CN-121977877-A - Method, device and system for monitoring faults of shield spiral machine

CN121977877ACN 121977877 ACN121977877 ACN 121977877ACN-121977877-A

Abstract

The invention discloses a method, a device and a system for monitoring faults of a shield spiral machine. The invention relates to the technical field of shield screw machine fault monitoring, and solves the problem that the prior art lacks a complete processing flow from the discovery of screw machine system faults and the judgment of faults to the solution of faults and the follow-up review optimization. The method comprises the steps of obtaining measurement data of a plurality of groups of sensors installed on a shield spiral machine, processing the measurement data to obtain characteristic parameters, comparing the characteristic parameters with a preset threshold value, extracting historical data in a corresponding time window when the parameters are over-limit, calculating and extracting a characteristic set of a current state based on the historical data, matching the characteristic set with a pre-stored fault characteristic library, calculating the matching degree, and executing corresponding early warning or fault processing operation according to the matching degree. The method realizes on-line real-time continuous monitoring and automatic diagnosis of the running state of the shield screw machine, and reduces the dependence on complex hardware, high calculation force and stable network.

Inventors

LI BO
LAN HAO
DENG RONG

Assignees

湖南浩拓机电科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260408

Claims (10)

1. The shield spiral machine fault monitoring method is characterized by comprising the following steps of: The method comprises the steps of firstly, acquiring measurement data of a plurality of sensors installed on a shield screw machine, wherein the plurality of sensors comprise an eddy current sensor for measuring the abrasion loss and the offset of the shield screw machine blade, a temperature sensor for monitoring the working temperature of the shield screw machine and a rotating speed sensor for acquiring the rotating speed of the shield screw machine; step two, processing the measurement data to obtain characteristic parameters of the running state of the shield spiral machine, wherein the characteristic parameters comprise the abrasion loss of the blade, the offset of the central shaft of the blade, the offset direction angle of the central shaft of the blade, the temperature value and the rotation angular velocity; step three, comparing the characteristic parameters with preset parameter thresholds, and judging whether any characteristic parameter exceeds the corresponding parameter threshold; step four, when any characteristic parameter exceeds a corresponding parameter threshold, extracting historical measurement data in a time window corresponding to the characteristic parameter exceeding the parameter threshold; step five, calculating and extracting a feature set of the current state based on the historical measurement data in the time window; Step six, matching the feature set of the current state with a pre-stored fault feature library, and calculating the matching degree of each fault type in the fault feature library, wherein the fault feature library comprises feature templates corresponding to multiple fault types; and step seven, executing corresponding early warning or fault processing operation according to the matching degree.
2. The method for monitoring faults of a shield spiral machine according to claim 1, wherein in the second step, processing the measurement data includes: And filtering the real-time distance data acquired by the eddy current sensor to obtain an effective distance signal, wherein the filtering for removing the noise interference comprises the steps of repeatedly sampling at each sampling time point, removing abnormal values in the sampled data based on the characteristic that noise obeys Gaussian distribution, applying a triple standard deviation criterion, and applying a moving average filtering algorithm to the data after the abnormal values are removed.
3. The method of claim 1, wherein the historical measurement data in the time window includes all measurement data in a first predetermined time period before the occurrence of a characteristic parameter data point exceeding a parameter threshold and measurement data in a second predetermined time period after the occurrence of the data point.
4. The method for monitoring the faults of the shield screw machine according to claim 1, wherein in the fifth step, the feature set of the current state is calculated and extracted, and the feature set comprises the numerical value of each feature parameter and the variation trend of each feature parameter in the time window.
5. The method for monitoring faults of the shield screw machine according to claim 1, wherein in the step six, matching the feature set of the current state with a pre-stored fault feature library comprises: And in the similarity calculation process, different weights are distributed to the fault judgment importance according to different features, and the deviation degree of the feature values is evaluated by combining a preset allowable deviation range.
6. The method for monitoring faults of a shield screw machine according to claim 5, wherein in the step six, the method for calculating the matching degree comprises the step of comprehensively evaluating at least one of weighted euclidean distance similarity, cosine similarity and fuzzy membership function, and finally fusing to generate the matching degree in a percentage form.
7. The method for monitoring faults of a shield spiral machine according to claim 1, wherein the step seven comprises: if the matching degree is smaller than a first threshold value, judging that the matching degree is abnormal fluctuation, recording data and returning to the step one; if the matching degree is larger than or equal to a first threshold value and smaller than a second threshold value, triggering a first-stage early warning and improving the data monitoring frequency; If the matching degree is greater than or equal to a second threshold value, judging that a fault corresponding to the fault type with the highest matching degree occurs, and triggering a second-stage early warning comprising fault positioning information.
8. The method of claim 7, further comprising generating a fault report including the determined fault type and fault location after triggering the second level of pre-warning, and sending out an audible and visual alarm until the fault is handled or the equipment is shut down for maintenance.
9. The utility model provides a shield constructs screw machine fault monitoring device which characterized in that includes: The system comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is used for acquiring measurement data of a plurality of sensors installed on a shield screw machine, the plurality of sensors comprise an eddy current sensor used for measuring the abrasion amount and the offset amount of a shield screw machine blade, a temperature sensor used for monitoring the working temperature of the shield screw machine and a rotating speed sensor used for acquiring the rotating speed of the shield screw machine; the processing unit is used for processing the measurement data to obtain characteristic parameters of the running state of the shield spiral machine, wherein the characteristic parameters comprise the abrasion loss of the blade, the offset of the central shaft of the blade, the offset direction angle of the central shaft of the blade, the temperature value and the rotation angular velocity; the comparison unit is used for comparing the characteristic parameters with preset parameter thresholds and judging whether any characteristic parameter exceeds the corresponding parameter threshold; the extraction unit is used for extracting historical measurement data in a time window corresponding to the characteristic parameter exceeding the parameter threshold when any characteristic parameter exceeding the corresponding parameter threshold exists; A calculating unit, configured to calculate and extract a feature set of a current state based on the history measurement data in the time window; the matching unit is used for matching the feature set of the current state with a pre-stored fault feature library, calculating the matching degree of each fault type in the fault feature library, wherein the fault feature library comprises feature templates corresponding to multiple fault types; and the execution unit is used for executing corresponding early warning or fault processing operation according to the matching degree.
10. The shield screw machine fault monitoring system is characterized by comprising an eddy current sensor, a temperature sensor, a rotating speed sensor and a controller; The system comprises a shield spiral machine barrel, an eddy current sensor, a sensor module and a sensor module, wherein the eddy current sensor is used for measuring the abrasion loss and the offset of the shield spiral machine blade, and comprises a plurality of groups of eddy current sensors arranged on the shield spiral machine barrel, and each group of eddy current sensors is uniformly arranged at the corresponding positions of the two ends of the shield spiral machine blade; the temperature sensor is used for monitoring the working temperature of the shield screw machine; the rotating speed sensor is used for acquiring the rotating speed of the shield spiral machine; The controller is used for acquiring measurement data of a plurality of sensors installed on the shield spiral machine, processing the measurement data to obtain characteristic parameters of the running state of the shield spiral machine, wherein the characteristic parameters comprise blade abrasion amount, blade central axis deviation size, blade central axis deviation direction angle, temperature value and rotation angle speed, comparing the characteristic parameters with preset parameter thresholds to judge whether any characteristic parameter exceeds the corresponding parameter threshold, extracting historical measurement data in a time window corresponding to the characteristic parameter exceeding the parameter threshold when any characteristic parameter exceeds the corresponding parameter threshold, calculating and extracting a characteristic set of the current state based on the historical measurement data in the time window, matching the characteristic set of the current state with a pre-stored fault characteristic library, calculating the matching degree of each fault type in the fault characteristic library, wherein the fault characteristic library comprises the characteristic templates corresponding to multiple fault types, and executing corresponding early warning or fault processing operation according to the matching degree.

Description

Method, device and system for monitoring faults of shield spiral machine Technical Field The invention relates to the technical field of shield screw machine fault monitoring, in particular to a shield screw machine fault monitoring method, device and system. Background Shield tunneling machine/full face hard rock tunnel boring machine (Tunnel Boring Machine, TBM) is a common and important mechanical device in tunneling, one of its core components being a screw system. The main function of the screw machine system is to convey the dregs from the soil bin, when the dregs are conveyed, hard matters in the dregs and the dregs can be continuously rubbed and collided with the screw blades of the screw machine, so that the screw blades are worn after being used for a long time, gaps between the blades and the pipe body are larger and larger along with the wear, the risk of the central shaft deviation of the blades is increased due to the increase of the gaps, if the front end of the central shaft of the screw blades is slightly deviated, the geometrical amplification effect of rigid body deflection is caused, the offset of the rear end of the blades is obviously amplified, the rear end of the screw machine blades and the machine barrel are scratched, the vibration of the whole machine is abnormal, the abnormal wear of the blades, the machine barrel and the bearings is caused, and when the faults occur, the whole screw machine system is extremely paralyzed, even the whole shield machine system is caused, so that the faults of the screw machine system can be accurately found and processed in time, and the tunneling efficiency, the service life of the screw machine and the construction safety are directly influenced. Therefore, accurate and timely monitoring of the overall state of the screw machine in the tunneling process is of great significance in improving construction efficiency and avoiding equipment faults. At present, various actual operation means and related technology are applied to detection of the running state of a screw machine system in an industrial production scene, so that basic acquisition and preliminary detection of core running parameters such as working temperature of the screw machine, abrasion loss of blades, offset of central shafts of the blades and the like can be realized, for example, the abrasion loss of the blades is detected by a voltage signal output by a sensing unit arranged on a cylinder body in a nondestructive screw blade abrasion real-time monitoring system and detection method disclosed as CN120948602A, for example, the abrasion loss of the blades is detected by an algorithm in the detection method, for example, the cloud-edge-collaboration-based shield machine fault monitoring diagnosis method and system disclosed as CN117786392A is realized, and the core is compiled through cloud training molds, edge-end deployment diagnosis and cloud-edge data interaction iteration to realize the whole fault monitoring diagnosis of the shield machine. In the scheme, the CN120948602A has no standard system, and the technical and maintenance complexity required by the CN117786392A is high, the required hardware and calculation cost is low, and because the edge end model in the scheme periodically synchronizes the cloud version by the network, the model update is delayed when the underground network is poor, the model cannot iterate due to network interruption, and the diagnosis accuracy is influenced for a long time. Disclosure of Invention The invention aims to provide a method, a device and a system for monitoring faults of a shield screw machine, which are used for solving the problem that the existing screw machine state monitoring technology lacks a complete processing flow from the discovery of faults of a screw machine system, the judgment of the faults to the solution of the faults and the follow-up recheck optimization. In a first aspect, the invention provides a method for monitoring faults of a shield spiral machine, which comprises the following steps: The method comprises the steps of firstly, obtaining measurement data of a plurality of sensors installed on a shield screw machine, wherein the plurality of sensors comprise eddy current sensors used for measuring abrasion loss and offset of the shield screw machine blades, temperature sensors used for monitoring working temperature of the shield screw machine and rotating speed sensors used for obtaining rotating speed of the shield screw machine, each eddy current sensor comprises a plurality of groups of eddy current sensors installed on a shield screw machine barrel, and each group of eddy current sensors are uniformly arranged at corresponding positions at two ends of the shield screw machine blades. And step two, processing the measurement data to obtain characteristic parameters of the running state of the shield spiral machine, wherein the characteristic parameters comprise the abrasion loss of the blade, the offset of t