CN115203014A - Ecological service abnormity restoration system and restoration method based on deep learning - Google Patents
Ecological service abnormity restoration system and restoration method based on deep learning Download PDFInfo
- Publication number
- CN115203014A CN115203014A CN202210531037.1A CN202210531037A CN115203014A CN 115203014 A CN115203014 A CN 115203014A CN 202210531037 A CN202210531037 A CN 202210531037A CN 115203014 A CN115203014 A CN 115203014A
- Authority
- CN
- China
- Prior art keywords
- module
- log
- data
- abnormal
- service
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/362—Software debugging
- G06F11/3628—Software debugging of optimised code
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3688—Test management for test execution, e.g. scheduling of test suites
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an ecological service abnormity restoration system based on deep learning and a restoration method thereof. The invention relates to the technical field of service abnormity repair, wherein a log module is used for collecting and analyzing logs and primarily screening abnormal information data for a model training module; the model training module is used for model design and model training and provides a judgment basis for the matching abnormity restoration method; the abnormal repairing module is used for method construction and method matching and provides a repairing scheme for code release and test; the code release testing module is used for code release and test verification to ensure that the repair matched by the exception is correct and reliable. The invention provides an ecological service abnormity restoration system based on deep learning, which integrates abnormity records of vehicle-side access cloud ecological service, a cloud self-built ecological service abnormity restoration method library, a test script and a cloud code management and release system.
Description
Technical Field
The invention relates to the technical field of deep learning abnormity repair, in particular to a deep learning-based ecological service abnormity repair system and a deep learning-based ecological service abnormity repair method.
Background
With the continuous development of the automobile industry, an automobile cabin gradually becomes a carrier for fusing multi-ecological services, and a user can enjoy more and more functions in an automobile, but the stability of the functions is very important, the user has relatively intense spirit in the automobile using process, and once the functional problem is met, the emotion of the user is fluctuated, so that how to ensure the stability of the ecological service function is very important.
The technical problems to be solved by the invention are as follows: the existing ecological service system has huge related micro-services and codes, and meanwhile, the stock vehicles in the market are more and more, and the ecological services comprise navigation, music, videos, radio stations and the like and are high-frequency use functions on the vehicles, so that once abnormal services occur, the problems of the same kind need to be solved as soon as possible to avoid the occurrence of the problems. In the existing operation and maintenance system, the problem needs to be manually monitored or input after sale, and then the abnormity is analyzed and repaired, so that the timeliness is insufficient.
At present, all common implementation schemes for abnormal repair need to be completed manually by operation and maintenance personnel and developers, and a mature and reliable automatic system is not formed.
Patent document 1 (CN 113240011A) relates to the field of data processing, and in particular relates to a deep learning driven anomaly identification and repair method and an intelligent system. The method comprises the following steps: s1: data structure identification, S2: data feature transformation, S3: training an anomaly detection and repair neural network, S4: abnormal data identification and abnormal repair, S5: and restoring the data characteristics. Using a deep learning method, using a two-component hybrid model for each feature, wherein one component is used for interpreting clean units (i.e. normal values) and the other component is used for interpreting abnormal units (i.e. abnormal values); simulating potential normal data distribution by reducing the effect of abnormal cells, providing abnormal value scores for the data cells and an estimate of cell repair; the variation self-encoder and the generation countermeasure network are combined, so that a better repair result is generated; finally, cell-level (unit level) abnormity identification and repair are carried out on the mixed attribute data in an unsupervised learning mode.
Patent document 2 (CN 110991659A) provides an abnormal node identification method, apparatus, electronic device, and storage medium, where the method includes: the method comprises the steps of inputting feature data of a test image into a deep learning model to be recognized, wherein the deep learning model to be recognized comprises a plurality of nodes, monitoring the processing time of a designated node in the plurality of nodes in the process of processing the feature data by the deep learning model to be recognized, wherein the processing time of the designated node is the time for processing received data by the designated node, and when the processing time of the designated node is greater than a preset time threshold value, determining the designated node as an abnormal node. Abnormal nodes are identified from a plurality of nodes contained in the deep learning model, and after the abnormal nodes are identified, the abnormal nodes can be further processed, the deep learning model reasoning acceleration performance method can be deeply researched, and the running speed of the deep learning model is accelerated.
Patent document 3 (CN 111552609A) discloses a method, a system, a storage medium, a program, and a server for detecting an abnormal state, which perform unsupervised learning on LSTM1, supervised learning on LSTM2, and fast solving a server failure by using log information; predicting log information with time sequence characteristics by adopting an LSTM1 cyclic neural network, sending an alarm by identifying the abnormal occurrence of the log information, and assisting operation and maintenance personnel to check failure reasons; LSTM2 gives the cause of the failure that caused the current situation by logging information for a period of time before the failure occurred.
Disclosure of Invention
The invention provides a system and a method for restoring ecological service abnormity based on deep learning, aiming at overcoming the defects of the prior art, and the invention provides the following technical scheme:
a deep learning based ecological services anomaly remediation system, the system comprising:
the log module is used for collecting and analyzing logs and primarily screening abnormal information data for the model training module;
the model training module is used for model design and model training and provides judgment basis for the matching abnormity restoration method;
the abnormal repairing module is used for method construction and method matching and provides a repairing scheme for code release and test; the model training module model selects a network structure and determines a loss function, and the network structure selects an RNN/LSTM neural network;
and the code release testing module is used for releasing codes and testing and verifying to ensure that the repair matched by the abnormity is correct and reliable.
Preferably, the log module designs data to be collected according to requirements, forms a data model, and then realizes embedding points in each abnormal capturing logic in the ecological service;
the log module is realized based on kafka during data acquisition, and the embedded point log data generated in each ecological service are synchronized to the data storage module in real time;
and the log module is used for storing data, and persistently storing the log data synchronized by the kafka, wherein the storage time can be adjusted according to resources, and induction compression processing is required to be performed on historical data.
Preferably, the log module classifies the logs, and preliminarily classifies the persistent logs according to service names, log grades, log acquisition time, abnormal types in log information and the like; and log sorting: sorting is carried out according to the exception types preferentially, exceptions needing to be processed preferentially are screened, sorting is carried out according to service names, and ecological services with unstable service quality are screened for optimization; and finally, outputting a log result: and inputting the sorted high-frequency abnormal information into a model training module.
Preferably, the determination of the loss function is specifically: firstly, selecting a predicted restoration method according to the forward direction of input log information, then calculating loss according to a prediction result and a model actual output result, and then selecting a standard cross entropy loss function according to a loss updating parameter:
where x represents the sample, y represents the actual label, a represents the predicted output, and n represents the total number of samples.
Preferably, the abnormality repairing module performs method construction and abnormality matching, and the method construction needs to be realized by original input, dictionary table generation and method correction: original input: the method library is initially established by requiring a developer to input a batch of common abnormal processing methods according to experience; dictionary: extracting key information, such as abnormal types, method comments and other information according to the input method, and generating a corresponding field table for matching; the method comprises the following steps: finishing the method that the test result fails according to the test verification result;
and matching the abnormality with the method, adopting common similarity screening, and when the similarity between the model output result and the data in the dictionary table of the method library is more than 90%, using the corresponding method.
Preferably, the code issuing test module submits the code by submitting the code and automatically issuing, and the system can upload the modified abnormal code by pre-authorizing the git authority in the system; automatic issuing is realized based on a Jenkins production line, and when codes in git change, jenkins can be packed and issued;
the test verification is realized by monitoring the service state, triggering the test and judging the result: monitoring the service state, starting a monitor to monitor the service state of the ecological service cluster, and notifying a trigger test after the restart release of the existing service; triggering test, calling corresponding full-scale script to perform function test according to the result of monitoring state transmission; and judging a result, judging whether the repair is finished according to the test result of the script and a preset test result, and feeding the result back to the model training module.
An ecological service abnormity restoration method based on deep learning, comprising the following steps:
step 1: collecting and analyzing logs, and primarily screening abnormal information data for a model training module;
the log module designs data to be acquired according to requirements, forms a data model, and then realizes embedding in each abnormal capturing logic in the ecological service;
the log module is realized based on kafka during data acquisition, and the embedded point log data generated in each ecological service are synchronized to the data storage module in real time;
the log module stores data, performs persistent storage on the log data synchronized by the kafka, the storage time can be adjusted according to resources, and induction compression processing is performed on historical data
And 2, step: carrying out model design and model training, and providing a judgment basis for a matching anomaly restoration method;
model training optimization requires continuous optimization algorithm and resource allocation is adjusted; adam is used as an optimization algorithm, and the optimization algorithm is carried out based on momentum and a self-adaptive learning rate; allocating the resource allocation according to the actual condition of model operation;
and 3, step 3: the abnormal repairing module is used for method construction and method matching and provides a repairing scheme for code release and test, the abnormal repairing module carries out method construction and abnormal and method matching, and the method construction needs to be realized through original input, dictionary table generation and method correction;
and 4, step 4: and releasing codes and testing and verifying to ensure the correctness and reliability of the repair matched by the exception.
A vehicle, comprising: the system comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the program to realize the ecological service abnormity repairing method based on deep learning.
A computer-readable storage medium having stored thereon a computer program, which is executed by a processor, for implementing a deep learning-based ecological service anomaly restoration method.
A computer device comprises a storage and a processor, wherein the storage stores a computer program, and the processor realizes the ecological service abnormity restoration method based on deep learning when executing the computer program.
The invention has the following beneficial effects:
the invention provides an ecological service abnormity restoration system based on deep learning, which integrates abnormity records of vehicle-side access cloud ecological services, a cloud self-built ecological service abnormity restoration method library, a test script and a cloud code management and release system, and realizes automatic restoration, automatic test and automatic release of the ecological services based on a deep learning training model.
According to the method, a large amount of real data in the log are input into a preset model, the model is verified and optimized according to the forward calculation result of the log data, an optimal repairing method is matched according to the final output of the model, the abnormal code logic is automatically repaired, then the code is submitted through a code issuing production line, test verification is carried out, and the log is deployed to a production environment after verification is completed.
The purpose of the invention is as follows: the method can meet the requirement of abnormal restoration all the time, timely process the abnormality of the ecological service, avoid more market problems caused by one abnormality, and realize the automatic flow of abnormal analysis, abnormal restoration, code release and test regression. The system is integrally divided into four modules, namely a log module, a model training module, an exception recovery method library and a code release test module.
Both the patent 1 and the invention analyze the abnormity based on deep learning, the patent 1 is applied to abnormal data monitoring, and the invention introduces an abnormity repair method library and a code release verification module to carry out automatic abnormity repair on ecological services. The patent 2 identifies abnormal nodes in the image characteristic data processing process based on deep learning, and the invention trains a deep learning model based on text-type log information data to complete automatic abnormal restoration of ecological services. Both patent 3 and the invention carry out fault treatment based on deep learning, patent 3 is used for early warning and assisting operation and maintenance personnel to troubleshoot faults, and the invention carries out automatic abnormal repair on ecological services based on an abnormal analysis result, an abnormal repair method library and a code release verification module.
The advantages are that:
1. based on a log module of embedded points and real-time transmission acquisition, the effectiveness of source data is guaranteed before the source data is input into a training model by acquiring real log information and filtering the log information preliminarily;
2. the model training module based on deep learning continuously optimizes and iterates the model, so that the accuracy of the output result of the model can be greatly improved;
3. the system comprises an exception recovery method library based on a dictionary table and a similarity matching mechanism and a code release test module based on automatic release and test verification, so that manpower can be effectively liberated, and all-weather full-time processing can be performed on common exceptions easy to occur through a server;
4. after the whole system is on line, market complaints can be effectively reduced, and market problem complaints caused by similar problems can be avoided in advance.
Drawings
FIG. 1 is a schematic diagram of an ecological service anomaly repairing system architecture based on deep learning;
FIG. 2 is a schematic diagram of a log module;
FIG. 3 is a schematic diagram of a model training module;
FIG. 4 is a schematic diagram of an anomaly recovery module;
FIG. 5 is a schematic diagram of a code release and test module
Detailed Description
The present invention will be described in detail with reference to specific examples.
The first embodiment is as follows:
as shown in fig. 1 to 5, the specific optimized technical solution adopted to solve the above technical problems of the present invention is: the invention relates to an ecological service abnormity restoration system and method based on deep learning. FIG. 1 is a schematic diagram of an ecological service anomaly repairing system architecture based on deep learning according to an embodiment of the present invention
An ecological service anomaly remediation system based on deep learning, the system comprising:
the log module is used for collecting and analyzing logs and primarily screening abnormal information data for the model training module
The model training module is used for model design and model training and provides judgment basis for the matching abnormity restoration method
An exception recovery module for method construction and method matching, providing a recovery scheme for code release and testing
And the code release testing module is used for releasing codes and testing and verifying to ensure that the repair matched by the abnormity is correct and reliable.
The second embodiment is as follows:
the second embodiment of the present application differs from the first embodiment only in that:
fig. 2 is a schematic structural diagram of a log module according to an embodiment of the present invention, please refer to fig. 2.
The log module designs data to be acquired according to requirements, forms a data model, and then realizes embedding points in each abnormal capturing logic in the ecological service;
the log module is realized based on kafka during data acquisition, and the embedded point log data generated in each ecological service are synchronized to the data storage module in real time;
and the log module is used for storing data, and persistently storing the log data synchronized by the kafka, wherein the storage time can be adjusted according to resources, and induction compression processing is required to be performed on historical data.
The third concrete embodiment:
the difference between the third embodiment and the second embodiment of the present application is only that:
the log module classifies logs, and performs preliminary classification on the persistent logs according to service names, log grades, log acquisition time, abnormal types in log information and the like; and log sorting: sorting is carried out according to the exception types preferentially, exceptions needing to be processed preferentially are screened, sorting is carried out according to the service names, and ecological services with unstable service quality are screened for optimization; and finally, outputting a log result: and inputting the sorted high-frequency abnormal information into a model training module.
The fourth concrete embodiment:
the difference between the fourth embodiment and the third embodiment is only that:
fig. 3 is a schematic structural diagram of a model training module according to an embodiment of the present invention, please refer to fig. 3.
The model training module includes two types of functions: model design and model optimization.
The realization of model design needs to select network structure and design loss function: 1. the network structure refers to a network framework in a neural network algorithm, and in the system, as model training needs to be carried out based on a large amount of log information texts, an RNN/LSTM neural network is selected;
the fifth concrete embodiment:
the difference between the fifth embodiment and the fourth embodiment is only that:
loss function: the loss function has the function of identifying the optimal parameter from a plurality of parameter values, and the determination of the loss function specifically comprises the following steps: firstly, selecting a predicted repairing method according to the forward direction of input log information, then calculating loss according to a predicted result and a model actual output result, and then selecting a standard cross entropy loss function according to a loss updating parameter:
where x denotes the samples, y denotes the actual label, a denotes the predicted output, and n denotes the total number of samples.
The realization of model training optimization requires continuous optimization algorithm and resource allocation adjustment. The optimization algorithm uses Adam, and the optimization algorithm is performed based on momentum and adaptive learning rate. The resource allocation can be distributed according to the actual condition of the model operation.
The sixth specific embodiment:
the difference between the sixth embodiment and the fifth embodiment is only that:
the abnormity repairing module carries out method construction and abnormity and method matching, and the method construction needs to be realized through original input, dictionary table generation and method correction: original input: the method library is initially established by requiring a developer to input a batch of common abnormal processing methods according to experience; a dictionary: extracting key information, such as abnormal types, method comments and other information according to the input method, and generating a corresponding field table for matching; the method comprises the following steps: finishing the method that the test result fails according to the test verification result;
and matching the abnormality with the method, adopting common similarity screening, and when the similarity between the model output result and the data in the dictionary table of the method library is more than 90%, using the corresponding method.
The seventh specific embodiment:
the seventh embodiment of the present application differs from the sixth embodiment only in that:
the code issuing test module submits the codes by submitting the codes and automatically issuing, and the system can upload the modified abnormal codes by pre-authorizing git authority in the system; automatic issuing, which is realized based on a Jenkins production line, and when codes in git change, jenkins can be packed and issued;
the test verification is realized by monitoring the service state, triggering the test and judging the result: monitoring the service state, starting a monitor to monitor the service state of the ecological service cluster, and notifying a trigger test after the restart release of the existing service; triggering test, calling corresponding full-scale script to perform function test according to the result of monitoring state transmission; and judging a result, judging whether the repair is finished according to the test result of the script and a preset test result, and feeding the result back to the model training module.
The invention provides a scheme of an ecological service abnormity restoration system based on deep learning, which integrates a plurality of modules of log information, model training, a method library and test verification to form a whole set of complete automatic abnormity restoration scheme, and can effectively solve the problems of operation and maintenance and human resource development.
The eighth embodiment:
the eighth embodiment of the present application differs from the seventh embodiment only in that:
an ecological service abnormity restoration method based on deep learning, comprising the following steps:
step 1: collecting and analyzing logs, and primarily screening abnormal information data for a model training module;
the log module designs data to be acquired according to requirements, forms a data model, and then realizes embedding points in each abnormal capturing logic in the ecological service;
the log module is realized based on kafka during data acquisition, and the embedded point log data generated in each ecological service are synchronized to the data storage module in real time;
the log module stores data, performs persistent storage on the log data synchronized by the kafka, the storage time can be adjusted according to resources, and induction compression processing is performed on historical data
Step 2: carrying out model design and model training, and providing a judgment basis for a matching anomaly restoration method;
model training optimization requires continuous optimization algorithm and resource allocation is adjusted; adam is used as an optimization algorithm, and the optimization algorithm is carried out based on momentum and a self-adaptive learning rate; allocating the resource allocation according to the actual condition of the model operation;
and 3, step 3: the abnormal repairing module is used for method construction and method matching and provides a repairing scheme for code release and test, the abnormal repairing module carries out method construction and abnormal and method matching, and the method construction needs to be realized through original input, dictionary table generation and method correction;
and 4, step 4: and releasing codes and testing and verifying to ensure the correctness and reliability of the repair matched by the exception.
The specific embodiment is nine:
the difference between the ninth embodiment and the eighth embodiment is only that:
the present invention provides a vehicle, including: the system comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the program to realize the ecological service abnormity repairing method based on deep learning.
The specific example is ten:
the embodiment ten of the present application differs from the embodiment nine only in that:
the present invention provides a computer-readable storage medium having stored thereon a computer program which is executed by a processor for implementing a deep learning-based ecological service abnormality repairing method.
The concrete example eleven:
the difference between the eleventh embodiment and the tenth embodiment is only that:
the invention provides computer equipment which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes an ecological service abnormity restoration method based on deep learning when executing the computer program.
The specific example twelve:
the twelfth embodiment of the present application differs from the eleventh embodiment only in that:
the invention provides a remote control parking system, which comprises: the log module is mainly used for collecting and analyzing logs and primarily screening abnormal information data for the model training module; the model training module is mainly used for model design and model training and provides a judgment basis for the matching abnormity restoration method; the abnormal repairing method library is mainly used for method construction and method matching and provides a repairing scheme for code release and test; and the code issuing and verifying module is mainly used for issuing codes and testing and verifying to ensure the correctness and reliability of the repairing method matched by the exception.
The above description is only a preferred embodiment of the ecological service abnormity repair system and repair method based on deep learning, and the protection scope of the ecological service abnormity repair system and repair method based on deep learning is not limited to the above embodiments, and all technical solutions belonging to the idea belong to the protection scope of the present invention. It should be noted that modifications and variations which do not depart from the gist of the invention will be those skilled in the art to which the invention pertains and which are intended to be within the scope of the invention.
Claims (10)
1. An ecological service abnormity restoration system based on deep learning is characterized in that: the system comprises:
the log module is used for collecting and analyzing logs and primarily screening abnormal information data for the model training module;
the model training module is used for model design and model training and provides a judgment basis for the matching anomaly restoration method;
the abnormal repairing module is used for method construction and method matching and provides a repairing scheme for code release and test; the model training module model selects a network structure and determines a loss function, and the network structure selects an RNN/LSTM neural network;
and the code release testing module is used for releasing codes and testing and verifying to ensure the correctness and reliability of the repair matched by the exception.
2. The ecological service abnormity restoration system based on deep learning as claimed in claim 1, wherein: the log module designs data to be acquired according to requirements, forms a data model, and then realizes embedding in each abnormal capturing logic in the ecological service;
the log module is realized based on kafka during data acquisition, and the embedded point log data generated in each ecological service are synchronized to the data storage module in real time;
and the log module is used for storing data, and persistently storing the log data synchronized by the kafka, wherein the storage time can be adjusted according to resources, and induction compression processing is required to be performed on historical data.
3. The ecological service abnormity restoration system based on deep learning as claimed in claim 2, wherein: the log module classifies logs, and performs preliminary classification on the persistent logs according to service names, log grades, log acquisition time, abnormal types in log information and the like; and log sorting: sorting is carried out according to the exception types preferentially, exceptions needing to be processed preferentially are screened, sorting is carried out according to the service names, and ecological services with unstable service quality are screened for optimization; and finally, outputting a log result: and inputting the sorted high-frequency abnormal information into a model training module.
4. The ecological service abnormity restoration system based on deep learning as claimed in claim 3, wherein: the determination of the loss function is specifically: firstly, selecting a predicted repairing method according to the forward direction of input log information, then calculating loss according to a predicted result and a model actual output result, and then selecting a standard cross entropy loss function according to a loss updating parameter:
where x represents the sample, y represents the actual label, a represents the predicted output, and n represents the total number of samples.
5. The ecological service abnormity restoration system based on deep learning as claimed in claim 4, wherein:
the abnormity repair module carries out method construction and abnormity and method matching, and the method construction needs to be realized through original input, dictionary table generation and method correction: original input: the method library is initially established by requiring a developer to input a batch of common abnormal processing methods according to experience; a dictionary: extracting key information, such as abnormal types, method comments and other information according to the input method, and generating a corresponding field table for matching; the method comprises the following steps: finishing the method that the test result fails according to the test verification result;
and matching the abnormality with the method, adopting common similarity screening, and when the similarity between the model output result and the data in the dictionary table of the method library is more than 90%, using the corresponding method.
6. The ecological service abnormity restoration system based on deep learning of claim 5, which is characterized in that: the code issuing test module submits the codes by submitting the codes and automatically issuing, and the system can upload the modified abnormal codes by pre-authorizing git authority in the system; automatic issuing is realized based on a Jenkins production line, and when codes in git change, jenkins can be packed and issued;
the test verification is realized by monitoring the service state, triggering the test and judging the result: monitoring the service state, starting a monitor to monitor the service state of the ecological service cluster, and notifying a trigger test after the restart release of the existing service; triggering test, calling corresponding full-scale script to perform function test according to the result of monitoring state transmission; and judging a result, judging whether the repair is finished according to the test result of the script and a preset test result, and feeding the result back to the model training module.
7. An ecological service abnormity restoration method based on deep learning is characterized in that: the method comprises the following steps:
step 1: collecting and analyzing logs, and primarily screening abnormal information data for a model training module;
the log module designs data to be acquired according to requirements, forms a data model, and then realizes embedding in each abnormal capturing logic in the ecological service;
the log module is realized based on kafka during data acquisition, and the embedded point log data generated in each ecological service are synchronized to the data storage module in real time;
the log module stores data, performs persistent storage on the log data synchronized by the kafka, the storage time can be adjusted according to resources, and induction compression processing is performed on historical data
Step 2: carrying out model design and model training, and providing a judgment basis for a matching anomaly restoration method;
model training optimization requires a continuous optimization algorithm, and resource allocation is adjusted; adam is used as an optimization algorithm, and the optimization algorithm is carried out based on momentum and a self-adaptive learning rate; allocating the resource allocation according to the actual condition of the model operation;
and step 3: the abnormal repairing module is used for method construction and method matching and provides a repairing scheme for code release and test, the abnormal repairing module carries out method construction and abnormal and method matching, and the method construction needs to be realized through original input, dictionary table generation and method correction;
and 4, step 4: and releasing codes and testing and verifying to ensure the correctness and reliability of the repair matched by the exception.
8. A vehicle, characterized by comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the deep learning based ecological service anomaly remediation method of claim 7.
9. A computer-readable storage medium having stored thereon a computer program, wherein the program is executed by a processor for implementing the deep learning based ecological service anomaly restoration method according to claim 7.
10. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: the processor, when executing the computer program, implements the deep learning based ecological service anomaly remediation method of claim 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210531037.1A CN115203014A (en) | 2022-05-16 | 2022-05-16 | Ecological service abnormity restoration system and restoration method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210531037.1A CN115203014A (en) | 2022-05-16 | 2022-05-16 | Ecological service abnormity restoration system and restoration method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115203014A true CN115203014A (en) | 2022-10-18 |
Family
ID=83574819
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210531037.1A Pending CN115203014A (en) | 2022-05-16 | 2022-05-16 | Ecological service abnormity restoration system and restoration method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115203014A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116501531A (en) * | 2023-06-19 | 2023-07-28 | 成都移信通科技有限公司 | Software plug-in configuration method and system for monitoring software operation data security |
-
2022
- 2022-05-16 CN CN202210531037.1A patent/CN115203014A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116501531A (en) * | 2023-06-19 | 2023-07-28 | 成都移信通科技有限公司 | Software plug-in configuration method and system for monitoring software operation data security |
| CN116501531B (en) * | 2023-06-19 | 2023-09-08 | 成都移信通科技有限公司 | Software plug-in configuration method and system for monitoring software operation data security |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111241154B (en) | Storage battery fault early warning method and system based on big data | |
| CN108038049B (en) | Real-time log control system and control method, cloud computing system and server | |
| CN111352971A (en) | Bank system monitoring data anomaly detection method and system | |
| CN110333710B (en) | System and method for detecting and processing on-orbit fault of spacecraft | |
| CN105577440A (en) | Network fault time location method and analyzing device | |
| CN110471945B (en) | Active data processing method, system, computer equipment and storage medium | |
| CN114880312B (en) | Flexibly-set application system service data auditing method | |
| CN111126820A (en) | Electricity stealing prevention method and system | |
| CN111666978B (en) | Intelligent fault early warning system for IT system operation and maintenance big data | |
| CN114707401A (en) | Fault early warning method and device for signal system equipment | |
| CN109145319A (en) | Key equipment cutting tool method for predicting residual useful life based on deep neural network | |
| CN110658905A (en) | Early warning method, early warning system and early warning device for equipment running state | |
| CN115203014A (en) | Ecological service abnormity restoration system and restoration method based on deep learning | |
| CN114612011A (en) | Risk prevention and control decision method and device | |
| CN111062827B (en) | Engineering supervision method based on artificial intelligence mode | |
| CN115984158A (en) | Defect analysis method and device, electronic equipment and computer readable storage medium | |
| CN113093695A (en) | Data-driven SDN controller fault diagnosis system | |
| CN117310500A (en) | Battery state classification model construction method and battery state classification method | |
| CN115358348B (en) | Vehicle straight-through rate influence characteristic determination method, device, equipment and medium | |
| CN116308679A (en) | Supply chain abnormal order processing method, device, equipment and storage medium | |
| CN115452376A (en) | Bearing fault diagnosis method based on improved lightweight deep convolution neural network | |
| CN110673843B (en) | Software module behavior state machine restoration method based on data | |
| CN114385472A (en) | Abnormal data detection method, device, equipment and storage medium | |
| CN117667497B (en) | Automatic fault analysis method and system for dispatching centralized system | |
| CN112765034B (en) | Software defect prediction method based on neural network |
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 |