Fault online diagnosis method of scraper conveyer

Abstract

The invention relates to a fault online diagnosis method for a scraper conveyer, and belongs to the technical field of scraper conveyer operation state monitoring. Comprising the steps that operation state initial data collected by various types of sensors arranged on the scraper conveyor in the operation process are obtained, the various types of sensors comprise vibration sensors, pressure sensors and ammeters, and the operation state initial data comprise initial vibration signals, initial pressure signals and initial current signals; preprocessing the initial data of the running state to obtain target data of the running state; and inputting the operation state target data into a pre-trained abnormal working condition identification model of the scraper conveyer, and identifying the fault type of the scraper conveyer according to an output result of the abnormal working condition identification model of the scraper conveyer. According to the invention, faults can be found in time, so that serious accidents are avoided, and the fault detection efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring the operating state of scraper conveyors, and particularly to an online fault diagnosis method for scraper conveyors. Background Art

[0002] Scraper conveyors are the main transportation devices in underground coal mines. During the daily operation of scraper conveyors, due to the complex and harsh working conditions in the mine environment, the scraper conveyor often malfunctions during operation due to reasons such as incomplete discharging, material impact, and large debris jamming. For example, when the scraper conveyor fails to discharge completely, large pieces of materials or debris enter the sprocket support nests along with the chain, causing the chain to derail or the chain to shift; another example is that large foreign objects and excessive wear will cause the chain or scraper to jam, resulting in excessive stretching of the chain or deformation of the scraper. When the scraper conveyor malfunctions, it cannot complete the material transportation task normally. If the scraper conveyor continues to operate, it will cause unnecessary consumption and may further expand the failure of the scraper conveyor. Therefore, during the operation of the scraper conveyor, it is necessary to promptly detect its abnormal operating conditions.

[0003] At present, when the scraper conveyor is running, on-site personnel are mostly used to conduct periodic inspections on its operating state to avoid the occurrence of the above-mentioned faults. However, due to the occasional nature of the faults of the scraper conveyor, manual inspections cannot detect the faults in a timely manner, which will lead to relatively serious accidents, and the efficiency of manual inspections is low. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides an online fault diagnosis method for scraper conveyors. The technical solution of the present invention is as follows:

[0005] An online fault diagnosis method for a scraper conveyor, which includes:

[0006] S1, obtaining initial operating state data collected by various types of sensors configured on the scraper conveyor during its operation. The various types of sensors include vibration sensors, pressure sensors, and ammeters, and the initial operating state data includes initial vibration signals, initial pressure signals, and initial current signals;

[0007] S2, preprocessing the initial operating state data to obtain target operating state data;

[0008] S3, inputting the target operating state data into a pre-trained abnormal condition recognition model for the scraper conveyor, and identifying the fault type of the scraper conveyor according to the output result of the abnormal condition recognition model for the scraper conveyor.

[0009] Optionally, the vibration sensor is installed on the reducer, gear oil pump, planetary gear reducer and bearing seat of the scraper conveyor; the pressure sensor is installed at the oil inlet of the hydraulic motor of the scraper conveyor; and the ammeter is installed on the three-phase asynchronous motor of the scraper conveyor.

[0010] Optionally, when preprocessing the initial operation state data in S2, it includes:

[0011] S21, performing de-noising and filtering processing on the initial vibration signal in the initial operation state data to obtain a target vibration signal.

[0012] S22, performing outlier processing on the initial pressure signal and initial current signal in the initial operation state data to obtain a target pressure signal and a target current signal.

[0013] Optionally, the abnormal working condition identification model of the scraper conveyor is a convolutional neural network model.

[0014] Optionally, when S3 identifies the fault type of the scraper conveyor according to the output result of the abnormal working condition identification model of the scraper conveyor, it determines that the fault type corresponding to the maximum output probability is the current fault type of the scraper conveyor.

[0015] Optionally, after S3 identifies the fault type of the scraper conveyor according to the output result of the abnormal working condition identification model of the scraper conveyor, it further includes:

[0016] Displaying the fault type and giving an alarm.

[0017] Optionally, after S3 identifies the fault type of the scraper conveyor according to the output result of the abnormal working condition identification model of the scraper conveyor, it further includes:

[0018] Recommending repair or maintenance suggestions for the scraper conveyor according to the fault type.

[0019] All of the above optional technical solutions can be arbitrarily combined, and the present invention does not elaborate on the structures after combination one by one.

[0020] By means of the above solution, the beneficial effects of the present invention are as follows:

[0021] By obtaining the initial operation status data collected by various types of sensors on the scraper conveyor, setting various types of sensors including vibration sensors, pressure sensors and ammeters, and determining the fault type of the scraper conveyor according to the initial operation status data and the pre-trained abnormal working condition identification model of the scraper conveyor, a method capable of timely determining the fault type of the scraper conveyor is provided. By performing fault diagnosis on the scraper conveyor through this method, not only can faults be detected in a timely manner, thereby avoiding more serious accidents, but also the fault detection efficiency can be improved.

[0022] The above description is only an overview of the technical solution of the present invention. In order to be able to understand the technical means of the present invention more clearly and implement it in accordance with the content of the description, the following describes in detail with reference to the preferred embodiments of the present invention and the accompanying drawings. Brief Description of the Drawings

[0023] Figure 1 is a flowchart of the on-line fault diagnosis method for the scraper conveyor provided by the present invention. Detailed Description of the Preferred Embodiments

[0024] The following further describes in detail the specific embodiments of the present invention with reference to the accompanying drawings and embodiments. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0025] As Figure 1 shown, the on-line fault diagnosis method for the scraper conveyor provided by the embodiment of the present invention can be implemented by an on-board processing terminal configured on the scraper conveyor. The method includes the following steps:

[0026] S1. Obtain the initial operation status data collected by various types of sensors configured on the scraper conveyor during operation. The various types of sensors include vibration sensors, pressure sensors and ammeters. The initial operation status data includes an initial vibration signal, an initial pressure signal and an initial current signal.

[0027] Specifically, the vibration sensor is installed on the reducer, gear oil pump, planetary gear reducer and bearing seat of the scraper conveyor; the pressure sensor is installed at the oil inlet of the hydraulic motor of the scraper conveyor; the ammeter is installed on the three-phase asynchronous motor of the scraper conveyor.

[0028] By installing these sensors at the above positions of the scraper conveyor, the operation data of these positions can be collected in real time, and then the faults at these positions of the scraper conveyor can be detected in a timely manner through subsequent steps.

[0029] S2. Preprocess the initial operation status data to obtain the target operation status data.

[0030] Optionally, when preprocessing the initial operation status data, S2 includes:

[0031] S21, perform noise reduction filtering on the initial vibration signal in the initial operation status data to obtain a target vibration signal.

[0032] Specifically, when performing noise reduction filtering, it can be implemented through relevant filtering algorithms. Through noise reduction filtering, the initial operation status data that obviously does not have research value can be removed, thereby reducing the data volume of the initial operation status data, reducing invalid calculations, and improving calculation efficiency.

[0033] S22, perform outlier processing on the initial pressure signal and the initial current signal in the initial operation status data to obtain a target pressure signal and a target current signal.

[0034] Specifically, data anomaly means that a wrong value is transmitted by the sensor, generally an outlier, which is common in the damage of internal components of the sensor. When performing outlier processing, in order to ensure the integrity of the data, the embodiment of the present invention adopts the method of filling with adjacent time-domain values. For example, if the initial pressure signal is missing, the pressure value at the previous moment is used for filling.

[0035] S3, input the target operation status data into a pre-trained abnormal working condition recognition model of the scraper conveyor, and identify the fault type of the scraper conveyor according to the output result of the abnormal working condition recognition model of the scraper conveyor.

[0036] Among them, the abnormal working condition recognition model of the scraper conveyor is a classification model. Preferably, the abnormal working condition recognition model of the scraper conveyor is a convolutional neural network model. On this basis, when S3 identifies the fault type of the scraper conveyor according to the output result of the abnormal working condition recognition model of the scraper conveyor, it determines the fault type corresponding to the maximum output probability of the convolutional neural network model as the current fault type of the scraper conveyor.

[0037] Specifically, the output result of the convolutional neural network model is the probability of various typical faults of the scraper conveyor, and the sum of these probability values is 1. When the probability value of a certain typical fault is the largest, the embodiment of the present invention determines the fault type as the current fault type of the scraper conveyor. For example, if the probability value of gear fault is the largest in the output result of the convolutional neural network model, it is determined that the gear fault is the current fault type of the scraper conveyor. It should be noted that the output result of the convolutional neural network model also includes the classification of no fault. When the probability of no fault is the largest in the output result of the convolutional neural network model, it is determined that the current operation status of the scraper conveyor is normal.

[0038] Further, before inputting the target data of the operating state into the pre-trained abnormal condition recognition model of the scraper conveyor, S3 needs to train the abnormal condition recognition model of the scraper conveyor first. The abnormal condition recognition model of the scraper conveyor is trained by the data collected by the above various sensors during the historical operation of the scraper conveyor. Specifically, the training data used during training includes both the data collected by various types of sensors during the normal operation of the scraper conveyor and the data collected by the above various types of sensors when various faults occur in the operation of the scraper conveyor.

[0039] Optionally, after S3 identifies the fault type of the scraper conveyor according to the output result of the abnormal condition recognition model of the scraper conveyor, it further includes: displaying the fault type and giving an alarm to remind the staff to handle the fault in time.

[0040] Further, after S3 identifies the fault type of the scraper conveyor according to the output result of the abnormal condition recognition model of the scraper conveyor, it further includes: recommending maintenance or repair suggestions for the scraper conveyor according to the fault type.

[0041] Among them, the maintenance or repair suggestions include regular maintenance or condition-based maintenance of the scraper conveyor, and the condition-based maintenance includes manual inspection, routine maintenance, and fault repair. Routine maintenance includes maintaining the star wheel reducer, increasing the lubrication of the hydraulic motor, and strengthening the foundation rigidity. Fault repair specifically includes repairing or replacing bearings, repairing or replacing gears, and repairing the reduction box.

[0042] The above is only the preferred embodiment of the present invention and is not used to limit the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A method for online fault diagnosis of a scraper conveyor, characterized in that: include: S1, obtaining initial operating status data collected by various types of sensors configured on the scraper conveyor during its operation, wherein the various types of sensors include a vibration sensor, a pressure sensor, and an ammeter, and the initial operating status data include an initial vibration signal, an initial pressure signal, and an initial current signal; S2, preprocessing the initial running status data to obtain target running status data; S3, inputting the operating status target data into a pre-trained abnormal operating condition identification model for a scraper conveyor, and identifying the fault type of the scraper conveyor according to an output result of the abnormal operating condition identification model for the scraper conveyor.

2. The method for online fault diagnosis of a scraper conveyor according to claim 1, characterized in that: The vibration sensor is installed on the reducer, gear oil pump, planetary reducer and bearing seat of the scraper conveyor; the pressure sensor is installed at the oil inlet of the hydraulic motor of the scraper conveyor; the ammeter is installed on the three-phase asynchronous motor of the scraper conveyor.

3. The method for online fault diagnosis of a scraper conveyor according to claim 1 or 2, characterized in that: When the S2 pre-processes the initial data of the operating state, it includes: S21, performing denoising and filtering processing on the initial vibration signal in the initial data of the running state to obtain a target vibration signal. S22, performing abnormal value processing on the initial pressure signal and the initial current signal in the initial data of the operating state to obtain a target pressure signal and a target current signal.

4. The method for online fault diagnosis of a scraper conveyor according to claim 1, characterized in that: The abnormal operating condition recognition model of the scraper conveyor is a convolutional neural network model.

5. The method for online fault diagnosis of a scraper conveyor according to claim 4, characterized in that: When identifying the fault type of the scraper conveyor according to the output result of the abnormal operating condition identification model of the scraper conveyor, S3 determines the fault type corresponding to the maximum output probability as the current fault type of the scraper conveyor.

6. The method for online fault diagnosis of a scraper conveyor according to claim 1, characterized in that: After identifying the fault type of the scraper conveyor according to the output result of the abnormal operating condition identification model of the scraper conveyor, S3 further includes: Display the fault type and issue an alarm.

7. The method for online fault diagnosis of a scraper conveyor according to claim 1, characterized in that: After identifying the fault type of the scraper conveyor according to the output result of the abnormal operating condition identification model of the scraper conveyor, S3 further includes: Recommendations for repair or maintenance of the scraper conveyor are given based on the type of fault described.

Images