Fault diagnosis method based on vibration trajectory profile map

Abstract

The invention relates to the field of equipment state monitoring and fault diagnosis, and discloses a fault diagnosis method based on a vibration track profile map. According to the method, vibration signals are collected and converted into displacement data, a vibration track profile map is drawn in a coordinate system, a graphical interaction interface is provided for a user to customize a graphical alarm area, a graphical boundary is converted into a mathematical criterion, and real-time out-of-limit judgment and alarm of track points are achieved. The method supports various alarm area shapes such as a circle, an ellipse and a polygon, track visualization can be enhanced through dragging dynamic adjustment in combination with a time sequence color coding technology, the dependence on a complex mechanism model is reduced, and the intuition, flexibility and accuracy of fault diagnosis are improved.

Description

Technical Field

[0001] This invention relates to the field of equipment condition monitoring and fault diagnosis, specifically to a method for diagnosing faults in rotating machinery based on vibration trajectory profile diagrams. Background Technology

[0002] With the rapid development of industrial production, large rotating machinery plays an increasingly important role in modern industry. However, these machines are susceptible to various factors during operation, such as wear, loosening, eccentricity, and imbalance, leading to abnormal vibration signals and affecting normal operation. Traditional methods for diagnosing rotating machinery faults mainly rely on manual experience and professional knowledge, which often makes accurate judgment difficult for complex fault situations. Currently, vibration sensor-based analysis has become the mainstream technology, mainly including spectral analysis and vibration intensity threshold methods.

[0003] However, existing fault diagnosis methods have several shortcomings. First, they suffer from poor model dependence and universality: vibration mechanism models vary greatly across different equipment and operating conditions, making it difficult to establish an accurate and universal model, resulting in low reliability and real-time performance of the diagnosis. Second, their abstract nature leads to low participation: the diagnostic process relies on abstract spectra or numerical values, which are difficult for equipment maintenance personnel to understand and participate in intuitively, hindering the rapid conversion of field experience into diagnostic criteria and making continuous optimization of the diagnostic system difficult. Furthermore, threshold setting is inflexible: traditional threshold setting methods often use fixed values ​​or simple linear relationships, failing to meet the need for flexible threshold adjustments during equipment operation, leading to inaccurate alarms.

[0004] Therefore, there is an urgent need for an intuitive, interactive fault diagnosis method that does not rely on complex mechanistic models to improve the accuracy and efficiency of fault diagnosis. Simultaneously, this method should also feature dynamic and personalized settings, enabling the rapid and flexible setting of alarm thresholds based on the actual operating status of the equipment and historical data, thereby achieving more personalized and higher diagnostic accuracy.

[0005] Several invention patents have been developed to address the issues of visualizing and intuitively interpreting equipment mechanical fault diagnosis, moving away from complex mechanistic models, and enabling dynamic and personalized settings.

[0006] For example, CN111307440A discloses a qualitative diagnostic method for power frequency faults in rotating machinery. This method involves installing measuring points at the rotor's fault-sensitive cross-section to collect shaft vibration displacement signals and their corresponding key phase signals. It then uses a two-dimensional holographic spectrum to extract the pure fault power frequency component from the vibration signal and finally constructs a two-dimensional holographic spectrum waterfall plot and its characteristic trend map corresponding to the power frequency to describe the changing characteristics of the pure fault precession trajectory. However, this method lacks an intuitive graphical interface when performing FFT analysis on the collected vibration signal samples, making it difficult to provide maintenance personnel with intuitive fault diagnosis information.

[0007] CN106226049A provides a method for fault diagnosis of rotating machinery based on waveform indices. This method divides real-time acquired vibration signals into fault-free vibration signals and mixed signals using standard vibration signals. Waveform indices are then constructed using these mixed and fault-free vibration signals, resulting in a more sensitive fault diagnosis and the ability to detect weak fault characteristic signals through signal changes. However, this method still has shortcomings in terms of improving the waveform indices, making it difficult to develop more sensitive and accurate waveform indices to improve the accuracy and reliability of rotating machinery fault diagnosis.

[0008] The existing technology has the following drawbacks:

[0009] 1. Poor model dependence and universality: Vibration mechanism models vary greatly under different equipment and operating conditions, making it difficult to establish an accurate and universal model, resulting in low reliability and real-time performance of the diagnosis.

[0010] 2. Abstractness leads to low participation: The diagnostic process relies on abstract spectrum or numerical values, which are difficult for equipment maintenance personnel to understand and participate in intuitively. They are unable to quickly transform on-site experience into diagnostic criteria, making it difficult to continuously optimize the diagnostic system.

[0011] 3. Inflexible threshold setting: Traditional threshold setting methods are mostly fixed values ​​or simple linear relationships, which cannot meet the needs of equipment to freely adjust the threshold during operation, resulting in inaccurate alarms.

[0012] 4. Lack of intuitive graphical interface: Existing methods lack an intuitive graphical interface when performing FFT analysis on collected vibration signal samples, making it difficult to provide maintenance personnel with intuitive fault diagnosis information.

[0013] 5. Lack of dynamic personalized settings: It is impossible to quickly and flexibly set the alarm threshold that best fits the actual operating status and historical data of the equipment, resulting in a lack of personalization and accuracy in diagnosis.

[0014] Purpose of the invention

[0015] Existing technologies for fault diagnosis based on vibration sensors suffer from poor model dependence, high abstraction, inflexible threshold settings, and a lack of intuitive graphical user interfaces. Therefore, to address these issues, this invention provides a method and system for diagnosing mechanical faults in equipment based on vibration trajectory profiles. The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for diagnosing mechanical faults in equipment based on vibration trajectory profiles, comprising:

[0016] Step 1: Data Acquisition and Signal Processing;

[0017] Step 2: Vibration trajectory drawing and visualization;

[0018] Step 3: Graphical alarm threshold setting and calculation;

[0019] Step 4: Trajectory-based auxiliary diagnosis.

[0020] Step 1 includes:

[0021] Step 101: Simultaneously acquire vibration acceleration signals in multiple directions using a vibration sensor at a preset sampling frequency;

[0022] Step 102: Verify the validity of the acquired acceleration signal, check whether the data is within the sensor's range, and whether the data packets are continuous and without loss.

[0023] Step 103: Filter the acceleration signal, using a high-pass filter to remove the DC component from the signal and a low-pass filter to remove high-frequency noise;

[0024] Step 2 includes:

[0025] Step 201: Establish a coordinate system. Take the position of the sensor when it is stationary as the origin and establish a two-dimensional or three-dimensional profile coordinate system.

[0026] Step 202: Draw the trajectory. Connect the series of displacement points obtained in Step 1 with line segments in the coordinate system to form the motion trajectory.

[0027] Step 203: Perform time-series color coding. To clearly show the trend of trajectory changes over time, different colors are used to distinguish trajectories in different time periods. For example, the trajectory segments of the most recent N points are drawn in red, and the trajectory segments of the previous M points are drawn in light blue.

[0028] Step 3 includes:

[0029] Step 301: Initial alarm area. The system provides a circular alarm area centered at the origin by default.

[0030] Step 302: User interaction adjustment. Users can directly drag the circular boundary to change it into an ellipse through the human-computer interaction interface, or define a complex alarm area by dragging the vertex of an irregular polygon.

[0031] Step 303: Threshold quantization calculation. For an ellipse, after the user drags, the software obtains the new semi-major axis 'a' and semi-minor axis 'b' parameters of the ellipse in real time, and the ellipse equation (X / a). 2 +(Y / b) 2 =1 is the alarm criterion; when the trajectory point satisfies (X / a) 2 +(Y / b) 2 When the value is greater than 1, an alarm is triggered.

[0032] Step 304: For irregular polygons, the software discretizes the user-drawn graphic into a polygon composed of a series of vertex coordinates. The algorithm for determining whether a trajectory point exceeds the limit uses the ray method. A ray is drawn horizontally to the right from the point. If the number of intersections with the polygon boundary is odd, then it is inside the polygon (safe); otherwise, it is outside (alarm).

[0033] Step 4 includes:

[0034] Step 401: Users can make a quick preliminary judgment on the fault type by observing the typical shape of the trajectory and combining it with knowledge of the mechanism model.

[0035] Step 402: Based on the equipment type and operating conditions, and combined with historical fault data, optimize alarm rules to improve the accuracy and reliability of diagnosis.

[0036] In step 1, the data acquisition card acquires acceleration signals in the X and Y directions at a sampling rate of 5120Hz. The sampling frequency can be set according to the equipment type and fault detection accuracy.

[0037] In step 2, the motion trajectory drawing software updates the XY profile at a refresh rate of 100ms. The trajectory line settings are as follows: the trajectory within the last 10 seconds is represented by a thick red line, and the trajectory within the previous 10-30 seconds is represented by a thin blue line.

[0038] In step 3, the radius of the default circular alarm area is 2-3 times the trajectory of the device during normal operation, the major semi-axis a of the ellipse is 1.2-1.5 times the diameter of the device, and the minor semi-axis b is 0.2-0.3 times the diameter of the device.

[0039] In step 4, different alarm threshold rules are set according to different fault types. For example, an ellipse indicates imbalance, a figure-eight indicates misalignment, and a petal indicates looseness.

[0040] In step 1, the data processing sub-features include a specific method for calculating displacement by performing two integrations on the acceleration signal (including DC removal and filtering steps), using the Simpson numerical integration method for two integrations.

[0041] In step 3, the graphical interaction sub-features include algorithms that convert user-dragped graphics (ellipses, polygons) into mathematical parameters or criteria in real time (such as elliptic parametric equations, ray tracing).

[0042] In step 2, the visualization sub-features include using different colors and / or line types to represent trajectories within different time windows, and using time-series color coding technology to display the trend of trajectory changes over time.

[0043] Compared with the prior art, the present invention provides a method and system for diagnosing mechanical faults of equipment based on vibration trajectory profile diagrams, which has the following beneficial effects:

[0044] 1. This invention transforms vibration data into intuitive two-dimensional / three-dimensional motion trajectories and uses a graphical drag-and-drop method to set alarm thresholds, overcoming the shortcomings of traditional methods that rely on complex mechanism models, reducing the technical threshold for fault diagnosis, enabling maintenance personnel to make fault judgments more intuitively, and greatly improving diagnostic efficiency.

[0045] 2. This invention adopts a diagnostic method based on trajectory morphology and trend, which does not rely on accurate and difficult-to-obtain vibration mechanism models, thus improving the universality and reliability of the diagnosis. It is applicable to equipment fault diagnosis under various complex working conditions and overcomes the problem of poor model applicability in the prior art.

[0046] 3. This invention provides a flexible alarm threshold setting function through a graphical interface. Users can quickly and flexibly set the alarm threshold that best fits the actual operating status of the equipment and historical data, making the diagnosis more personalized, improving the accuracy of the diagnosis, and overcoming the shortcomings of the traditional threshold setting method that is fixed and inflexible.

[0047] 4. This invention employs multiple visualization methods, such as different colors to distinguish between new and old trajectories and the shape of trajectory movement, which effectively enhances the ability to analyze fault trends, enabling users to more intuitively understand the vibration status of the equipment and achieve early warning of fault evolution.

[0048] 5. This invention perfectly combines human experience (through graphical settings) with the precise calculation of computers (threshold quantification and real-time monitoring), forming a continuously optimized and evolving intelligent diagnostic system, thereby improving the level of intelligence in fault diagnosis. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the system architecture and process of this solution;

[0050] Figure 2 This is a schematic diagram of the interactive alarm settings for this solution;

[0051] Figure 3 This is a schematic diagram of a cross-sectional view depicting a two-dimensional vibration trajectory in the XY direction, as shown in an embodiment of this solution.

[0052] Figure 4 This is a schematic diagram showing that the magnitudes of the X-axis vibration displacement and the Y-axis vibration displacement are inconsistent in an embodiment of this solution;

[0053] Figure 5 This diagram illustrates the setting of irregular alarm thresholds for the vibration trajectory determined during normal operation of the equipment in this scheme. Detailed Implementation

[0054] Example 1:

[0055] like Figure 1 As shown, this invention provides a method for diagnosing mechanical faults in equipment based on vibration trajectory profile diagrams. The specific implementation steps are as follows:

[0056] Step 1: Data Acquisition and Signal Processing

[0057] Step 101: Simultaneously acquire vibration acceleration signals in the X and Y directions using an industrial-grade triaxial IEPE accelerometer installed on the main shaft bearing housing of the wind turbine at a sampling frequency of 5120Hz.

[0058] Step 102: Use a data acquisition card to verify the validity of the acquired acceleration signal, check whether the data is within the sensor's range, and whether the data packets are continuous and without loss.

[0059] Step 103: Use a high-pass filter with a cutoff frequency of 1Hz to filter the acceleration signal to remove the DC component; use different cutoff frequencies to remove high-frequency noise depending on the operating parameters of the monitored object.

[0060] Step 2: Vibration Trajectory Drawing and Visualization

[0061] like Figure 2 As shown, step 201: Establish a two-dimensional XY coordinate system, with the position of the sensor when it is stationary as the origin.

[0062] Step 202: Connect the series of displacement points (X(i), Y(i)) obtained in Step 1 with line segments in the coordinate system to form the motion trajectory.

[0063] Step 203: Using time-series color coding technology, to clearly display the trajectory's changing trend over time, the trajectory segments within the most recent 10 seconds are drawn in red, and the trajectory segments from the previous 10-30 seconds are drawn in light blue. For example... Figure 4 As shown, this method is also applicable to users who can dynamically adjust alarm thresholds based on actual operating data, such as when the magnitude of X-axis vibration displacement is inconsistent with the magnitude of Y-axis vibration displacement.

[0064] like Figure 5 As shown, this method is also applicable to setting irregular alarm thresholds based on the vibration trajectory determined during normal equipment operation;

[0065] Step 3: Graphical alarm threshold setting and calculation

[0066] like Figure 3 As shown, step 301: The system provides a circular alarm area with the origin as the center by default, and the radius of the circle is set to 2.5 times the trajectory of the device during normal operation.

[0067] Step 302: The maintenance personnel drag the circular boundary through the human-machine interface to change it into an elliptical alarm area. The major semi-axis a of the ellipse is set to 1.3 times the diameter of the equipment, and the minor semi-axis b is set to 0.25 times the diameter of the equipment.

[0068] Step 303: The system calculates the parameters of the ellipse in real time, obtaining the new major semi-axis a and minor semi-axis b, and the equation of the ellipse (X / a). 2 +(Y / b) 2 =1 is used as the alarm criterion when the trajectory point satisfies (X / a). 2 +(Y / b) 2 An alarm is triggered when the value is greater than 1.

[0069] Step 304: The system uses the ray method to determine whether the trajectory point exceeds the limit. Draw a ray horizontally to the right from the trajectory point. If the intersection point with the ellipse boundary is not found, draw an infinitely extending horizontal ray from the measured point to the right (or left) and calculate the number of intersection points between the ray and the polygon boundary line.

[0070] The ray casting method is used to determine whether a trajectory point exceeds the limit.

[0071] If the number of intersections is odd, then the point is inside the polygon (safe, no alarm).

[0072] If the number of intersections is even (including 0), then the point is outside the polygon (out of bounds, triggering an alarm).

[0073] If the number is odd, it is inside the ellipse (safe); otherwise, it is outside (alarm).

[0074] Algorithm steps and details: Assume there is a polygon formed by connecting vertices P1, P2, ..., Pk, and a point Q to be measured.

[0075] Initialization: Set a counter count = 0.

[0076] Traverse each edge of the polygon: For each edge (The last edge is) ), perform the following judgment:

[0077] Coordinate check: Check if the ordinate Yq of point Q lies on the edge. Between the two endpoints, ordinates Yi and Yi+1 (i.e., whether Yq is within [min(Y i ,Y i+1 ),max(Y i ,Y i+1 Within this range.

[0078] If it's not there, it means the ray cannot intersect this edge, so skip it directly.

[0079] Calculate the x-coordinate of the intersection point: If Yq is within the interval, then calculate the intersection of the ray (horizontal line Y = Yq) and the edge. The x-coordinate of the intersection point is Xintersect.

[0080] The equation of the line along the side can be expressed in two-point form. Since it is a horizontal ray, the calculation can be simplified:

[0081] X intersect =X i +(Y q -Y i )*(X i-1 -X i ) / (Y i+1 -Y i )

[0082] 5. Determine the validity of the intersection points:

[0083] If XqXintersect > Xq (i.e., the intersection point is to the right of point Q), then this intersection point is counted. In this case, the counter count is incremented by 1.

[0084] 6. Handling special cases at the boundary:

[0085] Vertex intersection: If a ray happens to pass through a vertex of a polygon, it may lead to double or missed calculations. The standard approach is to establish a rule, such as: "When a ray passes through a vertex, only calculate the edge whose vertex is the higher endpoint" (or only calculate the edge whose vertex is the lower endpoint), ensuring that the same edge is calculated only once.

[0086] Edge coincides with ray: If an edge of a polygon lies entirely on a ray (horizontal edge), this edge can usually be ignored because it does not constitute a "crossing" from the inside to the outside or from the outside to the inside.

[0087] The conclusion is: after traversing all edges, check the count.

[0088] count%2==1 (odd number): Point inside the polygon -> safe.

[0089] count%2==0 (even number): Point outside the polygon -> alarm.

[0090] Step 4: Trajectory-based auxiliary diagnosis

[0091] Step 401: The maintenance personnel, combining their knowledge of the mechanism model, observe the typical shape of the trajectory and determine that it is an imbalance fault.

[0092] Step 402: Based on historical data of imbalance faults, optimize alarm rules and adjust alarm threshold values ​​to a specific range.

[0093] Example 2:

[0094] This invention provides a method for diagnosing mechanical faults in equipment based on vibration trajectory profile diagrams. The specific implementation steps are as follows:

[0095] Step 1: Data Acquisition and Signal Processing

[0096] Step 101: Simultaneously acquire vibration acceleration signals in the X, Y, and Z directions using a triaxial MEMS accelerometer mounted on the motor bracket at a sampling frequency of 11kHz.

[0097] Step 102: Use data acquisition software to verify the validity of the acquired acceleration signal, check whether the data is within the sensor's range, and whether the data packets are continuous and without loss.

[0098] Step 103: Use a high-pass filter with a cutoff frequency of 0.5Hz to filter the acceleration signal to remove the DC component; use a low-pass filter with a cutoff frequency of 1kHz to remove high-frequency noise.

[0099] Step 2: Vibration Trajectory Drawing and Visualization

[0100] Step 201: Establish a three-dimensional XYZ coordinate system, with the position of the sensor when it is stationary as the origin.

[0101] Step 202: Connect the series of displacement points (X(i), Y(i), Z(i)) obtained in Step 1 with line segments in the coordinate system to form a motion trajectory.

[0102] Step 203: Using time-series color coding technology, to clearly display the trajectory's changing trend over time, the trajectory segments within the most recent 5 seconds are drawn in red, and the trajectory segments from the previous 5-10 seconds are drawn in light blue. Step 3: Graphical Alarm Threshold Setting and Calculation

[0103] Step 301: The system provides a spherical alarm area centered at the origin by default, with the radius of the sphere set to 2.2 times the trajectory of the device during normal operation.

[0104] Step 302: The maintenance personnel drag the spherical boundary through the human-machine interface to change it into an elliptical alarm area. The major semi-axis a of the ellipse is set to 1.4 times the diameter of the equipment, and the minor semi-axis b is set to 0.28 times the diameter of the equipment.

[0105] Step 303: The system calculates the parameters of the ellipse in real time, obtaining the new major semi-axis a and minor semi-axis b, and the equation of the ellipse (X / a). 2 +(Y / b) 2 +(Z / c) 2 =1 is used as the alarm criterion when the trajectory point satisfies (X / a). 2 +(Y / b) 2 +(Z / c) 2 An alarm is triggered when the value is greater than 1.

[0106] Step 304: The system uses the ray casting method to determine whether a trajectory point exceeds the limit. A ray is drawn horizontally to the right from the trajectory point. If the number of intersections with the ellipse boundary is odd, the point is inside the ellipse (safe); otherwise, it is outside (alarm). Step 4: Auxiliary diagnosis based on trajectory morphology.

[0107] Step 401: The maintenance personnel, combining their knowledge of the mechanism model, observe the typical shape of the trajectory and determine that it is a misalignment fault.

[0108] Step 402: Based on historical data of misalignment faults, optimize the alarm rules and adjust the alarm threshold value to a specific range.

[0109] Example 3:

[0110] This invention provides a method for diagnosing mechanical faults in equipment based on vibration trajectory profile diagrams. The specific implementation steps are as follows:

[0111] Step 1: Data Acquisition and Signal Processing

[0112] Step 101: Simultaneously acquire vibration acceleration signals in the X, Y, and Z directions using a triaxial high-precision accelerometer mounted on the gearbox housing at a sampling frequency of 24kHz.

[0113] Step 102: Use a data acquisition system to verify the validity of the acquired acceleration signal, check whether the data is within the sensor's range, and whether the data packets are continuous and without loss.

[0114] Step 103: Use a high-pass filter with a cutoff frequency of 2Hz to filter the acceleration signal to remove the DC component; use a low-pass filter with a cutoff frequency of 2kHz to remove high-frequency noise.

[0115] Step 2: Vibration Trajectory Drawing and Visualization

[0116] Step 201: Establish a three-dimensional XYZ coordinate system, with the position of the sensor when it is stationary as the origin.

[0117] Step 202: Connect the series of displacement points (X(i), Y(i), Z(i)) obtained in Step 1 with line segments in the coordinate system to form a motion trajectory.

[0118] Step 203: Using time-series color coding technology, to clearly show the trend of trajectory changes over time, the trajectory segments within the most recent 0.2 seconds are drawn in red, and the trajectory segments from the previous 1-3 seconds are drawn in light blue.

[0119] Step 3: Graphical alarm threshold setting and calculation

[0120] Step 301: The system provides a spherical alarm area centered at the origin by default, and the radius of the sphere is set to 2.0 times the trajectory of the device during normal operation.

[0121] Step 302: The maintenance personnel drag the spherical boundary through the human-computer interaction interface to turn it into a polygonal alarm area. The vertex coordinates of the polygon are discretized.

[0122] Step 303: The system calculates the vertex coordinates of the polygon in real time and uses the ray method to determine whether the trajectory point exceeds the limit. When the number of intersections between the trajectory point and the polygon boundary is odd, it is inside the polygon (safe); otherwise, it is outside (alarm).

[0123] Step 304: The system performs adaptive processing on the polygons, dynamically adjusting their shapes according to the actual distribution of the trajectory.

[0124] Step 4: Trajectory-based auxiliary diagnosis

[0125] Step 401: The maintenance personnel, combining their knowledge of the mechanism model, observe the typical shape of the trajectory and determine that it is a loosening fault.

[0126] Step 402: Based on historical data of loosening faults, optimize the alarm rules and adjust the alarm threshold value to a specific range.

[0127] Finally, it should be noted that the above descriptions are merely preferred embodiments of this application, and this application is not limited to the above embodiments. It is understood that other improvements and variations that can be directly derived or conceived by those skilled in the art without departing from the spirit and concept of this application should be considered to be included within the protection scope of this application.

Claims

1. A fault diagnosis method based on vibration trajectory profile diagrams, characterized in that, Includes the following steps: The vibration signals of rotating machinery are collected and converted into displacement data; Based on the displacement data, a cross-sectional view of the vibration trajectory of the rotating machinery is drawn in a pre-established coordinate system; A graphical user interface is provided to respond to user operations and define a graphical alarm area in the vibration trajectory profile. The graphical boundaries of the graphical alarm area are converted into corresponding mathematical criteria. Based on the mathematical criteria, it is determined in real time whether the trajectory points in the vibration trajectory profile exceed the limit, and an alarm is triggered when the limit is exceeded.

2. The method according to claim 1, characterized in that, The step of providing a graphical interactive interface and defining a graphical alarm area in the vibration trajectory profile diagram in response to user operation includes: The initial geometry is displayed in the interactive interface as the default alarm area; In response to the user's dragging operation on the boundary of the geometric shape, the shape and / or size of the geometric shape are adjusted in real time to form the updated graphical alarm area.

3. The method according to claim 2, characterized in that, The geometric shape includes at least one of a circle, an ellipse, or a polygon.

4. The method according to claim 3, characterized in that, When the graphical alarm area is elliptical, the step of converting the graphical boundary of the graphical alarm area into a corresponding mathematical criterion includes: Obtain the semi-major axis parameter (a) and semi-minor axis parameter (b) of the ellipse; Based on the major and minor axis parameters, an ellipse equation is generated as the mathematical criterion. The ellipse equation is used to determine whether the coordinates (X, Y) of the trajectory point satisfy (X / a)² + (Y / b)² > 1.

5. The method according to claim 3, characterized in that, When the graphical alarm area is a polygon, the step of converting the graphical boundary of the graphical alarm area into a corresponding mathematical criterion, and determining whether the trajectory point exceeds the limit based on the mathematical criterion, includes: Obtain the coordinates of each vertex of the polygon; Based on the ray-mapping algorithm, it is determined whether the trajectory point is located outside the polygon according to the vertex coordinates.

6. The method according to any one of claims 1 to 5, characterized in that, The step of drawing a vibration trajectory profile of the rotating machinery in a pre-established coordinate system based on the displacement data includes: Using time-series color coding technology, different colors and / or line types are used to draw trajectory segments generated in different time periods.

7. The method according to any one of claims 1 to 5, characterized in that, The coordinate system is either a two-dimensional coordinate system or a three-dimensional coordinate system.

8. The method according to any one of claims 1 to 5, characterized in that, The step of acquiring vibration signals from rotating machinery and converting the vibration signals into displacement data includes: The collected vibration acceleration signals are filtered; and The filtered acceleration signal is integrated twice to obtain the displacement data.

9. The method according to any one of claims 1 to 5, characterized in that, It also includes the following steps: Based on the morphological characteristics of the vibration trajectory profile, auxiliary diagnosis of fault types is performed.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 9.

Images