Mine hoist head sheave group fault diagnosis method and system

Abstract

The invention discloses a fault diagnosis method and system for a head sheave group of a mine hoist, and aims to solve the problems that traditional lubrication is not uniform and fault diagnosis depends on a single parameter. According to the method, data such as temperature and pressure are collected through multiple sensors, the bearing operation temperature difference serves as a core index to weaken environment interference, a high-dimensional feature set is formed by combining vibration signal time-frequency domain features, after PSO-LPP dimensionality reduction, an SVDD model is input to achieve anomaly recognition, and a self-adaptive re-lubrication strategy is triggered to adjust grease injection parameters when anomaly occurs. The system comprises a monitoring module, a control module, a grease injection execution module, a diagnosis module and a remote monitoring module, and a closed-loop lubrication loop and a remote management system are constructed. The method improves the diagnosis sensitivity and abnormity recognition precision, optimizes the lubrication control effect, reduces the manual maintenance cost and safety risk, remarkably improves the operation stability and reliability of the head sheave group, and is suitable for the field of fault diagnosis of mine hoisting equipment.

Description

Technical Field

[0001] This invention relates to the field of fault diagnosis technology for mine hoisting equipment, specifically to a fault diagnosis method and system for mine hoist sheave groups. Background Technology

[0002] As a key load-bearing and guiding component, the operating status of the sheave assembly of a mine hoist directly affects the safety and stability of the hoisting system.

[0003] Traditional lubrication methods rely on manual operation, which can lead to uneven lubrication and poor timeliness, resulting in poor bearing lubrication and premature failure.

[0004] Existing fault diagnosis methods are mostly based on single parameter monitoring, lacking systematic and intelligent diagnostic means in complex mining environments, making it difficult to achieve early warning and precise control. Summary of the Invention

[0005] The purpose of this section is to outline some aspects of the embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0006] In view of the problems existing in the above and / or the fault diagnosis methods and systems for mine hoist sheave groups, this invention is proposed.

[0007] Therefore, the purpose of this invention is to provide a fault diagnosis method and system for mine hoist sheave groups, which effectively weakens environmental interference and improves diagnostic sensitivity by utilizing temperature difference characteristics; combines PSO-LPP and SVDD models to achieve high-precision anomaly identification in a low-dimensional feature space; realizes closed-loop control and adaptive optimization of lubrication, significantly improving the operational stability and reliability of the sheave group; and the system has remote monitoring and automatic alarm functions, reducing manual maintenance costs and safety risks.

[0008] To address the aforementioned technical problems, according to one aspect of the present invention, the present invention provides the following technical solution:

[0009] A method and system for diagnosing faults in mine hoist sheave units, comprising:

[0010] S1, through a multi-sensor system deployed in the bearings and lubrication pipelines of the sheave assembly, collects data on temperature, pressure, vibration and lubricant volume in real time;

[0011] S2, Based on the temperature data, calculate the bearing operating temperature difference, and use this temperature difference as the core state indicator to reduce the interference of ambient temperature.

[0012] S3, perform wavelet packet decomposition on the vibration signal, extract time-frequency domain features, and combine them with the temperature difference index to form a high-dimensional feature set;

[0013] S4. The high-dimensional feature set is reduced in dimensionality using a local preservation projection method optimized by particle swarm optimization, with the objective function being to minimize the information entropy difference, in order to obtain the optimal low-dimensional features.

[0014] S5, input the low-dimensional features into the pre-trained support vector data description model, and identify abnormal states by calculating the distance between the sample and the center of the hypersphere;

[0015] S6, when an abnormal state is identified, triggers control linkage, and adjusts the grease injection cycle and grease injection amount according to the adaptive relubrication strategy.

[0016] As a preferred embodiment of the mine hoist sheave group fault diagnosis method and system of the present invention, the adaptive relubrication strategy includes:

[0017] Based on formula Calculate the reference relubrication cycle, where Tz is the relubrication cycle, n is the sheave speed, d is the bearing inner diameter, and K is the correction coefficient;

[0018] Based on formula Calculate the amount of grease added in a single operation, where D is the outer diameter of the bearing and B is the width of the bearing;

[0019] The correction coefficient K is dynamically adjusted based on the effective operating time of the equipment and the temperature state during the bearing rest period.

[0020] Implement "small amount, multiple times" lubrication control, and divide the calculated grease replenishment amount into m times for uniform injection.

[0021] As a preferred embodiment of the fault diagnosis method and system for mine hoist sheave groups described in this invention, the training of the support vector data description model uses only sample data under normal operating conditions to construct a minimum hypersphere that surrounds normal samples in the feature space; during the identification process, samples falling outside the hypersphere are judged as abnormal.

[0022] The mine hoist sheave assembly fault diagnosis system includes:

[0023] Monitoring module: includes a temperature sensor, a pipeline pressure sensor, a vibration sensor, a liquid volume metering sensor installed in the bearing housing, and an end oil flow signal sensor for detecting the working status of the oil separator;

[0024] Control module: Based on a programmable logic controller, it coordinates the actions of the pump station, oil distributor and reversing valve through an intermediate controller, and integrates an RS485 communication interface.

[0025] The grease injection module includes a lubrication pump station, a quantitative oil distributor with multiple outlets, a reversing valve, and a grease pump for waste grease recovery, forming a closed-loop lubrication circuit of "single pump, multiple points, and quantitative differentiated grease supply".

[0026] Diagnostic module: Embedded in the monitoring host or cloud server, configured to perform feature extraction, local preserving projection (PSO-LPP) dimensionality reduction and support vector data description (SVDD) anomaly detection algorithms;

[0027] Remote monitoring and management module: including industrial computer or configuration software platform, used for data visualization, historical data query, remote parameter configuration, fault alarm and maintenance report generation.

[0028] As a preferred embodiment of the mine hoist sheave group fault diagnosis system of the present invention, the system connects the monitoring module, the control module and the remote monitoring and management module through an industrial network bridge to realize remote data transmission and control command issuance.

[0029] As a preferred embodiment of the mine hoist sheave group fault diagnosis system of the present invention, the liquid volumetric metering sensor is used to monitor and provide feedback on the amount of lubricant injected into each lubrication point in real time, so as to ensure precise control of the amount of grease injected.

[0030] Compared with the prior art, the beneficial effects of this invention are as follows: This method and system for diagnosing faults in mine hoist sheave groups effectively weakens environmental interference and improves diagnostic sensitivity by utilizing temperature difference characteristics; it combines PSO-LPP and SVDD models to achieve high-precision anomaly identification in a low-dimensional feature space; it realizes closed-loop control and adaptive optimization of lubrication, significantly improving the operational stability and reliability of the sheave group; and the system has remote monitoring and automatic alarm functions, reducing manual maintenance costs and safety risks. Attached Figure Description

[0031] To more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and detailed embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0032] Figure 1 A comparison chart showing the temperature difference between well-lubricated and long-term unlubricated conditions;

[0033] Figure 2 The image shows the PSO-LPP-SVDD classification results under three-dimensional features;

[0034] Figure 3 To maintain the comparison chart of the temperature difference before and after. Detailed Implementation

[0035] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0036] Secondly, the present invention is described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure will be partially enlarged and not at the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth.

[0037] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0038] This invention provides a method and system for diagnosing faults in the sheave assembly of a mine hoist. It effectively reduces environmental interference and improves diagnostic sensitivity by utilizing temperature difference characteristics. By combining PSO-LPP and SVDD models, it achieves high-precision anomaly identification in a low-dimensional feature space. It realizes closed-loop control and adaptive optimization of lubrication, significantly improving the operational stability and reliability of the sheave assembly. The system has remote monitoring and automatic alarm functions, reducing manual maintenance costs and safety risks.

[0039] This invention provides a method for diagnosing faults in the sheave assembly of a mine hoist, comprising the following steps:

[0040] S1, through a multi-sensor system deployed in the bearings and lubrication pipelines of the sheave assembly, collects data on temperature, pressure, vibration and lubricant volume in real time;

[0041] S2, Based on the temperature data, calculate the bearing operating temperature difference, and use this temperature difference as the core state indicator to reduce the interference of ambient temperature.

[0042] S3, perform wavelet packet decomposition on the vibration signal, extract time-frequency domain features, and combine them with the temperature difference index to form a high-dimensional feature set;

[0043] S4. The high-dimensional feature set is reduced in dimensionality using a local preservation projection method optimized by particle swarm optimization, with the objective function being to minimize the information entropy difference, in order to obtain the optimal low-dimensional features.

[0044] S5, input the low-dimensional features into the pre-trained support vector data description model, and identify abnormal states by calculating the distance between the sample and the center of the hypersphere;

[0045] S6, when an abnormal state is identified, triggers control linkage, and adjusts the grease injection cycle and grease injection amount according to the adaptive relubrication strategy.

[0046] The adaptive relubrication strategy includes:

[0047] Based on formula Calculate the reference relubrication cycle, where Tz is the relubrication cycle, n is the sheave speed, d is the bearing inner diameter, and K is the correction coefficient;

[0048] Based on formula Calculate the amount of grease added in a single operation, where D is the outer diameter of the bearing and B is the width of the bearing;

[0049] The correction coefficient K is dynamically adjusted based on the effective operating time of the equipment and the temperature state during the bearing rest period.

[0050] Implement "small amount, multiple times" lubrication control, and divide the calculated grease replenishment amount into m times for uniform injection.

[0051] The training of the support vector data description model uses only sample data under normal operating conditions to construct a minimum hypersphere that surrounds normal samples in the feature space; during the recognition process, samples falling outside the hypersphere are judged as abnormal.

[0052] The mine hoist sheave assembly fault diagnosis system includes:

[0053] Monitoring module: includes a temperature sensor, a pipeline pressure sensor, a vibration sensor, a liquid volume metering sensor installed in the bearing housing, and an end oil flow signal sensor for detecting the working status of the oil separator;

[0054] Control module: Based on a programmable logic controller, it coordinates the actions of the pump station, oil distributor and reversing valve through an intermediate controller, and integrates an RS485 communication interface.

[0055] The grease injection module includes a lubrication pump station, a quantitative oil distributor with multiple outlets, a reversing valve, and a grease pump for waste grease recovery, forming a closed-loop lubrication circuit of "single pump, multiple points, and quantitative differentiated grease supply".

[0056] Diagnostic module: Embedded in the monitoring host or cloud server, configured to perform feature extraction, PSO-LPP dimensionality reduction and SVDD anomaly identification algorithm. In the PSO-LPP dimensionality reduction method, the optimal number of low-dimensional features is determined to be 3.

[0057] Remote monitoring and management module: including industrial computer or configuration software platform, used for data visualization, historical data query, remote parameter configuration, fault alarm and maintenance report generation.

[0058] The system connects the monitoring module, control module and remote monitoring and management module through an industrial network bridge to realize remote data transmission and control command issuance.

[0059] The liquid volumetric metering sensor is used to monitor and provide feedback on the amount of lubricant injected into each lubrication point in real time, ensuring precise control of the grease injection amount.

[0060] System hardware deployment and integration

[0061] This invention was implemented on a JKMD-5×4Ⅲ type multi-rope friction hoisting system in the auxiliary shaft of a certain mine. The sheave assembly of this system is distributed on four platforms ranging from 23m to 41m on the headframe.

[0062] Pump station and control cabinet installation: The core pump station and PLC control cabinet of the intelligent lubrication system are installed in the equipment room at the bottom of the derrick for easy power supply and maintenance.

[0063] Pipeline installation: 32×26mm stainless steel pipes are used as the main oil passages, fixedly laid along the derrick guardrail, leading to each level of the sheave platform. Protective sleeves are installed on the outside of the pipelines to prevent corrosion and physical damage from the harsh downhole environment.

[0064] Sensor and actuator installation:

[0065] Temperature sensor: A PT100 platinum resistance temperature sensor is used, which is directly installed in the outer ring seat hole of the sheave bearing to directly measure the bearing's operating temperature.

[0066] Vibration sensor: An industrial-grade accelerometer vibration sensor is used, which is fixed to the bearing housing by a magnetic base to collect vibration signals.

[0067] Pressure sensor: Installed at the inlet of the main oil circuit and each layer of oil distributor to monitor the system oil supply pressure.

[0068] Liquid volumetric metering sensor: Built into each outlet branch of the oil distributor, used to accurately measure the volume of grease injected into each bearing.

[0069] End-of-line oil supply signal sensor: Installed at the end of each oil distributor, it is used to confirm whether the grease has been delivered to the end of the pipeline, serving as the basis for determining whether the cycle is complete.

[0070] Oil distributor and reversing valve: According to the layout of the sheave, they are reasonably set on each platform to achieve independent and quantitative grease supply from one pump to multiple points.

[0071] Test protocol and evaluation indicators

[0072] To verify the diagnostic method and lubrication control strategy proposed in this paper, comparative tests under different operating conditions were designed:

[0073] Comparing well-lubricated conditions with long-term unlubricated conditions—verifying the distinguishing ability of temperature difference as a sensitive indicator;

[0074] Before and after maintenance operation comparison – used to evaluate the improvement effect of intelligent lubrication system on temperature difference stability and operational reliability;

[0075] Anomaly detection comparison under different dimensionality reductions—used to examine the diagnostic performance of PSO-LPP and SVDD models on real samples.

[0076] The data collected during the experiment included: pump station oil level, pipeline pressure and temperature, sheave bearing temperature and vibration signal, end-point oil supply signal, and alarm records. Temperature fluctuation amplitude, anomaly identification accuracy, and fault alarm results were used as core evaluation indicators.

[0077] Results and Analysis

[0078] Distinguishing ability of temperature difference index

[0079] Comparative analysis of two operating conditions—one with good lubrication and the other without lubrication for an extended period—revealed that temperature difference, as a diagnostic indicator, clearly reflects differences in lubrication status. Under good lubrication, the maximum temperature difference of the sheave bearing was approximately 0.8℃, stabilizing after about 1.5 hours of operation with a mean close to zero. However, under conditions of prolonged lack of lubrication, the maximum temperature difference reached 1.9℃, and even after stabilization, it remained significantly higher than normal. This indicates that insufficient lubrication significantly increases frictional heat, keeping the temperature difference at a higher level. Therefore, the temperature difference indicator effectively mitigates the interference of ambient temperature and possesses strong capability in differentiating operating conditions.

[0080] Model recognition performance analysis

[0081] After feature extraction via wavelet packet decomposition, this paper employs the PSO-LPP method for feature dimensionality reduction and combines it with the SVDD model for anomaly identification. A total of 50 samples were collected in the experiment, with the first 10 being normal samples and the remaining 40 being faulty samples. The results show that the highest identification accuracy (82%) is achieved when the dimensionality is reduced to three dimensions; while the accuracy is approximately 74%–76% for five, seven, and nine dimensions, respectively. Classification results show that most faulty samples are clearly separated from normal samples in the feature space, with only a small number of samples falling near the boundary. This indicates that PSO-LPP optimization effectively maintains the discriminative power of the features, and SVDD possesses strong anomaly identification capabilities under single-class sample conditions.

[0082] Before the intelligent lubrication system was implemented, the temperature fluctuation of the sheave bearing was significant, with obvious abnormal peaks during operation. After adopting intelligent lubrication control, the temperature fluctuation amplitude was significantly reduced, and the system remained stable during operation without any sudden increases. The comparative results show that intelligent lubrication not only improves lubrication uniformity but also significantly reduces the risk of frictional heat accumulation, thereby enhancing the reliability and stability of the sheave assembly. The system's remote monitoring and automatic alarm functions effectively reduce manual maintenance workload, providing a reliable guarantee for the safe operation of the mine hoist.

[0083] To address the problem of poor lubrication in the sheave assembly of mine hoists, a fault diagnosis technology path of "lubrication status monitoring - feature extraction - intelligent identification - control linkage" was constructed and verified in an actual auxiliary shaft hoisting system.

[0084] The study yielded the following conclusions:

[0085] By comparing two operating conditions—one with good lubrication and the other without lubrication for a long time—it was verified that temperature difference can significantly reduce the interference of ambient temperature and intuitively reflect the difference in lubrication status, providing a reliable basis for early fault identification of sheave assembly.

[0086] The intelligent model provides accurate diagnosis: the diagnostic method based on wavelet packet decomposition, PSO-LPP dimensionality reduction and SVDD has the highest recognition accuracy of about 82% when the feature dimension is three, which can effectively distinguish between abnormal and normal operating conditions.

[0087] The intelligent lubrication system achieves "single pump, multiple points, and quantitatively differentiated" grease supply and waste grease recovery. The comparison before and after maintenance shows that the temperature difference fluctuation is significantly reduced after lubrication, and the operational stability is significantly improved.

[0088] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the disclosed embodiments can be combined with each other in any manner. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for diagnosing faults in the sheave assembly of a mine hoist, characterized in that, Includes the following steps: S1, through a multi-sensor system deployed in the bearings and lubrication pipelines of the sheave assembly, collects data on temperature, pressure, vibration and lubricant volume in real time; S2, Based on the temperature data, calculate the bearing operating temperature difference, and use this temperature difference as the core state indicator to reduce the interference of ambient temperature. S3, perform wavelet packet decomposition on the vibration signal, extract time-frequency domain features, and combine them with the temperature difference index to form a high-dimensional feature set; S4. The high-dimensional feature set is reduced in dimensionality using a local preservation projection method optimized by particle swarm optimization, with the objective function being to minimize the information entropy difference, in order to obtain the optimal low-dimensional features. S5, input the low-dimensional features into the pre-trained support vector data description model, and identify abnormal states by calculating the distance between the sample and the center of the hypersphere; S6, when an abnormal state is identified, triggers control linkage, and adjusts the grease injection cycle and grease injection amount according to the adaptive relubrication strategy.

2. The method for diagnosing faults in the sheave assembly of a mine hoist according to claim 1, characterized in that, The adaptive relubrication strategy includes: Based on formula Calculate the reference relubrication cycle, where Tz is the relubrication cycle, n is the sheave speed, d is the bearing inner diameter, and K is the correction coefficient; Based on formula Calculate the amount of grease added in a single operation, where D is the outer diameter of the bearing and B is the width of the bearing; The correction coefficient K is dynamically adjusted based on the effective operating time of the equipment and the temperature state during the bearing rest period. Perform lubrication control by applying small amounts of lubricant multiple times, dividing the calculated grease replenishment amount into m uniform injections.

3. The method for diagnosing faults in the sheave assembly of a mine hoist according to claim 1, characterized in that, The training of the support vector data description model uses only sample data under normal operating conditions to construct a minimum hypersphere that surrounds normal samples in the feature space; during the recognition process, samples falling outside the hypersphere are judged as abnormal.

4. A mine hoist sheave assembly fault diagnosis system, used to implement the mine hoist sheave assembly fault diagnosis method as described in any one of claims 1-3, characterized in that, include: Monitoring module: includes a temperature sensor, a pipeline pressure sensor, a vibration sensor, a liquid volume metering sensor installed in the bearing housing, and an end oil flow signal sensor for detecting the working status of the oil separator; Control module: Based on a programmable logic controller, it coordinates the actions of the pump station, oil distributor and reversing valve through an intermediate controller, and integrates an RS485 communication interface. The grease injection module includes a lubrication pump station, a quantitative oil distributor with multiple outlets, a reversing valve, and a grease pump for waste grease recovery, forming a closed-loop lubrication circuit of "single pump, multiple points, and quantitative differentiated grease supply". Diagnostic module: Embedded in the monitoring host or cloud server, configured to perform feature extraction, local preserving projection dimensionality reduction and support vector data description anomaly identification algorithms; Remote monitoring and management module: including industrial computer or configuration software platform, used for data visualization, historical data query, remote parameter configuration, fault alarm and maintenance report generation.

5. The mine hoist sheave assembly fault diagnosis system according to claim 4, characterized in that, The system connects the monitoring module, control module and remote monitoring and management module through an industrial network bridge to realize remote data transmission and control command issuance.

6. The mine hoist sheave assembly fault diagnosis system according to claim 4, characterized in that, The liquid volumetric metering sensor is used to monitor and provide feedback on the amount of lubricant injected into each lubrication point in real time, ensuring precise control of the grease injection amount.

Images