A method and system for diagnosing the health state of a dredger sensor

Abstract

The application provides a kind of dredger sensor health state diagnosis method and system, belong to ship equipment health management technical field, dredger sensor health state diagnosis method, comprising the following steps: selecting recent consecutive N sampling points, calculate the data mean and standard deviation in sliding window, and define threshold confidence interval;If current sampling value falls into threshold confidence interval for M times in succession;The application uses data acquisition unit to collect mud pump power, slurry density, pressure, flow rate and other data in real time, calculates mean and standard deviation through signal preprocessing unit analysis, and then combines dynamic threshold adaptation module, soil classification strategy module and redundancy check module to determine fault type, then triggers broken line, water ingress, out-of-boundary and other alarm signals according to priority through alarm output unit, uses dynamic threshold adjustment, multi-dimensional data fusion and redundancy check mechanism, solves the problems of static threshold mismatch, poor soil adaptability, false alarm rate and other problems in traditional technology.

Description

Technical Field

[0001] This invention relates to the field of ship equipment health management technology, specifically to a method and system for diagnosing the health status of sensors on a dredger. Background Technology

[0002] In trailing suction hopper dredging operations, various sensors such as mud density sensors, pressure sensors, and flow sensors are core components ensuring the safe operation of equipment and process control. However, due to the complex dredging environment, such as mud fluctuations, equipment vibrations, and diverse soil types, sensors are susceptible to external interference or internal faults, leading to data anomalies. Traditional fault diagnosis methods have the following drawbacks: 1. Static threshold dependence: Using fixed thresholds to judge anomalies cannot adapt to the dynamic characteristics of parameters such as mud density and flow velocity changing with soil type; 2. Sensitivity to single-point faults: Lacking redundancy verification mechanisms, a single sensor fault can easily lead to false alarms or missed alarms; 3. Poor environmental adaptability: Not considering the differentiated impact of different soil types (clay, sand, silt) on sensor data, resulting in a high false alarm rate; 4. Insufficient real-time performance: Fixed sliding window parameters make it difficult to balance noise suppression and anomaly response speed. Therefore, existing technologies urgently need an adaptive and highly robust sensor health status diagnosis method to improve the reliability and safety of dredging operations. Summary of the Invention

[0003] This invention provides an intelligent and adaptive method and system for diagnosing the health status of sensors on dredgers. By using dynamic threshold adjustment, multi-dimensional data fusion, and redundancy verification mechanisms, it solves the problems of static threshold mismatch, poor soil adaptability, and high false alarm rate in traditional technologies.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for diagnosing the health status of sensors on a dredger includes the following steps: Select the most recent N consecutive sampling points and calculate the mean of the data within the sliding window. and standard deviation And define the threshold confidence interval; If the current sample value If the signal falls within the threshold confidence interval M times consecutively, it is considered stable; otherwise, it is marked as abnormal. Combined with historical thresholds and real-time standard deviation The threshold for determining the sensor's health status at the current moment is dynamically adjusted using an exponentially weighted moving average model. And determine the threshold based on the current sensor health status. This forms the updated threshold confidence interval; If the current sample value If the sensor falls within the update threshold confidence interval M times consecutively, it is determined that the sensor is working normally; if the current sample value If the value remains below the lower limit of the update threshold confidence interval, a disconnection fault is determined to have occurred; if the current sampled value... If the value remains above the upper limit of the update threshold confidence interval, a water ingress fault is determined to have occurred.

[0005] Preferably, the average value of the data Use the following formula: The standard deviation Use the following formula: Where N is the number of sampling points within the sliding window. For the first time at time t One real-time sampled value.

[0006] Preferably, the threshold confidence interval satisfies: .in The threshold confidence interval coefficient, The threshold is the number of consecutive stable occurrences.

[0007] Preferably, the current sensor health status determination threshold Use the following formula: ,in, It is the attenuation factor, and , The threshold for determining the sensor's health status at the previous moment.

[0008] Preferably, the update threshold confidence interval satisfies: .

[0009] Preferably, the dredging vessel sensor health status diagnosis method further includes the steps of soil classification and identification and sensor fault judgment, including the following steps: If the detected signal continuously exceeds the upper limit of the threshold confidence interval, water ingress suspicion is triggered; If the soil type is identified as clay at this time, it is directly confirmed as a water ingress fault; If the soil type is identified as sandy soil, the corrected threshold is obtained according to the sandy soil adaptation formula, and the confidence interval of the updated threshold for sandy soil is obtained. Then, the fault judgment is performed. If the signal still exceeds the upper limit of the confidence interval of the updated threshold for sandy soil at this time, it is confirmed as a water ingress fault. If the detected signal is consistently below the lower limit of the threshold confidence interval, a suspected disconnection fault is triggered. If the soil type is identified as sandy soil at this time, it is directly confirmed as a broken line fault; If the soil type is identified as clay, the corrected threshold is obtained according to the clay adaptation formula, and the confidence interval of the clay update threshold is obtained. Then, the fault judgment is performed. If the signal is still lower than the lower limit of the confidence interval of the clay update threshold at this time, it is confirmed as a disconnection fault.

[0010] Preferably, obtaining the confidence interval for the sand soil renewal threshold includes the following steps: Determine the threshold for judging the current sensor health status under sandy soil conditions. ;in, The threshold for determining the sensor's health status at the current moment. As a correction factor, for sandy soil, take... ; Based on the sensor health status determination threshold under sandy soil conditions at the current moment, the confidence interval for the sandy soil update threshold is obtained: .

[0011] Preferably, obtaining the confidence interval for the clay update threshold includes the following steps: Determine the threshold for judging the current sensor health status under clay working conditions. .in, The threshold for determining the sensor's health status at the current moment. As a correction factor, for clay, take... ; Based on the sensor health status determination threshold under clay conditions at the current moment, the confidence interval for the clay update threshold is obtained: .

[0012] A dredger sensor health status diagnosis system according to the aforementioned method includes: The data acquisition unit is used to collect the working parameters of the mud pump in real time. A signal preprocessing unit, wherein the signal preprocessing unit calculates the mean and standard deviation based on sliding window variance analysis; A diagnostic decision unit, wherein the diagnostic decision unit is used to determine the type of sensor fault; and An alarm output unit is used to output an alarm signal.

[0013] Preferably, the diagnostic decision unit includes a dynamic threshold adaptation module, a soil classification strategy module, and a redundancy verification module.

[0014] As can be seen from the above technical solutions, the present invention has the following beneficial effects: 1. In this invention, the data acquisition unit collects data such as mud pump power, mud density, pressure, and flow rate in real time. The mean and standard deviation are analyzed and calculated by the signal preprocessing unit. Then, the fault type is determined by combining the dynamic threshold adaptation module, the soil classification strategy module, and the redundancy verification module. Subsequently, the alarm output unit triggers alarm signals such as disconnection, water ingress, and boundary crossing according to priority.

[0015] 2. In this invention, sensors such as mud density sensors, pressure sensors, and flow sensors are used to monitor the stability of sensor signals in real time. The threshold for judging the health status of the sensors at the current moment is dynamically adjusted based on the exponential weighted moving average model to adapt to the fluctuation characteristics of mud parameters with soil quality changes. The diagnostic decision unit is used to distinguish fault types such as line breakage, water ingress, and boundary crossing.

[0016] 3. In this invention, diagnostic parameters are optimized by soil classification to reduce the impact of environmental interference, while dual-sensor redundancy verification is used to improve the reliability of health status determination. Attached Figure Description

[0017] Figure 1 A block diagram of the dredging vessel sensor health status diagnosis method provided by the present invention; Figure 2 The structural block diagram of the dredging vessel sensor health status diagnosis system provided by the present invention; Figure 3 The flowchart illustrates the health status diagnosis method for dredging vessel sensors provided by this invention.

[0018] In the diagram: 10, Data Acquisition Unit; 20, Signal Preprocessing Unit; 30, Diagnostic Decision Unit; 310, Dynamic Threshold Adaptation Module; 320, Soil Classification Strategy Module; 330, Redundancy Verification Module; 340, Alarm Decision Module; 40, Alarm Output Unit. Detailed Implementation

[0019] A preferred embodiment of the present invention will now be described in detail with reference to the accompanying drawings.

[0020] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions: (Refer to...) Figure 1 , Figure 3 This invention provides a dredging vessel sensor health status diagnosis method M10. The dredging vessel sensor health status diagnosis method M10 of this embodiment includes the following steps: S100, S200, 300, and S400. Furthermore, S100: Select the most recent N consecutive sampling points and calculate the average value of the data within the sliding window. and standard deviation And define the threshold confidence interval.

[0021] Specifically, select the most recent consecutive Each sampling point (e.g.) =15), calculate the mean of the data within the sliding window. and standard deviation ,in, (1) (2) In equations (1) and (2), Current moment The average of the data; : Number of sampling points within the sliding window; At time t, the first Real-time sampled values ​​(i.e., at time 1) The first time (data points) Current moment The standard deviation of the data.

[0022] Furthermore, S200: If the current sample value If the signal falls within the threshold confidence interval for M consecutive times, it is considered stable; otherwise, it is marked as abnormal (such as disconnection or water ingress).

[0023] Specifically, the threshold confidence interval satisfies: (3) In equation (3), Threshold confidence interval coefficient (usually 3, corresponding to a 99.7% confidence level); : Threshold for the number of consecutive stable occurrences to prevent misjudgment of transient fluctuations.

[0024] Based on the aforementioned threshold confidence interval and Further, proceed to the next step, S300: combine historical thresholds. and real-time standard deviation The threshold for determining the sensor's health status at the current moment is dynamically adjusted using an exponentially weighted moving average model. .

[0025] Specifically, the threshold for determining the current sensor health status. Satisfy the following formula: (4) In equation (4), The threshold for determining the sensor's health status at the current moment; The threshold for determining the sensor's health status at the previous moment (based on historical data weighting) has an initial value of 0. Attenuation factor ( ), to control the influence of historical thresholds; The real-time standard deviation of the sensor data at the current moment (reflecting the degree of data fluctuation).

[0026] It should be noted that, pass Instantaneous fluctuations and historical trends in parameter balance: Low Value: High weighting for real-time fluctuations → rapid response to changes in confidence intervals → sensitive anomaly detection; high Value: High weighting of historical data → smooth changes in confidence interval → strong noise resistance.

[0027] The updated threshold confidence interval is formed based on the calculation result of equation (4), and the updated threshold confidence interval is: Thus, proceed to the next step, S400: when the sensor detects values ​​in real time... If the sensor falls within the update threshold confidence interval multiple times (M times, as defined above), it is determined that the sensor is working normally; when the sensor's real-time detection value... When the value remains below the lower limit of the update threshold confidence interval, a disconnection fault is determined to have occurred; when the real-time sensor detection value... If the value remains above the upper limit of the update threshold confidence interval, a water ingress fault is determined to have occurred.

[0028] Furthermore, it should be noted that the physical properties of different soil types (such as clay, sand, silt, etc.) vary significantly, directly affecting the distribution pattern of sensor data and the operating status of the equipment. Based on parameters such as mud density, flow velocity, and sand content, the soil type (clay, sand, etc.) is identified in real time, and key parameters are adjusted. Therefore, the dredging vessel sensor health status diagnosis method also includes the steps of soil classification and identification, and sensor fault diagnosis, including the following steps: If the detected signal continuously exceeds the upper limit of the threshold confidence interval, water ingress suspicion is triggered; If the soil type is identified as clay at this time, it is directly confirmed as a water ingress fault; If the soil type is identified as sandy soil, the corrected threshold is obtained according to the sandy soil adaptation formula, and the confidence interval of the updated threshold for sandy soil is obtained. Then, the fault judgment is performed. If the signal still exceeds the upper limit of the confidence interval of the updated threshold for sandy soil at this time, it is confirmed as a water ingress fault. If the detected signal is consistently below the lower limit of the threshold confidence interval, a suspected disconnection fault is triggered. If the soil type is identified as sandy soil at this time, it is directly confirmed as a broken line fault; If the soil type is identified as clay, the corrected threshold is obtained according to the clay adaptation formula, and the confidence interval of the clay update threshold is obtained. Then, the fault judgment is performed. If the signal is still lower than the lower limit of the confidence interval of the clay update threshold at this time, it is confirmed as a disconnection fault.

[0029] Furthermore, obtaining the confidence interval for the updated sand soil threshold includes the following steps: When the sand is suitable for the sensor, the vibration frequency of the sensor increases due to the friction of sand particles, causing the threshold to rise. Therefore, the threshold for determining the sensor's health status at the current moment under sandy soil conditions is... Satisfy the following formula: (5) In equation (5), The threshold for determining the sensor's health status at the current moment; Correction factor, take (Threshold increased by 20%) to suppress false alarms caused by particle friction.

[0030] Subsequently, based on the sensor health status threshold under the sandy soil condition at the current moment, the confidence interval for the sandy soil update threshold is obtained: .

[0031] Similarly, obtaining the confidence interval for the clay update threshold includes the following steps: When clay is used, the signal fluctuation is small and the threshold is lower due to the poor fluidity of clay. Therefore, the threshold for determining the sensor's health status at the current moment under clay working conditions is... Satisfy the following formula: (6) In formula (6) The threshold for determining the sensor's health status at the current moment; Correction factor, take (Threshold lowered by 10%) to adapt to low volatility characteristics.

[0032] Subsequently, based on the sensor health status threshold at the current moment under clay conditions, the confidence interval for the clay update threshold is obtained: .

[0033] In this way, after obtaining the judgment threshold through formula (4), if the detected signal continuously exceeds the upper limit of the confidence interval, water ingress is suspected; if the soil type is identified as clay at this time, it is directly confirmed as a water ingress fault; if the soil type is identified as sandy soil, the modified threshold is obtained according to the soil type adaptation strategy of formula (5), and the updated threshold confidence interval for sandy soil is obtained again. Then, fault diagnosis is performed. If the signal still exceeds the upper limit of the confidence interval at this time, it is confirmed as a water ingress fault.

[0034] If the detected signal is consistently below the lower limit of the confidence interval, a suspected disconnection fault is triggered; if the soil type is identified as clay at this time, the threshold is corrected according to formula (6) to obtain the clay updated threshold confidence interval: Then, fault diagnosis is performed. If the signal is still below the lower limit of the confidence interval, it is confirmed as a line breakage fault; if the soil is identified as sandy soil, it is directly identified as a line breakage fault.

[0035] Furthermore, in some embodiments, redundant sensors are deployed for key parameters (such as mud pump power, pressure, etc.). That is, two sensors are configured for the same measurement signal. For easy distinction, the two sensors can be named sensor A and sensor B, respectively. In use, if the data of sensor A continuously exceeds the threshold confidence interval... If sensor B's data is stable, then sensor A is marked as "faulty," its data is isolated, and the data input diagnostic process is suspended.

[0036] Reference Figure 2 In some embodiments, the present invention also provides a dredging vessel sensor health status diagnosis system, including a data acquisition unit 10, a signal preprocessing unit 20, a diagnostic decision unit 30, and an alarm output unit 40. Specifically, the data acquisition unit 10 is used to acquire dredging pump operating parameters in real time, the signal preprocessing unit 20 calculates the mean and standard deviation based on sliding window variance analysis, the diagnostic decision unit 30 is used to determine the sensor fault type, and the alarm output unit 40 is used to output an alarm signal.

[0037] In addition, the dredger sensor health status diagnosis system also includes an alarm decision module 340, a dynamic threshold adaptation module 320, a soil classification strategy module 330, and a redundancy verification module 330, all of which are electrically connected to the alarm decision module 340. This allows the diagnostic decision results to be transmitted to the alarm decision module 340 and then transmitted to the alarm output unit 40 for output.

[0038] Furthermore, the data acquisition unit includes a mud pump power sensor, a mud density sensor, a pressure sensor, etc., used to collect sensor data such as mud pump power, mud density, pressure, and flow rate in real time.

[0039] Furthermore, the diagnostic decision unit 30 includes a dynamic threshold adaptation module 310, a soil classification strategy module 320, and a redundancy verification module 330. In use, the data acquisition unit collects data such as mud pump power, mud density, pressure, and flow rate in real time, and the signal preprocessing unit analyzes and calculates the mean. and standard deviation Then, by combining the dynamic threshold adaptation module, the soil classification strategy module, and the redundancy verification module, the fault type is determined, and then the alarm output unit triggers alarm signals such as disconnection, water ingress, and boundary crossing according to priority.

[0040] The above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for diagnosing the health status of sensors on a dredger, characterized in that, Includes the following steps: Select the most recent N consecutive sampling points and calculate the mean of the data within the sliding window. and standard deviation And define the threshold confidence interval; If the current sample value If a signal falls within the threshold confidence interval M times consecutively, it is considered stable; otherwise, it is marked as abnormal. The threshold for the number of consecutive stable occurrences; Combining historical thresholds and real-time standard deviation The threshold for determining the sensor's health status at the current moment is dynamically adjusted using an exponentially weighted moving average model. And determine the threshold based on the current sensor health status. This forms the confidence interval for the updated threshold; If the current sample value If the sensor falls within the update threshold confidence interval M times consecutively, it is determined that the sensor is working normally; if the current sample value If the value remains below the lower limit of the update threshold confidence interval, a disconnection fault is determined to have occurred; if the current sampled value... If the value remains above the upper limit of the update threshold confidence interval, a water ingress fault is determined to have occurred. The dredging vessel sensor health status diagnosis method also includes the steps of soil classification and identification and sensor fault judgment, including the following steps: If the detected signal continuously exceeds the upper limit of the threshold confidence interval, water ingress suspicion is triggered; If the soil type is identified as clay at this time, it is directly confirmed as a water ingress fault; If the soil type is identified as sandy soil, the corrected threshold is obtained according to the sandy soil adaptation formula, and the confidence interval of the updated threshold for sandy soil is obtained. Then, the fault judgment is performed. If the signal still exceeds the upper limit of the confidence interval of the updated threshold for sandy soil at this time, it is confirmed as a water ingress fault. If the detected signal is consistently below the lower limit of the threshold confidence interval, a suspected disconnection fault is triggered. If the soil type is identified as sandy soil at this time, it is directly confirmed as a broken line fault; If the soil type is identified as clay, the corrected threshold is obtained according to the clay adaptation formula, and the clay update threshold confidence interval is obtained. Then, the fault judgment is performed. If the signal is still lower than the lower limit of the clay update threshold confidence interval at this time, it is confirmed as a disconnection fault. The process of obtaining the confidence interval for the sand soil renewal threshold includes the following steps: Determine the threshold for judging the current sensor health status under sandy soil conditions. ;in, The threshold for determining the sensor's health status at the current moment. As a correction factor, for sandy soil, take... ; Based on the sensor health status determination threshold under sandy soil conditions at the current moment, the confidence interval for the sandy soil update threshold is obtained: ,in, The threshold confidence interval coefficient; The process of obtaining the confidence interval for the clay update threshold includes the following steps: Determine the threshold for judging the current sensor health status under clay working conditions. ; in, The threshold for determining the sensor's health status at the current moment. As a correction factor, for clay, take... ; Based on the sensor health status determination threshold under clay conditions at the current moment, the confidence interval for the clay update threshold is obtained: .

2. The method for diagnosing the health status of sensors on a dredger according to claim 1, characterized in that, The average of the data Use the following formula: The standard deviation Use the following formula: Where N is the number of sampling points within the sliding window. For the first time at time t A real-time sampled value.

3. The method for diagnosing the health status of sensors on a dredger according to claim 2, characterized in that, The threshold confidence interval satisfies: .

4. The method for diagnosing the health status of sensors on a dredger according to claim 3, characterized in that, The current sensor health status determination threshold Use the following formula: ,in, It is the attenuation factor, and , The threshold for determining the sensor's health status at the previous moment.

5. The method for diagnosing the health status of sensors on a dredger according to claim 4, characterized in that, The update threshold confidence interval satisfies: .

6. A dredger sensor health status diagnosis system using the dredger sensor health status diagnosis method according to any one of claims 1-5, characterized in that, include: Data acquisition unit (10), the data acquisition unit (10) is used to acquire the working parameters of the mud pump in real time; Signal preprocessing unit (20), which calculates the mean and standard deviation based on sliding window variance analysis; Diagnostic decision unit (30), the diagnostic decision unit (30) is used to determine the type of sensor fault; and An alarm output unit (40) is used to output an alarm signal.

7. The dredging vessel sensor health status diagnosis system according to claim 6, characterized in that, The diagnostic decision unit (30) includes a dynamic threshold adaptation module (310), a soil classification strategy module (320), and a redundancy verification module (330).

Images