Mountain fire identification and monitoring method for power transmission line environment and related equipment

Abstract

The invention relates to a forest fire identification and monitoring method for a power transmission line environment and related equipment. The method comprises the following steps: monitoring sensor data of a to-be-monitored power transmission line environment in real time; performing environment analysis on the sensor data by using a preset vector machine to obtain the environment condition of the to-be-tested power transmission line environment; determining a shooting strategy of a camera located in the to-be-tested power transmission line environment based on the environment condition, and controlling the camera to shoot the to-be-tested power transmission line environment based on the shooting strategy to obtain a target image; performing feature recognition on the target image by using a preset YOLO model, and determining a forest fire condition of the to-be-detected power transmission line environment; and determining a target emergency response mechanism based on the forest fire condition, and using the target emergency response mechanism to deal with the forest fire occurring in the to-be-tested power transmission line environment. According to the method, the forest fire condition of the power transmission line environment can be identified and monitored in time while resources are saved.

Description

Background Art

[0002] Transmission lines pass through complex terrain areas such as high mountains and steep hills. Due to high vegetation coverage, accumulation of combustible materials and great influence of climatic conditions, they are prone to cause wildfires. The complex terrain and inconvenient transportation make management and maintenance difficult, increasing the risk of wildfires. Wildfires are more likely to occur in the afternoon, especially in spring and autumn when the climate is dry, the temperature is high and the wind is strong.

[0003] With the widespread promotion and in-depth application of image video monitoring technology, it has become possible to use video images to implement all-weather, no-blind-angle monitoring of transmission lines and their surroundings. The application of this technology not only significantly reduces the frequency of manual inspections, but also effectively reduces the occurrence of potential accidents, greatly improving the safety and stability of transmission line operation.

[0004] However, due to the particularity of the working environment of the transmission line, all-weather monitoring means that the equipment needs to run continuously, which undoubtedly brings huge energy consumption challenges. In most cases, the transmission channel is in normal operation, and a large number of video monitoring equipment is idle, occupying a large amount of communication traffic resources. In addition, most of the current transmission line video monitoring equipment is powered by solar energy and batteries, which makes it difficult to ensure long-term continuous and stable operation of the equipment. Therefore, there is an urgent need for a method that can save resources while being able to timely identify and monitor wildfire conditions in the transmission line environment in order to cope with the energy consumption challenges brought about by all-weather monitoring, which is of vital importance to ensuring the stable operation of the transmission line monitoring system. Summary of the invention

[0005] In order to solve the above-mentioned problems, the present application provides a method and related equipment for identifying and monitoring wildfires in a power transmission line environment.

[0006] In the first aspect, in order to solve the above technical problems, the present application provides a method for identifying and monitoring wildfires in a power transmission line environment, comprising:

[0007] Real-time monitoring of sensor data of the transmission line environment to be tested;

[0008] Use the preset vector machine to perform environmental analysis on the sensor data to obtain the environmental conditions of the transmission line environment to be tested;

[0009] Based on the environmental conditions, determine the shooting strategy of the camera located in the environment of the power transmission line to be tested, and control the camera to shoot the environment of the power transmission line to be tested based on the shooting strategy to obtain a target image;

[0010] Use the preset YOLO model to perform feature recognition on the target image and determine the wildfire situation of the power transmission line environment to be tested;

[0011] The target emergency response mechanism is determined based on the wildfire situation, so as to utilize the target emergency response mechanism to respond to wildfires occurring in the environment of the transmission line to be tested.

[0012] On the second aspect, the present application also provides a wildfire identification and monitoring device for a transmission line environment, including a sensor, a camera and a processor. The sensor is used to monitor the sensor data of the transmission line environment to be tested in real time, the camera is used to shoot the transmission line environment to be tested based on a shooting strategy, and the processor is used to implement a wildfire identification and monitoring method for a transmission line environment.

[0013] In a third aspect, the present application also provides a wildfire identification and monitoring system for a power transmission line environment, comprising:

[0014] A real-time monitoring module, used to monitor sensor data of the power transmission line environment to be tested in real time;

[0015] An analysis module, used to perform environmental analysis on sensor data using a preset vector machine to obtain environmental conditions of the power transmission line environment to be tested;

[0016] A shooting module is used to determine a shooting strategy of a camera located in the environment of the power transmission line to be tested based on the environmental conditions, and control the camera to shoot the environment of the power transmission line to be tested based on the shooting strategy to obtain a target image;

[0017] The recognition module is used to perform feature recognition on the target image using a preset YOLO model to determine the wildfire situation of the power transmission line environment to be tested;

[0018] The determination module is used to determine the target emergency response mechanism based on the wildfire situation, so as to use the target emergency response mechanism to respond to the wildfire occurring in the environment of the transmission line to be tested.

[0019] In a fourth aspect, the present application also provides a computing device, including a memory, a processor, and a program stored in the memory and running on the processor, and when the processor executes the program, the steps of a method for identifying and monitoring wildfires in a power transmission line environment as described above are implemented.

[0020] In a fifth aspect, the present application also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal device, the terminal device executes the steps of a method for identifying and monitoring wildfires in a power transmission line environment.

[0021] The beneficial effects of the present application are as follows: the sensor data is the environmental data of the transmission line environment to be tested collected in real time by a variety of sensors. First, by real-time monitoring of the sensor data of the transmission line environment to be tested, comprehensive monitoring of the transmission line environment to be tested is achieved. Secondly, based on the environmental conditions obtained by environmental analysis of the sensor data by a preset vector machine, the shooting strategy of the camera located in the transmission line environment to be tested is determined, and the camera is controlled to shoot the transmission line environment to be tested based on the shooting strategy to obtain the target image, so that the camera only needs to intermittently shoot the transmission line environment to be tested based on the shooting strategy, and there is no need to shoot all day, which effectively reduces the monitoring energy consumption of the camera. Then, the preset YOLO model is used to perform feature recognition on the target image, determine the wildfire situation of the transmission line environment to be tested, and determine the target emergency response mechanism based on the wildfire situation, so that the wildfire situation of the transmission line environment to be tested can be monitored in time, so as to use the target emergency response mechanism to respond to the wildfire that occurs in the transmission line environment to be tested. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic flow chart of a method for identifying and monitoring wildfires in a power transmission line environment is shown as an exemplary embodiment of the present application;

[0023] Figure 2 A flowchart of a wildfire identification and monitoring method for a power transmission line environment provided in an exemplary embodiment of the present application is provided;

[0024] Figure 3 This is a schematic structural diagram of a wildfire identification and monitoring device for a power transmission line environment in an exemplary embodiment of the present application;

[0025] Figure 4 This is a schematic diagram of an application scenario of a wildfire identification and monitoring device in a power transmission line environment in an exemplary embodiment of the present application;

[0026] Figure 5 A schematic structural diagram of a wildfire identification and monitoring system for a power transmission line environment is shown as an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0027] The following examples are provided to further explain and supplement the present application and do not constitute any limitation to the present application.

[0028] The following describes a method for identifying and monitoring wildfires in a power transmission line environment and related equipment according to an embodiment of the present application in conjunction with the accompanying drawings.

[0029] An embodiment of the present application provides a method for identifying and monitoring wildfires in a power transmission line environment. The method is applied to a terminal device. In the present application scheme, the terminal device is used as the execution subject to illustrate the present application scheme. The terminal device is used to execute the steps of a method for identifying and monitoring wildfires in a power transmission line environment.

[0030] See also Figure 1 , Figure 1 A method for identifying and monitoring wildfires in a power transmission line environment is shown as an exemplary embodiment of the present application. Figure 1 As shown, the present application provides a method for identifying and monitoring wildfires in a power transmission line environment, comprising:

[0031] Step S11, real-time monitoring of sensor data of the power transmission line environment to be tested;

[0032] Step S12, using a preset vector machine to perform environmental analysis on the sensor data to obtain the environmental conditions of the power transmission line to be tested;

[0033] Step S13, determining a shooting strategy of a camera located in the environment of the power transmission line to be tested based on the environmental conditions, and controlling the camera to shoot the environment of the power transmission line to be tested based on the shooting strategy to obtain a target image;

[0034] Step S14, using a preset YOLO model to perform feature recognition on the target image to determine the wildfire situation of the power transmission line environment to be tested;

[0035] Step S15, determining a target emergency response mechanism based on the wildfire situation, so as to utilize the target emergency response mechanism to respond to the wildfire occurring in the environment of the transmission line to be tested.

[0036] In a method for identifying and monitoring wildfires in a power transmission line environment of the present embodiment, sensor data is environmental data of the power transmission line environment to be tested collected in real time by a variety of sensors. Sensor data is environmental data of the power transmission line environment to be tested collected in real time by a variety of sensors. First, by real-time monitoring of the sensor data of the power transmission line environment to be tested, comprehensive monitoring of the power transmission line environment to be tested is achieved. Secondly, based on the environmental conditions obtained by environmental analysis of the sensor data by a preset vector machine, the shooting strategy of the camera located in the power transmission line environment to be tested is determined, and the camera is controlled to shoot the power transmission line environment to be tested based on the shooting strategy to obtain the target image, so that the camera only needs to intermittently shoot the power transmission line environment to be tested based on the shooting strategy, and there is no need to shoot all day, which effectively reduces the monitoring energy consumption of the camera. Then, the preset YOLO model is used to perform feature recognition on the target image, determine the wildfire situation of the power transmission line environment to be tested, and determine the target emergency response mechanism based on the wildfire situation, so that the wildfire situation of the power transmission line environment to be tested can be monitored in time, so as to use the target emergency response mechanism to respond to the wildfire that occurs in the power transmission line environment to be tested.

[0037] In an exemplary embodiment, a plurality of sensors include temperature sensors, humidity sensors, smoke sensors, thermal imaging sensors, and current sensors, and a plurality of sensors are arranged in the environment of the transmission line to be tested, and the operation status of the transmission line in the environment of the transmission line to be tested can be fully monitored through a plurality of sensors. The preset vector machine and the preset YOLO model are both stored in the host computer. The sensor data of the transmission line environment to be tested collected in real time by a plurality of sensors are transmitted to the host computer through the LoRa communication technology, so that these sensor data are stored in the environment data sample set in the host computer, and the preset vector machine and the preset YOLO model are used for higher-level data processing and analysis, so as to realize real-time wildfire monitoring of the transmission line environment to be tested, and timely respond to wildfires in the transmission line environment to be tested. The LoRa communication technology has low power consumption, long-distance transmission and anti-interference capabilities, and is suitable for reliable data transmission in complex environments.

[0038] Optionally, a preset vector machine is used to perform environmental analysis on the sensor data to obtain the environmental conditions of the power transmission line environment to be tested, including:

[0039] Use a preset vector machine to extract features from sensor data to obtain a feature vector;

[0040] Analyze the environment of the transmission line environment to be tested based on the characteristic vector to obtain the environmental state of the transmission line environment to be tested;

[0041] Calculate the wildfire risk probability of the transmission line environment to be tested using the environmental status;

[0042] The formula for calculating wildfire risk probability is as follows:

[0043]

[0044] Wherein, P(y=1|x) represents the wildfire risk probability, x represents the sensor data, f(x) represents the environmental state obtained by classifying x using a preset vector machine, and A and B represent scaling parameters obtained by fitting the training set data of the preset vector machine;

[0045] The environmental conditions of the transmission line environment to be tested are formed based on the environmental status and wildfire risk probability.

[0046] In this embodiment, the feature vector obtained by extracting features from sensor data using a preset vector machine, and the optimal decision boundary constructed by the preset vector machine are used to analyze the environment of the power transmission line environment to be tested, so that the changing trend of the power transmission line environment to be tested can be understood, and the environmental state corresponding to the changing trend can be obtained. The wildfire risk probability of the power transmission line environment to be tested is calculated using the environmental state, and the environmental condition of the power transmission line environment to be tested is obtained, which is convenient for the subsequent determination of the shooting strategy of the camera located in the power transmission line environment to be tested based on the environmental condition, so as to determine the wildfire condition based on the target image, so as to realize real-time wildfire monitoring of the power transmission line environment to be tested. Among them, the preset vector machine is a support vector machine (SVM).

[0047] In an exemplary embodiment, first, in order to scale the data of different features to the same range so as to be more effectively processed in the machine learning model, the data of different features need to be normalized. Then, a support vector machine (SVM) is used to extract features from the sensor data, and the single feature vectors of all sensor data are sequentially concatenated into a long feature vector, and then the long feature vector is normalized into a feature vector.

[0048] The calculation formula for normalizing a long eigenvector to a eigenvector is as follows:

[0049] x' i ={x i -x min} / {x max -x min};

[0050] Among them, in the formula: x' i is the normalized eigenvector, x i is the long eigenvector before normalization, x min and x max are the minimum and maximum eigenvectors in the long eigenvector before normalization.

[0051] Secondly, the process of using support vector machine (SVM) to analyze the environment of the transmission line environment to be tested based on feature vectors and obtaining the environmental state of the transmission line environment to be tested is actually a process of multi-data feature fusion classification. In the process of multi-data feature fusion classification using support vector machine (SVM), when the existing spatial data is linearly inseparable, the data is mapped to a high-dimensional space, the inseparable data set is converted into a separable data set, and a hyperplane is created so that the distance from the points closest to the plane to the hyperplane is the largest for the two sets of data. The support vector machine (SVM) is a vector machine trained using data sets from N different sensors, and its training process is as follows:

[0052] Get data sets from N different sensors. For the i-th sensor, the feature vector of its data set can be expressed as where x ij is the jth feature vector of the ith sensor, m i is the number of data samples of the i-th sensor.

[0053] The feature vectors of N different sensors are integrated into a unified feature space. The feature vector X after feature fusion fused Expressed as N is the number of fused feature vectors;

[0054] The feature vector X fused As the input of SVM, it is trained and classified, and the goal of SVM is an optimal hyperplane to separate data points of different categories f(x) = w T Φ(x)+b, where f(x) represents the eigenvector X fused Classification result: environment state, f(x) reflects the distance between the input data and the SVM hyperplane. The larger the value, the higher the confidence of the model in the classification. Φ(x) is the distance between the feature vector X and the SVM hyperplane. fused The kernel function is mapped to a high-dimensional space (possibly infinite), w is the normal vector of the hyperplane, and b is the bias term. The nonlinear vector space of the data feature is mapped to a linearly separable space. Once the hyperplane is determined, the w value and the b value can be infinitely enlarged or reduced in proportion. Therefore, the geometric interval is introduced. To solve this problem, γ is x i The distance to the hyperplane, y i Represents x i The hyperplane can be obtained by solving the following optimization problem:

[0055]

[0056] sty i (w T x i +b)-1≥0,i=1,...,N;

[0057] For the above optimization problem, Lagrange duality is usually used to transform it into a dual problem for solution. Construct the Lagrangian function and introduce the Lagrangian multiplier α i ≥0, we get the dual problem. By solving the dual problem, we can get the optimal solution w of the original problem. * and b * , and then obtain the classification decision function and use the Lagrange multiplier method to solve sty i (w T x i +b)-1≥0,i=1,...,N, the process is as follows:

[0058]

[0059] Finally, find w * , b * :

[0060]

[0061]

[0062] The optimal classification function can be obtained

[0063] When training the SVM model, we use labeled historical data samples to mark "non-wildfire risk state" and "wildfire risk state" respectively, and use these samples to train the SVM model to obtain an SVM classification model that can distinguish between non-wildfire risk state and wildfire risk state corresponding to wildfire precursors. Through model training and verification, we adjust the network parameters of the SVM model (such as the penalty parameter CCC and kernel function parameters) to improve the classification accuracy and generalization ability of the model.

[0064] Optionally, the environmental conditions include environmental status and wildfire risk probability; based on the environmental conditions, determining a shooting strategy of a camera located in the power transmission line environment to be tested includes:

[0065] When the environmental state is a wildfire risk state and the wildfire risk probability is greater than a probability threshold, determining that the shooting strategy of the camera located in the power transmission line environment to be tested is an instant shooting strategy;

[0066] When the environmental state is a wildfire risk state and the wildfire risk probability is less than or equal to the probability threshold, determining that the shooting strategy of the camera is a first timed shooting strategy;

[0067] When the environmental state is a non-wildfire risk state, determining the shooting strategy of the camera to be a second timed shooting strategy;

[0068] The shooting time interval of the first timing shooting strategy is smaller than the shooting time interval of the second timing shooting strategy.

[0069] In this embodiment, when the environmental state is a wildfire risk state, and the wildfire risk probability is greater than the probability threshold, it means that a wildfire is very likely to occur or has already occurred in the environment of the transmission line to be tested. At this time, the camera located in the environment of the transmission line to be tested must shoot immediately to facilitate the staff to timely and at any time understand the actual situation of the environment of the transmission line to be tested, so as to control the spread of the wildfire in advance. When the environmental state is a wildfire risk state, and the wildfire risk probability is less than or equal to the probability threshold, and when the environmental state is a non-wildfire risk state, it means that no wildfire has occurred in the environment of the transmission line to be tested or there is a very small possibility of a wildfire. It only needs to select different timed shooting strategies in a targeted manner to meet the real-time monitoring needs of the environment of the transmission line to be tested. Therefore, the camera only needs to intermittently shoot the environment of the transmission line to be tested based on the shooting strategy, and there is no need to shoot all day, which effectively reduces the monitoring energy consumption of the camera.

[0070] In an exemplary embodiment, the image resolution when the camera uses the instant shooting strategy is higher than the image resolution when the camera uses the first timed shooting strategy and the second timed shooting strategy. The image resolution when the camera uses the first timed shooting strategy and the second timed shooting strategy is the same.

[0071] Optionally, a preset YOLO model is used to perform feature recognition on the target image to determine the wildfire situation of the power transmission line environment to be tested, including:

[0072] The preset YOLO algorithm is used to identify the wildfire features in the target image, and the wildfire feature prediction box and the confidence of the wildfire feature prediction box are obtained;

[0073] Identify the wildfire feature prediction frame to obtain the wildfire category of the transmission line environment to be tested;

[0074] Based on the wildfire category and confidence level, the wildfire occurrence probability of the transmission line environment to be tested is obtained;

[0075] Based on the wildfire category and wildfire occurrence probability, the wildfire situation of the transmission line environment to be tested is formed.

[0076] In this embodiment, the wildfire features include wildfire-related features such as flames and smoke. By using the trained preset YOLO model to directly identify the wildfire features in the target image, the wildfire category and wildfire occurrence probability of the transmission line environment to be tested can be directly obtained with the accuracy that meets the requirements, which is convenient for the subsequent determination of the target emergency response mechanism according to the wildfire situation, so as to use the target emergency response mechanism to respond to the wildfire occurring in the transmission line environment to be tested, which can enhance the safety monitoring effect of the transmission line environment to be tested, and at the same time improve the efficiency of emergency response to the transmission line environment to be tested.

[0077] Optionally, the wildfire feature prediction frame is identified to obtain the wildfire category of the transmission line environment to be tested, including:

[0078] Perform category recognition on the wildfire feature prediction frame to obtain the feature category of the wildfire feature prediction frame;

[0079] The feature category is used to perform a table lookup operation in a preset category table to obtain the wildfire category of the power transmission line environment to be tested.

[0080] In this embodiment, the feature categories include wildfire-related feature categories such as flame category and smoke category. The trained preset YOLO model can directly obtain the wildfire category of the power transmission line environment to be tested. The trained model can automatically predict the wildfire category, which can not only improve the detection accuracy of the wildfire situation of the power transmission line environment to be tested, but also improve the detection efficiency, and meet the timeliness of monitoring the wildfire situation of the power transmission line environment to be tested.

[0081] Optionally, based on the wildfire category and confidence level, the wildfire occurrence probability of the transmission line environment to be tested is obtained, including:

[0082] Calculate predicted probabilities for wildfire categories;

[0083] The formula for calculating the predicted probability is as follows:

[0084]

[0085] Among them, P(class=fire|x) represents the predicted probability, x i Indicates the K Assume that the YOLO model has an unnormalized prediction score for i wildfire categories, K represents the number of wildfire categories, is the exponential sum of the prediction scores for all wildfire categories, so that the sum of the probabilities of all wildfire categories is 1;

[0086] The probability of wildfire occurrence in the transmission line environment to be tested is calculated using the predicted probability and confidence level;

[0087] The calculation formula for the probability of wildfire occurrence is as follows:

[0088] P(fire|x)=C·P(class=fire|x);

[0089] Among them, P(fire|x) represents the probability of wildfire occurrence, C represents the confidence, which represents the confidence of the preset YOLO model on whether the detected target is a real object, and P(class=fire|x) represents the predicted probability.

[0090] In this embodiment, the trained preset YOLO model is used to obtain the probability of wildfire occurrence in the power transmission line environment to be tested based on the wildfire category and confidence level. The trained model is used to automatically predict the probability of wildfire occurrence in the power transmission line environment to be tested, which can not only improve the detection accuracy of the wildfire situation in the power transmission line environment to be tested, but also improve the detection efficiency, thereby meeting the timeliness of monitoring the wildfire situation in the power transmission line environment to be tested.

[0091] In an exemplary embodiment, the preset YOLO model uses an image analysis algorithm to confirm the wildfire situation in the target image. The preset YOLO model can effectively detect wildfire-related features such as flames and smoke in the image. The preset YOLO model is a trained model, and the sample set used to train the preset YOLO model is a data set containing various wildfire flames and smoke images. Each sample image in the data set is annotated with wildfire features so that the model can learn wildfire features such as wildfire flames and smoke. At the same time, the data set covers smoke conditions or wildfire flame conditions at different times and under different weather conditions to enhance the generalization ability of the model.

[0092] When training the preset YOLO model, a total loss function is used to guide the training process of the neural network. The prediction performance of the model is improved by optimizing the loss function. The function comprehensively considers the location of the wildfire feature prediction box, feature category, and wildfire occurrence probability. The formula for the total loss is as follows:

[0093] Total Loss=Position Loss+Dimension Loss+Object ConfidenceLoss+ClassLoss;

[0094] Among them, Total Loss represents the total loss, Position Loss represents the position loss, Dimension Loss represents the size loss, Object Confidence Loss represents the confidence loss, and Class Loss represents the category loss;

[0095] The position loss is calculated using the mean squared error (MSE) by comparing the difference between the predicted box and the actual target box. For each bounding box, the position loss is calculated as follows:

[0096]

[0097] Among them, Position Loss represents the position loss, λ coord represents the weight of coordinate loss, S 2 represents the number of grid cells, B is the number of bounding boxes predicted for each grid cell, is an indicator function indicating whether the jth bounding box in the i-th grid cell is responsible for detecting the target. The mean square error is calculated for each pixel box and the number of predicted boxes. x1 and y1 represent the x-axis and y-axis positions of the predicted position, respectively, and x and y represent the x-axis and y-axis positions of the actual position, respectively.

[0098] Dimension Regression Loss measures the difference between the width and height of the predicted bounding box and the actual target bounding box. The calculation formula of the dimension loss is as follows:

[0099]

[0100] Among them, Dimension Loss represents the size loss, λ coord represents the weight of the size loss, ω1 and h1 represent the width and height of the predicted bounding box, respectively, and ω and h represent the width and height of the actual target bounding box.

[0101] Object Confidence Loss is used to measure the accuracy of the model's confidence prediction of whether the bounding box contains an object. The calculation formula for size loss is as follows:

[0102]

[0103] Among them, Object Confidence Loss represents the confidence loss, C1 represents the confidence that the predicted bounding box contains the target, and C is the label of the actual target. is an indicator function indicating whether the jth bounding box in the i-th grid cell is responsible for detecting the target, λ noobj Represents the weight of the confidence loss.

[0104] Class Probability Loss measures the difference between the model's prediction of the target category and the actual target category. The calculation formula for class loss is as follows:

[0105]

[0106] Among them, Class Loss represents the category loss, λ class represents the weight of the category loss, p1(c) represents the target category probability predicted by the jth bounding box in the i-th grid cell, and p(c) is the one-hot encoded label of the actual target.

[0107] In the above formula, λ coord , noobj , classThey are the weights used to balance different loss terms. By adjusting these weights, you can adjust the YOLO model's attention to position, size, confidence, and category.

[0108] Optionally, the wildfire situation includes wildfire category and wildfire occurrence probability; determining a target emergency response mechanism based on the wildfire situation includes:

[0109] Using the wildfire category and wildfire occurrence probability to perform a table lookup operation in a preset wildfire level table to determine the wildfire risk level corresponding to the wildfire situation;

[0110] Find the target emergency response mechanism corresponding to the wildfire risk level in the preset strategy table.

[0111] In this embodiment, the wildfire risk level corresponding to the wildfire situation is obtained by directly looking up the table, and the target emergency response mechanism corresponding to the wildfire risk level is obtained by looking up the table. Compared with the method of manually formulating emergency measures on site, the target emergency response mechanism obtained by the method of this embodiment is more targeted and more efficient.

[0112] Optionally, the method further comprises:

[0113] When the wildfire risk level is greater than the level threshold, an alarm message is sent to the user terminal, and a target image corresponding to the wildfire occurrence probability is sent.

[0114] In this embodiment, when the wildfire risk level is greater than the level threshold, it means that a wildfire is likely to occur or has occurred in the power transmission line environment to be tested. At this time, the alarm mechanism is immediately triggered, and an alarm message is sent to the user terminal, and a target image corresponding to the probability of wildfire occurrence is sent. This can promptly remind the staff of the wildfire situation in the power transmission line environment to be tested, and at the same time, the actual on-site situation of the wildfire situation can be queried in time, so as to respond to the wildfire situation in a timely manner.

[0115] Optionally, the method further comprises:

[0116] The wildfire category and wildfire occurrence probability are used to perform a table lookup operation in the preset strategy table to determine the camera adjustment strategy corresponding to the wildfire situation, so as to control the camera to shoot the transmission line environment to be tested based on the camera adjustment strategy.

[0117] In this embodiment, a corresponding camera adjustment strategy is determined based on the wildfire situation to control the camera to shoot the transmission line environment to be tested based on the camera adjustment strategy, so that the staff can accurately capture the time period when measures need to be taken in the transmission line environment to be tested, so as to improve the efficiency of tracing the cause of the wildfire in the transmission line environment to be tested.

[0118] See also Figure 2 , Figure 2In an exemplary embodiment of the present application, the process of the method for identifying and monitoring wildfires in a power transmission line environment provided is applied, such as Figure 2 As shown, a variety of sensors set in the environment of the transmission line to be tested are used to collect environmental data in real time to obtain sensor data, and then a preset vector machine is used to perform environmental analysis on the sensor data to obtain an environmental situation including environmental status and wildfire risk probability. When the environmental status is a wildfire risk state and the wildfire risk probability is greater than the probability threshold, it means that the environmental status of the environment of the transmission line to be tested is in a wildfire risk state, that is, the environmental status is abnormal, then the shooting strategy of the camera (image sensor) located in the environment of the transmission line to be tested is determined to be an instant shooting strategy, and the camera is controlled to perform image acquisition on the environment of the transmission line to be tested according to the instant shooting strategy to obtain a target image. Otherwise, environmental data is collected using a variety of sensors set in the environment of the transmission line to be tested to monitor the environment of the transmission line to be tested in real time. Then, the preset YOLO model is used to identify the target image, determine the wildfire situation of the environment of the transmission line to be tested, so as to realize the identification and monitoring of the wildfire of the environment of the transmission line to be tested. Then, a risk assessment is performed based on the wildfire situation to determine the target emergency response mechanism, so as to use the target emergency response mechanism to respond to the wildfire that occurs in the environment of the transmission line to be tested.

[0119] The method provided by the present application monitors the state of the power transmission line environment through a variety of sensors, and uses the abnormal amount of a variety of sensors to wake up the video monitoring, reducing the energy consumption of the video monitoring process throughout the day, and the sensor not only has low power consumption, but also can continuously monitor the key parameters such as temperature, humidity, and wind speed in the environment. By monitoring the changes of these data, especially under conditions such as high temperature, low humidity and strong wind, and then using support vector machine (SVM) to classify the environmental state and calculate the probability of wildfire risk through data fusion, the potential risk of wildfire can be identified in advance, and by combining image video monitoring, when the sensor data exceeds the normal range, SVM will react quickly, that is, analyze that the environmental state is a wildfire risk state, and the probability of wildfire risk is greater than the probability threshold, then trigger the camera in the power transmission line environment to start shooting immediately, so as to realize the timely capture and processing of abnormal situations with wildfire risk in the power transmission line environment, and realize a more efficient, comprehensive and stable wildfire identification and monitoring method. In this way, the camera will be started only when necessary, which not only improves the sensitivity of wildfire monitoring, but also greatly reduces the energy consumption of video monitoring. Then combined with advanced image recognition algorithm (YOLO model), fast and accurate identification of wildfires is realized. The YOLO model has excellent performance in the field of image processing. It can efficiently process a large amount of image data in a short time and accurately determine whether there is a wildfire. The application of this technology enables the present invention to achieve a high level of accuracy and efficiency in wildfire monitoring. The method of the present application can not only solve the energy consumption problem of long-term video monitoring, but also intelligently identify the situation of wildfires when wildfires occur, and can quickly initiate early warnings, further improving the reliability and accuracy of wildfire monitoring of transmission lines. The implementation of this method will provide important support for the intelligent management of transmission lines, accident prevention, and improving the overall safety, reliability and stability of the transmission system.

[0120] See also Figure 3 , Figure 3 An exemplary embodiment of the present application shows a wildfire identification and monitoring device for a power transmission line environment, such as Figure 3 As shown, the present application provides a wildfire identification and monitoring device for a transmission line environment, including a sensor, a camera and a processor ARM. The sensor is used to monitor the sensor data of the transmission line environment to be tested in real time, the camera is used to photograph the transmission line environment to be tested based on a shooting strategy, and the processor ARM is used to implement the above-mentioned wildfire identification and monitoring method for a transmission line environment.

[0121] Figure 3In the invention, the sensors include TMR sensor (Tunneling Magnetoresistance), thermal imaging sensor, smoke sensor, temperature and humidity sensor, and gas sensor. The sensor data collected by the sensor is sent to the host computer for storage through LoRa communication, and is processed by a preset vector machine and a preset YOLO model. The camera is controlled to shoot based on the environmental conditions obtained by the preset vector machine, and the camera shooting strategy is adjusted based on the mountain fire conditions obtained by identifying the target image shot by the camera based on the preset YOLO model, so as to obtain the camera adjustment strategy, so as to control the camera to shoot the environment of the power transmission line to be tested based on the camera adjustment strategy.

[0122] The application scenarios of the wildfire identification and monitoring device for the power transmission line environment of this embodiment are as follows: Figure 4 As shown, Figure 4 In the present invention, the application scenario includes a transmission line environment to be tested, and a wildfire identification and monitoring device for the transmission line environment arranged in the transmission line environment to be tested.

[0123] See also Figure 5 , Figure 5 An exemplary embodiment of the present application shows a wildfire identification and monitoring device for a power transmission line environment, such as Figure 5 As shown, the present application provides a wildfire identification and monitoring system for a power transmission line environment, comprising:

[0124] A real-time monitoring module, used to monitor sensor data of the power transmission line environment to be tested in real time;

[0125] An analysis module, used to perform environmental analysis on sensor data using a preset vector machine to obtain environmental conditions of the power transmission line environment to be tested;

[0126] A shooting module is used to determine a shooting strategy of a camera located in the environment of the power transmission line to be tested based on the environmental conditions, and control the camera to shoot the environment of the power transmission line to be tested based on the shooting strategy to obtain a target image;

[0127] The recognition module is used to perform feature recognition on the target image using a preset YOLO model to determine the wildfire situation of the power transmission line environment to be tested;

[0128] The determination module is used to determine the target emergency response mechanism based on the wildfire situation, so as to use the target emergency response mechanism to respond to the wildfire occurring in the environment of the transmission line to be tested.

[0129] In the present embodiment, a wildfire identification and monitoring system for a power transmission line environment is provided, and the sensor data is the environmental data of the power transmission line environment to be tested collected in real time by a variety of sensors. First, the sensor data of the power transmission line environment to be tested is monitored in real time by the real-time monitoring module, thereby realizing comprehensive monitoring of the power transmission line environment to be tested. Secondly, the environmental conditions are obtained by performing environmental analysis on the sensor data based on the preset vector machine by the analysis module, and the shooting strategy of the camera located in the power transmission line environment to be tested is determined in the shooting module, and the camera is controlled to shoot the power transmission line environment to be tested based on the shooting strategy to obtain the target image, so that the camera only needs to shoot the power transmission line environment to be tested intermittently based on the shooting strategy, and there is no need to shoot all day, which effectively reduces the monitoring energy consumption of the camera. Then, the preset YOLO model is used in the recognition module to perform feature recognition on the target image, determine the wildfire situation of the power transmission line environment to be tested, and use the determination module to determine the target emergency response mechanism based on the wildfire situation, so that the wildfire situation of the power transmission line environment to be tested can be monitored in time, so as to use the target emergency response mechanism to respond to the wildfire that occurs in the power transmission line environment to be tested.

[0130] Optionally, the analysis module is specifically used to:

[0131] Use a preset vector machine to extract features from sensor data to obtain a feature vector;

[0132] Analyze the environment of the transmission line environment to be tested based on the characteristic vector to obtain the environmental state of the transmission line environment to be tested;

[0133] Calculate the wildfire risk probability of the transmission line environment to be tested using the environmental status;

[0134] The environmental conditions of the transmission line environment to be tested are formed based on the environmental status and wildfire risk probability.

[0135] Optionally, the environmental conditions include environmental status and wildfire risk probability; the shooting module is specifically used for:

[0136] When the environmental state is a wildfire risk state and the wildfire risk probability is greater than a probability threshold, determining that the shooting strategy of the camera located in the power transmission line environment to be tested is an instant shooting strategy;

[0137] When the environmental state is a wildfire risk state and the wildfire risk probability is less than or equal to the probability threshold, determining that the shooting strategy of the camera is a first timed shooting strategy;

[0138] When the environmental state is a non-wildfire risk state, determining the shooting strategy of the camera to be a second timed shooting strategy;

[0139] The shooting time interval of the first timing shooting strategy is smaller than the shooting time interval of the second timing shooting strategy.

[0140] Optionally, the identification module is specifically used to:

[0141] The preset YOLO algorithm is used to identify the wildfire features in the target image, and the wildfire feature prediction box and the confidence of the wildfire feature prediction box are obtained;

[0142] Identify the wildfire feature prediction frame to obtain the wildfire category of the transmission line environment to be tested;

[0143] Based on the wildfire category and confidence level, the wildfire occurrence probability of the transmission line environment to be tested is obtained;

[0144] Based on the wildfire category and wildfire occurrence probability, the wildfire situation of the transmission line environment to be tested is formed.

[0145] Optionally, the identification module is specifically used to:

[0146] Perform category recognition on the wildfire feature prediction frame to obtain the feature category of the wildfire feature prediction frame;

[0147] The feature category is used to perform a table lookup operation in a preset category table to obtain the wildfire category of the power transmission line environment to be tested.

[0148] Optionally, the identification module is specifically used to:

[0149] Calculate predicted probabilities for wildfire categories;

[0150] The probability of wildfire occurrence in the tested transmission line environment is calculated using the predicted probability and confidence level.

[0151] It should be noted that the wildfire identification and monitoring system for a power transmission line environment provided in the above embodiment and the wildfire identification and monitoring method for a power transmission line environment provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In actual applications, the wildfire identification and monitoring system for a power transmission line environment provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0152] A computing device according to an embodiment of the present application includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, some or all steps of the above-mentioned method for identifying and monitoring wildfires in a power transmission line environment are implemented.

[0153] Among them, the computing device can be a computer, and correspondingly, its program is computer software. The above-mentioned parameters and steps in a computing device of the present application can refer to the parameters and steps in the embodiment of a method for identifying and monitoring wildfires in a power transmission line environment mentioned above, and will not be repeated here.

[0154] In an embodiment of the present application, a computer-readable storage medium is provided, in which instructions are stored. When the instructions are executed, the steps of the above-mentioned method for identifying and monitoring wildfires in a power transmission line environment are executed.

[0155] The computer-readable storage medium may be a transient computer-readable storage medium or a non-transitory computer-readable storage medium.

[0156] The technical solution of the embodiment of the present disclosure can be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method of the embodiment of the present disclosure. The aforementioned computer-readable storage medium may be a non-transitory computer-readable storage medium, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk, and other media that can store program codes, or a transient computer-readable storage medium.

[0157] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0158] Those skilled in the art know that the present application can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, generally referred to as "module" or "system" herein. In addition, in some embodiments, the present application can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code. Computer-readable storage media can be, for example, but not limited to - electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above.

[0159] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0160] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for identifying and monitoring wildfires in a power transmission line environment, characterized in that: include: Real-time monitoring of sensor data of the transmission line environment to be tested; Performing environmental analysis on the sensor data using a preset vector machine to obtain environmental conditions of the power transmission line environment to be tested; Based on the environmental conditions, determining a shooting strategy of a camera located in the power transmission line environment to be tested, and controlling the camera to shoot the power transmission line environment to be tested based on the shooting strategy to obtain a target image; Using a preset YOLO model to perform feature recognition on the target image to determine the wildfire situation of the power transmission line environment to be tested; A target emergency response mechanism is determined based on the wildfire situation, so as to utilize the target emergency response mechanism to respond to the wildfire occurring in the environment of the power transmission line to be tested.

2. The method according to claim 1, characterized in that The step of performing environmental analysis on the sensor data using a preset vector machine to obtain the environmental conditions of the power transmission line environment to be tested includes: Using a preset vector machine to extract features from the sensor data to obtain a feature vector; Analyze the environment of the power transmission line environment to be tested based on the characteristic vector to obtain the environmental state of the power transmission line environment to be tested; Calculating the wildfire risk probability of the power transmission line environment to be tested using the environmental state; The environmental condition of the power transmission line environment to be tested is formed based on the environmental state and the wildfire risk probability.

3. The method according to claim 1, characterized in that The environmental conditions include environmental status and wildfire risk probability; and determining the shooting strategy of the camera located in the power transmission line environment to be tested based on the environmental conditions includes: When the environmental state is a wildfire risk state and the wildfire risk probability is greater than a probability threshold, determining that the shooting strategy of the camera located in the power transmission line environment to be tested is an instant shooting strategy; When the environmental state is a wildfire risk state and the wildfire risk probability is less than or equal to a probability threshold, determining that the shooting strategy of the camera is a first timed shooting strategy; When the environmental state is a non-wildfire risk state, determining that the shooting strategy of the camera is a second timed shooting strategy; The shooting time interval of the first timing shooting strategy is smaller than the shooting time interval of the second timing shooting strategy.

4. The method according to any one of claims 1 to 3, characterized in that: The using a preset YOLO model to perform feature recognition on the target image to determine the wildfire situation of the power transmission line environment to be tested includes: Using a preset YOLO algorithm to identify wildfire features in the target image, and obtaining a wildfire feature prediction frame and a confidence level of the wildfire feature prediction frame; Identifying the wildfire feature prediction frame to obtain the wildfire category of the power transmission line environment to be tested; Based on the wildfire category and the confidence level, obtaining a wildfire occurrence probability of the transmission line environment to be tested; Based on the wildfire category and the wildfire occurrence probability, a wildfire situation of the power transmission line environment to be tested is formed.

5. The method according to claim 4, characterized in that The step of identifying the wildfire feature prediction frame to obtain the wildfire category of the power transmission line environment to be tested includes: Performing category recognition on the wildfire feature prediction frame to obtain a feature category of the wildfire feature prediction frame; The feature category is used to perform a table lookup operation in a preset category table to obtain the wildfire category of the power transmission line environment to be tested.

6. The method according to claim 4, characterized in that The obtaining, based on the wildfire category and the confidence level, the wildfire occurrence probability of the transmission line environment to be tested includes: Calculating predicted probabilities of the wildfire categories; The predicted probability and the confidence level are used to calculate the probability of wildfire occurrence in the power transmission line environment to be tested.

7. A wildfire identification and monitoring device for a power transmission line environment, characterized in that: It includes a sensor, a camera and a processor, the sensor is used to monitor the sensor data of the transmission line environment to be tested in real time, the camera is used to shoot the transmission line environment to be tested based on a shooting strategy, and the processor is used to implement a forest fire identification and monitoring method for a transmission line environment as described in any one of claims 1 to 6.

8. A wildfire identification and monitoring system for a power transmission line environment, characterized in that: include: A real-time monitoring module, used to monitor sensor data of the power transmission line environment to be tested in real time; An analysis module, used to perform environmental analysis on the sensor data using a preset vector machine to obtain the environmental conditions of the power transmission line environment to be tested; A shooting module, used for determining a shooting strategy of a camera located in the power transmission line environment to be tested based on the environmental conditions, and controlling the camera to shoot the power transmission line environment to be tested based on the shooting strategy to obtain a target image; An identification module, used to perform feature recognition on the target image using a preset YOLO model to determine the wildfire situation of the power transmission line environment to be tested; A determination module is used to determine a target emergency response mechanism based on the wildfire situation, so as to use the target emergency response mechanism to respond to the wildfire occurring in the power transmission line environment to be tested.

9. A computing device comprising a memory, a processor, and a program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the steps of a method for identifying and monitoring wildfires in a power transmission line environment as described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a terminal device, the terminal device executes the steps of a method for identifying and monitoring wildfires in a power transmission line environment as described in any one of claims 1 to 6.

Images