Radio equipment fault diagnosis method and system based on cloud platform

Abstract

The invention discloses a radio equipment fault diagnosis method and system based on a cloud platform, relates to the technical field of fault diagnosis, and aims to scientifically and accurately obtain a fault score value of radio equipment in a layer-by-layer progressive mode based on data driving so as to improve the fault diagnosis efficiency. Therefore, the fault state of the radio equipment output based on the confidence degree of the fault diagnosis model is further verified to obtain the fault state of the radio equipment with the fault, so that the accuracy of the output result of the fault diagnosis model can be further confirmed, the risk of misdiagnosis is reduced, and the fault diagnosis efficiency is improved. And the formulation of the final maintenance plan can be better and accurately grasped, that is, the more accurate and efficient maintenance plan can be formulated, so that the unnecessary manpower and material resource expenditure can be reduced to the greatest extent, the time cost can be saved, and the availability and stability of the radio equipment can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a radio equipment fault diagnosis method and system based on a cloud platform. Background Art

[0002] With the development of modern communication technology, radio equipment plays a vital role in the fields of military, civil aviation, navigation, satellite communication, etc. The reliability and stability of these devices directly affect the performance of the entire communication system. Therefore, it is particularly important to ensure that the radio equipment is in good working condition and to detect and eliminate faults in a timely manner. Traditional radio equipment fault diagnosis relies on manual inspection by experienced technicians, which is not only time-consuming and labor-intensive, but also easily affected by human factors.

[0003] In recent years, with the advancement of cloud computing and big data technology, radio equipment fault diagnosis technology based on cloud platforms has developed rapidly. This technology uses the powerful computing power and storage resources of cloud platforms to efficiently process large amounts of monitoring data and achieve real-time monitoring and prediction of radio equipment faults.

[0004] At present, most cloud-based radio equipment fault diagnosis technologies rely on machine learning algorithms, such as support vector machines, decision trees, random forests, neural networks, etc. These algorithms are trained to establish a fault diagnosis model that can output fault status based on the input radio equipment operation data. However, the existing technologies still have some problems in practical application, which are mainly reflected in the following aspects:

[0005] First, the output of the model is often based on the confidence level in the model to achieve accurate output, but the output often has a certain degree of uncertainty. Even if the accuracy of the model is very high, there may be false positives, especially when the confidence level is low. Therefore, there is still a certain degree of misdiagnosis in the output of the model.

[0006] Second, existing diagnostic methods often rely only on a single model output result and lack a further verification mechanism for the model output, which may lead to inaccurate judgment of the fault status;

[0007] 3. When the fault diagnosis results based on the model output are directly used to formulate maintenance plans, the specific cause and severity of the fault are not fully considered, which will lead to the maintenance plan being not detailed enough and thus failing to achieve the optimal maintenance effect.

[0008] Therefore, the prior art urgently needs a technical solution for a radio equipment fault diagnosis method and system based on a cloud platform. Summary of the invention

[0009] In order to solve the above technical problems, the present invention provides a radio equipment fault diagnosis method based on a cloud platform, which specifically includes the following steps:

[0010] Step S1, acquiring real-time operation data through a plurality of sensors arranged on the radio equipment;

[0011] Step S2: uploading the real-time operation data to the cloud platform, and preprocessing the real-time operation data through the cloud platform;

[0012] Step S3, performing a fault diagnosis and verification process based on the preprocessed real-time operation data to obtain a determined fault state of the radio equipment;

[0013] Step S3a, performing fault diagnosis on the real-time operation data to obtain the current fault status of the radio equipment;

[0014] Step S3a1, extracting features from the pre-processed real-time operation data to obtain a feature vector;

[0015] Step S3a2: construct a fault diagnosis model, and use the feature vector to train the fault diagnosis model until the fault diagnosis model converges to obtain a trained fault diagnosis model;

[0016] Step S3a3, reacquiring a set of real-time operation data, and inputting the real-time operation data into the fault diagnosis model to identify the fault state of the current radio equipment;

[0017] Step S3b, performing a verification process on the fault state of the current radio device to obtain a determined fault state of the radio device;

[0018] Step S3b1, calling the confidence score of the fault state of the current radio equipment through the fault diagnosis model;

[0019] Step S3b2, calling the real-time operation data of the fault status of the current radio equipment;

[0020] Step S3b3, obtaining the parameters and parameter data included in the real-time operation data obtained in step S3b2, and analyzing the influence of each parameter on the fault of the current radio equipment to obtain the fault score value of the current radio equipment;

[0021] Step S3b31: based on the data of the parameters included in the real-time operation data, the data volume of each parameter is counted to form a data set of each parameter, the mean of the data set of each parameter is calculated and obtained, and the standard deviation of the data set of each parameter is obtained through the mean;

[0022] The calculation formula for the mean of the data set for each parameter is:

[0023]

[0024] in, represents the mean of the data set x of the i-th parameter; n represents the number of data points in the data set; x ik The kth data point in the dataset x representing the i-th parameter;

[0025] Step S3b32, obtaining the skewness and kurtosis of the data set of each parameter through the mean and standard deviation of the data set of each parameter;

[0026] Among them, the calculation formula for the skewness of the data set for each parameter is:

[0027]

[0028] in, represents the skewness of the data set x of the i-th parameter; n represents the number of data points in the data set; x ik The kth data point in the dataset x representing the i-th parameter; represents the mean of the data set x of the i-th parameter; The sum of the cube of the differences between all data points and the mean in the data set x representing the i-th parameter; represents the standard deviation of the data set x of the i-th parameter; The cube of the standard deviation of the data set x representing the i-th parameter;

[0029] Among them, the calculation formula for the kurtosis of the data set for each parameter is:

[0030]

[0031] in, The kurtosis of the data set x is the i-th parameter; n represents the number of data points in the data set; x ik The kth data point in the dataset x representing the i-th parameter; represents the mean of the data set x of the i-th parameter; The sum of the cube of the differences between all data points and the mean in the data set x representing the i-th parameter; represents the standard deviation of the data set x of the i-th parameter;

[0032] The fourth power of the standard deviation of the data set x representing the i-th parameter;

[0033] Step S3b33: Obtain the maximum value in the data set of each parameter through the data set of each parameter, and obtain the first influence coefficient of each parameter on the fault of the current radio device according to the maximum value in the data set of each parameter, the mean, standard deviation, skewness, and kurtosis of the data set of each parameter;

[0034] The calculation formula for obtaining the first influence coefficient of each parameter on the failure of the current radio equipment is:

[0035]

[0036] Among them, Y i represents the first influence coefficient of the ith parameter on the failure of the current radio equipment; λ represents the shape parameter, which is used to adjust the steepness of the function; x i，max Represents the maximum value in the data set of the i-th parameter; represents the mean of the data set x of the i-th parameter; represents the standard deviation of the data set x of the i-th parameter; represents the skewness of the data set x of the i-th parameter; The kurtosis of the data set x of the i-th parameter; α represents the weight coefficient of the skewness of the data set of each parameter; β represents the weight coefficient of the kurtosis of the data set of each parameter;

[0037] Step S3b34, performing weighted average processing on the first influence coefficient of each parameter on the failure of the current radio device, to obtain the second influence coefficient of all parameters on the failure of the current radio device;

[0038]

[0039] Where D represents the second influence coefficient of all parameters on the fault of the current radio equipment; L represents the number of parameters; δ represents the weight coefficient of the first influence coefficient; Y i represents the first influence coefficient of the i-th parameter on the failure of the current radio equipment;

[0040] Step S3b35, analyzing the influence degree of each parameter combination on the fault of the current radio device, and obtaining a third influence coefficient of each parameter combination on the fault of the current radio device;

[0041] Step S3b351, obtaining a data set of each parameter, combining them one by one, and performing standardization on the combined data sets to obtain a data set of combined parameters on the same measurement dimension;

[0042] Step S3b352, according to step S3b31 and step S3b32, calculate the mean, standard deviation, skewness and kurtosis of each data set of the combined parameters in turn;

[0043] Step S3b353: according to each one-to-one combination parameter data set, obtain the maximum value in each one-to-one combination parameter data set, and according to the maximum value in each one-to-one combination parameter data set, the mean, standard deviation, skewness and kurtosis of each one-to-one combination parameter data set, obtain the third influence coefficient of each one-to-one combination parameter on the fault of the current radio device;

[0044] The calculation formula for obtaining the third influence coefficient of each one-to-one combination parameter on the fault of the current radio equipment is:

[0045]

[0046] Among them, Y {i，j} represents the third influence coefficient of the combination of the i-th parameter and the j-th parameter on the failure of the current radio equipment; λ represents the shape parameter, which is used to adjust the steepness of the function; x {i，j}，max Represents the maximum value in the data set when the i-th parameter is combined with the j-th parameter; Represents the mean of the data set x when the i-th parameter is combined with the j-th parameter; Represents the standard deviation of the data set x when the i-th parameter is combined with the j-th parameter; Represents the skewness of the data set x when the i-th parameter is combined with the j-th parameter; The kurtosis of the data set x when the i-th parameter is combined with the j-th parameter; ε represents the weight coefficient of the skewness of the data set of the one-to-one combination parameters; τ represents the weight coefficient of the kurtosis of the data set of the one-to-one combination parameters;

[0047] Step S3b36, performing weighted average processing on the third influence coefficient of each one-to-one combination parameter on the fault of the current radio device, to obtain the fourth influence coefficient of all one-to-one combination parameters on the fault of the current radio device;

[0048] The calculation formula for obtaining the fourth influence coefficient of all the combined parameters on the fault of the current radio equipment is:

[0049]

[0050] Wherein, T represents the fourth influence coefficient of all one-to-one combination parameters on the fault of the current radio equipment; Q represents the number of one-to-one combination parameters; θ represents the weight coefficient of the third influence coefficient; Y {i，j} represents the third influence coefficient of the combination of the ith parameter and the jth parameter on the failure of the current radio device;

[0051] Step S3b37, obtaining a fault score value of the current radio device according to the second influence coefficients of all parameters on the fault of the current radio device and the fourth influence coefficients of all one-to-one combination parameters on the fault of the current radio device;

[0052] The calculation formula for obtaining the fault score value of the current radio equipment is:

[0053] FSV=φ·D+ω·T;

[0054] Wherein, FSV represents the fault score value of the current radio equipment; D represents the second influence coefficient of all parameters on the fault of the current radio equipment; T represents the fourth influence coefficient of all one-to-one combination parameters on the fault of the current radio equipment; φ represents the weight coefficient of the second influence coefficient; T represents the weight coefficient of the fourth influence coefficient;

[0055] Step S3b4, performing standardization processing on the fault score value, so that the fault score value and the confidence score are in the same measurement dimension, and obtaining the standard fault score value of the current radio equipment. If the confidence score of the fault state of the current radio equipment is greater than or equal to the standard fault score value of the current radio equipment, it means that the fault state of the current radio equipment is determined to have a fault, that is, it is the determined fault state of the radio equipment;

[0056] If the confidence score of the fault state of the current radio device is less than the standard fault score of the current radio device, it means that the fault state of the current radio device is determined to be non-existent, that is, it is a negative fault state of the radio device. At this time, step S3a3 is re-executed until the fault state of the determined radio device is obtained;

[0057] Step S4: formulate and execute a maintenance plan based on the determined fault status of the radio equipment.

[0058] A radio equipment fault diagnosis system based on a cloud platform, using a radio equipment fault diagnosis method based on a cloud platform as described in any one of the above items, comprising the following modules:

[0059] Data acquisition module: used to obtain real-time operation data through multiple sensors set on the radio equipment;

[0060] Preprocessing module: connected to the data acquisition module, used to upload the real-time operation data to the cloud platform and preprocess the real-time operation data through the cloud platform;

[0061] Diagnosis and verification module: connected to the preprocessing module, used to perform fault diagnosis and verification process on the real-time operation data after preprocessing to obtain the determined fault status of the radio equipment;

[0062] Plan formulation and execution module: connected to the diagnosis and verification module, used to formulate and execute a maintenance plan based on the determined fault status of the radio equipment.

[0063] The embodiments of the present invention have the following technical effects:

[0064] The present invention aims to obtain the fault score value of the radio equipment scientifically and accurately in a data-driven and progressive manner, so as to further verify the fault state of the radio equipment based on the confidence output of the fault diagnosis model, so as to obtain the fault state of the radio equipment with a confirmed fault. This can not only further confirm the accuracy of the output result of the fault diagnosis model and reduce the risk of misdiagnosis, but also better grasp the formulation of the final maintenance plan, that is, it can formulate a more accurate and efficient maintenance plan, thereby reducing unnecessary human and material expenditures to the greatest extent, saving time costs, and improving the availability and stability of the radio equipment;

[0065] The present invention adopts a data-driven and progressive approach in the process of obtaining a scientific and accurate fault score value for radio equipment. First, when calculating the first influence coefficient of each parameter on the fault of the current radio equipment, by using the maximum value, mean, standard deviation, skewness and kurtosis, the influence of each parameter on the radio equipment fault can be comprehensively evaluated, thereby providing a more comprehensive information basis for subsequent fault scores, helping to improve the accuracy of the fault score value and reduce the possibility of misdiagnosis. Then, by performing weighted averaging processing on the first influence coefficient, the influence of all parameters on the fault can be comprehensively considered. The obtained second influence coefficient can provide a more accurate fault score value because it integrates the influence of all parameters, which is helpful for further Improve the accuracy of fault diagnosis, and then by using the maximum value, mean, standard deviation, skewness and kurtosis in the data set of each one-to-one combination of parameters, the characteristics of the parameter combination can be fully considered, so as to better evaluate the influence of the parameter combination on the radio equipment fault, so that the obtained third influence coefficient helps to further improve the accuracy of the fault score value and reduce the possibility of misdiagnosis. Finally, by weighted averaging the third influence coefficient, the influence of all parameter combinations on the fault can be comprehensively considered, so as to better reflect the influence of different parameter combinations on the fault. The fourth influence coefficient obtained can provide a more accurate and reliable fault score value because it integrates the influence of all parameter combinations, which helps to further improve the accuracy of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0067] Figure 1 is a flow chart of a radio equipment fault diagnosis method based on a cloud platform provided by an embodiment of the present invention;

[0068] Figure 2 It is a framework diagram of a cloud platform-based radio equipment fault diagnosis system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0070] Embodiment 1: Figure 1 As shown, the present invention provides a radio equipment fault diagnosis method based on a cloud platform, comprising the following steps:

[0071] Step S1, acquiring real-time operation data through a plurality of sensors arranged on the radio equipment;

[0072] It is worth noting that the sensors installed on the radio equipment include but are not limited to: temperature sensor: used to monitor the temperature changes inside the radio equipment to help identify overheating; voltage sensor: used to measure the voltage level of each component inside the radio equipment to detect voltage fluctuations or anomalies; current sensor: used to measure the current consumption of each component inside the radio equipment to help identify power consumption anomalies; vibration sensor: used to detect the vibration of the radio equipment to help identify mechanical failures or loose parts; signal strength sensor: used to monitor the strength of the signal transmitted and received by the radio equipment to help identify the situation of signal quality degradation; humidity sensor: used to monitor the humidity of the environment where the radio equipment is located to prevent damage to the equipment due to excessive humidity;

[0073] The real-time operating data collected above include but are not limited to: temperature data, voltage data, current data, vibration data, signal strength data, and humidity data; regarding the frequency setting of the above sensors for data collection: for key parameters such as temperature, voltage, and current, set a higher collection frequency, such as collecting data once per second to ensure that rapidly changing situations can be captured in a timely manner; for parameters that change slowly, such as signal strength and humidity, set a lower collection frequency, such as collecting data once every five minutes, to reduce the pressure of data processing.

[0074] Step S2: uploading the real-time operation data to the cloud platform, and preprocessing the real-time operation data through the cloud platform;

[0075] It is worth mentioning that regarding the method of uploading real-time operation data, the current real-time operation data is uploaded to the cloud platform using the RESTful API or MQTT protocol. At the same time, to ensure data security, HTTPS encrypted transmission is used. Then the cloud platform uses the message queue to temporarily store the uploaded data to ensure reliable data transmission. At the same time, the data is stored in a distributed database for efficient retrieval and processing. Subsequently, preprocessing operations are performed on the temporarily stored real-time operation data. The specific process includes but is not limited to: data cleaning stage, through outlier detection and missing value processing to ensure data quality; data conversion stage, unified data units and formats for subsequent analysis.

[0076] Step S3, performing a fault diagnosis and verification process based on the preprocessed real-time operation data to obtain a determined fault state of the radio equipment;

[0077] Step S3a, performing a fault diagnosis process on the real-time operation data to obtain the current fault status of the radio equipment;

[0078] Step S3a1, extracting features from the pre-processed real-time operation data to obtain a feature vector;

[0079] It is worth noting that in the feature extraction stage, the above-mentioned temperature data, voltage data, current data, vibration data, signal strength data, and humidity data are still taken as examples; Fourier transform is used to extract the time series features of temperature data, such as frequency components; wavelet transform is used to extract the time series features of voltage data, such as wavelet coefficients; the first-order derivative and second-order derivative of current data are calculated as dynamic change features; empirical mode decomposition (EMD) is used to extract the fundamental mode functions (IMFs) of vibration data, and the first few IMFs are selected as features; the autocorrelation function is used to analyze the periodicity of signal strength data and extract the autocorrelation peak position; the differential sequence of humidity data is calculated as a change trend feature;

[0080] Exemplarily, the different types of features extracted above are combined into a feature vector, such as: the temperature feature vector contains the amplitudes of the first N main frequency components; the voltage feature vector may contain the first N wavelet coefficients; the current feature vector may contain the maximum value, minimum value and maximum value of the first-order derivative and the second-order derivative; the vibration feature vector may contain the average energy of the first M IMFs; the signal strength feature vector may contain the first K peak positions of the autocorrelation function; the humidity feature vector may contain the trend coefficient of humidity change.

[0081] Step S3a2: construct a fault diagnosis model, and use the feature vector to train the fault diagnosis model until the fault diagnosis model converges to obtain a trained fault diagnosis model;

[0082] A support vector machine is used as a fault diagnosis model, and then the feature vector extracted in step S3a1 is used and divided into a training set and a test set in a ratio of 80%:20%. 80% of the data is used as training data, and known fault data is used as labels for supervised learning. The best hyperparameter combination is then selected through cross-validation. At the same time, a grid search or random search method is used for hyperparameter tuning, and convergence conditions are set, such as the number of iterations or the rate of change of the loss function, until the model converges, that is, the set convergence conditions are reached. At the same time, in order to ensure the accuracy of model training, real-time operating data after preprocessing by the cloud platform should be repeatedly obtained during this process, and after feature extraction, it should be used as part of model training to ensure sufficient data support so that the model can accurately output results.

[0083] Step S3a3, reacquiring a set of real-time operation data, and inputting the real-time operation data into the fault diagnosis model to identify the fault state of the current radio equipment;

[0084] Acquire new real-time operation data from the sensor on the radio equipment, perform the same preprocessing operation as step S2 on the newly acquired real-time operation data, perform feature extraction on the preprocessed real-time operation data to obtain a new feature vector, and input the new feature vector into the trained fault diagnosis model. The model will output the fault status of the current radio equipment based on the confidence level.

[0085] Step S3b, performing a verification process on the fault state of the current radio device to obtain a determined fault state of the radio device;

[0086] It is worth noting that, since the fault status of the radio equipment in the above-mentioned example is based on the confidence output of the fault diagnosis model, there is uncertainty in the output result, especially when the confidence level is low. Therefore, by executing the verification process, the accuracy of the fault status of the radio equipment can be further confirmed to ensure the reliability of the fault diagnosis. It is worth noting that even if the accuracy of the fault diagnosis model is very high, false alarms may occur, especially when the confidence level is low. Therefore, through the verification process, the output results of the model can be further confirmed to reduce the risk of misdiagnosis. This is for the formulation of the final maintenance plan, such as the maintenance and repair of radio equipment. A higher confidence level is required to support the formulation of the plan, so as to reduce unnecessary manpower and material expenditures and save time costs.

[0087] Step S3b1, calling the confidence score of the fault state of the current radio equipment through the fault diagnosis model;

[0088] Step S3b2, calling the real-time operation data of the fault status of the current radio equipment;

[0089] Step S3b3, obtaining the parameters and parameter data included in the real-time operation data obtained in step S3b2, and analyzing the influence of each parameter on the fault of the current radio equipment to obtain the fault score value of the current radio equipment;

[0090] Step S3b31: based on the data of the parameters included in the real-time operation data, the data volume of each parameter is counted to form a data set of each parameter, the mean of the data set of each parameter is calculated and obtained, and the standard deviation of the data set of each parameter is obtained through the mean;

[0091] The calculation formula for the mean of the data set for each parameter is:

[0092]

[0093] in, represents the mean of the data set x of the i-th parameter; n represents the maximum number of data points in the data set; x ik The kth data point in the dataset x representing the i-th parameter;

[0094] Step S3b32, obtaining the skewness and kurtosis of the data set of each parameter through the mean and standard deviation of the data set of each parameter;

[0095] Among them, the calculation formula for the skewness of the data set for each parameter is:

[0096]

[0097] in, represents the skewness of the data set x of the i-th parameter; The sum of the cube of the differences between all data points and the mean in the data set x representing the i-th parameter;

[0098] represents the standard deviation of the data set x of the i-th parameter;

[0099] The cube of the standard deviation of the data set x representing the i-th parameter;

[0100] Among them, the calculation formula for the kurtosis of the data set for each parameter is:

[0101]

[0102] in, The kurtosis of the data set x of the i-th parameter; The sum of the cube of the differences between all data points and the mean in the data set x representing the i-th parameter; The fourth power of the standard deviation of the data set x representing the i-th parameter;

[0103] Step S3b33: Obtain the maximum value in the data set of each parameter through the data set of each parameter, and obtain the first influence coefficient of each parameter on the fault of the current radio device according to the maximum value in the data set of each parameter, the mean, standard deviation, skewness, and kurtosis of the data set of each parameter;

[0104] The calculation formula for obtaining the first influence coefficient of each parameter on the failure of the current radio equipment is:

[0105]

[0106] Among them, Y i represents the first influence coefficient of the ith parameter on the failure of the current radio equipment; λ represents the shape parameter, which is used to adjust the steepness of the function; x i，max Represents the maximum value of the parameter data in the data set of the i-th parameter; α represents the weight coefficient of the skewness of the data set of each parameter; β represents the weight coefficient of the kurtosis of the data set of each parameter;

[0107] in, Equivalent to Both represent the standard deviation of the data set x of the i-th parameter;

[0108] Step S3b34, performing weighted average processing on the first influence coefficient to obtain a second influence coefficient of all parameters on the fault of the current radio device;

[0109]

[0110] Where D represents the second influence coefficient of all parameters on the fault of the current radio equipment; L represents the number of parameters; δ represents the weight coefficient of the first influence coefficient; Y i represents the first influence coefficient of the i-th parameter on the failure of the current radio equipment;

[0111] Step S3b35, analyzing the influence degree of each parameter combination on the fault of the current radio device, and obtaining a third influence coefficient of each parameter combination on the fault of the current radio device;

[0112] Step S3b351, obtaining a data set of each parameter, combining them one by one, and performing standardization on the combined data sets to obtain a data set of combined parameters on the same measurement dimension;

[0113] Step S3b352, according to step S3b31 and step S3b32, calculate the mean, standard deviation, skewness and kurtosis of each data set of the combined parameters in turn;

[0114] Step S3b353, according to each one-to-one combination parameter data set, obtain the maximum value in each one-to-one combination parameter data set, and according to the maximum value in each one-to-one combination parameter data set, the mean, standard deviation, skewness and kurtosis of each one-to-one combination parameter data set, obtain the third influence coefficient of each one-to-one combination parameter on the fault of the current radio equipment.

[0115] The calculation formula for obtaining the third influence coefficient of each one-to-one combination parameter on the fault of the current radio equipment is:

[0116]

[0117] Among them, Y {i，j} represents the third influence coefficient of the combination of the i-th parameter and the j-th parameter on the failure of the current radio equipment; x {i，j}，max Represents the maximum value in the data set when the i-th parameter is combined with the j-th parameter; Represents the mean of the data set x when the i-th parameter is combined with the j-th parameter; Represents the standard deviation of the data set x when the i-th parameter is combined with the j-th parameter; Represents the skewness of the data set x when the i-th parameter is combined with the j-th parameter; The kurtosis of the data set x when the i-th parameter is combined with the j-th parameter; ε represents the weight coefficient of the skewness of the data set of the one-to-one combination parameters; τ represents the weight coefficient of the kurtosis of the data set of the one-to-one combination parameters;

[0118] Step S3b36, performing weighted average processing on the third influence coefficient to obtain a fourth influence coefficient of all the combined parameters on the fault of the current radio device;

[0119] Among them, the calculation formula of the fourth influence coefficient is:

[0120]

[0121] Wherein, T represents the fourth influence coefficient of all one-to-one combination parameters on the fault of the current radio equipment; Q represents the maximum number of one-to-one combination parameters; θ represents the weight coefficient of the third influence coefficient; Y {i，j} represents the third influence coefficient of the combination of the ith parameter and the jth parameter on the failure of the current radio device;

[0122] Step S3b37, obtaining a fault score value of the current radio equipment according to the second influence coefficient and the fourth influence coefficient;

[0123] The calculation formula for obtaining the fault score value of the current radio equipment is:

[0124] FSV=φ·D+ω·T;

[0125] Among them, FSV represents the fault score value of the current radio equipment; D represents the second influence coefficient of all parameters on the fault of the current radio equipment; T represents the fourth influence coefficient of all one-to-one combination parameters on the fault of the current radio equipment; φ represents the weight coefficient of the second influence coefficient.

[0126] Step S3b4, performing standardization processing on the fault score value, so that the fault score value and the confidence score are in the same measurement dimension, and obtaining the standard fault score value of the current radio equipment. If the confidence score of the fault state of the current radio equipment is greater than or equal to the standard fault score value of the current radio equipment, it means that the fault state of the current radio equipment is determined to have a fault, that is, it is the determined fault state of the radio equipment;

[0127] If the confidence score of the fault state of the current radio equipment is less than the standard fault score value of the current radio equipment, it means that the fault state of the current radio equipment is determined to be non-existent, that is, it is a negative fault state of the radio equipment. At this time, step S3a3 is re-executed until the fault state of the radio equipment is determined.

[0128] It is worth noting that the specific process of normalizing the fault score value so that the fault score value and the confidence score are in the same measurement dimension includes:

[0129] 1. Determine the scope of standardization:

[0130] First, the confidence range of the model output is between 0 and 1. Then, according to this standardization, the minimum fault score value is determined to be 0 and the maximum fault score value is 100.

[0131] 2. Apply the standardized formula:

[0132]

[0133] Among them, FSV normalized FSV stands for Failure Score Value; FSV stands for Failure Score Value; min Represents the minimum value of the fault score value; FSV max Represents the maximum value of the fault score;

[0134] The standard fault score value calculated by the above formula is between 0 and 1, which ensures the comparability between the fault score value and the confidence score, facilitates the subsequent determination of the fault status, and through standardization, can better reflect the true meaning of the fault score value and improve the accuracy of fault diagnosis.

[0135] Step S4: formulate and execute a maintenance plan based on the determined fault status of the radio equipment.

[0136] Embodiment 2: Figure 2 As shown, the present invention also proposes a radio equipment fault diagnosis system based on a cloud platform, which executes the above-mentioned radio equipment fault diagnosis method based on a cloud platform, including the following modules:

[0137] Data acquisition module: used to obtain real-time operation data through multiple sensors set on the radio equipment;

[0138] Preprocessing module: connected to the data acquisition module, used to upload the real-time operation data to the cloud platform and preprocess the real-time operation data through the cloud platform;

[0139] Diagnosis and verification module: connected to the preprocessing module, used to perform fault diagnosis and verification process on the real-time operation data after preprocessing to obtain the determined fault status of the radio equipment;

[0140] Plan formulation and execution module: connected to the diagnosis and verification module, used to formulate and execute a maintenance plan based on the determined fault status of the radio equipment.

[0141] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0142] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A radio equipment fault diagnosis method based on a cloud platform, characterized in that: The following steps are involved: Step S1, acquiring real-time operation data through a plurality of sensors arranged on the radio equipment; Step S2: uploading the real-time operation data to the cloud platform, and preprocessing the real-time operation data through the cloud platform; Step S3, performing a fault diagnosis and verification process based on the preprocessed real-time operation data to obtain a determined fault state of the radio equipment; Step S4: formulate and execute a maintenance plan based on the determined fault status of the radio equipment.

2. The radio equipment fault diagnosis method based on cloud platform according to claim 1, characterized in that: The step S3, performing a fault diagnosis and verification process based on the pre-processed real-time operation data to obtain a determined fault state of the radio equipment, includes: Step S3a, performing fault diagnosis on the real-time operation data to obtain the current fault status of the radio equipment; Step S3b: Perform a verification process on the fault state of the current radio device to obtain a determined fault state of the radio device.

3. The radio equipment fault diagnosis method based on cloud platform according to claim 2 is characterized in that: Step S3a, performing fault diagnosis on the real-time operation data to obtain the current fault status of the radio equipment, including: Step S3a1, extracting features from the pre-processed real-time operation data to obtain a feature vector; Step S3a2: construct a fault diagnosis model, and use the feature vector to train the fault diagnosis model until the fault diagnosis model converges to obtain a trained fault diagnosis model; Step S3a3: reacquire a set of real-time operation data, and input the real-time operation data into the fault diagnosis model to identify the fault status of the current radio equipment.

4. The radio equipment fault diagnosis method based on cloud platform according to claim 3 is characterized in that: Step S3b, performing a verification process on the fault state of the current radio device to obtain a determined fault state of the radio device, including: Step S3b1, calling the confidence score of the fault state of the current radio equipment through the fault diagnosis model; Step S3b2, calling the real-time operation data of the fault status of the current radio equipment; Step S3b3, obtaining the parameters and parameter data included in the real-time operation data obtained in step S3b2, and analyzing the influence of each parameter on the fault of the current radio equipment to obtain the fault score value of the current radio equipment; Step S3b4, performing standardization processing on the fault score value, so that the fault score value and the confidence score are in the same measurement dimension, and obtaining the standard fault score value of the current radio equipment. If the confidence score of the fault state of the current radio equipment is greater than or equal to the standard fault score value of the current radio equipment, it means that the fault state of the current radio equipment is determined to have a fault, that is, it is the determined fault state of the radio equipment; If the confidence score of the fault state of the current radio equipment is less than the standard fault score value of the current radio equipment, it means that the fault state of the current radio equipment is determined to be non-existent, that is, it is a negative fault state of the radio equipment. At this time, step S3a3 is re-executed until the fault state of the radio equipment is determined.

5. The radio equipment fault diagnosis method based on cloud platform according to claim 4 is characterized in that: Step S3b3, obtaining the parameters and parameter data included in the real-time operation data through the real-time operation data obtained in step S3b2, and analyzing the influence of each parameter on the fault of the current radio equipment to obtain the fault score value of the current radio equipment, including: Step S3b31: based on the data of the parameters included in the real-time operation data, the data volume of each parameter is counted to form a data set of each parameter, the mean of the data set of each parameter is calculated and obtained, and the standard deviation of the data set of each parameter is obtained through the mean; The calculation formula for the mean of the data set for each parameter is: in, represents the mean of the data set x of the i-th parameter; n represents the number of data points in the data set; x ik The kth data point in the dataset x representing the i-th parameter; Step S3b32, obtaining the skewness and kurtosis of the data set of each parameter through the mean and standard deviation of the data set of each parameter; Among them, the calculation formula for the skewness of the data set for each parameter is: in, represents the skewness of the data set x of the i-th parameter; n represents the number of data points in the data set; x ik The kth data point in the dataset x representing the i-th parameter; represents the mean of the data set x of the i-th parameter; The sum of the cube of the differences between all data points and the mean in the data set x representing the i-th parameter; represents the standard deviation of the data set x of the i-th parameter; The cube of the standard deviation of the data set x representing the i-th parameter; Among them, the calculation formula for the kurtosis of the data set for each parameter is: in, represents the kurtosis of the data set x of the i-th parameter; n represents the number of data points in the data set; x ik The kth data point in the dataset x representing the i-th parameter; represents the mean of the data set x of the i-th parameter; The sum of the cube of the differences between all data points and the mean in the data set x representing the i-th parameter; represents the standard deviation of the data set x of the i-th parameter; The fourth power of the standard deviation of the data set x representing the i-th parameter; Step S3b33: Obtain the maximum value in the data set of each parameter through the data set of each parameter, and obtain the first influence coefficient of each parameter on the fault of the current radio device according to the maximum value in the data set of each parameter, the mean, standard deviation, skewness, and kurtosis of the data set of each parameter; Step S3b34, performing weighted average processing on the first influence coefficient to obtain a second influence coefficient of all parameters on the fault of the current radio device; Where D represents the second influence coefficient of all parameters on the fault of the current radio equipment; L represents the number of parameters; δ represents the weight coefficient of the first influence coefficient; Y i represents the first influence coefficient of the i-th parameter on the failure of the current radio equipment; Step S3b35, analyzing the influence degree of each parameter combination on the fault of the current radio device, and obtaining a third influence coefficient of each parameter combination on the fault of the current radio device; Step S3b36, performing weighted average processing on the third influence coefficient of each one-to-one combination parameter on the fault of the current radio device, to obtain the fourth influence coefficient of all one-to-one combination parameters on the fault of the current radio device; The calculation formula for obtaining the fourth influence coefficient of all the combined parameters on the fault of the current radio equipment is: Wherein, T represents the fourth influence coefficient of all one-to-one combination parameters on the fault of the current radio equipment; Q represents the number of one-to-one combination parameters; θ represents the weight coefficient of the third influence coefficient; Y {i，j} represents the third influence coefficient of the combination of the ith parameter and the jth parameter on the failure of the current radio device; Step S3b37, obtaining a fault score value of the current radio device according to the second influence coefficients of all parameters on the fault of the current radio device and the fourth influence coefficients of all one-to-one combination parameters on the fault of the current radio device; The calculation formula for obtaining the fault score value of the current radio equipment is: Wherein, FSV represents the fault score value of the current radio equipment; D represents the second influence coefficient of all parameters on the fault of the current radio equipment; T represents the fourth influence coefficient of all one-to-one combination parameters on the fault of the current radio equipment; represents the weight coefficient of the second influence coefficient; ω represents the weight coefficient of the fourth influence coefficient.

6. The radio equipment fault diagnosis method based on cloud platform according to claim 5, characterized in that: Step S3b35, analyzing the influence of each parameter combination on the fault of the current radio device, and obtaining a third influence coefficient of each parameter combination on the fault of the current radio device, including: Step S3b351, obtaining a data set of each parameter, combining them one by one, and performing standardization on the combined data sets to obtain a data set of combined parameters on the same measurement dimension; Step S3b352, according to step S3b31 and step S3b32, calculate the mean, standard deviation, skewness and kurtosis of each data set of the combined parameters one by one in sequence; Step S3b353, according to each one-to-one combination parameter data set, obtain the maximum value in each one-to-one combination parameter data set, and according to the maximum value in each one-to-one combination parameter data set, the mean, standard deviation, skewness and kurtosis of each one-to-one combination parameter data set, obtain the third influence coefficient of each one-to-one combination parameter on the fault of the current radio equipment.

7. The radio equipment fault diagnosis method based on cloud platform according to claim 5, characterized in that: in, The calculation formula for obtaining the first influence coefficient of each parameter on the current radio equipment failure is: Among them, Y i represents the first influence coefficient of the ith parameter on the failure of the current radio equipment; λ represents the shape parameter, which is used to adjust the steepness of the function; x i，max Represents the maximum value in the data set of the i-th parameter; represents the mean of the data set x of the i-th parameter; represents the standard deviation of the data set x of the i-th parameter; represents the skewness of the data set x of the i-th parameter; The kurtosis of the data set x of the i-th parameter; α represents the weight coefficient of the skewness of the data set of each parameter; β represents the weight coefficient of the kurtosis of the data set of each parameter.

8. The radio equipment fault diagnosis method based on cloud platform according to claim 6, characterized in that: in, The calculation formula for obtaining the third influence coefficient of each one-to-one combination parameter on the fault of the current radio equipment is: Among them, Y {i，j} represents the third influence coefficient of the combination of the i-th parameter and the j-th parameter on the failure of the current radio equipment; λ represents the shape parameter, which is used to adjust the steepness of the function; x {i，j}，max Represents the maximum value in the data set when the i-th parameter is combined with the j-th parameter; σ x{i，j} Represents the mean of the data set x when the i-th parameter is combined with the j-th parameter; σ x{i，j} Represents the standard deviation of the data set x when the i-th parameter is combined with the j-th parameter; γ x{i，j} represents the skewness of the data set x when the i-th parameter is combined with the j-th parameter; k x{i，j} The kurtosis of the data set x when the i-th parameter is combined with the j-th parameter; ε represents the weight coefficient of the skewness of the data set of the one-to-one combination parameters; τ represents the weight coefficient of the kurtosis of the data set of the one-to-one combination parameters.

9. A radio equipment fault diagnosis system based on a cloud platform, used to execute the radio equipment fault diagnosis method based on a cloud platform according to any one of claims 1 to 8, characterized in that: The system comprises: Data acquisition module: used to obtain real-time operation data through multiple sensors set on the radio equipment; Preprocessing module: connected to the data acquisition module, used to upload the real-time operation data to the cloud platform and preprocess the real-time operation data through the cloud platform; Diagnosis and verification module: connected to the preprocessing module, used to perform fault diagnosis and verification process on the real-time operation data after preprocessing to obtain the determined fault status of the radio equipment; Plan formulation and execution module: connected to the diagnosis and verification module, used to formulate and execute a maintenance plan based on the determined fault status of the radio equipment.

Images