CN112492567B - Fault analysis and solution method and device in emergency command communication - Google Patents
Fault analysis and solution method and device in emergency command communication Download PDFInfo
- Publication number
- CN112492567B CN112492567B CN202011304841.3A CN202011304841A CN112492567B CN 112492567 B CN112492567 B CN 112492567B CN 202011304841 A CN202011304841 A CN 202011304841A CN 112492567 B CN112492567 B CN 112492567B
- Authority
- CN
- China
- Prior art keywords
- data
- solution
- abnormal
- fault
- analysis model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/90—Services for handling of emergency or hazardous situations, e.g. earthquake and tsunami warning systems [ETWS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/04—Arrangements for maintaining operational condition
Abstract
The application discloses a method and a device for analyzing and solving faults in emergency command communication. Collecting a large amount of abnormal operation historical data stored in each communication base station, communication node and terminal equipment, learning according to a large amount of operation historical log data, and constructing a fault analysis model capable of automatically analyzing and predicting the fault reasons of the newly added communication base station, communication node and terminal equipment; collecting operation data of each newly added communication base station, communication nodes and terminal equipment, and inputting a fault analysis model for analysis to obtain a solution; and starting a corresponding emergency response mechanism according to the analyzed solution, and performing exception alarm processing. By adopting the fault analysis and solution method, the abnormal equipment can be found more quickly and accurately during emergency command communication, an accurate solution is provided for abnormal conditions, and the time for abnormal recovery is shortened.
Description
Technical Field
The application relates to the field of emergency communication, in particular to a method and a device for analyzing and solving faults in emergency command communication.
Background
The internet of things, as a typical representative of a new generation of information communication technology, has become a key basis and an important engine for the core drive of a new scientific and technological revolution and industrial revolution around the world and the green, intelligent and sustainable development of the economic society. Currently, the internet of things accelerates penetration and multidimensional fusion to each link of upstream and downstream of the traditional industry in the form of a ubiquitous sensing, lean control, data decision and other capability element sets, promotes industry upgrading and structure optimization, and promotes emerging industry states to emerge continuously. The method aims at realizing comprehensive integrated application of the technology of the Internet of things by taking automatic perception as a basis, taking data acquisition as a means and intelligent control as a core and carrying out fine management and service promotion, and the global Internet of things is wholly entering a new stage of substantial promotion and large-scale development.
In many fields of application of the internet of things, most of deployment implementation main bodies of the fields are government and public service departments and have high safety level requirements in multiple fields of national defense construction, emergency command and communication guarantee, national security, environmental data acquisition, forest fire prevention, geological disaster detection and the like. Therefore, the technology of the special communication internet of things is applied to intelligent information management systems such as emergency command communication systems and data link systems, and the development trend of the current technology is to provide services such as special equipment management, resource allocation, information perception and transmission, emergency communication, real-time decision, early warning and emergency management schemes.
Due to environmental reasons, abnormal operation conditions easily occur in emergency communication, and the emergency rescue effect is seriously influenced due to too low manual monitoring efficiency of the conditions.
Disclosure of Invention
The application provides a fault analysis and solution method in emergency command communication, which comprises the following steps:
collecting a large amount of abnormal operation historical data stored in each communication base station, communication node and terminal equipment, learning according to a large amount of operation historical log data, and constructing a fault analysis model capable of automatically analyzing and predicting the fault reasons of the newly added communication base station, communication node and terminal equipment;
collecting operation data of each communication base station, each communication node and each terminal device, and inputting the operation data into a fault analysis model for analysis to obtain a solution;
and starting a corresponding emergency response mechanism according to the analyzed solution, and performing exception alarm processing.
The method for analyzing and solving the fault in the emergency command communication comprises the following steps of:
carrying out data preprocessing on a large amount of abnormal operation log data;
performing characteristic weight calculation on the log data to obtain a text vector set;
randomly dividing a text vector set into a training set, a verification set and a test set, pairing the training sets in pairs, inputting the training sets into a fault analysis model for training, adjusting the trained fault analysis model by using the verification set, and testing by using the test set and the trained fault analysis model to obtain a finally trained fault analysis model;
and determining a corresponding solution according to the analysis result of the fault analysis model.
The method for analyzing and solving the fault in the emergency command communication comprises the following steps that after the cloud platform acquires the operation data, the cloud platform performs the following sub-steps:
s1, acquiring a data identifier in the running data, judging whether the running data is normal running data or abnormal running data according to the data identifier, if the running data is abnormal running data, executing a step S2, and if the running data is normal running data, continuing to wait for the running data uploaded by each device;
s2, inputting the data identification into the fault analysis model, judging whether a solution can be obtained from the fault analysis model according to the data identification, if so, starting a corresponding emergency response mechanism according to the solution obtained by analysis, otherwise, executing the step S3;
and S3, sending an exception alarm to the staff member, and requesting the corresponding exception equipment to upload the exception handling solution after subsequent exception handling.
The method for analyzing and solving the fault in the emergency command communication comprises the steps that the operation abnormity historical data comprises data identification, data length, equipment identification, equipment position and solution; the operation data of each newly added communication base station, communication node and terminal equipment comprises data identification, data length, equipment identification and equipment position.
The method for analyzing and solving the fault in the emergency command communication, wherein the corresponding emergency response mechanism is started to perform the abnormal alarm processing, specifically comprises: if the solution is solved by the on-duty personnel, sending an abnormal alarm to the on-duty personnel mobile terminal; if the solution is solved by the professional, sending an abnormal alarm to a professional mobile terminal; if the solution is self-recovery solution of the program, the cloud platform sends a program solution code to the corresponding abnormal device, and the abnormal device runs the solution code to achieve the solution of the abnormal solution; wherein, the abnormal alarm information comprises an equipment identifier and an equipment position.
The application provides an unusual monitoring devices of operation in emergency command communication, includes: the system comprises a fault analysis model construction module, an operation data acquisition module and an abnormal alarm processing module;
the fault analysis model building module is used for collecting a large amount of abnormal operation historical data stored in each communication base station, communication node and terminal equipment, learning according to a large amount of operation historical log data, and building a fault analysis model capable of automatically analyzing and predicting the fault reasons of the newly added communication base station, communication node and terminal equipment;
the operation data acquisition module is used for acquiring operation data of each communication base station, each communication node and each terminal device, and inputting the operation data into the fault analysis model for analysis to obtain a solution;
and the abnormity alarm processing module is used for starting a corresponding emergency response mechanism according to the analyzed solution and processing abnormity alarm.
The operation abnormity monitoring device in emergency command communication comprises a fault analysis model building module, a fault analysis module and a fault analysis module, wherein the fault analysis model building module is specifically used for preprocessing a large amount of abnormal operation data; performing characteristic weight calculation on the log data to obtain a text vector set; randomly dividing a text vector set into a training set, a verification set and a test set, pairing the training sets in pairs, inputting the training sets into a fault analysis model for training, adjusting the trained fault analysis model by using the verification set, and testing by using the test set and the trained fault analysis model to obtain a finally trained fault analysis model; and determining a corresponding solution according to the analysis result of the fault analysis model.
The device for monitoring the operation abnormity in the emergency command communication comprises an operation data acquisition module, a solution query module and an abnormity processing module, wherein the operation data acquisition module specifically comprises an operation data type identification submodule, a solution query submodule and an abnormity processing submodule;
the operation data type identification submodule acquires a data identifier in the operation data, judges whether the operation data is normal operation data or abnormal operation data according to the data identifier, if the operation data is judged to be abnormal operation data, triggers the solution query submodule, and otherwise, continuously waits for the operation data uploaded by each device;
the solution query submodule inputs the data identification into the fault analysis model, judges whether a solution can be obtained from the fault analysis model according to the data identification, if so, starts a corresponding emergency response mechanism according to the solution obtained by analysis, and if not, triggers the exception handling submodule;
and the exception handling submodule sends an exception alarm to the on-duty personnel and requests the corresponding exception equipment to upload an exception handling solution after subsequent exception handling.
The device for monitoring the abnormal operation in the emergency command communication comprises a data identification, a data length, an equipment identification, an equipment position and a solution, wherein the historical data of the abnormal operation comprises the data identification, the data length, the equipment identification, the equipment position and the solution; the operation data of each newly added communication base station, communication node and terminal equipment comprises data identification, data length, equipment identification and equipment position.
The device for monitoring the abnormal operation in the emergency command communication comprises an abnormal alarm processing module, a monitoring module and a monitoring module, wherein the abnormal alarm processing module is specifically used for sending an abnormal alarm to a mobile terminal of a duty worker if a solution is solved by the duty worker; if the solution is solved by the professional, sending an abnormal alarm to a professional mobile terminal; if the solution is self-recovery solution of the program, the cloud platform sends a program solution code to the corresponding abnormal device, and the abnormal device runs the solution code to achieve the solution of the abnormal solution; wherein, the abnormal alarm information comprises an equipment identifier and an equipment position.
The beneficial effect that this application realized is as follows: by adopting the fault analysis and solution method, the abnormal equipment can be found more quickly and accurately during emergency command communication, an accurate solution is provided for abnormal conditions, and the time for abnormal recovery is shortened.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings can be obtained by those skilled in the art according to the drawings.
Fig. 1 is a flowchart of a method for analyzing and resolving a fault in emergency command communication according to an embodiment of the present disclosure;
FIG. 2 is a flowchart of a specific method for constructing a fault analysis model according to a large amount of abnormal operation data by the cloud platform;
fig. 3 is a specific operation flowchart after the cloud platform acquires the operation data;
fig. 4 is a schematic diagram of a fault analysis and resolution device in emergency command communication according to the second embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
An embodiment of the present application provides a method for analyzing and solving a fault in emergency command communication, which is applied to a cloud platform, and as shown in fig. 1, the method includes:
recording historical operation data of each device in a log mode in each communication base station, each communication node and each terminal device, and uploading the operation data to the cloud platform periodically; in addition, each communication base station, each communication node and each terminal device also upload a solution to the cloud platform after repairing the abnormality; the cloud platform stores the operation data and the abnormal repair solution of each communication base station, each communication node and each terminal device, and learns the log data of the historical faults by means of the learning mode of the operation data and the abnormal repair solution, so that the reason for automatically analyzing the faults of the newly added devices is precipitated, and the faults are predicted;
the operation data of each communication base station, communication node and terminal device includes normal operation data and abnormal operation data, and is specifically shown in table 1 below:
table 1 statistical table of operational data
As shown in table 1 above, the operation data of each communication base station, communication node, and terminal device is counted in the cloud platform, where the operation data of each communication base station includes normal operation data (i.e., the received data is in the form of 0xA101) and abnormal operation data, and the abnormal operation data includes multiple abnormal types, including but not limited to three abnormal types described in table 1, that is, the first received data identifier 0xA101 indicates a hardware fault type, such as an ac breakpoint, an idle trip, a power phase loss/phase failure, and other hardware abnormalities, and after the data identifier, is a data length + a base station identifier (i.e., data from which base station) + a base station position; the second received data identifier 0xA102 indicates an inter-base station communication interruption anomaly, and is data length + base station identifier (i.e., data from which base station) + base station position after the data identifier; the third received data identifier 0xA103 represents an ECC communication interruption anomaly, that is, a base station network anomaly due to an unreasonable ECC networking; similarly, each communication node and terminal device in table 1 are also provided with corresponding data identifiers to indicate the type of operation data, which is not described in detail herein;
it should be noted that the several exception types described in table 1 are only several exception forms in emergency communication, in addition, a plurality of exception events occur, each device sets corresponding exception data identifiers for different exception events, and the cloud platform determines the exception types according to the exception data identifiers;
further, table 1 shows that communication between devices is possible but an abnormal condition occurs, and in addition, an abnormality that communication between devices is impossible is also included, specifically, because each communication base station, communication node, and terminal device upload data to the cloud platform at regular time, and the cloud platform periodically detects upload data of each communication base station, communication node, and terminal device, if the cloud platform still does not receive the upload data after exceeding the detection period, the cloud platform actively sends a data acquisition request to the device that does not upload data, and if a response is not received for a long time, it is determined that a communication failure occurs in the current device, and the cloud platform sets a corresponding data identifier for such an abnormality, such as 0x 00;
in order to increase the accuracy of the operation exception and the model thereof, the cloud platform constructs a fault analysis model according to a large amount of abnormal operation data by taking the large amount of abnormal operation data of the various types as samples of the fault analysis model, as shown in fig. 2, the method specifically comprises the following substeps:
the preprocessing specifically comprises the steps of screening and cleaning abnormal operation data;
specifically, the data screening includes removing log data which has no value on fault location, and selecting the log data which can be used as machine learning model classification, namely the log data with a large number of logs corresponding to fault categories;
the data cleaning comprises the steps of removing special symbols, various format type useless information and the like in the log data after data screening, and separating out data identification, equipment position and solution; in addition, the log data in different formats needs to be washed respectively.
in the embodiment of the application, data identification features and solution features are extracted from the description of the solution of abnormal operation data, the data identification features represent abnormal types, and the solution features represent solutions for solving corresponding abnormalities, wherein the solutions comprise solutions solved by a person on duty, solutions solved by a professional and solutions solved by a program; taking the data identification features and the solution features as feature words to perform feature weight calculation;
specifically, the manner of calculating the feature weight includes, but is not limited to, a boolean weight calculation method, a frequency weight calculation method, a word frequency-inverse file frequency calculation method, and the like;
the Boolean weight calculation method specifically comprises the following steps: searching whether a certain fault characteristic word appears in the text, if so, setting the weight to be 1, and if not, setting the weight to be 0;
the frequency weight calculation method specifically comprises the following steps: acquiring the frequency weight of a certain fault feature word in a text, namely the feature word with more occurrence times, wherein the set weight number is larger, namely the importance of the feature word is larger;
the word frequency-inverse file frequency calculation method specifically comprises the following steps: the more times a certain fault feature word appears in one log data, and the less times the certain fault feature word appears in all log data, the better the feature word has the classification distinguishing capability, and the more the fault feature word can represent that the log data is fault data; specifically, the word frequency-inverse file frequency is calculated by the following formula:
TF_IDF＝TF×IDF
TF＝m/n
wherein, TF _ IDF is the word frequency-inverse file frequency, TF is the frequency of a certain fault characteristic word appearing in a piece of log data, m is the frequency of the fault characteristic word appearing in the certain log data, and n is the total vocabulary of the certain log data with the fault characteristic word; the IDF is the inverse file frequency, x is the total number of all abnormal operation log data after data screening and cleaning, and y is all the abnormal operation log data after data screening and cleaning of a certain fault feature word.
After the feature weight calculation in the above several ways, a text vector of each piece of log data is obtained, and a text vector set is formed.
in order to prevent overfitting, text vectors are subjected to combined grouping training, for example, the text vectors are divided into a training set, a verification set and a test set according to the proportion of 6:2: 2; optionally, in practical application, only the text vectors may be divided into a training set and a test set;
the training set is a learning sample data set and is used for training a fault analysis model; the verification set is a parameter for determining the complexity of a network structure or a control model and is used for adjusting the classifier parameter of the learned fault analysis model; the test set is used for verifying the performance of the finally selected optimal model and measuring the recognition rate of the trained fault analysis model;
the fault analysis model training comprises two inputs (a training set is divided into two groups, wherein one input data is used as data to be trained, the other input data is used as fault template data), the data labels from the same class are 1, and the data labels from different classes are 0; and the fault analysis model outputs the similarity of the two data, wherein the output is the similarity of the current input data.
specifically, the value range of the similarity output by the fault analysis model is 0 to 1, the analysis result is output in a probability form, the output analysis result is compared with a preset threshold value, if the output similarity is 1, the trained text vector is more likely to be the same type of fault, if the output similarity is 0, the fault types are different, the fault type to which the trained text vector belongs is output after different fault feature words are input into the fault analysis model, and a corresponding solution is determined according to the fault type;
for example, for a hardware fault exception, the hardware fault exception is recovered by a duty personnel; for the communication abnormality between the devices, recovering by a professional; for ECC communication exception, the program tries to recover itself.
Referring back to fig. 1, step 120, collecting operation data of each newly added communication base station, communication node and terminal device, and inputting a fault analysis model for analysis to obtain a solution;
in the embodiment of the application, each communication base station, each communication node and each terminal device upload the operation data to the cloud platform in real time or at regular time, or the cloud platform sends acquisition instructions to each communication base station, each communication node and each terminal device in real time or at regular time to acquire the operation data of each device;
as shown in fig. 3, after acquiring the operation data of each newly added communication base station, communication node, and terminal device, the cloud platform performs the following substeps:
step 310, acquiring a data identifier in the operating data, judging whether the operating data is normal operating data or abnormal operating data according to the data identifier, if the operating data is abnormal operating data, executing step 320, and if the operating data is normal operating data, continuing to wait for the operating data uploaded by each device;
because the abnormal conditions recorded in the cloud platform may be only part of the abnormal conditions of each communication base station, each communication node and each terminal device, after the cloud platform receives the operation data, as long as the data identifier of the operation data is not the data identifier of the normal operation data, the operation data is all treated as abnormal operation data.
Step 320, inputting the data identifier into the fault analysis model, and judging whether a solution can be acquired from the fault analysis model according to the data identifier, if so, executing step 130, otherwise, executing step 330;
after the data identification is input into the fault analysis model, if the corresponding solution can be found, the data identification is determined to be the recorded abnormal condition, and if the corresponding solution cannot be found, the data identification is determined to be the new abnormal condition.
And step 330, sending an exception alarm to the on-duty personnel, and requesting corresponding exception equipment to upload an exception handling solution after subsequent exception handling.
Referring back to fig. 1, step 130, starting a corresponding emergency response mechanism according to the solution obtained by analysis, and performing exception alarm processing;
specifically, if the solution is solved by the on-duty personnel, an abnormity alarm is sent to the on-duty personnel mobile terminal; if the solution is solved by the professional, sending an abnormal alarm to a professional mobile terminal; if the solution is self-recovery solution of the program, the cloud platform sends a program solution code to the corresponding abnormal device, and the abnormal device runs the solution code to achieve the solution of the abnormal solution;
wherein, including equipment identification and equipment position in the abnormal alarm information, the staff can confirm which kind of equipment takes place unusually according to the equipment identification, confirms equipment place according to the equipment position, makes things convenient for the staff to accurately find unusual equipment. In addition, for notifying the staff on duty or professionals, the solution can be sent to an emergency command center by the cloud platform, and the emergency command center carries out exception handling scheduling on corresponding workers according to the solution.
Example two
An embodiment of the present application provides a fault analysis and solution device in emergency command communication, as shown in fig. 4, including: a fault analysis model building module 410, an operation data acquisition module 420 and an abnormal alarm processing module 430;
the fault analysis model building module 410 is configured to collect a large amount of operation abnormal history data stored in each communication base station, communication node, and terminal device, learn according to a large amount of operation history log data, and build a fault analysis model capable of automatically analyzing and predicting the fault cause of the newly added communication base station, communication node, and terminal device; the abnormal operation data comprises data identification, data length, equipment identification, equipment position and solution;
the operation data acquisition module 420 is configured to acquire operation data of each communication base station, communication node, and terminal device, and input the operation data to a fault analysis model for analysis to obtain a solution; the operation data comprises a data identifier, a data length, an equipment identifier and an equipment position;
the exception alarm processing module 430 is configured to start a corresponding emergency response mechanism according to the analyzed solution, and perform exception alarm processing.
Specifically, the fault analysis model building module 410 is specifically configured to perform data preprocessing on a large amount of abnormal operation log data; performing characteristic weight calculation on the log data to obtain a text vector set; randomly dividing a text vector set into a training set, a verification set and a test set, pairing the training sets in pairs, inputting the training sets into a fault analysis model for training, adjusting the trained fault analysis model by using the verification set, and testing by using the test set and the trained fault analysis model to obtain a finally trained fault analysis model; and determining a corresponding solution according to the analysis result of the fault analysis model.
The operation data acquisition module 420 specifically includes an operation data type identification submodule 421, a solution query submodule 422, and an exception handling submodule 423;
the operation data type identification submodule 421 obtains a data identifier in the operation data, determines whether the operation data is normal operation data or abnormal operation data according to the data identifier, if the operation data is determined to be abnormal operation data, triggers the solution query submodule 422, otherwise, continues to wait for the operation data uploaded by each device;
the solution query sub-module 422 inputs the data identifier into the fault analysis model, determines whether a solution can be obtained from the fault analysis model according to the data identifier, if so, starts a corresponding emergency response mechanism according to the solution obtained by analysis, and if not, triggers the exception handling sub-module 423;
the exception handling sub-module 423 sends an exception alarm to the on-duty staff, and requests the corresponding exception device to upload the solution of exception handling after subsequent exception handling.
Further, the exception alarm processing module 430 is specifically configured to send an exception alarm to the mobile terminal of the duty worker if the solution is solved by the duty worker; if the solution is solved by the professional, sending an abnormal alarm to a professional mobile terminal; if the solution is self-recovery solution of the program, the cloud platform sends a program solution code to the corresponding abnormal device, and the abnormal device runs the solution code to achieve the solution of the abnormal solution; wherein, the abnormal alarm information comprises an equipment identifier and an equipment position.
The above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the present disclosure, which should be construed in light of the above teachings. Are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (8)
1. A method for analyzing and resolving a fault in emergency command communications, comprising:
collecting a large amount of abnormal operation historical data stored in each communication base station, communication node and terminal equipment, learning according to a large amount of operation historical log data, and constructing a fault analysis model capable of automatically analyzing and predicting the fault reasons of the newly added communication base station, communication node and terminal equipment;
collecting operation data of each newly added communication base station, communication nodes and terminal equipment, and inputting a fault analysis model for analysis to obtain a solution;
starting a corresponding emergency response mechanism according to the analyzed solution, and performing exception alarm processing;
the method comprises the following steps of constructing a fault analysis model according to a large amount of abnormal operation data, and specifically comprising the following substeps:
carrying out data preprocessing on a large amount of abnormal operation log data; the preprocessing specifically comprises the steps of carrying out data screening and data cleaning on abnormal operation data, eliminating special symbols and various format useless information in the log data after the data screening, and separating out data identification, equipment position and solution;
performing characteristic weight calculation on the log data to obtain a text vector set; extracting data identification features and solution features from the description of the solution of the abnormal operation data, wherein the data identification features represent abnormal types, and the solution features represent solutions for solving corresponding abnormalities; taking the data identification features and the solution features as feature words to perform feature weight calculation; the characteristic weight calculation mode comprises a word frequency-inverse file frequency calculation method, which specifically comprises the following steps: the more times a certain fault feature word appears in one log data, and the less times the certain fault feature word appears in all log data, the better the feature word has the classification distinguishing capability, and the more the fault feature word can represent that the log data is fault data; specifically, the word frequency-inverse file frequency is calculated by the following formula:
TF_IDF＝TF×IDF
TF＝m/n
wherein, TF _ IDF is the word frequency-inverse file frequency, TF is the frequency of a certain fault characteristic word appearing in a piece of log data, m is the frequency of the fault characteristic word appearing in the certain log data, and n is the total vocabulary of the certain log data with the fault characteristic word; IDF is the frequency of the inverse file, x is the total number of all abnormal operation log data after data screening and cleaning, and y is all abnormal operation log data after data screening and cleaning of a certain fault feature word;
randomly dividing a text vector set into a training set, a verification set and a test set, pairing the training sets in pairs, inputting the training sets into a fault analysis model for training, adjusting the trained fault analysis model by using the verification set, and testing by using the test set and the trained fault analysis model to obtain a finally trained fault analysis model;
and determining a corresponding solution according to the analysis result of the fault analysis model.
2. The method of claim 1, wherein the cloud platform performs the following sub-steps after obtaining the operational data:
s1, acquiring a data identifier in the running data, judging whether the running data is normal running data or abnormal running data according to the data identifier, if the running data is abnormal running data, executing a step S2, and if the running data is normal running data, continuing to wait for the running data uploaded by each device;
s2, inputting the data identification into the fault analysis model, judging whether a solution can be obtained from the fault analysis model according to the data identification, if so, starting a corresponding emergency response mechanism according to the solution obtained by analysis, otherwise, executing the step S3;
and S3, sending an exception alarm to the staff member, and requesting the corresponding exception equipment to upload the exception handling solution after subsequent exception handling.
3. The method of fault analysis and resolution in emergency command communications of claim 1 wherein operational anomaly history data includes data identification, data length, equipment identification, equipment location and resolution; the operation data of each newly added communication base station, communication node and terminal equipment comprises data identification, data length, equipment identification and equipment position.
4. The method for analyzing and resolving the fault in the emergency command communication according to claim 1, wherein the step of starting a corresponding emergency response mechanism and performing the abnormal alarm processing specifically comprises the steps of: if the solution is solved by the on-duty personnel, sending an abnormal alarm to the on-duty personnel mobile terminal; if the solution is solved by the professional, sending an abnormal alarm to a professional mobile terminal; if the solution is self-recovery solution of the program, the cloud platform sends a program solution code to the corresponding abnormal device, and the abnormal device runs the solution code to achieve the solution of the abnormal solution; wherein, the abnormal alarm information comprises an equipment identifier and an equipment position.
5. An abnormal operation monitoring device in emergency command communication, comprising: the system comprises a fault analysis model construction module, an operation data acquisition module and an abnormal alarm processing module;
the fault analysis model building module is used for collecting a large amount of abnormal operation historical data stored in each communication base station, communication node and terminal equipment, learning according to a large amount of operation historical log data, and building a fault analysis model capable of automatically analyzing and predicting the fault reasons of the newly added communication base station, communication node and terminal equipment;
the operation data acquisition module is used for acquiring operation data of each newly added communication base station, communication node and terminal equipment, and inputting the operation data into the fault analysis model for analysis to obtain a solution;
the abnormal alarm processing module is used for starting a corresponding emergency response mechanism according to the solution obtained by analysis and processing abnormal alarm;
the method comprises the following steps of constructing a fault analysis model according to a large amount of abnormal operation data, and specifically comprising the following substeps:
carrying out data preprocessing on a large amount of abnormal operation log data; the preprocessing specifically comprises the steps of carrying out data screening and data cleaning on abnormal operation data, eliminating special symbols and various format useless information in the log data after the data screening, and separating out data identification, equipment position and solution;
performing characteristic weight calculation on the log data to obtain a text vector set; extracting data identification features and solution features from the description of the solution of the abnormal operation data, wherein the data identification features represent abnormal types, and the solution features represent solutions for solving corresponding abnormalities; taking the data identification features and the solution features as feature words to perform feature weight calculation; the characteristic weight calculation mode comprises a word frequency-inverse file frequency calculation method, which specifically comprises the following steps: the more times a certain fault feature word appears in one log data, and the less times the certain fault feature word appears in all log data, the better the feature word has the classification distinguishing capability, and the more the fault feature word can represent that the log data is fault data; specifically, the word frequency-inverse file frequency is calculated by the following formula:
TF_IDF＝TF×IDF
TF＝m/n
wherein, TF _ IDF is the word frequency-inverse file frequency, TF is the frequency of a certain fault characteristic word appearing in a piece of log data, m is the frequency of the fault characteristic word appearing in the certain log data, and n is the total vocabulary of the certain log data with the fault characteristic word; IDF is the frequency of the inverse file, x is the total number of all abnormal operation log data after data screening and cleaning, and y is all abnormal operation log data after data screening and cleaning of a certain fault feature word;
randomly dividing a text vector set into a training set, a verification set and a test set, pairing the training sets in pairs, inputting the training sets into a fault analysis model for training, adjusting the trained fault analysis model by using the verification set, and testing by using the test set and the trained fault analysis model to obtain a finally trained fault analysis model;
and determining a corresponding solution according to the analysis result of the fault analysis model.
6. The device for monitoring the abnormal operation in the emergency command communication according to claim 5, wherein the operation data acquisition module specifically comprises an operation data type identification sub-module, a solution query sub-module and an abnormality processing sub-module;
the operation data type identification submodule acquires a data identifier in the operation data, judges whether the operation data is normal operation data or abnormal operation data according to the data identifier, if the operation data is judged to be abnormal operation data, triggers the solution query submodule, and otherwise, continuously waits for the operation data uploaded by each device;
the solution query submodule inputs the data identification into the fault analysis model, judges whether a solution can be obtained from the fault analysis model according to the data identification, if so, starts a corresponding emergency response mechanism according to the solution obtained by analysis, and if not, triggers the exception handling submodule;
and the exception handling submodule sends an exception alarm to the on-duty personnel and requests the corresponding exception equipment to upload an exception handling solution after subsequent exception handling.
7. The abnormal operation monitoring device in emergency command communication according to claim 5, wherein the history data of abnormal operation includes data identification, data length, equipment identification, equipment position and solution; the operation data of each newly added communication base station, communication node and terminal device comprises a data identifier, a data length, a device identifier and a device position.
8. The device for monitoring abnormal operation in emergency command communication according to claim 7, wherein the abnormal alarm handling module is specifically configured to send an abnormal alarm to the mobile terminal of the duty person if the solution is solved by the duty person; if the solution is solved by the professional, sending an abnormal alarm to a professional mobile terminal; if the solution is self-recovery solution of the program, the cloud platform sends a program solution code to the corresponding abnormal device, and the abnormal device runs the solution code to achieve the solution of the abnormal solution; wherein, the abnormal alarm information comprises an equipment identifier and an equipment position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011304841.3A CN112492567B (en) | 2020-11-19 | 2020-11-19 | Fault analysis and solution method and device in emergency command communication |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011304841.3A CN112492567B (en) | 2020-11-19 | 2020-11-19 | Fault analysis and solution method and device in emergency command communication |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112492567A CN112492567A (en) | 2021-03-12 |
| CN112492567B true CN112492567B (en) | 2022-03-18 |
Family
ID=74932151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011304841.3A Active CN112492567B (en) | 2020-11-19 | 2020-11-19 | Fault analysis and solution method and device in emergency command communication |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112492567B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113091817A (en) * | 2021-04-06 | 2021-07-09 | 重庆大学 | Triethylene glycol dewatering device state monitoring and fault diagnosis system |
| CN113505039A (en) * | 2021-07-13 | 2021-10-15 | 河北建筑工程学院 | Communication fault analysis method, device and system |
| CN114666882B (en) * | 2022-04-25 | 2024-01-02 | 浙江省通信产业服务有限公司 | Power control method, device, base station and storage medium |
| CN115083093A (en) * | 2022-05-16 | 2022-09-20 | 浙江台谊消防股份有限公司 | Electrical fire monitoring system |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708454B (en) * | 2012-05-14 | 2016-06-29 | 北京奇虎科技有限公司 | Solution of terminal fault provides method and device |
| CN103914735B (en) * | 2014-04-17 | 2017-03-29 | 北京泰乐德信息技术有限公司 | A kind of fault recognition method and system based on Neural Network Self-learning |
| CN105631596B (en) * | 2015-12-29 | 2020-12-29 | 山东鲁能软件技术有限公司 | Equipment fault diagnosis method based on multi-dimensional piecewise fitting |
| US10410115B2 (en) * | 2017-04-28 | 2019-09-10 | Intel Corporation | Autonomous machines through cloud, error corrections, and predictions |
| US10536370B2 (en) * | 2017-08-08 | 2020-01-14 | Dell Products Lp | Method and system to avoid temporary traffic loss with BGP ethernet VPN multi-homing with data-plane MAC address learning |
| CN109034368B (en) * | 2018-06-22 | 2021-10-15 | 北京航空航天大学 | DNN-based complex equipment multiple fault diagnosis method |
| CN111209131A (en) * | 2019-12-30 | 2020-05-29 | 航天信息股份有限公司广州航天软件分公司 | Method and system for determining fault of heterogeneous system based on machine learning |
-
2020
- 2020-11-19 CN CN202011304841.3A patent/CN112492567B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| 基于知识图谱的智能故障诊断研究;刘瑞宏，谢国强，苑宗港，宋文婷，王高虎;《邮电设计技术》;20201031;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112492567A (en) | 2021-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112492567B (en) | Fault analysis and solution method and device in emergency command communication | |
| CN110210512B (en) | Automatic log anomaly detection method and system | |
| WO2021184630A1 (en) | Method for locating pollutant discharge object on basis of knowledge graph, and related device | |
| CN106209405B (en) | Method for diagnosing faults and device | |
| CN115496625A (en) | Pipe network safety linkage disposal method for intelligent gas and Internet of things system | |
| CN111290900B (en) | Software fault detection method based on micro service log | |
| CN104363106A (en) | Electric power information communication fault early warning analysis method based on big-data technique | |
| CN113190421A (en) | Detection and analysis method for equipment health state of data center | |
| CN113313280B (en) | Cloud platform inspection method, electronic equipment and nonvolatile storage medium | |
| CN113542039A (en) | Method for positioning 5G network virtualization cross-layer problem through AI algorithm | |
| CN115809183A (en) | Method for discovering and disposing information-creating terminal fault based on knowledge graph | |
| CN115357418A (en) | Microservice fault detection method and device, storage medium and computer equipment | |
| CN112988843B (en) | SMT chip mounter fault management and diagnosis system based on SQL Server database | |
| CN113468022B (en) | Automatic operation and maintenance method for centralized monitoring of products | |
| CN112070180B (en) | Power grid equipment state judging method and device based on information physical bilateral data | |
| CN110855489B (en) | Fault processing method and device and fault processing device | |
| CN115587312B (en) | Fault point positioning method and system based on multidimensional big data analysis | |
| CN116934304A (en) | Intelligent power distribution room equipment operation maintenance management system and method thereof | |
| CN113676343B (en) | Fault source positioning method and device for power communication network | |
| CN114548769A (en) | Intelligent power grid IT asset big data monitoring system and method | |
| CN114398947A (en) | Expert system-based power grid fault automatic classification method and system | |
| CN117728582A (en) | 5G power monitoring system with self-adaptive judging capability | |
| CN113010394B (en) | Machine room fault detection method for data center | |
| CN117667497B (en) | Automatic fault analysis method and system for dispatching centralized system | |
| CN113591937B (en) | Power system key node identification method based on local distance coding |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |