Base station thunder and lightning risk assessment method and device and electronic equipment

Abstract

The invention discloses a base station lightning risk assessment method and device and electronic equipment, and the method is applied to the technical field of communication, and the method comprises the steps: collecting a plurality of lightning stroke parameters on a ground wire of base station equipment through a target sensor, and carrying out the feature extraction of the plurality of lightning stroke parameters, and obtaining a plurality of key features; constructing an association relationship among the plurality of key features, and fusing the association relationship and the plurality of lightning stroke parameters by adopting a fusion algorithm to obtain fused data; calculating and deleting abnormal data points in the fused data based on a local abnormal factor algorithm to obtain target parameters; and calculating the target parameter based on a risk evaluation algorithm, evaluating the lightning risk at the current moment, and obtaining a lightning risk index. According to the method and the device, the problem that the assessment result is inaccurate due to the fact that the considered factor is relatively single and the lightning stroke energy and equipment damage actually invading the interior of the base station are not considered when the base station is assessed to be subjected to the lightning stroke risk in the prior art is solved.

Description

Technical Field

[0001] This application relates to the field of communication technologies, and in particular, to a method, apparatus, and electronic device for lightning risk assessment of a base station. Background Technique

[0002] Lightning strikes are one of the main causes of communication base station failures. Currently, historical data of lightning strike failures of communication base stations can be used, and some data mining algorithms can be adopted for data analysis to study the correlation relationships between data, so as to evaluate the lightning strike failure risk of communication base stations. For example, the statistical method. However, compared with the method of analyzing the historical data of lightning strike failures of overhead lines in a certain area by using data mining algorithms, the method of lightning strike mechanism analysis lacks pertinence; the statistical method only analyzes a single influencing factor and lightning strike failures, and the considered factors are relatively single, resulting in incomplete analysis results. The traditional monitoring of the atmosphere and environmental dimensions, and the geographical environment are based on the data analysis outside the building, which is a relative evaluation method. There is no data support for the actual lightning strike energy invading the base station interior and the damage to equipment. Therefore, the actual risk of the base station interior equipment cannot be quantitatively analyzed by data, and lightning strike failure analysis and precise lightning protection cannot be realized.

[0003] Aiming at the problem that when evaluating the lightning strike risk of a base station in the related art, the considered factors are relatively single, and the actual lightning strike energy invading the base station interior and equipment damage are not considered, resulting in inaccurate evaluation results, no effective solution has been proposed yet. Summary of the Invention

[0004] The main purpose of this application is to provide a method, apparatus, and electronic device for lightning risk assessment of a base station, so as to solve the problem that when evaluating the lightning strike risk of a base station in the related art, the considered factors are relatively single, and the actual lightning strike energy invading the base station interior and equipment damage are not considered, resulting in inaccurate evaluation results.

[0005] To achieve the above object, according to one aspect of this application, a method for lightning risk assessment of a base station is provided. The method includes: collecting a variety of lightning strike parameters on the grounding wire of the base station equipment through a target sensor, and performing feature extraction processing on the variety of lightning strike parameters to obtain a variety of key features, where the target sensor is a sensor including a variety of sensing elements; constructing an association relationship between the variety of key features, and using a fusion algorithm to fuse the association relationship and the variety of lightning strike parameters to obtain fused data; calculating and deleting abnormal data points in the fused data based on the local outlier factor algorithm to obtain target parameters; calculating the target parameters based on a risk assessment algorithm to evaluate the lightning risk at the current moment, and obtaining a lightning risk index, where the risk assessment algorithm includes at least one of the following: fuzzy logic algorithm, Bayesian network algorithm, absolute method risk assessment algorithm.

[0006] Further, a fusion algorithm is used to fuse the association relationship and the multiple lightning strike parameters to obtain fused data, including: determining the prior probability distribution of each sensor in the target sensor based on the historical measurement information of the target sensor, adjusting the prior probability distribution of each sensor using the multiple lightning strike parameters to obtain the posterior probability model of each sensor; determining the weight of each sensor through a neural network model based on the association relationship, the historical accuracy of each sensor, and / or the correlation degree between the current lightning risk assessment task and the parameters collected by each sensor; fusing the posterior probability models of each sensor according to the weight of each sensor to obtain a target probability model; and fusing the lightning strike parameters corresponding to each sensor in the multiple lightning strike parameters based on the target probability model to obtain the fused data.

[0007] Further, based on the local outlier factor algorithm, abnormal data points in the fused data are calculated and deleted to obtain target parameters, including: for each first data point in the fused data, determining second data points whose distances from the first data point are less than a preset distance threshold, and determining a set of neighborhood data points of the first data point according to the second data points; calculating the reciprocal of the average reachable distance between each second data point in the set of neighborhood data points and the first data point to obtain the local reachability density of the first data point; calculating the ratio of the local reachability density of each second data point in the set of neighborhood data points to the local reachability density of the first data point respectively, and calculating the average value of the ratios to obtain the outlier value corresponding to each first data point; in the case where the outlier value of the first data point is greater than a preset outlier value threshold, determining the first data point as an abnormal data point, and deleting the abnormal data points in the fused data to obtain the target parameters.

[0008] Further, feature extraction processing is performed on the multiple lightning strike parameters to obtain multiple key features, including: using a digital filter to remove noise and interference signals in the multiple lightning strike parameters to obtain the filtered multiple lightning strike parameters; checking the integrity and correctness of the filtered multiple lightning strike parameters through cyclic redundancy check to obtain the verified multiple lightning strike parameters; and extracting the key features from the verified multiple lightning strike parameters, where the key features at least include: peak voltage, current waveform, resistance value change trend, and temperature and humidity change.

[0009] Further, the multiple lightning strike parameters at least include: ground potential counterattack overvoltage, lightning strike frequency, lightning energy magnitude, and grounding status of base station equipment; the risk assessment algorithm is the absolute method risk assessment algorithm. Based on the risk assessment algorithm, the target parameters are calculated to evaluate the lightning risk at the current moment, and a lightning risk index is obtained, including: standardizing the index value of each parameter in the target parameters to obtain the standardized target parameters, where the lightning energy magnitude is characterized by a preset unit, and the grounding status of the base station equipment is characterized by the integrity of the grounding system of the base station equipment; determining the influence degree of each parameter in the target parameters on the current lightning risk assessment task, and determining the weight of each parameter according to the influence degree of each parameter; using the method of weighted summation to calculate the standardized target parameters and the weight of each parameter in the target parameters to obtain the lightning risk index.

[0010] Further, after collecting multiple lightning strike parameters on the grounding wire of the base station equipment through the target sensor, the method further includes: constructing a decision tree model based on the parameters with a relatively high degree of correlation with the lightning risk in the multiple lightning strike parameters; training the decision tree model with the multiple lightning strike parameters to obtain a trained decision tree model; evaluating the trained decision tree model based on the cross-validation algorithm, and optimizing the trained decision tree model based on the reinforcement learning strategy according to the evaluation result to obtain an optimized decision tree model; calculating the multiple lightning strike parameters collected within a preset time period based on the optimized decision tree model to generate and execute a warning strategy.

[0011] Further, after collecting multiple lightning strike parameters on the grounding wire of the base station equipment through the target sensor, the method further includes: transmitting the multiple lightning strike parameters to multiple headers of a message queue; transmitting the multiple lightning strike parameters stored in the multiple headers to multiple data processing modules according to the target mapping relationship, where each data processing module in the multiple data processing modules is responsible for performing different data processing operations on the multiple lightning strike parameters; the target mapping relationship refers to the mapping relationship between the multiple headers and the multiple data processing modules; the transmission process of the multiple lightning strike parameters is transmitted using parallel compressive sensing technology.

[0012] Further, after calculating the target parameter based on the risk assessment algorithm, evaluating the lightning risk at the current moment, and obtaining the lightning risk index, the method further includes: determining the lightning risk level at the current moment according to the lightning risk index, where the lightning risk level includes at least one of the following: the first level and the second level, and the execution strategies corresponding to the first level and the second level are different; sending the lightning risk level and the execution strategy corresponding to the lightning risk level to the target object, and receiving the response information of the target object in response to the lightning risk level; generating a lightning risk processing report based on the lightning risk index, the lightning risk level, the execution strategy corresponding to the lightning risk level, and the response information.

[0013] To achieve the above object, according to another aspect of the present application, there is provided a base station lightning risk assessment device, including: a collection unit, configured to collect a variety of lightning strike parameters on the grounding wire of the base station equipment through a target sensor, and perform feature extraction processing on the variety of lightning strike parameters to obtain a variety of key features, where the target sensor is a sensor including a variety of types of sensing elements; a fusion unit, configured to construct an association relationship between the variety of key features, and use a fusion algorithm to fuse the association relationship and the variety of lightning strike parameters to obtain fused data; a deletion unit, configured to calculate and delete abnormal data points in the fused data based on the local outlier factor algorithm to obtain a target parameter; an evaluation unit, configured to calculate the target parameter based on a risk assessment algorithm, evaluate the lightning risk at the current moment, and obtain a lightning risk index, where the risk assessment algorithm includes at least one of the following: a fuzzy logic algorithm, a Bayesian network algorithm, and an absolute method risk assessment algorithm.

[0014] Further, the fusion unit includes: a first determination subunit, configured to determine the prior probability distribution of each sensor among the variety of sensors based on the historical measurement information of the variety of sensors, and adjust the prior probability distribution of each sensor using the variety of lightning strike parameters to obtain the posterior probability model of each sensor; a second determination subunit, configured to determine the weight of each sensor through a neural network model based on the association relationship, the historical accuracy rate of each sensor, and / or the correlation degree between the current lightning risk assessment task and the parameters collected by each sensor; a first fusion subunit, configured to fuse the posterior probability models of each sensor according to the weight of each sensor to obtain a target probability model; a second fusion subunit, configured to fuse the lightning strike parameters corresponding to each sensor among the variety of lightning strike parameters based on the target probability model to obtain the fused data.

[0015] Further, the deletion unit includes: a third determination subunit, configured to, for each first data point in the fused data, determine second data points whose distances from the first data point are less than a preset distance threshold, and determine a set of neighborhood data points of the first data point according to the second data points; a first calculation subunit, configured to calculate the reciprocal of the average reachable distance between each second data point in the set of neighborhood data points and the first data point, to obtain the local reachable density of the first data point; a second calculation subunit, configured to calculate the ratio between the local reachable density of each second data point in the set of neighborhood data points and the local reachable density of the first data point respectively, and calculate the average value of the ratios, to obtain an outlier corresponding to each first data point; a deletion subunit, configured to, when the outlier of the first data point is greater than a preset outlier threshold, determine the first data point as an abnormal data point, and delete the abnormal data points in the fused data, to obtain the target parameter.

[0016] Further, the acquisition unit includes: a filtering subunit, configured to use a digital filter to remove noise and interference signals in the multiple lightning strike parameters, to obtain the filtered multiple lightning strike parameters; an inspection subunit, configured to check the integrity and correctness of the filtered multiple lightning strike parameters through cyclic redundancy check, to obtain the verified multiple lightning strike parameters; an extraction subunit, configured to extract the key features from the verified multiple lightning strike parameters, where the key features at least include: peak voltage, current waveform, resistance value change trend, temperature and humidity change.

[0017] Further, the multiple lightning strike parameters at least include: ground potential counterattack overvoltage, lightning strike frequency, lightning energy magnitude, grounding status of base station equipment; the risk assessment algorithm is the absolute method risk assessment algorithm, and the assessment unit includes: a processing subunit, configured to perform normalization processing on the index value of each parameter in the target parameter, to obtain the normalized target parameter, where the lightning energy magnitude is characterized by a preset unit, and the grounding status of the base station equipment is characterized by the integrity of the grounding system of the base station equipment; a fourth determination subunit, configured to determine the influence degree of each parameter in the target parameter on the current lightning strike risk assessment task, and determine the weight of each parameter according to the influence degree of each parameter; a third calculation subunit, configured to calculate the lightning risk index by using the method of weighted summation for the normalized target parameter and the weight of each parameter in the target parameter.

[0018] Further, the device further includes: a construction unit, configured to construct a decision tree model according to parameters with a relatively high degree of relevance to lightning risks among the multiple lightning strike parameters collected by a target sensor on the grounding wire of a base station device; a training unit, configured to train the decision tree model by using the multiple lightning strike parameters to obtain a trained decision tree model; a second evaluation unit, configured to evaluate the trained decision tree model based on a cross-validation algorithm, and optimize the trained decision tree model according to a reinforcement learning strategy based on the evaluation result to obtain an optimized decision tree model; and a calculation unit, configured to calculate multiple lightning strike parameters collected within a preset time period based on the optimized decision tree model, so as to generate and execute a warning strategy.

[0019] Further, the device further includes: a first transmission unit, configured to transmit the multiple lightning strike parameters to multiple headers of a message queue after collecting the multiple lightning strike parameters by a target sensor; a second transmission unit, configured to transmit the multiple lightning strike parameters stored in the multiple headers to multiple data processing modules according to a target mapping relationship, where each data processing module among the multiple data processing modules is responsible for performing different data processing operations on the multiple lightning strike parameters; the target mapping relationship refers to the mapping relationship between the multiple headers and the multiple data processing modules; and the transmission process of the multiple lightning strike parameters is performed by using a parallel compressive sensing technology.

[0020] Further, the device further includes: a determination unit, configured to determine a lightning risk level at the current moment according to the lightning risk index after calculating the target parameter based on a risk assessment algorithm to evaluate the lightning risk at the current moment, where the lightning risk level includes at least one of the following: a first level and a second level, and the execution strategies corresponding to the first level and the second level are different; a receiving unit, configured to send the lightning risk level and the execution strategy corresponding to the lightning risk level to a target object, and receive response information of the target object to the lightning risk level; and a generating unit, configured to generate a lightning risk processing report according to the lightning risk index, the lightning risk level, the execution strategy corresponding to the lightning risk level, and the response information.

[0021] To achieve the above object, according to one aspect of the present application, there is provided a computer program product, including a computer program, where when the computer program is executed by a processor, it implements the base station lightning risk assessment method described in any one of the above, and when the computer program is executed by a processor, it implements the steps of the base station lightning risk assessment method in various embodiments of the present application.

[0022] To achieve the above object, according to one aspect of the present application, there is provided a computer-readable storage medium, the computer-readable storage medium including stored computer instructions, wherein when the computer instructions are executed by a processor, the base station lightning risk assessment method described in any one of the above is implemented.

[0023] To achieve the above object, according to one aspect of the present application, there is provided an electronic device, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the base station lightning risk assessment method described in any one of the above.

[0024] Through the present application, the following steps are adopted: collecting a variety of lightning strike parameters on the grounding wire of the base station equipment through a target sensor, and performing feature extraction processing on the variety of lightning strike parameters to obtain a variety of key features, wherein the target sensor is a sensor including a variety of sensing elements; constructing an association relationship between the variety of key features, and using a fusion algorithm to fuse the association relationship and the variety of lightning strike parameters to obtain fused data; calculating and deleting abnormal data points in the fused data based on the local outlier factor algorithm to obtain target parameters; calculating the target parameters based on a risk assessment algorithm to evaluate the lightning risk at the current moment to obtain a lightning risk index, wherein the risk assessment algorithm includes at least one of the following: fuzzy logic algorithm, Bayesian network algorithm, absolute method risk assessment algorithm, which solves the problem in the related technology that when evaluating the lightning strike risk of a base station, the considered factors are relatively single, and the actual lightning energy invading the base station interior and equipment damage are not considered, resulting in inaccurate evaluation results.

[0025] Through the above steps, a comprehensive, accurate, and real-time assessment of the lightning risk of the base station is achieved, significantly improving the accuracy and reliability of early warning. First, by integrating multiple sensors to collect lightning strike parameters and perform feature extraction and processing, the diversity and real-time nature of the data are ensured, key information is refined, the data volume is simplified, the processing efficiency is improved, and a data foundation is provided for subsequent feature extraction and risk assessment. Then, by constructing the correlation relationships between key features and integrating multivariate data using a fusion algorithm, the complexity of the lightning environment of the base station can be comprehensively understood, and the accuracy of risk assessment is improved. The local outlier factor algorithm is used to eliminate abnormal data points, ensuring the purity of the target parameters and avoiding the influence of outliers on the risk assessment results. Finally, based on algorithms such as the absolute method or fuzzy logic and Bayesian network, the lightning risk index is calculated to achieve a quantitative assessment of the risk. Warnings can be quickly and accurately issued according to the size of the index, providing immediate lightning protection measures for base station equipment, effectively avoiding lightning damage, and ensuring communication security. In summary, the base station lightning risk assessment method provided in the first embodiment of the present application improves the real-time response ability and decision support ability of the base station lightning risk early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings forming a part of this application are used to provide a further understanding of this application. The schematic embodiments and descriptions thereof of this application are used to explain this application and do not constitute an improper limitation of this application. In the drawings:

[0027] Figure 1 is a flowchart of the base station lightning risk assessment method provided in the first embodiment of the present application;

[0028] Figure 2 is a schematic flowchart of optional data collection provided in the first embodiment of the present application;

[0029] Figure 3 is a schematic flowchart of an optional decision tree for implementing the lightning risk early warning strategy algorithm provided in the first embodiment of the present application;

[0030] Figure 4 is a schematic flowchart of an optional message queue for processing and analyzing real-time data streams of lightning risks provided in the first embodiment of the present application;

[0031] Figure 5 is a schematic flowchart of an optional lightning risk assessment algorithm provided in the first embodiment of the present application;

[0032] Figure 6 is a schematic structural diagram of an optional lightning risk assessment system provided in the first embodiment of the present application;

[0033] Figure 7 is a schematic diagram of the base station lightning risk assessment device provided in the second embodiment of the present application;

[0034] Figure 8 It is a schematic diagram of the base station lightning risk assessment electronic device provided in Embodiment 5 of the present application. Detailed implementation manners

[0035] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0036] It should be noted that the user information involved in the present application (including but not limited to user equipment information, user personal information, collected data, used data, generated data, processed data, etc.) and data (including but not limited to data for analysis, stored data, displayed data, collected information, used information, generated information, processed information, etc.) are all information and data authorized by the user or fully authorized by all parties. Moreover, the processing of relevant data such as collection, storage, use, processing, transmission, provision, disclosure, and application complies with the relevant laws, regulations, and standards of relevant countries and regions, takes necessary confidentiality measures, does not violate public order and good customs, and provides corresponding operation entrances for users to choose to authorize or refuse. For example, an interface is set between the present system and relevant users or institutions. Before obtaining relevant information, a request for obtaining needs to be sent to the aforementioned users or institutions through the interface, and after receiving the consent information feedback from the aforementioned users or institutions, the relevant information can be obtained.

[0037] It should be noted that the present application provides corresponding operation entrances for users to choose to agree or refuse the automated decision-making results; if the user chooses to refuse, the expert decision-making process will be entered.

[0038] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0039] It should be noted that the terms "first", "second", etc. in the description, claims and the above-mentioned drawings of the present application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so as to implement the embodiments of the present application described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0040] Embodiment 1

[0041] The present invention will be described below in conjunction with the preferred implementation steps. Figure 1 It is a flowchart of a base station lightning risk assessment method provided by Embodiment 1 of the present application. As Figure 1 shown, the method includes the following steps:

[0042] Step S101, collect a variety of lightning strike parameters on the grounding wire of the base station equipment through a target sensor, and perform feature extraction processing on the variety of lightning strike parameters to obtain a variety of key features, where the target sensor is a sensor containing a variety of sensing elements.

[0043] In this Embodiment 1, by installing a composite sensor (i.e., the above-mentioned target sensor) integrating a variety of sensing elements (such as voltage, current, frequency and grounding state sensors) on the grounding wire of the base station equipment, a variety of lightning strike parameters including ground potential counterattack overvoltage, lightning activity frequency and intensity are collected in real time. Subsequently, data processing technology is used to perform feature extraction on the collected raw data, and key features such as peak voltage, current waveform, resistance fluctuation and environmental temperature and humidity are refined to accurately reflect the changes in the base station lightning environment.

[0044] Step S102, construct the correlation relationship between a variety of key features, and use a fusion algorithm to fuse the correlation relationship and a variety of lightning strike parameters to obtain the fused data.

[0045] In the first embodiment, after collecting and processing key features including peak voltage, current waveform, resistance change, temperature and humidity, etc., a correlation algorithm is used to analyze the relationships between these features. For example, the direct correlation between the voltage peak and lightning intensity is identified, or the indirect influence of the change in grounding resistance on environmental humidity is recognized. Then, a fusion algorithm, such as the Bayesian algorithm or neural network model, is adopted to combine these correlation relationships with various collected lightning strike parameters for data integration. The fused data not only contains the direct measurement values of each sensor but also incorporates the interaction information between the features, forming a more comprehensive and accurate comprehensive dataset reflecting the lightning strike risk status of the base station, providing a data basis for the formulation of subsequent early warning strategies.

[0046] Step S103, calculate and delete the abnormal data points in the fused data based on the Local Outlier Factor algorithm to obtain the target parameters.

[0047] In the first embodiment, the Local Outlier Factor (LOF for short) algorithm is used to analyze the fused multi - parameter lightning strike data (i.e., the above - mentioned fused data), that is, to detect local anomalies and identify the abnormal points that significantly deviate from the normal pattern. These abnormal data points are automatically removed according to the LOF value to improve the data quality of the dataset. The fused data after removing the abnormal data is the target parameter, which will be used for the precise assessment of lightning strike risk and the formulation of early warning strategies, thereby improving the accuracy and stability of the early warning system.

[0048] Step S104, calculate the target parameters based on the risk assessment algorithm to evaluate the lightning risk at the current moment and obtain the lightning risk index. Among them, the risk assessment algorithm includes at least one of the following: fuzzy logic algorithm, Bayesian network algorithm, and absolute method risk assessment algorithm.

[0049] In the first embodiment, a risk assessment algorithm, such as fuzzy logic, Bayesian network, or absolute method, is used to process the fused data (target parameters) after removing the abnormal points to quantitatively evaluate the current lightning risk. Among them, the absolute method directly calculates the risk index according to specific indicators such as lightning strike frequency, energy, and grounding status without comparing with historical data; fuzzy logic is suitable for dealing with uncertainties and evaluating risk factors that are difficult to accurately quantify; the Bayesian network calculates the occurrence probability of lightning risk by constructing a probability relationship model between variables. Through these algorithms, the target parameters can be comprehensively analyzed and the lightning risk index can be output, thus providing a data basis for warning level classification and lightning protection measures.

[0050] In summary, the lightning risk assessment method for a base station provided in the first embodiment of the present application collects various lightning strike parameters on the grounding wire of the base station equipment through a target sensor, and performs feature extraction processing on the various lightning strike parameters to obtain various key features, where the target sensor is a sensor including various sensing elements; constructs the association relationship between the various key features, and uses a fusion algorithm to fuse the association relationship and the various lightning strike parameters to obtain the fused data; calculates and deletes the abnormal data points in the fused data based on the local outlier factor algorithm to obtain the target parameters; calculates the target parameters based on the risk assessment algorithm to evaluate the lightning risk at the current moment, and obtains the lightning risk index, where the risk assessment algorithm includes at least one of the following: fuzzy logic algorithm, Bayesian network algorithm, and absolute method risk assessment algorithm, which solves the problem that in the related technology, when evaluating the lightning strike risk of a base station, the considered factors are relatively single, and the actual lightning energy invading the base station interior and equipment damage are not considered, resulting in inaccurate evaluation results.

[0051] Through the above steps, the comprehensive, accurate, and real-time assessment of the base station lightning risk is realized, and the accuracy and reliability of the early warning are significantly improved. First, by integrating multiple sensors to collect lightning strike parameters and perform feature extraction processing, the diversity and real-time nature of the data are ensured, the key information is refined, the data volume is simplified, the processing efficiency is improved, and a data basis is provided for subsequent feature extraction and risk assessment. Then, by constructing the association relationship between the key features and using the fusion algorithm to integrate the multi-source data, the complexity of the base station lightning environment can be comprehensively understood, and the accuracy of the risk assessment is improved. By removing the abnormal data points through the local outlier factor algorithm, the data quality of the target parameters is improved, and the influence of abnormal values on the risk assessment results is avoided. Finally, based on algorithms such as the absolute method or fuzzy logic and Bayesian network, the lightning risk index is calculated, realizing the quantitative assessment of the risk. The early warning can be quickly and accurately issued according to the size of the index, providing immediate lightning protection measures for the base station equipment, effectively avoiding lightning damage, and ensuring communication security. In summary, the lightning risk assessment method for a base station provided in the first embodiment of the present application improves the real-time response ability and decision support ability of the base station lightning risk early warning.

[0052] Optionally, in the base station lightning risk assessment method provided in Embodiment 1 of this application, a fusion algorithm is used to fuse the association relationship and various lightning strike parameters to obtain fused data, including: determining the prior probability distribution of each sensor in the target sensor based on the historical measurement information of the target sensor, and adjusting the prior probability distribution of each sensor by using various lightning strike parameters to obtain the posterior probability model of each sensor; determining the weight of each sensor through a neural network model based on the association relationship, the historical accuracy rate of each sensor, and / or the correlation degree between the current lightning risk assessment task and the parameters collected by each sensor; fusing the posterior probability models of each sensor according to the weight of each sensor to obtain a target probability model; and fusing the lightning strike parameters corresponding to each sensor in various lightning strike parameters based on the target probability model to obtain fused data.

[0053] In Embodiment 1, an advanced method combining Bayesian theory and neural network technology can be used to optimize the accuracy and efficiency of data processing.

[0054] First, based on the historical measurement information of multiple sensors, determine the prior probability distribution of each sensor for the lightning strike parameters it monitors. The prior probability distribution reflects the prediction of the lightning strike parameter distribution by each sensor without considering the current environmental conditions.

[0055] Then, use various lightning strike parameters monitored in real time, such as ground potential counterattack overvoltage, lightning activity frequency and magnitude, etc., to adjust the prior probability distribution of each sensor and generate a posterior probability model that is closer to the current environmental conditions. This adjustment process is based on Bayes' theorem, combines new observation data, updates the prediction distribution of the parameters, and improves the real-time performance and accuracy of data processing.

[0056] Secondly, comprehensively consider the association relationship, historical accuracy rate between each sensor, and the correlation degree between the current lightning risk assessment task and the parameters collected by each sensor, and automatically adjust the weight of each sensor through a neural network model. The neural network assigns a weight value to each sensor according to historical performance and correlation analysis. This weight reflects the importance and reliability of the sensor in the current risk assessment, making the result of data fusion more accurate and reliable. Exemplarily, a neural network model can be constructed based on a three-dimensional attention mechanism network, including: calculating the correlation degree between each sensor and the base station topology through spatial attention; calculating the parameter change rate through temporal attention and assigning a temporal weight to the parameter based on the parameter change rate; calculating the mutual information amount between the parameter corresponding to each sensor and the lightning risk feature through semantic attention. Train the above neural network model with various lightning strike parameters collected, and use the trained neural network model to output the weight of each sensor.

[0057] Next, according to the weights of each sensor, all posterior probability models are weighted and fused to generate an object probability model that synthesizes the information of all sensors. This fusion process ensures that the observation results of each sensor are fully considered, and the data of sensors with higher weights have a greater impact on the final model, thereby improving the accuracy and stability of the model.

[0058] Finally, based on the object probability model, secondary fusion of various lightning strike parameters is performed to eliminate redundant information and integrate key data. This step further improves the purity and comprehensiveness of the data, making the fused data more accurate and reliable, and achieving the effect of improving the accuracy of lightning risk assessment and lightning risk warning.

[0059] Through the above steps, the fusion processing of the base station lightning risk data is realized, which not only improves the efficiency and accuracy of data fusion, but also enhances the real-time response ability and decision-making support ability of lightning risk assessment, improves the safety and reliability of base station equipment under lightning strike risks, and thus helps to improve the overall level of base station lightning risk warning.

[0060] Optionally, in the base station lightning risk assessment method provided in Embodiment 1 of this application, based on the local outlier factor algorithm, abnormal data points in the fused data are calculated and deleted to obtain target parameters, including: for each first data point in the fused data, determining second data points whose distances from the first data point are less than a preset distance threshold, and determining a set of neighborhood data points of the first data point according to the second data points; calculating the reciprocal of the average reachable distance between each second data point in the neighborhood data point set and the first data point to obtain the local reachability density of the first data point; calculating the ratio of the local reachability density of each second data point in the neighborhood data point set to the local reachability density of the first data point respectively, and calculating the average value of the ratios to obtain the outlier value corresponding to each first data point; in the case where the outlier value of the first data point is greater than a preset outlier value threshold, determining the first data point as an abnormal data point, and deleting the abnormal data points in the fused data to obtain target parameters.

[0061] In Embodiment 1, the local outlier factor algorithm can be used to identify and remove abnormal points from the fused data to purify the data set and improve the accuracy of lightning risk assessment.

[0062] Exemplarily, for each data point (referred to as the first data point) in the fused dataset, first determine all data points whose distance from it is less than a preset threshold (referred to as the second data points), and form a set of neighborhood data points for this first data point. By setting a reasonable distance threshold, it is possible to select data points that truly belong to the same local environment and avoid interference from irrelevant data. In an alternative embodiment, the distance between data points can also be dynamically calculated based on the peak voltage and the physical sensitivity weights of multiple lightning strike parameters.

[0063] Then, for the selected first data point, calculate the average value of the reachable distances between each second data point and the first data point in its set of neighborhood data points, and obtain the reciprocal of this average value to quantify the local reachability density of the first data point. The local reachability density reflects the closeness of neighboring data points. Data points with high density are more likely to represent normal environmental states, while low-density points may indicate abnormal conditions or abnormal data. In an alternative embodiment, a sliding window dynamic clustering can also be introduced to dynamically adjust the neighborhood radius, thereby optimizing the set of neighborhood data points for the first data point.

[0064] Secondly, calculate the ratio between the local reachability density of each second data point in the set of neighborhood data points and the local reachability density of the first data point to quantify the degree of abnormality of the first data point relative to its neighborhood. By taking the average of all ratios, an abnormality index for the first data point is obtained, that is, the local outlier factor value (which can be abbreviated as the LOF value), which is also the above-mentioned outlier value.

[0065] Finally, when the outlier value of a certain first data point exceeds a preset outlier threshold, determine that this data point is an abnormal data point. These abnormal points may be caused by sensor failures, extreme environmental changes, or other conditions. The existence of abnormal points will affect the accuracy of subsequent risk assessments. Therefore, automatically delete these abnormal data points to generate a cleaned dataset from the fused dataset, that is, the target parameter, for lightning risk assessment.

[0066] Through the above steps, it is ensured that lightning risk assessment is carried out based on the most reliable and accurate data, thereby improving the effectiveness of early warning information and the scientific nature of decision-making. Through the fine screening of the LOF algorithm, abnormal points can be effectively eliminated without affecting normal data, providing high-quality data support for the lightning risk warning of the base station and further ensuring the safe operation of the base station equipment.

[0067] Optionally, in the base station lightning risk assessment method provided in Embodiment 1 of this application, feature extraction processing is performed on multiple lightning strike parameters to obtain multiple key features, including: using a digital filter to remove noise and interference signals in the multiple lightning strike parameters to obtain the filtered multiple lightning strike parameters; checking the integrity and correctness of the filtered multiple lightning strike parameters through cyclic redundancy check to obtain the verified multiple lightning strike parameters; extracting key features from the verified multiple lightning strike parameters, where the key features at least include: peak voltage, current waveform, resistance value change trend, temperature and humidity changes.

[0068] In Embodiment 1, in order to ensure the efficient and accurate operation of the lightning risk warning system, data preprocessing needs to be performed on the multiple lightning strike parameters collected, including three steps: digital filtering, data verification, and key feature extraction.

[0069] First, a digital filter is used to remove noise and interference signals from the multiple lightning strike parameters collected. The digital filter can effectively identify and eliminate signal noise caused by environmental factors, equipment aging, or electromagnetic interference, etc., ensure the purity of the data, reduce the false alarm and missed alarm rates, and improve the accuracy of lightning activity monitoring. By optimizing the filter parameters, it can quickly respond to different types of noise and guarantee the reliability of subsequent analysis.

[0070] Then, cyclic redundancy check (which can be abbreviated as CRC, by calculating the check code of the data and comparing it with the preset value, quickly detecting any errors or data packet damage during the transmission process) is performed on the filtered multiple lightning strike parameters to check the integrity and correctness of the data.

[0071] Finally, key features are extracted from the verified multiple lightning strike parameters, including but not limited to peak voltage, current waveform, resistance value change trend, temperature change, humidity change, lightning pulse breakdown probability, multi-conductor coupling coefficient, and grounding body degradation rate, etc., to comprehensively reflect the lightning environment state of the base station. By extracting these key features, data can be analyzed and processed more efficiently, thereby achieving the effect of identifying the patterns and trends of lightning activities, and further achieving the effect of improving the accuracy of lightning risk assessment and lightning risk warning. Exemplarily, adaptive variational mode decomposition (AVMD) can be used to process the verified multiple lightning strike parameters, and the processed data is input into the trained lightweight capsule network (CapsNet-Lite), setting the dynamic routing to iterate 3 times, and the output feature vector can include: lightning pulse breakdown probability, multi-conductor coupling coefficient, and grounding body degradation rate, etc.

[0072] Through the above steps, the data quality of various lightning strike parameters and the data processing efficiency are improved, providing pure, complete and critical information for the lightning risk warning system, making the generation of warning signals more timely and accurate, thereby effectively enhancing the lightning protection ability of the base station and ensuring communication security.

[0073] Optionally, in the base station lightning risk assessment method provided in Embodiment 1 of this application, the various lightning strike parameters at least include: ground potential counterattack overvoltage, lightning strike frequency, lightning energy magnitude, and grounding status of base station equipment; the risk assessment algorithm is an absolute method risk assessment algorithm, and based on the risk assessment algorithm, the target parameters are calculated to evaluate the lightning risk at the current moment, and a lightning risk index is obtained, including: standardizing the index value of each parameter in the target parameters to obtain the standardized target parameters, where the lightning energy magnitude is characterized by a preset unit, and the grounding status of the base station equipment is characterized by the integrity of the grounding system of the base station equipment; determining the influence degree of each parameter in the target parameters on the current lightning risk assessment task, and determining the weight of each parameter according to the influence degree of each parameter; using the method of weighted summation to calculate the standardized target parameters and the weights of each parameter in the target parameters to obtain the lightning risk index.

[0074] In Embodiment 1, for the various lightning strike parameters collected and preprocessed, the absolute method risk assessment algorithm can be used for quantitative analysis of lightning risk.

[0075] First, standardize each index value in the target parameters to eliminate the differences in different parameter dimensions and magnitudes, and ensure the accuracy of subsequent calculations. In an optional embodiment, the lightning energy magnitude can be characterized in joules (i.e., the above-mentioned preset unit) for direct comparison of numerical values; while the grounding status of the base station equipment is measured by the integrity of the grounding system, reflecting the effectiveness of its lightning protection performance; the lightning strike frequency can be the number of lightning strikes within a preset time period. The standardized target parameters are more suitable as the input for risk assessment, improving the applicability of the algorithm and the accuracy of the results. Exemplarily, the lightning energy E can be subjected to hierarchical fuzzy standardization, for example, using an S-shaped function to process the lightning energy E.

[0076] Then, analyze the influence degree of each parameter on the current lightning risk assessment task, and then determine the weight of each parameter. When analyzing the influence degree of each parameter, it can be analyzed according to the historical performance of the sensor corresponding to each parameter, physical meaning, and relative importance in risk assessment. For example, the integrity of the grounding system may be given a higher weight because it is directly related to the strength of the base station's lightning protection ability; and the weight of the lightning energy magnitude may also be relatively high, reflecting the potential destructive power of the lightning strike. Exemplarily, a deep cross network (DCN) can be constructed to compress the output dimension to 5 dimensions and map it to the weight space.

[0077] Finally, a weighted summation method is adopted to combine the standardized target parameters with their respective weights to calculate the lightning risk index. The calculation formula can be expressed as: , where is the weight of the i-th parameter, is the standardized index value corresponding to the i-th parameter, and n is the total number of parameters.

[0078] Through this series of steps, a comprehensive assessment of lightning strike risks can be carried out based on multi-dimensional data. The generated lightning risk index provides an intuitive quantitative basis for formulating early warning strategies, improves the accuracy and reliability of base station lightning assessments, and effectively ensures the stable operation of base station equipment.

[0079] Optionally, in the base station lightning risk assessment method provided in Embodiment 1 of this application, after collecting various lightning strike parameters on the grounding wire of the base station equipment through multiple sensors, the above method further includes: constructing a decision tree model based on the parameters with a relatively high degree of correlation with lightning risks among the various lightning strike parameters; training the decision tree model with the various lightning strike parameters to obtain a trained decision tree model; evaluating the trained decision tree model based on a cross-validation algorithm, and optimizing the trained decision tree model based on a reinforcement learning strategy according to the evaluation results to obtain an optimized decision tree model; calculating the various lightning strike parameters collected within a preset time period based on the optimized decision tree model to generate and execute an early warning strategy.

[0080] In this Embodiment 1, in order to realize the automatic generation of personalized early warning strategies and the continuous optimization of early warning strategies, a decision tree model can be used to analyze and process various lightning strike parameters to obtain real-time lightning early warning strategies.

[0081] Exemplarily, a basic decision tree model is constructed based on the parameters with a relatively high degree of correlation with lightning risks among the various lightning strike parameters, such as ground potential counterattack overvoltage, lightning energy magnitude, etc., ensuring that the construction of the model focuses on the parameters most influential for lightning risk assessment and improving the accuracy of the model.

[0082] Then, the constructed decision tree model is trained using a historical data set. Through continuous learning and adjustment, the model gradually learns and understands the patterns and laws of lightning risk occurrences, forming a trained decision tree model. The historical data set can be various lightning strike parameters collected in the past, as well as the corresponding lightning early warning strategies when lightning phenomena occur.

[0083] Secondly, the cross-validation algorithm is used to evaluate the trained decision tree model. By dividing the historical dataset into a training set and a test set, the generalization ability of the model on unseen data is examined to ensure the stability and accuracy of the model. According to the evaluation results, the model is further optimized based on the reinforcement learning strategy. By simulating the reward or punishment mechanism in the decision-making process, the model parameters are automatically adjusted to better adapt to the actual lightning environment, and the optimized decision tree model is obtained.

[0084] Finally, based on the optimized decision tree model, various lightning strike parameters collected within a preset time period (i.e., the time period for which the lightning warning strategy is required) are calculated in real time. Based on the prediction results of the model, a warning strategy for the current environment is automatically generated and immediately executed. The generation and execution of this strategy significantly improve the response speed and adaptability of the warning system, enabling timely measures to be taken to reduce the potential damage of lightning to the base station equipment.

[0085] In an alternative embodiment, digital twin technology can also be used to enhance the effectiveness and response speed of base station lightning strike risk warnings through high-precision virtual models and intelligent strategy generation. The relationship between the magnetic field change caused by the electric field change and the electric displacement vector and charge density is constructed using Maxwell's equations in electromagnetic field theory to simulate the real-time electromagnetic environment changes around the base station. The VR visualization interface allows users to observe the lightning strike path from any angle, providing an intuitive 360° view to help understand and analyze the impact of lightning activities on the base station. The warning strategy is ensured to follow industry norms according to the standards issued by authoritative organizations such as the International Electrotechnical Commission (IEC) to improve its practicality and safety. The learning and decision-making of the intelligent system are guided by setting the reward function R. The reward function R can comprehensively consider lightning strike parameters such as response time, false alarm rate, and miss rate. Finally, with the help of discrete event simulation (DES) technology, the execution effects of the strategy in different lightning scenarios are simulated to achieve the dynamic evaluation and rapid optimization of the warning strategy.

[0086] Through the above steps, the most suitable warning strategy can be generated for the specific environment of different base stations, providing efficient and accurate decision-making support for base station lightning protection and effectively ensuring the safe and stable operation of the base station.

[0087] Optionally, in the base station lightning risk assessment method provided in Embodiment 1 of this application, after collecting multiple lightning strike parameters on the grounding wire of the base station equipment through multiple sensors, the above method further includes: transmitting the multiple lightning strike parameters to multiple headers of a message queue; transmitting the multiple lightning strike parameters stored in the multiple headers to multiple data processing modules according to a target mapping relationship, where each data processing module among the multiple data processing modules is responsible for performing different data processing operations on the multiple lightning strike parameters; the target mapping relationship refers to the mapping relationship between the multiple headers and the multiple data processing modules; the transmission process of the multiple lightning strike parameters is carried out by using parallel compressive sensing technology.

[0088] In Embodiment 1, in the real-time monitoring module, parallel compressive sensing technology and message queue mechanism can be adopted to efficiently process and transmit multiple lightning strike parameter data, ensuring the real-time nature and integrity of information.

[0089] Exemplarily, multiple collected lightning strike parameters, such as ground potential counterattack overvoltage, lightning activity frequency, lightning energy magnitude, and the grounding state of base station equipment, etc., are transmitted to different headers of the message queue. Each header corresponds to a specific type of parameter data. For example, the "overvoltage data" header is used to store the ground potential counterattack overvoltage information, or headers such as "data filtering processing", "data verification", "data fusion", and "abnormal data elimination" are used to store the ground potential counterattack overvoltage information to ensure clear data classification. Each header serves as a container for data, ensuring the orderliness and structuring of data during the transmission process, facilitating parallel access and processing by multiple data processing modules (or processes).

[0090] Then, based on the target mapping relationship, the multiple lightning strike parameters stored in the multiple headers are quickly and accurately transmitted to multiple data processing modules. Each data processing module is assigned a specific data processing task, such as data filtering processing, data verification, data fusion, and abnormal data elimination, etc. The parallel processing method is adopted to improve the speed and efficiency of data processing. Among them, the parallel compressive sensing technology based on the message queue can reduce the data volume during the data transmission process, reduce the transmission delay, and at the same time maintain the integrity of key information, ensuring the efficient transmission of data from the message queue to each data processing module.

[0091] Through the above steps, the real-time collection, classified storage, and efficient processing of multiple lightning strike parameter data are realized, providing a solid data foundation for the formulation of risk assessment and early warning strategies. By combining the parallel compressive sensing technology and the message queue mechanism, the real-time nature and accuracy of data processing are ensured, improving the overall efficiency of the lightning risk early warning system, and further achieving the effect of improving the reliability of lightning protection for the base station.

[0092] Optionally, in the base station lightning risk assessment method provided in Embodiment 1 of this application, after calculating the target parameters based on the risk assessment algorithm to evaluate the lightning risk at the current moment and obtaining the lightning risk index, the above method further includes: determining the lightning risk level at the current moment according to the lightning risk index, where the lightning risk level includes at least one of the following: the first level, the second level, and the execution strategies corresponding to the first level and the second level are different; sending the lightning risk level and the execution strategy corresponding to the lightning risk level to the target object, and receiving the response information of the target object in response to the lightning risk level; generating a lightning risk handling report based on the lightning risk index, the lightning risk level, the execution strategy corresponding to the lightning risk level, and the response information.

[0093] In this Embodiment 1, after determining the lightning risk index, the lightning risk level of the base station at the current moment can be determined according to this lightning risk index. The lightning risk level at the current moment is determined according to the high or low of the risk index. The lightning risk level includes at least the first level and the second level. Exemplarily, the lightning risk level may include three levels: low risk, medium risk, and high risk. The execution strategy corresponding to low risk may focus on routine monitoring and preparation, while the execution strategy corresponding to high risk requires immediate adoption of emergency protection measures, such as powering off equipment, starting lightning protection devices, etc.

[0094] Then, after determining the lightning risk level, the relevant information of the lightning risk level and the corresponding execution strategy are quickly sent to the target object by means of text messages, emails, mobile environmental control systems, or directly sending to the mobile device of a certain object, etc. The target object may include base station maintenance personnel, network operation management centers, etc. At the same time, receive the response information of the target object to the lightning risk level, that is, whether the warning information has been received, and confirm the status of executing the corresponding strategy to ensure the effective transmission and execution of the warning information.

[0095] Finally, based on the lightning risk index, the lightning risk level, the execution strategy, and the response information, a lightning risk handling report is automatically generated, including but not limited to: the time when the warning is triggered, the specific level of the lightning risk, the protection measures taken, the response information, and the final processing result.

[0096] In an alternative embodiment, a dynamic hierarchical response can be performed based on the lightning risk index. A spatio-temporal adaptive threshold model is constructed, and the risk levels are dynamically divided by combining the lightning risk index, thunderstorm season, and base station altitude (0.7 + 0.1×altitude / 1000), forming a four-level response mechanism, including: Red (≥0.9): Cut off non-essential loads and activate the lightning elimination device; Orange (0.75 - 0.89): Enhance SPD protection and start the backup power supply; Yellow (0.6 - 0.74): Limit the communication bandwidth and activate the electromagnetic shielding; Blue (0.3 - 0.59): Automatically generate equipment inspection work orders. When the lightning risk index is greater than 0.8, switch the grounding mode (TN-S↔IT) and control the on / off timing of multiple-stage SPDs (with an accuracy of 25ns); when the lightning risk index in the communication machine room is greater than 0.7, automatically connect the metal layer of the optical cable; when the lightning risk index in the energy storage power station is greater than 0.6, force the battery pack to be isolated. Project a three-dimensional lightning field intensity map through an augmented reality (AR) head-mounted device, and generate a real-time evacuation path in combination with the Dijkstra algorithm.

[0097] Through the above steps, not only can the timely transmission of lightning risk-related information be ensured, but also the improvement of the lightning warning strategy for base stations can be promoted by generating detailed processing reports, providing closed-loop management for the lightning protection of base stations, ensuring the long-term safe and stable operation of base station equipment, and further achieving the effect of improving the security and reliability of the communication network.

[0098] Optionally, in the first embodiment, Figure 2 is a schematic diagram of the data collection process. The data obtained by the composite sensor (i.e., the above-mentioned target sensor) is first transmitted to the adaptive sampling technology and low-latency circuit. The adaptive sampling technology can automatically adjust the sampling rate according to the signal change to ensure that detailed enough information is captured during the peak of lightning activity; while the low-latency circuit ensures the rapid transfer of data from collection to processing, reduces signal delay, and improves the immediacy of the warning. The data flows through the data filtering link to eliminate noise and interference and keep the signal pure; the preprocessing and verification steps further ensure the integrity and accuracy of the data. Through the data fusion module, the data from different sensors is integrated to form a unified data view; the abnormal data rejection algorithm identifies and excludes those readings that significantly deviate from the normal range to avoid abnormal values misleading subsequent risk assessments. The processed data is uploaded through a secure and reliable channel in combination with the current temperature information of the base station for further analysis and risk assessment.

[0099] Optionally, in the first embodiment, Figure 3It is a schematic diagram of the algorithm process for implementing the lightning strike risk warning strategy using a decision tree. Historical lightning strike data including lightning frequency, intensity, ground potential counterattack overvoltage, equipment leakage current, grounding resistance, etc. is collected. By screening the collected historical lightning strike data, features crucial for lightning strike risk warning, such as geographical location, season, meteorological conditions, historical records of lightning activities, etc., are selected. Using the previously selected feature data, the decision tree model is started to be trained. The historical lightning strike data is divided into a training set and a validation set, and the accuracy and generalization ability of the decision tree are evaluated through multiple iterations to ensure reliable prediction on unseen data, thereby enhancing the practicality and effectiveness of the warning strategy. According to the rules output by the decision tree model, based on the current lightning environment and the status of the base station, personalized warning strategies are automatically generated, such as adjusting the working mode of protection equipment, notifying maintenance personnel to strengthen inspections, etc., to quickly respond to lightning activities and reduce the potential damage of lightning strikes to base station equipment.

[0100] Optionally, in the first embodiment, Figure 4 It is a schematic diagram of the process for the message queue to process and analyze the real-time data stream of lightning strike risk. The composite sensor continuously monitors the base station environment, including ground potential counterattack overvoltage, lightning activity parameters, and equipment status, etc. The data collected by the sensor is immediately sent to the message queue through a reliable data transmission channel, such as a double-layer shielded coaxial cable, to ensure the immediacy and integrity of the data and avoid the delay problem in traditional data transmission. Through the stream processing ability of the message queue, the real-time lightning strike data is cleaned, transformed, and aggregated. At the same time, using parallel computing resources, the patterns and trends of lightning strike risks are quickly identified to provide real-time basis for risk assessment. To ensure the long-term retention and traceability of the data, the preliminarily analyzed data is stored in the database as a valuable resource for subsequent research and auditing. At the same time, it also provides data support for historical data analysis and the optimization of warning strategies. According to the analysis results, when a lightning strike risk pattern or trend is detected, a lightning strike warning will be immediately issued to the base station management personnel and the maintenance team through channels such as text messages and the mobile environment monitoring system to ensure prompt action and reduce the damage of lightning strikes to base station equipment.

[0101] Optionally, in the first embodiment, Figure 5It is a schematic flow diagram of the lightning strike risk assessment algorithm. Determine the key parameters required for risk assessment (i.e., various lightning strike parameters mentioned above), such as lightning strike frequency, lightning strike energy, and grounding status. Convert various lightning strike parameters into specific numerical values, and give a quantitative score to the magnitude of lightning strike frequency, lightning strike energy, and the integrity of the grounding system. Based on the importance of each indicator to the lightning strike risk, assign weights to ensure that more dangerous indicators occupy a more important position in the overall assessment. For example, the effectiveness of the grounding system is often given a higher weight because it is directly related to the lightning protection ability of the base station. Use the method of weighted summation to calculate the overall risk index by combining the quantitative values of each indicator and the assigned weights. Map the calculated risk index to predefined risk levels, such as low risk, medium risk, and high risk, which helps the base station take corresponding protection measures according to the risk level to ensure the safety of equipment.

[0102] Optionally, in the first embodiment, Figure 6 It is a schematic structural diagram of the lightning risk assessment system. The data acquisition module collects the base station environment and equipment status parameters in real time through a composite sensor (i.e., the target sensor mentioned above), including ground potential counterattack overvoltage, lightning activity data, and grounding status. The data fusion module receives multi-source data from the data acquisition module, and integrates the scattered information through strategies such as feature extraction and association, and intelligent weight assignment. The data rejection module uses the LOF algorithm to identify and remove outliers in the fused data. The early warning strategy formulation module formulates personalized early warning strategies through the decision tree algorithm based on historical lightning strike data and base station characteristics. The real-time monitoring module continuously monitors the base station environment, updates the data input of the risk assessment module in a timely manner, and ensures the real-time nature of the assessment results. The risk index calculated by the risk assessment module is processed by the early warning release module, determines the lightning risk level based on the risk index, and sends the early warning information and execution strategy to the target object, and at the same time receives the response information to generate a risk handling report.

[0103] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0104] Embodiment 2

[0105] The second embodiment of the present application also provides a base station lightning risk assessment device. It should be noted that the base station lightning risk assessment device in the second embodiment of the present application can be used to execute the method for base station lightning risk assessment provided in the first embodiment of the present application. The base station lightning risk assessment device provided in the second embodiment of the present application is introduced below.

[0106] Figure 7It is a schematic diagram of the base station lightning risk assessment device according to Embodiment 2 of the present application. As Figure 7 shown, the device includes: a collection unit 701, a fusion unit 702, a deletion unit 703, and a first evaluation unit 704.

[0107] Specifically, the collection unit 701 is configured to collect various lightning strike parameters on the grounding wire of the base station equipment through a target sensor, and perform feature extraction processing on the various lightning strike parameters to obtain various key features, where the target sensor is a sensor including various types of sensing elements.

[0108] The fusion unit 702 is configured to construct an association relationship between various key features, and use a fusion algorithm to fuse the association relationship and various lightning strike parameters to obtain fused data.

[0109] The deletion unit 703 is configured to calculate and delete abnormal data points in the fused data based on the local outlier factor algorithm to obtain target parameters.

[0110] The first evaluation unit 704 is configured to calculate the target parameters based on a risk assessment algorithm, evaluate the lightning risk at the current moment, and obtain a lightning risk index, where the risk assessment algorithm includes at least one of the following: a fuzzy logic algorithm, a Bayesian network algorithm, and an absolute method risk assessment algorithm.

[0111] The base station lightning risk assessment device provided in Embodiment 2 of the present application collects various lightning strike parameters on the grounding wire of the base station equipment through the collection unit 701 through a target sensor, and performs feature extraction processing on the various lightning strike parameters to obtain various key features, where the target sensor is a sensor including various types of sensing elements; the fusion unit 702 constructs an association relationship between various key features, and uses a fusion algorithm to fuse the association relationship and various lightning strike parameters to obtain fused data; the deletion unit 703 calculates and deletes abnormal data points in the fused data based on the local outlier factor algorithm to obtain target parameters; the first evaluation unit 704 calculates the target parameters based on a risk assessment algorithm, evaluates the lightning risk at the current moment, and obtains a lightning risk index, where the risk assessment algorithm includes at least one of the following: a fuzzy logic algorithm, a Bayesian network algorithm, and an absolute method risk assessment algorithm, which solves the problem that in the related art, when evaluating the lightning strike risk of a base station, the considered factors are relatively single, and the actual lightning energy and equipment damage invading the base station interior are not considered, resulting in inaccurate evaluation results.

[0112] Through the above steps, a comprehensive, accurate, and real-time assessment of the lightning risk of the base station is achieved, significantly improving the accuracy and reliability of early warning. First, by integrating multiple sensors to collect lightning strike parameters and perform feature extraction and processing, the diversity and real-time nature of the data are ensured, key information is refined, the data volume is simplified, the processing efficiency is improved, and a data foundation is provided for subsequent feature extraction and risk assessment. Then, by constructing the correlation relationships between key features and using a fusion algorithm to integrate multi-source data, the complexity of the base station lightning environment can be comprehensively understood, and the accuracy of risk assessment is improved. The local outlier factor algorithm is used to eliminate abnormal data points, ensuring the purity of the target parameters and avoiding the influence of outliers on the risk assessment results. Finally, based on algorithms such as the absolute method or fuzzy logic, Bayesian network, etc., the lightning risk index is calculated to achieve a quantitative assessment of the risk. Warnings can be quickly and accurately issued according to the magnitude of the index, providing immediate lightning protection measures for base station equipment, effectively avoiding lightning damage, and ensuring communication security. In summary, the base station lightning risk assessment method provided in the first embodiment of the present application improves the real-time response ability and decision-making support ability of the base station lightning risk early warning.

[0113] Optionally, in the base station lightning risk assessment device provided in the second embodiment of the present application, the above-mentioned fusion unit 702 includes: a first determination subunit, configured to determine the prior probability distribution of each sensor among multiple sensors based on the historical measurement information of the multiple sensors, and adjust the prior probability distribution of each sensor by using multiple lightning strike parameters to obtain the posterior probability model of each sensor; a second determination subunit, configured to determine the weight of each sensor through a neural network model based on the correlation relationship, the historical accuracy rate of each sensor, and / or the degree of correlation between the current lightning risk assessment task and the parameters collected by each sensor; a first fusion subunit, configured to fuse the posterior probability models of each sensor according to the weight of each sensor to obtain a target probability model; a second fusion subunit, configured to fuse the lightning strike parameters corresponding to each sensor among multiple lightning strike parameters based on the target probability model to obtain the fused data.

[0114] Optionally, in the base station lightning risk assessment device provided in Embodiment 2 of this application, the above-mentioned deletion unit 703 includes: a third determination subunit, configured to, for each first data point in the fused data, determine a second data point whose distance from the first data point is less than a preset distance threshold, and determine a neighborhood data point set of the first data point according to the second data point; a first calculation subunit, configured to calculate the reciprocal of the average reachable distance between each second data point in the neighborhood data point set and the first data point, to obtain the local reachable density of the first data point; a second calculation subunit, configured to calculate the ratio between the local reachable density of each second data point in the neighborhood data point set and the local reachable density of the first data point respectively, and calculate the average value of the ratios, to obtain an outlier corresponding to each first data point; a deletion subunit, configured to, when the outlier of the first data point is greater than a preset outlier threshold, determine the first data point as an abnormal data point, and delete the abnormal data points in the fused data, to obtain target parameters.

[0115] Optionally, in the base station lightning risk assessment device provided in Embodiment 2 of this application, the above-mentioned acquisition unit 701 includes: a filtering subunit, configured to use a digital filter to remove noise and interference signals in a variety of lightning strike parameters, to obtain filtered various lightning strike parameters; an inspection subunit, configured to check the integrity and correctness of the filtered various lightning strike parameters through cyclic redundancy check, to obtain verified various lightning strike parameters; an extraction subunit, configured to extract key features from the verified various lightning strike parameters, where the key features at least include: peak voltage, current waveform, resistance value change trend, temperature and humidity change.

[0116] Optionally, in the base station lightning risk assessment device provided in Embodiment 2 of this application, the above-mentioned various lightning strike parameters at least include: ground potential counterattack overvoltage, lightning strike frequency, lightning energy magnitude, grounding status of base station equipment; the risk assessment algorithm is an absolute method risk assessment algorithm, and the first assessment unit 704 includes: a processing subunit, configured to perform standardization processing on the index value of each parameter in the target parameters, to obtain standardized target parameters, where the lightning energy magnitude is characterized by a preset unit, and the grounding status of base station equipment is characterized by the integrity of the grounding system of the base station equipment; a fourth determination subunit, configured to determine the influence degree of each parameter in the target parameters on the current lightning risk assessment task, and determine the weight of each parameter according to the influence degree of each parameter; a third calculation subunit, configured to calculate the standardized target parameters and the weight of each parameter in the target parameters by using the method of weighted summation, to obtain a lightning risk index.

[0117] Optionally, in the base station lightning risk assessment device provided in Embodiment 2 of the present application, the above device further includes: a construction unit, configured to construct a decision tree model based on the parameters with a relatively high degree of relevance to lightning risk among the multiple lightning strike parameters after collecting the multiple lightning strike parameters on the grounding wire of the base station equipment through a target sensor; a training unit, configured to train the decision tree model with the multiple lightning strike parameters to obtain a trained decision tree model; a second evaluation unit, configured to evaluate the trained decision tree model based on a cross-validation algorithm and optimize the trained decision tree model based on a reinforcement learning strategy according to the evaluation result to obtain an optimized decision tree model; a calculation unit, configured to calculate the multiple lightning strike parameters collected within a preset time period based on the optimized decision tree model to generate and execute a warning strategy.

[0118] Optionally, in the base station lightning risk assessment device provided in Embodiment 2 of the present application, the above device further includes: a first transmission unit, configured to transmit the multiple lightning strike parameters to multiple headers of a message queue after collecting the multiple lightning strike parameters on the grounding wire of the base station equipment through a target sensor; a second transmission unit, configured to transmit the multiple lightning strike parameters stored in the multiple headers to multiple data processing modules according to a target mapping relationship, where each data processing module among the multiple data processing modules is responsible for performing different data processing operations on the multiple lightning strike parameters; the target mapping relationship refers to the mapping relationship between the multiple headers and the multiple data processing modules; the transmission process of the multiple lightning strike parameters is performed using parallel compressive sensing technology.

[0119] Optionally, in the base station lightning risk assessment device provided in Embodiment 2 of the present application, the above device further includes: a determination unit, configured to determine the lightning risk level at the current moment according to the lightning risk index after calculating the target parameters based on a risk assessment algorithm to evaluate the lightning risk at the current moment, where the lightning risk level includes at least one of the following: a first level and a second level, and the execution strategies corresponding to the first level and the second level are different; a receiving unit, configured to send the lightning risk level and the execution strategy corresponding to the lightning risk level to a target object and receive the response information of the target object in response to the lightning risk level; a generating unit, configured to generate a lightning risk processing report based on the lightning risk index, the lightning risk level, the execution strategy corresponding to the lightning risk level, and the response information.

[0120] The base station lightning risk assessment device includes a processor and a memory. The above acquisition unit 701, fusion unit 702, deletion unit 703, first evaluation unit 704, etc. are all stored in the memory as program units, and the processor executes the above program units stored in the memory to implement corresponding functions.

[0121] The processor contains cores, which retrieve corresponding program units from the memory. One or more cores can be set, and the accuracy of the lightning risk assessment result of the base station can be improved by adjusting the core parameters.

[0122] The memory may include non-permanent memory in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.

[0123] Embodiment 3 of the present invention provides a computer-readable storage medium, on which a program is stored, and when the program is executed by a processor, it implements the base station lightning risk assessment method.

[0124] Embodiment 4 of the present invention provides a processor, which is used to run a program. When the program runs, it executes the base station lightning risk assessment method.

[0125] As Figure 8 shown, Embodiment 5 of the present invention provides an electronic device, which includes a processor, a memory, and a program stored on the memory and executable on the processor. When the processor executes the program, the following steps are implemented: collecting various lightning strike parameters on the grounding wire of the base station equipment through a target sensor, and performing feature extraction processing on the various lightning strike parameters to obtain various key features, where the target sensor is a sensor containing various sensing elements; constructing the association relationship between the various key features, and using a fusion algorithm to fuse the association relationship and the various lightning strike parameters to obtain the fused data; calculating and deleting the abnormal data points in the fused data based on the local outlier factor algorithm to obtain the target parameters; calculating the target parameters based on the risk assessment algorithm, and evaluating the lightning risk at the current moment to obtain the lightning risk index, where the risk assessment algorithm includes at least one of the following: fuzzy logic algorithm, Bayesian network algorithm, and absolute method risk assessment algorithm.

[0126] When the processor executes the program, the following steps are also implemented: using a fusion algorithm to fuse the association relationship and the various lightning strike parameters to obtain the fused data, including: determining the prior probability distribution of each sensor in the target sensor based on the historical measurement information of the target sensor, and adjusting the prior probability distribution of each sensor using the various lightning strike parameters to obtain the posterior probability model of each sensor; determining the weight of each sensor through a neural network model based on the association relationship, the historical accuracy of each sensor, and / or the correlation degree between the current lightning risk assessment task and the parameters collected by each sensor; fusing the posterior probability models of each sensor according to the weight of each sensor to obtain the target probability model; and fusing the lightning strike parameters corresponding to each sensor in the various lightning strike parameters based on the target probability model to obtain the fused data.

[0127] When the processor executes the program, the following steps are also implemented: calculating and removing abnormal data points in the fused data based on the local outlier factor algorithm to obtain target parameters, including: for each first data point in the fused data, determining second data points whose distances from the first data point are less than a preset distance threshold, and determining a set of neighborhood data points of the first data point based on the second data points; calculating the reciprocal of the average reachability distance between each second data point in the set of neighborhood data points and the first data point to obtain the local reachability density of the first data point; calculating the ratio between the local reachability density of each second data point in the set of neighborhood data points and the local reachability density of the first data point respectively, and calculating the average value of the ratios to obtain the outlier value corresponding to each first data point; in the case where the outlier value of the first data point is greater than a preset outlier value threshold, determining the first data point as an abnormal data point, and removing the abnormal data points in the fused data to obtain target parameters.

[0128] When the processor executes the program, the following steps are also implemented: performing feature extraction processing on multiple lightning strike parameters to obtain multiple key features, including: using a digital filter to remove noise and interference signals in the multiple lightning strike parameters to obtain the filtered multiple lightning strike parameters; checking the integrity and correctness of the filtered multiple lightning strike parameters through cyclic redundancy check to obtain the verified multiple lightning strike parameters; extracting key features from the verified multiple lightning strike parameters, where the key features at least include: peak voltage, current waveform, resistance value change trend, temperature and humidity change.

[0129] When the processor executes the program, the following steps are also implemented: the multiple lightning strike parameters at least include: ground potential counterattack overvoltage, lightning strike frequency, lightning energy magnitude, grounding status of base station equipment; the risk assessment algorithm is an absolute method risk assessment algorithm, calculating the target parameters based on the risk assessment algorithm to evaluate the lightning risk at the current moment to obtain a lightning risk index, including: performing standardization processing on the index value of each parameter in the target parameters to obtain the standardized target parameters, where the lightning energy magnitude is characterized by a preset unit, and the grounding status of base station equipment is characterized by the integrity of the grounding system of the base station equipment; determining the influence degree of each parameter in the target parameters on the current lightning risk assessment task, and determining the weight of each parameter according to the influence degree of each parameter; using the method of weighted summation to calculate the standardized target parameters and the weight of each parameter in the target parameters to obtain the lightning risk index.

[0130] When the processor executes the program, the following steps are also implemented: After collecting various lightning strike parameters on the grounding wire of the base station device through various sensors, the above method further includes: constructing a decision tree model based on the parameters with a relatively high degree of relevance to the lightning risk among the various lightning strike parameters; training the decision tree model with the various lightning strike parameters to obtain a trained decision tree model; evaluating the trained decision tree model based on the cross-validation algorithm, and optimizing the trained decision tree model based on the evaluation results and the reinforcement learning strategy to obtain an optimized decision tree model; calculating the various lightning strike parameters collected within a preset time period based on the optimized decision tree model to generate and execute a warning strategy.

[0131] When the processor executes the program, the following steps are also implemented: After collecting various lightning strike parameters on the grounding wire of the base station device through various sensors, the above method further includes: transmitting the various lightning strike parameters to multiple headers of the message queue; transmitting the various lightning strike parameters stored in the multiple headers to multiple data processing modules according to the target mapping relationship, where each data processing module among the multiple data processing modules is responsible for performing different data processing operations on the various lightning strike parameters; the target mapping relationship refers to the mapping relationship between the multiple headers and the multiple data processing modules; the transmission process of the various lightning strike parameters is transmitted using the parallel compressive sensing technology.

[0132] When the processor executes the program, the following steps are also implemented: After calculating the target parameters based on the risk assessment algorithm, evaluating the lightning risk at the current moment, and obtaining the lightning risk index, the above method further includes: determining the lightning risk level at the current moment according to the lightning risk index, where the lightning risk level includes at least one of the following: the first level, the second level, and the execution strategies corresponding to the first level and the second level are different; sending the lightning risk level and the execution strategy corresponding to the lightning risk level to the target object, and receiving the response information of the target object to the lightning risk level; generating a lightning risk processing report based on the lightning risk index, the lightning risk level, the execution strategy corresponding to the lightning risk level, and the response information.

[0133] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0134] The present application also provides a computer program product which, when executed on a data processing device, is adapted to execute a program initialized with the following method steps: collecting a variety of lightning strike parameters on the grounding wire of a base station device through a target sensor, and performing feature extraction processing on the variety of lightning strike parameters to obtain a variety of key features, wherein the target sensor is a sensor including a variety of sensing elements; constructing an association relationship between the variety of key features, and using a fusion algorithm to fuse the association relationship and the variety of lightning strike parameters to obtain fused data; calculating and deleting abnormal data points in the fused data based on the local outlier factor algorithm to obtain target parameters; calculating the target parameters based on a risk assessment algorithm to evaluate the lightning risk at the current moment and obtain a lightning risk index, wherein the risk assessment algorithm includes at least one of the following: fuzzy logic algorithm, Bayesian network algorithm, absolute method risk assessment algorithm.

[0135] When executed on a data processing device, it is also adapted to execute a program initialized with the following method steps: determining the prior probability distribution of each sensor in the target sensor based on the historical measurement information of the target sensor, and using a variety of lightning strike parameters to adjust the prior probability distribution of each sensor to obtain the posterior probability model of each sensor; determining the weight of each sensor through a neural network model based on the association relationship, the historical accuracy of each sensor, and / or the correlation degree between the current lightning risk assessment task and the parameters collected by each sensor; fusing the posterior probability models of each sensor according to the weight of each sensor to obtain a target probability model; fusing the lightning strike parameters corresponding to each sensor in the variety of lightning strike parameters based on the target probability model to obtain fused data.

[0136] When executed on a data processing device, it is also adapted to execute a program initialized with the following method steps: for each first data point in the fused data, determining a second data point whose distance from the first data point is less than a preset distance threshold, and determining a set of neighborhood data points of the first data point according to the second data point; calculating the reciprocal of the average reachable distance between each second data point in the set of neighborhood data points and the first data point to obtain the local reachability density of the first data point; calculating the ratio between the local reachability density of each second data point in the set of neighborhood data points and the local reachability density of the first data point respectively, and calculating the average value of the ratios to obtain the outlier value corresponding to each first data point; in the case where the outlier value of the first data point is greater than a preset outlier value threshold, determining the first data point as an abnormal data point, and deleting the abnormal data points in the fused data to obtain target parameters.

[0137] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: using a digital filter to remove noise and interference signals from various lightning strike parameters to obtain filtered various lightning strike parameters; checking the integrity and correctness of the filtered various lightning strike parameters through cyclic redundancy check to obtain verified various lightning strike parameters; extracting key features from the verified various lightning strike parameters, where the key features at least include: peak voltage, current waveform, resistance value change trend, temperature and humidity change.

[0138] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: performing normalization processing on the index value of each parameter in the target parameters to obtain the normalized target parameters, where the magnitude of lightning energy is characterized by a preset unit, and the grounding state of the base station equipment is characterized by the integrity of the grounding system of the base station equipment; determining the influence degree of each parameter in the target parameters on the current lightning strike risk assessment task, and determining the weight of each parameter according to the influence degree of each parameter; calculating the normalized target parameters and the weight of each parameter in the target parameters by using the method of weighted summation to obtain the lightning risk index.

[0139] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: constructing a decision tree model based on the parameters with a relatively high degree of correlation with lightning risk among various lightning strike parameters; training the decision tree model with various lightning strike parameters to obtain a trained decision tree model; evaluating the trained decision tree model based on the cross-validation algorithm, and optimizing the trained decision tree model based on the evaluation result based on the reinforcement learning strategy to obtain an optimized decision tree model; calculating various lightning strike parameters collected within a preset time period based on the optimized decision tree model to generate and execute a warning strategy.

[0140] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: transmitting various lightning strike parameters to multiple headers of a message queue; transmitting the various lightning strike parameters stored in the multiple headers to multiple data processing modules according to the target mapping relationship, where each data processing module in the multiple data processing modules is responsible for performing different data processing operations on the various lightning strike parameters; the target mapping relationship refers to the mapping relationship between the multiple headers and the multiple data processing modules; the transmission process of the various lightning strike parameters is transmitted by using the parallel compressive sensing technology.

[0141] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: determining the lightning risk level at the current moment according to the lightning risk index, where the lightning risk level includes at least one of the following: the first level, the second level, and the execution strategies corresponding to the first level and the second level are different; sending the lightning risk level and the execution strategy corresponding to the lightning risk level to the target object, and receiving the response information of the target object in response to the lightning risk level; generating a lightning risk handling report according to the lightning risk index, the lightning risk level, the execution strategy corresponding to the lightning risk level, and the response information.

[0142] 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 storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0143] 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 flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented 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 a device for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 a block or multiple blocks.

[0144] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 a block or multiple blocks. These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 a block or multiple blocks.

[0145] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory. Memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium. Computer-readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. According to the definition in this article, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0146] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0147] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0148] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various modifications and changes can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for assessing lightning risk in a base station, characterized in that: include: Collecting a variety of lightning strike parameters on the ground wire of the base station equipment through a target sensor, and performing feature extraction processing on the multiple lightning strike parameters to obtain a variety of key features, wherein the target sensor is a sensor including a variety of sensor elements; Constructing a correlation relationship between the multiple key features, and fusing the correlation relationship and the multiple lightning strike parameters using a fusion algorithm to obtain fused data; Calculate and delete abnormal data points in the fused data based on a local abnormal factor algorithm to obtain target parameters; The target parameters are calculated based on a risk assessment algorithm to assess the lightning risk at the current moment and obtain a lightning risk index, wherein the risk assessment algorithm includes at least one of the following: a fuzzy logic algorithm, a Bayesian network algorithm, and an absolute method risk assessment algorithm.

2. The method according to claim 1, characterized in that The association relationship and the multiple lightning strike parameters are fused using a fusion algorithm to obtain fused data, including: Determining a priori probability distribution of each sensor in the target sensors based on historical measurement information of the target sensors, adjusting the priori probability distribution of each sensor using the multiple lightning strike parameters, and obtaining a posterior probability model of each sensor; Determine the weight of each sensor through a neural network model based on the association relationship, the historical accuracy of each sensor and / or the correlation between the current lightning risk assessment task and the parameters collected by each sensor; The posterior probability model of each sensor is fused according to the weight of each sensor to obtain the target probability model; The lightning strike parameters corresponding to each sensor in the plurality of lightning strike parameters are fused based on the target probability model to obtain the fused data.

3. The method according to claim 1, characterized in that Calculate and delete abnormal data points in the fused data based on the local abnormal factor algorithm to obtain target parameters, including: For each first data point in the fused data, determine a second data point whose distance to the first data point is less than a preset distance threshold, and determine a neighborhood data point set of the first data point based on the second data point; Calculating the inverse of the average reachable distance between each second data point in the neighborhood data point set and the first data point to obtain a local reachable density of the first data point; Calculate the ratio of the local reachable density of each second data point in the neighborhood data point set to the local reachable density of the first data point, and calculate the average value of the ratio to obtain the outlier corresponding to each first data point; When the outlier value of the first data point is greater than a preset outlier value threshold, the first data point is determined to be an outlier data point, and the outlier data point in the fused data is deleted to obtain the target parameter.

4. The method according to claim 1, characterized in that: The plurality of lightning strike parameters are subjected to feature extraction processing to obtain a plurality of key features, including: Using a digital filter to remove noise and interference signals from the plurality of lightning strike parameters to obtain the plurality of lightning strike parameters after filtering; Checking the integrity and correctness of the filtered multiple lightning strike parameters through cyclic redundancy check to obtain the verified multiple lightning strike parameters; The key features are extracted from the verified multiple lightning parameters, wherein the key features at least include: peak voltage, current waveform, resistance value change trend, and temperature and humidity changes.

5. The method according to claim 1, characterized in that The multiple lightning parameters include at least: ground potential counter-attack overvoltage, lightning frequency, lightning energy, and grounding status of base station equipment; the risk assessment algorithm is the absolute risk assessment algorithm, and the target parameters are calculated based on the risk assessment algorithm to assess the lightning risk at the current moment and obtain a lightning risk index, including: Standardizing the index value of each parameter in the target parameter to obtain a standardized target parameter, wherein the lightning energy is characterized by a preset unit, and the grounding state of the base station equipment is characterized by the integrity of the grounding system of the base station equipment; Determine the influence of each parameter in the target parameters on the current lightning risk assessment task, and determine the weight of each parameter according to the influence of each parameter; The standardized target parameter and the weight of each parameter in the target parameter are calculated by using a weighted summation method to obtain the lightning risk index.

6. The method according to claim 1, characterized in that After collecting various lightning parameters on the grounding wire of the base station equipment through the target sensor, the method further includes: Constructing a decision tree model based on the parameters with a higher degree of correlation with lightning risk among the multiple lightning strike parameters; Using the multiple lightning strike parameters to train the decision tree model to obtain a trained decision tree model; Evaluate the trained decision tree model based on a cross-validation algorithm, and optimize the trained decision tree model based on a reinforcement learning strategy according to the evaluation result to obtain an optimized decision tree model; Based on the optimized decision tree model, various lightning strike parameters collected within a preset time period are calculated to generate and execute an early warning strategy.

7. The method according to claim 1, characterized in that After collecting various lightning parameters on the grounding wire of the base station equipment through the target sensor, the method further includes: Transmitting the plurality of lightning strike parameters to a plurality of titles in a message queue; The multiple lightning strike parameters stored in the multiple titles are transmitted to multiple data processing modules according to the target mapping relationship, wherein each of the multiple data processing modules is responsible for performing different data processing operations on the multiple lightning strike parameters; the target mapping relationship refers to the mapping relationship between the multiple titles and the multiple data processing modules; the transmission process of the multiple lightning strike parameters adopts parallel compressed sensing technology for transmission.

8. The method according to claim 1, characterized in that After calculating the target parameter based on the risk assessment algorithm, assessing the lightning risk at the current moment, and obtaining the lightning risk index, the method further includes: Determining the lightning risk level at the current moment according to the lightning risk index, wherein the lightning risk level includes at least one of the following: a first level and a second level, and the execution strategy corresponding to the first level is different from the execution strategy corresponding to the second level; Sending the lightning risk level and the execution strategy corresponding to the lightning risk level to the target object, and receiving response information of the target object in response to the lightning risk level; A lightning risk handling report is generated according to the lightning risk index, the lightning risk level, the execution strategy corresponding to the lightning risk level and the response information.

9. A base station lightning risk assessment device, characterized in that: include: A collection unit, used to collect multiple lightning parameters on the ground wire of the base station equipment through a target sensor, and perform feature extraction processing on the multiple lightning parameters to obtain multiple key features, wherein the target sensor is a sensor including multiple types of sensor elements; A fusion unit, used for constructing a correlation relationship between the multiple key features, and fusing the correlation relationship and the multiple lightning stroke parameters using a fusion algorithm to obtain fused data; A deletion unit, used for calculating and deleting abnormal data points in the fused data based on a local abnormal factor algorithm to obtain target parameters; An evaluation unit is used to calculate the target parameters based on a risk evaluation algorithm, evaluate the lightning risk at the current moment, and obtain a lightning risk index, wherein the risk evaluation algorithm includes at least one of the following: a fuzzy logic algorithm, a Bayesian network algorithm, and an absolute method risk evaluation algorithm.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the base station lightning risk assessment method described in any one of claims 1 to 8.

Images