A method, device, equipment and medium for evaluating the probability of wildfires caused by lightning strike wire breakage

Abstract

The present application discloses a method, device, equipment and medium for evaluating the probability of wildfires caused by lightning strike line breaks, which relates to the technical field of wildfire prevention and control of power transmission and distribution lines. The method includes: obtaining a data set composed of data of wildfire risk assessment characteristic indicators after lightning strike line breaks; based on the data set, using the entropy weight method to calculate the first probability of each wildfire risk assessment characteristic indicator causing a wildfire; based on the data set, using the gradient boosting tree method to calculate the second probability of each wildfire risk assessment characteristic indicator causing a wildfire; fusing the first probabilities and the second probabilities of all wildfire risk assessment characteristic indicators to obtain a comprehensive probability of causing a wildfire and outputting it. The present application realizes a reliable evaluation of the risk probability of wildfires caused by lightning strike line breaks by combining the entropy weight method and the gradient boosting tree algorithm.

Description

Technical Field

[0001] The present application relates to the technical field of prevention of wildfires on transmission and distribution lines, and specifically relates to a method, device, equipment and medium for evaluating the probability of wildfires caused by lightning strike and wire breakage. Background Art

[0002] In the early days of China's power distribution network, bare conductors were used to transmit electric energy. Since there was no surface insulation on the transmission and distribution lines, short-circuit faults were likely to occur when trees or animals came into direct contact with the bare conductors, affecting the reliability of power supply. To avoid phase-to-phase short circuits, the electrical distance between two phases was relatively large, resulting in a wide line corridor and wasting land resources. To solve the above problems, overhead insulated conductors are widely used in China's 10 kV medium-voltage distribution network at present. After a lightning strike on an insulated conductor, the stable arc generated by the power frequency follow current continuously burns the lightning strike point, which may lead to wire breakage. Especially in areas such as Sichuan, there are many mountains and lush forests. During the process of the arc burning and eroding the insulated conductor, the dripping of the fire-carrying insulation skin or the high-temperature grounding arc generated after the wire breakage contacting the vegetation is extremely likely to cause wildfires, which will cause great damage to the natural environment and power equipment, seriously affect the safety and stability of the power grid and the reliability of power supply, and a large amount of manpower, material and financial resources are required to extinguish the fire, posing a great threat to the safety of personnel and property.

[0003] If the probability of wildfire risk under the combined action of many factors after lightning strike and wire breakage can be obtained, it can provide a targeted plan for wildfire prevention, which has important reference significance for the safe and reliable operation of the power distribution network.

[0004] Most of the existing research on risk assessment of wildfires caused by transmission lines adopts the method of dividing risk levels. For example, the entropy weight method is used to assign weights, and the average weighted sum of each index is obtained to obtain the wildfire occurrence risk index, and the natural break method is simply used to divide it into 4 risk levels. This method is difficult to comprehensively and objectively identify the probability of fire risk hidden dangers in the power corridor, and the entropy weight method mechanically considers the weights of each quantitative index on the output result, while ignoring the interaction relationship between the output result and each index. Summary of the Invention

[0005] In order to accurately and reliably evaluate the probability of wildfire risk caused by lightning strike and wire breakage and provide more reliable data support for wildfire prevention, the present application proposes a method, device, equipment and medium for evaluating the probability of wildfires caused by lightning strike and wire breakage.

[0006] The present application is realized through the following technical solutions:

[0007] A method for evaluating the probability of wildfires caused by lightning strike and wire breakage includes:

[0008] Obtaining a data set composed of data of characteristic indexes for wildfire risk assessment after lightning strike and wire breakage;

[0009] Based on the dataset, the first probability of a wildfire being triggered by each wildfire risk assessment characteristic index is calculated using the entropy weight method;

[0010] Based on the dataset, the second probability of a wildfire being triggered by each wildfire risk assessment characteristic index is calculated using the gradient boosting tree method;

[0011] The first probabilities and the second probabilities of all wildfire risk assessment characteristic indexes are fused to obtain the comprehensive probability of a wildfire being triggered and output.

[0012] In some embodiments, the wildfire risk assessment characteristic index data is obtained through satellite remote sensing data, weather information of meteorological stations, collection of vegetation sample information, and real-time monitoring system of the distribution network. The wildfire risk assessment characteristic indexes include vegetation moisture content, temperature, relative humidity, and power frequency follow current value. The wildfire risk assessment characteristic index data refers to the data corresponding to the wildfire risk assessment characteristic indexes.

[0013] In some embodiments, the calculation of the first probability of a wildfire being triggered by each wildfire risk assessment characteristic index using the entropy weight method includes:

[0014] Establish a sample evaluation matrix after lightning strike and wire break. The sample evaluation matrix is composed of a number of data samples, and each data sample is composed of a number of characteristic index data;

[0015] Normalize the data in the sample evaluation matrix;

[0016] Calculate the proportion of each sample value under each characteristic index;

[0017] According to the proportion of each sample value under each characteristic index, calculate the entropy value of the corresponding characteristic index;

[0018] According to the entropy value of each characteristic index, calculate the entropy weight factor of the corresponding characteristic index;

[0019] According to the entropy weight factor of each characteristic index and combined with the risk coefficient corresponding to the characteristic index, calculate the first probability of the corresponding characteristic index triggering a wildfire.

[0020] In some embodiments, the calculation of the second probability of a wildfire being triggered by each wildfire risk assessment characteristic index using the gradient boosting tree method includes:

[0021] Construct a gradient boosting tree model: An ensemble model composed of multiple regression trees is constructed through an iterative method. Each tree sequentially fits the residual between the current model prediction result and the true value, and optimizes the loss function in the gradient descent direction;

[0022] Calculate the feature importance of each tree: During the construction of each tree, record the contribution degree of each characteristic index to the loss function decrease when each node splits and quantify it;

[0023] Evaluating feature importance: Accumulate the contribution degrees of each feature index in all trees according to the feature dimension to obtain the feature importance of each feature index;

[0024] According to the feature importance of each feature index and in combination with the risk coefficient corresponding to the feature index, calculate the second probability of a wildfire caused by the corresponding feature index.

[0025] In some embodiments, the method of using the entropy weight method to calculate the first probability of a wildfire caused by each wildfire risk assessment feature index further includes:

[0026] Visualize the entropy weight factors of each feature index;

[0027] And / or, the method of using the gradient boosting tree method to calculate the second probability of a wildfire caused by each wildfire risk assessment feature index further includes:

[0028] Visualize the feature importance of each feature index.

[0029] In some embodiments, the calculation method of the comprehensive probability is:

[0030]

[0031] Wherein, represents the comprehensive probability; , both represent adjustment parameters; represents the function describing the interaction between the i th feature index and the j th feature index; , respectively represent the first probability and the second probability of a wildfire caused by the i th feature index; , respectively represent the first probability and the second probability of a wildfire caused by the j th feature index; n represents the number of feature indices.

[0032] In some embodiments, the function describing the interaction between feature indices is assigned according to the general principles of physics. The value for a positive correlation is 1, the value for a negative correlation is -1, and the value for no obvious interaction is 0.

[0033] In a second aspect, the present application proposes a device for evaluating the probability of a wildfire caused by lightning strike and wire breakage, including:

[0034] An acquisition unit, configured to acquire a data set composed of wildfire risk assessment feature index data after lightning strike and wire breakage;

[0035] The first evaluation unit calculates the first probability of a wildfire caused by each wildfire risk assessment feature index based on the dataset using the entropy weight method;

[0036] The second evaluation unit calculates the second probability of a wildfire caused by each wildfire risk assessment feature index based on the dataset using the gradient boosting tree;

[0037] And a fusion unit for fusing the first probabilities and the second probabilities of all wildfire risk assessment feature indexes to obtain a comprehensive probability of a wildfire and outputting it.

[0038] In a third aspect, the present application proposes an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of any one of the above-mentioned methods.

[0039] In a fourth aspect, the present application proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of any one of the above-mentioned methods.

[0040] A method for evaluating the probability of a wildfire caused by lightning strike and wire break not only considers the objective weight of a single index inducing a wildfire, but also considers the interaction relationship between the evaluation result and each index. Taking whether a wildfire is caused as a feature and inputting it into the gradient boosting tree algorithm, the importance of each feature index for the evaluation result is obtained. Finally, the two parts of the evaluation results are fused, and the interaction relationship between the feature indexes is also considered during the fusion, so as to obtain a more accurate and reliable comprehensive probability, providing more real and reliable data support for wildfire prevention and control;

[0041] Correspondingly, a device, equipment and medium for evaluating the probability of a wildfire caused by lightning strike and wire break proposed by the present application also have the same above-mentioned technical effects. Description of the Drawings

[0042] The drawings described herein are used to provide a further understanding of the embodiments of the present application, constitute a part of the present application, and do not limit the embodiments of the present application. In the drawings:

[0043] Figure 1 is a schematic flow chart of the evaluation method proposed by the embodiment of the present application;

[0044] Figure 2 is a schematic architecture diagram of the evaluation device proposed by the embodiment of the present application;

[0045] Figure 3 is a schematic principle diagram of the evaluation system proposed by the embodiment of the present application;

[0046] Figure 4 is a schematic principle diagram of the electronic device proposed by the embodiment of the present application;

[0047] Figure 5 Schematic diagram of a computer-readable storage medium proposed in an embodiment of the present application;

[0048] Figure 6 Example diagram of the entropy weight factors of each characteristic index calculated by the entropy weight method in an embodiment of the present application;

[0049] Figure 7 Example diagram of the feature importance of each characteristic index calculated by the gradient boosting tree method in an embodiment of the present application;

[0050] Figure 8 Example of a function assignment diagram of the interaction relationship between characteristic indexes in an embodiment of the present application;

[0051] Reference numerals and corresponding component names:

[0052] 200 - Evaluation device, 201 - Acquisition unit, 202 - First evaluation unit, 203 - Second evaluation unit, 204 - Fusion unit, 300 - Evaluation system, 301 - Input device, 302 - Output device, 303 - Processor A, 304 - Memory A, 400 - Electronic device, 410 - Memory B, 420 - Processor B, 411 - Computer program A, 500 - Computer-readable storage medium, 511 - Computer program B. Detailed implementation manners

[0053] In the following, the term "comprise" or "may comprise" that may be used in various embodiments of the present application indicates the presence of the invented functions, operations or elements, and does not limit the addition of one or more functions, operations or elements. Further, as used in various embodiments of the present application, the terms "comprise", "have" and their cognates are only intended to represent the presence of specific features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be construed as precluding the existence or addition of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items first.

[0054] In various embodiments of the present application, the expression "or" or "at least one of A or / and B" includes any combination or all combinations of the recited words. For example, the expression "A or B" or "at least one of A or / and B" may include A, may include B, or may include both A and B.

[0055] Expressions (such as "first", "second", etc.) used in various embodiments of the present application may modify various components in the various embodiments, but do not limit the corresponding components. For example, the above expressions do not limit the order and / or importance of the components. The above expressions are only for the purpose of distinguishing one component from other components. For example, the first user device and the second user device indicate different user devices, although both are user devices. For example, without departing from the scope of the various embodiments of the present application, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component.

[0056] It should be noted that: if it is described that one component is "connected" to another component, the first component may be directly connected to the second component, and a third component may be "connected" between the first component and the second component. Conversely, when one component is "directly connected" to another component, it can be understood that there is no third component between the first component and the second component.

[0057] The terms used in the various embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit the various embodiments of the present application. As used herein, the singular form is intended to also include the plural form, unless the context clearly indicates otherwise. Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the various embodiments of the present application belong. The terms (such as those defined in a commonly used dictionary) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized meaning or an overly formal meaning, unless clearly defined in the various embodiments of the present application.

[0058] To make the purpose, technical solution and advantages of the present application more clear and understandable, the present application will be further described in detail below in conjunction with embodiments and drawings. The illustrative embodiments of the present application and their descriptions are only used to explain the present application and do not limit the present application.

[0059] Embodiment 1: This embodiment proposes a method for evaluating the probability of wildfires caused by lightning strikes and wire breaks, and realizes a reliable evaluation of the risk probability of wildfires caused by lightning strikes and wire breaks by combining the entropy weight method and the gradient boosting tree algorithm.

[0060] As Figure 1 shown, the method proposed in this embodiment includes the following steps:

[0061] Step 110, obtain a data set composed of wildfire risk assessment characteristic index data after lightning strikes and wire breaks.

[0062] Step 120: Based on this data set, use the entropy weight method to calculate the first probability of each wildfire risk assessment characteristic index causing a wildfire.

[0063] Step 130: Based on this data set, use the gradient boosting tree method to calculate the second probability of each wildfire risk assessment characteristic index causing a wildfire.

[0064] Step 140: Fuse the first probability and the second probability of all wildfire risk assessment characteristic indexes to obtain the comprehensive probability of causing a wildfire and output it.

[0065] In an alternative implementation, step 110 can obtain wildfire risk assessment characteristic index data through satellite remote sensing data, weather information of weather stations, collection of vegetation sample information, and real-time monitoring system of the distribution network, and form a data set accordingly. Among them, the wildfire risk assessment characteristic indexes mainly include characteristic indexes such as vegetation moisture content, temperature, relative humidity, and the magnitude of power frequency follow current value. These characteristic indexes can be determined by analyzing historical data, which belongs to the prior art in this field and will not be elaborated here.

[0066] In an alternative implementation, in step 120, the entropy weight method is used to calculate the first probability of each wildfire risk assessment characteristic index causing a wildfire. The specific process is as follows:

[0067] Step 121: Establish a sample evaluation matrix after lightning strike and wire break. This sample evaluation matrix can be composed of m samples, and each sample includes n characteristic index data. This sample evaluation matrix can be specifically expressed as:

[0068]

[0069] Among them, represents the i th characteristic index data of the j th sample ( i = 1, 2,..., m ; j = 1, 2,..., n ), m represents the number of samples; n represents the number of characteristic indexes.

[0070] Step 122: Normalize the data in the sample evaluation matrix.

[0071] Since the measurement units of various characteristic indicators are not unified, before calculating the indicators, standardization processing (normalization processing) needs to be carried out first, that is, converting the absolute values of the characteristic indicators into relative values to solve the homogenization problem of various heterogeneous characteristic indicators. In addition, the meanings represented by the values of positive indicators and negative indicators are different (the higher the value of the positive indicator, the better; the lower the value of the negative indicator, the better). Therefore, different algorithms need to be used for the normalization processing of positive indicators and negative indicators, and the processing methods are as follows:

[0072] Positive indicator:

[0073]

[0074] Negative indicator:

[0075]

[0076] For the convenience of description, the data after normalization is still denoted as , that is, the data used in the subsequent calculation process are all the data after normalization.

[0077] Step 123, calculate the proportion of each sample value under each characteristic indicator, and the calculation formula is as follows:

[0078]

[0079] Among them, represents the proportion of the j th sample value under the i th indicator in this indicator.

[0080] Step 124, calculate the entropy value of each characteristic indicator, and the calculation formula is as follows:

[0081]

[0082] Among them, represents the entropy value of the j th indicator; k is a constant used to standardize the entropy value, .

[0083] Step 125, calculate the information entropy redundancy (difference) of each characteristic indicator, and the calculation formula is as follows:

[0084]

[0085] Among them, represents the information entropy redundancy of the j th indicator.

[0086] Step 126, calculate the entropy weight factor of each characteristic indicator, and the calculation formula is as follows:

[0087]

[0088] Among them, represents the entropy weight factor of the j th index.

[0089] Step 127, according to the entropy weight factors of each characteristic index, calculate the first probability of wildfire caused by this characteristic index. The calculation formula is as follows:

[0090]

[0091] Among them, represents the first probability of wildfire caused by the i th index; represents the entropy weight factor of the i th index; represents the risk coefficient corresponding to the i th index.

[0092] The risk coefficient R is an adjustment coefficient for the weight (entropy weight factor K). According to expert experience, assign scores to various situations of different indexes causing wildfires. Combine with the wildfire database, count the number of wildfires occurring in various specific situations under a certain index, and multiply the frequency of occurrence in the same type of situations in a specific index by the corresponding assigned score to obtain the risk coefficient.

[0093] Optionally, after calculating the entropy weight factors of each characteristic index, it further includes visualizing the entropy weight factors of each characteristic index. Specifically, a horizontal bar chart or a radar chart can be used to display the entropy weight factors of each characteristic index.

[0094] In an alternative implementation manner, in step 130, the gradient boosting tree method is used to calculate the second probability of wildfire caused by each wildfire risk assessment characteristic index. The specific process is as follows:

[0095] Step 131, construct a gradient boosting tree model, and construct an ensemble model composed of multiple regression trees through an iterative method. Each tree sequentially fits the residual between the current model prediction result and the true value, and optimizes the loss function in the gradient descent direction. Specifically, by sequentially generating multiple weak decision trees, in each round of iteration, the newly constructed decision tree uses the prediction residual of the previous combined model as the learning target, and updates the parameters by moving along the negative gradient direction of the loss function, gradually reducing the prediction error of the overall model.

[0096] Step 132: Calculate the feature importance of each tree. During the construction of each tree, record the contribution degree of each feature index to the decrease of the loss function when splitting at each node and quantify it. Specifically, it is reflected in the loss reduction amount brought by each feature index when it is selected as the splitting point in all trees. Optionally, the feature importance can also be evaluated through a permutation test, that is, randomly shuffle the values of a certain feature index and then observe the attenuation degree of the model prediction performance. The greater the performance decline, the more significant the impact of the feature index on the model.

[0097] Step 133: Evaluate the feature importance. Accumulate the contribution degrees of each feature index in all trees according to the feature dimension. Since the gradient boosting tree optimizes the model through multiple rounds of iteration, the importance of the feature index needs to comprehensively consider its cumulative contribution in all trees and finally be transformed into a relative importance index through normalization processing. According to the summarized feature importance, evaluate the contribution of each feature to the model prediction ability. The higher the feature importance of a feature index, the greater the impact on the prediction result of the model.

[0098] Step 134: Calculate the second probability of a wildfire caused by each feature according to the importance of each feature. The calculation formula is as follows:

[0099]

[0100] where, represents the second probability of a wildfire caused by the i th index; represents the importance of the i th index; represents the risk coefficient corresponding to the i th index.

[0101] In this embodiment, the gradient boosting tree algorithm is adopted, which not only captures the direct role of features in gradient direction optimization but also reflects the potential correlation between its features and the prediction target through global perturbation analysis, thus realizing the evaluation of feature importance from multiple perspectives.

[0102] Optionally, after evaluating the importance of each feature, it further includes visualizing the feature importance. Specifically, a horizontal bar chart or a radar chart can be used to display the feature importance ranking, highlighting the contrast relationship between key features and secondary features.

[0103] In an alternative embodiment, the comprehensive probability of a wildfire is calculated by the following formula:

[0104]

[0105] where, represents the comprehensive probability; , both represent adjustment parameters. Preferably, a takes 0.5 and b takes 0.1. Indicates a function that describes the interaction between the i th characteristic index and the j th characteristic index. It is assigned according to the general principles of physics. The value for positive correlation is 1, the value for negative correlation is -1, and the value for no obvious interaction is 0.

[0106] In this embodiment, the entropy weight method is first used to objectively determine the weights according to the dispersion degree of the data, reflecting the variation information of the indicators. At the same time, the gradient boosting tree is used to evaluate the importance of the indicators, so as to realize the comprehensive evaluation of the indicators. It not only considers the objective weights of individual indicators inducing wildfires, but also considers the interaction relationship between the output results and each indicator. Whether a wildfire is triggered is used as a feature to input into the gradient boosting tree algorithm to obtain the importance of each characteristic indicator for the evaluation result. Finally, the two parts of the evaluation results are fused, and the interaction relationship between the characteristic indicators is also considered during the fusion to obtain a more accurate and reliable comprehensive probability, providing more real and reliable data support for wildfire prevention and control.

[0107] This embodiment also proposes an embodiment of a device for evaluating the probability of a wildfire caused by lightning strike and wire breakage. As Figure 2 shown, the evaluation device 200 includes:

[0108] An acquisition unit 201, configured to acquire a data set composed of wildfire risk assessment characteristic index data after lightning strike and wire breakage. The specific data set acquisition method is as described in the above method, and will not be elaborated here.

[0109] A first evaluation unit 202, based on this data set, calculates the first probability of each wildfire risk assessment characteristic index causing a wildfire by using the entropy weight method. The specific process is as described in the above method, and will not be elaborated here.

[0110] A second evaluation unit 203, based on this data set, calculates the second probability of each wildfire risk assessment characteristic index causing a wildfire by using the gradient boosting tree method. The specific process is as described in the above method, and will not be elaborated here.

[0111] And a fusion unit 204, configured to fuse the first probabilities and the second probabilities of all wildfire risk assessment characteristic indicators to obtain a comprehensive probability of causing a wildfire and output it. The specific process is as described in the above method, and will not be elaborated here.

[0112] This embodiment also proposes an embodiment of a system for evaluating the probability of a wildfire caused by lightning strike and wire breakage. As Figure 3 shown, the evaluation system 300 proposed in this embodiment includes:

[0113] An input device 301, an output device 302, a processor A 303, and a memory A 304; wherein, the number of the processor A 303 and the memory A 304 can be one or more. Figure 3 For illustration purposes, one processor A 303 and one memory A 304 are taken as an example. The input device 301, the output device 302, the processor A 303, and the memory A 304 can be connected via a bus or other means. Figure 3 For illustration purposes, connection via a bus is taken as an example.

[0114] Wherein, by invoking the operation instructions stored in the memory A 304, the processor A 303 is configured to perform the following steps:

[0115] Obtain a data set composed of wildfire risk assessment characteristic index data after lightning strike and wire breakage;

[0116] Based on this data set, use the entropy weight method to calculate the first probability of each wildfire risk assessment characteristic index causing a wildfire;

[0117] Based on this data set, use the gradient boosting tree method to calculate the second probability of each wildfire risk assessment characteristic index causing a wildfire;

[0118] Fuse the first probabilities and the second probabilities of all wildfire risk assessment characteristic indexes to obtain a comprehensive probability of causing a wildfire and output it.

[0119] Optionally, by invoking the operation instructions stored in the memory A 304, the processor A 303 is further configured to execute any one of the implementation manners in the corresponding embodiments of the above evaluation method.

[0120] This embodiment also proposes an electronic device 400, as Figure 4 shown. The electronic device 400 includes: a memory B 410, a processor B 420, and a computer program A 411 stored on the memory B 410 and executable on the processor B 420. When the processor B 420 executes the computer program A 411, the following steps are implemented:

[0121] Obtain a data set composed of wildfire risk assessment characteristic index data after lightning strike and wire breakage;

[0122] Based on this data set, use the entropy weight method to calculate the first probability of each wildfire risk assessment characteristic index causing a wildfire;

[0123] Based on this data set, use the gradient boosting tree method to calculate the second probability of each wildfire risk assessment characteristic index causing a wildfire;

[0124] Fuse the first probabilities and the second probabilities of all wildfire risk assessment characteristic indexes to obtain a comprehensive probability of causing a wildfire and output it.

[0125] Optionally, when the processor B420 executes the computer program A411, any implementation manner in the corresponding embodiment of the above evaluation method can be implemented.

[0126] It should be noted that the electronic device proposed in this embodiment is a device adopted to implement the above evaluation method. Therefore, based on the above evaluation method proposed in this embodiment, those skilled in the art can understand the specific implementation manner and various variation forms of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the above evaluation method will not be introduced in detail here. As long as the electronic device adopted by those skilled in the art to implement the above evaluation method belongs to the scope to be protected by this application.

[0127] In another embodiment, this embodiment also proposes a computer-readable storage medium 500, as Figure 5 shown. A computer program B511 is stored on the computer-readable storage medium 500. When the computer program B511 is executed by a processor, the following steps are implemented:

[0128] Obtain a data set composed of lightning strike disconnection-induced wildfire risk assessment characteristic index data;

[0129] Based on this data set, use the entropy weight method to calculate the first probability of each wildfire risk assessment characteristic index causing a wildfire;

[0130] Based on this data set, use the gradient boosting tree method to calculate the second probability of each wildfire risk assessment characteristic index causing a wildfire;

[0131] Fuse the first probability and the second probability of all wildfire risk assessment characteristic indexes to obtain the comprehensive probability of causing a wildfire and output it.

[0132] Optionally, when the computer program B511 is executed by a processor, any implementation manner in the corresponding embodiment of the above warning method can be implemented.

[0133] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0134] Embodiment 2: The following is the specific process of using the evaluation method proposed in Embodiment 1 above for example evaluation:

[0135] S1, organize the historical data of each characteristic index of lightning strike disconnection-induced wildfire, and obtain the entropy weight factor by using the entropy weight method; as Figure 6 shown. It can be seen from the figure that the entropy weight factor of the power frequency follow current value is the largest, and the entropy weight factor of the temperature is the smallest.

[0136] S2. Then, import the sorted data into the gradient boosting tree model to obtain the feature importance. As Figure 7 shown, it can be seen from the figure that the power frequency follow - current value has the highest importance, and the temperature has the lowest importance.

[0137] S3. Substitute the entropy weight factors and importance of each feature into the corresponding probability calculation formula to obtain the first probability and the second probability of each feature index causing wildfires.

[0138] S4. a Take the value of 0.5, b Take the value of 0.1, The interaction factor is assigned values according to the general principles of physics. The value for positive - correlation interaction is 1, the value for negative - correlation interaction is - 1, and the value for no obvious interaction is 0. The assignment diagram is as Figure 8 shown.

[0139] S5. Collect each feature index. Taking the vegetation with a moisture content of 31%, a temperature of 19.5°C, a humidity of 47.0%, and a power - frequency follow - current value of 9A in a certain south - west region as an example, the wildfire probability value under the comprehensive action of this feature index after lightning - induced line break is obtained as 65.77% through the comprehensive probability formula. The result differs from the historical data statistical result of 62.28% in this region by 5.30%, and it is considered to have sufficient reliability.

[0140] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer - usable storage media (including but not limited to disk memory, CD - ROM, optical memory, etc.) containing computer - usable program code.

[0141] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general - purpose computer, a special - purpose computer, an embedded processor, or other programmable data - processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data - processing devices generate means for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0142] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the flows Figure 1 one or more of the flows and / or boxes Figure 1 specified in one or more of the boxes.

[0143] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or boxes Figure 1 specified in one or more of the boxes.

[0144] The specific embodiments described above further elaborate on the objectives, technical solutions, and beneficial effects of the present application. It should be understood that the above description is only for the specific embodiments of the present application and is not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

Claims

1. A method for evaluating the probability of wildfires caused by lightning strikes and wire breaks, characterized in that, Including: Obtain a data set composed of characteristic index data for post-lightning strike wire break wildfire risk assessment; the characteristic indexes for wildfire risk assessment include vegetation moisture content, temperature, relative humidity, and power frequency follow current value, and the characteristic index data for wildfire risk assessment refers to the data corresponding to the characteristic indexes for wildfire risk assessment; Based on the data set, use the entropy weight method to calculate the first probability of each characteristic index for wildfire risk assessment to cause a wildfire; Based on the data set, use the gradient boosting tree method to calculate the second probability of each characteristic index for wildfire risk assessment to cause a wildfire; Fuse the first probabilities and the second probabilities of all characteristic indexes for wildfire risk assessment to obtain the comprehensive probability of causing a wildfire and output it; The calculation method of the comprehensive probability is: Among them, P represents the comprehensive probability; both a and b represent adjustment parameters; f(i, j) represents a function describing the interaction between the i-th feature index and the j-th feature index; P 1i , P 2i respectively represent the first probability and the second probability of the i-th feature index triggering a wildfire; P 1j , P 2j respectively represent the first probability and the second probability of the j-th feature index triggering a wildfire; n represents the number of feature indices.

2. The method for evaluating the probability of wildfires caused by lightning strikes and wire breaks according to claim 1, wherein The characteristic index data for wildfire risk assessment is obtained through satellite remote sensing data, weather information of weather stations, collection of vegetation sample information, and real-time monitoring system of the distribution network.

3. The method for evaluating the probability of wildfires caused by lightning strike and wire breakage according to claim 1, wherein The step of using the entropy weight method to calculate the first probability of each characteristic index for wildfire risk assessment to cause a wildfire includes: Establish a sample evaluation matrix after lightning strike wire break, the sample evaluation matrix is composed of several data samples, and each data sample is composed of several characteristic index data; Normalize the data in the sample evaluation matrix; Calculate the proportion of each sample value under each characteristic index; According to the proportion of each sample value under each characteristic index, calculate the entropy value of the corresponding characteristic index; According to the entropy value of each characteristic index, calculate the entropy weight factor of the corresponding characteristic index; According to the entropy weight factor of each characteristic index and in combination with the risk coefficient corresponding to the characteristic index, calculate the first probability of the corresponding characteristic index to cause a wildfire.

4. The method for evaluating the probability of wildfires caused by lightning strike wire breaks according to claim 1, wherein The step of using the gradient boosting tree method to calculate the second probability of each characteristic index for wildfire risk assessment to cause a wildfire includes: Construct a gradient boosting tree model: construct an ensemble model composed of multiple regression trees through an iterative method, and each tree sequentially fits the residual between the current model prediction result and the true value, and optimizes the loss function in the gradient descent direction; Calculate the feature importance of each tree: during the construction of each tree, record the contribution degree of each characteristic index to the decrease of the loss function when splitting each node and quantify it; Evaluate the feature importance: accumulate the contribution degrees of each characteristic index in all trees according to the feature dimension to obtain the feature importance of each characteristic index; According to the feature importance of each characteristic index and in combination with the risk coefficient corresponding to the characteristic index, calculate the second probability of the corresponding characteristic index to cause a wildfire.

5. A method for evaluating the probability of wildfires caused by lightning strike wire breaks according to claim 3 or 4, characterized in that, The step of using the entropy weight method to calculate the first probability of each characteristic index for wildfire risk assessment to cause a wildfire further includes: Visualize the entropy weight factors of each characteristic index; And / or, the step of using the gradient boosting tree method to calculate the second probability of each characteristic index for wildfire risk assessment to cause a wildfire further includes: Visualize the feature importance of each characteristic index.

6. The method for evaluating the probability of wildfires caused by lightning strike wire breaks according to claim 1, characterized in that, A function describing the interaction between characteristic indexes, which is assigned values according to the general principles of physics, with a positive correlation effect taking a value of 1, a negative correlation effect taking a value of -1, and no obvious interaction taking a value of 0.

7. A device for evaluating the probability of wildfires caused by lightning strike and wire breakage, characterized in that, Including: An acquisition unit, configured to acquire a data set composed of post-lightning strike and wire-break wildfire risk assessment characteristic index data; the wildfire risk assessment characteristic indexes include vegetation moisture content, temperature, relative humidity, and power frequency follow current value, and the wildfire risk assessment characteristic index data refers to the data corresponding to the wildfire risk assessment characteristic indexes; A first evaluation unit, based on the data set, calculates the first probability of each wildfire risk assessment characteristic index causing a wildfire by using the entropy weight method; A second evaluation unit, based on the data set, calculates the second probability of each wildfire risk assessment characteristic index causing a wildfire by using the gradient boosting tree; And a fusion unit, configured to fuse the first probabilities and the second probabilities of all wildfire risk assessment characteristic indexes to obtain a comprehensive probability of causing a wildfire and output the comprehensive probability; The calculation method of the comprehensive probability is: Among them, P represents the comprehensive probability; both a and b represent adjustment parameters; f(i, j) represents a function describing the interaction between the i-th characteristic index and the j-th characteristic index; P 1i and P 2i respectively represent the first probability and the second probability of wildfires caused by the i-th characteristic index; P 1j and P 2j respectively represent the first probability and the second probability of wildfires caused by the j-th characteristic index; n represents the number of characteristic indexes.

8. An electronic device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1-6.

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

Images