Intelligent recommendation system and method based on artificial intelligence large model

Abstract

The embodiment of the invention provides an intelligent recommendation method based on an artificial intelligence large model, and relates to the technical field of intelligent recommendation technologies. The method comprises the steps of obtaining dialogue information input by a user; the dialogue information is analyzed through a preset first model, equipment information and fault features are extracted according to an analysis result, and the equipment information comprises an equipment key level; acquiring working information of the target equipment according to the equipment information; carrying out fault detection according to the working information and the fault features, triggering scheme matching under the condition that a fault detection result meets a first condition, and enabling a first model to carry out fault scheme matching according to the working information and the fault features to obtain a first scheme; and recommending the first scheme to a target terminal, and indicating the target terminal to display the first scheme. According to the invention, the problem of low maintenance scheme recommendation precision is solved, and the effect of improving the maintenance scheme recommendation precision is achieved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of maintenance plan recommendation, and more specifically, to an intelligent recommendation system and method based on an artificial intelligence large model. Background Art

[0002] Existing full life cycle health management of key equipment such as port shore power, wind power generation, and chemical production lines usually relies on a single signal (such as temperature or vibration) for threshold alarms. It lacks the integrated analysis of multi-source heterogeneous data such as electrical, vibration, and temperature, resulting in high rates of missed reports and false alarms. It also ignores economic factors such as maintenance time, downtime losses, and spare parts costs, resulting in excessive maintenance of high-value equipment or insufficient maintenance of low-value equipment.

[0003] To address the above issues, a new maintenance plan recommendation system is needed to enhance the fusion analysis of heterogeneous data, improve the accuracy of the plan and reduce costs. Summary of the Invention

[0004] The embodiments of the present invention provide an intelligent recommendation system and method based on an artificial intelligence large model to at least solve the problem of low accuracy in maintenance plan recommendations in related technologies.

[0005] According to one embodiment of the present invention, an intelligent recommendation method based on an artificial intelligence large model is provided, comprising: Acquiring dialogue information input by the user, wherein the dialogue information includes text information and / or image information describing a fault of the target device; Parsing the conversation information using a preset first model, and extracting device information and fault characteristics based on the parsing results, wherein the device information includes a device criticality level; Acquire operating information of the target device according to the device information, wherein the operating information includes electrical parameters, vibration parameters, and temperature timing parameters; performing fault detection according to the operating information and the fault characteristics, triggering solution matching when a fault detection result meets a first condition, and causing the first model to perform fault solution matching according to the operating information and the fault characteristics to obtain a first solution; The first solution is recommended to a target terminal, and the target terminal is instructed to display the first solution.

[0006] In an exemplary embodiment, the first model matches a fault solution according to the operating information and the fault characteristics to obtain a first solution, including: Perform fault matching and state similarity calculation based on the electrical parameters, the vibration parameters, the temperature time series, and the fault characteristics; Determine the initial solution based on the fault matching results and the state similarity calculation results; The initial solution is screened and adjusted according to a preset cost constraint to obtain the first solution.

[0007] In an exemplary embodiment, performing fault matching according to the electrical parameters, the vibration parameters, the temperature sequence, and the fault characteristics includes: Determining a fault criterion based on the fault characteristics, wherein the fault criterion is used to indicate a probability range of a device fault corresponding to the fault characteristics; Calculating a fault probability based on the electrical parameters, the vibration parameters, and the temperature time series; The failure probability is compared with the failure criterion, and when the failure probability meets the second condition, the first solution is determined according to the failure probability and state similarity; otherwise, the failure criterion is re-determined according to the failure characteristics.

[0008] In an exemplary embodiment, the calculating of state similarity based on the electrical parameters, the vibration parameters, the temperature time series, and the fault characteristics includes: Determining a similarity standard based on the fault characteristics, wherein the similarity standard is used to indicate a similarity range of the device fault corresponding to the fault characteristics; Calculating state similarity based on the electrical parameters, the vibration parameters, and the temperature time series; The similarity value is compared with the similarity standard, and when the fault probability meets the third condition, the first solution is determined according to the fault probability and the state similarity; otherwise, the similarity standard is re-determined according to the fault feature.

[0009] In an exemplary embodiment, after determining the initial solution according to the fault matching result and the state similarity calculation result, the method further includes: determining a maintenance risk value based on the work information and the maintenance duration included in the initial plan; Determining a recommended value of the initial solution based on the maintenance risk value, the similarity calculation result, and the cost constraint; The initial solution with the largest recommendation value is determined as the first solution.

[0010] According to another embodiment of the present invention, an intelligent recommendation system based on an artificial intelligence large model is provided, comprising: A conversation collection module is used to obtain conversation information input by the user, wherein the conversation information includes text information and / or image information describing the fault of the target device; a semantic parsing module, configured to parse the conversation information using a preset first model and extract device information and fault characteristics based on the parsing results, wherein the device information includes a device criticality level; A working information acquisition module, configured to obtain working information of a target device according to the device information, wherein the working information includes electrical parameters, vibration parameters, and temperature timing parameters; a matching module, configured to perform fault detection based on the operating information and the fault characteristics, and trigger solution matching when a fault detection result meets a first condition, so that the first model performs fault solution matching based on the operating information and the fault characteristics to obtain a first solution; The recommendation module is configured to recommend the first solution to a target terminal and instruct the target terminal to display the first solution.

[0011] In an exemplary embodiment, the first model matches a fault solution according to the operating information and the fault characteristics to obtain a first solution, including: Perform fault matching and state similarity calculation based on the electrical parameters, the vibration parameters, the temperature time series, and the fault characteristics; Determine the initial solution based on the fault matching results and the state similarity calculation results; The initial solution is screened and adjusted according to a preset cost constraint to obtain the first solution.

[0012] In an exemplary embodiment, performing fault matching according to the electrical parameters, the vibration parameters, the temperature sequence, and the fault characteristics includes: Determining a fault criterion based on the fault characteristics, wherein the fault criterion is used to indicate a probability range of a device fault corresponding to the fault characteristics; Calculating a fault probability based on the electrical parameters, the vibration parameters, and the temperature time series; The failure probability is compared with the failure criterion, and when the failure probability meets the second condition, the first solution is determined according to the failure probability and state similarity; otherwise, the failure criterion is re-determined according to the failure characteristics.

[0013] According to yet another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored. The computer program is configured to execute the steps of any one of the above method embodiments when run.

[0014] According to another embodiment of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any one of the above method embodiments.

[0015] Through the present invention, since the dialogue extraction of fault content and the like is combined with data analysis and the actual perception of the staff, the accuracy of the maintenance plan recommendation is greatly improved. Therefore, the problem of low accuracy of the maintenance plan recommendation can be solved, and the effect of improving the accuracy of the maintenance plan recommendation is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of an intelligent recommendation method based on an artificial intelligence large model according to an embodiment of the present invention; Figure 2 This is a structural block diagram of an intelligent recommendation system based on an artificial intelligence large model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments.

[0018] Hereinafter, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature identified with "first," "second," etc., may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0019] In addition, in this application, directional terms such as "up", "down", "left", and "right" may be defined including but not limited to the orientation relative to the schematic placement of the components in the drawings. It should be understood that these directional terms may be relative concepts. They are used for relative descriptions and clarifications, and they may change accordingly according to changes in the orientation of the components in the drawings.

[0020] In this application, unless otherwise specified or limited, the term "connection" should be understood broadly. For example, "connection" can mean fixed connection, detachable connection, or integration; it can mean direct connection or indirect connection through an intermediate medium. In addition, the term "coupling" can refer to the manner in which electrical connection is achieved for signal transmission.

[0021] As used herein, "about," "substantially," or "approximately" includes the stated value and an average value that is within an acceptable range of deviation from the particular value as determined by one of ordinary skill in the art taking into account the measurements in question and the errors associated with the measurement of the particular quantity (i.e., the limitations of the measurement system).

[0022] In this embodiment, an intelligent recommendation method based on an artificial intelligence large model is provided. Figure 1 This is a flow chart of an intelligent recommendation method based on an artificial intelligence large model according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps: Step S11, obtaining dialogue information input by the user, wherein the dialogue information includes text information and / or image information describing the fault of the target device; In this embodiment, the user continuously inputs the fault condition of the target device into the dialogue interface according to the actual situation, and the dialogue robot inquires and analyzes the fault condition. For example, the user inputs "the sealing temperature of the packaging machine suddenly drops", and then the dialogue robot responds "please enter the location and model (or number) of the packaging machine, and the temperature drop rate is how many seconds per degree?" and so on; or the user can also directly upload photos or audio and video of the equipment failure into the dialogue interface, and then the dialogue robot analyzes the photos and / or audio and video.

[0023] Step S12: parsing the conversation information using a preset first model, and extracting device information and fault characteristics based on the parsing results, wherein the device information includes a device criticality level; In this embodiment, when the conversation information contains only text information, a semantic model can be used for parsing. When the conversation information contains image information, the semantic information and image recognition model (such as the YOLOv8-Seg model) are used together for parsing. Among them, the equipment information includes the equipment name (such as a packaging machine, etc.), the equipment model / number (such as MQ1035 or #M0105, etc.), the equipment purpose (such as heating, processing, drilling, cutting, etc.), the equipment criticality (the criticality can be set to 1-5 levels, with level 5 being the highest level. This criticality involves the economic loss per unit time during equipment maintenance), etc.; the fault characteristics include temperature drop, abnormal vibration, abnormal noise, and other fault characteristics used to indicate equipment failure.

[0024] Step S13, acquiring operating information of the target device according to the device information, wherein the operating information includes electrical parameters, vibration parameters, and temperature timing parameters; In this embodiment, after obtaining the device information, data collected by the sensors of the device are retrieved and cluster analysis is performed on the data, thereby facilitating subsequent fault analysis based on the data.

[0025] Among them, electrical parameters include current unbalance rate, which is used to indicate motor winding fault ( , usually less than 2%, is the average current value, is the maximum current, Current minimum value), harmonic distortion rate (usually less than 5%) used to indicate inverter abnormalities, and power factor (usually greater than 0.9) used to indicate power changes. Vibration parameters include effective vibration value (usually around 2.8 mm / s) used to indicate equipment structural looseness or bearing wear, crest factor (usually less than 5) used to indicate alluvial faults, kurtosis (usually 3-3.5) used to indicate bearing spalling, and rotational frequency amplitude (usually less than baseline value × 1.5) used to indicate balance and alignment. Temperature timing parameters include temperature rise / fall rate and hot spot temperature difference. In addition, operating parameters may also include pressure pulsation (usually less than 5%), flow deviation (usually less than ±3%) used to indicate pipe blockage, and oil particle count used to indicate lubrication wear.

[0026] For example, the exhaust temperature of the air compressor rises from 95°C to 120°C within a single unit time, and the vibration amplitude increases from 2.1 mm / s to 6.3 mm / s. Since the amplitude is greater than 4.5 mm / s and the temperature difference ΔT is greater than 15K, it is determined whether the air compressor has a bearing or lubrication failure. And so on.

[0027] Step S14, performing fault detection based on the operating information and the fault characteristics, triggering solution matching when the fault detection result meets a first condition, causing the first model to perform fault solution matching based on the operating information and the fault characteristics to obtain a first solution; In this embodiment, corresponding processing solutions are quickly matched to different faults, which greatly improves processing efficiency.

[0028] The first model matches a fault solution according to the working information and the fault characteristics to obtain a first solution, including: Step S141, performing fault matching and state similarity calculation based on the electrical parameters, the vibration parameters, the temperature time series, and the fault characteristics; Step S142: determining an initial solution based on the fault matching result and the state similarity calculation result; Step S143 : Screening and adjusting the initial solution according to the preset cost constraint to obtain the first solution.

[0029] In this embodiment, multiple parameters such as electrical parameters and vibration parameters are incorporated into fault matching to improve the matching accuracy of the fault solution.

[0030] The performing fault matching according to the electrical parameters, the vibration parameters, the temperature sequence, and the fault characteristics includes: Step S1411, determining a fault criterion based on the fault characteristics, wherein the fault criterion is used to indicate a probability range of the device fault corresponding to the fault characteristics; Step S1412, calculating the fault probability based on the electrical parameters, the vibration parameters, and the temperature time series; Step S1413 : compare the fault probability with the fault criterion, and if the fault probability meets the second condition, determine the first solution according to the fault probability and state similarity; otherwise, redetermine the fault criterion according to the fault feature.

[0031] In this embodiment, because the electrical parameters and vibration parameters under different faults are different, it is necessary to first determine the possible faults based on different characteristics, and then determine the corresponding parameter range standards. Specifically, each fault characteristic (such as current THD, vibration RMS, temperature slope, etc.) is divided into intervals, and the frequency of faults within each interval is counted. Specifically, a probability threshold map (PTM) is first constructed. This map defines three probability thresholds for each type of fault characteristic (such as bearing wear, winding overheating, harmonic anomalies, etc.): a lower threshold: a value below which the system is considered "healthy"; a median threshold: used to trigger a "warning"; and an upper threshold: a value above which a "fault" is triggered. The corresponding calculation formulas include: Among them, for the j-th type of fault characteristics, the offline statistics of the historical fault database are obtained: Where, is the standard deviation corresponding to the parameter, is the mean failure probability corresponding to the feature in the historical samples, and k is the confidence coefficient (usually k=1.5, and the specific range can be adjusted between 1.2-2.0).

[0032] Assuming that the historical mean failure probability of the "bearing wear" feature is μ = 0.65 and σ = 0.12, then: =0.47, =0.65, =0.83 Then, the three types of heterogeneous data (electrical, vibration, and temperature) are input into the Lightweight Hybrid Net (LHN), and the network outputs an 8-dimensional fault probability vector , and then determine the probability of each dimension. If , then it is considered that the second condition is met, otherwise it triggers the update of the fault standard.

[0033] After the judgment is completed, the comprehensive fault score S is calculated: Where, is the state similarity, which is calculated by the weighted Euclidean distance-Gaussian kernel function, which is not described here. w is the corresponding weight. When the fault score S of the solution meets the preset range, the solution is determined to be the first solution, and so on.

[0034] The calculating of state similarity according to the electrical parameters, the vibration parameters, the temperature time series, and the fault characteristics includes: Step S1414: determining a similarity standard based on the fault feature, wherein the similarity standard is used to indicate a similarity range of the device fault corresponding to the fault feature; Step S1415, calculating state similarity based on the electrical parameters, the vibration parameters, and the temperature time series; Step S1416 , comparing the similarity value with the similarity standard, and determining the first solution based on the fault probability and state similarity when the fault probability meets the third condition; otherwise, re-determine the similarity standard based on the fault characteristics.

[0035] In this embodiment, after the state similarity is calculated by the weighted Euclidean distance-Gaussian kernel function, the obtained state similarity is compared with the preset similarity standard. The comparison method can refer to the calculation method of the aforementioned fault score S, or other comparison methods can be used, which will not be repeated here.

[0036] Step S15: recommend the first solution to a target terminal, and instruct the target terminal to display the first solution.

[0037] In this embodiment, after obtaining the relevant solution, the time sequence content corresponding to the solution is sent to a mobile terminal (such as a mobile phone, etc.) to instruct relevant personnel to perform fault maintenance on relevant equipment.

[0038] Through the above steps, by extracting conversations about faults and other content and combining data analysis with the actual perceptions of staff, the accuracy of maintenance plan recommendations is greatly improved, the problem of low accuracy in maintenance plan recommendations is solved, and the accuracy of maintenance plan recommendations is improved.

[0039] Example 2 The difference from embodiment 1 is that, in an optional embodiment, after determining the initial solution based on the fault matching result and the state similarity calculation result, the method further includes: Step S1421: determining a maintenance risk value based on the work information and the maintenance duration included in the initial plan; Step S1422, determining a recommended value of the initial solution based on the maintenance risk value, the similarity calculation result, and the cost constraint; Step S1423: Determine the initial solution with the largest recommendation value as the first solution.

[0040] In this embodiment, since there will be production losses when the equipment is under maintenance, the losses caused by the duration of the equipment maintenance need to be included in the calculation. Specifically: In the formula, Ri is the loss value of any device, C is the criticality level of the device, δ is the device weight, Mi is the maintenance duration of any plan, maxMi is the maximum maintenance duration in history, and the final recommended value R is calculated as follows: In the formula, λ, μ, ν are the corresponding weights, Q is the cost of the solution, and Qmax is the historical maximum cost (i.e., cost constraint). The solution that best meets the requirements is then determined through the recommendation.

[0041] Through the description of the above embodiments, those skilled in the art will clearly understand that the methods according to the above embodiments can be implemented using software plus the necessary general-purpose hardware platform. Of course, hardware can also be used, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, or optical disk) and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0042] In this embodiment, an intelligent recommendation system based on an artificial intelligence large model is also provided. The system is used to implement the above-mentioned embodiments and preferred embodiments. The details that have been described will not be repeated here. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

[0043] Figure 2This is a structural diagram of an intelligent recommendation system based on an artificial intelligence large model according to an embodiment of the present invention. Figure 2 As shown, the system includes: The conversation collection module 21 is used to obtain the conversation information input by the user, wherein the conversation information includes text information and / or image information describing the fault of the target device; a semantic parsing module 22 for parsing the conversation information using a preset first model and extracting device information and fault characteristics based on the parsing results, wherein the device information includes a device criticality level; A working information acquisition module 23 is configured to obtain working information of a target device according to the device information, wherein the working information includes electrical parameters, vibration parameters, and temperature timing parameters; a matching module 24 configured to perform fault detection based on the operating information and the fault characteristics, and trigger solution matching when a fault detection result meets a first condition, so that the first model performs fault solution matching based on the operating information and the fault characteristics to obtain a first solution; The recommendation module 25 is configured to recommend the first solution to a target terminal and instruct the target terminal to display the first solution.

[0044] In an optional embodiment, the first model matches a fault solution according to the working information and the fault characteristics to obtain a first solution, including: Perform fault matching and state similarity calculation based on the electrical parameters, the vibration parameters, the temperature time series, and the fault characteristics; Determine the initial solution based on the fault matching results and the state similarity calculation results; The initial solution is screened and adjusted according to a preset cost constraint to obtain the first solution.

[0045] In an optional embodiment, the performing fault matching according to the electrical parameter, the vibration parameter, the temperature sequence, and the fault feature includes: Determining a fault criterion based on the fault characteristics, wherein the fault criterion is used to indicate a probability range of a device fault corresponding to the fault characteristics; Calculating a fault probability based on the electrical parameters, the vibration parameters, and the temperature time series; The failure probability is compared with the failure criterion, and when the failure probability meets the second condition, the first solution is determined according to the failure probability and state similarity; otherwise, the failure criterion is re-determined according to the failure characteristics.

[0046] It should be noted that the above modules can be implemented through software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0047] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any one of the above method embodiments when running.

[0048] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0049] An embodiment of the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0050] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0051] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0052] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0053] Units described as separate components may or may not be physically separate, and components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0054] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0055] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application, or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product. The software product is stored in a storage medium and includes several instructions for enabling a device (which can be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0056] The above content is only a specific embodiment of this application, but the scope of protection of this application is not limited to this. Any changes or replacements within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An intelligent recommendation method based on an artificial intelligence large model, characterized in that: include: Acquiring dialogue information input by the user, wherein the dialogue information includes text information and / or image information describing a fault of the target device; Parsing the conversation information using a preset first model, and extracting device information and fault characteristics based on the parsing results, wherein the device information includes a device criticality level; Acquire operating information of the target device according to the device information, wherein the operating information includes electrical parameters, vibration parameters, and temperature timing parameters; performing fault detection according to the operating information and the fault characteristics, triggering solution matching when a fault detection result meets a first condition, and causing the first model to perform fault solution matching according to the operating information and the fault characteristics to obtain a first solution; The first solution is recommended to a target terminal, and the target terminal is instructed to display the first solution.

2. The method according to claim 1, characterized in that The first model matches a fault solution according to the working information and the fault characteristics to obtain a first solution, including: Perform fault matching and state similarity calculation based on the electrical parameters, the vibration parameters, the temperature time series, and the fault characteristics; Determine the initial solution based on the fault matching results and the state similarity calculation results; The initial solution is screened and adjusted according to a preset cost constraint to obtain the first solution.

3. The method according to claim 2, characterized in that The performing fault matching according to the electrical parameters, the vibration parameters, the temperature sequence and the fault characteristics includes: Determining a fault criterion based on the fault characteristics, wherein the fault criterion is used to indicate a probability range of a device fault corresponding to the fault characteristics; Calculating a fault probability based on the electrical parameters, the vibration parameters, and the temperature time series; The failure probability is compared with the failure criterion, and when the failure probability meets the second condition, the first solution is determined according to the failure probability and state similarity; otherwise, the failure criterion is re-determined according to the failure characteristics.

4. The method according to claim 3, characterized in that The calculating of state similarity according to the electrical parameters, the vibration parameters, the temperature time series and the fault characteristics includes: Determining a similarity standard based on the fault characteristics, wherein the similarity standard is used to indicate a similarity range of the device fault corresponding to the fault characteristics; Calculating state similarity based on the electrical parameters, the vibration parameters, and the temperature time series; The similarity value is compared with the similarity standard, and when the fault probability meets the third condition, the first solution is determined according to the fault probability and the state similarity; otherwise, the similarity standard is re-determined according to the fault feature.

5. The method according to claim 2, characterized in that After determining the initial solution according to the fault matching result and the state similarity calculation result, the method further includes: determining a maintenance risk value based on the work information and the maintenance duration included in the initial plan; Determining a recommended value of the initial solution based on the maintenance risk value, the similarity calculation result, and the cost constraint; The initial solution with the largest recommendation value is determined as the first solution.

6. An intelligent recommendation system based on an artificial intelligence large model, characterized in that: include: A conversation collection module is used to obtain conversation information input by the user, wherein the conversation information includes text information and / or image information describing the fault of the target device; a semantic parsing module, configured to parse the conversation information using a preset first model and extract device information and fault characteristics based on the parsing results, wherein the device information includes a device criticality level; A working information acquisition module, configured to obtain working information of a target device according to the device information, wherein the working information includes electrical parameters, vibration parameters, and temperature timing parameters; a matching module, configured to perform fault detection based on the operating information and the fault characteristics, and trigger solution matching when a fault detection result meets a first condition, so that the first model performs fault solution matching based on the operating information and the fault characteristics to obtain a first solution; The recommendation module is configured to recommend the first solution to a target terminal and instruct the target terminal to display the first solution.

7. The system according to claim 6, characterized in that The first model matches a fault solution according to the working information and the fault characteristics to obtain a first solution, including: Perform fault matching and state similarity calculation based on the electrical parameters, the vibration parameters, the temperature time series, and the fault characteristics; Determine the initial solution based on the fault matching results and the state similarity calculation results; The initial solution is screened and adjusted according to a preset cost constraint to obtain the first solution.

8. The system according to claim 7, characterized in that The performing fault matching according to the electrical parameters, the vibration parameters, the temperature sequence and the fault characteristics includes: Determining a fault criterion based on the fault characteristics, wherein the fault criterion is used to indicate a probability range of a device fault corresponding to the fault characteristics; Calculating a fault probability based on the electrical parameters, the vibration parameters, and the temperature time series; The failure probability is compared with the failure criterion, and when the failure probability meets the second condition, the first solution is determined according to the failure probability and state similarity; otherwise, the failure criterion is re-determined according to the failure characteristics.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 5 when executed.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 5.

Images