy facility sample in quantity, but also shows a certain degree of clustering in spatial distribution, then this sample can be considered to have a higher confidence probability of energy facilities.

[0101] Furthermore, other attribute information of the sample can be introduced, such as the frequency of use of the facility, the functional type, etc., to comprehensively evaluate its relevance to the energy facility. These refined steps help to improve the accuracy and practicality of the confidence probability of energy facilities.

[0102] In some embodiments, the sample weights of all samples to be processed are updated by the following formula:

[0103]

[0104] in, is the sample weight of the juth sample to be processed in the kth 3D point cloud model in the auxiliary sample set, w ju is the original weight of the juth sample to be processed in the auxiliary sample set, p ju It is the proportion of the number of samples whose original label is energy facilities in the sample set of each sample to be processed.

[0105] It should be noted that the sample weights of all samples to be processed are updated by combining the proportion of the number of samples whose original label is energy facilities in the sample set to which they belong. The samples to be processed here refer to infrastructure samples whose classification labels have not yet been finalized, and the sample set refers to the set of all relevant infrastructure samples collected in a specific area. The proportion of the number of samples whose original label is energy facilities refers to the proportion of samples in the sample set that have been labeled as energy facilities.

[0106] Specifically, the update of sample weights can be achieved by the following formula:

[0107]

[0108] in represents the new weight of the juth sample to be processed, w ju is the original weight, and r is the proportion of the number of samples with the original label of energy facilities in the sample set where the sample to be processed is located. This proportion r can be obtained by counting the ratio of the number of energy facility samples in the sample set to the total number of samples. The update of sample weight is intended to adjust its influence in the model according to the correlation between the sample and the energy facility.

[0109] Preferably, the updating process of the sample weight can be further refined to more accurately reflect the relevance of the sample to the energy facility. For example, a confidence threshold can be introduced, and the weight of the sample is updated only when the confidence probability of the energy facility of the sample exceeds this threshold.

[0110] Furthermore, other characteristics of the sample can be considered, such as its spatial distance from the labeled energy facility samples, functional similarity, etc., to comprehensively determine the adjustment of the weight. These refined steps help improve the accuracy and robustness of the model in classifying infrastructure samples.

[0111] In some embodiments, each piece of data in the data set to be analyzed includes the latitude and longitude coordinates of energy facilities and infrastructure, energy facility feature vectors, and original labels; mapping the data to be analyzed to the gridded area set to analyze the spatial location and operating status of energy facilities and infrastructure in the rural smart energy area includes:

[0112] For the data to be analyzed that are classified as energy facilities, the grid area where each energy facility is located is obtained according to the longitude and latitude coordinates of each piece of data to be analyzed; the connection relationship between the energy facilities corresponding to the grid area is incremented by 1;

[0113] The connection relationship between any two points in the energy facilities is summarized to analyze the spatial location and operation status of the energy facilities and infrastructure in the rural smart energy area.

[0114] It should be noted that the dataset to be analyzed is mapped to the gridded area set in order to analyze the spatial location and operating status of energy facilities and infrastructure in the rural smart energy area. The dataset to be analyzed refers to the dataset containing the latest data of energy facilities and infrastructure, which is ready to be mapped to the established gridded area set. The gridded area set refers to dividing the rural smart energy area into multiple grids, each of which contains a specific geographical range, for more detailed management and analysis of data.

[0115] Specifically, the mapping process can be carried out through the following steps: first, determine the grid area where each piece of data to be analyzed is located according to its latitude and longitude coordinates; then, for the data to be analyzed that is classified as energy facilities, update the connection relationship between the energy facilities in the grid area, such as increasing the connection counter; finally, by summarizing the connection relationship between any two points in all energy facilities, comprehensively analyze the spatial location and operating status of the energy facilities and infrastructure in the entire rural smart energy area. Parameter settings may include the size and shape of the grid, and how to define and update the connection relationship.

[0116] Preferably, the mapping and analysis process can be refined by incorporating more contextual information and analytical techniques. For example, machine learning algorithms can be used to predict the operating status of energy facilities, or spatial analysis methods can be used to identify potential optimization locations for energy facilities and infrastructure.

[0117] Furthermore, real-time data streams can be introduced to dynamically update information in the gridded area collection and timely reflect the current status of energy facilities and infrastructure. These alternatives help improve the real-time and accuracy of analysis, making the management of rural smart energy areas more efficient and intelligent.

[0118] The above-mentioned embodiments of the present invention have the following beneficial effects: the three-dimensional data acquisition technology of the present invention can provide accurate data support for rural smart energy areas. By integrating energy facilities and infrastructure sample sets, combining the construction of gridded regional sets, and iteratively optimized data processing algorithms, the generated three-dimensional point cloud model can highly restore the actual situation of rural energy facilities. This method refines the sample set processing, continuously adjusts the model weights and sample weights, and updates the sample weights and original labels of the infrastructure sample set, thereby improving the fitting accuracy of the model, making the final three-dimensional data model more accurate, which helps to achieve effective monitoring and management of energy facilities and infrastructure.

[0119] At the same time, this method can more accurately map and analyze the energy facilities and infrastructure data of rural smart energy areas by weighted combination of various three-dimensional point cloud models to obtain the final data model. This mapping and analysis process can reveal the spatial location and operating status of energy facilities and infrastructure, and provide a detailed data basis for energy management and optimization in rural areas. The application of this method can enhance the data collection accuracy of rural energy areas, improve the level of intelligent energy management, help promote the rational development and effective utilization of rural energy, and provide strong technical support for the sustainable development of rural areas.

[0120] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0121] The above descriptions are only some preferred embodiments of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention to form a technical solution.

Claims

1. A method for collecting three-dimensional data of rural smart energy areas, characterized in that: The steps include: Based on the same rural area, a data collection sample set is constructed, wherein the data collection sample set includes an energy facility sample set and an infrastructure sample set in the rural smart energy area; Construct a gridded regional set based on the location of energy facilities in the rural smart energy area; Based on the optimized data processing algorithm, the data collection sample set is iteratively processed to obtain a high-precision three-dimensional point cloud model; Each time the processing is performed, the weight of the current three-dimensional point cloud model and the sample weight of the energy facility sample set are set according to the fitting error of the current three-dimensional point cloud model to the energy facility sample set, and the sample weight and original label of the infrastructure sample set are updated in combination with the spatial position information of the samples in the infrastructure sample set, and the next three-dimensional point cloud model is processed according to the updated data acquisition sample set; The final 3D data model is obtained by weighted combination of the 3D point cloud models; The energy facilities and infrastructure data of the rural smart energy area are obtained again as the data set to be analyzed and passed into the final three-dimensional data model. The data to be analyzed are mapped to the gridded area set to analyze the spatial location and operation status of the energy facilities and infrastructure in the rural smart energy area.

2. The rural smart energy regional three-dimensional data acquisition method according to claim 1 is characterized in that: Each sample in the data collection sample set includes an energy facility feature vector and an original label, wherein the energy facility feature vector includes: facility type, facility size, facility location, facility operating status, and facility connection relationship; the original labels of the energy facility sample set include: distribution network, distributed photovoltaic, and wind power generation facilities; the original labels of the infrastructure sample set include: transportation facilities, communication facilities, and water conservancy facilities.

3. The rural smart energy regional three-dimensional data acquisition method according to claim 2 is characterized in that: The fitting error of the current three-dimensional point cloud model to the energy facility sample set is the sum of the weights of samples in the energy facility sample set whose classification labels are inconsistent with the original labels.

4. The rural smart energy regional three-dimensional data acquisition method according to claim 3 is characterized in that: The weight of the current 3D point cloud model is set by the following formula: Among them, w k is the weight of the kth 3D point cloud model, e k is the fitting error of the kth 3D point cloud model for the energy facility sample set.

5. The rural smart energy regional three-dimensional data acquisition method according to claim 4 is characterized in that: The sample weights of the energy facility sample set are updated by the following formula: in, is the sample weight of the i-th sample in the k-th 3D point cloud model in the target sample set, w i is the original weight of the i-th sample in the target sample set, e k is the fitting error of the kth 3D point cloud model for the energy facility sample set.

6. The rural smart energy regional three-dimensional data acquisition method according to claim 2 is characterized in that: The updating of the sample weights and original labels of the infrastructure sample set in combination with the spatial location information of the samples in the infrastructure sample set includes: Update sample weights for samples in the infrastructure sample set that are classified as non-energy facilities; The samples classified as energy facilities in the infrastructure sample set are taken as samples to be processed, the sample set in the leaf node to which each sample to be processed is classified by the current three-dimensional point cloud model is obtained, and the energy facility confidence probability of each sample to be processed is calculated; after sorting the samples to be processed from large to small according to the energy facility confidence probability, the original labels of the samples to be processed before the energy facility confidence probability are updated to energy facilities, and the sample weights of all samples to be processed are updated, where Z∈1,2,...,Zmax-1, and Zmax is the number of samples to be processed.

7. The rural smart energy regional three-dimensional data acquisition method according to claim 6 is characterized in that: The sample weights of samples classified as non-energy facilities in the infrastructure sample set are updated by the following formula: in, is the sample weight of the jnth sample in the auxiliary sample set classified as a non-energy facility in the kth 3D point cloud model, w jn is the original weight of the jnth sample in the auxiliary sample set classified as a non-energy facility, d jn is the spatial position error of the jnth sample classified as a non-energy facility in the auxiliary sample set.

8. The rural smart energy regional three-dimensional data acquisition method according to claim 6 is characterized in that: The energy facility confidence probability of each sample to be processed is the ratio of the number of samples whose original label is energy facility in the sample set where each sample to be processed is located.

9. The rural smart energy regional three-dimensional data acquisition method according to claim 6 is characterized in that: Update the sample weights of all samples to be processed by the following formula: in, is the sample weight of the juth sample to be processed in the kth 3D point cloud model in the auxiliary sample set, w ju is the original weight of the juth sample to be processed in the auxiliary sample set, p ju It is the proportion of the number of samples whose original label is energy facilities in the sample set of each sample to be processed.

10. The rural smart energy regional three-dimensional data acquisition method according to claim 2 is characterized in that: Each piece of data in the data set to be analyzed includes the latitude and longitude coordinates of energy facilities and infrastructure, energy facility feature vectors and original labels; the data to be analyzed is mapped to the gridded area set to analyze the spatial location and operation status of energy facilities and infrastructure in the rural smart energy area, including: For the data to be analyzed that are classified as energy facilities, the grid area where each energy facility is located is obtained according to the longitude and latitude coordinates of each piece of data to be analyzed; the connection relationship between the energy facilities corresponding to the grid area is incremented by 1; The connection relationship between any two points in the energy facilities is summarized to analyze the spatial location and operation status of the energy facilities and infrastructure in the rural smart energy area.

Images