CN118552035A - Safety monitoring method and system for artificial intelligent target identification - Google Patents

Safety monitoring method and system for artificial intelligent target identification Download PDF

Info

Publication number
CN118552035A
CN118552035A CN202410650077.7A CN202410650077A CN118552035A CN 118552035 A CN118552035 A CN 118552035A CN 202410650077 A CN202410650077 A CN 202410650077A CN 118552035 A CN118552035 A CN 118552035A
Authority
CN
China
Prior art keywords
safety key
safety
abnormal
abnormality
monitoring
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
Application number
CN202410650077.7A
Other languages
Chinese (zh)
Inventor
唐俊
衣杨
罗良艺
叶彩仙
周贤来
黄雪敏
吴勇达
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Xinhua College
Original Assignee
Guangzhou Xinhua College
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Xinhua College filed Critical Guangzhou Xinhua College
Priority to CN202410650077.7A priority Critical patent/CN118552035A/en
Publication of CN118552035A publication Critical patent/CN118552035A/en
Pending legal-status Critical Current

Links

Landscapes

  • Testing And Monitoring For Control Systems (AREA)

Abstract

The invention discloses a safety monitoring method and a system for artificial intelligent target identification, which relate to the technical field of safety and comprise the following steps: obtaining a monitoring partition; acquiring at least one safety key parameter in a monitoring range; constructing a risk monitoring index system; obtaining conventional safety key parameters and key safety key parameters; obtaining an abnormal grade of the safety key parameter; determining the check deadline of safety key parameters with abnormality; finishing the checking and the abnormal processing within the checking deadline; dynamically adjusting the checking deadline of the safety key parameters with abnormality; and checking and exception processing are carried out on the safety key parameters with the exception according to the dynamically adjusted checking deadline. By arranging the abnormality evaluation module, the time generation module, the dynamic adjustment module and the abnormality processing module, the priority processing of the emergency abnormality is always guaranteed, the abnormality is prevented from being delayed, and serious production accidents are prevented from being caused.

Description

Safety monitoring method and system for artificial intelligent target identification
Technical Field
The invention relates to the technical field of safety, in particular to a safety monitoring method and system for artificial intelligent target identification.
Background
By means of tools such as instruments, meters, sensors, detection equipment and the like, the types, the hazard degrees, the ranges and the dynamic changes of dangerous factors and toxic and harmful factors in a production system and an operation environment are known rapidly and accurately, occupational safety is evaluated, safety technologies and facilities are monitored, the effect of safety technical measures is monitored, reliable and accurate information is provided for improving labor operation conditions, improving production process, controlling accidents (faults) of the system or the equipment, and all the operation processes are called safety monitoring and monitoring technologies.
The existing safety monitoring generally processes single abnormal conditions, considers the concurrent multiple abnormal conditions, and performs indiscriminate processing during processing, so that urgent abnormality is easily delayed to be processed, and serious production accidents can be caused.
Disclosure of Invention
In order to solve the technical problems, the technical scheme provides a safety monitoring method and a system for artificial intelligent target recognition, which solve the problems that the existing safety monitoring provided in the background art generally processes single abnormal conditions, and the concurrent multiple abnormal conditions are not considered, and indiscriminate processing is carried out during processing, so that more urgent abnormal conditions are delayed to be processed, and more serious production accidents are possibly caused.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a security monitoring method for artificial intelligence object recognition, comprising:
Partitioning the monitoring range to obtain at least one monitoring partition, and simultaneously monitoring in different monitoring partitions during monitoring;
Acquiring at least one safety key parameter in a monitoring range based on operation history data of the monitoring range;
Constructing a risk monitoring index system, and evaluating safety key parameters by the risk monitoring index system;
Classifying the safety key parameters to obtain conventional safety key parameters and key safety key parameters;
collecting real-time data of safety key parameters, and reporting the data collected in real time;
When the real-time data of at least one safety key parameter is abnormal, carrying out risk early warning, and evaluating the abnormal safety key parameter by the risk early warning to obtain an abnormal grade of the safety key parameter;
Sorting the safety key parameters with the abnormality according to the abnormality level of the safety key parameters, and determining the checking deadline of the safety key parameters with the abnormality;
Finishing the checking and the abnormal processing within the checking deadline;
In the checking process, the safety key parameters are detected in real time, and the checking deadline of the safety key parameters with abnormality is dynamically adjusted;
Checking and exception processing are carried out on the safety key parameters with exceptions according to the dynamically adjusted checking deadline;
After the checking and abnormal processing are finished, real-time processing feedback is carried out;
The distributed database technology is used for storing data generated in the checking process, and the data generated in the checking process are used for periodic checking of potential safety hazards.
Preferably, the acquiring the at least one safety key parameter in the monitoring range based on the operation history data of the monitoring range includes the following steps:
Classifying the operation history data in the monitoring range to obtain at least one data classification;
Judging whether the historical data in the data classification causes safety abnormality, if not, not performing any processing, and if so, taking the monitoring item corresponding to the data classification as a safety key parameter.
Preferably, the construction of the risk monitoring index system includes the following steps:
acquiring data causing safety abnormality in data classification corresponding to the safety key parameters as abnormal data;
Acquiring the distribution range of the abnormal data, and equally dividing the distribution range of the abnormal data to obtain at least one abnormal representation point;
Numbering the abnormal representation points in the order from small to large, wherein the grades of the abnormal representation points are the numbers;
And summarizing the safety key parameters and the numerical value of at least one abnormal representation point corresponding to the safety key parameters to obtain a risk monitoring index system.
Preferably, the classifying the safety critical parameters to obtain the conventional safety critical parameters and the key safety critical parameters includes the following steps:
Acquiring an anomaly representing point with the number of 1 corresponding to the safety key parameter as a characteristic anomaly representing point;
And when the numerical value of the safety key parameter reaches the characteristic abnormal representation point, evaluating the safety abnormal condition, if the safety abnormal condition exceeds a preset critical value, taking the safety key parameter as a key safety key parameter, and if not, taking the safety key parameter as a conventional safety key parameter.
Preferably, the risk early warning evaluates an abnormal safety key parameter, and obtaining an abnormal level of the safety key parameter includes the following steps:
Acquiring a first abnormality representing point and a second abnormality representing point from at least one abnormality representing point corresponding to the abnormal safety key parameter, wherein the value of the first abnormality representing point is smaller than that of the second abnormality representing point, and real-time data of the abnormal safety key parameter is between the value of the first abnormality representing point and the value of the second abnormality representing point;
And assigning the grade of the first abnormal representation point to the safety key parameter to obtain the abnormal grade of the safety key parameter.
Preferably, the sorting the safety critical parameters with the abnormality, determining the check deadline of the safety critical parameters with the abnormality includes the following steps:
The ordering follows the principle that: when the two abnormal safety key parameters are the conventional safety key parameters or key safety key parameters, sorting according to the abnormal grades of the abnormal safety key parameters from large to small;
When the two abnormal safety key parameters are the conventional safety key parameter and the key safety key parameter respectively, the abnormal safety key parameter serving as the conventional safety key parameter is ranked on the abnormal safety key parameter serving as the key safety key parameter;
after the sequencing is completed, obtaining the sequence of the abnormal safety key parameters;
And distributing check deadlines for the abnormal safety key parameters according to the sequence order of the abnormal safety key parameters, wherein the check deadline at the front of the sequence is smaller than the check deadline at the back of the sequence.
Preferably, the dynamically adjusting the check deadline of the safety key parameter with the abnormality comprises the following steps:
Detecting the safety key parameters in real time, and updating the abnormal level of the safety key parameters with abnormality by using the updated data;
and reordering the safety key parameters with the abnormality according to the updated abnormality level, and reassigning the check deadline.
Preferably, the real-time treatment feedback after the checking and the abnormal treatment are finished includes the following steps:
after checking and processing the monitoring items corresponding to the safety key parameters with the abnormality are finished, re-detecting the safety key parameters with the abnormality to obtain the repair values of the safety key parameters with the abnormality;
Judging whether the repair value of the abnormal safety key parameter is smaller than the value of the abnormal representation point with the number of 1 corresponding to the abnormal safety key parameter, if so, repairing the monitoring item corresponding to the abnormal safety key parameter, and if not, sending out early warning.
Preferably, the storing the data generated during the checking process using the distributed database technology includes the steps of:
dividing and slicing data generated in the checking process, and uniformly storing the data on a plurality of distributed database nodes;
Setting data replication and redundancy backup strategies in the distributed database, and replicating data to a plurality of distributed database nodes by adopting a master-slave replication or multi-master replication mode;
adopting a distributed transaction processing technology to realize data consistency and synchronization in a distributed database;
A data fragment routing method is adopted, and load balancing and performance optimization strategies are implemented in a distributed database;
in the distributed database, disaster recovery and failure recovery mechanisms including failure detection and automatic switching are set.
A safety monitoring system for artificial intelligent object recognition, which is used for realizing the safety monitoring method for artificial intelligent object recognition, comprising the following steps:
The partition monitoring module is used for partitioning the monitoring range to obtain at least one monitoring partition, and monitoring is carried out in different monitoring partitions simultaneously during monitoring;
The parameter acquisition module is used for acquiring at least one safety key parameter in the monitoring range based on the operation history data of the monitoring range;
the index construction module is used for constructing a risk monitoring index system;
the parameter classification module classifies the safety key parameters to obtain conventional safety key parameters and key safety key parameters;
the abnormality evaluation module performs risk early warning and evaluates abnormal safety key parameters to obtain abnormal grades of the safety key parameters;
the time generation module is used for sequencing the safety key parameters with the abnormality and determining the check deadline of the safety key parameters with the abnormality;
The dynamic adjustment module is used for detecting the safety key parameters in real time and dynamically adjusting the check deadline of the safety key parameters with abnormality;
The abnormality processing module is used for checking and processing the safety key parameters with the abnormality according to the dynamically adjusted checking deadline;
The feedback module is used for carrying out real-time treatment feedback after the checking and abnormal processing are finished;
and the distributed storage module is used for storing data generated in the checking process by using a distributed database technology.
Compared with the prior art, the invention has the beneficial effects that:
By arranging the abnormality evaluation module, the time generation module, the dynamic adjustment module and the abnormality processing module, multiple concurrent abnormal conditions are synchronously considered, different abnormal conditions are classified and considered during processing, the abnormal conditions are further processed according to the classified conditions, the serious abnormal conditions can be guaranteed to be processed preferentially, in the processing process, the abnormal conditions are monitored in real time and are dynamically ordered, and according to the ordered conditions updated in real time, check deadlines are allocated to the abnormal conditions which are not processed yet again, so that the urgent abnormal conditions are always guaranteed to be processed preferentially, the delayed processing of the abnormal conditions is avoided, and serious production accidents are avoided.
Drawings
FIG. 1 is a schematic flow chart of a safety monitoring method for artificial intelligent object recognition according to the present invention;
FIG. 2 is a schematic flow chart of acquiring at least one safety key parameter in a monitoring range based on operation history data in the monitoring range;
FIG. 3 is a schematic flow chart of a risk monitoring index system constructed according to the present invention;
FIG. 4 is a flow chart of classifying safety critical parameters to obtain conventional safety critical parameters and key safety critical parameters according to the present invention;
FIG. 5 is a schematic flow chart of an abnormal level of the safety critical parameter obtained by evaluating the safety critical parameter of the abnormality in the risk early warning of the present invention;
FIG. 6 is a schematic diagram of a checking deadline flow chart for sorting security critical parameters with anomalies and determining the security critical parameters with anomalies according to the present invention;
FIG. 7 is a schematic flow chart of dynamically adjusting the check deadline of the safety critical parameter with abnormality according to the present invention;
Fig. 8 is a schematic diagram of a real-time processing feedback flow after the completion of checking and exception handling in accordance with the present invention.
Detailed Description
The following description is presented to enable one of ordinary skill in the art to make and use the invention. The preferred embodiments in the following description are by way of example only and other obvious variations will occur to those skilled in the art.
Referring to fig. 1, a security monitoring method for artificial intelligence object recognition includes:
Partitioning the monitoring range to obtain at least one monitoring partition, and simultaneously monitoring in different monitoring partitions during monitoring;
Acquiring at least one safety key parameter in a monitoring range based on operation history data of the monitoring range;
Constructing a risk monitoring index system, and evaluating safety key parameters by the risk monitoring index system;
Classifying the safety key parameters to obtain conventional safety key parameters and key safety key parameters;
collecting real-time data of safety key parameters, and reporting the data collected in real time;
When the real-time data of at least one safety key parameter is abnormal, carrying out risk early warning, and evaluating the abnormal safety key parameter by the risk early warning to obtain an abnormal grade of the safety key parameter;
Sorting the safety key parameters with the abnormality according to the abnormality level of the safety key parameters, and determining the checking deadline of the safety key parameters with the abnormality;
Finishing the checking and the abnormal processing within the checking deadline;
In the checking process, the safety key parameters are detected in real time, and the checking deadline of the safety key parameters with abnormality is dynamically adjusted;
Checking and exception processing are carried out on the safety key parameters with exceptions according to the dynamically adjusted checking deadline;
After the checking and abnormal processing are finished, real-time processing feedback is carried out;
The distributed database technology is used for storing data generated in the checking process, and the data generated in the checking process are used for periodic checking of potential safety hazards.
Referring to fig. 2, based on the operation history data of the monitoring range, acquiring at least one safety-critical parameter in the monitoring range includes the steps of:
Classifying the operation history data in the monitoring range to obtain at least one data classification;
Judging whether the historical data in the data classification causes safety abnormality, if not, not performing any processing, and if so, taking the monitoring item corresponding to the data classification as a safety key parameter.
The historical data does not have an influence on the safety, so that the data which has little influence on the safety needs to be removed, and then the monitoring items corresponding to the data which has influence on the safety are selected as the safety key parameters, the method for selecting the data is to judge whether the data in the data classification causes safety abnormality or not, and the data classification has enough data, so that when the corresponding monitoring item can influence the safety, the safety is influenced when the monitoring item takes the value of part of the data in the data classification.
Referring to fig. 3, constructing a risk monitoring index system includes the steps of:
acquiring data causing safety abnormality in data classification corresponding to the safety key parameters as abnormal data;
Acquiring the distribution range of the abnormal data, and equally dividing the distribution range of the abnormal data to obtain at least one abnormal representation point;
Numbering the abnormal representation points in the order from small to large, wherein the grades of the abnormal representation points are the numbers;
And summarizing the safety key parameters and the numerical value of at least one abnormal representation point corresponding to the safety key parameters to obtain a risk monitoring index system.
The risk monitoring index system aims to evaluate the real-time data of the safety key parameters and judge whether the safety key parameters are abnormal or not and the degree of the abnormality, so that the abnormality representing points are numbered in the order from small to large, the level of the abnormality representing points is the number of the abnormality representing points, and when the real-time data of the safety key parameters exceeds the corresponding value of the abnormality representing points with the number of 1, the abnormality is indicated.
Referring to fig. 4, classifying the safety-critical parameters to obtain the conventional safety-critical parameters and the key safety-critical parameters includes the steps of:
Acquiring an anomaly representing point with the number of 1 corresponding to the safety key parameter as a characteristic anomaly representing point;
And when the numerical value of the safety key parameter reaches the characteristic abnormal representation point, evaluating the safety abnormal condition, if the safety abnormal condition exceeds a preset critical value, taking the safety key parameter as a key safety key parameter, and if not, taking the safety key parameter as a conventional safety key parameter.
The criterion for classifying the safety critical parameters is to evaluate the safety influence degree according to the safety critical parameters, so that the condition consistency needs to be ensured during evaluation, namely the abnormal degree of the safety critical parameters needs to be kept consistent, otherwise, the judgment cannot be performed, and the abnormal degree of all abnormal safety critical parameters is set to be 1 level.
Referring to fig. 5, the risk early warning evaluates an abnormal safety critical parameter, and obtaining an abnormal level of the safety critical parameter includes the following steps:
Acquiring a first abnormality representing point and a second abnormality representing point from at least one abnormality representing point corresponding to the abnormal safety key parameter, wherein the value of the first abnormality representing point is smaller than that of the second abnormality representing point, and real-time data of the abnormal safety key parameter is between the value of the first abnormality representing point and the value of the second abnormality representing point;
And assigning the grade of the first abnormal representation point to the safety key parameter to obtain the abnormal grade of the safety key parameter.
The evaluation of the level of the safety-critical parameter can evaluate the degree of abnormality of the safety-critical parameter, and the processing time of the monitoring item corresponding to the safety-critical parameter can be determined according to the degree of abnormality of the safety-critical parameter, because the number of operators for processing is limited, when the concurrent abnormality is more, the waiting situation is necessarily occurred, and therefore, the emergency abnormality needs to be pre-solved, but the abnormality which is not particularly important is post-arranged.
Referring to fig. 6, sorting safety-critical parameters in which an abnormality exists, determining a check deadline of the safety-critical parameters in which an abnormality exists includes the steps of:
The ordering follows the principle that: when the two abnormal safety key parameters are the conventional safety key parameters or key safety key parameters, sorting according to the abnormal grades of the abnormal safety key parameters from large to small;
When the two abnormal safety key parameters are the conventional safety key parameter and the key safety key parameter respectively, the abnormal safety key parameter serving as the conventional safety key parameter is ranked on the abnormal safety key parameter serving as the key safety key parameter;
after the sequencing is completed, obtaining the sequence of the abnormal safety key parameters;
And distributing check deadlines for the abnormal safety key parameters according to the sequence order of the abnormal safety key parameters, wherein the check deadline at the front of the sequence is smaller than the check deadline at the back of the sequence.
In the sorting, the situation of the conventional safety critical parameter and the key safety critical parameter is also considered, the safety critical parameter may be the conventional safety critical parameter or the key safety critical parameter, when the safety critical parameter is the same parameter, the sorting is performed according to the grade, but when the safety critical parameter is the conventional safety critical parameter and the key safety critical parameter, the respective grade is ignored, because the abnormality caused by the key safety critical parameter with low abnormality degree is more serious than the conventional safety critical parameter with high abnormality degree.
Referring to fig. 7, the dynamic adjustment of the check deadline of the safety critical parameter for which abnormality exists includes the steps of:
Detecting the safety key parameters in real time, and updating the abnormal level of the safety key parameters with abnormality by using the updated data;
and reordering the safety key parameters with the abnormality according to the updated abnormality level, and reassigning the check deadline.
The purpose of the dynamic adjustment is that the abnormal situation has burstiness and can suddenly appear in a serious abnormal situation, so that the serious abnormal situation must be solved preferentially, and the existing abnormal situation cannot be solved in a stiff way, and then the serious abnormal situation is solved.
Referring to fig. 8, after the process of checking and exception is completed, performing real-time treatment feedback includes the steps of:
after checking and processing the monitoring items corresponding to the safety key parameters with the abnormality are finished, re-detecting the safety key parameters with the abnormality to obtain the repair values of the safety key parameters with the abnormality;
Judging whether the repair value of the abnormal safety key parameter is smaller than the value of the abnormal representation point with the number of 1 corresponding to the abnormal safety key parameter, if so, repairing the monitoring item corresponding to the abnormal safety key parameter, and if not, sending out early warning.
The reason for the real-time treatment feedback is that not all the treatments solve the abnormal situation, so that secondary detection is needed to ensure that the abnormality is solved, and when the abnormality is not solved, early warning is carried out, and an operator can treat the abnormality after receiving the early warning.
Storing data generated during the verification process using a distributed database technique includes the steps of:
dividing and slicing data generated in the checking process, and uniformly storing the data on a plurality of distributed database nodes;
Setting data replication and redundancy backup strategies in the distributed database, and replicating data to a plurality of distributed database nodes by adopting a master-slave replication or multi-master replication mode;
adopting a distributed transaction processing technology to realize data consistency and synchronization in a distributed database;
A data fragment routing method is adopted, and load balancing and performance optimization strategies are implemented in a distributed database;
in the distributed database, disaster recovery and failure recovery mechanisms including failure detection and automatic switching are set.
A safety monitoring system for artificial intelligent object recognition, which is used for realizing the safety monitoring method for artificial intelligent object recognition, comprising the following steps:
The partition monitoring module is used for partitioning the monitoring range to obtain at least one monitoring partition, and monitoring is carried out in different monitoring partitions simultaneously during monitoring;
The parameter acquisition module is used for acquiring at least one safety key parameter in the monitoring range based on the operation history data of the monitoring range;
the index construction module is used for constructing a risk monitoring index system;
the parameter classification module classifies the safety key parameters to obtain conventional safety key parameters and key safety key parameters;
the abnormality evaluation module performs risk early warning and evaluates abnormal safety key parameters to obtain abnormal grades of the safety key parameters;
the time generation module is used for sequencing the safety key parameters with the abnormality and determining the check deadline of the safety key parameters with the abnormality;
The dynamic adjustment module is used for detecting the safety key parameters in real time and dynamically adjusting the check deadline of the safety key parameters with abnormality;
The abnormality processing module is used for checking and processing the safety key parameters with the abnormality according to the dynamically adjusted checking deadline;
The feedback module is used for carrying out real-time treatment feedback after the checking and abnormal processing are finished;
and the distributed storage module is used for storing data generated in the checking process by using a distributed database technology.
Still further, the present disclosure also provides a storage medium having a computer readable program stored thereon, the computer readable program when invoked performing the above-described security monitoring method for artificial intelligence object recognition.
It is understood that the storage medium may be a magnetic medium, e.g., floppy disk, hard disk, magnetic tape; optical media such as DVD; or a semiconductor medium such as a solid state disk SolidStateDisk, SSD, etc.
In summary, the invention has the advantages that: by arranging the abnormality evaluation module, the time generation module, the dynamic adjustment module and the abnormality processing module, multiple concurrent abnormal conditions are synchronously considered, different abnormal conditions are classified and considered during processing, the abnormal conditions are further processed according to the classified conditions, the serious abnormal conditions can be guaranteed to be processed preferentially, in the processing process, the abnormal conditions are monitored in real time and are dynamically ordered, and according to the ordered conditions updated in real time, check deadlines are allocated to the abnormal conditions which are not processed yet again, so that the urgent abnormal conditions are always guaranteed to be processed preferentially, the delayed processing of the abnormal conditions is avoided, and serious production accidents are avoided.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made therein without departing from the spirit and scope of the invention, which is defined by the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (10)

1. A security monitoring method for artificial intelligence object recognition, comprising:
Partitioning the monitoring range to obtain at least one monitoring partition, and simultaneously monitoring in different monitoring partitions during monitoring;
Acquiring at least one safety key parameter in a monitoring range based on operation history data of the monitoring range;
Constructing a risk monitoring index system, and evaluating safety key parameters by the risk monitoring index system;
Classifying the safety key parameters to obtain conventional safety key parameters and key safety key parameters;
collecting real-time data of safety key parameters, and reporting the data collected in real time;
When the real-time data of at least one safety key parameter is abnormal, carrying out risk early warning, and evaluating the abnormal safety key parameter by the risk early warning to obtain an abnormal grade of the safety key parameter;
Sorting the safety key parameters with the abnormality according to the abnormality level of the safety key parameters, and determining the checking deadline of the safety key parameters with the abnormality;
Finishing the checking and the abnormal processing within the checking deadline;
In the checking process, the safety key parameters are detected in real time, and the checking deadline of the safety key parameters with abnormality is dynamically adjusted;
Checking and exception processing are carried out on the safety key parameters with exceptions according to the dynamically adjusted checking deadline;
After the checking and abnormal processing are finished, real-time processing feedback is carried out;
The distributed database technology is used for storing data generated in the checking process, and the data generated in the checking process are used for periodic checking of potential safety hazards.
2. The safety monitoring method for artificial intelligence object recognition according to claim 1, wherein the acquiring at least one safety critical parameter in the monitoring range based on the operation history data of the monitoring range comprises the steps of:
Classifying the operation history data in the monitoring range to obtain at least one data classification;
Judging whether the historical data in the data classification causes safety abnormality, if not, not performing any processing, and if so, taking the monitoring item corresponding to the data classification as a safety key parameter.
3. The method for safety monitoring for artificial intelligence object recognition according to claim 2, wherein the constructing a risk monitoring index system comprises the steps of:
acquiring data causing safety abnormality in data classification corresponding to the safety key parameters as abnormal data;
Acquiring the distribution range of the abnormal data, and equally dividing the distribution range of the abnormal data to obtain at least one abnormal representation point;
Numbering the abnormal representation points in the order from small to large, wherein the grades of the abnormal representation points are the numbers;
And summarizing the safety key parameters and the numerical value of at least one abnormal representation point corresponding to the safety key parameters to obtain a risk monitoring index system.
4. A safety monitoring method for artificial intelligent object recognition according to claim 3, wherein the classifying the safety critical parameters to obtain the conventional safety critical parameters and the key safety critical parameters comprises the steps of:
Acquiring an anomaly representing point with the number of 1 corresponding to the safety key parameter as a characteristic anomaly representing point;
And when the numerical value of the safety key parameter reaches the characteristic abnormal representation point, evaluating the safety abnormal condition, if the safety abnormal condition exceeds a preset critical value, taking the safety key parameter as a key safety key parameter, and if not, taking the safety key parameter as a conventional safety key parameter.
5. The safety monitoring method for artificial intelligent object recognition according to claim 4, wherein the risk early warning evaluates an abnormal safety key parameter, and the obtaining of an abnormal level of the safety key parameter comprises the following steps:
Acquiring a first abnormality representing point and a second abnormality representing point from at least one abnormality representing point corresponding to the abnormal safety key parameter, wherein the value of the first abnormality representing point is smaller than that of the second abnormality representing point, and real-time data of the abnormal safety key parameter is between the value of the first abnormality representing point and the value of the second abnormality representing point;
And assigning the grade of the first abnormal representation point to the safety key parameter to obtain the abnormal grade of the safety key parameter.
6. The method for security monitoring for artificial intelligence object recognition according to claim 5, wherein the sorting the security critical parameters having an abnormality, determining the check deadline of the security critical parameters having an abnormality comprises the steps of:
The ordering follows the principle that: when the two abnormal safety key parameters are the conventional safety key parameters or key safety key parameters, sorting according to the abnormal grades of the abnormal safety key parameters from large to small;
When the two abnormal safety key parameters are the conventional safety key parameter and the key safety key parameter respectively, the abnormal safety key parameter serving as the conventional safety key parameter is ranked on the abnormal safety key parameter serving as the key safety key parameter;
after the sequencing is completed, obtaining the sequence of the abnormal safety key parameters;
And distributing check deadlines for the abnormal safety key parameters according to the sequence order of the abnormal safety key parameters, wherein the check deadline at the front of the sequence is smaller than the check deadline at the back of the sequence.
7. The method for security monitoring for artificial intelligence object recognition according to claim 6, wherein the dynamically adjusting the check deadline of the security critical parameter having an abnormality comprises the steps of:
Detecting the safety key parameters in real time, and updating the abnormal level of the safety key parameters with abnormality by using the updated data;
and reordering the safety key parameters with the abnormality according to the updated abnormality level, and reassigning the check deadline.
8. The method for safety monitoring for artificial intelligence object recognition according to claim 7, wherein the real-time treatment feedback after the completion of the checking and the abnormal treatment comprises the steps of:
after checking and processing the monitoring items corresponding to the safety key parameters with the abnormality are finished, re-detecting the safety key parameters with the abnormality to obtain the repair values of the safety key parameters with the abnormality;
Judging whether the repair value of the abnormal safety key parameter is smaller than the value of the abnormal representation point with the number of 1 corresponding to the abnormal safety key parameter, if so, repairing the monitoring item corresponding to the abnormal safety key parameter, and if not, sending out early warning.
9. The method of claim 8, wherein the step of storing data generated during verification using distributed database techniques comprises the steps of:
dividing and slicing data generated in the checking process, and uniformly storing the data on a plurality of distributed database nodes;
Setting data replication and redundancy backup strategies in the distributed database, and replicating data to a plurality of distributed database nodes by adopting a master-slave replication or multi-master replication mode;
adopting a distributed transaction processing technology to realize data consistency and synchronization in a distributed database;
A data fragment routing method is adopted, and load balancing and performance optimization strategies are implemented in a distributed database;
in the distributed database, disaster recovery and failure recovery mechanisms including failure detection and automatic switching are set.
10. A security monitoring system for artificial intelligence object recognition, for implementing a security monitoring method for artificial intelligence object recognition according to any one of claims 1 to 9, comprising:
The partition monitoring module is used for partitioning the monitoring range to obtain at least one monitoring partition, and monitoring is carried out in different monitoring partitions simultaneously during monitoring;
The parameter acquisition module is used for acquiring at least one safety key parameter in the monitoring range based on the operation history data of the monitoring range;
the index construction module is used for constructing a risk monitoring index system;
the parameter classification module classifies the safety key parameters to obtain conventional safety key parameters and key safety key parameters;
the abnormality evaluation module performs risk early warning and evaluates abnormal safety key parameters to obtain abnormal grades of the safety key parameters;
the time generation module is used for sequencing the safety key parameters with the abnormality and determining the check deadline of the safety key parameters with the abnormality;
The dynamic adjustment module is used for detecting the safety key parameters in real time and dynamically adjusting the check deadline of the safety key parameters with abnormality;
The abnormality processing module is used for checking and processing the safety key parameters with the abnormality according to the dynamically adjusted checking deadline;
The feedback module is used for carrying out real-time treatment feedback after the checking and abnormal processing are finished;
and the distributed storage module is used for storing data generated in the checking process by using a distributed database technology.
CN202410650077.7A 2024-05-24 2024-05-24 Safety monitoring method and system for artificial intelligent target identification Pending CN118552035A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410650077.7A CN118552035A (en) 2024-05-24 2024-05-24 Safety monitoring method and system for artificial intelligent target identification

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410650077.7A CN118552035A (en) 2024-05-24 2024-05-24 Safety monitoring method and system for artificial intelligent target identification

Publications (1)

Publication Number Publication Date
CN118552035A true CN118552035A (en) 2024-08-27

Family

ID=92443555

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410650077.7A Pending CN118552035A (en) 2024-05-24 2024-05-24 Safety monitoring method and system for artificial intelligent target identification

Country Status (1)

Country Link
CN (1) CN118552035A (en)

Similar Documents

Publication Publication Date Title
CN111209131B (en) Method and system for determining faults of heterogeneous system based on machine learning
KR101856543B1 (en) Failure prediction system based on artificial intelligence
WO2021179574A1 (en) Root cause localization method, device, computer apparatus, and storage medium
EP3105644B1 (en) Method of identifying anomalies
CN110704231A (en) Fault processing method and device
EP1810091A2 (en) Method, system and storage medium for managing automated system events
US10775270B2 (en) System and method for monitoring a turbomachine, with indicator merging for the synthesis of an alarm confirmation
CN112785131A (en) Optimization method for preventive maintenance task of airplane
CN112286771A (en) Alarm method for monitoring global resources
CN117310422B (en) Initiating explosive device resistor performance test method and system
WO2019019429A1 (en) Anomaly detection method, device and apparatus for virtual machine, and storage medium
CN117041029A (en) Network equipment fault processing method and device, electronic equipment and storage medium
CN110705787A (en) Aircraft maintenance task optimization method
CN115840915A (en) Automatic identification method, system, terminal and storage medium for electromechanical equipment fault
CN113723637B (en) Economic repair level analysis method and system for ship repair system
CN108122061A (en) Aircraft equipment software reusage degree stage division based on Danger Indexes index matrix
CN117785538A (en) Operation and maintenance system fault positioning method based on multi-model fusion
CN113553352A (en) Method and system for identifying familial defects of power secondary equipment
CN117313012A (en) Fault management method, device, equipment and storage medium of service orchestration system
CN118552035A (en) Safety monitoring method and system for artificial intelligent target identification
CN112819262A (en) Memory, process pipeline inspection and maintenance decision method, device and equipment
JP7499168B2 (en) Cause estimation system and cause estimation method
CN114139740A (en) Method for evaluating average repair time and/or maximum repair time
CN110703730A (en) Data acquisition testing method and device based on MES
CN117668677B (en) Data detection method

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