Industrial equipment multi-source defect fusion scoring system driven by AI large model

Abstract

The invention relates to the technical field of industrial equipment multi-source defect fusion scoring, and particularly discloses an AI large model-driven industrial equipment multi-source defect fusion scoring system, which comprises an equipment operation environment confirmation module, a data source weight adjustment module, a defect grade score generation module, a visual real-time display module and a database, according to the method, the real-time input power and the equipment temperature are extracted, the load and the temperature environment are accurately divided, the basic fusion distribution weight of each data source is calculated by adopting a product fusion method, the weight is dynamically corrected in combination with the measurement deviation degree and the operation and maintenance delay degree of the sensor, and finally the defect degree is calculated in a weighted mode and matched with the grade score. According to the system, the problems of static weight distribution, sensor deviation correction deficiency and the like in a traditional scheme are solved, dynamic fusion and accurate diagnosis of multi-source data under complex working conditions are realized, the defect detection accuracy and the equipment operation and maintenance efficiency are remarkably improved, and the system is suitable for intelligent state monitoring and fault diagnosis of industrial equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-source defect fusion scoring of industrial equipment, and in particular to a multi-source defect fusion scoring system for industrial equipment driven by an AI large model. Background Art

[0002] The stable operation of industrial equipment is the core guarantee for efficient and safe industrial production. Timely and accurate detection of equipment defects is a key component of preventive maintenance. With the development of intelligent industry, the operating environment of equipment is becoming increasingly complex, and factors such as load fluctuations and temperature changes have a significantly greater impact on defect characteristics. Therefore, to meet the demand for real-time and accurate assessment of equipment status in the Industrial Internet era, it is necessary to introduce large AI models as the core driving engine to process and analyze multi-source heterogeneous data and provide equipment visualization.

[0003] The following problems still exist in the existing technology: First, the existing technology uses fixed weights and does not adjust the importance of data sources in combination with dynamic working conditions such as real-time load rate and temperature of the equipment, resulting in low accuracy of defect fusion scoring under complex working conditions.

[0004] Secondly, the sensor’s measurement deviation and operation and maintenance delay were not taken into consideration, resulting in unreliable data collection and an inability to provide effective data support for subsequent defect fusion scoring. Summary of the Invention

[0005] In view of this, in order to solve the problems raised in the above background technology, an AI large model-driven multi-source defect fusion scoring system for industrial equipment is proposed.

[0006] The objectives of the present invention can be achieved through the following technical solutions: The present invention provides an AI large model-driven industrial equipment multi-source defect fusion scoring system, including: an equipment operating environment confirmation module, which extracts the rated power of the target industrial equipment, monitors the operating environment parameters of the target industrial equipment in real time, and confirms the current operating environment of the target industrial equipment based on this.

[0007] The data source weight adjustment module obtains the measurement data and operation and maintenance data of each data source sensor during the operation of the target industrial equipment, and automatically adjusts the weight of each data source based on the correction rules of sensor deviation. The data sources include vibration data source, temperature data source and current data source.

[0008] The defect level score generation module monitors the vibration data, temperature data and current data of the target industrial equipment in real time during operation, and generates a defect level score for the target industrial equipment based on the adjusted weights of each data source.

[0009] The visual real-time display module uses a large AI model to drive the real-time display of the operating environment of the target industrial equipment, the weight of each data source, and the defect level score.

[0010] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention accurately divides the load environment (low load / medium load / high load) and the temperature environment (low temperature / normal temperature / high temperature) by extracting the real-time input power and equipment temperature, and adopts the product fusion method of the basic weights of the load and temperature environment. The basic fusion weight is obtained by normalizing the proportion, highlighting the highly sensitive data source under the dual working condition coupling, effectively avoiding the single working condition weight from masking the key abnormal characteristics, and improving the detection accuracy of complex working conditions.

[0011] (2) The present invention constructs a weight correction factor based on measurement deviation and operation and maintenance delay, and dynamically adjusts the basic fusion weight after coupling analysis of the two, thereby reducing the misjudgment rate caused by sensor abnormalities, improving data reliability, and enhancing data reliability.

[0012] (3) The present invention combines the two-dimensional deviation of vibration amplitude and frequency with vibration anomaly, adopts the average of temperature deviations at adjacent time points for temperature anomaly, and uses the current anomaly based on current fluctuations during the monitoring period. The three are weighted and summed up by the adjusted data source weights to obtain defectivity, and are matched with the preset defectivity interval to generate a grade score, covering multiple types of anomalies such as mechanical vibration, temperature drift, and current fluctuation, realizing the upgrade from "single-point detection" to "multi-feature fusion", and reducing the error in defect level determination. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 This is a schematic diagram of the system module structure connection of the present invention.

[0015] Figure 2 Schematic diagram of the process of obtaining the weight correction factor of the present invention.

[0016] Figure 3 A schematic diagram of the hardware structure of a computer provided in the second embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0018] Example 1

[0019] See also Figure 1 As shown, the present invention provides an AI large model-driven industrial equipment multi-source defect fusion scoring system, including: an equipment operating environment confirmation module, a data source weight adjustment module, a defect level scoring generation module and a visual real-time display module.

[0020] It should be noted that the present invention also includes a database for storing the load rate range corresponding to each load environment, the equipment temperature range corresponding to each temperature environment, the basic allocation weights of each data source corresponding to each load environment and each temperature environment, and the defect degree range corresponding to each defect level score.

[0021] The device operating environment confirmation module and the data source weight adjustment module are connected, and both the device operating environment confirmation module and the data source weight adjustment module are connected to the defect level score generation module. The device operating environment confirmation module, the data source weight adjustment module and the defect level score generation module are all connected to the visual real-time display module, and the device operating environment confirmation module, the data source weight adjustment module and the defect level score generation module are all connected to the database.

[0022] The equipment operating environment confirmation module extracts the rated power of the target industrial equipment, monitors the operating environment parameters of the target industrial equipment in real time, and confirms the current operating environment of the target industrial equipment based on the power.

[0023] It should be noted that the rated power of the target industrial equipment is extracted from the factory settings of the target industrial equipment.

[0024] In a specific embodiment of the present invention, the specific process of confirming the current operating environment of the target industrial equipment is: extracting the real-time input power and the temperature values corresponding to each temperature distribution area from the operating environment parameters of the target industrial equipment.

[0025] It should be noted that the real-time input power is collected by a power sensor installed at the input end of the target industrial equipment, and the temperature values corresponding to each temperature distribution area are obtained by collecting a thermal image of the target industrial equipment using a thermal imager and extracting it from the thermal image.

[0026] The ratio between the real-time input power of the target industrial equipment and the rated power is taken as the load rate of the target industrial equipment.

[0027] The temperature values corresponding to each temperature distribution area are averaged to obtain the equipment temperature of the target industrial equipment.

[0028] The load rate of the target industrial equipment is compared with the load rate intervals corresponding to each load environment stored in the database. If the load rate of the target industrial equipment is within the load rate interval corresponding to a certain load environment, the load environment is used as the load environment of the target industrial equipment.

[0029] The device temperature of the target industrial device is compared with the device temperature ranges corresponding to each temperature environment stored in the database. If the device temperature of the target industrial device is within the device temperature range corresponding to a certain temperature environment, the temperature environment is used as the temperature environment of the target industrial device.

[0030] In a specific embodiment of the present invention, the load environment includes but is not limited to low load, medium load and high load, the temperature environment includes but is not limited to low temperature, normal temperature and high temperature, and the load rate range corresponding to each load environment and the equipment temperature range corresponding to each temperature environment are extracted from the safety operation specification manual of the target industrial equipment.

[0031] The data source weight adjustment module obtains the measurement data and operation and maintenance data of each data source sensor during the operation of the target industrial equipment, and automatically adjusts the weight of each data source based on the correction rules of sensor deviation, wherein the data sources include vibration data source, temperature data source and current data source.

[0032] In a specific embodiment of the present invention, the data source sensor includes but is not limited to a vibration sensor, a temperature sensor, and a current sensor.

[0033] In a specific embodiment of the present invention, the specific process of automatically adjusting the weights of each data source based on the correction rule of sensor deviation is: comparing the load environment of the target industrial equipment with the basic allocation weights of each data source corresponding to each load environment stored in the database, and obtaining the basic allocation weights of each data source corresponding to the load environment of the target industrial equipment.

[0034] The temperature environment of the target industrial equipment is compared with the basic allocation weights of each data source corresponding to each temperature environment stored in the database to obtain the basic allocation weights of each data source corresponding to the temperature environment of the target industrial equipment.

[0035] It should be noted that the basic weightings assigned to each data source for each temperature and load environment are designed to accommodate the characteristics of industrial equipment under varying operating conditions, highlighting the value of key data sources for defect detection. Because load and temperature changes can cause different responses in equipment components (for example, current and vibration data under high loads can better reflect motor status, while constant temperature data in high-temperature environments is particularly important for defect diagnosis), differentiated weighting allows the system to more accurately capture equipment anomalies when integrating multi-source data, improving the accuracy and specificity of defect detection. This ensures effective identification of equipment defects in complex operating conditions and prevents critical data from being overlooked or misjudged due to environmental differences.

[0036] The basic allocation weights of the data sources corresponding to the load environment and temperature environment of the target industrial equipment are fused to obtain the basic fusion allocation weights of the data sources of the target industrial equipment.

[0037] In a specific embodiment of the present invention, the specific process of obtaining the basic fusion allocation weights of each data source of the target industrial equipment is: multiplying the basic allocation weights of each data source corresponding to the load environment of the target industrial equipment and the basic allocation weights of each data source corresponding to the temperature environment to obtain the fusion allocation weights of each data source.

[0038] The ratio of the fusion allocation weight of each data source to the fusion allocation weight of all data sources is used as the basic fusion allocation weight of each data source of the target industrial equipment.

[0039] Based on the measurement data and operation and maintenance data of each data source sensor, deviation correction analysis is performed to obtain the weight correction factor corresponding to each data source.

[0040] See also Figure 2 As shown, in a specific embodiment of the present invention, the specific process of obtaining the weight correction factor corresponding to each data source is: based on the measured value and calibration value of each measurement in the measurement data of each data source sensor, the measurement deviation of each data source sensor is calculated to obtain the measurement deviation degree.

[0041] In a specific embodiment of the present invention, the specific process of obtaining the measurement deviation of each data source sensor is: subtracting the calibration value of each measurement of each data source sensor from the actual value to obtain the deviation value of each measurement of each data source sensor.

[0042] The deviation values of each measurement of each data source sensor are averaged to obtain the average deviation value of each data source sensor.

[0043] The difference between the average deviation value of each data source sensor and the deviation value of the set reference is compared with the deviation value of the set reference to obtain the measurement deviation degree of each data source sensor.

[0044] Calibration calculation is performed based on the time points corresponding to each equipment operation and maintenance in the operation and maintenance data of each data source sensor to obtain the operation and maintenance delay of each data source sensor.

[0045] In a specific embodiment of the present invention, the specific process of obtaining the operation and maintenance delay of each data source sensor is: the interval duration between the time points corresponding to the operation and maintenance of adjacent devices of each data source sensor is used as the operation and maintenance interval duration corresponding to each device operation and maintenance of each data source sensor.

[0046] The operation and maintenance interval duration corresponding to each device operation and maintenance of each data source sensor is compared with the set operation and maintenance interval threshold. If the operation and maintenance interval duration corresponding to a device operation and maintenance of a data source sensor is longer than the set operation and maintenance interval threshold, the device operation and maintenance is recorded as delayed operation and maintenance, and the number of delayed operations and maintenance of each data source sensor is counted.

[0047] The difference between the delayed operation and maintenance times of each data source sensor and the set reference delayed operation and maintenance times is compared with the set reference delayed operation and maintenance times to obtain the operation and maintenance delay degree of each data source sensor.

[0048] Based on the measurement deviation and operation and maintenance delay of each data source sensor, a coupling analysis is performed to obtain the weight correction factor corresponding to each data source.

[0049] It should be noted that the specific formula for obtaining the weight correction factor corresponding to each data source is: Among them, β i Indicates the weight correction factor corresponding to the i-th data source, χ i and δ i They represent the measurement deviation and operation delay corresponding to the i-th data source, respectively. a1 and a2 represent the weights of the weight correction factors corresponding to the measurement deviation and operation delay, respectively. a1+a2=1, e represents a natural constant, and i represents the number of the data source, i=1,2,...,n, and n=3.

[0050] It is also necessary to further explain that the logic of writing the above weight correction factor is: the measurement deviation χ i and operation and maintenance delay δ i As influencing factors, the influence ratio of the two on the weight correction factor is adjusted by the weight coefficients a1 and a2. The exponential term is constructed using the natural constant e Let the weight correction factor β i It is mapped to the range of 0-1 to ensure that its value is reasonable. The greater the measurement deviation or operation delay, the smaller the exponential term, and β i The smaller it is, the lower the weight of the corresponding data source is, and the quantitative correction of the reliability of the data source is achieved, so that the weight adjustment is more in line with the quality status of the data source under actual working conditions.

[0051] In a specific embodiment of the present invention, when evaluating the weight correction factor corresponding to each data source, the proportion of measurement deviation and operation and maintenance delay is equally important. Therefore, the setting value of a1 is 0.5 and the setting value of a2 is 0.5.

[0052] The embodiment of the present invention constructs a weight correction factor based on measurement deviation and operation and maintenance delay, and dynamically adjusts the basic fusion weight after coupling analysis of the two, thereby reducing the misjudgment rate caused by sensor abnormalities, improving data reliability, and enhancing data reliability.

[0053] The product of the basic fusion allocation weight of each data source of the target industrial equipment and the corresponding weight correction factor is obtained, and the product result is added to the basic fusion allocation weight of each data source to obtain the adjusted weight of each data source.

[0054] The embodiment of the present invention extracts real-time input power and equipment temperature to accurately divide the load environment (low load / medium load / high load) and temperature environment (low temperature / normal temperature / high temperature), and adopts the product fusion method of the basic weights of the load and temperature environment. The basic fusion weight is obtained by normalizing the proportion, highlighting the highly sensitive data source under the dual working condition coupling, effectively avoiding the single working condition weight from masking key abnormal characteristics, and improving the detection accuracy of complex working conditions.

[0055] The defect level score generation module monitors the vibration data, temperature data and current data of the target industrial equipment in real time during operation, and generates a defect level score for the target industrial equipment based on the adjusted weights of each data source.

[0056] In a specific embodiment of the present invention, the specific process of generating the defect grade score of the target industrial equipment is: extracting the vibration amplitude and vibration frequency from the vibration data of the target industrial equipment during operation, extracting the equipment temperature corresponding to each monitoring time point from the temperature data, and extracting the current corresponding to each monitoring time period from the current data, and accordingly obtaining the vibration abnormality, temperature abnormality and current abnormality of the target industrial equipment.

[0057] It should be noted that the vibration amplitude and vibration frequency are acquired through a vibration sensor, the device temperature corresponding to each monitoring time point is acquired through a temperature sensor, and the current corresponding to each monitoring time period is acquired through a current sensor.

[0058] In a specific embodiment of the present invention, the specific process of obtaining the vibration abnormality of the target industrial equipment is: subtracting the vibration amplitude and vibration frequency of the target industrial equipment during operation from the set reference vibration amplitude and vibration frequency, and respectively comparing the difference between the two with the set reference vibration amplitude and vibration frequency, and taking the sum of the two ratio results as the vibration abnormality of the target industrial equipment.

[0059] In a specific embodiment of the present invention, the specific process of obtaining the temperature anomaly of the target industrial equipment is: obtaining the equipment temperature deviation corresponding to adjacent monitoring time points during the operation of the target industrial equipment, taking the average of the equipment temperature deviations corresponding to all adjacent monitoring time points to obtain the average equipment temperature deviation of the target industrial equipment during the operation process, and comparing the difference between the average equipment temperature deviation of the target industrial equipment during the operation process and the set reference value with the set reference value to obtain the temperature anomaly of the target industrial equipment.

[0060] It should be noted that the specific method of obtaining the current abnormality of the target industrial equipment is: comparing the current corresponding to each monitoring time period with the current interval under normal operating conditions. If the current corresponding to a monitoring time period is not within the current interval under normal operating conditions, then the monitoring time period is recorded as an abnormal monitoring time period, and the number of abnormal monitoring time periods of the target industrial equipment is counted.

[0061] The ratio of the difference between the number of abnormality monitoring time periods of the target industrial equipment and the number of abnormality monitoring time periods set as the reference number to the number of abnormality monitoring time periods set as the current abnormality degree of the target industrial equipment.

[0062] The vibration abnormality, temperature abnormality and current abnormality of the target industrial equipment are weightedly calculated and summed with the corresponding adjusted data source weights to obtain the defect degree of the target industrial equipment.

[0063] The defect degree of the target industrial equipment is matched with the defect degree intervals corresponding to each defect grade score stored in the database. If the defect degree of the target industrial equipment is within the defect degree interval corresponding to a certain defect grade score, the defect grade score is used as the defect grade score of the target industrial equipment.

[0064] In the embodiment of the present invention, vibration anomaly is combined with the two-dimensional deviation of vibration amplitude and frequency. Temperature anomaly adopts the average of temperature deviations at adjacent time points. Current anomaly is based on current fluctuations during the monitoring period. The three are weighted and summed through the adjusted data source weights to obtain defectivity, and then matched with the preset defectivity interval to generate a grade score, covering multiple types of anomalies such as mechanical vibration, temperature drift, and current fluctuation, realizing the upgrade from "single-point detection" to "multi-feature fusion", and reducing the error in defect level determination.

[0065] The visual real-time display module drives the real-time display of the operating environment of the target industrial equipment, the weight of each data source and the defect level score through the AI big model.

[0066] Example 2

[0067] like Figure 3As shown, an embodiment of the present invention provides the following technical solution: a computer, comprising a memory 202, a processor 201, and a computer program stored in the memory 202 and executable on the processor 201; when the processor 201 executes the computer program, the multi-source defect fusion scoring system for industrial equipment driven by the AI large model as described above is implemented.

[0068] Specifically, the processor 201 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0069] Among them, the memory 202 may include a large-capacity memory for data or instructions. By way of example and not limitation, the memory 202 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 202 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 202 may be inside or outside the data processing device. In a specific embodiment, the memory 202 is a non-volatile memory. In a specific embodiment, the memory 202 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0070] The memory 202 may be used to store or cache various data files that need to be processed and / or used for communication, as well as possible computer program instructions executed by the processor 201 .

[0071] The processor 201 reads and executes computer program instructions stored in the memory 202 to implement the above-mentioned AI large model-driven industrial equipment multi-source defect fusion scoring system.

[0072] In some embodiments, the computer may further include a communication interface 203 and a bus 200. Figure 3 As shown, the processor 201 , the memory 202 , and the communication interface 203 are connected via a bus 200 and communicate with each other.

[0073] The communication interface 203 is used to implement communication between the various modules, devices, units, and / or devices in the embodiments of the present application. The communication interface 203 can also implement data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.

[0074] The bus 200 includes hardware, software, or both, and couples computer components together. The bus 200 includes, but is not limited to, at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example and not limitation, bus 200 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Bus 200 may include one or more buses, where appropriate. Although embodiments herein describe and illustrate a particular bus, this application contemplates any suitable bus or interconnect.

[0075] 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 described specific embodiments 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 should all fall within the scope of protection of the present invention.

Claims

1. The AI large model-driven industrial equipment multi-source defect fusion scoring system is characterized by: include: The equipment operating environment confirmation module extracts the rated power of the target industrial equipment, monitors the operating environment parameters of the target industrial equipment in real time, and confirms the current operating environment of the target industrial equipment based on this; The data source weight adjustment module obtains the measurement data and operation and maintenance data of each data source sensor during the operation of the target industrial equipment, and automatically adjusts the weight of each data source based on the correction rules of sensor deviation. The data sources include vibration data source, temperature data source and current data source; The defect grade score generation module monitors the vibration data, temperature data, and current data of the target industrial equipment in real time during operation, and generates a defect grade score for the target industrial equipment based on the adjusted weights of each data source; The visual real-time display module uses a large AI model to drive the real-time display of the operating environment of the target industrial equipment, the weight of each data source, and the defect level score.

2. The AI large model-driven industrial equipment multi-source defect fusion scoring system according to claim 1 is characterized by: The specific process of confirming the current operating environment of the target industrial equipment is as follows: Extracting real-time input power and temperature values corresponding to each temperature distribution area from the operating environment parameters of the target industrial equipment; The ratio between the real-time input power and the rated power of the target industrial equipment is used as the load rate of the target industrial equipment; Calculate the average of the temperature values corresponding to each temperature distribution area to obtain the equipment temperature of the target industrial equipment; Compare the load rate of the target industrial equipment with the load rate intervals corresponding to each load environment stored in the database. If the load rate of the target industrial equipment is within the load rate interval corresponding to a certain load environment, then use the load environment as the load environment of the target industrial equipment. The device temperature of the target industrial device is compared with the device temperature ranges corresponding to each temperature environment stored in the database. If the device temperature of the target industrial device is within the device temperature range corresponding to a certain temperature environment, the temperature environment is used as the temperature environment of the target industrial device.

3. The AI large model-driven industrial equipment multi-source defect fusion scoring system according to claim 2 is characterized by: The specific process of automatically adjusting the weight of each data source based on the correction rule of sensor deviation is as follows: Comparing the load environment of the target industrial equipment with the basic allocation weights of each data source corresponding to each load environment stored in the database to obtain the basic allocation weights of each data source corresponding to the load environment of the target industrial equipment; Comparing the temperature environment of the target industrial equipment with the basic allocation weights of each data source corresponding to each temperature environment stored in the database to obtain the basic allocation weights of each data source corresponding to the temperature environment of the target industrial equipment; The basic allocation weights of the data sources corresponding to the load environment and temperature environment of the target industrial equipment are fused to obtain the basic fusion allocation weights of the data sources of the target industrial equipment; Based on the measurement data and operation and maintenance data of each data source sensor, deviation correction analysis is performed to obtain the weight correction factor corresponding to each data source; The product of the basic fusion allocation weight of each data source of the target industrial equipment and the corresponding weight correction factor is obtained, and the product result is added to the basic fusion allocation weight of each data source to obtain the adjusted weight of each data source.

4. The AI large model-driven industrial equipment multi-source defect fusion scoring system according to claim 3 is characterized by: The specific process of obtaining the basic fusion allocation weights of each data source of the target industrial equipment is as follows: Multiply the basic allocation weight of each data source corresponding to the load environment of the target industrial equipment and the basic allocation weight of each data source corresponding to the temperature environment to obtain the fusion allocation weight of each data source; The ratio of the fusion allocation weight of each data source to the fusion allocation weight of all data sources is used as the basic fusion allocation weight of each data source of the target industrial equipment.

5. The AI large model-driven industrial equipment multi-source defect fusion scoring system according to claim 3 is characterized by: The specific process of obtaining the weight correction factor corresponding to each data source is as follows: Calculate the measurement deviation based on the actual value and calibration value of each measurement in the measurement data of each data source sensor to obtain the measurement deviation degree of each data source sensor; Calibration calculation is performed based on the time points corresponding to each device operation and maintenance in the operation and maintenance data of each data source sensor to obtain the operation and maintenance delay of each data source sensor; Based on the measurement deviation and operation and maintenance delay of each data source sensor, a coupling analysis is performed to obtain the weight correction factor corresponding to each data source.

6. The AI large model-driven industrial equipment multi-source defect fusion scoring system according to claim 5 is characterized by: The specific process of obtaining the measurement deviation of each data source sensor is as follows: Subtract the calibration value of each measurement of each data source sensor from the actual measured value to obtain the deviation value of each measurement of each data source sensor; Calculate the average of the deviation values of each measurement of each data source sensor to obtain the average deviation value of each data source sensor; The difference between the average deviation value of each data source sensor and the deviation value of the set reference is compared with the deviation value of the set reference to obtain the measurement deviation degree of each data source sensor.

7. The AI large model-driven industrial equipment multi-source defect fusion scoring system according to claim 5 is characterized by: The specific process of obtaining the operation and maintenance delay of each data source sensor is as follows: The interval between the time points corresponding to the operation and maintenance of adjacent devices of each data source sensor is used as the operation and maintenance interval corresponding to each device operation and maintenance of each data source sensor; Compare the operation and maintenance interval duration corresponding to each device operation and maintenance of each data source sensor with the set operation and maintenance interval threshold. If the operation and maintenance interval duration corresponding to a device operation and maintenance of a data source sensor is longer than the set operation and maintenance interval threshold, then record the device operation and maintenance as delayed operation and maintenance, and count the number of delayed operations and maintenance of each data source sensor; The difference between the delayed operation and maintenance times of each data source sensor and the set reference delayed operation and maintenance times is compared with the set reference delayed operation and maintenance times to obtain the operation and maintenance delay degree of each data source sensor.

8. The AI large model-driven industrial equipment multi-source defect fusion scoring system according to claim 1 is characterized by: The specific process of generating the defect grade score of the target industrial equipment is as follows: Extract the vibration amplitude and frequency from the vibration data of the target industrial equipment during operation, extract the equipment temperature corresponding to each monitoring time point from the temperature data, and extract the current corresponding to each monitoring time period from the current data. Based on this, the vibration anomaly, temperature anomaly, and current anomaly of the target industrial equipment are obtained. The vibration abnormality, temperature abnormality, and current abnormality of the target industrial equipment are weighted and summed with the corresponding adjusted data source weights to obtain the defect degree of the target industrial equipment; The defect degree of the target industrial equipment is matched with the defect degree intervals corresponding to each defect grade score stored in the database. If the defect degree of the target industrial equipment is within the defect degree interval corresponding to a certain defect grade score, the defect grade score is used as the defect grade score of the target industrial equipment.

9. The AI large model-driven industrial equipment multi-source defect fusion scoring system according to claim 8 is characterized by: The specific process of obtaining the vibration abnormality of the target industrial equipment is as follows: the vibration amplitude and vibration frequency of the target industrial equipment during operation are respectively subtracted from the vibration amplitude and vibration frequency of the set reference, and the difference between the two is respectively compared with the vibration amplitude and vibration frequency of the set reference, and the sum of the two ratio results is used as the vibration abnormality of the target industrial equipment.

10. The AI large model driven industrial equipment multi-source defect fusion scoring system according to claim 8 is characterized by: The specific process of obtaining the temperature anomaly of the target industrial equipment is as follows: obtaining the equipment temperature deviation corresponding to adjacent monitoring time points during the operation of the target industrial equipment, taking the average of the equipment temperature deviations corresponding to all adjacent monitoring time points to obtain the average equipment temperature deviation of the target industrial equipment during the operation, and comparing the difference between the average equipment temperature deviation of the target industrial equipment during the operation and the set reference value with the set reference value to obtain the temperature anomaly of the target industrial equipment.

Images