Unattended intelligent monitoring method and system for aluminum oxide scraper

Abstract

The invention discloses an unattended intelligent monitoring method and system for an aluminum oxide scraper conveyor, and the system comprises a sensing layer, an edge layer, an execution layer and a platform end, and the sensing layer comprises a collection device which is used for collecting the real-time operation data of equipment; the edge layer comprises a fault prediction model and is used for carrying out processing and health diagnosis on the data acquired by the sensing layer; the execution layer is used for executing fault processing according to the health diagnosis result; and the platform end comprises a visualization module, a remote control module and an alarm module, and is used for realizing video monitoring, online monitoring and historical playback of the sensing layer, a fault analysis function of the edge layer, equipment operation trend judgment, abnormity alarm and remote control execution fault processing. Through real-time monitoring and intelligent diagnosis, faults can be warned in advance, sudden shutdown can be avoided, slight abnormities can be automatically handled, delay caused by manual intervention is reduced, it is guaranteed that the production process is more stable, raw material supply is more continuous, and the efficiency of subsequent procedures is improved.

Description

Technical Field

[0001] The present invention relates to an unmanned intelligent monitoring method and system for an alumina scraper, belonging to the technical field of intelligent monitoring for alumina production equipment. Background Art

[0002] In the alumina production process, scraper conveyors are used for continuous conveying of powder, subject to prolonged exposure to high dust and heavy loads. The existing system relies on manual inspections, requiring regular checks of parameters such as chain tension, roller wear, and bearing temperature. This results in low efficiency and a high rate of missed detections (industry statistics indicate a missed fault detection rate of >30%). Specific deficiencies include: (1) Traditional monitoring is single: it cannot cover complex faults such as deviation, material blockage, and chain breakage; (2) High safety risks: Manual inspections in high-dust environments can easily lead to the inhalation of large amounts of dust; (3) Inefficient operation and maintenance: Fault response time > 2 hours, downtime losses reaching 10,000 yuan per minute; (4) Monitoring blind spots: lack of real-time perception of the global status of the equipment; (5) Control hysteresis: The PLC only responds to threshold alarms and cannot predict gradual faults such as chain wear; (6) Uneven distribution: Manual adjustment of the distribution trolley position leads to accumulation and segregation of materials in the tank. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide an unmanned intelligent monitoring method and system for an alumina scraper, so as to at least solve the technical problems mentioned in the background technology.

[0004] The purpose of the present invention is achieved through the following technical solutions: An unmanned intelligent monitoring system for an alumina scraper, characterized by comprising: a perception layer, an edge layer, an execution layer and a platform end, The perception layer includes a collection device for collecting real-time operation data of the equipment; The edge layer includes a fault prediction model, which is used to process and perform health diagnosis on the data collected by the perception layer; The execution layer is used to perform fault processing according to the health diagnosis results; The platform includes a visualization module, a remote control module and an alarm module. The visualization module is used to realize video monitoring, online monitoring and historical playback of the perception layer. The remote control module is used to realize the fault analysis function of the edge layer and equipment operation trend judgment. The alarm module is used for abnormal alarm and remote control to execute fault processing.

[0005] Furthermore, the acquisition device includes one or more of a temperature and vibration sensor, a voiceprint sensor, a video camera, a rotary encoder, a fabric trolley positioner, and a distance sensor.

[0006] Furthermore, the temperature and vibration sensor is arranged on the motor reducer of the scraper conveyor to monitor the bearing temperature and vibration intensity; the sound print sensor is arranged in the head sprocket cover to monitor the abnormal sound and breakage of the chain engagement; the video camera is arranged on both sides of the silo to monitor the material stacking status; the rotary encoder is arranged on the walking motor to monitor the walking speed matching; the material distribution trolley positioner is arranged on the material distribution trolley to monitor the real-time coordinates of the trolley; the distance sensor is arranged on the rake frame to monitor the distance from the rake frame to the material pile.

[0007] Furthermore, the fault prediction model includes: A data processing module, which is used for data alignment, feature extraction and data fusion; The health diagnosis module is used to preset a diagnosis basis, compare the processed data with the diagnosis basis, and judge the equipment operation trend.

[0008] Furthermore, the execution layer includes a blowing device, a motor speed regulator, a material distribution trolley driver and a PLC switch control, which are used for any one or more actions of silo blowing, trolley position adjustment, switching walking mode, and chain parking.

[0009] An unmanned intelligent monitoring method for an alumina scraper comprises the following steps: S1: Data acquisition, collecting real-time operation data of the equipment, the real-time operation data including one or more of equipment mechanical status data, electrical parameters, material distribution data and current parameters; S2: Build a fault prediction model based on the collected data, perform data processing and health diagnosis, and determine the development trend of equipment failures; S3: When the equipment shows a trend of failure, monitoring, early warning and real-time decision-making mechanisms are implemented.

[0010] Furthermore, in step S1, the mechanical state data includes vibration data and voiceprint data of the mechanical equipment, the electrical parameters include motor operating temperature and vibration data, and the material distribution data includes material distribution video and positioning data; the current parameter is the instantaneous working current of the scraper ( ).

[0011] Furthermore, the vibration data and voiceprint data of the mechanical equipment are the vibration data and voiceprint data of the scraper chain; the motor operating temperature is the motor bearing temperature ( ) and vibration data; the material distribution video and positioning data include silo video data, real-time coordinate data of the material distribution trolley, and distance data from the rake to the material pile.

[0012] Furthermore, in step S2, the specific steps of data processing and health diagnosis include: S2.1. Data alignment: Use timestamp synchronization technology to establish a unified data frame with a 500ms (t) time window. S2.2, Feature extraction, including, Vibration signal ( ): FFT spectrum analysis, focusing on the 1-3kHz bearing fault characteristic frequency band; Voiceprint signal ( ): MFCC (Mel-frequency cepstral coefficient) feature extraction; Video data ( ): YOLOv5 real-time target detection, the target is to identify the material accumulation area; Location Data ( ): Differential calculation of the cloth cart's moving trajectory; S2.3, feature fusion, establish device state matrix, ; S2.4. Determine the development trend of equipment failure based on the diagnostic evidence.

[0013] Furthermore, in step S3, when the device shows a fault development trend, the real-time decision-making mechanism is executed, including: (1) Chain fatigue: continuous monitoring is performed; chain tooth skipping or chain breakage: interlocking and parking are performed; (2) Severe blockage triggers injection; (3) If the material is unevenly distributed, the method of implementation is to adjust the position of the trolley; (4) If the vehicle deviates from the track, the execution mode is interlock parking; (5) Motor failure, the execution mode is interlock parking; (6) When the distance between the rake and the raw material is ≤10cm, switch the walking mode and start the rake and scraper.

[0014] Compared with the prior art, the present invention has the following beneficial effects: (1) Reduce downtime Through real-time monitoring (current, temperature, vibration, material level, etc.) and intelligent diagnosis, the system can provide early warning of faults (such as blockage, deviation, chain breakage, and bearing overheating) to avoid sudden downtime. It can also automatically handle minor anomalies, reducing delays caused by manual intervention, ensuring a more stable production process, a more continuous supply of raw materials, and improving the efficiency of subsequent processes (such as dissolution and sedimentation).

[0015] (2) Save labor costs Reduce the number of on-site operators and inspection personnel: no need for dedicated personnel to operate 24 hours a day and conduct frequent on-site inspections; Optimize maintenance staffing: Shift from reactive repairs to predictive maintenance to improve maintenance efficiency and reduce maintenance manpower requirements.

[0016] (3) Reduce maintenance costs Predictive maintenance: Based on sensor data analysis, accurately schedule maintenance at the early stages of equipment degradation, preventing minor problems from developing into major repairs and extending equipment life. Reduce unplanned maintenance costs: prevent sudden failures, reduce emergency repair costs and spare parts loss; Reduce excessive maintenance: Avoid unnecessary scheduled downtime for maintenance.

[0017] (4) Eliminate personnel safety risks Avoid exposure to hazardous areas: Scraper conveyors are often located in environments with high temperatures, high dust levels, high noise levels, confined spaces, or where there is a risk of mechanical injury. Unmanned operation completely eliminates the risk of mechanical injury, slips, dust inhalation, and noise exposure for operators and inspectors near the equipment.

[0018] (5) Improve the level of intrinsic safety The system is equipped with comprehensive safety interlocks (such as emergency stop, deviation interlock, material blockage interlock, and personnel intrusion detection) and automatic protection functions, and its response speed far exceeds that of manual operation.

[0019] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which: Figure 1 This is a structural block diagram of an unmanned intelligent monitoring system for an alumina scraper provided by one embodiment of the present invention; Figure 2 This is the structural diagram of the perception layer; Figure 3 This is the edge layer intelligent diagnosis flow chart. DETAILED DESCRIPTION

[0021] 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 some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0022] Example 1 like Figure 1-Figure 3 As shown in the figure, an unmanned intelligent monitoring system for an alumina scraper includes: a perception layer, an edge layer, an execution layer, and a platform end. The perception layer includes a collection device for collecting real-time operation data of the equipment; The edge layer includes a fault prediction model, which is used to process and perform health diagnosis on the data collected by the perception layer; The execution layer is used to perform fault processing according to the health diagnosis results; The platform includes a visualization module, a remote control module and an alarm module. The visualization module is used to realize video monitoring, online monitoring and historical playback of the perception layer. The remote control module is used to realize the fault analysis function of the edge layer and equipment operation trend judgment. The alarm module is used for abnormal alarm and remote control to execute fault processing.

[0023] The acquisition device includes one or more of a temperature and vibration sensor, a voiceprint sensor, a video camera, a rotary encoder, a cloth trolley positioner, and a distance sensor. Figure 2 ,The deployment of the perception layer is shown in Table 1: Table 1 The fault prediction model includes: A data processing module, which is used for data alignment, feature extraction and data fusion; The health diagnosis module is used to preset a diagnosis basis, compare the processed data with the diagnosis basis, and judge the equipment operation trend.

[0024] The execution layer includes a blowing device, a motor speed regulator, a material distribution trolley driver and a PLC switch control, which are used for any one or more actions of silo blowing, trolley position adjustment, switching walking mode, and chain parking.

[0025] The platform can be used to remotely view the real-time scene on site and the real-time operating data of each device, and the energy consumption and operating status of the device can be viewed through the operating trend. The operating data and trends of a single device in a certain time period can be viewed through the historical playback function. When the equipment operation is abnormal, the platform will sound and light alarms, and the administrator can view the fault problem through fault analysis and remotely control the scraper's travel, scraper, rake frame, and material distribution trolley.

[0026] An unmanned intelligent monitoring method for an alumina scraper based on the above system includes the following steps: S1: Data acquisition, collecting real-time operation data of the equipment, the real-time operation data including one or more of equipment mechanical status data, electrical parameters, material distribution data and current parameters; S2: Build a fault prediction model based on the collected data, perform data processing and health diagnosis, and determine the development trend of equipment failures; S3: When the equipment shows a trend of failure, monitoring, early warning and real-time decision-making mechanisms are implemented.

[0027] In step S1, the mechanical state data includes vibration data and voiceprint data of the mechanical equipment, the electrical parameters include motor operating temperature and vibration data, and the material distribution data includes material distribution video and positioning data; the current parameter is the instantaneous working current of the scraper ( Specifically, the vibration data and voiceprint data of the mechanical equipment are the vibration data and voiceprint data of the scraper chain; the motor operating temperature is the motor bearing temperature ( ) and vibration data; the material distribution video and positioning data include silo video data, real-time coordinate data of the material distribution trolley, and distance data from the rake to the material pile.

[0028] In step S2, the specific steps of data processing and health diagnosis include: S2.1. Data alignment: Use timestamp synchronization technology to establish a unified data frame with a 500ms (t) time window. S2.2, Feature extraction, including, Vibration signal ( ): FFT spectrum analysis, focusing on the 1-3kHz bearing fault characteristic frequency band; Voiceprint signal ( ): MFCC (Mel-frequency cepstral coefficient) feature extraction; Video data ( ): YOLOv5 real-time target detection, the target is to identify the material accumulation area; Location Data ( ): Differential calculation of the cloth cart's moving trajectory; S2.3, feature fusion, establish device state matrix, ; S2.4. Determine the development trend of equipment failure based on diagnostic evidence.

[0029] Specifically, one or more features can be fused according to different problem types, as follows: .

[0030] The fault prediction model includes a chain health diagnosis model, a material blockage identification module, a material distribution analysis model and a motor health diagnosis model.

[0031] The chain health diagnosis model includes chain tooth skipping diagnosis and chain life prediction: / / Chain tooth skipping diagnosis algorithm public boolean isChainSkipping(double[] audioSpectrum, doublevibration) { double peak3k = calculatePeak(audioSpectrum, 3000, 5000); / / 3-5kHz peak return (peak3k>0.85)&&(vibration>5.0); } / / Chain life prediction model public double predictRemainingLife(double elongation, doublespeedFluctuation) { double wearRate = (elongation - INITIAL_LENGTH) / OPERATING_HOURS; return (MAX_ELONGATION - elongation) / wearRate; / / Remaining hours }.

[0032] The blockage identification module includes visual and electrical blockage detection: / / Blockage detection based on vision + current public BlockageLevel checkBlockage(Mat videoFrame, double current) { double blockageRatio = calculateBlockageArea(videoFrame); double currentSurge = (current - BASELINE_CURRENT) / BASELINE_CURRENT; if (blockageRatio>0.8&¤tSurge>0.3) { return BlockageLevel.CRITICAL; } else if (blockageRatio>0.6&¤tSurge>0.15) { return BlockageLevel.WARNING; } return BlockageLevel.NORMAL; }.

[0033] The cloth analysis model includes the PID cloth control algorithm: / / PID cloth control algorithm public double calculateAdjustment(double[] heightProfile) { double error = 0; for (int i = 0; i <heightProfile.length; i++) { error += (heightProfile[i] -TARGET_HEIGHT) * POSITION_WEIGHT[i]; } / / PID control: Kp=0.8, Ki=0.05, Kd=0.1 integralError += error * DT; double derivative = (error - prevError) / DT; return 0.8*error + 0.05*integralError + 0.1*derivative; }.

[0034] The motor health diagnosis model includes bearing overheating diagnosis, bearing damage diagnosis, and electrical fault diagnosis: / / Rule 1: Bearing overheating diagnosis (combined severity levels) if (θ_t>MAX_TEMP * 1.2 || θ_t>MAX_TEMP) { return 1; / / OVERHEAT } / / Rule 2: Bearing damage diagnosis (prioritized spectrum analysis) if (v_m>MAX_VIBRATION) { double[]spectrum = FFT.analyze(vibrationSignal); double bpfo = calculateBPFO(1470); / / Example speed 1470RPM double baseline = getBaselineSpectrum(bpfo); / / Vibration exceeds the limit + characteristic frequency exceeds the limit if (v_m>MAX_VIBRATION * 1.5 || spectrum[(int)bpfo]>baseline * 3.0) { return 2; / / BEARING_DAMAGE } } / / Rule 3: Electrical Troubleshooting if (i_phases.length>= 3) { double imbalance = calculateCurrentImbalance(i_phases); if (imbalance>IMBALANCE_RATIO) { double thd = calculateTHD(i_phases); if (thd>0.15 || hasHarmonicSignature(i_phases)) { return 3; / / ELECTRICAL_FAULT } } } return 0; / / HEALTHY } / / Current imbalance rate calculation (keep the original implementation) private static double calculateCurrentImbalance(double[]phases) { double avg = (phases[0] +phases[1] + phases[2]) / 3; return Math.max( Math.abs(phases[0]-avg) / avg, Math.max( Math.abs(phases[1]-avg) / avg, Math.abs(phases[2]-avg) / avg ) ); } / / Bearing Fault Frequency Occurrence (BPFO) private static double calculateBPFO(double rpm) { double f_rpm = rpm / 60.0; return f_rpm * BEARING_BALL_COUNT * 0.5 * (1 - (BALL_DIAM / CAGE_DIAM)*Math.cos(Math.toRadians(CONTACT_ANGLE))); } / / Example implementation of harmonic analysis (need to replace the real algorithm) private static double calculateTHD(double[] phases) { / / Simplified THD calculation (actually requires FFT) double fundamental = (phases[0]+phases[1]+phases[2]) / 3; double harmonic = 0.15 * fundamental; / / example value return harmonic / fundamental; } private static boolean hasHarmonicSignature(double[]phases) { / / Detect 5th / 7th harmonics (sample pseudocode) / / double h5 = getHarmonic(phases, 5); / / double h7 = getHarmonic(phases, 7); / / return (h5>threshold) || (h7>threshold); return false; / / Actual implementation } private static double getBaselineSpectrum(double freq) { / / Get the baseline spectrum value (needs to be implemented based on historical data) return 0.05; / / Example baseline value } }.

[0035] In step S3, when the device shows a fault development trend, the real-time decision-making mechanism is executed, including: (1) Chain fatigue: continuous monitoring is performed; chain tooth skipping or chain breakage: interlocking and parking are performed; (2) Severe blockage triggers injection; (3) The material is unevenly distributed. The execution method is to adjust the position of the trolley; (4) If the vehicle deviates from the track, the execution mode is interlock parking; (5) Motor failure, the execution mode is interlock parking; (6) When the distance between the rake and the raw material is ≤10cm, switch the walking mode and start the rake and scraper.

[0036] See Table 2 for details: Table 2 The above description is only a preferred embodiment of the present invention and does not constitute any form of confidentiality restriction on the present invention. Any simple modification, equivalent change and modification of the above embodiment that does not deviate from the content of the technical solution of the present invention and is based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. An unmanned intelligent monitoring system for an alumina scraper, characterized in that: include: Perception layer, edge layer, execution layer and platform end, The perception layer includes a collection device for collecting real-time operation data of the equipment; The edge layer includes a fault prediction model, which is used to process and perform health diagnosis on the data collected by the perception layer; The execution layer is used to perform fault processing according to the health diagnosis results; The platform includes a visualization module, a remote control module and an alarm module. The visualization module is used to realize video monitoring, online monitoring and historical playback of the perception layer. The remote control module is used to realize the fault analysis function of the edge layer and equipment operation trend judgment. The alarm module is used for abnormal alarm and remote control to execute fault processing.

2. The intelligent monitoring system according to claim 1, characterized in that: The acquisition device includes one or more of a temperature and vibration sensor, a voiceprint sensor, a video camera, a rotary encoder, a fabric trolley positioner, and a distance sensor.

3. The intelligent monitoring system according to claim 2, characterized in that: The temperature and vibration sensor is arranged on the motor reducer of the scraper conveyor and is used to monitor the bearing temperature and vibration intensity; The sound pattern sensor is arranged in the head sprocket cover and is used to monitor abnormal chain meshing noise and breakage; The video cameras are set on both sides of the silo to monitor the material accumulation status; The rotary encoder is arranged on the walking motor and is used to monitor the walking speed matching; The cloth trolley positioner is arranged on the cloth trolley and is used to monitor the real-time coordinates of the trolley; The distance sensor is arranged on the rake frame and is used to monitor the distance from the rake frame to the pile of materials.

4. The intelligent monitoring system according to claim 1, characterized in that: The fault prediction model includes: A data processing module, which is used for data alignment, feature extraction and data fusion; The health diagnosis module is used to preset a diagnosis basis, compare the processed data with the diagnosis basis, and judge the equipment operation trend.

5. The intelligent monitoring system according to claim 1, characterized in that: The execution layer includes a blowing device, a motor speed regulator, a material distribution trolley driver and a PLC switch control, which are used for any one or more actions of silo blowing, trolley position adjustment, switching walking mode, and chain parking.

6. An unmanned intelligent monitoring method for an alumina scraper, characterized in that: The following steps are involved: S1: Data acquisition, collecting real-time operation data of the equipment, the real-time operation data including one or more of equipment mechanical status data, electrical parameters, material distribution data and current parameters; S2: Build a fault prediction model based on the collected data, perform data processing and health diagnosis, and determine the development trend of equipment failures; S3: When the equipment shows a trend of failure, monitoring, early warning and real-time decision-making mechanisms are implemented.

7. The unmanned intelligent monitoring method for an alumina scraper according to claim 6, characterized in that: In step S1, the mechanical state data includes vibration data and voiceprint data of the mechanical equipment, the electrical parameters include motor operating temperature and vibration data, and the material distribution data includes material distribution video and positioning data; the current parameter is the instantaneous working current of the scraper ( ).

8. The unmanned intelligent monitoring method for an alumina scraper according to claim 7, characterized in that: The vibration data and voiceprint data of the mechanical equipment are the vibration data and voiceprint data of the scraper chain; the motor operating temperature is the motor bearing temperature ( ) and vibration data; the material distribution video and positioning data include silo video data, real-time coordinate data of the material distribution trolley, and distance data from the rake to the material pile.

9. The unmanned intelligent monitoring method for an alumina scraper according to claim 6, characterized in that: In step S2, the specific steps of data processing and health diagnosis include: S2.

1. Data alignment: Use timestamp synchronization technology to establish a unified data frame with a 500ms (t) time window. S2.2, Feature extraction, including, Vibration signal ( ): FFT spectrum analysis, focusing on the 1-3kHz bearing fault characteristic frequency band; Voiceprint signal ( ): MFCC (Mel-frequency cepstral coefficient) feature extraction; Video data ( ): YOLOv5 real-time target detection, the target is to identify the material accumulation area; Location Data ( ): Differential calculation of the cloth cart's moving trajectory; S2.3, feature fusion, establish device state matrix, ； S2.

4. Determine the development trend of equipment failure based on the diagnostic evidence.

10. The unmanned intelligent monitoring method for an alumina scraper according to claim 6, characterized in that: In step S3, when the device shows a fault development trend, the real-time decision-making mechanism is executed, including: (1) Chain fatigue: continuous monitoring is performed; chain tooth skipping or chain breakage: interlocking and parking are performed; (2) Severe blockage triggers injection; (3) If the material is unevenly distributed, the method of implementation is to adjust the position of the trolley; (4) If the vehicle deviates from the track, the execution mode is interlock parking; (5) Motor failure, the execution mode is interlock parking; (6) When the distance between the rake and the raw material is ≤10cm, switch the walking mode and start the rake and scraper.

Images