Chemical equipment explosion-proof inspection robot fault early warning method based on vibration analysis

Abstract

The invention discloses a fault early warning method for a chemical equipment explosion-proof inspection robot based on vibration analysis, and belongs to the field of robot fault early warning. Comprising the steps of multi-source vibration data acquisition and anti-explosion signal processing, vibration-environment multi-source data fusion diagnosis, statistical process control (SPC) vibration health assessment, anti-explosion environment self-adaptive early warning mechanism and graded early warning and maintenance decision. According to the chemical equipment anti-explosion inspection robot fault early warning method based on vibration analysis, the Db4 wavelet denoising technology is adopted, the fault detection rate is greatly improved compared with a traditional method, mechanical faults and electrical risks are accurately distinguished, the ATEX highest anti-explosion standard is achieved, the sensor self-healing capacity is achieved, the system reliability is improved, and the fault early warning effect is good. Through the daily increasing trend of the entropy value of the vibration envelope spectrum, faults are pre-judged 72 hours ahead of time, most of accidental shutdown losses are reduced, meanwhile, three-level response strategy synergy is achieved, similar vibration modes are combined through the K-means clustering algorithm in each quarter, and the annual diagnosis precision is improved.

Description

Technical Field

[0001] This application relates to the field of robot fault early warning, and in particular to a fault early warning method for explosion-proof inspection robots for chemical equipment based on vibration analysis. Background Technology

[0002] The vibration analysis-based fault early warning method for explosion-proof inspection robots for chemical equipment is a system that uses vibration analysis of the explosion-proof inspection robot itself to provide early warning of robot faults, effectively reducing the failure rate.

[0003] However, existing fault early warning methods for explosion-proof inspection robots in chemical equipment face challenges in strong electromagnetic interference environments. The 1-3kHz high-frequency vibration characteristics of faults such as bearing wear are easily masked by noise, resulting in a detection rate of less than 50% for traditional Fourier transform methods. This leads to the problem of high-frequency feature masking. Existing systems exhibit a response delay >500ms to combustible gas explosion risks, far exceeding ATEX standards. Furthermore, they lack self-healing capabilities when sensors fail, resulting in a high accident rate. Fixed-threshold-based early warning strategies lead to over 60% false shutdowns. The lack of a lifespan prediction mechanism results in preventative maintenance costs exceeding 35% of total expenditures. Moreover, existing systems struggle to balance high-frequency feature preservation with noise reduction using conventional wavelet bases, leading to a >40% missed detection rate for bearing lubrication failures. Independent analysis of vibration and current data fails to distinguish between mechanical faults and electrical risks, resulting in a false alarm rate as high as 18-25%. When combustible gas concentrations exceed limits, only audible and visual alarms are triggered without coordinated shutdown measures, resulting in an average response time of over two seconds for explosion accidents. The use of a one-size-fits-all shutdown approach for maintenance also increases ineffective maintenance costs. Summary of the Invention

[0004] To address the problems mentioned in the background art, this application provides a fault early warning method for explosion-proof inspection robots for chemical equipment based on vibration analysis.

[0005] The fault early warning method for explosion-proof inspection robots for chemical equipment based on vibration analysis provided in this application adopts the following technical solution:

[0006] A fault early warning method for explosion-proof inspection robots for chemical equipment based on vibration analysis includes the following steps:

[0007] S1. Multi-source vibration data acquisition and explosion-proof signal processing:

[0008] S101. Vibration Sensor Deployment: Explosion-proof vibration sensors are installed on the robot joints, drive chassis, and end effector of the robotic arm to collect the vibration spectrum (0-10kHz), acceleration amplitude, and time-domain signals in real time during equipment operation.

[0009] S102, Explosion-proof signal processing:

[0010] Data cleaning: Remove abnormal vibration peaks such as electromagnetic interference and collision noise;

[0011] Normalization: The vibration amplitude is normalized to the [0,1] interval according to the equipment safety threshold. Formula:

[0012]

[0013] in Taken from the equipment's safety explosion-proof vibration limits;

[0014] Wavelet denoising: The vibration signal is decomposed using the Db4 wavelet basis to filter out nonlinear impact noise;

[0015] S2, Vibration-Environment Multi-Source Data Fusion Diagnosis:

[0016] S201, Fusion Logic: Real-time correlation analysis of vibration data with temperature, humidity, and current power;

[0017] S202, Fault Feature Database Matching: Establish a database of vibration faults for chemical equipment;

[0018] S3. Statistical Process Control (SPC) Vibration Health Assessment:

[0019] S301, Control Chart Analysis: XR control charts are used to monitor the fundamental frequency amplitude of vibration of key robot components (joints, drive chassis, robotic arm) in real time. The Shewhart judgment criterion triggers an early warning: when the vibration data of 7 consecutive sampling points exceeds the ±3σ control limit, it is determined that the system has an abnormal degradation risk.

[0020] Process capability index Cpk calculation:

[0021]

[0022] Where USL / LSL is the equipment safety explosion-proof vibration limit, μ is the mean vibration amplitude, and σ is the standard deviation. <1.33 (health threshold), indicating mechanical component degradation;

[0023] S302. Remaining life prediction: Based on the growth trend of vibration envelope spectrum entropy, a Weibull distribution model is fitted to predict the failure time.

[0024] S4. Explosion-proof environment adaptive early warning mechanism:

[0025] S401, Environmental Risk Linkage: When the concentration of combustible gas in the environment is >10%LEL and the robot's vibration amplitude increases sharply, an explosion-proof emergency shutdown is triggered. When the power monitoring shows abnormal vibration and current fluctuation >20%, the risk of short circuit due to aging of the circuit is determined.

[0026] S402, Explosion-proof sensor self-diagnosis: Vibration sensor failure determination, continuously outputting zero signal and constant amplitude for 10 minutes, and then generating a self-test report;

[0027] S5. Tiered Early Warning and Maintenance Decision: Generates early warning level classifications, generates maintenance plans, calls upon the expert knowledge base based on fault type, and outputs targeted maintenance instructions.

[0028] Preferably, the vibration sensor deployed in S1 is an explosion-proof vibration sensor that conforms to the ATEX / IECEx standard. The explosion-proof vibration sensor outputs a 4-20mA analog signal and a CAN bus digital signal, which are directly connected to the robot signal processing module. It is made of 316L stainless steel explosion-proof shell material and has an internal energy storage of ≤0.1mJ.

[0029] Preferably, the establishment of the chemical equipment vibration fault database in S2 is as follows:

[0030] Fault type Vibration characteristics Bearing wear due to lack of lubrication High-frequency resonance peaks (1-3 kHz sustained peaks), envelope spectral entropy > 0.85, kurtosis coefficient > 4.0 Broken gear teeth The amplitude of the sideband at the meshing frequency suddenly increased threefold, and the failure frequency ( Harmonic occurrence Explosion-proof casing loose Random impact signal (>5g), vibration signal variance suddenly increases by 200%. Motor rotor eccentricity Double power frequency (2× The amplitude accounts for 60% of the fundamental frequency, and the phase difference is >90°.

[0031] The K-means clustering algorithm automatically merges similar vibration modes, optimizing the accuracy of feature thresholds.

[0032] Preferably, the remaining life prediction in S3 is based on a degradation model constructed according to the growth trend of the envelope spectrum entropy of the vibration signal, and the failure time is fitted using a three-parameter Weibull distribution: When the entropy value of the envelope spectrum increases by more than 0.15 per day, a remaining lifetime report is automatically generated and pushed to the maintenance terminal.

[0033] Preferably, in the environmental risk linkage in S4, if the robot vibration amplitude suddenly increases by more than 30% within 5 seconds, the robot power supply will be cut off and the inert gas protection system will be activated simultaneously when the explosion-proof emergency stop command is triggered, with a response delay of less than 100ms.

[0034] Preferably, the explosion-proof sensor in S4 has a self-diagnostic output with a zero signal and a constant amplitude fluctuation range of <±0.05g. The self-diagnostic report includes hardware damage, signal interference, positioning code, and maintenance priority (P1-P3).

[0035] Preferably, the graded early warning and maintenance decision-making in S5 uses the following early warning grade classification table:

[0036] grade Triggering conditions Response measures Level 1 (P1) <1.0 or vibration / shock> 8g The machine was shut down immediately and maintenance personnel intervened. Level 2 (P2) 1.0≤ <1.33 and temperature exceeds the standard Mission suspended, maintenance to be completed within 72 hours. Level 3 (P3) Harmonic components continue to increase Maintenance before the next mission

[0037] When a Level 1 warning (P1) is triggered, the inert gas protection system is activated simultaneously. The maintenance plan for a Level 3 warning (P3) integrates the optimized results of the fault feature library.

[0038] In summary, this application includes the following beneficial technical effects:

[0039] 1. By utilizing high-frequency signal lossless extraction and employing Db4 wavelet denoising technology, the bearing wear characteristics of 1-3kHz are fully preserved in a strong electromagnetic interference environment. The fault detection rate is significantly improved compared with traditional methods. It has the advantage of multi-source collaborative analysis, and the vibration, current and temperature parameters are linked in real time. It can accurately distinguish between mechanical faults and electrical risks, and the false judgment rate is reduced to below 5%. The overall fault diagnosis capability is significantly improved.

[0040] 2. It has millisecond-level explosion prevention and control. When the concentration of combustible gas exceeds 10% LEL and the vibration amplitude increases by 30% within 5 seconds, it triggers 100ms-level emergency inerting protection, reaching the highest ATEX explosion-proof standard. It has sensor self-healing capability, automatically diagnoses sensor failure, switches to backup equipment and generates a location maintenance report, improves system reliability, and achieves a leapfrog upgrade in overall explosion-proof safety performance.

[0041] 3. It constitutes a precise life prediction mechanism, which predicts faults 72 hours in advance by observing the daily increase trend of vibration envelope spectrum entropy value, reducing most of the unexpected downtime losses. At the same time, it has a three-level response strategy to enhance efficiency. When the P1 level warning occurs, the machine is shut down for maintenance within 30 minutes to avoid major accidents. When the P3 level warning occurs, the maintenance plan is optimized by combining the feature library, reducing preventive maintenance costs and significantly reducing the overall maintenance cost.

[0042] 4. It can perform self-optimization of the fault feature library, merge similar vibration modes every quarter through K-means clustering algorithm, improve annual diagnostic accuracy, and adopt a double-blind verification mechanism to independently label field data to ensure that the false alarm rate of the feature library is consistently below 3%, thus realizing the system's intelligent evolution capability. Attached Figure Description

[0043] Figure 1 This is an overall structural block diagram of a fault early warning method for an explosion-proof inspection robot for chemical equipment based on vibration analysis, as described in an embodiment of this application. Detailed Implementation

[0044] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] like Figure 1 According to an embodiment of the present invention, a fault early warning method for an explosion-proof inspection robot for chemical equipment based on vibration analysis is provided, the method comprising the following steps:

[0046] S1. Multi-source vibration data acquisition and explosion-proof signal processing:

[0047] S101. Vibration Sensor Deployment: Explosion-proof vibration sensors are installed on the robot joints, drive chassis, and end effector of the robotic arm to collect the vibration spectrum (0-10kHz), acceleration amplitude, and time-domain signals in real time during equipment operation.

[0048] S102, Explosion-proof signal processing:

[0049] Data cleaning: Remove abnormal vibration peaks such as electromagnetic interference and collision noise;

[0050] Normalization: The vibration amplitude is normalized to the [0,1] interval according to the equipment safety threshold. Formula:

[0051]

[0052] in Taken from the equipment's safety explosion-proof vibration limits;

[0053] Wavelet denoising: The vibration signal is decomposed using the Db4 wavelet basis to filter out nonlinear impact noise;

[0054] S2, Vibration-Environment Multi-Source Data Fusion Diagnosis:

[0055] S201, Fusion Logic: Real-time correlation analysis of vibration data with temperature, humidity, and current power;

[0056] S202, Fault Feature Database Matching: Establish a database of vibration faults for chemical equipment;

[0057] S3. Statistical Process Control (SPC) Vibration Health Assessment:

[0058] S301, Control Chart Analysis: XR control charts are used to monitor the fundamental frequency amplitude of vibration of key robot components (joints, drive chassis, robotic arm) in real time. The Shewhart judgment criterion triggers an early warning: when the vibration data of 7 consecutive sampling points exceeds the ±3σ control limit, it is determined that the system has an abnormal degradation risk.

[0059] Process capability index Cpk calculation:

[0060]

[0061] Where USL / LSL is the equipment safety explosion-proof vibration limit, μ is the mean vibration amplitude, and σ is the standard deviation. <1.33 (health threshold), indicating mechanical component degradation;

[0062] S302. Remaining life prediction: Based on the growth trend of vibration envelope spectrum entropy, a Weibull distribution model is fitted to predict the failure time.

[0063] S4. Explosion-proof environment adaptive early warning mechanism:

[0064] S401, Environmental Risk Linkage: When the concentration of combustible gas in the environment is >10%LEL and the robot's vibration amplitude increases sharply, an explosion-proof emergency shutdown is triggered. When the power monitoring shows abnormal vibration and current fluctuation >20%, the risk of short circuit due to aging of the circuit is determined.

[0065] S402, Explosion-proof sensor self-diagnosis: Vibration sensor failure determination, continuously outputting zero signal and constant amplitude for 10 minutes, and then generating a self-test report;

[0066] S5. Tiered Early Warning and Maintenance Decision: Generates early warning level classifications, generates maintenance plans, calls upon the expert knowledge base based on fault type, and outputs targeted maintenance instructions.

[0067] In this embodiment, the vibration sensor deployed in S1 is an explosion-proof vibration sensor that conforms to the ATEX / IECEx standard. The explosion-proof vibration sensor outputs a 4-20mA analog signal and a CAN bus digital signal, which are directly connected to the robot signal processing module. It is made of 316L stainless steel explosion-proof shell material and has an internal energy storage of ≤0.1mJ.

[0068] In this embodiment, the chemical equipment vibration fault database is established in S2 as follows:

[0069] Fault type Vibration characteristics Bearing wear due to lack of lubrication High-frequency resonance peaks (1-3 kHz sustained peaks), envelope spectral entropy > 0.85, kurtosis coefficient > 4.0 Broken gear teeth The amplitude of the sideband at the meshing frequency suddenly increased threefold, and the failure frequency ( Harmonic occurrence Explosion-proof casing loose Random impact signal (>5g), vibration signal variance suddenly increases by 200%. Motor rotor eccentricity Double power frequency (2× The amplitude accounts for 60% of the fundamental frequency, and the phase difference is >90°.

[0070] The K-means clustering algorithm automatically merges similar vibration modes, optimizing the accuracy of feature thresholds.

[0071] In this embodiment, the remaining lifetime prediction in S3 is based on a degradation model constructed according to the growth trend of the envelope spectrum entropy of the vibration signal, and the failure time is fitted using a three-parameter Weibull distribution: When the entropy value of the envelope spectrum increases by more than 0.15 per day, a remaining lifetime report is automatically generated and pushed to the maintenance terminal.

[0072] In this embodiment, the environmental risk linkage in S4 is triggered when the robot's vibration amplitude suddenly increases by more than 30% within 5 seconds, which triggers the explosion-proof emergency stop command. At this time, the robot's power supply is cut off and the inert gas protection system is activated, with a response delay of less than 100ms.

[0073] In this embodiment, the explosion-proof sensor in S4 performs self-diagnosis, outputting zero signal and a constant amplitude fluctuation range of <±0.05g. The self-test report includes hardware damage, signal interference, location coding, and maintenance priority (P1-P3).

[0074] Preferably, the graded early warning and maintenance decision-making in S5 uses the following early warning level classification table:

[0075] grade Triggering conditions Response measures Level 1 (P1) <1.0 or vibration / shock> 8g The machine was shut down immediately and maintenance personnel intervened. Level 2 (P2) 1.0≤ <1.33 and temperature exceeds the standard Mission suspended, maintenance to be completed within 72 hours. Level 3 (P3) Harmonic components continue to increase Maintenance before the next mission

[0076] When a Level 1 warning (P1) is triggered, the inert gas protection system is activated simultaneously. The maintenance plan for a Level 3 warning (P3) integrates the optimized results of the fault feature library.

[0077] It should be noted that in the explosion-proof adaptability design principle of vibration signal processing, the wavelet denoising adopts the Db4 wavelet basis instead of the conventional Db6 / Haar. This is because, while preserving the high-frequency fault characteristics of 1-3kHz (such as bearing lubrication failure), it can effectively filter out electromagnetic interference (inverter radiation) and low-frequency noise (<100Hz) specific to chemical environments and explosion impacts. This design complies with the IEC60079-27 explosion-proof equipment signal processing specification and normalization processing. The values ​​are directly related to the vibration safety limits for explosion-proof equipment in GB 3836-2010, ensuring that the warning thresholds are legally compliant.

[0078] In the failure protection mechanism of multi-source fusion diagnosis, when the vibration sensor fails and outputs a constant amplitude of ±0.05g, the system automatically switches to the temperature-current coupling analysis mode. When the temperature exceeds the standard (>85℃) and the current fluctuation is >20%, the risk of short circuit in the motor winding is determined. The mechanism ensures that ASIL-D level functional safety is maintained when the sensor fails.

[0079] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may exist in actual implementation. Modules described as separate components may or may not be physically separated, and components shown as modules may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the method in this embodiment according to actual needs.

[0080] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A fault early warning method for explosion-proof inspection robots for chemical equipment based on vibration analysis, characterized in that: Includes the following steps: S1. Multi-source vibration data acquisition and explosion-proof signal processing: S101. Vibration Sensor Deployment: Explosion-proof vibration sensors are installed on the robot joints, drive chassis, and end effector of the robotic arm to collect the vibration spectrum (0-10kHz), acceleration amplitude, and time-domain signals in real time during equipment operation. S102, Explosion-proof signal processing: Data cleaning: Remove abnormal vibration peaks such as electromagnetic interference and collision noise; Normalization: The vibration amplitude is normalized to the [0,1] interval according to the equipment safety threshold. Formula: in Taken from the equipment's safety explosion-proof vibration limits; Wavelet denoising: The vibration signal is decomposed using the Db4 wavelet basis to filter out nonlinear impact noise; S2, Vibration-Environment Multi-Source Data Fusion Diagnosis: S201, Fusion Logic: Real-time correlation analysis of vibration data with temperature, humidity, and current power; S202, Fault Feature Database Matching: Establish a database of vibration faults for chemical equipment; S3. Statistical Process Control (SPC) Vibration Health Assessment: S301, Control Chart Analysis: XR control charts are used to monitor the fundamental frequency amplitude of vibration of key robot components (joints, drive chassis, robotic arm) in real time. The Shewhart judgment criterion triggers an early warning: when the vibration data of 7 consecutive sampling points exceeds the ±3σ control limit, it is determined that the system has an abnormal degradation risk. Process capability index Cpk calculation: Where USL / LSL is the equipment safety explosion-proof vibration limit, μ is the mean vibration amplitude, and σ is the standard deviation. <1.33 (health threshold), indicating mechanical component degradation; S302. Remaining life prediction: Based on the growth trend of vibration envelope spectrum entropy, a Weibull distribution model is fitted to predict the failure time. S4. Explosion-proof environment adaptive early warning mechanism: S401, Environmental Risk Linkage: When the concentration of combustible gas in the environment is >10%LEL and the robot's vibration amplitude increases sharply, an explosion-proof emergency shutdown is triggered. When the power monitoring shows abnormal vibration and current fluctuation >20%, the risk of short circuit due to aging of the circuit is determined. S402, Explosion-proof sensor self-diagnosis: Vibration sensor failure determination, continuously outputting zero signal and constant amplitude for 10 minutes, and then generating a self-test report; S5. Tiered Early Warning and Maintenance Decision: Generates early warning level classifications, generates maintenance plans, calls upon the expert knowledge base based on fault type, and outputs targeted maintenance instructions.

2. The fault early warning method for explosion-proof inspection robots of chemical equipment based on vibration analysis according to claim 1, characterized in that: The vibration sensor deployment in S1 is an explosion-proof vibration sensor that conforms to the ATEX / IECEx standard. The explosion-proof vibration sensor outputs a 4-20mA analog signal and a CAN bus digital signal, which are directly connected to the robot signal processing module. It is made of 316L stainless steel explosion-proof shell material and has an internal energy storage of ≤0.1mJ.

3. The method for fault early warning of explosion-proof inspection robot for chemical equipment based on vibration analysis according to claim 1, characterized in that: The chemical equipment vibration fault database established in S2 is as follows: The K-means clustering algorithm automatically merges similar vibration modes, optimizing the accuracy of feature thresholds.

4. The fault early warning method for an explosion-proof inspection robot for chemical equipment based on vibration analysis according to claim 1, characterized in that: The remaining life prediction in S3 is based on the degradation model constructed by the growth trend of the envelope spectrum entropy value of the vibration signal. The fault time is fitted by a three-parameter Weibull distribution: [#imgpt6#]. When the growth rate of the envelope spectrum entropy value exceeds the threshold of >0.15 per day, a remaining life report is automatically generated and pushed to the maintenance terminal.

5. A fault early warning method for an explosion-proof inspection robot for chemical equipment based on vibration analysis according to claim 3, characterized in that: In the S4 environmental risk linkage, if the robot vibration amplitude suddenly increases by more than 30% within 5 seconds, the explosion-proof emergency stop command will be triggered, and the robot power will be cut off simultaneously and the inert gas protection system will be activated, with a response delay of less than 100ms.

6. The fault early warning method for an explosion-proof inspection robot for chemical equipment based on vibration analysis according to claim 1, characterized in that: The explosion-proof sensor in S4 has a self-diagnostic output with a zero signal and a constant amplitude fluctuation range of <±0.05g. The self-diagnostic report includes hardware damage, signal interference, positioning code, and maintenance priority (P1-P3).

7. A fault early warning method for an explosion-proof inspection robot for chemical equipment based on vibration analysis according to claim 5, characterized in that: The graded early warning and maintenance decision-making in S5 includes the following early warning level classification table: When a Level 1 warning (P1) is triggered, the inert gas protection system is activated simultaneously. The maintenance plan for a Level 3 warning (P3) integrates the optimized results of the fault feature library.

Images