Vector database automatic operation and maintenance method and system based on artificial intelligence

Abstract

The invention discloses a vector database automatic operation and maintenance method and system based on artificial intelligence, and relates to the technical field of automatic operation and maintenance. Equipment operation states are analyzed, feature collection is carried out on abnormal operation equipment, and the similarity degree of the current equipment operation state and the historical fault equipment operation state is analyzed; the similar historical fault records are stored in a temporary comparison library, fault types are judged according to maintenance records, then predicted fault types and the number of predicted fault type records are obtained, the predicted fault types are analyzed, the predicted fault types are analyzed through the fault records and current characteristics, the predicted fault types are ranked, and the fault types are obtained. According to the method and the system, the complexity and the uncertainty of manual operation are reduced by matching historical fault records, and the troubleshooting time is shortened.

Description

Technical Field

[0001] This invention relates to the field of automated operation and maintenance technology, specifically to an automated operation and maintenance method and system for vector databases based on artificial intelligence. Background Technology

[0002] Traditional equipment maintenance relies heavily on manual inspections, the effectiveness of which is limited by personnel experience and energy. During inspections, visual fatigue and distraction can easily lead to missed anomalies, especially in complex equipment systems where comprehensive coverage is difficult. Furthermore, manual inspections are limited by weather conditions and environmental risks, often forced to stop in severe weather or high-risk scenarios, resulting in the failure to detect potential hazards in a timely manner. In addition, manually recorded data is prone to omissions and cannot be synchronized in real time, creating information silos and causing delayed equipment anomaly warnings. Small faults can easily escalate into major problems, threatening the continuous and stable operation of equipment. Current fault handling largely relies on maintenance personnel manually comparing historical records, lacking systematic feature analysis and intelligent matching mechanisms. During troubleshooting, maintenance personnel must exhaustively explore possible paths based on experience, easily getting bogged down in limited diagnostic directions, leading to lengthy and time-consuming troubleshooting processes. Moreover, subjective differences in human judgment can easily lead to misjudgments or ineffective repair attempts, resulting in a waste of human and material resources. Especially in complex equipment failure scenarios, traditional methods struggle to quickly pinpoint the root cause, often leading to prolonged equipment downtime due to repeated testing, significantly increasing maintenance costs. Traditional maintenance solutions are often designed specifically for particular equipment, lacking universal adaptability and failing to address the operational characteristics and failure modes of different equipment types. Furthermore, fault records and maintenance data are often scattered across paper documents or isolated systems, lacking a unified database. This data silo phenomenon prevents the effective accumulation of historical experience, hinders the application of past cases to new fault handling, and deprives maintenance models of iterative optimization. In the long run, existing solutions cannot adapt to the dynamic changes in equipment operating status, making it difficult to achieve continuous improvement in maintenance capabilities. Summary of the Invention

[0003] The purpose of this invention is to provide an automated operation and maintenance method and system for vector databases based on artificial intelligence, so as to solve the problems mentioned in the background art.

[0004] To address the aforementioned technical problems, this invention provides the following technical solution: an automated operation and maintenance method for vector databases based on artificial intelligence, comprising the following steps: S1. Real-time monitoring of equipment operation data, analysis of equipment operation status, and feature collection of equipment with abnormal operation; S2. Collect features from historical fault records and abnormal equipment, and analyze the similarity between the current equipment operating status and the operating status of historical faulty equipment. S3. Store similar historical fault records in a temporary comparison database, determine the fault type based on the maintenance records, and then obtain the predicted fault type and the number of predicted fault type records. S4. Call the equipment's factory data, analyze and predict the fault type, and analyze the predicted fault type through fault records and current characteristics; S5. Sort the predicted fault types, and then check the fault types in the predicted fault type sequence in order. S6. After completing the troubleshooting, update the database.

[0005] Furthermore, in step S1, for any type of equipment, equipment operation data is collected, and after preprocessing, equipment operation characteristics are obtained. The current equipment operation status is monitored. In one monitoring cycle, the product output of the equipment is α. If α / β < γ, the current equipment operation status is judged to be abnormal, where β is the predetermined standard product output per unit monitoring cycle, and γ is the predetermined allowable fluctuation percentage of the ratio of actual product output to standard product output. Otherwise, the current equipment operation status is judged to be normal. When the current equipment operation status is judged to be abnormal, the equipment operation characteristics are called as equipment operation anomaly vectors and stored in the operation anomaly vector database, and the equipment tags of the equipment to be processed are marked. The dimension of the equipment operation anomaly vector is the same as the number of equipment operation characteristics. By collecting equipment operation data and preprocessing to extract operation characteristics, and combining the allowable fluctuation range of the ratio of actual to standard product output to determine the status, compared with the traditional fuzzy judgment relying on experience, the accuracy of equipment operation status judgment is improved, and the situation of misjudgment or omission is reduced. When the equipment is abnormal, the operation characteristics can be stored as anomaly vectors in the database and marked with tags to be processed, and the key data at the time of the anomaly can be completely retained to avoid information loss and provide an accurate basis for subsequent fault matching and analysis. Real-time monitoring and timely anomaly identification can quickly incorporate abnormal equipment into the processing flow, preventing minor anomalies from escalating into major failures, effectively ensuring the continuous and stable operation of equipment, and reducing losses caused by the expansion of failures.

[0006] Furthermore, in step S2, when a device tag to be processed appears in the running anomaly vector database, the device running feature corresponding to the device tag to be processed is called as the current running feature, where the current running feature is {A1, A2, ..., A...}. n ,…,A N}, where N represents the number of equipment operating characteristics, A n This represents the nth current operating characteristic. It monitors equipment faults and retrieves historical fault records. The number of historical fault records is M, and the equipment operating characteristic corresponding to the mth fault record is {B1, B2, ..., B...}. n ,…,B N}, B nThis represents the operating feature of the nth device in the mth fault record. Then, considering the nth current operating feature and the operating feature of the nth device, we obtain the similarity Z between the nth feature of the current device and the device in the mth fault record. m_n The dissimilarity Z of the nth feature of the mth fault record m_n =|(A n -B n ) / B n Substitute each of the following into n=1,2,…,N to obtain the N feature dissimilarity between the current device and the device in the m-th fault record, and then obtain the device dissimilarity C between the current device and the device in the m-th fault record. m C m Let C be the average of the dissimilarity of the current device and the device with the m-th fault record among the N devices. m If the value is greater than C0, the system determines that the current device's operating characteristics are dissimilar to those of the m-th fault record, where C0 represents a pre-defined threshold for device dissimilarity. Otherwise, the system determines that the current device's operating characteristics are similar to those of the m-th fault record. By comparing the operating characteristics of the current abnormal device with those of historical fault records, and using feature dissimilarity and average dissimilarity as the basis, combined with a preset threshold, the system effectively avoids the subjectivity and bias of traditional manual comparisons, improving the objectivity and accuracy of similarity judgments. This comprehensive approach, considering the differences in various operating characteristics, can accurately filter out fault cases similar to the current device's state from a large number of historical records, providing a reliable reference for subsequent fault type judgment, reducing blind investigations, significantly improving the efficiency and targeting of fault analysis, and laying the foundation for quickly pinpointing the fault direction.

[0007] Furthermore, in step S3, when it is determined that the operating characteristics of the current device are similar to those of the m-th fault record, the m-th fault record is recorded in the temporary comparison database of the current device, and m=1,2,…,M is substituted one by one. Then, X fault records are recorded in the temporary comparison database of the current device, and X fault records are called as the comparison fault records of the current device. Then, the maintenance records of the X fault records are called. If the maintenance records of the X fault records belong to the same fault type V after monitoring and maintenance, then the fault type of the current device is determined to be fault V. If the maintenance records of the X fault records do not belong to the same fault type after monitoring and maintenance, then the fault type corresponding to the X fault records is called as the predicted fault type, and the predicted fault type is {D1,D2,…,D…}. y ,…,D Y}, where Y represents the number of predicted fault types, D yThe number of fault records corresponding to the predicted fault types is obtained by retrieving the temporary comparison database. The number of fault records corresponding to the Y predicted fault types in the temporary comparison database is {d1,d2,…,dy,…,dY}, where dy is the predicted fault type D. y The temporary comparison database contains the number of corresponding fault records. By centrally storing historical fault records with similar characteristics to the current equipment in the temporary comparison database, and automatically analyzing fault types based on maintenance records, the fault diagnosis process is effectively simplified. When similar records belong to the same fault type, the fault type of the current equipment can be directly determined, avoiding the tedious manual comparison. When the types are different, the predicted fault type and the corresponding number of records are automatically extracted, providing a clear direction for subsequent troubleshooting. This approach reduces the subjective bias of manual judgment, ensures a more systematic and comprehensive fault type analysis, and lays a data foundation for accurate troubleshooting, improving the efficiency and reliability of fault diagnosis.

[0008] Furthermore, in step S4, the normal operating characteristics of the device at the time of factory shipment are invoked, denoted as {E1, E2, ..., E n ,…,E N}, where E n This represents the nth normal operation characteristic. For the yth predicted fault type, analysis is performed by calling the dy-th fault records of the y-th predicted fault type from the temporary comparison database. The operating characteristic of the nth device among the dy-th fault records is then obtained as {F}. 1_n ,F 2_n ,…,F w_n ,…,F dy_n}, where F w_n Let be the operating feature of the nth device in the wth fault record, where w = 1, 2, ..., dy. Then, the similarity J between the operating feature of the nth device in the wth fault record and the operating feature of the nth normal operation record can be obtained. w_n =1-|(F w_n -E n ) / E n |, if J w_n If J0 >, determine that the operating characteristics of the nth device in the wth fault record are different from the operating characteristics of the nth normal operation record; otherwise, determine that the operating characteristics of the nth device in the wth fault record are similar to the operating characteristics of the nth normal operation record. Substitute each value of w = 1, 2, ..., dy into the list of fault records to obtain the similarity between the nth device's operating feature and the nth normal operation feature. Then, obtain the number L of dissimilar features between the nth device's operating feature and the nth normal operation feature in the dy fault records. If L / dy ≥ k0, then determine that the nth device's operating feature is an identifier feature of the dy fault records, where k0 is a pre-set identifier feature judgment threshold. Otherwise, determine that the nth device's operating feature is not an identifier feature of the dy fault records, and obtain R identifier features of the dy fault records. If R = 0, then determine that the dy fault records have no identifier features, and predict the fault type D. y If the fault is classified as random, then the R current operating features corresponding to the identifier features of the dy fault records are called. These R current operating features are used as judgment features, and are labeled as {a1, a2, ..., a...}. r ,…,a R}, where a r Let represent the r-th judgment feature, and obtain the similarity {j1,j2,…,j} of the R judgment features to the normal operation feature. r ,…,j R}, where j r Let j be the similarity between the r-th judgment feature and the normal operation feature. r =1-|(a r -E r ) / E r |,E r For the normal operation feature in the normal operation feature set corresponding to the r-th judgment feature, if j r If the value of the r-th judgment feature is greater than J0, then the r-th judgment feature is determined to conform to the y-th predicted fault type; otherwise, the r-th judgment feature is determined to not conform to the y-th predicted fault type. This leads to the fault judgment priority for the y-th predicted fault type, which is the ratio of the number of features conforming to the y-th predicted fault type to the number of judgment features. Based on the normal operating characteristics of the equipment at the factory, and through comparative analysis of the historical records of predicted fault types, identifying features are accurately selected, avoiding interference from irrelevant features in fault judgment. Fault priority is determined by judging the degree of conformity between the current operating characteristics and the identified features, making fault analysis more precise than it appears. This method ensures the uniformity of the judgment benchmark by relying on factory standards, and focuses on key differences through identified features, reducing invalid analysis and providing a clear priority basis for subsequent fault troubleshooting. This improves the scientific rigor and relevance of fault type judgment, effectively avoiding the confusion caused by complex features in traditional troubleshooting.

[0009] Furthermore, in step S5, the predicted fault types that satisfy R not equal to 0 are sorted from high to low according to the fault judgment priority, and the random fault types that satisfy R equal to 0 are sorted after the predicted fault types that satisfy R not equal to 0, to obtain the predicted fault type sequence, and then the predicted fault type sequence is checked sequentially for fault types.

[0010] Furthermore, in step S6, after completing the sequential fault type investigation of the fault type sequence, the temporary comparison database is deleted, and the current fault information is updated in the database.

[0011] The AI-based automated operation and maintenance system for vector databases includes: a data monitoring and anomaly collection module, a historical fault similarity analysis module, a similarity record and type initial determination module, a data in-depth analysis module, a fault type sorting and troubleshooting module, and a troubleshooting result data update module. The data monitoring and anomaly collection module is used to monitor equipment operation data in real time, analyze equipment operation status, and collect features of equipment with abnormal operation. The historical fault similarity analysis module is used to call historical fault records and abnormal equipment for feature collection, and analyze the degree of similarity between the current equipment operating status and the operating status of historical faulty equipment. The similarity record and type initial determination module is used to store similar historical fault records into a temporary comparison database, determine the fault type based on the maintenance records, and then obtain the predicted fault type and the number of predicted fault type records. The data depth analysis module is used to call the equipment's factory data, analyze and predict fault types, and analyze the predicted fault types through fault records and current characteristics; The fault type sorting and investigation module is used to sort the predicted fault types, and then perform sequential fault type investigation on the predicted fault type sequence. The troubleshooting result data update module is used to update the database after the fault type investigation is completed.

[0012] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: On the one hand, by collecting equipment operation data in real time and preprocessing it to extract features, and combining it with preset standards to accurately determine the operating status, abnormal equipment can be quickly identified and abnormal features can be retained. Compared with the traditional method of relying on manual inspection, it can avoid the delay in the discovery of abnormalities due to human oversight, promptly incorporate abnormal equipment into the handling process, prevent small faults from developing into big problems, effectively ensure the continuous and stable operation of equipment, reduce losses such as production interruptions and increased maintenance costs caused by equipment downtime or the expansion of faults, and provide proactive early warning support for reliable equipment operation; On the one hand, in the fault handling stage, the method matches historical fault records, analyzes feature similarity to filter similar cases, and then combines maintenance records and factory data to deepen the fault type analysis, finally troubleshooting in an orderly manner according to priority. This process eliminates the need for maintenance personnel to manually compare a large number of historical records or conduct blind testing, reducing the complexity and uncertainty of manual operations and shortening the time spent on fault troubleshooting. At the same time, it accurately pinpoints the fault direction based on historical data and feature analysis, avoiding ineffective repair attempts, reducing the waste of human and material resources, significantly reducing maintenance costs, and improving the overall efficiency of maintenance work. On the other hand, after each fault investigation is completed, the database is updated and new fault and maintenance data is accumulated, providing richer reference data for subsequent operation and maintenance. With data accumulation, historical fault matching becomes more accurate, and fault type judgments more closely reflect actual equipment operating conditions, gradually optimizing the operation and maintenance model. Furthermore, this method is adaptable to the operating characteristics and fault modes of different types of equipment, eliminating the need for separate operation and maintenance plans for individual devices, thus possessing strong versatility. In the long term, the data closed loop and universal adaptability enable the system to continuously adapt to changes in equipment operating status, constantly improving the reliability and intelligence of operation and maintenance, providing sustainable support for the long-term stable operation and maintenance of equipment. Attached Figure Description

[0013] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a structural diagram of the AI-based automated operation and maintenance system for vector databases according to the present invention. Figure 2 This is a flowchart of the automated operation and maintenance method for vector databases based on artificial intelligence, as described in this invention. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please see Figure 1 and Figure 2 This invention provides a technical solution: an automated operation and maintenance method for vector databases based on artificial intelligence, comprising the following steps: S1. Real-time monitoring of equipment operation data, analysis of equipment operation status, and feature collection of equipment with abnormal operation; S2. Collect features from historical fault records and abnormal equipment, and analyze the similarity between the current equipment operating status and the operating status of historical faulty equipment. S3. Store similar historical fault records in a temporary comparison database, determine the fault type based on the maintenance records, and then obtain the predicted fault type and the number of predicted fault type records. S4. Call the equipment's factory data, analyze and predict the fault type, and analyze the predicted fault type through fault records and current characteristics; S5. Sort the predicted fault types, and then check the fault types in the predicted fault type sequence in order. S6. After completing the troubleshooting, update the database.

[0016] In step S1, for any type of equipment, equipment operation data is collected, and after preprocessing, equipment operation characteristics are obtained. The current equipment operation status is monitored. In one monitoring cycle, the product output of the equipment is α. If α / β < γ, the current equipment operation status is judged to be abnormal, where β is the predetermined standard product output per unit monitoring cycle, and γ is the predetermined allowable fluctuation percentage of the ratio of actual product output to standard product output. Otherwise, the current equipment operation status is judged to be normal. When the current equipment operation status is judged to be abnormal, the equipment operation characteristics are called as equipment operation anomaly vectors and stored in the operation anomaly vector database. The equipment tag of the equipment to be processed is marked, where the dimension of the equipment operation anomaly vector is the same as the number of equipment operation characteristics. By collecting equipment operation data and preprocessing to extract operation characteristics, and combining the allowable fluctuation range of the ratio of actual to standard product output to determine the status, compared with the traditional fuzzy judgment relying on experience, the accuracy of equipment operation status judgment is improved, and the situation of misjudgment or omission is reduced. When the equipment is abnormal, the operation characteristics can be stored as anomaly vectors in the database and marked with tags to be processed, and the key data at the time of the anomaly can be completely retained to avoid information loss and provide an accurate basis for subsequent fault matching and analysis. Real-time monitoring and timely anomaly identification can quickly incorporate abnormal equipment into the processing flow, preventing minor anomalies from escalating into major failures, effectively ensuring the continuous and stable operation of equipment, and reducing losses caused by the expansion of failures.

[0017] In step S2, when a device tag to be processed appears in the anomaly vector database, the device operation feature corresponding to the device tag to be processed is called as the current operation feature. The current operation feature is {A1, A2, ..., A...} n ,…,A N}, where N represents the number of equipment operating characteristics, A n This represents the nth current operating characteristic. It monitors equipment faults and retrieves historical fault records. The number of historical fault records is M, and the equipment operating characteristic corresponding to the mth fault record is {B1, B2, ..., B...}.n ,…,B N}, B n This represents the operating feature of the nth device in the mth fault record. Then, considering the nth current operating feature and the operating feature of the nth device, we obtain the similarity Z between the nth feature of the current device and the device in the mth fault record. m_n The dissimilarity Z of the nth feature of the mth fault record m_n =|(A n -B n ) / B n Substituting each of the following into n=1,2,…,N, we obtain the N feature dissimilarity levels between the current device and the device in the m-th fault record, and then obtain the device dissimilarity C between the current device and the device in the m-th fault record. m C m Let C be the average of the dissimilarity of the current device and the device with the m-th fault record among the N devices. m If the value is greater than C0, the system determines that the current device's operating characteristics are dissimilar to those of the m-th fault record, where C0 represents a pre-defined threshold for device dissimilarity. Otherwise, the system determines that the current device's operating characteristics are similar to those of the m-th fault record. By comparing the operating characteristics of the current abnormal device with those of historical fault records, and using feature dissimilarity and average dissimilarity as the basis, combined with a preset threshold, the system effectively avoids the subjectivity and bias of traditional manual comparisons, improving the objectivity and accuracy of similarity judgments. This comprehensive approach, considering the differences in various operating characteristics, can accurately filter out fault cases similar to the current device's state from a large number of historical records, providing a reliable reference for subsequent fault type judgment, reducing blind investigations, significantly improving the efficiency and targeting of fault analysis, and laying the foundation for quickly pinpointing the fault direction.

[0018] In step S3, when the operating characteristics of the current device are similar to those of the m-th fault record, the m-th fault record is added to the temporary comparison database of the current device, and m=1,2,…,M is substituted one by one. Then, X fault records are added to the temporary comparison database of the current device, and these X fault records are used as the comparison fault records for the current device. The maintenance records of these X fault records are then called. If the maintenance records of these X fault records, after monitoring and maintenance, belong to the same fault type V, then the fault type of the current device is determined to be fault V. If the maintenance records of these X fault records, after monitoring and maintenance, do not belong to the same fault type, then the fault type corresponding to these X fault records is used as the predicted fault type, and the predicted fault type is {D1,D2,…,D…}. y ,…,D Y}, where Y represents the number of predicted fault types, D yThe number of fault records corresponding to the predicted fault types is obtained by retrieving the temporary comparison database. The number of fault records corresponding to the Y predicted fault types in the temporary comparison database is {d1,d2,…,dy,…,dY}, where dy is the predicted fault type D. y The temporary comparison database contains the number of corresponding fault records. By centrally storing historical fault records with similar characteristics to the current equipment in the temporary comparison database, and automatically analyzing fault types based on maintenance records, the fault diagnosis process is effectively simplified. When similar records belong to the same fault type, the fault type of the current equipment can be directly determined, avoiding the tedious manual comparison. When the types are different, the predicted fault type and the corresponding number of records are automatically extracted, providing a clear direction for subsequent troubleshooting. This approach reduces the subjective bias of manual judgment, ensures a more systematic and comprehensive fault type analysis, and lays a data foundation for accurate troubleshooting, improving the efficiency and reliability of fault diagnosis.

[0019] In step S4, the normal operating characteristics of the device at the time of factory shipment are invoked, denoted as {E1, E2, ..., E n ,…,E N}, where E n This represents the nth normal operation characteristic. For the yth predicted fault type, analysis is performed by calling the dy-th fault records of the y-th predicted fault type from the temporary comparison database. The operating characteristic of the nth device among the dy-th fault records is then obtained as {F}. 1_n ,F 2_n ,…,F w_n ,…,F dy_n}, where F w_n Let be the operating feature of the nth device in the wth fault record, where w = 1, 2, ..., dy. Then, the similarity J between the operating feature of the nth device in the wth fault record and the operating feature of the nth normal operation record can be obtained. w_n =1-|(F w_n -E n ) / E n |, if J w_n If J0 >, determine that the operating characteristics of the nth device in the wth fault record are different from the operating characteristics of the nth normal operation record; otherwise, determine that the operating characteristics of the nth device in the wth fault record are similar to the operating characteristics of the nth normal operation record. Substitute each value of w = 1, 2, ..., dy into the list of fault records to obtain the similarity between the nth device's operating feature and the nth normal operation feature. Then, obtain the number L of dissimilar features between the nth device's operating feature and the nth normal operation feature in the dy fault records. If L / dy ≥ k0, then determine that the nth device's operating feature is an identifier feature of the dy fault records, where k0 is a pre-set identifier feature judgment threshold. Otherwise, determine that the nth device's operating feature is not an identifier feature of the dy fault records, and obtain R identifier features of the dy fault records. If R = 0, then determine that the dy fault records have no identifier features, and predict the fault type D. y If the fault is classified as random, then the R current operating features corresponding to the identifier features of the dy fault records are called. These R current operating features are used as judgment features, and are labeled as {a1, a2, ..., a...}. r ,…,a R}, where a r Let represent the r-th judgment feature, and obtain the similarity {j1,j2,…,j} of the R judgment features to the normal operation feature. r ,…,j R}, where j r Let j be the similarity between the r-th judgment feature and the normal operation feature. r =1-|(a r -E r ) / E r |,E r For the normal operation feature in the normal operation feature set corresponding to the r-th judgment feature, if j r If the value of the r-th judgment feature is greater than J0, then the r-th judgment feature is determined to conform to the y-th predicted fault type; otherwise, the r-th judgment feature is determined to not conform to the y-th predicted fault type. This leads to the fault judgment priority for the y-th predicted fault type, which is the ratio of the number of features conforming to the y-th predicted fault type to the number of judgment features. Based on the normal operating characteristics of the equipment at the factory, and through comparative analysis of the historical records of predicted fault types, identifying features are accurately selected, avoiding interference from irrelevant features in fault judgment. Fault priority is determined by judging the degree of conformity between the current operating characteristics and the identified features, making fault analysis more precise than it appears. This method ensures the uniformity of the judgment benchmark by relying on factory standards, and focuses on key differences through identified features, reducing invalid analysis and providing a clear priority basis for subsequent fault troubleshooting. This improves the scientific rigor and relevance of fault type judgment, effectively avoiding the confusion caused by complex features in traditional troubleshooting.

[0020] In step S5, the predicted fault types that satisfy R not equal to 0 are sorted from high to low according to the fault judgment priority. The random fault types that satisfy R equal to 0 are sorted after the predicted fault types that satisfy R not equal to 0, thus obtaining a sequence of predicted fault types. Then, the predicted fault type sequence is used to check the fault types in sequence.

[0021] In step S6, after completing the sequential fault type investigation of the fault type sequence, the temporary comparison database is deleted, and the current fault information is updated in the database.

[0022] An AI-based automated operation and maintenance system for vector databases includes: a data monitoring and anomaly collection module, a historical fault similarity analysis module, a similarity record and type initial determination module, a data in-depth analysis module, a fault type sorting and investigation module, and an investigation result data update module. The data monitoring and anomaly collection module is used to monitor equipment operation data in real time, analyze equipment operation status, and collect features of equipment with abnormal operation. The historical fault similarity analysis module is used to call historical fault records and abnormal equipment for feature collection, and analyze the degree of similarity between the current equipment operating status and the operating status of historical faulty equipment. The similarity record and type initial determination module is used to store similar historical fault records into a temporary comparison database, determine the fault type based on the maintenance records, and then obtain the predicted fault type and the number of predicted fault type records. The data depth analysis module is used to call the equipment's factory data, analyze and predict fault types, and analyze the predicted fault types through fault records and current characteristics; The fault type sorting and investigation module is used to sort the predicted fault types, and then perform sequential fault type investigation on the predicted fault type sequence. The troubleshooting result data update module is used to update the database after the fault type investigation is completed.

[0023] Example 1: In step S1, for the processing equipment, data such as temperature, rotation speed, energy consumption, and vibration frequency are collected in real time during its operation. Outliers due to temporary fluctuations are removed through data cleaning, and then key features reflecting the equipment's operating status are obtained through feature extraction and integration. Simultaneously, the number of qualified parts produced by the equipment within a complete processing cycle is monitored and compared with a pre-set standard number of qualified products per unit cycle. If the ratio of the actual qualified quantity to the standard quantity is lower than the allowable fluctuation range, the equipment is deemed to be operating abnormally. The extracted equipment operating features are then stored as an anomaly vector in the operating anomaly vector database, and the equipment is labeled as pending processing. If the ratio is within the allowable range, the equipment is deemed to be operating normally, and real-time monitoring continues.

[0024] Upon entering step S2, when the system detects a pending label for the device in the operational anomaly vector database, it immediately retrieves the current operational characteristics of the device corresponding to the label and simultaneously retrieves past fault records of this type of processing equipment stored in the database. The system compares the current device's temperature, rotational speed, and other characteristics with the corresponding characteristics in each historical fault record, analyzes the degree of difference between the two, calculates the average of all characteristic differences, and combines this with a preset difference judgment threshold to filter out historical fault records similar to the current device's state.

[0025] In step S3, the selected similar historical fault records are stored in the temporary comparison database of the equipment, and then the maintenance and repair files corresponding to these historical records are retrieved. If the files show that these records all correspond to the "bearing wear" fault type, the current equipment fault type is initially determined to be bearing wear; if the files contain multiple fault types such as bearing wear and motor overload, these types are used as the predicted fault types of the current equipment, and the number of historical records corresponding to each predicted type in the temporary comparison database is counted.

[0026] In step S4, the normal operating characteristic data of the processing equipment at the time of its manufacture is retrieved, including standard temperature range, rated speed, and normal energy consumption range. For each predicted fault type, the differences between the equipment characteristics in the corresponding historical records and the normal characteristics at the time of manufacture are analyzed. Features that commonly deviate from the normal range in this type of fault are selected as identification features, such as the abnormal temperature rise feature that often occurs in bearing wear faults. Then, the degree of conformity between the corresponding features of the current equipment and the identification features is compared, and the judgment priority of each predicted fault type is calculated.

[0027] In step S5, the predicted fault types are sorted from high to low according to their judgment priority, with random fault types without clear identifying characteristics placed at the end, forming a fault investigation sequence. Maintenance personnel conduct investigations based on this sequence, prioritizing high-priority bearing wear faults by disassembling the equipment bearings to check for wear. If no problems are found, other types such as motor overload are then investigated sequentially.

[0028] After troubleshooting and repairing the equipment, proceed to step S6, delete the temporary comparison database for the equipment, and update the system database with information such as the equipment fault type, troubleshooting process, and repair plan to provide data support for the operation and maintenance of similar equipment in the future.

[0029] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary sensing device embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An automated operation and maintenance method for vector databases based on artificial intelligence, characterized by: The method includes the following steps: S1. Real-time monitoring of equipment operation data, analysis of equipment operation status, and feature collection of equipment with abnormal operation; S2. Collect features from historical fault records and abnormal equipment, and analyze the similarity between the current equipment operating status and the operating status of historical faulty equipment. S3. Store similar historical fault records in a temporary comparison database, determine the fault type based on the maintenance records, and then obtain the predicted fault type and the number of predicted fault type records. S4. Call the equipment's factory data, analyze and predict the fault type, and analyze the predicted fault type through fault records and current characteristics; S5. Sort the predicted fault types, and then check the fault types in the predicted fault type sequence in order. S6. After completing the troubleshooting, update the database.

2. The automated operation and maintenance method for vector databases based on artificial intelligence according to claim 1, characterized in that: In step S1, for any type of equipment, equipment operation data is collected, and after preprocessing, equipment operation characteristics are obtained. The current equipment operation status is monitored. In one monitoring cycle, the product output of the equipment is α. If α / β < γ, the current equipment operation status is judged to be abnormal, where β is the standard product output per unit monitoring cycle and γ is the allowable fluctuation percentage of the ratio of actual product output to standard product output. Otherwise, the current equipment operation status is judged to be normal. When the current equipment operation status is judged to be abnormal, the equipment operation characteristics are called as equipment operation abnormality vectors and stored in the operation abnormality vector database. The equipment label of the equipment to be processed is marked, where the dimension of the equipment operation abnormality vector is the same as the number of equipment operation characteristics.

3. The automated operation and maintenance method for vector databases based on artificial intelligence according to claim 2, characterized in that: In step S2, when a device tag to be processed appears in the anomaly vector database, the device operation feature corresponding to the device tag to be processed is called as the current operation feature. The current operation feature is {A1, A2, ..., A...} n ,…,A N }, where N represents the number of equipment operating characteristics, A n This represents the nth current operating characteristic. It monitors equipment faults and retrieves historical fault records. The number of historical fault records is M, and the equipment operating characteristic corresponding to the mth fault record is {B1, B2, ..., B...}. n ,…,B N }, B n This represents the operating feature of the nth device in the mth fault record. Then, considering the nth current operating feature and the operating feature of the nth device, we obtain the similarity Z between the nth feature of the current device and the device in the mth fault record. m_n The dissimilarity Z of the nth feature of the mth fault record m_n =|(A n -B n ) / B n Substitute each of the following into n=1,2,…,N to obtain the N feature dissimilarity between the current device and the device in the m-th fault record, and then obtain the device dissimilarity C between the current device and the device in the m-th fault record. m C m Let C be the average of the dissimilarity of the current device and the device with the m-th fault record among the N devices. m If >C0, it is determined that the operating characteristics of the current device are not similar to those of the device in the m-th fault record. C0 represents the pre-set threshold for determining the dissimilarity of the devices. Otherwise, it is determined that the operating characteristics of the current device are similar to those of the device in the m-th fault record.

4. The automated operation and maintenance method for vector databases based on artificial intelligence according to claim 3, characterized in that: In step S3, when the operating characteristics of the current device are similar to those of the m-th fault record, the m-th fault record is added to the temporary comparison database of the current device, and m=1,2,…,M is substituted one by one. Then, X fault records are added to the temporary comparison database of the current device, and these X fault records are used as the comparison fault records for the current device. The maintenance records of these X fault records are then called. If the maintenance records of these X fault records, after monitoring and maintenance, belong to the same fault type V, then the fault type of the current device is determined to be fault V. If the maintenance records of these X fault records, after monitoring and maintenance, do not belong to the same fault type, then the fault type corresponding to these X fault records is used as the predicted fault type, and the predicted fault type is {D1,D2,…,D…}. y ,…,D Y }, where Y represents the number of predicted fault types, and D y The number of fault records corresponding to the predicted fault types is obtained by retrieving the temporary comparison database. The number of fault records corresponding to the Y predicted fault types in the temporary comparison database is {d1,d2,…,dy,…,dY}, where dy is the predicted fault type D. y The number of corresponding fault records is compared in the temporary comparison database.

5. The automated operation and maintenance method for vector databases based on artificial intelligence according to claim 4, characterized in that: In step S4, the normal operating characteristics of the device at the time of factory shipment are invoked, denoted as {E1, E2, ..., E n ,…,E N }, where E n This represents the nth normal operation characteristic. For the yth predicted fault type, analysis is performed by calling the dy-th fault records of the y-th predicted fault type from the temporary comparison database. The operating characteristic of the nth device among the dy-th fault records is then obtained as {F}. 1_n ,F 2_n ,…,F w_n ,…,F dy_n }, where F w_n Let be the operating feature of the nth device in the wth fault record, where w = 1, 2, ..., dy. Then, the similarity J between the operating feature of the nth device in the wth fault record and the operating feature of the nth normal operation record can be obtained. w_n =1-|(F w_n -E n ) / E n |, if J w_n >J0, determine that the operating characteristics of the nth device in the wth fault record are different from the operating characteristics of the nth normal operation record; Otherwise, determine that the operating characteristics of the nth device in the wth fault record are similar to the operating characteristics of the nth normal operation record.

6. The automated operation and maintenance method for vector databases based on artificial intelligence according to claim 5, characterized in that: Substitute w=1,2,…,dy one by one to obtain the similarity between the nth device operation feature and the nth normal operation feature in the dy fault records. Then obtain the number L of the nth device operation feature and the nth normal operation feature in the dy fault records that are different. If L / dy≥k0, then the nth device operation feature is judged to be the identification feature of the dy fault records, where k0 is the pre-set identification feature judgment threshold. Otherwise, determine that the operating characteristic of the nth device is not the identifier feature of the dy fault record, and then obtain R identifier features of the dy fault record. If R=0, then determine that the dy fault records have no identifier features, and predict the fault type D. y If the fault is classified as random, then the R current operating features corresponding to the identifier features of the dy fault records are called. These R current operating features are used as judgment features, and are labeled as {a1, a2, ..., a...}. r ,…,a R }, where a r Let represent the r-th judgment feature, and obtain the similarity {j1,j2,…,j} of the R judgment features to the normal operation feature. r ,…,j R }, where j r Let j be the similarity between the r-th judgment feature and the normal operation feature. r =1-|(a r -E r ) / E r |,E r For the normal operation feature in the normal operation feature set corresponding to the r-th judgment feature, if j r If the value of the r-th judgment feature is greater than J0, then the r-th judgment feature is determined to conform to the y-th predicted fault type; otherwise, the r-th judgment feature is determined to not conform to the y-th predicted fault type, and the fault judgment priority of the y-th predicted fault type is obtained. The fault judgment priority of the y-th predicted fault type is the ratio of the number of features conforming to the y-th predicted fault type to the number of judgment features.

7. The automated operation and maintenance method for vector databases based on artificial intelligence according to claim 6, characterized in that: In step S5, the predicted fault types that satisfy R not equal to 0 are sorted from high to low according to the fault judgment priority. The random fault types that satisfy R equal to 0 are sorted after the predicted fault types that satisfy R not equal to 0, thus obtaining a sequence of predicted fault types. Then, the predicted fault type sequence is used to check the fault types in sequence.

8. The automated operation and maintenance method for vector databases based on artificial intelligence according to claim 6, characterized in that: In step S6, after completing the sequential fault type investigation of the fault type sequence, the temporary comparison database is deleted, and the current fault information is updated in the database.

9. An AI-based automated operation and maintenance system for vector databases, wherein the system is applied to the AI-based automated operation and maintenance method for vector databases as described in any one of claims 1-8, characterized in that: The system includes: a data monitoring and anomaly collection module, a historical fault similarity analysis module, a similarity record and type initial determination module, a data in-depth analysis module, a fault type sorting and investigation module, and an investigation result data update module; The data monitoring and anomaly collection module is used to monitor equipment operation data in real time, analyze equipment operation status, and collect features of equipment with abnormal operation. The historical fault similarity analysis module is used to call historical fault records and abnormal equipment for feature collection, and analyze the degree of similarity between the current equipment operating status and the operating status of historical faulty equipment. The similarity record and type initial determination module is used to store similar historical fault records into a temporary comparison database, determine the fault type based on the maintenance records, and then obtain the predicted fault type and the number of predicted fault type records. The data depth analysis module is used to call the equipment's factory data, analyze and predict fault types, and analyze the predicted fault types through fault records and current characteristics; The fault type sorting and investigation module is used to sort the predicted fault types, and then perform sequential fault type investigation on the predicted fault type sequence. The troubleshooting result data update module is used to update the database after the fault type investigation is completed.

Images