A method and system for detecting an abnormality of an eccentric shaft of a jaw crusher

Abstract

The application belongs to the technical field of image processing, and particularly relates to a jaw crusher eccentric shaft abnormality detection method and system, which comprises the following steps: S1, an original gray-scale image of a jaw crusher eccentric shaft end face is acquired, the center coordinates of the eccentric shaft end face are determined through circular contour positioning, and the original gray-scale image is converted into a polar coordinate system image with the center coordinates as the origin; S2, the radial gradient and the tangential gradient of each pixel point in the analysis window on the polar coordinate system image are evaluated and analyzed, and the edge response stability index is determined according to the distribution relationship between the radial gradient and the tangential gradient; and S3, the global motion vector of the computer rack background feature points and the local motion vector of the eccentric shaft end face feature region are calculated. The application effectively offsets strong background vibration interference, can identify the trace slip in the budding stage, and significantly improves the safety and monitoring accuracy of equipment operation.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology, specifically relating to a method and system for detecting abnormalities in the eccentric shaft of a jaw crusher. Background Technology

[0002] Jaw crushers are core equipment in mining crushing production lines, and their power output is highly dependent on the stable operation of the eccentric shaft assembly. When crushing high-strength materials such as hard rock and basalt, the eccentric shaft drives the moving jaw plate to operate at high frequency and with a large stroke. The key fasteners such as the end flywheel and the withdrawal sleeve need to withstand huge alternating loads and impact stresses. Even a slight movement or slippage can lead to wear on the mating surfaces, and in severe cases, it can induce accidents such as shaft breakage, causing downtime and threatening personnel safety.

[0003] However, the extremely harsh operating environment of crushers presents two major technological challenges to the condition monitoring of eccentric shaft assemblies. First, the extremely high concentration of dust at the work site obscures the image. Dust particles easily adhere to the surface of the monitoring lens and, while diffused in the air, interfere with imaging, significantly reducing the image quality acquired by conventional visual monitoring equipment. This results in blurred edge features of the assemblies and difficulty in capturing subtle loosening displacement signals. Second, the intense vibrations generated during rock crushing cause global high-frequency oscillations that superimpose and interfere with localized loosening displacement signals of the assemblies. This makes it difficult for various sensors to effectively distinguish between normal operating vibrations and abnormal mechanical loosening, hindering the accurate extraction of fault characteristics.

[0004] In existing technologies, traditional methods relying on manual inspections and shutdown measurements suffer from significant delays, failing to achieve dynamic real-time monitoring during equipment operation. Problems are often only discovered after they manifest, missing the optimal maintenance window. Furthermore, traditional contact displacement sensors have extremely short lifespans under harsh environments of strong impact and high-frequency vibration, and are difficult to adapt to high-speed rotating shaft structures, making long-term stable operation impossible. Moreover, existing image monitoring algorithms lack effective anti-interference mechanisms when handling such complex scenarios of strong vibration, often misinterpreting overall machine sway as component loosening, resulting in a very high false alarm rate. This fundamentally fails to meet the actual needs of mining enterprises for precise operation and maintenance, cost reduction, and efficiency improvement. Summary of the Invention

[0005] This invention provides a method and system for detecting abnormalities in the eccentric shaft of a jaw crusher, in order to solve the technical problem in the prior art that it is difficult to accurately identify loose fasteners in crushers under strong vibration and high dust environments.

[0006] In a first aspect, the present invention provides a method for detecting abnormalities in the eccentric shaft of a jaw crusher, comprising the following steps:

[0007] S1. Obtain the original grayscale image of the eccentric shaft end face of the jaw crusher, determine the center coordinates of the eccentric shaft end face through circular contour positioning, and convert the original grayscale image into a polar coordinate system image with the center coordinates as the origin.

[0008] S2, evaluate and analyze the radial and tangential gradients of each pixel within the window on the polar coordinate system image, and determine the edge response stability index based on the distribution relationship between the radial and tangential gradients.

[0009] S3, the global motion vector of the background feature points of the computer frame and the local motion vector of the feature region of the eccentric shaft end face, the global and local motion decoupling coefficients are determined based on the difference magnitude of the local motion vector and the global motion vector and the dot product of the two.

[0010] S4. Establish a sliding time window, obtain the edge response stability index and global and local motion decoupling coefficients corresponding to each frame within the time window, obtain the dynamic cumulative loosening risk index through time dimension integration, and trigger an alarm when the dynamic cumulative loosening risk index exceeds the preset threshold.

[0011] Its effects are as follows: In response to the harsh environment of high dust and strong vibration at the jaw crusher operation site, the complex rotational motion is transformed into linear features through polar coordinate transformation, and the dust and noise interference is effectively eliminated by combining the edge response stability index; more importantly, by using the global and local motion decoupling mechanism, the independent micro-slippage of the eccentric shaft assembly can be accurately separated from the violent background shaking of the machine body. With the help of the time window integration algorithm, false alarms caused by instantaneous impact are eliminated, and early, accurate and non-contact online monitoring of loosening faults of key components is realized.

[0012] Furthermore, the formula for calculating the edge response stability index is as follows:

[0013]

[0014] in, As an indicator of edge response stability; To analyze the total number of pixels within the window; To analyze the first in the window Radial gradient magnitude of each pixel; To analyze the first in the window The magnitude of the tangential gradient at each pixel; It is the basic stability constant.

[0015] Its effect is as follows: by using a specific edge response stability calculation formula and comparing the distribution differences of radial and tangential gradients, the feature weights of real mechanical contours are significantly enhanced from a mathematical perspective. At the same time, the disordered gradient signals caused by suspended dust are suppressed, ensuring that the system can still lock onto physically meaningful monitoring targets even when the imaging clarity is impaired, thus improving the robustness of feature extraction.

[0016] Furthermore, the formula for calculating the global and local motion decoupling coefficients is as follows:

[0017]

[0018] in, These are the decoupling coefficients between global and local motion; The magnitude of the difference between the local motion vector and the global motion vector; It is the dot product of the local motion vector and the global motion vector; is the damping constant.

[0019] Its effect is as follows: by using a motion decoupling coefficient model that includes difference modulus and dot product operation, the correlation between the local motion of the component and the global background vibration of the rack is quantified; the algorithm can output a low value when the two vibrate synchronously, and output a high value quickly when the component has asynchronous relative displacement, thereby effectively offsetting strong common mode vibration interference and realizing the keen capture of small relative motion in dynamic background.

[0020] Furthermore, the formula for calculating the dynamic cumulative loosening risk index is as follows:

[0021]

[0022] in, This is a dynamically accumulated risk index for easing. The frame length of the time window; The current moment; The time step for backtracking; for Edge response stability metrics at any given time; for The decoupling coefficients of global and local motion at time t.

[0023] Its effects are as follows: a dynamic cumulative risk index model based on the time dimension is established, and the weak abnormal signals of a single frame are transformed into statistically significant trend indicators by using logarithmic functions and integral processing; compared with simple single-frame threshold determination, this method not only retains the sensitivity to loosening faults in the early stage, but also greatly smooths the numerical fluctuations caused by random noise or instantaneous mechanical impact, significantly reduces the false alarm rate of the system, and improves the reliability of alarms.

[0024] Furthermore, after acquiring the original grayscale image of the eccentric shaft end face of the jaw crusher, the process also includes enhancing the original grayscale image. The process uses a dark channel prior algorithm for dehazing to eliminate image blurring caused by dust.

[0025] Its effects are as follows: In response to the problem of extremely high dust concentration at the mining crushing site, the dark channel prior dehazing algorithm is introduced to preprocess and enhance the original image, which effectively improves the contrast and clarity of the image. This not only improves the quality of visual observation, but also provides a high-quality data foundation for subsequent edge detection and feature extraction, and reduces the risk of missed detection due to low environmental visibility.

[0026] Furthermore, the center coordinates of the eccentric shaft end face are determined by circular contour positioning, and the Hough transform algorithm is used to retrieve the circular boundary of the eccentric shaft end cap in the image, thereby obtaining the two-dimensional coordinates of the center.

[0027] The effect is as follows: the Hough transform is used to accurately locate the center coordinates of the eccentric shaft end cap, providing an accurate geometric origin for the subsequent polar coordinate transformation; this step ensures that the annular mechanical structure can be regularly unfolded into linear features in the polar coordinate system, simplifying the complex circumferential rotation matching problem into linear displacement calculation, which reduces the computational complexity of the algorithm and improves the geometric accuracy of the state analysis of rotating parts.

[0028] Furthermore, the global motion vector of the background feature points of the rack is obtained by selecting the bolt head as a reference point in the static area of ​​the rack and using optical flow to trace the displacement vector of the reference point between consecutive frames.

[0029] Its effects are as follows: by tracking the optical flow displacement of the stationary area of ​​the rack to represent the global background vibration vector, a real physical reference benchmark is provided for motion decoupling; the method cleverly utilizes the rigidity of the rack itself, and can obtain the overall vibration state of the fuselage in real time without the need to deploy additional contact accelerometers, ensuring that the reference system for relative motion calculation is accurate and saving hardware costs.

[0030] Furthermore, the local motion vector of the characteristic region of the eccentric shaft end face is obtained by calculating the abnormal displacement vector of the region other than the rotation component of the eccentric shaft in the region with the highest edge response stability index.

[0031] Its effect is that it establishes a feature point optimization mechanism based on edge quality, calculates local motion vectors only in the region with the highest edge response stability, automatically avoids feature blurring areas caused by oil stains, wear or reflection, and ensures that the displacement data input to the decoupled model comes from the clearest and most reliable mechanical contour, thereby further improving the credibility of the final loosening judgment result.

[0032] Furthermore, when the dynamic cumulative loosening risk index exceeds a preset threshold, an alarm is triggered, which also includes outputting the location information of the components determined to be loose and sending an emergency stop command to the crusher control system.

[0033] Secondly, the present invention provides a jaw crusher eccentric shaft abnormality detection system, including a memory and a processor. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned jaw crusher eccentric shaft abnormality detection method is implemented.

[0034] The beneficial effects are:

[0035] This invention constructs a multi-dimensional motion feature evaluation system. Through cross-validation and collaborative judgment of edge response stability index and global and local motion decoupling coefficient, it breaks through the strict dependence of traditional visual monitoring technology on static background. It successfully solves the technical bottleneck of non-contact monitoring in the strong vibration and high dust operation environment of jaw crusher, realizes accurate perception of eccentric shaft assembly loosening under extreme working conditions, greatly reduces the manpower and material resources for equipment fault diagnosis and maintenance, and significantly reduces the equipment operation and maintenance costs of mining enterprises.

[0036] Furthermore, this invention employs a sliding time window accumulation mechanism to transform the sub-pixel-level micro-displacement signal obtained from a single frame detection into a statistically significant dynamic cumulative loosening risk index through time-dimensional integration. This scheme ensures high detection sensitivity for loosening faults in their nascent stage while effectively smoothing numerical fluctuations caused by instantaneous environmental impacts and random noise by leveraging the nonlinear compression characteristics of the logarithmic function. This fundamentally avoids the problem of instantaneous false alarms in the system, providing stable and reliable technical support and implementation path for the intelligent, refined, and unmanned operation and maintenance of large-scale crushing equipment in mines. Attached Figure Description

[0037] Figure 1 This is a flowchart of the abnormal detection method for the eccentric shaft of a jaw crusher in this invention.

[0038] Figure 2 This is a schematic diagram comparing the feature space distribution in this invention.

[0039] Figure 3 This is a schematic diagram of motion correlation analysis in this invention.

[0040] Figure 4 This is a schematic diagram of the health status assessment of multiple components in this invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] An embodiment of the jaw crusher eccentric shaft anomaly detection method provided by the present invention:

[0043] like Figure 1 As shown, the method for detecting abnormalities in the eccentric shaft of a jaw crusher includes the following steps:

[0044] S1. Obtain the original grayscale image of the eccentric shaft end face of the jaw crusher, determine the center coordinates of the eccentric shaft end face by circular contour positioning, and convert the original grayscale image into a polar coordinate system image with the center coordinates as the origin.

[0045] In practice, a high frame rate industrial camera is mounted on a vibration-damping bracket on the side of the jaw crusher frame, ensuring that the lens optical axis is vertically aligned with the center of the eccentric shaft end face. Due to the extremely high dust concentration on site, the images often appear hazy. This step first uses a dark channel prior algorithm to dehaze and enhance the original grayscale image. Then, Hough transform is used to retrieve the circular outline of the eccentric shaft end cap in the first frame image and obtain the coordinates of the center. Using this coordinate as the origin, the image is moved from the Cartesian coordinate system. Mapping to polar coordinates In the polar coordinate system, the concentric circular ring structures such as the round nut and the shaft end face are unfolded into horizontal straight lines, which simplifies the subsequent monitoring of component rotational slip from complex curve matching to linear displacement detection.

[0046] By transforming the polar coordinates, the original annular structure and circular rotation of the eccentric shaft end face are regularly unfolded into linear features and linear displacements. This significantly simplifies the algorithmic logic for subsequent feature extraction and motion analysis, effectively reducing computational overhead and power consumption. Simultaneously, this transformation method allows the mechanical contour edges of the eccentric shaft assembly to exhibit a clearer and more regular distribution in the polar coordinate image, providing a standardized and regularized image representation for the subsequent accurate extraction of stable mechanical edge features, thus greatly improving the accuracy and robustness of edge detection.

[0047] S2 evaluates and analyzes the radial and tangential gradients of each pixel within the window on the polar coordinate system image, and determines the edge response stability index based on the distribution relationship between the radial and tangential gradients.

[0048] Considering that gradient features alone are highly unstable under dust interference, this step aims to filter out the true mechanical edges in the polar coordinate image. The radial gradient along the radial direction is calculated using operators (such as the Sobel operator). and tangential gradient along the circumference Edge response stability index The calculation formula is as follows:

[0049]

[0050] in, To analyze the total number of pixels within the window, for example, Pick Pixel window; To analyze the first in the window Radial gradient magnitude of each pixel; To analyze the first in the window The magnitude of the tangential gradient at each pixel; It is the basic stability constant.

[0051] Taking a certain calculation as an example: assuming the average pixel brightness within the analysis window is 100, then... If a window is located at the actual edge of a nut, its radial gradient sum is... The value is 500, due to edge smoothing, the sum of the tangential gradients. The value is only 2, which is calculated by substituting it into the formula. If the window is located in a dusty / noise zone, the total radial gradient is 200, but due to the disordered noise, the total tangential gradient is also as high as 150. The calculated values ​​are... As can be seen, the index value of the real edge is much higher than that of the noisy area.

[0052] By quantitatively evaluating and thresholding edge response stability indicators, the system can accurately distinguish between effective gradient signals of real mechanical contours and disordered gradient signals caused by dust and noise, efficiently filtering various environmental noise interferences from complex image backgrounds. This evaluation method relies on the distribution relationship between radial and tangential gradients to enhance and highlight physically meaningful fastener edge features, while weakening and suppressing irregular noise pixels. This completely eliminates interference from invalid signals on the monitoring results, ensuring that even under harsh conditions with compromised imaging quality, the system can still accurately lock and continuously track the real contours of key fasteners such as the eccentric shaft end face flywheel, withdrawal sleeve, and round nut. This lays a precise and stable feature foundation for subsequent extraction of local motion vectors, significantly improving the accuracy and anti-interference capability of feature tracking.

[0053] S3, the global motion vector of the background feature points of the computer frame and the local motion vector of the feature region of the eccentric shaft end face, the global and local motion decoupling coefficients are determined based on the difference magnitude of the local motion vector and the global motion vector and the dot product of the two.

[0054] To isolate loose component displacements from severe fuselage swaying, this step introduces a motion decoupling mechanism. First, a reference point is selected in the stationary region of the frame, and its motion vector is calculated using optical flow as the global motion vector. Simultaneously, the motion vectors of the highly stable edge regions determined in step S2 are calculated as local motion vectors. Global and local motion decoupling coefficients The calculation formula is as follows:

[0055]

[0056] in, The magnitude of the difference between the local motion vector and the global motion vector; It is the dot product of the local motion vector and the global motion vector; is the damping constant.

[0057] Set the damping constant Assuming the frame vibration velocity is under normal operating conditions... The components vibrate synchronously. At this point, the square of the difference modulus is... dot product Calculations yielded The value is extremely small; if the component becomes loose, its local movement becomes... At this time, the difference modulus squared dot product Calculations yielded The value increased significantly.

[0058] By quantifying the decoupling coefficients of global and local motions, the correlation between the local motion of the eccentric shaft assembly and the global vibration of the frame can be accurately quantified. With the synergistic effect of the difference modulus and dot product operation, the correlation between the two motions can be used to effectively cancel the global vibration interference under strong background, strip away the invalid signals caused by the overall shaking of the fuselage, and accurately lock the independent micro-relative displacement of the eccentric shaft assembly relative to the frame from the superimposed complex motions, providing accurate and reliable motion characteristic data for subsequent loosening fault determination.

[0059] S4. Establish a sliding time window, obtain the edge response stability index and global and local motion decoupling coefficients corresponding to each frame within the time window, obtain the dynamic cumulative loosening risk index through time dimension integration, and trigger an alarm when the dynamic cumulative loosening risk index exceeds the preset threshold.

[0060] To avoid false alarms caused by noise in a single frame, this step uses integration over time to make the determination, and sets the sliding window length accordingly. Frame, dynamic cumulative loosening risk index The calculation formula is as follows:

[0061]

[0062] in, The frame length of the time window; The current moment; The time step for backtracking; for Edge response stability metrics at any given time; for The decoupling coefficients of global and local motion at time t.

[0063] Assume that over a period of time, the mean edge stability remains around 10, while the decoupling coefficient between global and local motion increases from 0.001 to 0.05. (Logarithmic term) After 50 frames of accumulation and root mean square calculation, The index will rise smoothly from an extremely low value, and when it exceeds a preset warning threshold, the system will immediately issue an alarm. For example, if the alarm threshold is set to 0.5, the index will quickly exceed this value when significant asynchronous displacements occur continuously.

[0064] By using the quantitative assessment and trend determination of the dynamic cumulative loosening risk index, weak and discrete sub-pixel level slip signals in single-frame detection can be integrated and accumulated over time, transforming them into a statistically significant continuous risk trend indicator. This serves as a reliable alarm basis for equipment loosening faults, accurately capturing minute loosening hazards in their nascent stages, and truly achieving early identification and prediction of loosening faults in eccentric shaft components.

[0065] Reference Figure 2 The diagram illustrates the feature space distribution after applying this invention. The left side shows the feature distribution of the prior art, and the right side shows the feature distribution of this invention. After applying the edge response stability index and the global and local motion decoupling coefficient, the normally operating sample points are tightly confined within the safe operating feature domain, while the loosened sample points are clearly deviated from the safe domain. This intuitively demonstrates the superior performance of this solution in distinguishing between vibration and loosening.

[0066] Reference Figure 3 The synchronous vibration data points closely follow the ideal synchronous line with a slope of 1, indicating that the component and the frame vibrate synchronously; the deviation of the asynchronous drift data points reveals that the component has generated independent relative motion.

[0067] Reference Figure 4 By dynamically accumulating the loosening risk index, the health status of different components can be divided into safe, warning, and shutdown zones, verifying the effectiveness of the graded warning system.

[0068] An embodiment of the jaw crusher eccentric shaft anomaly detection system provided by the present invention:

[0069] The jaw crusher eccentric shaft anomaly detection system includes a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned jaw crusher eccentric shaft anomaly detection method is implemented.

[0070] The jaw crusher eccentric shaft abnormality detection system also includes other components well known to those skilled in the art, such as communication interfaces. Their settings and functions are known in the art and will not be described in detail here.

[0071] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions stored or otherwise maintained by such a computer-readable medium.

[0072] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for detecting abnormalities in the eccentric shaft of a jaw crusher, characterized in that, Includes the following steps: S1. Obtain the original grayscale image of the eccentric shaft end face of the jaw crusher, determine the center coordinates of the eccentric shaft end face through circular contour positioning, and convert the original grayscale image into a polar coordinate system image with the center coordinates as the origin. S2, evaluate and analyze the radial and tangential gradients of each pixel within the window on the polar coordinate system image, and determine the edge response stability index based on the distribution relationship between the radial and tangential gradients. S3, the global motion vector of the background feature points of the computer frame and the local motion vector of the feature region of the eccentric shaft end face, the global and local motion decoupling coefficients are determined based on the difference magnitude of the local motion vector and the global motion vector and the dot product of the two. S4. Establish a sliding time window, obtain the edge response stability index and global and local motion decoupling coefficients corresponding to each frame within the time window, obtain the dynamic cumulative loosening risk index through time dimension integration, and trigger an alarm when the dynamic cumulative loosening risk index exceeds the preset threshold.

2. The method for detecting abnormalities in the eccentric shaft of a jaw crusher according to claim 1, characterized in that, The formula for calculating the edge response stability index is: in, As an indicator of edge response stability; To analyze the total number of pixels within the window; To analyze the first in the window Radial gradient magnitude of each pixel; To analyze the first in the window The magnitude of the tangential gradient at each pixel; It is the basic stability constant.

3. The method for detecting abnormalities in the eccentric shaft of a jaw crusher according to claim 1, characterized in that, The formula for calculating the global and local motion decoupling coefficients is as follows: in, These are the decoupling coefficients between global and local motion; The magnitude of the difference between the local motion vector and the global motion vector; It is the dot product of the local motion vector and the global motion vector; is the damping constant.

4. The method for detecting abnormalities in the eccentric shaft of a jaw crusher according to claim 1, characterized in that, The formula for calculating the dynamic cumulative loosening risk index is as follows: in, This is a dynamically accumulated risk index for easing. The frame length of the time window; The current moment; The time step for backtracking; for Edge response stability metrics at any given time; for The decoupling coefficients of global and local motion at time t.

5. The method for detecting abnormalities in the eccentric shaft of a jaw crusher according to claim 1, characterized in that, After acquiring the original grayscale image of the eccentric shaft end face of the jaw crusher, the process also includes enhancing the original grayscale image. The process uses a dark channel prior algorithm to remove fogging in order to eliminate image blurring caused by dust.

6. The method for detecting abnormalities in the eccentric shaft of a jaw crusher according to claim 1, characterized in that, The center coordinates of the eccentric shaft end face are determined by locating the circular contour. The Hough transform algorithm is then used to retrieve the circular boundary of the eccentric shaft end cap in the image, thereby obtaining the two-dimensional coordinates of the center.

7. The method for detecting abnormalities in the eccentric shaft of a jaw crusher according to claim 1, characterized in that, The global motion vector of the background feature points of the rack is obtained by selecting the bolt head as a reference point in the static area of ​​the rack and using optical flow to trace the displacement vector of the reference point between consecutive frames.

8. The method for detecting abnormalities in the eccentric shaft of a jaw crusher according to claim 1, characterized in that, The local motion vector of the characteristic region on the end face of the eccentric shaft is obtained by calculating the abnormal displacement vector of the region other than the rotation component of the eccentric shaft in the region with the highest edge response stability index.

9. The method for detecting abnormalities in the eccentric shaft of a jaw crusher according to claim 1, characterized in that, An alarm is triggered when the dynamic cumulative loosening risk index exceeds a preset threshold. The alarm also includes outputting the location information of components determined to be loose and sending an emergency stop command to the crusher control system.

10. A detection system for abnormal eccentric shafts in a jaw crusher, characterized in that, The method includes a memory and a processor. The memory stores computer program instructions, which, when executed by the processor, implement the abnormal detection method for the eccentric shaft of a jaw crusher as described in any one of claims 1-9.

Images