A system and method for intelligent status monitoring and life prediction of industrial equipment

Abstract

The present invention discloses a system and method for intelligent status monitoring and life prediction of industrial equipment, including a multi-physics field sensor module, a data processing module, a model building module, an analysis and prediction platform, and an output and alarm module. The infrared sensor submodule, the ultrasonic sensor submodule, and the electromagnetic sensor submodule are used to detect surface temperature changes, internal damage, and material property changes of parts and components, respectively. The data processing module receives sensor data and performs data standardization through ultrasonic and infrared processing submodules. The model building module builds and updates the wear model in real time based on the monitoring data, and adjusts the model parameters in combination with the wear status. The analysis and prediction platform fuses historical and real-time data, calculates the current wear rate, predicts the remaining service life, and generates a wear report. The present invention improves the accuracy and stability of wear monitoring through multi-physics field data fusion and intelligent analysis, and is suitable for industrial equipment in extreme temperature and high dust environments.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical monitoring, and in particular to a system and method for intelligent state monitoring and life prediction of industrial equipment. Background Art

[0002] Intelligent monitoring and lifespan prediction technologies for industrial equipment are gradually developing to improve equipment reliability and optimize maintenance costs. Traditional condition monitoring methods rely primarily on single sensor data, such as vibration, temperature, and current, combined with statistical analysis or physical modeling to assess equipment health. However, in complex operating conditions, a single data source cannot fully reflect the operating status of the equipment.

[0003] While existing industrial equipment condition monitoring and life prediction technologies have made some progress, key challenges remain in extreme environments (such as those characterized by high temperatures, high dust levels, and strong vibrations). First, data acquisition stability is limited. Traditional sensors are easily affected by environmental factors like heat and dust, resulting in data distortion and inaccurate monitoring data, thus reducing the reliability of the monitoring system. For example, vibration sensors can experience errors due to thermal drift in high-temperature environments. Infrared sensors, affected by extreme temperatures, can make it difficult to discern temperature differences caused by actual component wear. Dust accumulation can also affect the accuracy of optical sensors. Second, data analysis accuracy is limited. Current life prediction technologies mostly rely on single-sensor data or historical failure data, ignoring the impact of environmental variables on the equipment aging process. This results in insufficient generalization of prediction models. Furthermore, existing intelligent monitoring systems lack adaptability. Due to the complex and ever-changing operating environments of equipment, fixed-parameter models are difficult to adjust to dynamic conditions, leading to significant deviations in prediction results and compromising the accuracy of maintenance decisions. More importantly, existing maintenance strategies often rely on preset thresholds rather than the actual health status of the equipment, which can easily lead to problems of "over-maintenance" or "under-maintenance", increasing operation and maintenance costs and potentially causing sudden failures.

[0004] Based on this, the present invention proposes a system and method for intelligent status monitoring and life prediction of industrial equipment. It utilizes environmental adaptive data acquisition, multi-source data fusion, deep learning life prediction and intelligent maintenance decision optimization technology to build an efficient, accurate and stable equipment health management system to solve the problems of poor data stability, low monitoring accuracy and inaccurate maintenance strategies in existing technologies under extreme environments. Summary of the Invention

[0005] The purpose of the present invention is to provide a system for intelligent status monitoring and life prediction of industrial equipment, which has the advantage of effectively improving the accuracy of wear detection.

[0006] The above technical objectives of the present invention are achieved through the following technical solutions:

[0007] A system for intelligent condition monitoring and life prediction of industrial equipment, comprising:

[0008] A multi-physics sensor module, comprising an infrared sensor submodule, an ultrasonic sensor submodule, and an electromagnetic sensor submodule. The infrared sensor submodule is used to detect temperature changes on the surface of a component; the ultrasonic sensor submodule is used to detect damage inside a component; and the electromagnetic sensor submodule is used to detect performance changes of a component.

[0009] A data processing module, the data processing module is used to receive and process real-time data from the multi-physics field sensor module to obtain monitoring data;

[0010] Model building module, used to build the wear model of components and update the wear model in real time based on monitoring data;

[0011] An analysis and prediction platform, used to predict the remaining service life of components based on wear models and generate wear reports;

[0012] Output and alarm module for sending wear reports to the user and issuing alarms when the remaining service life reaches a predefined threshold.

[0013] Further configuration: the ultrasonic sensing submodule specifically includes an ultrasonic sensor and a correction unit, the ultrasonic sensor is arranged at a position facing the target component, the correction unit is arranged on the propagation path of the ultrasonic sensor, the ultrasonic sensor is used to transmit ultrasonic waves and receive ultrasonic signals reflected back from the component to obtain ultrasonic detection data; the correction unit is used to generate an electrical signal based on the received ultrasonic wave, and generate ultrasonic correction data based on the electrical signal.

[0014] Further configuration: the correction unit is made of piezoelectric material.

[0015] Further configuration: the data processing module includes an ultrasonic processing submodule, and the ultrasonic processing submodule analyzes and calculates the ultrasonic detection data and the ultrasonic correction data to obtain ultrasonic standard data, and the specific steps are:

[0016] The original flight time and original echo amplitude are obtained by extracting the ultrasonic detection data, and the time correction parameter and amplitude correction parameter are obtained by extracting the ultrasonic correction data;

[0017] Based on the time correction parameter, calculating and compensating for the delay deviation of the propagation speed of the ultrasonic wave in the current environment, thereby correcting the original flight time to obtain the standardized time;

[0018] Based on the amplitude correction parameter, calculating and compensating for the energy attenuation or scattering loss of the ultrasonic wave in the current environment, thereby correcting the original echo amplitude to obtain a standardized amplitude;

[0019] The ultrasound standard data is obtained according to the normalized amplitude and the normalized magnitude.

[0020] Further configuration: the infrared sensor submodule detects the temperature change of the component surface to obtain infrared image data, the data processing module includes an infrared processing submodule, and the infrared processing submodule analyzes and calculates the infrared detection data to obtain infrared standard data. The specific steps are:

[0021] Preprocessing the infrared image data to obtain an infrared analysis image for enhancing the quality of the infrared image;

[0022] Divide the infrared analysis image into multiple grid areas and calculate the average temperature and temperature fluctuation coefficient of each grid area;

[0023] According to the preset temperature fluctuation threshold, the grid area with a temperature fluctuation coefficient less than the temperature fluctuation threshold is selected as the stable area;

[0024] Eliminate noise regions in stable regions according to a morphological analysis method, wherein the noise regions represent grid regions that are too small or irregular in shape;

[0025] Calculate the average temperature of the stable area as the reference temperature;

[0026] The measured temperature of the edge wear area is calculated according to the reference temperature, and the measured temperature is corrected based on the heat conduction equation to obtain infrared standard data. The edge wear area is characterized as the area where the working surface of the component is located.

[0027] Further configuration: the model building module is used to build a wear model of components and update the wear model in real time according to monitoring data, specifically including the following steps:

[0028] Establish a wear model based on the preset wear calculation framework and material parameters of components;

[0029] Determine the wear state of the component based on the monitoring data, including normal wear state, accelerated wear state and critical failure state;

[0030] The wear model is updated and corrected according to the monitoring data and wear status to obtain a new wear model.

[0031] Further configuration: The analysis and prediction platform is used to predict the remaining service life of parts based on the wear model and generate a wear report specifically including:

[0032] According to the wear model, the accumulated wear amount, current wear rate and wear status are obtained;

[0033] The remaining service life is calculated based on the accumulated wear, current wear rate and wear status;

[0034] Generate maintenance plans and wear reports based on remaining service life.

[0035] Another object of the present invention is to provide a method for intelligent status monitoring and life prediction of industrial equipment, which is applied to the intelligent status monitoring and life prediction system for industrial equipment described above.

[0036] The above technical objectives of the present invention are achieved through the following technical solutions:

[0037] A method for intelligent state monitoring and life prediction of industrial equipment, comprising the following steps:

[0038] The multi-physics field sensor module is used to obtain the surface temperature changes, internal damage conditions, and material performance changes of components to form multi-physics field monitoring data.

[0039] The monitoring data is received and ultrasonic signal correction and infrared temperature data calibration are performed respectively, wherein, for the ultrasonic detection data, the original flight time and echo amplitude are extracted, and the time delay and amplitude attenuation are corrected based on the ultrasonic correction data of the correction unit to obtain ultrasonic standard data; for the infrared detection data, the temperature fluctuation coefficients of multiple grid areas are calculated, the stable areas are screened out, and the temperature data of the edge wear area is corrected based on the average temperature of the stable areas and the heat conduction equation to obtain infrared standard data;

[0040] Build a wear model for components and update it in real time based on monitoring data;

[0041] Predict the remaining service life of components based on wear models and generate wear reports;

[0042] Send wear reports to users and issue alerts when remaining service life reaches a pre-defined threshold.

[0043] In summary, the present invention has the following beneficial effects:

[0044] 1. Compared with a single sensing method, the multi-physics field sensor module that combines infrared, ultrasonic and electromagnetic sensors can effectively reduce the impact of factors such as ambient temperature fluctuations and dust interference on the measurement results, improve the reliability of monitoring data, and enable the system to operate stably in high and low temperature and dust-intensive environments.

[0045] 2. The infrared processing submodule and ultrasonic processing submodule of the data processing module correct and standardize temperature measurements and ultrasonic signals, respectively. The infrared processing submodule uses temperature gridding analysis, morphological analysis, and heat conduction model correction to ensure that temperature data is free of ambient temperature influences, improving the accuracy of temperature measurements in worn areas of components. The ultrasonic processing submodule uses time correction and amplitude compensation to eliminate the effects of dust on ultrasonic propagation characteristics, ensuring more accurate internal damage detection.

[0046] 3. Utilizing a model-building module, the wear model is updated in real time based on monitoring data, dynamically adjusting wear parameters to improve the accuracy of component status characterization. The system can identify three states: normal wear, accelerated wear, and critical failure, and adjusts the wear calculation method accordingly, ensuring the model's adaptability to complex wear conditions.

[0047] 4. The calculation logic of the remaining service life can be adjusted according to different wear stages to ensure that the prediction results are more in line with actual working conditions, thereby providing a scientific basis for equipment maintenance and avoiding losses caused by premature replacement or delayed maintenance.

[0048] 5. Through the output and alarm module, the system automatically sends an alert when the remaining service life reaches a preset threshold, reminding users to take proactive maintenance measures and reduce the risk of equipment downtime due to sudden failures. Combined with wear reports, users can optimize maintenance plans based on system recommendations, improve equipment management efficiency, and extend component life. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is an overall structural block diagram of the embodiment. DETAILED DESCRIPTION

[0050] The present invention will be further described in detail below with reference to the accompanying drawings.

[0051] Example:

[0052] like Figure 1 As shown, a system for intelligent condition monitoring and life prediction of industrial equipment includes:

[0053] A multi-physics sensor module, comprising an infrared sensor submodule, an ultrasonic sensor submodule, and an electromagnetic sensor submodule. The infrared sensor submodule is used to detect temperature changes on the surface of a component; the ultrasonic sensor submodule is used to detect damage inside a component; and the electromagnetic sensor submodule is used to detect performance changes of a component.

[0054] A data processing module, the data processing module is used to receive and process real-time data from the multi-physics field sensor module to obtain monitoring data;

[0055] Model building module, used to build the wear model of components and update the wear model in real time based on monitoring data;

[0056] An analysis and prediction platform, used to predict the remaining service life of components based on wear models and generate wear reports;

[0057] Output and alarm module for sending wear reports to users and issuing alarms when the remaining service life reaches a predetermined threshold.

[0058] Among them, the ultrasonic sensing submodule specifically includes an ultrasonic sensor and a correction unit. The ultrasonic sensor is arranged at a position facing the target component, and the correction unit is arranged on the propagation path of the ultrasonic sensor. The ultrasonic sensor is used to transmit ultrasonic waves and receive ultrasonic signals reflected from the component to obtain ultrasonic detection data; the correction unit is used to generate an electrical signal based on the received ultrasonic wave and generate ultrasonic correction data based on the electrical signal.

[0059] The correction unit is made of piezoelectric material, which has a piezoelectric effect. That is, piezoelectric material generates electric charge when subjected to pressure or tension. Conversely, when an electric field is applied, the piezoelectric material deforms and produces mechanical vibration. When ultrasound propagates to the piezoelectric material in a dusty environment, the energy it carries causes the piezoelectric material to generate an electrical signal. By detecting changes in these electrical signals, the propagation characteristics of ultrasound can be obtained, similar to how magneto-optical materials indirectly obtain ultrasound information through the magneto-optical effect. Therefore, piezoelectric materials are used to capture changes in ultrasonic signals and compensate for the effects of factors such as dust in the environment on ultrasonic propagation, thereby improving detection accuracy.

[0060] The data processing module includes an ultrasonic processing submodule, an infrared processing submodule and an electromagnetic processing submodule. The ultrasonic processing submodule analyzes and calculates the ultrasonic detection data and the ultrasonic correction data to obtain the ultrasonic standard data. The specific steps are as follows:

[0061] The original flight time and original echo amplitude are obtained by extracting the ultrasonic detection data, and the time correction parameter and amplitude correction parameter are obtained by extracting the ultrasonic correction data;

[0062] Based on the time correction parameter, the delay deviation of the propagation speed of the ultrasonic wave in the current environment is calculated and compensated, thereby correcting the original flight time to obtain the standardized time. In an environment with high dust or low / high temperature, the actual propagation speed of the ultrasonic wave deviates from the reference speed, resulting in the echo flight time not being consistent with the ideal value. Since the pre-set correction unit knows its setting position and the distance between it and the ultrasonic sensor, the obtained time correction parameter can reflect this deviation. This delay deviation in propagation speed can be calculated based on the pre-known distance, the time when the ultrasonic sensor emits the ultrasonic wave, and the time when the correction unit generates an electrical signal to obtain the actual detected ultrasonic wave propagation speed and the theoretical ultrasonic wave propagation speed. Based on this wave speed ratio, the time correction parameter can be obtained. By correcting the original flight time according to the time correction parameter, the true time from the ultrasonic sensor emitting the ultrasonic wave to the echo being received when not affected by the dust environment can be obtained, that is, the standardized time.

[0063] Based on the amplitude correction parameters, the energy attenuation or scattering loss of the ultrasonic wave in the current environment is calculated and compensated, thereby correcting the original echo amplitude to obtain a standardized amplitude. Dust or temperature changes not only affect the propagation speed of the ultrasonic wave but also cause increased energy attenuation or waveform distortion. Therefore, the actual intensity of the ultrasonic wave at the correction unit's location is measured by performing peak amplitude detection, RMS (root mean square) calculation, or integrated energy analysis on the electrical signal generated by the correction unit when receiving the ultrasonic wave. This intensity indicator (such as peak voltage, RMS value, etc.) is mapped into an amplitude correction parameter for attenuation correction. When correcting the original echo amplitude, a multiplicative method is used to correct the original echo amplitude.

[0064] The ultrasound standard data is obtained according to the normalized amplitude and the normalized magnitude.

[0065] The infrared sensor submodule detects the temperature change on the surface of the component to obtain infrared image data. The data processing module includes an infrared processing submodule. The infrared processing submodule analyzes and calculates the infrared detection data to obtain infrared standard data. The specific steps are as follows:

[0066] 1. Preprocess the infrared image data to obtain an infrared analysis image to enhance the quality of the infrared image;

[0067] Preprocessing specifically includes:

[0068] Denoising: Adaptive filtering algorithms (such as bilateral filtering or wavelet transform denoising) are used to remove noise from the image to ensure that the temperature data is not affected by the environment.

[0069] Contrast enhancement: Use histogram equalization (HE) or contrast-limited adaptive histogram equalization (CLAHE) to enhance image contrast, making the temperature distribution more clearly discernible.

[0070] Edge detection: Apply the Canny operator or Sobel operator to enhance the edge of the image to make the structural features of the parts more obvious

[0071] The infrared analysis image is divided into multiple grid areas and the average temperature and temperature standard deviation of each grid area are calculated. The temperature fluctuation coefficient of each grid area is calculated according to the temperature standard deviation of each grid area. The calculation formula is:

[0072]

[0073] Among them, σ i is the temperature standard deviation, T i is the average temperature, W i is the temperature fluctuation coefficient.

[0074] 2. Based on the preset temperature fluctuation threshold, the grid area with a temperature fluctuation coefficient less than the temperature fluctuation threshold is selected as the stable area to represent the influence of the ambient temperature;

[0075] 4. Based on the morphological analysis method, the noise area in the stable area is eliminated. The noise area represents the grid area with too small area or irregular shape, ensuring that the stable area tested conforms to the structural characteristics of the component.

[0076] 5. Calculate the average temperature of the stable area as the reference temperature, calculate the measured temperature of the edge wear area based on the reference temperature, and correct the measured temperature based on the heat conduction equation to obtain infrared standard data. The edge wear area is characterized as the area where the working surface of the component is located.

[0077] The calculation formula for the reference temperature is:

[0078]

[0079] Among them, T s is the reference temperature, M is the number of pixels in the stable area, T j is the average temperature of each stable region.

[0080] The calculation formula for calculating the measured temperature is:

[0081] T b =T measured -T s

[0082] Among them, T bis the deviation temperature, T measured The temperature detected at the pixel in the edge wear area is obtained by subtracting the reference temperature from the detected temperature. This can remove the offset of the overall ambient temperature, but it still includes the heat conduction effect. Therefore, the temperature needs to be corrected in combination with the heat conduction equation.

[0083] Correct the temperature by combining the heat conduction equation:

[0084]

[0085] Among them, q is the heat flux density, that is, the heat transferred from the environment to the component, k is the thermal conductivity of the component material, d is the thickness of the heat conduction path, and the distance from the working surface at the edge of the component to the temperature stable area is taken. It is the final infrared standard data, which can reflect the temperature increase caused by wear of parts.

[0086] The model building module is used to build a wear model of components and update the wear model in real time based on monitoring data. Specifically, it includes the following steps:

[0087] The model building module is used to build a wear model of components and update the wear model in real time based on monitoring data. Specifically, it includes the following steps:

[0088] Establish a wear model based on the preset wear calculation framework and material parameters of components;

[0089] Determine the wear state of the component based on the monitoring data, including normal wear state, accelerated wear state and critical failure state;

[0090] The wear model is updated and corrected according to the monitoring data and wear status to obtain a new wear model.

[0091] In this embodiment, the wear calculation framework adopts the Archard wear model, and the initial parameters are set according to the specific type of parts, including material hardness H, wear coefficient K, load P, and relative motion speed v.

[0092] Wear status is identified by analyzing monitoring data using LSTM (Long Short-Term Memory) or Bayesian update models. The infrared sensor module generates standard infrared data, representing temperature T(t), to correct for wear caused by thermal stress. The ultrasonic sensor module generates standard ultrasonic data, representing damage depth d(t), reflecting material loss due to crack propagation. The electromagnetic sensor module generates standard electromagnetic data, representing material property change Δσ(t), reflecting the impact of metal fatigue on wear.

[0093] In the normal wear state: T(t) changes slowly, d(t) is approximately 0, and Δσ(t) is stable. At this time, the wear model parameters are adjusted to use a low wear coefficient K1, mainly using the friction term of the Archard model:

[0094]

[0095] In the accelerated wear state: T(t) increases, d(t) increases, Δσ(t) decreases, and the wear model is adjusted to:

[0096]

[0097] Among them, K2 is greater than K1, which means that the wear rate increases.

[0098] In the critical failure state: T(t) changes drastically, d(t) increases sharply, and Δσ(t) changes suddenly, further increasing the wear model weight:

[0099]

[0100] This means that the early warning mechanism will be activated and an alarm will be issued to recommend maintenance.

[0101] Kalman filtering (KF) is used to dynamically adjust the wear model parameters. The state equation is:

[0102] X t+1 =AX t +BU t +w t

[0103] Among them, the state vector:

[0104]

[0105] The wear state is used to adjust the state transfer matrix A to adapt the model to different wear stages.

[0106] The analysis and prediction platform is used to predict the remaining service life of components based on the wear model and generate a wear report, specifically including:

[0107] According to the wear model, the cumulative wear amount W(t) and the current wear rate are obtained. and wear status;

[0108] The remaining service life is calculated based on the accumulated wear, current wear rate and wear status;

[0109] The current wear rate is obtained based on the wear amount changes at consecutive time points of the monitoring data:

[0110]

[0111] Under normal wear conditions, the current wear rate changes slowly, and the remaining service life can be calculated using a linear wear rate:

[0112]

[0113] Among them, W failure The critical wear threshold is determined according to the design standard of the component and represents the maximum amount of wear that the component can withstand before failure.

[0114] In the accelerated wear state, the current wear rate increases, and the remaining service life needs to be included in the wear trend prediction:

[0115]

[0116] in, is the rate of change (acceleration) of wear rate, and λ1 is the adjustment coefficient to make RUL prediction more stable.

[0117] In the critical failure state, the current wear rate reaches the highest, and the calculation method is adjusted:

[0118]

[0119] Among them, f warning It is a nonlinear correction term based on the fault threshold to make the prediction more accurate. λ2 and λ3 are adaptive adjustment coefficients used to adjust the sensitivity of RUL calculation.

[0120] Generate maintenance plans and wear reports based on the remaining useful life. The analysis and prediction platform needs to automatically generate wear reports, provide data visualization, and provide maintenance recommendations. Using tools such as Power BI, Matplotlib, or D3.js, it can plot wear trend curves, temperature-damage correlation curves, and predicted RUL curves in real time.

[0121] Based on the wear prediction results, the analysis and prediction platform generates a maintenance plan: in normal status (RUL>1000 hours), no maintenance is required; in warning status (100≤RUL≤1000 hours), inspection is recommended; in emergency status (RUL<100 hours), replacement of parts is recommended.

[0122] A method for intelligent state monitoring and life prediction of industrial equipment, comprising the following steps:

[0123] The multi-physics field sensor module is used to obtain the surface temperature changes, internal damage conditions, and material performance changes of components to form multi-physics field monitoring data.

[0124] The monitoring data is received and ultrasonic signal correction and infrared temperature data calibration are performed respectively, wherein, for the ultrasonic detection data, the original flight time and echo amplitude are extracted, and the time delay and amplitude attenuation are corrected based on the ultrasonic correction data of the correction unit to obtain ultrasonic standard data; for the infrared detection data, the temperature fluctuation coefficients of multiple grid areas are calculated, the stable areas are screened out, and the temperature data of the edge wear area is corrected based on the average temperature of the stable areas and the heat conduction equation to obtain infrared standard data;

[0125] Build a wear model for components and update it in real time based on monitoring data;

[0126] Predict the remaining service life of components based on wear models and generate wear reports;

[0127] Send wear reports to users and issue alerts when remaining service life reaches a pre-defined threshold.

[0128] In summary, this embodiment has the following beneficial effects:

[0129] Compared with a single sensing method, the multi-physics field sensor module that combines infrared, ultrasonic and electromagnetic sensors can effectively reduce the impact of factors such as ambient temperature fluctuations and dust interference on measurement results, improve the reliability of monitoring data, and enable the system to operate stably in high temperature, low temperature and dust-intensive environments.

[0130] The infrared processing submodule and the ultrasonic processing submodule of the data processing module correct and standardize temperature measurements and ultrasonic signals, respectively. The infrared processing submodule utilizes temperature gridding analysis, morphological analysis, and heat conduction model correction to ensure that temperature data is free of ambient temperature influences, improving the accuracy of temperature measurements in worn areas of components. The ultrasonic processing submodule uses time correction and amplitude compensation to eliminate the effects of dust on ultrasonic propagation characteristics, enabling more accurate internal damage detection.

[0131] Using a model-building module, the wear model is updated in real time based on monitoring data, dynamically adjusting wear parameters to improve the accuracy of component status. The system can identify three states: normal wear, accelerated wear, and critical failure, and adjust the wear calculation method accordingly to ensure the model's adaptability to complex wear conditions.

[0132] The calculation logic of the remaining service life can be adjusted according to different wear stages to ensure that the prediction results are more in line with actual working conditions, thereby providing a scientific basis for equipment maintenance and avoiding losses caused by premature replacement or delayed maintenance.

[0133] The output and alert module automatically sends an alert when the remaining service life reaches a preset threshold, prompting users to take proactive maintenance measures and reduce the risk of equipment downtime due to unexpected failures. Combined with wear and tear reports, users can optimize maintenance plans based on the system's recommendations, improving equipment management efficiency and extending component life.

[0134] The above-described embodiments do not constitute a limitation on the scope of protection of this technical solution. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the above-described embodiments shall be included in the scope of protection of this technical solution.

Claims

1. A system for intelligent status monitoring and life prediction of industrial equipment, characterized in that: include: A multi-physics sensor module, comprising an infrared sensor submodule, an ultrasonic sensor submodule, and an electromagnetic sensor submodule. The infrared sensor submodule is used to detect temperature changes on the surface of a component; the ultrasonic sensor submodule is used to detect damage inside a component; and the electromagnetic sensor submodule is used to detect performance changes of a component. A data processing module, the data processing module is used to receive and process real-time data from the multi-physics field sensor module to obtain monitoring data; Model building module, used to build the wear model of components and update the wear model in real time based on monitoring data; An analysis and prediction platform, used to predict the remaining service life of components based on wear models and generate wear reports; Output and alarm module for sending wear reports to users and issuing alarms when the remaining service life reaches a predetermined threshold; The ultrasonic sensor submodule specifically includes an ultrasonic sensor and a correction unit. The ultrasonic sensor is arranged at a position facing the target component, and the correction unit is arranged on the propagation path of the ultrasonic sensor. The ultrasonic sensor is used to transmit ultrasonic waves and receive ultrasonic signals reflected by the component to obtain ultrasonic detection data; the correction unit is used to generate an electrical signal based on the received ultrasonic wave and generate ultrasonic correction data based on the electrical signal. The data processing module includes an ultrasonic processing submodule, which performs analysis and calculation based on ultrasonic detection data and ultrasonic correction data to obtain ultrasonic standard data. The specific steps are as follows: The original flight time and original echo amplitude are obtained by extracting the ultrasonic detection data, and the time correction parameter and amplitude correction parameter are obtained by extracting the ultrasonic correction data; Based on the time correction parameter, calculating and compensating for the delay deviation of the propagation speed of the ultrasonic wave in the current environment, thereby correcting the original flight time to obtain the standardized time; Based on the amplitude correction parameter, calculating and compensating for the energy attenuation or scattering loss of the ultrasonic wave in the current environment, thereby correcting the original echo amplitude to obtain a standardized amplitude; Obtaining ultrasound standard data according to the normalized amplitude and the normalized magnitude; The infrared sensor submodule detects the temperature change on the surface of the component to obtain infrared image data. The data processing module includes an infrared processing submodule. The infrared processing submodule analyzes and calculates the infrared detection data to obtain infrared standard data. The specific steps are as follows: Preprocessing the infrared image data to obtain an infrared analysis image for enhancing the quality of the infrared image; Divide the infrared analysis image into multiple grid areas and calculate the average temperature and temperature fluctuation coefficient of each grid area; According to the preset temperature fluctuation threshold, the grid area with a temperature fluctuation coefficient less than the temperature fluctuation threshold is selected as the stable area; Eliminate noise regions in stable regions according to a morphological analysis method, wherein the noise regions represent grid regions that are too small or irregular in shape; Calculate the average temperature of the stable area as the reference temperature; Calculating the measured temperature of the edge wear area according to the reference temperature, and correcting the measured temperature based on the heat conduction equation to obtain infrared standard data, wherein the edge wear area is characterized as the area where the working surface of the component is located; The model building module is used to build a wear model of components and update the wear model in real time based on monitoring data. Specifically, it includes the following steps: Establish a wear model based on the preset wear calculation framework and material parameters of components; Determine the wear state of the component based on the monitoring data, including normal wear state, accelerated wear state and critical failure state; The wear model is updated and corrected according to the monitoring data and wear status to obtain a new wear model.

2. The intelligent state monitoring and life prediction system for industrial equipment according to claim 1 is characterized in that: The correction unit is made of piezoelectric material.

3. The intelligent state monitoring and life prediction system for industrial equipment according to claim 1 is characterized in that: The analysis and prediction platform is used to predict the remaining service life of components based on the wear model and generate a wear report, specifically including: According to the wear model, the accumulated wear amount, current wear rate and wear status are obtained; The remaining service life is calculated based on the accumulated wear, current wear rate and wear status; Generate maintenance plans and wear reports based on remaining service life.

4. A method for intelligent state monitoring and life prediction of industrial equipment, applied to the system for intelligent state monitoring and life prediction of industrial equipment according to any one of claims 1 to 3, characterized in that: The specific steps include: The multi-physics field sensor module is used to obtain the surface temperature changes, internal damage conditions, and material property changes of components to form multi-physics field monitoring data. The monitoring data is received and ultrasonic signal correction and infrared temperature data calibration are performed respectively, wherein, for the ultrasonic detection data, the original flight time and echo amplitude are extracted, and the time delay and amplitude attenuation are corrected based on the ultrasonic correction data of the correction unit to obtain ultrasonic standard data; for the infrared detection data, the temperature fluctuation coefficients of multiple grid areas are calculated, the stable areas are screened out, and the temperature data of the edge wear area is corrected based on the average temperature of the stable areas and the heat conduction equation to obtain infrared standard data; Build a wear model for components and update the wear model in real time based on monitoring data; Predict the remaining service life of components based on wear models and generate wear reports; Send wear reports to users and issue alerts when remaining service life reaches a pre-defined threshold.

Images