A control method and system for a bearing press-fitting machine

Abstract

This invention relates to the field of mechanical automation control technology, and more particularly to a control method and system for a bearing press-fitting machine. The method includes: acquiring multi-dimensional monitoring data during the press-fitting process; constructing a deviation sequence for each dimension at the current moment based on the differences between the monitoring data at the current moment and the data at each moment within a preset window; adjusting the length of the preset window based on the correlation of the deviation sequence to obtain the target window at the current moment; determining the cutoff distance for cluster analysis based on the sparsity of the data point distribution of the monitoring data within the target window; using the cutoff distance as a parameter, performing cluster analysis on the data points within the target window using a density clustering algorithm to identify abnormal data points; and controlling the bearing press-fitting machine to execute a preset response action when the monitoring data at the current moment is abnormal. This invention improves the accuracy of abnormal state identification under multi-stage operating conditions by adaptively adjusting the analysis window and the cluster cutoff distance.

Description

Technical Field

[0001] This invention relates to the field of mechanical automation control technology, and in particular to a control method and system for a bearing press-fitting machine. Background Technology

[0002] Bearing press-fitting is a crucial step in the assembly process. Its core function is to smoothly and accurately press the bearing into the shaft or bearing housing using a hydraulic or servo system. During press-fitting, changes in parameters such as pressure and displacement directly reflect the assembly quality. In a healthy press-fitting process, these parameters exhibit different physical relationships at different stages, but the trajectories formed in the state space at the same stage show a high degree of consistency.

[0003] However, if problems such as jamming or uneven loading occur during the pressing process, the physical correlation and performance between parameters will be disrupted, leading to abnormal fluctuations in pressing force, which can cause damage to parts or even equipment. Therefore, it is necessary to assess the pressing status by analyzing the parameter variation characteristics during the pressing process, identify possible pressing anomalies, and immediately control the equipment to stop for maintenance in order to avoid serious damage to the assembled components.

[0004] Considering that abnormal fluctuations will cause data to deviate from the normal pattern, resulting in significantly lower local density of abnormal data points compared to normal data points, density peak clustering algorithms can be used to cluster data points to identify potential anomalies. However, the actual pressing process involves multiple stages, including idle stroke, contact pressing, and pressure holding. For example, during the idle stroke stage, the collected multi-dimensional monitoring data fluctuates drastically, often exhibiting a low-density distribution. Existing density peak clustering algorithms classify data based on a globally fixed cutoff distance threshold. This strategy easily mislabels low-density data under normal operating conditions as anomalies. This can lead to normal operating conditions being misjudged as abnormal, generating numerous false alarms and impacting production efficiency. Summary of the Invention

[0005] To address the technical problem that the density peak clustering algorithm, which uses a globally fixed cutoff distance, is unable to adapt to the multi-stage variable density conditions in the bearing press-fitting process, leading to the misjudgment of normal low-density data as abnormal, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a method for controlling a bearing press-fitting machine, the method comprising the steps of:

[0007] The system acquires multi-dimensional monitoring data at multiple moments during the bearing press-fitting process. For the monitoring data at the current moment, based on the difference between the current moment's monitoring data and the monitoring data at each moment within a preset window, it determines the deviation degree of each moment within the preset window relative to the current moment, thus constructing a deviation degree sequence for each dimension at the current moment. Based on the correlation between the deviation degree sequences of each dimension at the current moment within the preset window, it adjusts the window length of the preset window to obtain a target window at the same press-fitting stage as the current moment. According to the sparsity of the data point distribution of the monitoring data within the target window, it determines the cutoff distance for cluster analysis. Using the cutoff distance as a parameter, it performs cluster analysis on the data points within the target window using a density clustering algorithm to identify abnormal data points. When the monitoring data at the current moment is identified as an abnormal data point, it controls the bearing press-fitting machine to execute a preset abnormal response action.

[0008] This invention proposes a dual adaptive control method. Existing technologies often misclassify normal sparse data as anomalies due to their inability to adapt to changes in data density at different stages of the pressing process, resulting in a high false alarm rate. This invention first dynamically adjusts the target window based on data correlation to ensure that the analyzed data originates from the same pressing stage. Then, it adaptively determines the cutoff distance for cluster analysis based on the sparsity of the data within this window. This method intelligently adapts to the dynamic changes in the pressing process, significantly improving the accuracy and reliability of anomaly detection under multi-stage, variable-density conditions, effectively avoiding false alarms, and enhancing production efficiency and the intelligence level of the control system.

[0009] Preferably, the deviation of each time point within the preset window relative to the current time point satisfies the following relationship: ;in, It is the first Within the preset window of time, the first At the time of Dimensional deviation; It is the first At the time of Dimensional monitoring data values; It is the first Within the preset window of time, the first At the time of Dimensional monitoring data values; It is the first Within the preset window of time, the first The maximum value of the dimension monitoring data; It is a maximum value function; It is the first At any given time and within its preset window, the first The time difference between moments; This is the total number of data points in the preset window; It is the absolute value symbol; It is the preset first minute value; It is the preset second minute value.

[0010] This invention introduces time difference as a weighting factor. This makes the influence of data points closer to the current time on the deviation greater, while the influence of data points further away is weakened. It can more accurately reflect the dynamic evolution of the pressing state, and make the deviation calculation not only focus on numerical deviation, but also take into account temporal proximity, thereby more accurately quantifying the difference between the current state and recent historical states.

[0011] Preferably, adjusting the window length of the preset window based on the correlation between the deviation sequences at each time point within the preset window to obtain a target window that is in the same pressing stage as the current time includes: calculating the correlation coefficient of the deviation sequences in each dimension within the preset window; obtaining the standard deviation of the correlation coefficient of the target window corresponding to each historical time point in the historical pressing process of the pressing workpiece at the current time, forming a standard deviation set; determining the distinction threshold for being in the same pressing stage based on the standard deviation set; and adjusting the size of the preset window so that the standard deviation of the correlation coefficient within the preset window meets the distinction threshold for the same pressing stage to obtain the target window.

[0012] This invention determines whether data within a window belongs to the same pressing stage by calculating the standard deviation of the correlation coefficient. This method provides a data-driven, automated means to identify stage boundaries in the pressing process, ensuring data consistency in subsequent analyses, avoiding analytical errors caused by window fragmentation or mixing of data from different stages, and improving the accuracy of stage division.

[0013] Preferably, adjusting the preset window size to make the standard deviation of the correlation coefficient within the preset window meet the distinction threshold of the same pressing stage includes: increasing the preset window length by a preset step size when the standard deviation of the preset correlation coefficient at the current time is less than or equal to the distinction threshold; and decreasing the preset window length by a preset step size when the standard deviation of the preset correlation coefficient at the current time is greater than the distinction threshold.

[0014] This invention establishes a closed-loop feedback adjustment mechanism by adjusting the window length accordingly when the standard deviation of the correlation coefficient is less than or greater than a threshold. Compared to fuzzy adjustment strategies, this explicit control logic makes the adaptive adjustment process of the target window more efficient and stable, and can quickly converge to the optimal window size that matches the current pressing stage.

[0015] Preferably, determining the distinction threshold for data in the same pressing stage based on the standard deviation set includes: sorting the data in the standard deviation set in ascending order; calculating the preset quantile of the sorted data using linear interpolation; and using the preset quantile as the distinction threshold.

[0016] Preferably, the cutoff distance satisfies the following relationship: ;in, It is the first The cutoff distance at any given moment; It is the preset global cutoff distance; It is the first Within the target window, the first Time and the The Euclidean distance between data points at time points; It is the first The total number of data points in the target window at any given time; It is the standard normalized function. It is a preset value adjustment coefficient.

[0017] The cutoff distance calculated in this invention is directly related to the sparsity of the data point distribution within the target window. This is the key step in addressing the core pain points of existing technologies. It enables the clustering algorithm to automatically adjust its sensitivity based on whether the data is sparse or dense, thereby fundamentally solving the misjudgment problem caused by the traditional fixed cutoff distance and greatly improving the anomaly identification accuracy of density clustering algorithms under variable density conditions.

[0018] Preferably, the step of using the cutoff distance as a parameter to perform cluster analysis on the data points within the target window using a density clustering algorithm to identify anomalous data points includes: calculating the local density of each data point based on the cutoff distance; determining the cluster center based on the local density and the relative distance between the data points; and identifying data points that do not belong to any cluster after the cluster center is determined as anomalous data points.

[0019] Preferably, the multi-dimensional monitoring data includes at least: pressure data applied by the press machine and displacement data generated by the press head.

[0020] Preferably, the abnormal response action includes at least one of the following: triggering an audible and visual alarm device; controlling the switching valve to cut off the pressing process.

[0021] In a second aspect, the present invention provides a bearing press-fitting machine control system, which includes a memory and a processor. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the bearing press-fitting machine control method of the first aspect of the present invention is implemented.

[0022] By adopting the above technical solution, a computer program is generated from the bearing press-fitting machine control method of the first aspect of the present invention and stored in a memory so that it can be loaded and executed by a processor. A terminal device can then be made based on the memory and the processor for convenient use.

[0023] The beneficial effects of this invention are as follows: By analyzing the correlation stability of local deviation sequences, this invention can dynamically adjust the target window length, ensuring that data within a segment are in the same pressing stage, effectively avoiding cross-stage data interference, thereby improving the accuracy of state quantification. This invention adaptively adjusts the clustering cutoff distance by quantifying the sparsity of data point distribution within the current segment. It expands the neighborhood in sparse regions during idle travel and shrinks the neighborhood in dense regions during pressure holding, significantly improving the accuracy of density peak clustering in identifying abnormal states under variable density conditions. This invention feeds back the anomaly identification results to the control mechanism, achieving adaptive equipment control. When an abnormal state is identified, immediate control measures can be taken to ensure equipment and product quality. This invention, by adaptively adjusting the cutoff distance, avoids the situation where normal low-density data is misjudged as anomalies using traditional methods, reducing false alarms and improving production efficiency. Attached Figure Description

[0024] Figure 1 A flowchart of a bearing press-fitting machine control method provided in an embodiment of the present invention;

[0025] Figure 2 This is a structural block diagram of a bearing press-fitting machine control system provided in an embodiment of the present invention. Detailed Implementation

[0026] The first aspect of this invention provides a control method for a bearing press-fitting machine, such as... Figure 1 As shown, the method includes steps S100-S500:

[0027] Step S100: Obtain multi-dimensional monitoring data at multiple moments during the bearing press-fitting process.

[0028] It should be noted that during the operation of the bearing press-fitting machine, changes in parameters such as pressure and displacement directly reflect whether the press-fitting is smooth and whether there are problems such as jamming or uneven loading. Therefore, in order to achieve precise control, high-frequency data acquisition is required for the press-fitting process to capture transient changes.

[0029] Specifically, sensors integrated into the pressing head and hydraulic cylinder collect real-time status values ​​across multiple monitoring dimensions. These monitoring dimensions include: the axial pressure applied by the pressing machine, which can be acquired by a pressure sensor, and the displacement of the pressing head relative to its initial position, which can be acquired by a displacement sensor or encoder. To ensure the capture of transient changes during the pressing process, such as sudden changes in pressure or displacement caused by jamming or off-center loading, data is collected at a preset acquisition frequency.

[0030] As a preferred implementation, the preset data acquisition frequency can be set to acquire data once every 0.1 to 0.3 seconds. When the acquisition frequency is lower than once every 0.1 seconds, transient abnormal data may be smoothed out or missed, reducing the timeliness of anomaly identification; when the acquisition frequency is higher than once every 0.3 seconds, it will significantly increase the data processing burden and storage requirements, resulting in computational redundancy. Therefore, controlling the acquisition frequency at once every 0.2 seconds can ensure data granularity while taking into account system response speed and computational efficiency.

[0031] Thus, multi-dimensional monitoring data at multiple moments during the bearing press-fitting process were obtained.

[0032] Step S200: For the monitoring data at the current moment, based on the difference between the monitoring data at the current moment and the monitoring data at each moment within the preset window, determine the deviation degree of each moment within the preset window relative to the current moment, so as to construct a deviation degree sequence of each dimension at the current moment.

[0033] It should be noted that the normal bearing press-fitting process has distinct phases, such as the no-load phase, the pressurization phase, and the pressure holding phase. The data distribution characteristics of these phases differ significantly. To accurately quantify the current state, it is necessary to obtain the degree of change of the current state data relative to recent historical data. By constructing a deviation sequence, the changes in the current state can be quantified more precisely, thus providing a basis for subsequent phase division and anomaly identification.

[0034] Specifically, a preset number of time intervals are sampled backwards from the current time as the preset window for the current time. For example, sampling backwards from the current time... The current moment is used as the preset window. The preset number of data points is 20. If the number of data points in the window is insufficient due to a lack of data, it can be supplemented using the mean or known data.

[0035] Furthermore, regarding the first The first time in the preset window At the time of Calculate its deviation in dimensions Its deviation satisfies the following relationship:

[0036] ;

[0037] in, It is the first Within the preset window of time, the first At the time of Dimensional deviation; It is the first At the time of Dimensional monitoring data values; It is the first Within the preset window of time, the first At the time of Dimensional monitoring data values; It is the first Within the preset window of time, the first The maximum value of the dimension monitoring data; It is a maximum value function; It is the first At any given time and within its preset window, the first The time difference between moments; It is the total number of data points in the preset window. It is the absolute value symbol; It is a preset first tiny value used to prevent It can be set to 0, or 0.01, or as needed; It is a preset second tiny value used to prevent It can be set to 0, or 0.01, or as needed.

[0038] In this formula, This reflects the numerical difference between the current time and other times within the preset window; the larger the value, the more significant the deviation in state. Part Two This reflects the temporal proximity. Through weighted processing, points closer to the current time have higher weights, thereby reducing the interference of data from different stages on the deviation calculation and more accurately quantifying the deviation of the current time relative to each time point within the window.

[0039] Repeat the above calculations to obtain the deviation of the current time from each other time in the preset window in each dimension, and construct the deviation sequence of each dimension in the preset window of the current time. The data in the deviation sequence are arranged in ascending order of collection time.

[0040] At this point, the deviation sequence for each dimension at the current moment has been obtained.

[0041] Step S300: Based on the correlation between the deviation sequences of each dimension at the current moment within the preset window, adjust the window length of the preset window to obtain a target window that is in the same pressing stage as the current moment.

[0042] It should be noted that the phased characteristics of the pressing process lead to different data change patterns at each stage. Using a fixed-length window for analysis might fragment data from the same stage or mix in information from different stages, thus affecting the accuracy of the analysis. By introducing correlation analysis of the deviation sequences of each dimension within the window, the characteristics of different stages can be distinguished, and the window length can be adaptively adjusted based on the consistency of the correlation to ensure that the data within the window belong to the same pressing stage.

[0043] Specifically, firstly, the correlation coefficients of the deviation sequences for each dimension within the preset window at the current time are calculated; for example, the correlation coefficients of the deviation sequences for pressure and displacement. For parameter changes within the same pressing stage, the correlation coefficients in the deviation sequences should be relatively concentrated. This step utilizes the Pearson correlation coefficient to obtain the correlation coefficients of the deviation sequences for the two dimensions within the segment to which the current time belongs. The Pearson correlation coefficient is existing technology and will not be elaborated upon here.

[0044] Secondly, obtain the standard deviation of the correlation coefficient of the target window corresponding to each historical moment in the entire pressing history, and form a standard deviation set.

[0045] Next, the data in the standard deviation set are sorted in ascending order, and a preset quantile is calculated using linear interpolation. This quantile is used as the distinguishing threshold for the same pressing stage. The preset quantile can be set to the 10th percentile, or it can be set according to requirements.

[0046] Finally, by adjusting the window length of the preset window, the standard deviation of the correlation coefficient within the target window is made to meet the discrimination threshold.

[0047] The specific adjustment process is as follows: when the standard deviation of the correlation coefficient of the current window is less than or equal to the discrimination threshold, the length of the target window is increased by a preset step size, such as 1; when it is greater than the discrimination threshold, the length of the target window is decreased by a preset step size. This process is repeated until the condition is met, and the data points within the final target window are considered to be in the same pressing stage.

[0048] At this point, the target window that is in the same pressing stage as the current moment has been obtained, eliminating the possibility of misjudgment caused by data differences across stages.

[0049] Step S400: Determine the cutoff distance for cluster analysis based on the sparsity of the data point distribution of the monitored data within the target window.

[0050] It should be noted that although the data has been divided into the same pressing stage, the data density characteristics still differ between stages. For example, the data is sparse in the empty stroke stage and dense in the pressure holding stage. Density peak clustering is an efficient clustering method that identifies cluster centers by calculating the local density and relative distance of data points. It can quickly and accurately cluster the data, and its advantage is that it can effectively discover natural clustering structures in the data without pre-setting the number of clusters, providing strong support for identifying outliers in the data.

[0051] However, traditional density peak clustering algorithms use a globally fixed cutoff distance, making them unsuitable for scenarios with variable density, such as bearing press-fitting. Therefore, after stage division, the cutoff distance needs to be adaptively adjusted according to the sparsity of each stage to improve the clustering algorithm's ability to identify the true state.

[0052] Specifically, firstly, the data for each dimension at all times within the target window are normalized. Then, data points are constructed using the normalized data, and the sparsity of the data points within the target window is obtained by calculating the mean of the Euclidean distance between any two data points. Based on this sparsity, a preset global truncation distance is adaptively adjusted to obtain the truncation distance corresponding to the current time step. The truncation distance corresponding to the current time step satisfies the following relationship:

[0053] ;

[0054] in, It is the first The cutoff distance at any given moment; It is the preset global cutoff distance; It is the first Within the target window, the first Time and the The Euclidean distance between data points at time points; It is the first The total number of data points in the target window at any given time; It is the standard normalized function. It is a preset value adjustment coefficient.

[0055] In this formula, This represents the mean Euclidean distance between any two data points within the target window. A larger value indicates a sparser data distribution, which may be in the initial empty run phase. In this case, the cutoff distance needs to be increased to ensure that the neighborhood contains enough data points. Conversely, a smaller value indicates a denser data distribution, which may be in the holding phase. In this case, the cutoff distance needs to be decreased to improve the accuracy of anomaly detection.

[0056] It should be added that, It is the standard normalization function, used to normalize... Quantization For the interval, specific methods such as minimum-maximum normalization and Z-score standardization can be used, which are all existing technologies and will not be elaborated on here.

[0057] It should also be noted that, Yes The value of the adjustment coefficient will be... Setting it to 0.5 allows you to adjust the value to a range. , so that the calculated Based on this value, we can achieve two-way optimization by increasing or decreasing it. It is the preset global cutoff distance, which can be set to 0.5 or as needed.

[0058] At this point, the cutoff distance used for cluster analysis at each time step has been obtained, hereinafter referred to as the adaptive cutoff distance.

[0059] Step S500: Using the cutoff distance as a parameter, perform cluster analysis on the data points within the target window using a density clustering algorithm to identify abnormal data points; when the monitoring data at the current moment is identified as the abnormal data point, control the bearing press to execute a preset abnormal response action.

[0060] It should be noted that this step is the core step in identifying abnormal states during the pressing process. Using the adaptive cutoff distance obtained in step S400, this step replaces the fixed cutoff distance in the traditional density peak clustering algorithm, thereby improving the accuracy of the algorithm in identifying abnormal states under variable density conditions. Traditional density peak clustering algorithms use a fixed cutoff distance, which can easily misclassify data points in normal sparse areas as abnormal under variable density conditions. By using the adaptive cutoff distance, this problem can be effectively solved, thus improving the accuracy of the clustering algorithm in identifying abnormal states during the pressing process.

[0061] Specifically, the data points within the target window corresponding to the current moment and its historical moments are used as clustering samples. An adaptive cutoff distance at each moment is used to replace the fixed cutoff distance in the traditional density peak clustering algorithm for clustering.

[0062] First, the local density of each data point is calculated based on the adaptive cutoff distance. Then, cluster centers are determined based on the local density and the relative distance between data points. Finally, data points that do not belong to any cluster are identified as outliers. The clustering process of the traditional density peak clustering algorithm is existing technology. This invention only modifies the cutoff distance. The detailed clustering process will not be elaborated here.

[0063] If the monitoring data at the current moment is identified as abnormal, it indicates that the bearing press-fitting machine is in an abnormal operating state. At this time, the corresponding audible and visual alarm devices should be activated immediately, and the corresponding switching valves should be activated to immediately cut off the press-fitting process to prevent damage to the workpiece or equipment.

[0064] The second aspect of this embodiment provides a bearing press-fitting machine control system, such as... Figure 2 As shown, the bearing press-fitting machine control system includes a memory and a processor. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the bearing press-fitting machine control method of the first aspect of the present invention is implemented.

[0065] The bearing press-fitting machine control system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces. Their settings and functions are known in the art and will not be described in detail here.

[0066] 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 (DRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (DRAM), high-bandwidth memory, hybrid memory cube, 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.

[0067] 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 control method for a bearing press-fitting machine, characterized in that, Including the following steps: Acquire multi-dimensional monitoring data at multiple moments during the bearing press-fitting process; Based on the monitoring data at the current moment and the difference between the monitoring data at the current moment and the monitoring data at each moment within the preset window, the deviation of each moment within the preset window relative to the current moment is determined, so as to construct the deviation sequence of each dimension at the current moment. Based on the correlation between the deviation sequences of each dimension within the preset window at the current moment, the window length of the preset window is adjusted to obtain a target window that is in the same pressing stage as the current moment; this includes: calculating the correlation coefficient of the deviation sequences of each dimension within the preset window; obtaining the standard deviation of the correlation coefficient of the target window corresponding to each historical moment in the historical pressing process of the pressing workpiece at the current moment, forming a standard deviation set; determining the distinction threshold for being in the same pressing stage based on the standard deviation set; and adjusting the size of the preset window so that the standard deviation of the correlation coefficient within the preset window meets the distinction threshold for the same pressing stage to obtain the target window; The cutoff distance for cluster analysis is determined based on the sparsity of the data point distribution of the monitored data within the target window. Using the cutoff distance as a parameter, a density clustering algorithm is used to cluster data points within the target window to identify abnormal data points; when the monitoring data at the current moment is identified as an abnormal data point, the bearing press machine is controlled to execute a preset abnormal response action; The preset window size is adjusted so that the standard deviation of the correlation coefficient within the preset window meets the distinction threshold of the same pressing stage. This includes: when the standard deviation of the preset correlation coefficient at the current time is less than or equal to the distinction threshold, the preset window length is increased by a preset step size; when the standard deviation of the preset correlation coefficient at the current time is greater than the distinction threshold, the preset window length is decreased by a preset step size. Determining the distinction threshold for data in the same pressing stage based on the standard deviation set includes: sorting the data in the standard deviation set in ascending order; calculating the preset quantile of the sorted data using linear interpolation; and using the preset quantile as the distinction threshold.

2. The bearing press-fitting machine control method according to claim 1, characterized in that, The deviation of each time point within the preset window relative to the current time point satisfies the following relationship: ； in, It is the first Within the preset window of time, the first At the time of Dimensional deviation; It is the first At the time of Dimensional monitoring data values; It is the first Within the preset window of time, the first At the time of Dimensional monitoring data values; It is the first Within the preset window of time, the first The maximum value of the dimension monitoring data; It is a maximum value function; It is the first At any given time and within its preset window, the first The time difference between moments; This is the total number of data points in the preset window; It is the absolute value symbol; It is the preset first minute value; It is the preset second minute value.

3. The bearing press-fitting machine control method according to claim 1, characterized in that, The cutoff distance satisfies the following relationship: ; in, It is the first The cutoff distance at any given moment; It is the preset global cutoff distance; It is the first Within the target window, the first Time and the Euclidean distance between data points at time points; It is the first The total number of data points in the target window at any given time; It is the standard normalized function. It is a preset value adjustment coefficient.

4. The bearing press-fitting machine control method according to claim 1, characterized in that, The step of using the cutoff distance as a parameter to perform cluster analysis on the data points within the target window using a density clustering algorithm to identify outlier data points includes: Based on the cutoff distance, calculate the local density of each data point; Cluster centers are determined based on the local density and the relative distance between data points; Once the cluster centers are determined, data points that do not belong to any cluster are identified as outliers.

5. The bearing press-fitting machine control method according to claim 1, characterized in that, The multi-dimensional monitoring data includes at least: pressure data applied by the press machine and displacement data generated by the press head.

6. The bearing press-fitting machine control method according to claim 1, characterized in that, The control of the bearing press machine to execute preset abnormal response actions includes: Trigger the audible and visual alarm device; The control valve shuts off the pressing process.

7. A control system for a bearing press-fitting machine, characterized in that, The bearing press-fitting machine control system includes a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a bearing press-fitting machine control method according to any one of claims 1-6 is implemented.

Images