Industrial equipment fault identification and maintenance system based on AR and AI technology

Abstract

The present invention belongs to the technical field of industrial equipment fault identification and operation and maintenance, and specifically discloses an industrial equipment fault identification and operation and maintenance system based on AR and AI technology, the system comprising: an AR model construction module, a monitoring node labeling module, an equipment operation monitoring module, a monitoring node identification module, a monitoring node confirmation module, a database and an operation and maintenance identification display terminal; the present invention performs fault triggering trend analysis through three dimensions of mechanical monitoring nodes, electrical monitoring nodes and operating environment monitoring nodes, and confirms the preventive maintenance level of the mechanical monitoring node by setting maintenance requirement weights according to the operation and maintenance levels of the electrical monitoring nodes and the operating environment monitoring nodes, fully considering the correlation between the monitoring nodes, and at the same time, by constructing a three-dimensional AR model, the operation and maintenance level and operation and maintenance information of each monitoring node are made into operation and maintenance labels and imported into the three-dimensional AR model, and corresponding fault display is performed, thereby improving the efficiency of fault location confirmation.

Description

Technical Field

[0001] The present invention belongs to the technical field of industrial equipment fault identification and operation and maintenance, and relates to an industrial equipment fault identification and operation and maintenance system based on AR and AI technologies. Background Art

[0002] Industrial equipment is a general term for a wide range of machinery used to support industrial production and processing processes, covering all aspects from raw material processing to finished product manufacturing. Therefore, in order to ensure the efficient operation of industrial equipment, fault identification and maintenance of industrial equipment are required.

[0003] The process of industrial equipment fault identification and maintenance mainly includes data collection, data processing, fault detection and diagnosis, and fault maintenance. There are still the following deficiencies: 1. Currently, during data processing, a single numerical comparison is mainly performed according to the set fixed fault indicators, and it is judged whether a fault exists, and no predictive maintenance analysis is performed. There are certain identification limitations, which leads to insufficient effectiveness of fault identification and maintenance.

[0004] 2. The current fault identification is a one-way identification, which does not take into account the correlation between components and faults, resulting in the inability to guarantee the accuracy of fault identification. At the same time, the maintenance decision is not targeted enough.

[0005] 3. Currently, the fault analysis results are displayed in the form of reports. A three-dimensional AR fault identification and maintenance model is not constructed based on the fault analysis results of the target equipment for fault display. This makes it difficult to quickly understand and find the fault location and fault handling methods, and it is impossible to highlight its detailed performance, making it difficult for subsequent staff to carry out corresponding maintenance and operation work, thereby increasing the tediousness of subsequent work. Summary of the invention

[0006] In view of this, in order to solve the problems raised in the above background technology, an industrial equipment fault identification and operation and maintenance system based on AR and AI technology is proposed.

[0007] The objective of the present invention can be achieved through the following technical solutions: The present invention provides an industrial equipment fault identification and operation and maintenance system based on AR and AI technology, the system comprising: an AR model construction module, used to record the industrial equipment to be fault identified and maintained as the target equipment, perform a three-dimensional scan on the target equipment, obtain the three-dimensional outline of the target equipment, and construct a three-dimensional AR model based on this.

[0008] The monitoring node marking module is used to set the fault monitoring nodes of the target equipment, including mechanical monitoring nodes, electrical monitoring nodes and operating environment monitoring nodes, and mark the corresponding monitoring nodes in the three-dimensional AR model.

[0009] The device operation monitoring module is used to monitor the operation of the target device and obtain the operation information of the target device in the current monitoring time period.

[0010] The monitoring node identification module is used to identify the operation and maintenance level of each monitoring node according to the operation information of the target device in the current monitoring time period, and mark the operation and maintenance level;

[0011] The monitoring node confirmation module is used to confirm the operation and maintenance information of each monitoring node according to the operation and maintenance level of each monitoring node.

[0012] The database is used to store reference fault triggering trends of target equipment at various operating times, store mechanical maintenance frequencies set for target equipment, and store associated electrical components of various electrical indicators.

[0013] The operation and maintenance identification and display terminal is used to convert the operation and maintenance level and operation and maintenance information of each monitoring node into an operation and maintenance label, and import it into the three-dimensional AR model for corresponding display.

[0014] Furthermore, the identification of the operation and maintenance level of each monitoring node includes: S1, identifying the monitoring index value of each mechanical index corresponding to each mechanical monitoring part from the operation information of the target equipment in the current monitoring time period as the fault assessment index of the mechanical monitoring node, and then calculating the mechanical fault triggering tendency degree. .

[0015] S2. Identify the monitoring index values ​​of each electrical index at each monitoring time point from the operating information of the target device in the current monitoring time period, and use them as the fault assessment index of the electrical monitoring node, and then calculate the electrical fault triggering trend. .

[0016] S3. Identify the monitoring index value of each operating environment index from the operating information of the target device in the current monitoring time period as the fault assessment index of the operating environment monitoring node, and then calculate the operating environment fault triggering tendency degree. .

[0017] S4. Setting each operation and maintenance level and the fault triggering trend interval of each operation and maintenance level, wherein the operation and maintenance level includes level one operation and maintenance, level two operation and maintenance, and level three operation and maintenance.

[0018] S5. Match and compare the electrical fault trigger trend with the fault trigger trend interval of each operation and maintenance level, and use the matched operation and maintenance level as the operation and maintenance level of the electrical monitoring node.

[0019] S6. The operation and maintenance levels of the mechanical gas monitoring node and the operating environment gas monitoring node are identified in the same manner as the operation and maintenance level of the electrical monitoring node.

[0020] Furthermore, the statistical mechanical fault triggering tendency degree includes: comparing the monitoring index value of each mechanical indicator corresponding to each mechanical monitoring part with the set reference index value interval of each mechanical indicator, counting the deviation index value of each mechanical indicator corresponding to each mechanical monitoring part, marking the mechanical indicator with a deviation index value not equal to 0 as an abnormal index, and screening out the maximum deviation value of each mechanical monitoring part from the deviation index values ​​of each mechanical indicator, recorded as , is the serial number of the mechanical monitoring part, .

[0021] Count the number of abnormal indicators of each mechanical monitoring part, recorded as , and the number of mechanical indicators of each mechanical monitoring part is recorded as .

[0022] Statistics of abnormal trends of each mechanical monitoring part , , and Set the The number ratio of abnormal indicators of each mechanical monitoring part and the reference deviation value are used to select the maximum abnormal trend degree of the mechanical monitoring part as the mechanical fault triggering trend degree. .

[0023] Furthermore, the statistical electrical fault triggering tendency degree includes: selecting the maximum monitoring index value of each electrical index from the monitoring index values ​​of each electrical index, recorded as , is the serial number of the electrical index, .

[0024] With the monitoring time point as the horizontal axis and the monitoring index value of each electrical index as the vertical axis, a change curve of each electrical index is constructed.

[0025] The horizontal distance and vertical distance between each fluctuation point are extracted from the change curve, and the minimum horizontal distance and the maximum vertical distance are screened out, which are recorded as and , and the number of fluctuation points is recorded as .

[0026] Will As the fluctuation abnormality of the change curve, , and The horizontal distance, vertical distance and number of fluctuation points of the set reference are respectively used to obtain the fluctuation abnormality of each electrical index. .

[0027] Statistical electrical fault triggering trend , , and The set The reference index value and reference fluctuation frequency of each electrical index.

[0028] Furthermore, the statistical operation environment fault triggering tendency degree includes: locating the monitoring index values ​​of sound and temperature from the monitoring index values ​​of each operation environment index, and recording them as and .

[0029] Statistical operating environment fault triggering tendency , , and They are the set sound reference index value and temperature reference index value respectively.

[0030] Furthermore, the confirmation of the operation and maintenance information of each monitoring node includes: Q1. When the operation and maintenance level of the mechanical monitoring node is level one, the operation and maintenance type is recorded as emergency maintenance, the emergency maintenance index is confirmed, and the operation and maintenance type and the emergency maintenance index are used as operation and maintenance information.

[0031] Q2. When the operation and maintenance level of the mechanical monitoring node is level 2, the operation and maintenance type is recorded as preventive maintenance, the preventive maintenance level of the mechanical monitoring node is confirmed, and the operation and maintenance type and preventive maintenance level are used as operation and maintenance information.

[0032] Q3. When the operation and maintenance level of the mechanical monitoring node is level three, the operation and maintenance type is recorded as predictive maintenance, the correction maintenance frequency of the mechanical monitoring node is confirmed, and the operation and maintenance type and the correction maintenance frequency are used as operation and maintenance information.

[0033] Q4. Confirm the operation and maintenance information of the electrical monitoring node according to the operation and maintenance level of the electrical monitoring node.

[0034] Q5. According to the operation and maintenance level of the operating environment monitoring node, the operation and maintenance information of the operating environment monitoring node is identified through the environmental operation and maintenance diagnosis model.

[0035] Furthermore, the confirmation of the emergency maintenance index includes: taking the machine monitoring parts with an abnormal tendency greater than or equal to 0 as emergency maintenance parts, counting the number of emergency maintenance parts, and comparing it with the number of machine monitoring parts to obtain the emergency maintenance part number ratio.

[0036] If the ratio of the number of emergency maintenance parts is greater than the set reference ratio of the number of emergency maintenance parts, overall maintenance is used as the maintenance mode, otherwise, local maintenance is used as the maintenance mode.

[0037] When the maintenance mode is overall maintenance, each mechanical monitoring part is regarded as a maintenance part, and the abnormal trend degree of each maintenance part is matched with the abnormal trend degree corresponding to each set maintenance level to obtain the maintenance level matched by each mechanical monitoring part, which is made into a maintenance label and added to each mechanical monitoring part.

[0038] The monitored parts after adding labels are grouped into maintenance part sets, and the overall maintenance and maintenance part sets are used as emergency maintenance indicators.

[0039] When the maintenance mode is local maintenance, each emergency maintenance part is taken as each maintenance part, and the emergency maintenance index of the local maintenance is obtained by analyzing it in the same way as the analysis method of the emergency maintenance index during overall maintenance.

[0040] Further, the determination of the preventive maintenance level of the mechanical monitoring node includes: if the operation and maintenance level of the electrical monitoring node is level 1 or level 2 and the operation and maintenance level of the operating environment monitoring node is level 3, setting the maintenance requirement weight to .

[0041] If the maintenance level of the electrical monitoring node is level 3 and the maintenance level of the operating environment monitoring node is level 1 or level 2, set the maintenance requirement weight to .

[0042] If the operation and maintenance level of the electrical monitoring node and the operating environment monitoring node are level 1 or level 2 respectively, set the maintenance requirement weight to .

[0043] If the operation and maintenance level of the electrical monitoring node and the operating environment monitoring node are both level three, set the maintenance requirement weight to , and thus obtain the maintenance demand weight , The value is or or or , .

[0044] Statistics of maintenance needs , , is a natural constant.

[0045] The maintenance requirement degree is matched and compared with the maintenance requirement degree interval corresponding to each preventive maintenance level to obtain the preventive maintenance level of the mechanical monitoring node.

[0046] Further, the confirming of the correction maintenance frequency of the mechanical monitoring node includes: extracting from a database a reference mechanical fault triggering trend of the target equipment under the current accumulated operating time.

[0047] The mechanical fault triggering trend degree is subtracted from the reference mechanical fault triggering trend degree, and the difference is recorded as the mechanical fault triggering trend degree difference.

[0048] If the mechanical fault trigger trend degree difference is greater than 0, the mechanical fault trigger trend degree is matched and compared with the corrected mechanical fault trigger trend degree corresponding to each corrected maintenance frequency to obtain the matched corrected maintenance frequency, and the matched corrected maintenance frequency is used as the corrected maintenance frequency of the mechanical monitoring node; otherwise, the mechanical maintenance frequency set for the target device is extracted from the database, and the set mechanical maintenance frequency is used as the corrected maintenance frequency of the mechanical monitoring node.

[0049] Furthermore, the confirmation of operation and maintenance information of the electrical monitoring node includes: matching and comparing the monitoring index value of each electrical index with its set reference index value interval, marking the electrical index whose monitoring index value is not within the reference index value interval as an abnormal electrical index, and then obtaining each abnormal electrical index.

[0050] The electrical components associated with each abnormal electrical indicator are extracted from the database.

[0051] When the operation and maintenance level of the electrical monitoring node is level one, the associated electrical components corresponding to the abnormal electrical indicators will be replaced as operation and maintenance information.

[0052] When the operation and maintenance level of the electrical monitoring node is level 2, the associated electrical components corresponding to the abnormal electrical indicators are maintained as operation and maintenance information.

[0053] When the operation and maintenance level of the electrical monitoring node is level three, the operation and maintenance information of the electrical monitoring node when the operation and maintenance level is level three is confirmed in the same way as the operation and maintenance information of the mechanical monitoring node when the operation and maintenance level is level three.

[0054] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention performs fault triggering trend analysis through three dimensions: mechanical monitoring nodes, electrical monitoring nodes, and operating environment monitoring nodes, avoiding the limitation of single numerical comparison and judgment of set fixed fault indicators, while improving the effectiveness of fault identification and maintenance.

[0055] When confirming the preventive maintenance level of the mechanical monitoring node, the present invention sets the maintenance requirement weight according to the operation and maintenance level of the electrical monitoring node and the operating environment monitoring node, fully considering the correlation between the monitoring nodes, thereby ensuring the accuracy of fault diagnosis and improving the pertinence of maintenance decisions.

[0056] The present invention constructs a three-dimensional AR model, and imports the operation and maintenance level and operation and maintenance information of each monitoring node into the three-dimensional AR model as an operation and maintenance label to display the corresponding faults, thereby improving the efficiency of fault location confirmation, and at the same time improving the rapid understanding of fault handling methods, enhancing the detailed expression of the fault location, and improving the subsequent maintenance and operation work of subsequent staff, thereby reducing the tediousness of subsequent work. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0058] Figure 1 It is a schematic diagram of the connection of each module of the system of the present invention. DETAILED DESCRIPTION

[0059] 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.

[0060] See also Figure 1 As shown, the present invention provides an industrial equipment fault identification and operation and maintenance system based on AR and AI technology, the system comprising: an AR model building module, a monitoring node labeling module, an equipment operation monitoring module, a monitoring node identification module, a monitoring node confirmation module, a database and an operation and maintenance identification display terminal.

[0061] In the above, the monitoring node labeling module is respectively connected to the AR model construction module and the equipment operation monitoring module, the monitoring node identification module is respectively connected to the equipment operation monitoring module, the monitoring node confirmation module and the operation and maintenance identification display terminal, and the monitoring node confirmation module is also respectively connected to the database and the operation and maintenance identification display terminal.

[0062] The AR model building module is used to record the industrial equipment to be operated and maintained for fault identification as the target equipment, perform three-dimensional scanning on the target equipment, obtain the three-dimensional outline of the target equipment, and build a three-dimensional AR model based on the three-dimensional outline.

[0063] The monitoring node marking module is used to set the fault monitoring nodes of the target equipment, including mechanical monitoring nodes, electrical monitoring nodes and operating environment monitoring nodes, and mark the corresponding monitoring nodes in the three-dimensional AR model.

[0064] The device operation monitoring module is used to monitor the operation of the target device and obtain the operation information of the target device in the current monitoring time period.

[0065] The monitoring node identification module is used to identify the operation and maintenance level of each monitoring node according to the operation information of the target device in the current monitoring time period, and to mark the operation and maintenance level.

[0066] Exemplarily, the identification of the operation and maintenance level of each monitoring node includes: S1, identifying the monitoring index value of each mechanical index corresponding to each mechanical monitoring part from the operation information of the target equipment in the current monitoring time period as the fault assessment index of the mechanical monitoring node, and then calculating the mechanical fault triggering tendency degree. .

[0067] It should be added that the mechanical monitoring parts include but are not limited to bearings, blades, gear boxes and engine casings.

[0068] It should be added that the mechanical indicators include but are not limited to wear and corrosion, and the monitoring index values ​​of the mechanical indicators include but are not limited to the number of mechanical wear points, the maximum mechanical wear depth, the total mechanical wear area, the number of rust points and the total area of ​​rust points.

[0069] Furthermore, the statistical mechanical fault triggering tendency degree includes: S1-1, comparing the monitoring index value of each mechanical indicator corresponding to each mechanical monitoring part with the set reference index value interval of each mechanical indicator, counting the deviation index value of each mechanical indicator corresponding to each mechanical monitoring part, marking the mechanical indicator with a deviation index value not equal to 0 as an abnormal index, and screening out the maximum deviation value of each mechanical monitoring part from the deviation index values ​​of each mechanical indicator, recorded as , is the serial number of the mechanical monitoring part, .

[0070] S1-2. Count the number of abnormal indicators of each mechanical monitoring part, recorded as , and the number of mechanical indicators of each mechanical monitoring part is recorded as .

[0071] It should be added that the specific statistical process of counting the deviation values ​​of each abnormal indicator corresponding to each mechanical monitoring part is as follows: if the indicator value of the abnormal indicator of a certain mechanical indicator is greater than the upper limit value of its reference indicator value interval, the difference between the two is taken as the deviation value of the abnormal indicator of the mechanical indicator.

[0072] If the index value of the abnormal index of a certain mechanical index is less than the lower limit of its reference index value interval, the two are subtracted, and the absolute value of the difference is used as the deviation value of the abnormal index of the mechanical index, and then the deviation value of each abnormal index corresponding to each mechanical monitoring part is statistically obtained.

[0073] S1-3. Statistical analysis of abnormal trends of each mechanical monitoring part , , and Set the The number ratio of abnormal indicators of each mechanical monitoring part and the reference deviation value are used to select the maximum abnormal trend degree of the mechanical monitoring part as the mechanical fault triggering trend degree. .

[0074] It should be supplemented that the ratio of the number of abnormal indicators of each mechanical monitoring part is obtained by setting the weight of its impact on the operation of the target equipment, among which the weight of the blade is greater than the weight of the bearing, which is greater than the weight of the gearbox, which is greater than the weight of the engine casing. Therefore, when the mechanical monitoring part is the blade, the ratio of the number of abnormal indicators can be specifically taken as 0.6, when the mechanical monitoring part is the bearing, the ratio of the number of abnormal indicators can be specifically taken as 0.5, when the mechanical monitoring part is the gearbox, the ratio of the number of abnormal indicators can be specifically taken as 0.4, and when the mechanical monitoring part is the engine casing, the ratio of the number of abnormal indicators can be specifically taken as 0.3. The reference deviation values ​​of each mechanical monitoring part are extracted from the technical specification table of fault identification and operation and maintenance of industrial equipment.

[0075] S2. Identify the monitoring index values ​​of each electrical index at each monitoring time point from the operating information of the target device in the current monitoring time period, and use them as the fault assessment index of the electrical monitoring node, and then calculate the electrical fault triggering trend. .

[0076] It should be supplemented that the associated electrical indicators corresponding to the electrical monitoring nodes include current, voltage and resistance.

[0077] Furthermore, the statistical electrical fault triggering tendency degree includes: S2-1, selecting the maximum monitoring index value of each electrical index from the monitoring index values ​​of each electrical index, and recording it as , is the serial number of the electrical index, .

[0078] S2-2. Construct a change curve of each electrical indicator by taking the monitoring time point as the horizontal axis and the monitoring index value of each electrical indicator as the vertical axis.

[0079] S2-3, extract the horizontal distance and vertical distance between each fluctuation point from the change curve, and select the minimum horizontal distance and the maximum vertical distance, which are recorded as and , and the number of fluctuation points is recorded as .

[0080] S2-4, As the fluctuation abnormality of the change curve, , and The horizontal distance, vertical distance and number of fluctuation points of the set reference are respectively used to obtain the fluctuation abnormality of each electrical index. .

[0081] It should be added that the horizontal distance between the fluctuation points describes the fluctuation frequency. The larger the horizontal distance, the smaller the fluctuation frequency. The vertical distance between the fluctuation points describes the fluctuation amplitude. The larger the vertical distance, the larger the fluctuation amplitude. When the fluctuation frequency and fluctuation amplitude are larger, the fluctuation is more obvious, and thus more abnormal.

[0082] It should also be added that the horizontal distance is set according to the total length of the curve. Specifically, the curve can be divided into three equal parts. When the smallest horizontal distance is greater than one-third of the curve length, it indicates that the electrical index fluctuation is normal at this time. The vertical distance is set according to the total length of the curve. Specifically, the curve can be divided into three equal parts. When the largest vertical distance is less than one-third of the curve length, it indicates that the electrical index fluctuation is normal at this time.

[0083] It should be added that when the number of fluctuation points is 0, it indicates that the fluctuation of the electrical indicators is in a normal state. For the convenience of analysis, the number of fluctuation points can be specifically set to 1.

[0084] S2-5. Statistical electrical fault triggering trend , , and The set The reference index value and reference fluctuation frequency of each electrical index.

[0085] It should be added that and They are all extracted from the technical specification table of industrial equipment fault identification and operation and maintenance.

[0086] S3. Identify the monitoring index value of each operating environment index from the operating information of the target device in the current monitoring time period as the fault assessment index of the operating environment monitoring node, and then calculate the operating environment fault triggering tendency degree. .

[0087] Furthermore, the statistical operation environment fault triggering tendency degree includes: S3-1, locating the monitoring index values ​​of sound and temperature from the monitoring index values ​​of each operation environment index, and recording them as and .

[0088] It should be added that the monitoring index value of sound is loudness, and the monitoring index value of temperature is temperature value.

[0089] S3-2. Statistical operating environment fault triggering tendency , , and They are the set sound reference index value and temperature reference index value respectively.

[0090] It should be added that and They are all extracted from the technical specification table of industrial equipment fault identification and operation and maintenance.

[0091] S4. Setting each operation and maintenance level and the fault triggering trend interval of each operation and maintenance level, wherein the operation and maintenance level includes level one operation and maintenance, level two operation and maintenance, and level three operation and maintenance.

[0092] It should be added that the setting of the operation and maintenance levels as level 1, level 2 and level 3 is based on the analysis of three situations: actual trigger, critical trigger and critical trigger tendency. The value range of level 1 operation and maintenance is ,,The value range of secondary operation and maintenance is , the value range of the third-level operation and maintenance is .

[0093] S5. Match and compare the electrical fault trigger trend with the fault trigger trend interval of each operation and maintenance level, and use the matched operation and maintenance level as the operation and maintenance level of the electrical monitoring node.

[0094] S6. The operation and maintenance levels of the mechanical gas monitoring node and the operating environment gas monitoring node are identified in the same manner as the operation and maintenance level of the electrical monitoring node.

[0095] The embodiment of the present invention performs fault triggering trend analysis through three dimensions of mechanical monitoring nodes, electrical monitoring nodes and operating environment monitoring nodes, avoiding the limitation of single numerical comparison and judgment of set fixed fault indicators, and improving the effectiveness of fault identification and operation and maintenance.

[0096] The monitoring node confirmation module is used to confirm the operation and maintenance information of each monitoring node according to the operation and maintenance level of each monitoring node.

[0097] Exemplarily, the confirmation of the operation and maintenance information of each monitoring node includes: Q1. When the operation and maintenance level of the mechanical monitoring node is level one, the operation and maintenance type is recorded as emergency maintenance, the emergency maintenance index is confirmed, and the operation and maintenance type and the emergency maintenance index are used as operation and maintenance information.

[0098] Furthermore, the confirmation of the emergency maintenance index includes: Q1-1, taking the mechanical monitoring parts with abnormal tendency greater than or equal to 0 as emergency maintenance parts, counting the number of emergency maintenance parts, and comparing it with the number of mechanical monitoring parts to obtain the emergency maintenance part number ratio.

[0099] Q1-2. If the ratio of the number of emergency maintenance parts is greater than the set reference ratio of the number of emergency maintenance parts, overall maintenance is used as the maintenance mode; otherwise, local maintenance is used as the maintenance mode.

[0100] Q1-3. When the maintenance mode is overall maintenance, each mechanical monitoring part is regarded as a maintenance part, and the abnormal trend degree of each maintenance part is matched with the abnormal trend degree corresponding to each set maintenance level to obtain the maintenance level matched by each mechanical monitoring part, which is made into a maintenance label and added to each mechanical monitoring part.

[0101] Q1-4. The monitored parts with labels added are grouped into maintenance part sets, and the overall maintenance and maintenance part sets are used as emergency maintenance indicators.

[0102] Q1-5. When the maintenance mode is local maintenance, each emergency maintenance part is taken as each maintenance part, and the emergency maintenance index of the local maintenance is obtained by analyzing it in the same way as the analysis method of the emergency maintenance index during overall maintenance.

[0103] Further, the preventive maintenance level of the mechanical monitoring node is determined, including: Q2-1, if the operation and maintenance level of the electrical monitoring node is level 1 or level 2 and the operation and maintenance level of the operating environment monitoring node is level 3, the maintenance requirement weight is set to .

[0104] Q2-2. If the operation and maintenance level of the electrical monitoring node is level 3 and the operation and maintenance level of the operating environment monitoring node is level 1 or level 2, set the maintenance requirement weight to .

[0105] Q2-3. If the operation and maintenance level of the electrical monitoring node and the operating environment monitoring node are level 1 or level 2 respectively, set the maintenance requirement weight to .

[0106] Q2-4. If the operation and maintenance level of the electrical monitoring node and the operating environment monitoring node are both level 3, set the maintenance requirement weight to , and thus obtain the maintenance demand weight , The value is or or or , .

[0107] It should be added that the setting This is because an electrical failure may cause the machinery to stop working immediately or cause a serious failure, while changes in the operating environment usually have a gradual impact on the machinery and may take a long time to appear.

[0108] Q2-5. Statistics on maintenance needs , , is a natural constant.

[0109] Q2-6. Match and compare the maintenance requirement degree with the maintenance requirement degree interval corresponding to each preventive maintenance level to obtain the preventive maintenance level of the mechanical monitoring node.

[0110] Q3. When the operation and maintenance level of the mechanical monitoring node is level three, the operation and maintenance type is recorded as predictive maintenance, the correction maintenance frequency of the mechanical monitoring node is confirmed, and the operation and maintenance type and the correction maintenance frequency are used as operation and maintenance information.

[0111] The embodiment of the present invention performs fault triggering trend analysis based on monitoring indicators, and further identifies each monitoring node. When confirming the preventive maintenance level of the mechanical monitoring node, the maintenance requirement weight is set according to the operation and maintenance level of the electrical monitoring node and the operating environment monitoring node, and the correlation between the monitoring nodes is fully considered, thereby ensuring the accuracy of fault diagnosis and improving the pertinence of maintenance decisions.

[0112] Furthermore, the confirmation of the correction maintenance frequency of the mechanical monitoring node includes: Q3-1, extracting from the database a reference mechanical fault triggering trend of the target equipment under the current accumulated operating time.

[0113] Q3-2. Subtract the mechanical fault trigger trend from the reference mechanical fault trigger trend, and record the difference as the mechanical fault trigger trend difference.

[0114] Q3-3. If the mechanical fault trigger trend degree difference is greater than 0, the mechanical fault trigger trend degree is matched and compared with the corrected mechanical fault trigger trend degree corresponding to each corrected maintenance frequency to obtain the matched corrected maintenance frequency, and the matched corrected maintenance frequency is used as the corrected maintenance frequency of the mechanical monitoring node. Otherwise, the mechanical maintenance frequency set for the target device is extracted from the database, and the set mechanical maintenance frequency is used as the corrected maintenance frequency of the mechanical monitoring node.

[0115] Q4. Confirm the operation and maintenance information of the electrical monitoring node according to the operation and maintenance level of the electrical monitoring node.

[0116] Furthermore, the confirmation of the operation and maintenance information of the electrical monitoring node includes: Q4-1, matching and comparing the monitoring index value of each electrical indicator with its set reference index value interval, marking the electrical indicator whose monitoring index value is not within the reference index value interval as an abnormal electrical indicator, and then obtaining each abnormal electrical indicator.

[0117] Q4-2. Extract the electrical components associated with each abnormal electrical indicator from the database.

[0118] Q4-3. When the operation and maintenance level of the electrical monitoring node is level one, the associated electrical components corresponding to the abnormal electrical indicators will be replaced as operation and maintenance information.

[0119] Q4-4. When the operation and maintenance level of the electrical monitoring node is level 2, the associated electrical components corresponding to the abnormal electrical indicators are maintained as operation and maintenance information.

[0120] It should be supplemented that the electrical components whose electrical indicators are current include but are not limited to generators, starters and wires, the electrical components whose electrical indicators are voltage include but are not limited to generators, voltage regulators and voltage sensors, and the electrical components whose electrical indicators are resistance include but are not limited to resistors, actuators and circuit breakers.

[0121] Q4-5. When the operation and maintenance level of the electrical monitoring node is level three, the operation and maintenance information of the electrical monitoring node when the operation and maintenance level is level three shall be confirmed in the same way as the operation and maintenance information of the mechanical monitoring node when the operation and maintenance level is level three.

[0122] Q5. According to the operation and maintenance level of the operating environment monitoring node, the operation and maintenance information of the operating environment monitoring node is identified through the environmental operation and maintenance diagnosis model.

[0123] It should be added that the specific diagnostic process of the environmental operation and maintenance diagnostic model is as follows: , as the fault trigger tendency of sound , and Fault trigger tendency as temperature .

[0124] When the operation and maintenance level of the operating environment monitoring node is level 1, if , lubricating oil will be added as operation and maintenance information.

[0125] like , then coolant will be added as operation and maintenance information.

[0126] like and , then the lubricating oil and coolant operation and maintenance information will be added.

[0127] When the operation and maintenance level of the operating environment monitoring node is level 2, the fault triggering tendency of sound and temperature is matched and compared with the value range of each operation and maintenance level.

[0128] like It is within the range of the second-level operation and maintenance. If the value is within the range of level 3 operation and maintenance, dust cleaning is used as the operation and maintenance information.

[0129] like It is within the range of the third-level operation and maintenance. If it is within the value range of the second-level operation and maintenance, cleaning the radiator is used as the operation and maintenance information.

[0130] like and If both are within the value range of the second-level operation and maintenance, dust cleaning and radiator cleaning are taken as the operation and maintenance information.

[0131] When the operation and maintenance level of the operating environment monitoring node is level three, the operation and maintenance information of the operating environment monitoring node when the operation and maintenance level is level three is confirmed in the same way as the confirmation method of the operation and maintenance information of the mechanical monitoring node when the operation and maintenance level is level three.

[0132] The embodiment of the present invention constructs a three-dimensional AR model, and imports the operation and maintenance level and operation and maintenance information of each monitoring node into an operation and maintenance label into the three-dimensional AR model to display the corresponding faults, thereby improving the efficiency of confirming the fault location, and at the same time improving the rapid understanding of fault handling methods, enhancing the detailed expression of the fault location, and improving the subsequent staff's corresponding maintenance and operation work, thereby reducing the tediousness of subsequent work.

[0133] The database is used to store reference fault triggering trends of target equipment at various operating times, store mechanical maintenance frequencies set for target equipment, and store associated electrical components of various electrical indicators.

[0134] The operation and maintenance identification and display terminal is used to convert the operation and maintenance level and operation and maintenance information of each monitoring node into an operation and maintenance label, and import it into the three-dimensional AR model for corresponding display.

[0135] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. Industrial equipment fault identification and maintenance system based on AR and AI technology, characterized by: The system includes: The AR model building module is used to record the industrial equipment to be fault identified and maintained as the target equipment, perform three-dimensional scanning on the target equipment, obtain the three-dimensional outline of the target equipment, and build a three-dimensional AR model based on it; A monitoring node annotation module is used to set the fault monitoring nodes of the target equipment, including mechanical monitoring nodes, electrical monitoring nodes and operating environment monitoring nodes, and annotate the corresponding monitoring nodes in the three-dimensional AR model; The device operation monitoring module is used to monitor the operation of the target device and obtain the operation information of the target device in the current monitoring time period; The monitoring node identification module is used to identify the operation and maintenance level of each monitoring node according to the operation information of the target device in the current monitoring time period, and mark the operation and maintenance level; The monitoring node confirmation module is used to confirm the operation and maintenance information of each monitoring node according to the operation and maintenance level of each monitoring node; A database for storing reference fault triggering trends of target equipment at various operating times, storing mechanical maintenance frequencies set for target equipment, and storing associated electrical components of various electrical indicators; The operation and maintenance identification and display terminal is used to convert the operation and maintenance level and operation and maintenance information of each monitoring node into an operation and maintenance label, and import it into the 3D AR model for corresponding display; The confirming of the operation and maintenance information of each monitoring node includes: Q1. When the operation and maintenance level of the mechanical monitoring node is level 1, the operation and maintenance type is recorded as emergency maintenance, the emergency maintenance index is confirmed, and the operation and maintenance type and emergency maintenance index are used as operation and maintenance information; Q2. When the operation and maintenance level of the mechanical monitoring node is level 2, the operation and maintenance type is recorded as preventive maintenance, the preventive maintenance level of the mechanical monitoring node is confirmed, and the operation and maintenance type and preventive maintenance level are used as operation and maintenance information; Q3. When the operation and maintenance level of the mechanical monitoring node is level 3, the operation and maintenance type is recorded as predictive maintenance, the corrective maintenance frequency of the mechanical monitoring node is confirmed, and the operation and maintenance type and corrective maintenance frequency are used as operation and maintenance information; Q4. Confirm the operation and maintenance information of the electrical monitoring node according to the operation and maintenance level of the electrical monitoring node; Q5. According to the operation and maintenance level of the operation environment monitoring node, the operation and maintenance information of the operation environment monitoring node is identified through the environment operation and maintenance diagnosis model; The confirmation of the preventive maintenance level of the mechanical monitoring node includes: If the maintenance level of the electrical monitoring node is level 1 or 2 and the maintenance level of the operating environment monitoring node is level 3, set the maintenance requirement weight to ; If the maintenance level of the electrical monitoring node is level 3 and the maintenance level of the operating environment monitoring node is level 1 or level 2, set the maintenance requirement weight to ; If the operation and maintenance level of the electrical monitoring node and the operating environment monitoring node are level 1 or level 2 respectively, set the maintenance requirement weight to ; If the operation and maintenance level of the electrical monitoring node and the operating environment monitoring node are both level three, set the maintenance requirement weight to , and thus obtain the maintenance demand weight , The value is or or or , ; Statistics of maintenance needs , , is a natural constant, Indicates the mechanical fault triggering tendency; The maintenance requirement degree is matched and compared with the maintenance requirement degree interval corresponding to each preventive maintenance level to obtain the preventive maintenance level of the mechanical monitoring node.

2. The industrial equipment fault identification and operation and maintenance system based on AR and AI technology according to claim 1 is characterized by: The identifying the operation and maintenance level of each monitoring node includes: S1. Identify the monitoring index values ​​of each mechanical index corresponding to each mechanical monitoring part from the operating information of the target equipment in the current monitoring time period, and use them as the fault assessment index of the mechanical monitoring node, and then calculate the mechanical fault triggering trend degree. ; S2. Identify the monitoring index values ​​of each electrical index at each monitoring time point from the operating information of the target device in the current monitoring time period, and use them as the fault assessment index of the electrical monitoring node, and then calculate the electrical fault triggering trend. ; S3. Identify the monitoring index value of each operating environment index from the operating information of the target device in the current monitoring time period as the fault assessment index of the operating environment monitoring node, and then calculate the operating environment fault triggering tendency degree. ; S4. Setting each operation and maintenance level and the fault triggering trend interval of each operation and maintenance level, wherein the operation and maintenance level includes level 1 operation and maintenance, level 2 operation and maintenance, and level 3 operation and maintenance; S5. Match and compare the electrical fault trigger trend with the fault trigger trend interval of each operation and maintenance level, and use the matched operation and maintenance level as the operation and maintenance level of the electrical monitoring node; S6. The operation and maintenance levels of the mechanical monitoring nodes and the operating environment monitoring nodes are identified in the same manner as the operation and maintenance level of the electrical monitoring nodes.

3. The industrial equipment fault identification and operation and maintenance system based on AR and AI technology according to claim 2 is characterized by: The statistical mechanical fault triggering tendency includes: The monitoring index values ​​of each mechanical monitoring part corresponding to each mechanical index are compared with the reference index value intervals of each mechanical index, and the deviation index values ​​of each mechanical monitoring part corresponding to each mechanical index are counted. The mechanical index with a deviation index value not equal to 0 is marked as an abnormal index, and the maximum deviation value of each mechanical monitoring part is selected from the deviation index values ​​of each mechanical index, which is recorded as , is the serial number of the mechanical monitoring part, ; Count the number of abnormal indicators of each mechanical monitoring part, recorded as , and the number of mechanical indicators of each mechanical monitoring part is recorded as ; Statistics of abnormal trends of each mechanical monitoring part , , and Set the The number ratio of abnormal indicators of each mechanical monitoring part and the reference deviation value are used to select the maximum abnormal trend degree of the mechanical monitoring part as the mechanical fault triggering trend degree. .

4. The industrial equipment fault identification and operation and maintenance system based on AR and AI technology according to claim 2 is characterized by: The statistical electrical fault triggering trend includes: The maximum monitoring index value of each electrical index is selected from the monitoring index values ​​of each electrical index, and recorded as , is the serial number of the electrical index, ; With the monitoring time point as the horizontal coordinate and the monitoring index value of each electrical index as the vertical coordinate, the change curve of each electrical index is constructed; The horizontal distance and vertical distance between each fluctuation point are extracted from the change curve, and the minimum horizontal distance and the maximum vertical distance are screened out, which are recorded as and , and the number of fluctuation points is recorded as ; Will As the fluctuation abnormality of the change curve, , and The horizontal distance, vertical distance and number of fluctuation points of the set reference are respectively used to obtain the fluctuation abnormality of each electrical index. ; Statistical electrical fault triggering trend , , and The set The reference index value and reference fluctuation frequency of each electrical index.

5. The industrial equipment fault identification and operation and maintenance system based on AR and AI technology according to claim 2 is characterized in that: The statistical operation environment fault triggering tendency degree includes: The monitoring index values ​​of sound and temperature are located from the monitoring index values ​​of each operating environment index and recorded as and ; Statistical operating environment fault triggering tendency , , and They are the set sound reference index value and temperature reference index value respectively.

6. The industrial equipment fault identification and operation and maintenance system based on AR and AI technology according to claim 3 is characterized by: The indicators for confirming emergency maintenance include: The machine monitoring parts with abnormal tendency greater than or equal to 0 are regarded as emergency maintenance parts, the number of emergency maintenance parts is counted, and compared with the number of machine monitoring parts to obtain the emergency maintenance part number ratio; If the ratio of the number of emergency maintenance parts is greater than the set reference ratio of the number of emergency maintenance parts, the overall maintenance is used as the maintenance mode, otherwise, the local maintenance is used as the maintenance mode; When the maintenance mode is overall maintenance, each machine monitoring part is used as each maintenance part, and the abnormal tendency degree of each maintenance part is matched with the abnormal tendency degree corresponding to each maintenance level, so as to obtain the maintenance level matched by each machine monitoring part, make it into a maintenance label, and add it to each machine monitoring part; The monitored parts with labels are grouped into maintenance part sets, and the overall maintenance and maintenance part sets are used as emergency maintenance indicators; When the maintenance mode is local maintenance, each emergency maintenance part is taken as each maintenance part, and the emergency maintenance index of the local maintenance is obtained by analyzing it in the same way as the analysis method of the emergency maintenance index during overall maintenance.

7. The industrial equipment fault identification and operation and maintenance system based on AR and AI technology according to claim 6 is characterized by: The confirmation of the corrective maintenance frequency of the mechanical monitoring nodes includes: Extracting the reference mechanical fault triggering trend of the target equipment under the current accumulated running time from the database; Subtract the mechanical fault triggering trend degree from the reference mechanical fault triggering trend degree, and record the difference as the mechanical fault triggering trend degree difference; If the mechanical fault trigger trend degree difference is greater than 0, the mechanical fault trigger trend degree is matched and compared with the corrected mechanical fault trigger trend degree corresponding to each corrected maintenance frequency to obtain the matched corrected maintenance frequency, and the matched corrected maintenance frequency is used as the corrected maintenance frequency of the mechanical monitoring node; otherwise, the mechanical maintenance frequency set for the target device is extracted from the database, and the set mechanical maintenance frequency is used as the corrected maintenance frequency of the mechanical monitoring node.

8. The industrial equipment fault identification and operation and maintenance system based on AR and AI technology according to claim 1 is characterized in that: The confirming the operation and maintenance information of the electrical monitoring node includes: The monitoring index value of each electrical index is matched and compared with the reference index value interval set therein, and the electrical index whose monitoring index value is not within the reference index value interval is marked as an abnormal electrical index, thereby obtaining each abnormal electrical index; Extracting electrical components associated with each abnormal electrical indicator from a database; When the operation and maintenance level of the electrical monitoring node is level one, the associated electrical components corresponding to the abnormal electrical indicators are replaced as operation and maintenance information; When the operation and maintenance level of the electrical monitoring node is level 2, the associated electrical components corresponding to the abnormal electrical indicators are maintained as operation and maintenance information; When the operation and maintenance level of the electrical monitoring node is level three, the operation and maintenance information of the electrical monitoring node when the operation and maintenance level is level three is confirmed in the same way as the operation and maintenance information of the mechanical monitoring node when the operation and maintenance level is level three.

Images