Electric power data acquisition method and apparatus, device, and readable storage medium

Abstract

The present application relates to the technical field of electric power line inspection, and relates to an electric power data acquisition method and apparatus, a device, and a readable storage medium. The method comprises: determining a risk prediction result on the basis of multi-modal data of an electric power line and a preset risk prediction model, wherein the multi-modal data comprises image data, temperature data, and electrical quantity data; re-acquiring target multi-modal data according to a data acquisition strategy corresponding to the risk prediction result, and updating the risk prediction model on the basis of an actual fault, a hidden hazard resolution situation and the target multi-modal data; and acquiring current multi-modal data, performing risk prediction on the current multi-modal data on the basis of the updated risk prediction model, so as to obtain a risk detection result of the electric power line, and updating the data acquisition strategy on the basis of the risk detection result, wherein the updated data acquisition strategy is used for multi-modal data acquisition. By using the method, the utilization efficiency of acquisition resources can be improved.

Description

Power data collection method, device, equipment and readable storage medium TECHNICAL FIELD

[0001] The present application relates to the technical field of power line inspection, in particular to a power data collection method, device, equipment and readable storage medium. BACKGROUND

[0002] With the development of the power industry, the scale of the power grid has rapidly expanded, and the demand for the power system in daily life has increased. Power inspection aims to detect the state of power transmission equipment and timely discover equipment defects and safety hazards, playing an important role in ensuring the normal operation of the power system. Safety inspection technology has been widely researched and applied at home and abroad. By combining technologies such as unmanned aerial vehicles, mobile sensors, data analysis and prediction models, safety inspection technology can improve the safety and reliability of the line, reduce the risk of accidents and the likelihood of failure, and has important significance for the operation and maintenance of the power system.

[0003] However, the power line is long, it is difficult to obtain power with fixed collection terminals, unmanned aerial vehicle inspection is time-consuming and labor-intensive, and the cost of data collection is high, which limits the inspection frequency. Defects and hidden dangers must be discovered in a timely manner, otherwise they can cause large-scale power outages and serious consequences, and also require timely inspection. Therefore, there is an urgent need for a power data collection method that can improve the utilization of collection resources. SUMMARY

[0004] Therefore, it is necessary to provide a power data collection method, device, computer equipment, computer readable storage medium and computer program product that can improve the utilization of collection resources.

[0005] In a first aspect, the present application provides a power data collection method, comprising:

[0006] Based on the multi-modal data of the power line and the preset risk prediction model, a risk prediction result is determined; the multi-modal data includes image data, temperature data and electrical quantity data;

[0007] According to the data collection strategy corresponding to the risk prediction result, target multi-modal data is re-collected, and the risk prediction model is updated based on the actual fault, hidden danger elimination situation and target multi-modal data;

[0008] Obtain the current multi-modal data, based on the updated risk prediction model, the risk prediction of the current multi-modal data is carried out, the risk detection result of the power line is obtained, and the data collection strategy is updated according to the risk detection result, and the updated data collection strategy is used for multi-modal data collection.

[0009] In one of the embodiments, the risk prediction result is determined based on the power line multi-modal data and a preset risk prediction model, including:

[0010] The multi-modal data of the power line is identified by using a preset identification analysis system to obtain at least one hidden danger identification result corresponding to each hidden danger of the power line;

[0011] The at least one hidden danger identification result corresponding to each hidden danger is input into a preset risk prediction model to obtain a risk prediction result.

[0012] In one of the embodiments, the target multi-modal data is re-collected according to the data collection strategy corresponding to the risk prediction result, including:

[0013] The collection cycle of the collection device is adjusted according to the data collection strategy corresponding to the risk prediction result;

[0014] The collection device re-collects the target multi-modal data according to the adjusted collection cycle.

[0015] In one of the embodiments, the risk prediction model is updated based on the actual fault, the hidden danger elimination situation and the target multi-modal data, including:

[0016] The risk prediction result is updated based on the target multi-modal data and the risk prediction model;

[0017] The correctness of the updated risk prediction result is checked according to the actual fault and the hidden danger elimination situation;

[0018] The correctness checking result, the actual fault and the hidden danger elimination situation are input into a preset reinforcement learning model to obtain a parameter change amount, and the risk prediction model is updated based on the parameter change amount.

[0019] In one of the embodiments, the correctness checking result, the actual fault and the hidden danger elimination situation are input into a preset reinforcement learning model to obtain a parameter change amount, including:

[0020] The correctness checking result, the actual fault and the hidden danger elimination situation are preprocessed by using a reward and punishment function in the preset reinforcement learning model to obtain a preprocessing result;

[0021] The preprocessing result is converted into a parameter change amount based on a value function in the preset reinforcement learning model.

[0022] In one of the embodiments, the power data collection method further includes:

[0023] The to-be-processed multi-modal data is collected according to the updated data collection strategy;

[0024] The updated risk prediction model is updated based on the to-be-processed multi-modal data.

[0025] In a second aspect, the present application also provides an electric power data acquisition device, comprising:

[0026] The acquisition module is configured to determine a risk prediction result based on the multi-modal data of the electric power line and a preset risk prediction model; the multi-modal data comprises image data, temperature data and electrical quantity data;

[0027] The first updating module is configured to re-acquire target multi-modal data according to a data acquisition strategy corresponding to the risk prediction result, and update the risk prediction model based on actual faults, hidden danger elimination conditions and the target multi-modal data;

[0028] The second updating module is configured to acquire current multi-modal data, perform risk prediction on the current multi-modal data based on the updated risk prediction model, obtain a risk detection result of the electric power line, and update the data acquisition strategy according to the risk detection result; the updated data acquisition strategy is used for multi-modal data acquisition.

[0029] In a third aspect, the present application also provides a computer device, comprising a memory and a processor; the memory stores a computer program; and the processor implements the following steps when executing the computer program:

[0030] Determine a risk prediction result based on multi-modal data of the electric power line and a preset risk prediction model; the multi-modal data comprises image data, temperature data and electrical quantity data;

[0031] Re-acquire target multi-modal data according to a data acquisition strategy corresponding to the risk prediction result, and update the risk prediction model based on actual faults, hidden danger elimination conditions and the target multi-modal data;

[0032] Acquire current multi-modal data, perform risk prediction on the current multi-modal data based on the updated risk prediction model, obtain a risk detection result of the electric power line, and update the data acquisition strategy according to the risk detection result; the updated data acquisition strategy is used for multi-modal data acquisition.

[0033] In a fourth aspect, the present application also provides a computer readable storage medium, which stores a computer program; the computer program is executed by a processor to implement the following steps:

[0034] Determine a risk prediction result based on multi-modal data of the electric power line and a preset risk prediction model; the multi-modal data comprises image data, temperature data and electrical quantity data;

[0035] Re-acquire target multi-modal data according to a data acquisition strategy corresponding to the risk prediction result, and update the risk prediction model based on actual faults, hidden danger elimination conditions and the target multi-modal data;

[0036] obtain current multi-modal data, perform risk prediction on the current multi-modal data based on the updated risk prediction model to obtain a risk detection result of the power line, and update the data collection strategy according to the risk detection result, wherein the updated data collection strategy is used for multi-modal data collection.

[0037] In a fifth aspect, the present application further provides a computer program product comprising a computer program which, when executed by a processor, implements the following steps:

[0038] determining a risk prediction result based on the multi-modal data of the power line and a preset risk prediction model, wherein the multi-modal data comprises image data, temperature data and electrical quantity data;

[0039] recollecting target multi-modal data according to a data collection strategy corresponding to the risk prediction result, and updating the risk prediction model based on actual faults, hidden danger elimination conditions and the target multi-modal data;

[0040] obtain current multi-modal data, perform risk prediction on the current multi-modal data based on the updated risk prediction model to obtain a risk detection result of the power line, and update the data collection strategy according to the risk detection result, wherein the updated data collection strategy is used for multi-modal data collection.

[0041] The power data collection method, device, computer device, computer readable storage medium and computer program product described above obtain a risk prediction result based on multi-modal data of a power line and a preset risk prediction model, and then recollect target multi-modal data according to a data collection strategy corresponding to the risk prediction result. The target multi-modal data is recollected in combination with the risk prediction result of the power line, and therefore, the utilization rate of collection resources can be improved. The risk prediction model can be continuously updated according to actual faults, hidden danger elimination conditions and target multi-modal data, and therefore, the risk detection accuracy of the risk prediction model for the power line can be improved. The data collection strategy is updated by using a risk detection result detected by the updated risk prediction model, and therefore, the dynamic adjustment of the data collection strategy is realized. The updated data collection strategy is used for multi-modal data collection, and therefore, a self-adaptive collection closed loop of power data is formed, and the utilization rate of collection resources is improved. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0043] Fig. 1 is a diagram of an application environment of a power data collection method in an embodiment;

[0044] Fig. 2 is a flow diagram of a power data collection method in an embodiment;

[0045] Fig. 3 is a flow diagram of a power data collection method in another embodiment;

[0046] Fig. 4 is a flow diagram of a power data collection method in yet another embodiment;

[0047] Fig. 5 is a structural block diagram of a power data collection device in an embodiment;

[0048] Fig. 6 is an internal structural diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0050] The power data collection method provided by the embodiments of the present application can be applied in an application environment as shown in FIG. 1. In the application environment, a terminal 102 communicates with a server 104 through a network. A data storage system can store data required to be processed by the server 104. The data storage system can be integrated on the server 104, or placed on a cloud or other network server. The embodiments take the method applied to the terminal 102 as an example for illustration. It can be understood that the method can also be applied to the server 104, and can also be applied to a system including the terminal 102 and the server 104, and implemented through the interaction of the terminal 102 and the server 104. The terminal 102 determines a risk prediction result based on multi-modal data of a power line and a preset risk prediction model. The multi-modal data includes image data, temperature data and electrical quantity data. Target multi-modal data is re-collected according to a data collection strategy corresponding to the risk prediction result. The risk prediction model is updated based on actual faults, hidden danger elimination conditions and the target multi-modal data. Current multi-modal data is obtained. The current multi-modal data is risk predicted based on the updated risk prediction model, to obtain a risk detection result of the power line. The data collection strategy is updated according to the risk detection result. The updated data collection strategy is used for multi-modal data collection. The terminal 102 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things device can be a smart speaker, a smart television, a smart air conditioner, a smart vehicle-mounted device, a projection device, etc. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 can be a stand-alone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0051] In an exemplary embodiment, as shown in FIG. 2, a power data collection method is provided. The method is taken as an example for illustration applied to the terminal 102 in FIG. 1, and includes the following steps 202 to 206. In the steps:

[0052] In step 202, a risk prediction result is determined based on multi-modal data of a power line and a preset risk prediction model. The multi-modal data includes image data, temperature data and electrical quantity data.

[0053] The multi-modal data refers to multiple power-related data in the power line. For example, the multi-modal data includes image data, temperature data, and electrical quantity data, etc. The image data can be images obtained by using a patrol unmanned aerial vehicle or a monitoring camera to patrol and shoot the power line. The types of image data can include visible light images, infrared images, ultraviolet images, etc. When the image acquisition device is a three-dimensional camera, three-dimensional point cloud data can be acquired. The temperature data can be temperature acquired by an environmental sensor deployed in the power line. The electrical quantity data includes current data and voltage data, etc. The electrical quantity data can be data acquired by an electrical quantity sensor deployed in the power line.

[0054] The risk prediction model refers to a machine learning model preset in the terminal, which is used to predict the risk of the power line and obtain a risk prediction result.

[0055] The risk prediction result is used to indicate the risk type of the power line. Since the risk prediction result is obtained based on the multi-modal data and the risk prediction model, the accuracy of the risk prediction result depends on the real-time performance and accuracy of the multi-modal data, and the prediction ability of the risk prediction model.

[0056] In step 204, target multi-modal data is re-acquired according to a data acquisition strategy corresponding to the risk prediction result, and the risk prediction model is updated based on the actual fault, the hidden danger elimination situation, and the target multi-modal data.

[0057] The data acquisition strategy refers to a strategy for multi-modal data acquisition by each acquisition device. In some embodiments, a mapping relationship between the risk type and the data acquisition strategy is preset in the terminal. By searching the mapping relationship, the data acquisition strategy corresponding to the risk prediction result can be obtained.

[0058] The target multi-modal data is multi-modal data re-acquired according to the data acquisition strategy. Since the power line is long, the risk prediction model can obtain the risk prediction result of the power line. For different risk prediction results, different data acquisition strategies are used, which is beneficial to focus on monitoring the multi-modal data of the risk area and reduce the monitoring of the multi-modal data of the non-risk area, thereby improving the utilization rate of the acquisition resources.

[0059] The actual fault refers to a real fault of the power line confirmed by manual inspection. The hidden danger elimination situation refers to the situation of eliminating the discovered faults and hidden dangers.

[0060] With the collection of multi-modal data and the prediction of risks, the samples for fault detection increase, in order to improve the prediction accuracy of the risk prediction model, the embodiments of the present application propose to update the risk prediction model based on the actual fault, the hidden danger elimination situation and the target multi-modal data, so as to ensure that the risk prediction model can be dynamically updated according to the real-time collected multi-modal data and the real fault situation. In the field of power line fault detection, limited by the amount of fault sample data, this method of dynamically updating the risk prediction model is beneficial to improve the prediction ability of the risk prediction model.

[0061] In step 206, the current multi-modal data is obtained, the risk prediction model is updated based on the updated risk prediction model, the risk detection result of the power line is obtained, and the data acquisition strategy is updated according to the risk detection result. The updated data acquisition strategy is used for multi-modal data acquisition.

[0062] Among them, after the risk prediction model is updated, the terminal reacquires the current multi-modal data collected at the current time, and performs risk prediction on the current multi-modal data to obtain the risk detection result of the power line.

[0063] Since the multi-modal data and the risk prediction model are updated, the risk detection result is updated, the terminal can obtain the data acquisition strategy corresponding to the risk detection result, realize the dynamic update of the data acquisition strategy, and collect the multi-modal data according to the updated data acquisition strategy. The acquisition equipment can focus on the risk point, so as to ensure the efficient use of the acquisition resource.

[0064] In the above power data acquisition method, the risk prediction result is obtained through the multi-modal data of the power line and the preset risk prediction model, and the target multi-modal data is reacquired according to the data acquisition strategy corresponding to the risk prediction result. The target multi-modal data is reacquired in combination with the risk prediction result of the power circuit, so as to improve the utilization rate of the acquisition resource; the risk prediction model can be continuously updated according to the actual fault, the hidden danger elimination situation and the target multi-modal data, which is beneficial to improve the risk detection accuracy of the risk prediction model for the power line; the data acquisition strategy is updated by the risk detection result detected by the updated risk prediction model, the dynamic adjustment of the data acquisition strategy is realized, the updated data acquisition strategy is used for multi-modal data acquisition, and the self-adaptive acquisition closed loop of the power data is formed, which improves the utilization rate of the acquisition resource.

[0065] In an example embodiment, the risk prediction result is determined based on the power line multi-modal data and a preset risk prediction model, including: adopting a preset identification analysis system to identify hidden dangers in the power line multi-modal data to obtain at least one hidden danger each corresponding hidden danger identification result; inputting the at least one hidden danger each corresponding hidden danger identification result into the preset risk prediction model to obtain the risk prediction result.

[0066] The preset identification analysis system in the terminal can be a multi-modal large model or multiple small models. The multi-modal large model is used to uniformly process the multi-modal data, and the multiple small models are used to process each modal data respectively to identify possible hidden danger conditions in the power line and predict a hidden danger score of the power line.

[0067] For example, the terminal inputs the multi-modal data into the identification analysis system to obtain at least one hidden danger each corresponding hidden danger identification result. Each hidden danger corresponding hidden danger identification result can include a hidden danger score and a confidence. The confidence is a measure of the credibility of the hidden danger.

[0068] The terminal inputs the at least one hidden danger each corresponding hidden danger identification result into the preset risk prediction model to obtain a risk prediction result output by the risk prediction model. For example, the risk prediction result can be a risk prediction matrix. Each element in the matrix is a probability or intensity of a corresponding hidden danger.

[0069] In some embodiments, the terminal further abstracts the risk prediction matrix into a collection strategy matrix, and automatically generates a data collection strategy based on the collection strategy matrix. The collection strategy matrix and the risk prediction matrix have a preset corresponding relationship, and the data collection strategy and the collection strategy matrix have a preset corresponding relationship.

[0070] In this embodiment, the multi-modal data is identified by the preset identification analysis system, and the at least one hidden danger each corresponding hidden danger identification result is processed by the preset risk prediction model to obtain the risk prediction result. This way of processing the multi-modal data by the preset system and model is conducive to obtaining an accurate risk prediction result.

[0071] In an example embodiment, the target multi-modal data is re-collected according to the data collection strategy corresponding to the risk prediction result, including: adjusting a collection period of a collection device according to the data collection strategy corresponding to the risk prediction result; and controlling the collection device to re-collect the target multi-modal data according to the adjusted collection period.

[0072] The collection period refers to the interval time of each data collection of the collection device. Different data collection strategies correspond to different collection periods. For example, if the data collection strategy is to increase the collection period, the collection period of the collection device such as the inspection unmanned aerial vehicle, the monitoring camera, the electrical quantity sensor, and the environmental sensor in the power line will be increased by a preset time length. This method of adjusting the collection period according to the data collection strategy realizes the dynamic control of the monitoring strategy of the power line based on risk prediction, controls the collection device to re-collect the target multi-modal data according to the adjusted collection period, and is beneficial to greatly saving energy in the case of less defect hidden danger influence.

[0073] In this embodiment, the data collection strategy obtained by dynamic adjustment corresponds to the adjustment of the collection period of the collection device, realizes the dynamic adjustment of the monitoring strategy of the power line, is beneficial to greatly saving energy, and improves the utilization rate of collection resources.

[0074] In one exemplary embodiment, updating the risk prediction model based on actual faults, hidden danger elimination situations, and target multi-modal data includes: updating the risk prediction result based on the target multi-modal data and the risk prediction model; performing correctness checking on the updated risk prediction result according to the actual faults and the hidden danger elimination situations; inputting the correctness checking result, the actual faults, and the hidden danger elimination situations into a preset reinforcement learning model to obtain a parameter change amount, and updating the risk prediction model based on the parameter change amount.

[0075] Since the collection period of the collection device is adjusted, the collected target multi-modal data changes, and therefore, the risk prediction model can update the risk prediction result by performing risk prediction on the target multi-modal data.

[0076] The updated risk prediction result may differ from the actual fault situation, and therefore, correctness checking can be performed on the updated risk prediction result according to the actual faults and the hidden danger elimination situations. Since the actual faults and the hidden danger elimination situations are obtained by manual checking, for example, if the actual faults, the risk prediction result, and the updated risk prediction result are not identified, it is determined that the correctness checking result is incorrect, and it is necessary to correct the risk prediction model to improve the accuracy of the risk prediction result.

[0077] In some embodiments, the preset reinforcement learning model can be a preset corresponding relationship, and different input data correspond to different parameter change amounts. The terminal inputs the correctness checking result, the actual faults, and the hidden danger elimination situations into the preset reinforcement learning model to obtain the parameter change amount. The parameter change amount is the change amount of each model parameter in the risk prediction model. The terminal adjusts the model parameters of the risk prediction model according to the parameter change amount to obtain the updated risk prediction model.

[0078] In the embodiment, the target multi-modal data re-acquired is subjected to risk prediction by the risk prediction model to update the risk prediction result, and the updated risk prediction result is subjected to correctness checking according to the actual fault condition. Since the updated risk prediction result and the actual fault condition are different, the parameter variation amount can be determined by using the reinforcement learning model to update the risk prediction model according to the parameter variation amount, so that the risk prediction model is more consistent with the real fault condition, and the prediction capability of the risk prediction model is improved.

[0079] In an exemplary embodiment, the correctness checking result, the actual fault and the hidden danger elimination condition are input into a preset reinforcement learning model to obtain the parameter variation amount, including: using a reward and punishment function in the preset reinforcement learning model to pre-process the correctness checking result, the actual fault and the hidden danger elimination condition to obtain a pre-processing result; and converting the pre-processing result into the parameter variation amount based on a value function in the preset reinforcement learning model.

[0080] The reinforcement learning model includes the reward and punishment function and the value function. The reward and punishment function is used to pre-process the correctness checking result, the actual fault and the hidden danger elimination condition input into the reinforcement learning model to obtain a pre-processing result. The pre-processing mainly performs threshold judgment on each data to eliminate abnormal data to obtain a result after elimination of abnormal data as the pre-processing result.

[0081] The value function is a preset function relationship. The pre-processing result can be converted into the parameter variation amount by the value function.

[0082] In the embodiment, the reward and punishment function and the value function in the preset reinforcement learning model are used to process each input data to obtain the parameter variation amount, which is beneficial to determining the parameter variation amount of the risk prediction model according to the risk prediction result of the risk prediction model in combination with the real fault condition, and is used to adjust the model parameters of the risk prediction model to improve the risk prediction capability of the risk prediction model.

[0083] In an exemplary embodiment, the power data acquisition method further includes: acquiring the to-be-processed multi-modal data according to the updated data acquisition strategy; and updating the updated risk prediction model based on the to-be-processed multi-modal data.

[0084] After the risk prediction model is updated, the data acquisition strategy is updated, and the terminal control acquisition device performs data acquisition according to the updated data acquisition strategy to obtain the to-be-processed multi-modal data.

[0085] Since the multi-modal data has changed, the model samples of the risk prediction model are increased, and in the field of risk detection of power lines, the to-be-processed multi-modal data re-acquired can be used as model samples to further update the updated risk prediction model.

[0086] In this embodiment, by using the updated data acquisition strategy to reacquire the to-be-processed multi-modal data after updating the risk prediction model, the to-be-processed multi-modal data is used as a model sample to continue updating the updated risk prediction model, thereby realizing dynamic updating of the risk prediction model, ensuring that the risk prediction model learns more power line risk failure conditions, and being conducive to improving the risk prediction capability of the risk prediction model.

[0087] To illustrate the power data acquisition method and effects in this scheme in detail, a most detailed embodiment is described as follows:

[0088] The power line inspection scene is taken as an example for illustration. As shown in FIG. 3, it is a flowchart of the power data acquisition method in another embodiment. First, multi-modal data of the power line is acquired, hidden danger identification is performed on the multi-modal data to obtain hidden danger identification results to predict the current risk, the data acquisition strategy is dynamically updated, the environment data adaptive acquisition closed loop is formed, and efficient use of the acquisition resources is realized. Meanwhile, a feedback reward and punishment mechanism is established based on the actual fault, hidden danger elimination condition and risk prediction result, an online reinforcement learning method is used to update the risk prediction model, the risk prediction model adaptive updating closed loop is formed, and the acquisition system is guided to focus on the points prone to high-risk and rare defect hidden dangers.

[0089] As shown in FIG. 4, it is a flowchart of the power data acquisition method in another embodiment. The power data acquisition method includes two processes of the environment data adaptive acquisition closed loop and the risk model adaptive updating closed loop.

[0090] 1. The environment data adaptive acquisition closed loop process is as follows:

[0091] 1.1. Visible light images, infrared images, three-dimensional point clouds, current values, voltage values, temperatures and a series of multi-modal data of the power line are acquired through inspection unmanned aerial vehicles, monitoring cameras, electrical quantity sensors, environmental sensors and other devices.

[0092] 1.2. The multi-modal data is transmitted to the identification analysis system, the identification analysis system processes the multi-modal data through a multi-modal large model or multiple small models, and identifies at least one hidden danger of the power line to obtain a hidden danger identification result corresponding to each hidden danger, including possible abnormal conditions in the power line and a power line risk score.

[0093] 1.3. The hidden danger identification result is input into the risk prediction model to obtain a risk prediction matrix, i.e., a risk prediction result, and further abstracted into an acquisition strategy matrix.

[0094] 1.4, automatically generate the data collection strategy of the power line based on the collection strategy matrix, upload it to the power grid system, control and adjust the collection period of the inspection unmanned aerial vehicle, monitoring camera, electrical quantity sensor, environmental sensor and other equipment in the power line, realize dynamic control of the power line monitoring strategy based on risk prediction, and achieve significant energy saving under the condition of less impact on defect hidden danger discovery.

[0095] 2, the risk model adaptive update closed loop process is as follows:

[0096] 2.1, according to the data collection strategy corresponding to the risk prediction result, a series of target multi-modal data such as visible light image, infrared image, three-dimensional point cloud, current value, voltage value, temperature and the like of the power line are obtained through the inspection unmanned aerial vehicle, monitoring camera, electrical quantity sensor, environmental sensor and other equipment.

[0097] 2.2, the target multi-modal data is transmitted to the identification analysis system, the identification analysis system processes each modality data uniformly or respectively through a multi-modal large model or multiple small models, identifies the hidden danger identification result corresponding to each hidden danger of the power line, and updates the risk prediction result based on the hidden danger identification result and the risk prediction model.

[0098] 2.3, according to the actual fault and hidden danger elimination situation, the accuracy of the risk prediction result and the updated risk prediction result is judged.

[0099] 2.4, the online reinforcement learning model includes reward and punishment functions and value functions, the correctness check result, the actual fault and the hidden danger elimination situation are input into the online reinforcement learning model as high-risk feedback reward and punishment, the value function is updated, and then the risk prediction model is updated, realizing the online dynamic update of the risk prediction model.

[0100] 2.5, using the updated risk prediction model to evaluate the system risk for the subsequent collected current multi-modal data, adaptively updating the data collection strategy, and the updated data collection strategy is used for multi-modal data collection.

[0101] The power data acquisition method, by the multi-modal data of the power line and the preset risk prediction model, obtains a risk prediction result, and reacquires target multi-modal data according to a data acquisition strategy corresponding to the risk prediction result. The target multi-modal data is reacquired in combination with the risk prediction result of the power line, and therefore, the utilization rate of the acquisition resource is improved. The risk prediction model can be continuously updated according to actual faults, hidden danger elimination conditions and target multi-modal data, and the risk detection accuracy of the risk prediction model for the power line is improved. The risk detection result detected by the updated risk prediction model is used to update the data acquisition strategy, the dynamic adjustment of the data acquisition strategy is realized, the updated data acquisition strategy is used for multi-modal data acquisition, the adaptive acquisition closed loop of the power data is formed, and the utilization rate of the acquisition resource is improved. The dynamic adaptive acquisition architecture based on risk driving is established, and the balance between low-power active sensing and high-reliability risk prevention is realized. Since there are not many actual failure data of the power line, and the risk data before the failure is even more scarce, the online reinforcement learning method can realize the dynamic update of the risk prediction model, and realizes the double closed loop of adaptive acquisition of environmental data and adaptive update of the risk model.

[0102] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0103] Based on the same inventive concept, the embodiments of the present application also provide a power data acquisition device for implementing the power data acquisition method as described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, and therefore, the specific limitations in one or more power data acquisition device embodiments provided below can refer to the limitations of the power data acquisition method described above, which will not be described here.

[0104] In one exemplary embodiment, as shown in FIG. 5, a power data acquisition device 100 is provided, comprising: an acquisition module 120, a first update module 140 and a second update module 160, wherein:

[0105] The acquisition module 120 is configured to determine a risk prediction result based on the multi-modal data of the power line and a preset risk prediction model. The multi-modal data includes image data, temperature data, and electrical quantity data.

[0106] The first updating module 140 is configured to reacquire target multi-modal data according to a data acquisition strategy corresponding to the risk prediction result, and update the risk prediction model based on actual faults, hidden danger elimination conditions, and the target multi-modal data.

[0107] The second updating module 160 is configured to acquire current multi-modal data, perform risk prediction on the current multi-modal data based on the updated risk prediction model to obtain a risk detection result of the power line, and update the data acquisition strategy according to the risk detection result. The updated data acquisition strategy is used for multi-modal data acquisition.

[0108] The power data acquisition device described above determines a risk prediction result based on the multi-modal data of the power line and a preset risk prediction model, reacquires target multi-modal data according to a data acquisition strategy corresponding to the risk prediction result, and updates the risk prediction model based on actual faults, hidden danger elimination conditions, and the target multi-modal data. Therefore, the utilization rate of the acquisition resource is improved. The risk prediction model can be continuously updated based on actual faults, hidden danger elimination conditions, and the target multi-modal data, and the risk detection accuracy of the risk prediction model for the power line is improved. The risk detection result detected by using the updated risk prediction model is used to update the data acquisition strategy, the data acquisition strategy is dynamically adjusted, the updated data acquisition strategy is used for multi-modal data acquisition, a self-adaptive acquisition closed loop of power data is formed, and the utilization rate of the acquisition resource is improved.

[0109] In one embodiment, the acquisition module 120 is further configured to: perform hidden danger identification on the multi-modal data of the power line by using a preset identification analysis system to obtain at least one hidden danger identification result corresponding to each of the at least one hidden danger; and input the at least one hidden danger identification result corresponding to each of the at least one hidden danger into the preset risk prediction model to obtain the risk prediction result.

[0110] In one embodiment, the first updating module 140 is further configured to: adjust a collection period of the collection device according to the data acquisition strategy corresponding to the risk prediction result; and control the collection device to reacquire the target multi-modal data according to the adjusted collection period.

[0111] In an embodiment, based on the actual fault, the hidden danger elimination situation, and the target multi-modal data, the first updating module 140 is further configured to update the risk prediction result based on the target multi-modal data and the risk prediction model, perform correctness checking on the updated risk prediction result according to the actual fault and the hidden danger elimination situation, input the correctness checking result, the actual fault, and the hidden danger elimination situation into a preset reinforcement learning model to obtain a parameter change amount, and update the risk prediction model based on the parameter change amount.

[0112] In an embodiment, the first updating module 140 is further configured to: perform preprocessing on the correctness checking result, the actual fault, and the hidden danger elimination situation by using a reward and punishment function in the preset reinforcement learning model to obtain a preprocessing result, and convert the preprocessing result into the parameter change amount based on a value function in the preset reinforcement learning model.

[0113] In an embodiment, the power data acquisition device further includes a third updating module, and the third updating module is configured to: acquire the to-be-processed multi-modal data according to the updated data acquisition strategy, and update the updated risk prediction model based on the to-be-processed multi-modal data.

[0114] The above modules in the power data acquisition device can be all or partially implemented by software, hardware, or a combination thereof. The above modules can be embedded in or independent of a processor in a computer device in a hardware form, or can be stored in a memory in the computer device in a software form, so as to be called and executed by a processor to perform operations corresponding to the above modules.

[0115] In an exemplary embodiment, a computer device is provided, which can be a terminal, and an internal structure diagram thereof can be as shown in FIG. 6. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange data between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be achieved through WIFI, mobile cellular network, near field communication (Near Field Communication, NFC) or other technologies. The computer program is executed by the processor to implement a power data acquisition method. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0116] Those skilled in the art can understand that the structure shown in FIG. 6 is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0117] In an exemplary embodiment, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the following steps:

[0118] Based on the multi-modal data of the power line and the preset risk prediction model, a risk prediction result is determined; the multi-modal data includes image data, temperature data and electrical quantity data; target multi-modal data is re-acquired according to a data acquisition strategy corresponding to the risk prediction result, the risk prediction model is updated based on actual faults, hidden danger elimination and the target multi-modal data; current multi-modal data is acquired, the current multi-modal data is risk predicted based on the updated risk prediction model, a risk detection result of the power line is obtained, and the data acquisition strategy is updated according to the risk detection result, and the updated data acquisition strategy is used for multi-modal data acquisition.

[0119] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0120] The preset identification analysis system is used to identify hidden dangers of the power line based on the multi-modal data, to obtain at least one hidden danger of the power line corresponding to a hidden danger identification result; and the at least one hidden danger corresponding to the hidden danger identification result is input into a preset risk prediction model to obtain a risk prediction result.

[0121] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0122] According to the data acquisition strategy corresponding to the risk prediction result, the acquisition cycle of the acquisition device is adjusted; and the target multi-modal data is re-acquired by the acquisition device according to the adjusted acquisition cycle.

[0123] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0124] Based on the target multi-modal data and the risk prediction model, the risk prediction result is updated; the updated risk prediction result is checked for correctness according to actual faults and hidden danger elimination conditions; the correctness checking result, the actual faults and the hidden danger elimination conditions are input into a preset reinforcement learning model to obtain a parameter change amount, and the risk prediction model is updated based on the parameter change amount.

[0125] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0126] The correctness checking result, the actual faults and the hidden danger elimination conditions are preprocessed by using a reward and punishment function in the preset reinforcement learning model to obtain a preprocessing result; and the preprocessing result is converted into a parameter change amount based on a value function in the preset reinforcement learning model.

[0127] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0128] The to-be-processed multi-modal data is acquired according to the updated data acquisition strategy; and the updated risk prediction model is updated based on the to-be-processed multi-modal data.

[0129] In one embodiment, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the following steps:

[0130] Determine a risk prediction result based on power line multi-modal data and a preset risk prediction model; the multi-modal data includes image data, temperature data, and electrical quantity data; obtain target multi-modal data according to a data acquisition strategy corresponding to the risk prediction result; update the risk prediction model based on actual faults, hidden danger elimination situations, and the target multi-modal data; obtain current multi-modal data, perform risk prediction on the current multi-modal data based on the updated risk prediction model, obtain a risk detection result of the power line, and update the data acquisition strategy according to the risk detection result; and the updated data acquisition strategy is used for multi-modal data acquisition.

[0131] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0132] Perform hidden danger identification on the multi-modal data of the power line by using a preset identification analysis system, and obtain at least one hidden danger identification result corresponding to each of the at least one hidden danger of the power line; input the at least one hidden danger identification result corresponding to each of the at least one hidden danger into a preset risk prediction model, and obtain a risk prediction result.

[0133] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0134] Adjust a collection cycle of the collection device according to a data acquisition strategy corresponding to the risk prediction result; and control the collection device to re-collect target multi-modal data according to the adjusted collection cycle.

[0135] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0136] Update the risk prediction result based on the target multi-modal data and the risk prediction model; perform correctness checking on the updated risk prediction result according to actual faults and hidden danger elimination situations; input the correctness checking result, the actual faults, and the hidden danger elimination situations into a preset reinforcement learning model, obtain a parameter change amount, and update the risk prediction model based on the parameter change amount.

[0137] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0138] Preprocess the correctness checking result, the actual faults, and the hidden danger elimination situations by using a reward and punishment function in the preset reinforcement learning model, and obtain a preprocessing result; and convert the preprocessing result into a parameter change amount based on a value function in the preset reinforcement learning model.

[0139] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0140] The multi-modal data to be processed is collected according to the updated data collection strategy; and the updated risk prediction model is updated based on the multi-modal data to be processed.

[0141] In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the following steps:

[0142] Based on the multi-modal data of the power line and the preset risk prediction model, a risk prediction result is determined; the multi-modal data includes image data, temperature data and electrical quantity data; target multi-modal data is re-collected according to a data collection strategy corresponding to the risk prediction result, the risk prediction model is updated based on actual faults, hidden danger elimination conditions and the target multi-modal data; current multi-modal data is obtained, the risk prediction model is updated based on the updated risk prediction model, a risk detection result of the power line is obtained, and the data collection strategy is updated according to the risk detection result, and the updated data collection strategy is used for multi-modal data collection.

[0143] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0144] The multi-modal data of the power line is identified by using a preset identification analysis system, and at least one hidden danger identification result corresponding to each hidden danger of the power line is obtained; the at least one hidden danger identification result corresponding to each hidden danger is input into a preset risk prediction model, and a risk prediction result is obtained.

[0145] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0146] The collection cycle of the collection device is adjusted according to a data collection strategy corresponding to the risk prediction result; and the collection device is controlled to re-collect target multi-modal data according to the adjusted collection cycle.

[0147] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0148] Based on the target multi-modal data and the risk prediction model, the risk prediction result is updated; the updated risk prediction result is checked for correctness according to actual faults and hidden danger elimination conditions; the correctness checking result, the actual faults and the hidden danger elimination conditions are input into a preset reinforcement learning model, a parameter change amount is obtained, and the risk prediction model is updated based on the parameter change amount.

[0149] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0150] The reward and punishment function in the preset reinforcement learning model is used to preprocess the correctness check result, actual fault and hidden danger elimination situation to obtain a preprocessing result; and the preprocessing result is converted into a parameter variation based on the value function in the preset reinforcement learning model.

[0151] In one embodiment, the computer program, when executed by the processor, also implements the following steps:

[0152] The to-be-processed multi-modal data is collected according to the updated data collection strategy; and the updated risk prediction model is updated based on the to-be-processed multi-modal data.

[0153] It should be noted that the user data (including but not limited to user equipment data, user personal data, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all data and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0154] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0155] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combinations of the technical features do not exist contradictory, they should be considered as the scope of the present application.

[0156] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A method of electric power data acquisition, characterized by, The method comprises: determining a risk prediction result based on the multi-modal data of the power line and a preset risk prediction model; the multi-modal data comprises image data, temperature data and electrical quantity data; recollecting target multi-modal data according to a data collection strategy corresponding to the risk prediction result, and updating the risk prediction model based on actual faults, hidden danger elimination and the target multi-modal data; obtaining current multi-modal data, performing risk prediction on the current multi-modal data based on the updated risk prediction model to obtain a risk detection result of the power line, and updating the data collection strategy according to the risk detection result, wherein the updated data collection strategy is used for multi-modal data collection.

2. The method of claim 1, wherein, The method comprises: obtaining at least one hidden danger identification result corresponding to each hidden danger of the power line by identifying the multi-modal data of the power line using a preset identification analysis system; inputting the at least one hidden danger identification result corresponding to each hidden danger into a preset risk prediction model to obtain a risk prediction result.

3. The method of claim 1, wherein, The method comprises: adjusting the collection period of the collection device according to the data collection strategy corresponding to the risk prediction result; controlling the collection device to recollect target multi-modal data according to the adjusted collection period.

4. The method of claim 1, wherein, The method comprises: updating the risk prediction result based on the target multi-modal data and the risk prediction model; checking the correctness of the updated risk prediction result according to actual faults and hidden danger elimination; inputting the correctness checking result, actual faults and hidden danger elimination into a preset reinforcement learning model to obtain a parameter change amount, and updating the risk prediction model based on the parameter change amount.

5. The method of claim 4, wherein, The method comprises: preprocessing the correctness checking result, actual faults and hidden danger elimination using a reward and punishment function in the preset reinforcement learning model to obtain a preprocessing result; transforming the preprocessing result into a parameter change amount based on a value function in the preset reinforcement learning model.

6. The method of claim 1, wherein, The method further comprises: collecting to-be-processed multi-modal data according to the updated data collection strategy; updating the updated risk prediction model based on the to-be-processed multi-modal data.

7. An electric power data acquisition device, characterized by The device comprises: an acquisition module configured to determine a risk prediction result based on multi-modal data of a power line and a preset risk prediction model; the multi-modal data comprises image data, temperature data and electrical quantity data; a first updating module configured to recollect target multi-modal data according to a data collection strategy corresponding to the risk prediction result, and update the risk prediction model based on actual faults, hidden danger elimination and the target multi-modal data. A second updating module is configured to acquire current multi-modal data, perform risk prediction on the current multi-modal data based on the updated risk prediction model to obtain a risk detection result of the power line, and update the data collection strategy according to the risk detection result, wherein the updated data collection strategy is used for multi-modal data collection.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 6. The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 6.

Images