A power equipment maintenance management system and method based on artificial intelligence

Abstract

The present invention relates to the technical field of equipment maintenance safety management, and specifically to an artificial intelligence-based power equipment maintenance management system and method, comprising obtaining, according to the precautionary operations corresponding to each maintenance event, respectively an operation blacklist to be followed when executing each maintenance event, and a supervision center issuing safety warning prompts to maintenance personnel when executing each maintenance event based on the operation blacklist; extracting from the supervision center all historical monitoring video streams that have not caused the supervision center to issue safety warning prompts, and setting the extracted historical monitoring video streams as target video streams; identifying and extracting target maintenance operations existing when maintenance personnel execute various maintenance events; performing transition warning risk assessment on each execution transition node; and prioritizing warning prompts for precautionary operations sent to a mobile device worn by maintenance personnel based on the transition warning risk assessment results of each execution transition node.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment maintenance safety management, and in particular to an artificial intelligence-based power equipment maintenance management system and method. Background Art

[0002] Modern life is inseparable from electricity. The power department attaches great importance to the maintenance of power equipment. Regularly sending people to inspect and maintain power equipment is a necessary means to ensure that people can live a normal life. However, there are some safety hazards in the maintenance process of power equipment, which poses a huge threat to the safety of workers, work tools, and people and property around power equipment. It is necessary to pay attention to safety issues in the maintenance process of power equipment and take corresponding measures.

[0003] In actual work, identifying the danger point is the premise, controlling the danger point is the key, and resolving the danger point is the purpose. Through the analysis of danger points, we can point out the focus of prevention and achieve the purpose of safety. By carrying out the analysis of danger points, we can eliminate various possible accident hazards caused by man-made, equipment, environment, and management before they occur, which fully reflects the principle of prevention first, so as to achieve the purpose of preventing accidents before they happen and ensure the personal safety of maintenance workers. Summary of the invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based power equipment maintenance management system and method to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an artificial intelligence-based power equipment maintenance management method, the method comprising:

[0006] Step S100: Locking all maintenance areas arranged in the target area to carry out equipment maintenance on the power equipment in the target area; performing a maintenance task in one maintenance area as a maintenance event; installing monitoring equipment in each maintenance area, and uploading the monitoring video stream collected for each maintenance event to the supervision center in real time;

[0007] Step S200: According to the precautionary operation items corresponding to each maintenance event, an operation blacklist to be followed when executing each maintenance event is obtained, wherein the operation blacklist is composed of a number of maintenance operations; the supervision center issues safety warning prompts to maintenance personnel when executing each maintenance event according to the operation blacklist;

[0008] Step S300: extracting all historical monitoring video streams that have not caused the monitoring center to issue a safety warning prompt from the monitoring center, and setting the extracted historical monitoring video streams as target video streams; wherein one historical monitoring video stream corresponds to the entire execution process of a historical maintenance event; identifying and extracting the target maintenance operations existing in the maintenance personnel when executing various maintenance events;

[0009] Step S400: extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; perform transition warning risk assessment on each execution transition node according to the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node;

[0010] Step S500: According to the transition warning risk assessment results of each execution transition node, the warning prompts of the precautionary operations sent to the mobile device worn by the maintenance personnel are prioritized.

[0011] Furthermore, the maintenance operation includes the maintenance operations performed by maintenance personnel on the power equipment to be maintained indirectly through maintenance instruments, the usage operations performed on the maintenance instruments by maintenance personnel when they perform maintenance on the power equipment to be maintained through maintenance instruments, and the maintenance operations performed by maintenance personnel on the power equipment to be maintained directly.

[0012] Further, step S200 includes:

[0013] Step S201: Allocate a sequence of maintenance events to be executed by the maintenance personnel to the mobile device worn by the maintenance personnel. Before the maintenance personnel execute a maintenance event, the precautions that the maintenance personnel need to follow when executing a maintenance event are sent to the mobile device worn by the maintenance personnel. The corresponding monitoring equipment captures the video of the maintenance personnel when executing any maintenance event, and uploads the generated monitoring video stream to the supervision center.

[0014] Step S202: The supervision center identifies information on each monitoring video stream. When the supervision center identifies and captures from the monitoring video stream corresponding to a maintenance event that a maintenance personnel has performed a maintenance operation on a blacklist of operations that must be followed when performing a maintenance event, the supervision center sends a safety warning prompt to the mobile device worn by the maintenance personnel.

[0015] Furthermore, step S300 includes:

[0016] Step S301: performing information recognition on each historical monitoring video stream, capturing and extracting all maintenance operations performed by maintenance personnel in each historical monitoring video stream, and obtaining a maintenance operation set corresponding to each historical monitoring video stream; and collecting historical monitoring video streams corresponding to the same maintenance event, and obtaining a plurality of historical monitoring video stream sets;

[0017] Step S302: Extract the first characteristic maintenance operation set P1(a)=W1∩W2∩...∩W when the maintenance personnel perform the maintenance event a corresponding to the historical monitoring video stream set A from each historical monitoring video stream set A. n ; Among them, W1, W2, ..., W n are the 1st, 2nd, ..., ..nth maintenance operation sets in a historical monitoring video stream set A; collect any maintenance operation set W in a historical monitoring video stream set A i The extracted operation set U(W i )=W i -P1, obtain the second characteristic maintenance operation set P2(a) corresponding to maintenance event a;

[0018] The first characteristic maintenance operation set P1(a) extracted above is often some necessary maintenance operations when performing maintenance event a. According to P1(a), from any maintenance operation set W i The operation set U(W i ), refers to other maintenance operations performed by maintenance personnel in the corresponding historical monitoring video stream except for necessary maintenance operations. Because it is extracted from the historical monitoring video stream that has never caused the supervision center to issue a safety warning, it means that the other maintenance operations P1(a) are maintenance operations that are not dangerous when performing maintenance event a, that is, they are maintenance operations on the operation whitelist of maintenance event a; the process of calculating the characteristic index is to calculate the characteristic regularity of other maintenance operations performed by maintenance personnel in the process of performing maintenance event a except for necessary maintenance operations. The higher the characteristic index, the lower the randomness of the maintenance operation, the stronger the regularity, and the higher the probability of occurrence when performing maintenance event a;

[0019] Step S303: Calculate the characteristic index β(f)=(k / n)×H for any maintenance operation item f in the second characteristic maintenance operation set P2(a); wherein k represents the total number of maintenance operation sets containing any maintenance operation item f in a certain historical monitoring video stream set A; n represents the total number of maintenance operation sets contained in a certain historical monitoring video stream set A; H represents the highest occurrence frequency of any maintenance operation item f in a maintenance operation set among the k maintenance operation sets containing any maintenance operation item f;

[0020] Step S304: remove maintenance operation items whose characteristic index β is less than the index threshold from the second characteristic maintenance operation set P2(a) to obtain set P2(a)', and set P2(a)' is set as the target maintenance operation set when executing maintenance event a.

[0021] Furthermore, step S400 includes:

[0022] Step S401: extract the target maintenance operation set of two adjacent maintenance events constituting each execution transition node, and the maintenance operation set corresponding to the operation blacklist; suppose that in a certain execution transition node F, the target maintenance operation set of the maintenance event R1 with the previous execution order is X1, the maintenance operation set corresponding to the operation blacklist is Y1, the target maintenance operation set of the maintenance event R2 with the later execution order is X2, and the maintenance operation set corresponding to the operation blacklist is Y2;

[0023] Step S402: Calculate the transition warning risk value α of a certain execution transition node = [card(G1) / card(Y2)]×card(G2)×1 / T; wherein, G1=X1∩Y2, G2=Y1∩Y2; card(G1) represents the cumulative number of maintenance operations included in the set G1, card(Y2) represents the cumulative number of maintenance operations included in the set Y2, and card(G2) represents the cumulative number of maintenance operations included in the set G2; T represents the interval between the inspection personnel's execution of maintenance event R1 and maintenance event R2.

[0024] Among them, the higher the value of [card(G1) / card(Y2)], the more likely it is that the inspection personnel will make an error when transitioning from the maintenance event R1 with an earlier execution order to the maintenance event R2 with a later execution order, due to the execution habits in the previous execution event; the higher the value of card(G2), the higher the deviation between the precautions that the maintenance personnel need to follow in the two maintenance events, and the more likely it is that the inspection personnel will make an error; the shorter the T value, the shorter the time it takes for the maintenance personnel to transition between the two maintenance events, and the more likely it is that the inspection personnel will make an error.

[0025] Furthermore, step S500 includes:

[0026] Step S501: when the transition warning risk value α of a certain execution transition node F is greater than the risk threshold, obtain the maintenance operation set Y2 corresponding to the operation blacklist in the maintenance event R2 with a later execution order; mark the maintenance operations included in the set G1=X1∩Y2 in the maintenance operation set Y2;

[0027] Step S502: Specially mark the precautions corresponding to the maintenance operations marked in the maintenance operation set Y2. When the precautions are sent to the mobile device worn by the corresponding maintenance personnel, the precautions with special marks will be given priority as early warning reminders to the maintenance personnel.

[0028] In order to better implement the above method, a power equipment maintenance management system is also proposed, which includes an equipment maintenance management module, a safety warning prompt management module, a target maintenance operation extraction management module, a transition warning risk assessment management module, and a warning prompt management module;

[0029] The equipment maintenance management module is used to lock all maintenance areas arranged in the target area for equipment maintenance of power equipment in the target area; perform a maintenance task in a maintenance area and set it as a maintenance event; install monitoring equipment in each maintenance area and upload the monitoring video stream collected for each maintenance event to the supervision center in real time;

[0030] The safety warning prompt management module is used to obtain the operation blacklist that needs to be followed when executing each maintenance event according to the precautions corresponding to each maintenance event, where the operation blacklist consists of several maintenance operations; the supervision center issues safety warning prompts to maintenance personnel when executing each maintenance event based on the operation blacklist;

[0031] The target maintenance operation extraction management module is used to extract all historical monitoring video streams that have not caused the supervision center to issue a safety warning prompt from the supervision center, and set the extracted historical monitoring video streams as target video streams; wherein, a historical monitoring video stream corresponds to the entire execution process of a historical maintenance event; and the target maintenance operations existing in the maintenance personnel when executing various maintenance events are respectively identified and extracted;

[0032] The transition warning risk assessment management module is used to extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; according to the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node, the transition warning risk assessment is performed on each execution transition node;

[0033] The early warning prompt management module is used to prioritize the early warning prompts for the precautionary operations sent to the mobile devices worn by maintenance personnel according to the transition warning risk assessment results of each execution transition node.

[0034] Further, the transition warning risk assessment management module includes a cross-distribution information combing unit and a transition warning risk assessment unit;

[0035] The cross-distribution information combing unit is used to extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; comb the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node;

[0036] The transition warning risk assessment unit is used to receive the data in the cross-distribution information combing unit, calculate the transition warning risk value of each execution transition node, and complete the transition warning risk assessment of each execution transition node.

[0037] Compared with the prior art, the beneficial effect achieved by the present invention is as follows: the present invention performs transition warning risk assessment on the transition nodes between the execution of two adjacent maintenance events by analyzing the cross-distribution between the target maintenance operation and the operation blacklist when the maintenance personnel execute two adjacent maintenance events, thereby reducing the phenomenon of incorrect operations by inspection personnel due to the transition of execution events, and eliminating various possible man-made accident hazards before they occur, fully embodying the principle of prevention first, so as to achieve the purpose of preventing problems before they occur and ensuring the personal safety of maintenance workers. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0039] Figure 1 It is a flow chart of an artificial intelligence-based power equipment maintenance management method of the present invention;

[0040] Figure 2 It is a structural schematic diagram of an artificial intelligence-based power equipment maintenance management system of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1-Figure 2 The present invention provides a technical solution: an artificial intelligence-based power equipment maintenance management method, the method comprising:

[0043] Step S100: Locking all maintenance areas arranged in the target area to carry out equipment maintenance on the power equipment in the target area; performing a maintenance task in one maintenance area as a maintenance event; installing monitoring equipment in each maintenance area, and uploading the monitoring video stream collected for each maintenance event to the supervision center in real time;

[0044] Among them, the maintenance operation includes the maintenance operation indirectly performed by the maintenance personnel on the power equipment to be repaired through the maintenance instrument, the operation performed by the maintenance personnel on the maintenance instrument when the maintenance personnel perform the maintenance on the power equipment to be repaired through the maintenance instrument, and the maintenance operation directly performed by the maintenance personnel on the power equipment to be repaired;

[0045] Step S200: According to the precautionary operation items corresponding to each maintenance event, an operation blacklist to be followed when executing each maintenance event is obtained, wherein the operation blacklist is composed of a number of maintenance operations; the supervision center issues safety warning prompts to maintenance personnel when executing each maintenance event according to the operation blacklist;

[0046] Wherein, step S200 includes:

[0047] Step S201: Allocate a sequence of maintenance events to be executed by the maintenance personnel to the mobile device worn by the maintenance personnel. Before the maintenance personnel execute a maintenance event, the precautions that the maintenance personnel need to follow when executing a maintenance event are sent to the mobile device worn by the maintenance personnel. The corresponding monitoring equipment captures the video of the maintenance personnel when executing any maintenance event, and uploads the generated monitoring video stream to the supervision center.

[0048] Step S202: The supervision center identifies information from each monitoring video stream. When the supervision center identifies and captures, from the monitoring video stream corresponding to a maintenance event, that a maintenance personnel has performed a maintenance operation on an operation blacklist that must be followed when performing a maintenance event, the supervision center sends a safety warning prompt to the mobile device worn by the maintenance personnel;

[0049] Step S300: extracting all historical monitoring video streams that have not caused the monitoring center to issue a safety warning prompt from the monitoring center, and setting the extracted historical monitoring video streams as target video streams; wherein one historical monitoring video stream corresponds to the entire execution process of a historical maintenance event; identifying and extracting the target maintenance operations existing in the maintenance personnel when executing various maintenance events;

[0050] Wherein, step S300 includes:

[0051] Step S301: performing information recognition on each historical monitoring video stream, capturing and extracting all maintenance operations performed by maintenance personnel in each historical monitoring video stream, and obtaining a maintenance operation set corresponding to each historical monitoring video stream; and collecting historical monitoring video streams corresponding to the same maintenance event, and obtaining a plurality of historical monitoring video stream sets;

[0052] Step S302: Extract the first characteristic maintenance operation set P1(a)=W1∩W2∩...∩W when the maintenance personnel perform the maintenance event a corresponding to the historical monitoring video stream set A from each historical monitoring video stream set A. n ; Among them, W1, W2, ..., W n are the 1st, 2nd, ..., ..nth maintenance operation sets in a historical monitoring video stream set A; collect any maintenance operation set W in a historical monitoring video stream set A i The extracted operation set U(W i )=W i -P1, obtain the second characteristic maintenance operation set P2(a) corresponding to maintenance event a;

[0053] Step S303: Calculate the characteristic index β(f)=(k / n)×H for any maintenance operation item f in the second characteristic maintenance operation set P2(a); wherein k represents the total number of maintenance operation sets containing any maintenance operation item f in a certain historical monitoring video stream set A; n represents the total number of maintenance operation sets contained in a certain historical monitoring video stream set A; H represents the highest occurrence frequency of any maintenance operation item f in a maintenance operation set among the k maintenance operation sets containing any maintenance operation item f;

[0054] For example, a historical monitoring video stream set A is {historical monitoring video stream set 1, historical monitoring video stream set 2, historical monitoring video stream set 3, historical monitoring video stream set 4, historical monitoring video stream set 5, historical monitoring video stream set 6, historical monitoring video stream set 7, historical monitoring video stream set 8}; that is, n = 8;

[0055] For example, the second characteristic maintenance operation set P2(a) is {maintenance operation 1, maintenance operation 2, maintenance operation 3, maintenance operation 4};

[0056] Among them, the historical monitoring video stream set 1, the historical monitoring video stream set 2, the historical monitoring video stream set 4, and the historical monitoring video stream set 6 that contain the maintenance operation 1; that is, k=4;

[0057] And the frequency of maintenance operation 1 in historical monitoring video stream set 1 is 4, the frequency of maintenance operation 1 in historical monitoring video stream set 2 is 3, the frequency of maintenance operation 1 in historical monitoring video stream set 4 is 4, and the frequency of maintenance operation 1 in historical monitoring video stream set 6 is 8; that is, H=8;

[0058] In summary, the characteristic index β corresponding to maintenance operation 1 = (4 / 8) × 8 = 4;

[0059] Among them, the historical monitoring video stream set 1, the historical monitoring video stream set 2, the historical monitoring video stream set 3, the historical monitoring video stream set 5, and the historical monitoring video stream set 6 that contain the maintenance operation 2; that is, k=5;

[0060] And the frequency of maintenance operation 2 appearing in historical monitoring video stream set 1 is 5, the frequency of maintenance operation 2 appearing in historical monitoring video stream set 2 is 4, the frequency of maintenance operation 2 appearing in historical monitoring video stream set 3 is 4, the frequency of maintenance operation 2 appearing in historical monitoring video stream set 5 is 6, and the frequency of maintenance operation 2 appearing in historical monitoring video stream set 6 is 4; that is, H=6;

[0061] In summary, the characteristic index β corresponding to maintenance operation 2 = (5 / 8) × 6 = 3.75;

[0062] Among them, the historical monitoring video stream set 1, the historical monitoring video stream set 3, the historical monitoring video stream set 4, and the historical monitoring video stream set 7 that contain the maintenance operation 3; that is, k=4;

[0063] And the frequency of maintenance operation 3 appearing in historical monitoring video stream set 1 is 2, the frequency of maintenance operation 3 appearing in historical monitoring video stream set 3 is 4, the frequency of maintenance operation 3 appearing in historical monitoring video stream set 4 is 5, and the frequency of maintenance operation 3 appearing in historical monitoring video stream set 7 is 6; that is, H=6;

[0064] In summary, the characteristic index β corresponding to maintenance operation 3 = (4 / 8) × 6 = 3;

[0065] Among them, the historical monitoring video stream set 1, the historical monitoring video stream set 2, the historical monitoring video stream set 6, and the historical monitoring video stream set 8 that contain the maintenance operation 4; that is, k=4;

[0066] And the frequency of maintenance operation 4 appearing in historical monitoring video stream set 1 is 2, the frequency of maintenance operation 4 appearing in historical monitoring video stream set 2 is 5, the frequency of maintenance operation 4 appearing in historical monitoring video stream set 6 is 4, and the frequency of maintenance operation 4 appearing in historical monitoring video stream set 8 is 6; that is, H=6;

[0067] In summary, the characteristic index β corresponding to maintenance operation 4 = (4 / 8) × 6 = 3;

[0068] Step S304: remove the maintenance operation items whose characteristic index β is less than the index threshold from the second characteristic maintenance operation set P2(a) to obtain a set P2(a)', and set P2(a)' is set as the target maintenance operation set when executing maintenance event a;

[0069] Step S400: extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; perform transition warning risk assessment on each execution transition node according to the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node;

[0070] Wherein, step S400 includes:

[0071] Step S401: extract the target maintenance operation set of two adjacent maintenance events constituting each execution transition node, and the maintenance operation set corresponding to the operation blacklist; suppose that in a certain execution transition node F, the target maintenance operation set of the maintenance event R1 with the previous execution order is X1, the maintenance operation set corresponding to the operation blacklist is Y1, the target maintenance operation set of the maintenance event R2 with the later execution order is X2, and the maintenance operation set corresponding to the operation blacklist is Y2;

[0072] Step S402: Calculate the transition warning risk value α of a certain execution transition node = [card(G1) / card(Y2)]×card(G2)×1 / T; wherein, G1 = X1∩Y2, G2 = Y1∩Y2; card(G1) represents the cumulative number of maintenance operations contained in the set G1, card(Y2) represents the cumulative number of maintenance operations contained in the set Y2, and card(G2) represents the cumulative number of maintenance operations contained in the set G2; T represents the interval between the inspection personnel executing the maintenance event R1 and the maintenance event R2;

[0073] Step S500: Prioritize the warning prompts for the precautionary operations sent to the mobile device worn by the maintenance personnel according to the transition warning risk assessment results of each execution transition node;

[0074] Wherein, step S500 includes:

[0075] Step S501: when the transition warning risk value α of a certain execution transition node F is greater than the risk threshold, obtain the maintenance operation set Y2 corresponding to the operation blacklist in the maintenance event R2 with a later execution order; mark the maintenance operations included in the set G1=X1∩Y2 in the maintenance operation set Y2;

[0076] Step S502: Specially mark the precautions corresponding to the maintenance operations marked in the maintenance operation set Y2. When the precautions are sent to the mobile device worn by the corresponding maintenance personnel, the precautions with special marks will be given priority as early warning reminders to the maintenance personnel.

[0077] In order to better implement the above method, a power equipment maintenance management system is also proposed, which includes an equipment maintenance management module, a safety warning prompt management module, a target maintenance operation extraction management module, a transition warning risk assessment management module, and a warning prompt management module;

[0078] The equipment maintenance management module is used to lock all maintenance areas arranged in the target area for equipment maintenance of power equipment in the target area; perform a maintenance task in a maintenance area and set it as a maintenance event; install monitoring equipment in each maintenance area and upload the monitoring video stream collected for each maintenance event to the supervision center in real time;

[0079] The safety warning prompt management module is used to obtain the operation blacklist that needs to be followed when executing each maintenance event according to the precautions corresponding to each maintenance event, where the operation blacklist consists of several maintenance operations; the supervision center issues safety warning prompts to maintenance personnel when executing each maintenance event based on the operation blacklist;

[0080] The target maintenance operation extraction management module is used to extract all historical monitoring video streams that have not caused the supervision center to issue a safety warning prompt from the supervision center, and set the extracted historical monitoring video streams as target video streams; wherein, a historical monitoring video stream corresponds to the entire execution process of a historical maintenance event; and the target maintenance operations existing in the maintenance personnel when executing various maintenance events are respectively identified and extracted;

[0081] The transition warning risk assessment management module is used to extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; according to the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node, the transition warning risk assessment is performed on each execution transition node;

[0082] Among them, the transition warning risk assessment management module includes a cross-distribution information sorting unit and a transition warning risk assessment unit;

[0083] The cross-distribution information combing unit is used to extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; comb the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node;

[0084] A transition warning risk assessment unit is used to receive data from the cross-distribution information combing unit, calculate the transition warning risk value for each execution transition node, and complete the transition warning risk assessment for each execution transition node;

[0085] The early warning prompt management module is used to prioritize the early warning prompts for the precautionary operations sent to the mobile devices worn by maintenance personnel according to the transition warning risk assessment results of each execution transition node.

[0086] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0087] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A power equipment maintenance management method based on artificial intelligence, characterized in that: The method comprises: Step S100: Locking in all maintenance areas arranged in the target area to carry out equipment maintenance on the power equipment in the target area; performing a maintenance task in one maintenance area as a maintenance event; installing monitoring equipment in each maintenance area, and uploading the monitoring video stream collected for each maintenance event to the supervision center in real time; Step S200: According to the precautionary operation items corresponding to each maintenance event, an operation blacklist to be followed when executing each maintenance event is obtained, wherein the operation blacklist is composed of a number of maintenance operations; the supervision center issues a safety warning prompt to the maintenance personnel when executing each maintenance event according to the operation blacklist; Step S300: extracting all historical monitoring video streams that have not caused the monitoring center to issue a safety warning prompt from the monitoring center, and setting the extracted historical monitoring video streams as target video streams; wherein a segment of historical monitoring video streams corresponds to the entire execution process of a historical maintenance event; and identifying and extracting the target maintenance operations existing in the maintenance personnel when performing various maintenance events; Step S400: extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; perform transition warning risk assessment on each execution transition node according to the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node; Step S500: Prioritize the warning prompts for the precautionary operations sent to the mobile device worn by the maintenance personnel according to the transition warning risk assessment results of each execution transition node; The step S300 includes: Step S301: performing information recognition on each historical monitoring video stream, capturing and extracting all maintenance operations performed by maintenance personnel in each historical monitoring video stream, and obtaining a maintenance operation set corresponding to each historical monitoring video stream; and collecting historical monitoring video streams corresponding to the same maintenance event, and obtaining a plurality of historical monitoring video stream sets; Step S302: Extract the first characteristic maintenance operation set P1(a)=W1∩W2∩...∩W when the maintenance personnel perform the maintenance event a corresponding to the historical monitoring video stream set A from each of the historical monitoring video stream sets A. n ; Among them, W1, W2, ..., W n are the 1st, 2nd, ..., ..nth maintenance operation sets in the historical monitoring video stream set A; and any maintenance operation set W in the historical monitoring video stream set A is collected. i The extracted operation set U(W i )=W i -P1, obtain a second characteristic maintenance operation set P2(a) corresponding to the maintenance event a; Step S303: Calculate the characteristic index β(f)=(k / n)×H for any maintenance operation item f in the second characteristic maintenance operation set P2(a); wherein k represents the total number of maintenance operation sets that contain the arbitrary maintenance operation item f in the certain historical monitoring video stream set A; n represents the total number of maintenance operation sets contained in the certain historical monitoring video stream set A; H represents the highest occurrence frequency of the arbitrary maintenance operation item f in a maintenance operation set among the k maintenance operation sets that contain the arbitrary maintenance operation item f; Step S304: remove the maintenance operation items whose characteristic index β is less than the index threshold from the second characteristic maintenance operation set P2(a) to obtain a set P2(a)', and set the set P2(a)' as the target maintenance operation set when executing maintenance event a; The step S400 includes: Step S401: extract the target maintenance operation set of two adjacent maintenance events constituting each execution transition node, and the maintenance operation set corresponding to the operation blacklist; suppose that in a certain execution transition node F, the target maintenance operation set of the maintenance event R1 with the previous execution order is X1, the maintenance operation set corresponding to the operation blacklist is Y1, the target maintenance operation set of the maintenance event R2 with the later execution order is X2, and the maintenance operation set corresponding to the operation blacklist is Y2; Step S402: Calculate the transition warning risk value α=[card(G1) / card(Y2)]×card(G2)×1 / T of the execution transition node; wherein, G1=X1∩Y2, G2=Y1∩Y2; card(G1) represents the cumulative number of maintenance operations included in the set G1, card(Y2) represents the cumulative number of maintenance operations included in the set Y2, and card(G2) represents the cumulative number of maintenance operations included in the set G2; T represents the interval between the inspection personnel executing the maintenance event R1 and the maintenance event R2.

2. The method for power equipment maintenance management based on artificial intelligence according to claim 1 is characterized in that: The maintenance operations include maintenance operations performed indirectly on the power equipment to be maintained by maintenance personnel through maintenance instruments, maintenance operations performed on maintenance instruments when maintenance personnel use maintenance instruments to maintain the power equipment to be maintained, and maintenance operations performed directly on the power equipment to be maintained by maintenance personnel.

3. The method for power equipment maintenance management based on artificial intelligence according to claim 1 is characterized in that: The step S200 includes: Step S201: Allocate a sequence of maintenance events to be executed by a maintenance personnel to a mobile device worn by the maintenance personnel. Before the maintenance personnel execute a maintenance event, the maintenance personnel are sent the precautions to be followed when executing the maintenance event to the mobile device worn by the maintenance personnel. The corresponding monitoring equipment captures the video of the maintenance personnel when executing any maintenance event, and uploads the generated monitoring video stream to the supervision center. Step S202: The supervision center identifies information on each monitoring video stream. When the supervision center identifies and captures, from the monitoring video stream corresponding to a maintenance event, that a maintenance personnel has performed a maintenance operation on an operation blacklist that must be followed when performing the maintenance event, the supervision center sends a safety warning prompt to the mobile device worn by the maintenance personnel.

4. The method for power equipment maintenance management based on artificial intelligence according to claim 3 is characterized in that: The step S500 includes Step S501: when the transition warning risk value α of a certain execution transition node F is greater than the risk threshold, obtain the maintenance operation set Y2 corresponding to the operation blacklist in the maintenance event R2 with a later execution order; mark the maintenance operations included in the set G1=X1∩Y2 in the maintenance operation set Y2; Step S502: Specially mark the precautions corresponding to the maintenance operations marked in the maintenance operation set Y2. When the precautions are sent to the mobile device worn by the corresponding maintenance personnel, the precautions with special marks will be given priority as early warning reminders to the maintenance personnel.

5. An electric power equipment maintenance management system for executing an artificial intelligence-based electric power equipment maintenance management method as described in any one of claims 1 to 4, characterized in that: The system includes an equipment maintenance management module, a safety warning prompt management module, a target maintenance operation extraction management module, a transition warning risk assessment management module, and a warning prompt management module; The equipment maintenance management module is used to lock all maintenance areas arranged in the target area for carrying out equipment maintenance on the power equipment in the target area; Performing a maintenance task in a maintenance area is set as a maintenance event. Monitoring equipment is installed in each maintenance area, and the monitoring video stream collected for each maintenance event is uploaded to the supervision center in real time. The safety warning prompt management module is used to obtain the operation blacklist that needs to be followed when executing each maintenance event according to the precautionary operation items corresponding to each maintenance event, wherein the operation blacklist is composed of a number of maintenance operations; the supervision center provides safety warning prompts to the maintenance personnel when executing each maintenance event according to the operation blacklist; The target maintenance operation extraction management module is used to extract all historical monitoring video streams that have not caused the supervision center to issue a safety warning prompt from the supervision center, and set the extracted historical monitoring video streams as target video streams; wherein a historical monitoring video stream corresponds to the entire execution process of a historical maintenance event; and the target maintenance operations existing in the maintenance personnel when executing various maintenance events are respectively identified and extracted; The transition warning risk assessment management module is used to extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; according to the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node, a transition warning risk assessment is performed on each execution transition node; The warning prompt management module is used to prioritize the warning prompts for the precautionary operations sent to the mobile device worn by the maintenance personnel according to the transition warning risk assessment results of each execution transition node.

6. The power equipment maintenance management system according to claim 5, characterized in that: The transition warning risk assessment management module includes a cross-distribution information combing unit and a transition warning risk assessment unit; The cross-distribution information combing unit is used to extract the maintenance event sequence to be executed by the target assigned to each maintenance personnel from the mobile device end worn by each maintenance personnel, and set each two adjacent maintenance events in each maintenance event sequence as an execution transition node; comb the cross-distribution between the target maintenance operation and the operation blacklist between two adjacent maintenance events in each execution transition node; The transition warning risk assessment unit is used to receive the data in the cross-distribution information combing unit, calculate the transition warning risk value of each execution transition node, and complete the transition warning risk assessment of each execution transition node.

Images