Method for end face grinding of stepped shaft parts of high voltage switch

Abstract

The application discloses an end face grinding processing method for clamping a high-voltage switch stepped shaft part, and relates to the technical field of part grinding processing. The method comprises the following steps: obtaining an end face grinding force of the stepped shaft part in the end face grinding process of the stepped shaft part; analyzing the end face grinding force of the stepped shaft part to determine an actual vibration influence intensity of the stepped shaft part; the actual vibration influence intensity is used for representing the intensity of the actual vibration influence on the stepped shaft part in the end face grinding process; based on the actual vibration influence intensity of the stepped shaft part, the control parameters of proportional-integral-derivative control of the stepped shaft part in the end face grinding process are adjusted; and the control parameters are used for controlling the end face grinding of the stepped shaft part. The application can improve the end face grinding precision, thereby meeting the machining precision requirement of the high-voltage switch stepped shaft.

Description

Technical Field

[0001] This invention relates to the field of parts grinding technology, specifically to a method for grinding the end face of a stepped shaft part used for clamping high-voltage switches. Background Technology

[0002] In high-voltage switchgear, the stepped shaft is a core transmission component, and the grinding accuracy of its end face directly affects the operational reliability and safety of the high-voltage switch. Because the stepped shaft needs to withstand high voltage, high current, and frequent mechanical impacts, the requirements for its end face dimensional accuracy, geometric tolerances, and surface performance are extremely high. Poor end face grinding quality can easily lead to problems such as increased contact resistance and contact welding, which in turn can cause switch disconnection failure and safety accidents.

[0003] In existing end face grinding methods for stepped shaft parts, the machining of the stepped shaft end face is usually fixed with the central axis of the stepped shaft, the tool uses fixed machining parameters, and the grinding process is controlled by proportional-integral-derivative control to complete the end face grinding of the stepped shaft.

[0004] However, vibration interference can occur during the grinding process due to factors such as friction. The fixed proportional-integral-derivative control parameters cannot adapt to the vibration changes, resulting in insufficient end face grinding accuracy, which makes it difficult to meet the machining accuracy requirements of the high-voltage switch stepped shaft. Summary of the Invention

[0005] This invention provides a method for end face grinding of a stepped shaft part for clamping a high-voltage switch, which can improve the end face grinding accuracy and thus meet the machining accuracy requirements of the stepped shaft of the high-voltage switch.

[0006] A first aspect of the present invention provides a method for grinding the end face of a stepped shaft part for clamping a high-voltage switch, comprising:

[0007] During the end face grinding process of stepped shaft parts, the end face grinding force of stepped shaft parts is obtained;

[0008] The end face grinding force of the stepped shaft part is analyzed to determine the actual vibration influence intensity of the stepped shaft part; the actual vibration influence intensity is used to characterize the intensity of the vibration influence on the stepped shaft part during the end face grinding process.

[0009] Based on the actual vibration influence intensity of the stepped shaft parts, the control parameters of the proportional-integral-derivative control are adjusted during the end face grinding process of the stepped shaft parts; the control parameters are used to control the end face grinding of the stepped shaft parts.

[0010] Furthermore, this invention also proposes to analyze the end face grinding force of stepped shaft parts and determine the actual vibration influence intensity of stepped shaft parts, including:

[0011] The grinding force on the end face of the stepped shaft part is continuously rotated N times to form a grinding force analysis cycle; N is a positive integer.

[0012] Based on the maximum and minimum grinding force values ​​in each grinding force analysis cycle, a maximum value sequence and a minimum value sequence are constructed.

[0013] Obtain the target difference curve between the first envelope curve of the maximum value sequence and the second envelope curve of the minimum value sequence;

[0014] The actual vibration impact intensity of the stepped shaft component is determined by analyzing the target difference curve.

[0015] Furthermore, the present invention also proposes to analyze the target difference curve to determine the actual vibration influence intensity of the stepped shaft component, including:

[0016] The target difference curve is decomposed into a seasonal trend based on local weighted regression to obtain the target trend term and the target seasonal term of the target difference curve.

[0017] Obtain the target fitted slope of the target trend term and the average amplitude of the target seasonal term;

[0018] The actual vibration impact intensity of the stepped shaft component is determined based on the target fitting slope and average amplitude.

[0019] Furthermore, the present invention also proposes that the actual vibration influence intensity is multiple, and the control parameters include a proportionality coefficient;

[0020] Based on the actual vibration impact intensity of the stepped shaft parts, the control parameters of the proportional-integral-derivative control for the stepped shaft parts during end face grinding are adjusted, including:

[0021] Based on the influence intensity sequence composed of multiple actual vibration influence intensities, the first proportional adjustment parameter is determined.

[0022] Based on the first proportional adjustment parameter, the first proportional coefficient at the current moment is adjusted to obtain the second proportional coefficient.

[0023] Furthermore, the present invention also proposes determining a first proportional adjustment parameter based on an influence intensity sequence composed of multiple actual vibration influence intensities, including:

[0024] Based on the difference in influence intensity between two adjacent actual vibration influence intensities in the influence intensity sequence, an intensity difference sequence is constructed.

[0025] The historical control evaluation value is obtained by averaging the influence intensity differences in the intensity difference sequence.

[0026] Based on historical control evaluation values ​​and the target actual vibration influence intensity in the influence intensity sequence, the first proportional adjustment parameter is determined; the target actual vibration influence intensity is the last actual vibration influence intensity in the influence intensity sequence.

[0027] Furthermore, the present invention also proposes adjusting the first proportional coefficient at the current moment based on the first proportional adjustment parameter to obtain the second proportional coefficient, including:

[0028] The second proportional adjustment parameter is obtained by performing an exponential function on the negative of the first proportional adjustment parameter.

[0029] The second proportional adjustment parameter is multiplied by the first proportional coefficient to obtain the second proportional coefficient.

[0030] Furthermore, the present invention also proposes that the control parameters further include differential coefficients;

[0031] Based on the first proportional adjustment parameter, after adjusting the first proportional coefficient at the current moment to obtain the second proportional coefficient, the following steps are also included:

[0032] Simulation control is performed based on the second proportional coefficient to obtain the simulation control evaluation value corresponding to the simulated grinding force of the second proportional coefficient;

[0033] Based on the simulated control evaluation value, the first differential coefficient at the current moment is adjusted to obtain the second differential coefficient.

[0034] Furthermore, the present invention also proposes to adjust the first differential coefficient at the current moment based on the simulated control evaluation value to obtain the second differential coefficient, including:

[0035] The target fluctuation value is obtained by multiplying the difference between the simulated control evaluation value and the actual vibration influence intensity by the actual vibration influence intensity; the actual vibration influence intensity is the last actual vibration influence intensity in the influence intensity sequence.

[0036] The target fluctuation value is processed by an exponential function to obtain the differential adjustment parameter;

[0037] Multiply the differential adjustment parameter by the first differential coefficient to obtain the second differential coefficient.

[0038] Furthermore, the present invention also proposes that the control parameters further include integral coefficients;

[0039] Based on the simulated control evaluation value, after adjusting the first differential coefficient at the current moment to obtain the second differential coefficient, the following steps are also included:

[0040] Based on the second proportional coefficient and the second differential coefficient, the first integral coefficient at the current moment is adjusted to obtain the second integral coefficient.

[0041] Furthermore, the present invention also proposes to adjust the first integral coefficient at the current moment based on the second proportional coefficient and the second differential coefficient to obtain the second integral coefficient, including:

[0042] The difference between the first proportional coefficient and the second proportional coefficient is divided by the first proportional coefficient to obtain the proportional adjustment degree.

[0043] The difference between the first differential coefficient and the second differential coefficient is divided by the first differential coefficient to obtain the differential adjustment degree.

[0044] Based on the proportional adjustment degree and the differential adjustment degree, the first integral coefficient is adjusted to obtain the second integral coefficient.

[0045] The present invention has the following beneficial effects:

[0046] In the end-face grinding method for clamping a stepped shaft part of a high-voltage switch provided in this embodiment of the invention, the end-face grinding force is acquired in real time during the end-face grinding process of the stepped shaft part. The actual vibration influence intensity is determined by analyzing the end-face grinding force. This actual vibration influence intensity accurately characterizes the degree of vibration affecting the stepped shaft part during grinding. Based on this actual vibration influence intensity, the control parameters of the proportional-integral-derivative control are adjusted so that the control parameters can change in a timely manner with vibration changes, thereby effectively responding to vibration interference during the grinding process, improving the end-face grinding accuracy, and meeting the machining accuracy requirements of the stepped shaft of the high-voltage switch. Attached Figure Description

[0047] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A schematic flowchart illustrating a method for grinding the end face of a stepped shaft part for clamping a high-voltage switch, provided in an embodiment of the present invention.

[0049] Figure 2 This is a schematic diagram of an STL decomposition provided in an embodiment of the present invention. Detailed Implementation

[0050] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for grinding the end face of a stepped shaft part for clamping a high-voltage switch according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0052] In traditional high-voltage switch stepped shaft end face grinding, fixed proportional-integral-derivative control parameters are used for grinding process control. However, these control parameters cannot dynamically respond to vibration disturbances caused by friction and other factors during grinding, resulting in the end face dimensional accuracy, form and position tolerances, and surface properties failing to meet the operational reliability and safety requirements of high-voltage switchgear. Poor end face grinding quality leads to increased contact resistance and contact welding, which in turn affects the stepped shaft's function in high-voltage switch disconnection operations.

[0053] If the above problems are not resolved, insufficient grinding precision of the stepped shaft end face will directly lead to contact welding or arc reignition during high-voltage switch disconnection operations, threatening the stable operation of the power system and the safety of operators.

[0054] In this regard, such as Figure 1 As shown, this invention proposes a method for grinding the end face of a stepped shaft component for clamping a high-voltage switch. This method can be applied to electronic devices and includes the following steps S110 to S130:

[0055] S110, during the end face grinding process of the stepped shaft part, obtain the end face grinding force of the stepped shaft part;

[0056] S120, Analyze the end face grinding force of the stepped shaft part to determine the actual vibration influence intensity of the stepped shaft part; the actual vibration influence intensity is used to characterize the intensity of the vibration influence actually experienced by the stepped shaft part during the end face grinding process;

[0057] S130, based on the actual vibration influence intensity of the stepped shaft part, adjusts the control parameters of the proportional-integral-derivative control during the end face grinding process of the stepped shaft part; the control parameters are used to control the end face grinding of the stepped shaft part.

[0058] For ease of understanding, the following explains some key terms in this embodiment:

[0059] The stepped shaft component for clamping high-voltage switches refers to a shaft-like part with multiple diameter variations used in high-voltage switchgear. It requires a specialized fixture for fixation during the grinding process. The grinding accuracy of the end face of this stepped shaft component directly affects the operational reliability and safety of the high-voltage switch.

[0060] End face grinding force refers to the interaction force generated when the grinding wheel contacts the stepped shaft part during end face grinding. The magnitude and variation of this force can reflect the physical state of the grinding process, such as cutting conditions and vibration.

[0061] Actual vibration impact intensity refers to the degree of vibration disturbance actually experienced by a stepped shaft part during end face grinding. This actual vibration impact intensity is a quantitative indicator used to assess the potential impact of vibration on machining quality during grinding.

[0062] Proportional-Integral-Derivative (PID) control is a feedback control algorithm widely used in industrial control. It adjusts the control output by calculating the proportional, integral, and derivative terms to make the system output as close as possible to the target setpoint. In grinding, PID control is often used to precisely control process parameters such as grinding feed rate and spindle speed.

[0063] Control parameters refer to the various coefficients used to adjust the control effect in a proportional-integral-derivative (PID) control algorithm, mainly including proportional coefficient, integral coefficient, and derivative coefficient. The setting of these control parameters directly affects the response speed, stability, and error elimination capability of the control system.

[0064] This embodiment provides a method for grinding the end face of a stepped shaft component for clamping a high-voltage switch. This method aims to improve the grinding accuracy of the stepped shaft component's end face by dynamically adjusting control parameters during the grinding process to adapt to potential vibration interference.

[0065] Specifically, the method includes the following steps: during the end-face grinding of a stepped shaft part, acquiring the end-face grinding force. The end-face grinding force can be acquired in various ways. For example, force sensors, such as piezoelectric sensors or strain gauge sensors, can be installed on the fixture or spindle of the grinding machine to monitor the force acting on the workpiece in real time during grinding. These sensors can convert mechanical force into electrical signals and record them through a data acquisition system. Another approach is to indirectly infer the grinding force by monitoring the current or power consumption of the grinding spindle, as there is a correlation between the grinding force and the spindle load. These methods provide a continuous stream of grinding force data, laying the foundation for subsequent analysis.

[0066] Furthermore, the end-face grinding force of the stepped shaft part is analyzed to determine the actual vibration influence intensity of the stepped shaft part. This actual vibration influence intensity is used to characterize the intensity of the vibration influence actually experienced by the stepped shaft part during the end-face grinding process. Grinding force data usually contains vibration-related components. For example, vibration characteristics can be extracted by performing time-domain or frequency-domain analysis on the raw grinding force data. In the time domain, the root mean square value, peak-to-peak value, or standard deviation of the grinding force data within a specific time window can be calculated; these statistics can reflect the severity of force fluctuations. The greater the fluctuation, the stronger the vibration influence usually is. In the frequency domain, the spectrum of the grinding force data can be analyzed using methods such as Fourier transform to identify specific vibration frequencies and their amplitudes, thereby assessing the impact of vibration.

[0067] Based on this, and taking into account the actual vibration intensity of the stepped shaft part, the control parameters of the proportional-integral-derivative (PI-DE) control for the stepped shaft part during the end-face grinding process are adjusted. These control parameters are used to control the end-face grinding of the stepped shaft part. Traditional grinding processes often employ fixed-parameter PI-DE control, but when grinding conditions (such as workpiece material, tool wear, vibration, etc.) change, fixed parameters may not provide optimal control. This embodiment uses the acquired actual vibration intensity to dynamically adjust the PI-DE control parameters. By adjusting these control parameters, the feed rate, grinding depth, or spindle speed during the grinding process can be optimized, thereby effectively suppressing vibration and improving machining stability.

[0068] This embodiment acquires the end-face grinding force in real time during the end-face grinding process of the stepped shaft part. By analyzing the end-face grinding force, the actual vibration influence intensity is determined. This actual vibration influence intensity accurately characterizes the degree of vibration affecting the stepped shaft part during grinding. Based on this actual vibration influence intensity, the control parameters of the proportional-integral-derivative control are adjusted so that the control parameters can change in a timely manner with vibration changes, thereby effectively dealing with vibration interference during the grinding process, improving the end-face grinding accuracy, and meeting the machining accuracy requirements of the high-voltage switch stepped shaft.

[0069] In some embodiments of the present invention described above, during the end-face grinding process of a stepped shaft part, the end-face grinding force of the stepped shaft part is acquired, and the end-face grinding force is analyzed to determine the actual vibration influence intensity, thereby adjusting the control parameters of the proportional-integral-derivative control. However, in actual operation, grinding force data is often complex and susceptible to noise interference. Accurately and robustly extracting information that can effectively characterize the actual vibration influence intensity directly from the original grinding force data is a key challenge in achieving precise control. Currently, there is a lack of an effective and robust method to accurately extract key information that can characterize the actual vibration influence intensity from complex grinding force data, thus affecting the accuracy and timeliness of subsequent control parameter adjustments.

[0070] In this regard, the present invention further proposes that S120 includes:

[0071] The grinding force on the end face of the stepped shaft part is continuously rotated N times to form a grinding force analysis cycle; N is a positive integer.

[0072] Based on the maximum and minimum grinding force values ​​in each grinding force analysis cycle, a maximum value sequence and a minimum value sequence are constructed.

[0073] Obtain the target difference curve between the first envelope curve of the maximum value sequence and the second envelope curve of the minimum value sequence;

[0074] The actual vibration impact intensity of the stepped shaft component is determined by analyzing the target difference curve.

[0075] In this embodiment, the grinding force on the end face of the stepped shaft part during N consecutive rotations is defined as a grinding force analysis cycle. This aims to periodically divide the continuously collected grinding force data to facilitate subsequent feature extraction and analysis. By using the grinding force data of the stepped shaft part during N consecutive rotations as an analysis cycle, the variation pattern of the grinding force of the stepped shaft part within the complete rotation cycle can be captured, which is crucial for identifying vibration modes related to the rotation of the part.

[0076] Based on the maximum and minimum grinding force values ​​in each grinding force analysis cycle, a maximum value sequence and a minimum value sequence are constructed. The purpose is to extract key features reflecting the range of grinding force fluctuations—namely, the maximum and minimum values—from each grinding force analysis cycle. By constructing these maximum and minimum value sequences, the original continuous grinding force data can be simplified into two discrete sequences that more intuitively reflect the grinding force fluctuation trend, laying the foundation for subsequent envelope construction. For the grinding force data points collected within each grinding force analysis cycle, all data points can be directly traversed to find the maximum and minimum values. These maximum values ​​are then arranged in chronological order to form a maximum value sequence, and the minimum values ​​are arranged in chronological order to form a minimum value sequence.

[0077] The goal of obtaining the target difference curve between the first envelope curve of the maximum value sequence and the second envelope curve of the minimum value sequence is to further extract trend information of grinding force fluctuation amplitude through envelope line technology. The first envelope curve of the maximum value sequence and the second envelope curve of the minimum value sequence can respectively depict the upper and lower limits of grinding force fluctuation, while the target difference curve directly reflects the change in the grinding force fluctuation range over time. This change in fluctuation range is closely related to the actual vibration intensity of the stepped shaft part. Spline interpolation or moving average methods can be used to construct the envelope curve. For example, for the maximum value sequence, a smooth curve can be fitted using cubic spline interpolation as the first envelope curve; similarly, the minimum value sequence is processed to obtain the second envelope curve. Then, the difference between the points on the first envelope curve and the corresponding points on the second envelope curve is calculated to obtain the target difference curve. Specifically, when constructing the first and second envelope curves, the maximum and minimum value sequences are made equal in time dimension to ensure that the two envelope curves have perfectly matching time nodes. Thus, the points at each time node are the corresponding points between the first envelope curve and the second envelope curve.

[0078] Analyzing the target difference curve to determine the actual vibration impact intensity of the stepped shaft part is the core step. Through in-depth analysis of the target difference curve, the actual vibration impact intensity experienced by the stepped shaft part during grinding is quantified. The target difference curve contains dynamic information on the amplitude of grinding force fluctuations, and its shape, amplitude, and trend are closely related to the vibration state. Therefore, analyzing it can accurately characterize the actual vibration impact intensity.

[0079] This invention solves the problem of accurately and robustly extracting the actual vibration intensity from complex grinding force data through a series of refined data processing steps. First, the grinding force of the end face of the stepped shaft part during N consecutive rotations is defined as a grinding force analysis cycle. This periodic division effectively captures the grinding force variation pattern during the complete rotation process, providing a structured data foundation for subsequent vibration feature extraction, thereby reducing noise interference in the overall analysis. Second, based on the maximum and minimum grinding force values ​​in each grinding force analysis cycle, a maximum value sequence and a minimum value sequence are constructed. This operation transforms the original continuous grinding force data into discrete feature sequences that better reflect the fluctuation boundaries, effectively filtering out some high-frequency noise and highlighting the overall trend of grinding force fluctuations. Next, the target difference curve between the first envelope curve of the maximum value sequence and the second envelope curve of the minimum value sequence is obtained. The envelope curve smoothly depicts the upper and lower limits of grinding force fluctuations, while the difference curve between the two envelope curves directly quantifies the amplitude of grinding force fluctuations. This amplitude variation directly reflects the vibration state of the part and has a good suppression effect on random noise. Finally, the obtained target difference curve is analyzed in depth to accurately determine the actual vibration influence intensity of the stepped shaft part. Through the above steps, the present invention can systematically extract feature information highly correlated with the actual vibration influence intensity from complex grinding force data, overcoming the difficulty of directly analyzing the original signal. This provides a reliable basis for the accurate adjustment of subsequent proportional-integral-derivative control parameters, significantly improving the accuracy and stability of vibration influence intensity determination.

[0080] Through the above technical solution, this invention provides a more accurate and robust method for determining the actual vibration impact intensity. By periodically dividing the grinding force data and constructing an envelope curve based on the maximum and minimum values, noise interference can be effectively filtered out, and the true trend of grinding force fluctuations can be highlighted. This method makes extracting key vibration-related information from complex grinding force data more reliable, thus overcoming the limitations of traditional methods in handling noise and complex signals. Therefore, the determined actual vibration impact intensity can more accurately reflect the true vibration state of stepped shaft parts during end-face grinding, providing more precise input for subsequent adjustment of proportional-integral-derivative control parameters, thereby significantly improving the stability and machining quality of the grinding process.

[0081] In some embodiments of the present invention described above, the analysis of the target difference curve is proposed to determine the actual vibration impact intensity of the stepped shaft component. However, simply analyzing the target difference curve directly to determine the actual vibration impact intensity may lead to an inaccurate and unstable judgment of the vibration state, making it difficult to effectively distinguish the long-term trend and periodic characteristics of the vibration, thereby affecting the accuracy of subsequent control parameter adjustments.

[0082] To address this, the present invention further proposes analyzing the target difference curve to determine the actual vibration impact intensity of the stepped shaft component, including:

[0083] The target difference curve is decomposed into a seasonal trend based on local weighted regression to obtain the target trend term and the target seasonal term of the target difference curve.

[0084] Obtain the target fitted slope of the target trend term and the average amplitude of the target seasonal term;

[0085] The actual vibration impact intensity of the stepped shaft component is determined based on the target fitting slope and average amplitude.

[0086] In this embodiment, Seasonal Trend Decomposition using Loess (STL) based on Locally Weighted Regression is a statistical method that decomposes time series data into trend, seasonal, and residual components. Locally weighted regression, as a nonparametric regression technique, constructs a function describing the deterministic changes in the data by fitting a local subset of the data. This decomposition aims to separate the slowly changing long-term trend from the periodically repeating seasonal pattern in the target difference curve, thereby effectively filtering out random noise and irregular fluctuations, leading to a clearer understanding of the oscillation characteristics. As an example, such as... Figure 2 As shown, a schematic diagram of STL decomposition is provided. Figure 2 From top to bottom, the graph shows the original target difference curve and the trend term, seasonal term, and residual term obtained after STL decomposition.

[0087] The target trend term refers to the long-term, potential direction or movement of change in the target difference curve after removing seasonal and irregular components; it represents the gradual change of vibration effects over time. The target seasonal term refers to a predictable pattern in the target difference curve that repeats with a fixed period (e.g., each rotation cycle of a stepped shaft part); it represents the periodic component of vibration effects.

[0088] The target fit slope is a numerical value used to quantify the rate of change of the target trend term, indicating whether the trend term is rising, falling, or remaining stable. It can be obtained by performing linear regression analysis on the target trend term to obtain its slope, or by calculating the differences between consecutive points in the target trend term and averaging these differences over a specific time period. The average amplitude is an indicator that measures the typical amplitude of oscillations in the target seasonal term, quantifying the average degree of periodic change. It can be obtained by determining the peak and trough amplitudes of each period in the seasonal component and averaging these values.

[0089] Determining the actual vibration impact intensity of the stepped shaft component aims to comprehensively utilize information from the target fitting slope and average amplitude to derive an index that fully reflects the actual vibration state of the stepped shaft component. This ensures that the assessment of the vibration impact considers both its long-term evolution and its periodic fluctuations, providing a crucial basis for subsequent precise adjustment of control parameters. Specifically, the actual vibration impact intensity of the stepped shaft component can be determined using the following formula 1:

[0090] Formula 1

[0091] In formula 1, Used to characterize the actual vibration impact intensity of stepped shaft components. Used to characterize the average amplitude of the target seasonal term The target fitting slope is used to characterize the target trend term. Used to characterize activation functions.

[0092] The target fitting slope serves as the trend characteristic of the target difference curve, reflecting the stability of the grinding force. If the grinding force data shows no fluctuation, the trend characteristic of the target difference curve is 0; a negative trend characteristic indicates that the difference between the maximum and minimum values ​​of the grinding force is decreasing, suggesting that the fluctuation of the grinding force is weakening and gradually stabilizing; a positive trend characteristic indicates that the fluctuation of the grinding force is increasing. A larger average amplitude of the target seasonal term indicates a significant change in the grinding force during the processing, suggesting significant unevenness in the end face grinding, requiring adjustment of the grinding force in subsequent processing.

[0093] Building upon the aforementioned method of characterizing the vibration impact of stepped shaft parts by obtaining target difference curves, this invention, in order to more accurately identify and quantify vibration characteristics, first performs a seasonal trend decomposition of the target difference curves based on local weighted regression. This decomposition process effectively breaks down the complex information contained in the target difference curves into two more physically meaningful core components: a target trend term and a target seasonal term. The target trend term reveals the long-term, slow evolution of the vibration impact, while the target seasonal term captures vibration patterns periodically related to the rotation or grinding process of the stepped shaft parts. In this way, random noise and short-term disturbances are effectively separated, making the analysis of the vibration essence more focused and stable. Subsequently, the target fitting slope is extracted from the decomposed target trend term, which quantifies the long-term growth or decay rate of the vibration impact. Simultaneously, the average amplitude is obtained from the target seasonal term, which characterizes the intensity of the periodic vibration components. Finally, based on these two complementary quantitative indicators—the target fitting slope and the average amplitude—the actual vibration impact intensity of the stepped shaft parts is comprehensively determined. This multi-dimensional and refined analysis method avoids the inaccuracies that may arise from directly analyzing the original target difference curves, ensuring the comprehensiveness and robustness of the vibration state assessment. Therefore, the determined actual vibration intensity can more accurately reflect the true vibration state of the stepped shaft part during grinding, providing a solid data foundation for the precise adjustment of subsequent proportional-integral-derivative control parameters, thereby effectively improving the stability and processing quality of the end face grinding process.

[0094] By performing seasonal trend decomposition based on local weighted regression on the target difference curve, the present invention effectively separates the long-term trend of vibration impact from periodic fluctuations, thereby significantly reducing the interference of noise and short-term fluctuations in the original data on vibration intensity judgment. Based on the target fitting slope of the target trend term and the average amplitude of the target seasonal term, a more comprehensive, accurate, and robust quantitative assessment of the actual vibration impact intensity of stepped shaft parts can be performed. This refined method for determining vibration impact intensity provides a more reliable and accurate basis for subsequent adjustment of proportional-integral-derivative control parameters, enabling the control system to respond to actual vibration conditions more promptly and accurately, thereby effectively suppressing vibration during the grinding process and significantly improving the end-face grinding quality and stability of stepped shaft parts.

[0095] In some embodiments of the present invention described above, a method is proposed to acquire the end-face grinding force of a stepped shaft part during end-face grinding and analyze this force to determine the actual vibration impact intensity of the stepped shaft part. Based on the determined actual vibration impact intensity, the control parameters of the proportional-integral-derivative control are adjusted during the end-face grinding process of the stepped shaft part. However, in actual grinding, the actual vibration impact intensity over a single time period may not fully reflect the dynamic changes and trends of the grinding state, leading to insufficient precision or lag in the adjustment of control parameters, thereby affecting the stability and processing quality of the grinding process.

[0096] In response, this invention further proposes that the actual vibration influence intensity is multiple, and the control parameters include a proportionality coefficient; S130 includes:

[0097] Based on the influence intensity sequence composed of multiple actual vibration influence intensities, the first proportional adjustment parameter is determined.

[0098] Based on the first proportional adjustment parameter, the first proportional coefficient at the current moment is adjusted to obtain the second proportional coefficient.

[0099] In this embodiment, the actual vibration impact intensity is used to characterize the intensity of vibration actually experienced by the stepped shaft part during the end face grinding process. This actual vibration impact intensity reflects the vibration level experienced by the stepped shaft part during grinding and is an important basis for evaluating the grinding condition and guiding parameter adjustments.

[0100] Control parameters are adjustable variables in a proportional-integral-derivative (PID) controller, used to adjust the controller's output to achieve the desired control effect. The settings of these parameters directly affect the system's response speed, stability, and error elimination capability. The proportional gain is a core parameter in a PID controller, determining the strength of the controller's output response proportional to the current error. It's important to note that in PID control, to quickly offset the deviation caused by vibration, the proportional gain Kp plays a limited dominant role. Kp is used to quickly bring the grinding force close to the normal grinding force (because other PID parameters can only be fine-tuned to eliminate the negative impact of Kp). Therefore, the adjusted proportional gain Kp is determined based on the actual intensity of the vibration.

[0101] An influence intensity sequence refers to a sequence of multiple actual vibration influence intensity values ​​acquired consecutively over multiple time periods, arranged in chronological order. This influence intensity sequence reflects the trend and dynamic characteristics of actual vibration influence intensity changes over time, providing a data foundation for a more comprehensive assessment of grinding conditions and for predictive adjustments.

[0102] The first proportional adjustment parameter is an intermediate variable calculated based on the influence intensity sequence, used to guide the adjustment of the current first proportional coefficient. The determination of this first proportional adjustment parameter comprehensively considers historical vibration trends and the current vibration state, aiming to provide a more forward-looking and stable basis for adjustment.

[0103] The first proportional gain is the proportional gain at the current moment before adjustment; it is the proportional gain currently used by the PID controller. The second proportional gain is the new proportional gain obtained after adjustment based on the first proportional gain parameter. This new proportional gain will be used in subsequent proportional-integral-derivative control to achieve better grinding results.

[0104] The present invention overcomes the limitation of insufficient vibration intensity information in a single time period by dynamically adjusting the control parameters of proportional-integral-derivative control, particularly the proportional coefficient, by introducing multiple actual vibration influence intensities. Specifically, during the end face grinding of a stepped shaft part, the system continuously acquires and accumulates multiple actual vibration influence intensities. These continuous actual vibration influence intensities are organized into an influence intensity sequence, which not only contains the vibration information at the current moment but, more importantly, reflects the historical evolution trend of the vibration state. Based on this influence intensity sequence, the system can calculate a first proportional adjustment parameter. The calculation process of this first proportional adjustment parameter comprehensively considers the historical variation law of vibration intensity and the current state, enabling it to more accurately assess the vibration risk and trend during the grinding process. Subsequently, using this first proportional adjustment parameter, the first proportional coefficient currently in use is finely adjusted to obtain a new second proportional coefficient. In this way, the adjustment of the proportional coefficient no longer depends solely on the instantaneous vibration intensity, but is based on a comprehensive judgment of the vibration trend. This enables the control system to respond to changes in the grinding state more promptly and stably, avoiding excessive or delayed adjustments caused by fluctuations in a single data point, thereby improving the stability and machining accuracy of end face grinding.

[0105] Through the above technical solution, the adjustment of control parameters no longer relies solely on the actual vibration intensity over a single time period, but rather on a comprehensive judgment and dynamic adjustment based on a sequence of influence intensities composed of multiple actual vibration intensities. In particular, the adjustment of the proportional coefficient allows the proportional-integral-derivative control system to more comprehensively perceive the vibration state and its changing trends during the grinding process. This sequence-based adjustment mechanism effectively avoids misjudgments or delayed responses caused by instantaneous vibration fluctuations, thus making the adjustment of control parameters more precise and stable. Therefore, the solution of this invention can significantly improve the control accuracy and stability of the end face grinding process of stepped shaft parts, effectively suppress grinding vibration, and thereby improve machining quality and production efficiency.

[0106] In some embodiments of the present invention described above, a first proportional adjustment parameter is determined based on an influence intensity sequence composed of multiple actual vibration influence intensities. However, in actual grinding processes, no method is provided for determining the first proportional adjustment parameter based on the influence intensity sequence composed of multiple actual vibration influence intensities, resulting in insufficient precision or lag in the adjustment of the proportional coefficient, which affects the stability and processing quality of the grinding process.

[0107] To address this, the present invention further proposes determining a first proportional adjustment parameter based on an influence intensity sequence composed of multiple actual vibration influence intensities, including:

[0108] Based on the difference in influence intensity between two adjacent actual vibration influence intensities in the influence intensity sequence, an intensity difference sequence is constructed.

[0109] The historical control evaluation value is obtained by averaging the influence intensity differences in the intensity difference sequence.

[0110] Based on historical control evaluation values ​​and the target actual vibration influence intensity in the influence intensity sequence, the first proportional adjustment parameter is determined; the target actual vibration influence intensity is the last actual vibration influence intensity in the influence intensity sequence.

[0111] In this embodiment, constructing an intensity difference sequence aims to quantify the rate of change or trend of the actual vibration influence intensity over time. By calculating the difference between two adjacent actual vibration influence intensities in the influence intensity sequence, a series of values ​​reflecting the increase or decrease in vibration intensity can be obtained. For example, the actual vibration influence intensity can be continuously recorded over a period of time to form an ordered sequence, and then the difference between the previous value and the next value in the sequence can be calculated sequentially.

[0112] The purpose of averaging historical control evaluation values ​​is to eliminate short-term fluctuations and noise by averaging the influence intensity differences in the intensity difference sequence, thereby obtaining a more stable index that better represents the historical vibration trend. This averaging can be done using the arithmetic mean method, which involves summing all the differences in the intensity difference sequence and then dividing by the number of differences.

[0113] The determination of the first proportional adjustment parameter based on historical control evaluation values ​​and the actual vibration impact intensity of the target involves comprehensively judging the adjustment direction and magnitude of the proportional coefficient based on historical vibration trends and the latest vibration state (the actual vibration impact intensity of the target). The actual vibration impact intensity of the target is usually the most recently acquired value in the impact intensity sequence.

[0114] The first proportional adjustment parameter can be determined using the following formula 2:

[0115] Formula 2

[0116] In formula 2, The parameter A is used to characterize the first proportional adjustment parameter, and A is used to characterize the historical control evaluation value. Used to characterize the actual vibration impact intensity of the target. Used to characterize activation functions. This represents the normalization function, used to normalize... Dimensionlessness is performed to avoid the problem of dimensions in the input of the subsequent exponential function. For example, a minimax normalization method can be used, where the maximum and minimum values ​​are obtained based on the statistical analysis of the influence intensity sequence.

[0117] When the influence intensity sequence shows a weakening trend (i.e., A > 0), the more pronounced the decreasing trend and the smaller the intensity of the last actual vibration, the better the PID control effect. This means that in subsequent grinding force control processes, only the relevant parameters of the current grinding process need to be maintained. Conversely, if the influence intensity sequence shows no significant improvement over multiple time periods (i.e., A < 0), the adjustment range of the proportional coefficient will be larger.

[0118] This invention provides a refined strategy for determining the first proportional adjustment parameter based on the actual vibration influence intensity sequence. During end face grinding, the actual vibration influence intensity of the stepped shaft part is continuously monitored and acquired, and these are arranged in chronological order to form an influence intensity sequence. To gain a deeper understanding of the dynamic characteristics of vibration changes, this solution first constructs an intensity difference sequence by calculating the difference between two adjacent actual vibration influence intensities in the influence intensity sequence. This difference sequence intuitively reflects the increase or decrease of vibration intensity at various time points, thus revealing the instantaneous trend of vibration change. Subsequently, the influence intensity differences in the constructed intensity difference sequence are normalized, and the mean of all normalized results in the sequence is used as the historical control evaluation value. The normalization process effectively removes dimensions, avoiding the influence of dimensions on the subsequent exponential function input values ​​during the calculation process. The normalization method can employ a minimax normalization method, where the maximum and minimum values ​​are statistically derived from the intensity difference sequence.

[0119] Mean value processing effectively filters out random noise and short-term fluctuations that may occur during the grinding process, enabling historical control evaluation values ​​to more stably and accurately characterize the average vibration trend of the stepped shaft part over a period of time. Ultimately, this scheme combines this "historical control evaluation value," representing the historical vibration trend, with the latest vibration state—the last actual vibration influence intensity in the influence intensity sequence (the target actual vibration influence intensity)—to determine the first proportional adjustment parameter. This combination ensures that the determination of the proportional adjustment parameter comprehensively considers both the current state of vibration and its historical evolution trend, significantly improving the stability and accuracy of the control.

[0120] Through the above technical solution, this invention can more comprehensively and deeply analyze the vibration characteristics of stepped shaft parts during end face grinding. By constructing an intensity difference sequence and performing averaging, the system can not only sense the current vibration intensity but also accurately capture the changing trend and historical evolution of the vibration intensity, thereby obtaining a stable and representative historical control evaluation value. Combining this historical control evaluation value with the latest target actual vibration influence intensity makes the determination process of the first proportional adjustment parameter more intelligent and refined.

[0121] In some embodiments of the present invention described above, a first proportional adjustment parameter is determined based on an influence intensity sequence composed of multiple actual vibration influence intensities. A second proportional coefficient is then adjusted based on this first proportional adjustment parameter to obtain the first proportional coefficient at the current moment. However, in practical applications, how to effectively convert the first proportional adjustment parameter into a precise adjustment of the first proportional coefficient to ensure that the control system's response to vibration influences is both sensitive and stable is a key issue requiring further optimization.

[0122] To address this, the present invention further proposes adjusting the first proportional coefficient at the current moment based on the first proportional adjustment parameter to obtain the second proportional coefficient, including:

[0123] The second proportional adjustment parameter is obtained by performing an exponential function on the negative of the first proportional adjustment parameter.

[0124] The second proportional adjustment parameter is multiplied by the first proportional coefficient to obtain the second proportional coefficient.

[0125] In this embodiment, the nonlinear processing method of the exponential function allows for more flexible and precise adjustment of the proportional coefficient. It provides stronger adjustment capability, especially when the vibration intensity varies significantly, while offering smoother adjustment when the variation is small, thus avoiding over-adjustment or under-adjustment. The second proportional adjustment parameter is a multiplier factor processed by the exponential function.

[0126] Multiplying the second proportional adjustment parameter by the first proportional coefficient at the current moment directly yields the adjusted second proportional coefficient. This multiplicative adjustment method makes the adjustment of the proportional coefficient relative, that is, increasing or decreasing it as a percentage based on the current proportional coefficient, rather than increasing or decreasing its absolute value. This multiplicative adjustment method can maintain the dynamic range of the proportional coefficient, avoiding inconsistent adjustment effects at different magnitudes of proportional coefficients, and enabling the control system to maintain good adaptability under different operating conditions.

[0127] The first proportional coefficient at the current moment is the initial proportional coefficient of the PID control system for suppressing vibration during stepped shaft grinding, which can be obtained through calibration of the historical operating parameters of the PID control system.

[0128] Specifically, the second proportionality coefficient can be determined by the following formula 3:

[0129] Formula 3

[0130] In formula 3, Used to characterize the second proportionality coefficient Used to characterize the first proportionality coefficient Used to characterize the first proportional adjustment parameter Used to characterize the processing of exponential functions.

[0131] In cases of strong vibration during rotary grinding of stepped shaft end faces, it is necessary to reduce the proportional coefficient to minimize vibration and prevent further fluctuations in grinding force, which could exacerbate vibration during the machining process. Therefore, adjustments require sacrificing machining speed (e.g., reducing grinding force, performing multiple grinding passes on the same stepped shaft end face) to achieve stable grinding force and feed. Consequently, the proportional coefficient needs to be reduced in subsequent grinding control processes; the stronger the impact of vibration on the grinding process, the greater the reduction in the proportional coefficient.

[0132] The above technical solution transforms the first proportional adjustment parameter into a multiplier factor through an exponential function, and adjusts the first proportional coefficient in a multiplicative manner. This effectively solves the problems of inaccurate adjustment or sluggish response that may occur in traditional linear adjustment methods when dealing with complex vibration effects. This nonlinear adjustment mechanism makes the adjustment of the proportional coefficient more precise and dynamic, providing an appropriate adjustment range according to the degree of change in the intensity of vibration effects. This significantly improves the adaptability and robustness of the proportional-integral-derivative control system to vibration effects during the grinding of stepped shaft parts, ensuring the stability of the grinding process and the machining quality.

[0133] In some embodiments of the present invention described above, the control parameters of proportional-integral-derivative control are adjusted based on the actual vibration intensity of the stepped shaft part, and the method for adjusting the proportional coefficient is specifically described. However, in actual grinding processes, relying solely on the adjustment of the proportional coefficient may not be sufficient to cope with the complex dynamic response caused by grinding force fluctuations, especially when it is necessary to quickly suppress vibration or eliminate steady-state errors, which may lead to poor control performance or sluggish system response.

[0134] In this regard, the present invention further proposes that the control parameters also include differential coefficients;

[0135] Based on the first proportional adjustment parameter, after adjusting the first proportional coefficient at the current moment to obtain the second proportional coefficient, the following steps are also included:

[0136] Simulation control is performed based on the second proportional coefficient to obtain the simulation control evaluation value corresponding to the simulated grinding force of the second proportional coefficient;

[0137] Based on the simulated control evaluation value, the first differential coefficient at the current moment is adjusted to obtain the second differential coefficient.

[0138] In this embodiment, adjusting only the proportional gain of the PID parameters would slow down the system response and increase damping during grinding. Although the overshoot would decrease, the static deviation would increase, resulting in a longer time to reach a stable grinding state during end face grinding (i.e., the control process changes more slowly, requiring a longer time to achieve the control result). This means that a significant portion of the PID adjustment process would still be subject to vibration interference. Therefore, further adjustments to the remaining PID parameters are necessary.

[0139] The derivative coefficient is a crucial parameter in proportional-integral-derivative (PID) control, reflecting the influence of the controlled variable's rate of change on the control variable. By introducing the derivative term, the controller can predict the future trend of the controlled variable, thereby taking control actions in advance, effectively suppressing vibration, reducing overshoot, and accelerating the system's response speed. Properly setting the derivative coefficient is essential for improving the dynamic performance and stability of the control system.

[0140] Simulation control based on the second proportional coefficient refers to simulating the behavior of the control system after applying the second proportional coefficient by establishing a mathematical model, simulation system, or conducting small-scale, low-load test runs without actually performing physical grinding operations. The purpose of this simulation is to evaluate the impact of the currently adjusted proportional coefficient on the dynamic response of the system, especially its performance on key indicators such as grinding force and vibration, thereby providing a basis for subsequent adjustments to the differential coefficient.

[0141] The simulated control evaluation value corresponding to the simulated grinding force with the second proportional coefficient is obtained. The simulated grinding force refers to the virtual grinding force data obtained through model calculation or simulation during the simulated control process, based on the set second proportional coefficient. The simulated control evaluation value is then used to assess the intensity of vibration affecting the stepped shaft part under simulated conditions, based on these simulated grinding force data and following the same analytical method as historical control evaluation values. This allows for the prediction of the vibration suppression effect after adjusting the control parameters, even without actual grinding.

[0142] Based on the simulated control evaluation value, the first differential coefficient at the current moment is adjusted to obtain the second differential coefficient. This step refers to using the simulated control evaluation value obtained through simulated control as input to calculate or find new differential coefficients. Since the simulated control evaluation value reflects the dynamic characteristics of the system after applying the second proportional coefficient, the differential coefficients can be adjusted accordingly to further optimize the system's damping characteristics and response speed, thereby obtaining a better second differential coefficient.

[0143] The present invention, after adjusting the proportional coefficient to obtain a second proportional coefficient, does not directly apply it to actual grinding. Instead, it first performs simulation control based on this second proportional coefficient. Through simulation control, the system can predict the grinding force and the resulting vibration intensity under the current proportional coefficient. Specifically, the simulation control generates a series of simulated grinding force data, from which a simulation control evaluation value can be obtained. This simulation process avoids the processing quality problems or equipment damage risks that may occur in actual grinding due to improper parameter settings. Subsequently, the system uses this simulation control evaluation value as a basis for evaluating the current control effect, and then makes targeted adjustments to the first differential coefficient at the current moment to obtain a better second differential coefficient. This strategy of simulating first and then adjusting the differential coefficient allows the control system to more comprehensively consider the dynamic response after the proportional term adjustment. By introducing the differential term, vibration can be more accurately suppressed and predicted, thereby significantly improving the dynamic response speed and anti-interference ability of the grinding process while ensuring system stability. This effectively solves the problem of poor control effect or response lag that may occur when relying solely on proportional coefficient adjustment.

[0144] The above technical solution introduces a mechanism for further adjusting the derivative coefficient based on analog control after adjusting the proportional coefficient. This mechanism allows the system to pre-evaluate the dynamic effect of the proportional coefficient adjustment without affecting actual production, and to finely adjust the derivative coefficient accordingly. This not only avoids the machining quality risks that may arise from trial and error in actual grinding, but also significantly enhances the control system's ability to suppress transient vibrations and grinding force fluctuations during grinding by introducing and optimizing the derivative term. Therefore, this solution can effectively improve the machining accuracy and surface quality of the end face grinding of stepped shaft parts, while accelerating the system's response speed and ensuring the stability and efficiency of the grinding process, thus overcoming the control limitations that may exist when relying solely on proportional coefficient adjustment.

[0145] In some embodiments of the present invention described above, a scheme for adjusting the differential coefficients based on simulated control evaluation values ​​is proposed. However, in actual grinding processes, simply relying on simulated control evaluation values ​​for adjustment may result in insufficient precision in the adjustment of the differential coefficients, affecting the response speed and stability of the control system. Especially when facing complex and variable grinding conditions, it may not be able to effectively suppress vibration.

[0146] To address this, the present invention further proposes adjusting the first differential coefficient at the current moment based on the simulated control evaluation value to obtain the second differential coefficient, including:

[0147] The target fluctuation value is obtained by multiplying the difference between the simulated control evaluation value and the historical control evaluation value by the historical control evaluation value.

[0148] The target fluctuation value is processed by an exponential function to obtain the differential adjustment parameter;

[0149] Multiply the differential adjustment parameter by the first differential coefficient to obtain the second differential coefficient.

[0150] In this embodiment, the difference between the simulated control evaluation value and the historical control evaluation value is a key indicator for measuring the deviation between the simulated control effect and the current actual grinding state. It reflects the gap between the vibration level predicted by the system under the second proportional coefficient and the vibration level currently actually monitored.

[0151] Multiplying the aforementioned difference by the historical control evaluation value aims to weight this difference. By multiplying by the historical control evaluation value, the target fluctuation value can more sensitively reflect the deviation when the current actual vibration level is high, thus giving greater adjustment weight when the vibration is large, ensuring that the control system responds more promptly and effectively to the current actual vibration condition. The resulting target fluctuation value is a composite index that comprehensively considers the difference between simulated and actual vibration as well as the current actual vibration level. It quantifies the degree and direction of adjustment to the differential control required under the current grinding condition.

[0152] Applying an exponential function to the target fluctuation value allows for a nonlinear mapping of the target fluctuation value to the differential control parameter. The advantage of this nonlinear processing is that when the target fluctuation value is small, the change in the differential control parameter is also small, maintaining system stability; while when the target fluctuation value is large, the change in the differential control parameter increases significantly, enabling the system to respond quickly and suppress severe vibrations. This helps improve the robustness and adaptability of the control system. The resulting differential control parameter is a multiplicative factor used to adjust the current differential coefficients, directly determining the final value of the differential coefficients.

[0153] Finally, by multiplying the differential adjustment parameter by the first differential coefficient, dynamic adjustment of the differential coefficient can be achieved. This multiplicative adjustment method can maintain the proportional relationship between the differential coefficient and the inherent characteristics of the system, while making flexible corrections based on the actual vibration conditions, thereby obtaining a second differential coefficient that is more suitable for the current grinding state.

[0154] The first differential coefficient at the current moment is the initial differential coefficient of the PID control system for suppressing vibration during stepped shaft grinding, which can be obtained through calibration of the historical operating parameters of the PID control system.

[0155] Specifically, the second differential coefficient can be determined by the following formula 4:

[0156] Formula 4

[0157] In formula 4, Used to characterize the second differential coefficient Used to characterize the first differential coefficient Used to characterize historical control evaluation values The exp function is used to characterize the evaluation value of the simulation control.

[0158] Among them, if This indicates that the simulated control evaluation value is significantly weaker than before the adjustment, suggesting that the current adjustment of the proportional coefficient has significantly reduced the fluctuation of the grinding force data, resulting in higher stability of the grinding process. In this case, the effect of the derivative coefficient can be reduced.

[0159] Conversely, if The result indicates that the simulated control evaluation value has significantly increased compared to before the adjustment, suggesting that the derivative coefficient is insufficient. This means that the grinding process cannot promptly suppress overshoot after it occurs, leading to large amplitude fluctuations in the grinding force. In this case, the derivative coefficient should be increased to prevent further expansion of the vibration amplitude caused by changes in the proportional coefficient.

[0160] Basic deviation Multiply by historical control evaluation value The target fluctuation value is obtained. The more severe the historical vibration, the more significant the impact of the deviation on the adjustment. This avoids weak adjustment when the vibration is large or over-adjustment when the vibration is small. The nonlinear characteristics of the exponential function make the adjustment amplitude change exponentially with the increase of the deviation, which is suitable for the continuous and non-uniform variation characteristics of grinding vibration.

[0161] Through the above technical solution, this invention can accurately quantify the difference between simulated vibration and actual vibration, and perform weighted processing according to the current actual vibration level, making the adjustment of the differential coefficients more precise and timely. The introduction of exponential function processing further ensures that the control system can exhibit nonlinear adaptive capability when facing vibrations of different degrees, that is, it remains stable during small vibrations and responds rapidly during large vibrations. This refined and adaptive differential coefficient adjustment method significantly improves the vibration suppression effect of the proportional-integral-derivative control system in the end face grinding process of stepped shaft parts, effectively solving the problems of inaccurate differential coefficient adjustment, system response hysteresis, or overshoot in traditional methods.

[0162] In some embodiments of the present invention described above, the grinding force on the end face of the stepped shaft part is obtained and its vibration influence intensity is analyzed. The proportional and derivative coefficients in the proportional-integral-derivative control are then dynamically adjusted to adapt to changes during the grinding process. However, in actual grinding control, relying solely on adjusting the proportional and derivative coefficients may not completely eliminate the steady-state error of the system, or it may be difficult to avoid overshoot while maintaining a rapid response, thus affecting grinding accuracy and efficiency.

[0163] In this regard, the present invention further proposes that the control parameters also include integral coefficients;

[0164] Based on the simulated control evaluation value, after adjusting the first differential coefficient at the current moment to obtain the second differential coefficient, the following steps are also included:

[0165] Based on the second proportional coefficient and the second differential coefficient, the first integral coefficient at the current moment is adjusted to obtain the second integral coefficient.

[0166] In this embodiment, the integral coefficient is an important parameter in proportional-integral-derivative (PID) control, and its main function is to eliminate the steady-state error of the system. By integrating the error signal, the integral term can accumulate past errors and generate a control action, enabling the system output to accurately track the setpoint, thereby improving control accuracy. During end-face grinding, the introduction of the integral coefficient helps ensure that the grinding force or vibration intensity remains stable at the desired level, avoiding long-term deviations.

[0167] After the proportional and derivative coefficients have been dynamically adjusted based on the intensity of vibration during the grinding process, the integral coefficient adjustment is no longer done in isolation, but rather comprehensively considers the currently adjusted proportional and derivative coefficients. This adjustment method ensures a coordinated balance among the three parameters of the PID controller, avoiding control instability or performance degradation that may result from independent adjustments to a single parameter. One possible implementation is to pre-establish a parameter adjustment rule base or lookup table. This rule base recommends or directly provides the corresponding second integral coefficient based on different combinations of the second proportional and second derivative coefficients.

[0168] The present invention, based on the second proportional and second derivative coefficients already adjusted according to the simulation control evaluation values, further introduces and adjusts the integral coefficient. Specifically, after obtaining the second proportional and second derivative coefficients, the system uses these two adjusted parameters as inputs to calculate and determine the optimal second integral coefficient for the current moment. This strategy of adjusting the integral coefficient based on the adjusted proportional and derivative coefficients creates a coordinated and dynamically balanced adjustment mechanism among the three core parameters (proportional, integral, and derivative) of the PID controller. In this way, the adjustment of the integral coefficient is no longer independent but closely related to the transient response characteristics of the system (determined by the proportional and derivative coefficients). This holistic parameter adjustment method ensures that the PID controller maintains optimal control performance throughout the entire end-face grinding process, not only responding quickly to changes in grinding force and suppressing vibration, but also effectively eliminating long-term steady-state errors, making the grinding process more stable and precise.

[0169] Through the above technical solution, when performing end face grinding on stepped shaft parts, not only can the proportional and derivative coefficients be dynamically adjusted according to the actual vibration intensity to cope with transient changes, but furthermore, by coordinating the adjustment of the integral coefficient based on the adjusted second proportional and second derivative coefficients, the proportional-integral-derivative controller can achieve more precise and comprehensive control. This coordinated adjustment mechanism effectively solves the problem that steady-state errors may be difficult to eliminate or control performance may be poor if only the proportional and derivative coefficients are adjusted. It significantly improves the stability and accuracy of the grinding process, ensures the quality of end face grinding of stepped shaft parts, and reduces the scrap rate.

[0170] In some of the above embodiments, the vibration effects during the grinding process have been addressed by adjusting the proportional and derivative coefficients. However, in actual grinding operations, simply adjusting the proportional and derivative terms may not completely eliminate the accumulation of systematic errors, leaving room for improvement in control accuracy and stability, especially in precision grinding scenarios requiring long-term stable control.

[0171] To address this, the present invention further proposes adjusting the first integral coefficient at the current moment based on the second proportional coefficient and the second differential coefficient to obtain the second integral coefficient, including:

[0172] The difference between the first proportional coefficient and the second proportional coefficient is divided by the first proportional coefficient to obtain the proportional adjustment degree.

[0173] The difference between the first differential coefficient and the second differential coefficient is divided by the first differential coefficient to obtain the differential adjustment degree.

[0174] Based on the proportional adjustment degree and the differential adjustment degree, the first integral coefficient is adjusted to obtain the second integral coefficient.

[0175] In this embodiment, the difference between the first proportional coefficient and the second proportional coefficient is divided by the first proportional coefficient to obtain the proportional adjustment degree. This technical feature aims to quantify the relative change of the proportional control term before and after adjustment. The proportional adjustment degree can reflect the adjustment range experienced by the proportional coefficient within the current control cycle. Its function is to provide a standardized, dimensionless index for subsequent adjustment of the integral coefficient.

[0176] The difference between the first and second differential coefficients, divided by the first differential coefficient, yields the differential adjustment degree. This technical feature quantifies the relative change of the differential control term before and after adjustment. The differential adjustment degree reflects the adjustment magnitude experienced by the differential coefficients within the current control cycle, and its function is to provide a standardized, dimensionless index for subsequent adjustment of the integral coefficients.

[0177] Based on the proportional and derivative adjustment degrees, the first integral coefficient is adjusted to obtain the second integral coefficient. This technical feature describes how to dynamically adjust the integral coefficient using these two indicators. The integral coefficient is mainly used to eliminate the steady-state error of the system, and its adjustment is crucial for improving the long-term stability and accuracy of the control system. By comprehensively considering the adjustments of the proportional and derivative terms, the integral term can be optimized more precisely, making it better adapt to the dynamic changes during the grinding process.

[0178] The first integral coefficient at the current moment is the initial integral coefficient of the PID control system for suppressing vibration during stepped shaft grinding, which can be obtained through calibration of the historical operating parameters of the PID control system.

[0179] Specifically, the second integral coefficient can be determined using the following formula 5:

[0180] Formula 5

[0181] In formula 5, Used to characterize the second integral coefficient Used to characterize the first integral coefficient. Used to characterize the second differential coefficient Used to characterize the first differential coefficient Used to characterize the second proportionality coefficient Used to characterize the first proportionality coefficient.

[0182] When Kp is large, the system's open-loop gain is high, and the proportional action can quickly offset most of the grinding force deviation. Even if the integral action is slightly stronger, the proportional action can suppress the fluctuations caused by the integral, resulting in high system tolerance to integral forces. When Kp decreases, the system's open-loop gain decreases. If Ti is small (strong integral action), even small fluctuations in the grinding force (e.g., ±1%) will accumulate and transform into reciprocating abrupt changes in the grinding tool feed, leading to significant vibrations during grinding. Therefore, when Kp decreases, it is necessary to increase the integral Ti to weaken the integral action and avoid large fluctuations in grinding force due to integral oscillations.

[0183] As Td increases, the differential component's ability to suppress sudden changes in grinding force is enhanced (ensuring stable grinding force during end face grinding): before the grinding force exceeds the limit, the differential Td outputs a reverse adjustment (reducing the feed), preemptively curbing overshoot in the grinding force. At this point, the system's tolerance to integral action is significantly improved, and Ti can be moderately reduced (enhancing the integral), preventing vibration and accelerating the elimination of steady-state deviations. Conversely, if Td decreases while Ti is too small (stronger integral action), the PID control system will fall into a vicious cycle where grinding force fluctuations cannot be suppressed in advance, and the integral amplifies residual grinding force fluctuations. For example, small, stable fluctuations in grinding force are accumulated by the integral, transforming into sudden feed changes, which the differential cannot compensate for, thus amplifying the vibrational impact of the grinding force. Therefore, it is necessary to further increase the integral Ti to weaken the integral's effect.

[0184] Therefore, the greater the decrease in Kp, the greater the increase in the integral Ti; an increase in Td requires a decrease in Ti, and vice versa.

[0185] This invention, building upon the existing dynamic adjustment of the proportional and derivative coefficients, further introduces a refined adjustment mechanism for the integral coefficient. Specifically, by calculating the relative change in the proportional coefficient before and after adjustment (proportional adjustment degree) and the relative change in the derivative coefficient before and after adjustment (derivative adjustment degree), the contributions of the proportional and derivative terms to the system's dynamic response are quantified. These adjustment degrees reflect the system's immediate response to changes in the intensity of actual vibration. Subsequently, based on these two adjustment degrees, the system can comprehensively evaluate the adjustment trends and magnitudes of the proportional and derivative terms under the current control state, thereby making targeted corrections to the integral coefficient. This correction is not performed in isolation but is closely related to the adjustment of the proportional and derivative coefficients, forming a coordinated PID parameter adaptive adjustment closed loop. In this way, the integral term can better compensate for any steady-state errors or cumulative deviations that may exist after the proportional and derivative adjustments, enabling the entire PID controller to adapt more comprehensively and stably to the complex vibration environment during the grinding process. This significantly improves steady-state control accuracy and long-term operational stability while ensuring dynamic response speed. This linkage adjustment mechanism enables the three parameters of the PID controller to work together to optimize the control effect of the grinding process, effectively solving the limitations that may be caused by adjusting a single parameter.

[0186] Through the above technical solution, this invention, based on the dynamic adjustment of the proportional and derivative coefficients, further achieves refined adaptive adjustment of the integral coefficient. This integral coefficient adjustment mechanism based on the proportional and derivative adjustment degrees enables the PID controller to more comprehensively respond to dynamic changes and accumulated errors during the grinding process. Specifically, it can effectively compensate for transient or steady-state deviations that may arise from the adjustment of the proportional and derivative terms, avoiding the long-term accumulation of system errors, thereby significantly improving the control accuracy and long-term stability of grinding. Especially in high-precision, long-duration grinding operations, this solution can ensure that the end face grinding process of stepped shaft parts always remains in the optimal control state, reducing the scrap rate and improving the processing quality.

[0187] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0188] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for grinding the end face of a stepped shaft part for clamping a high-voltage switch, characterized in that, The method includes: During the end face grinding process of the stepped shaft part, the end face grinding force of the stepped shaft part is obtained; The end face grinding force of the stepped shaft part is analyzed to determine the actual vibration influence intensity of the stepped shaft part; the actual vibration influence intensity is used to characterize the intensity of the vibration influence actually experienced by the stepped shaft part during the end face grinding process; Based on the actual vibration impact intensity of the stepped shaft part, the control parameters of the proportional-integral-derivative control of the stepped shaft part during the end face grinding process are adjusted; the control parameters are used to control the end face grinding of the stepped shaft part. The analysis of the end face grinding force of the stepped shaft part to determine the actual vibration influence intensity of the stepped shaft part includes: The grinding force on the end face of the stepped shaft part, which is rotated N times continuously, constitutes one grinding force analysis cycle; N is a positive integer; Based on the maximum and minimum grinding force values ​​in each grinding force analysis cycle, a maximum value sequence and a minimum value sequence are constructed. Obtain the target difference curve between the first envelope curve of the maximum value sequence and the second envelope curve of the minimum value sequence; The actual vibration impact intensity of the stepped shaft component is determined by analyzing the target difference curve. The analysis of the target difference curve to determine the actual vibration impact intensity of the stepped shaft component includes: The target difference curve is decomposed into a seasonal trend based on local weighted regression to obtain the target trend term and the target seasonal term of the target difference curve. Obtain the target fitting slope of the target trend term and the average amplitude of the target seasonal term; Based on the target fitting slope and the average amplitude, the actual vibration impact intensity of the stepped shaft component is determined; The number of actual vibration impact intensities is multiple, and the control parameters include a proportionality coefficient; The adjustment of the control parameters of the proportional-integral-derivative control for the stepped shaft part during the end face grinding process, based on the actual vibration influence intensity of the stepped shaft part, includes: Based on the influence intensity sequence composed of multiple actual vibration influence intensities, a first proportional adjustment parameter is determined; Based on the first proportional adjustment parameter, the first proportional coefficient at the current moment is adjusted to obtain the second proportional coefficient; The determination of the first proportional adjustment parameter based on the influence intensity sequence composed of multiple actual vibration influence intensities includes: Based on the difference in influence intensity between two adjacent actual vibration influence intensities in the influence intensity sequence, an intensity difference sequence is constructed. The historical control evaluation value is obtained by averaging the influence intensity differences in the intensity difference sequence. Based on the historical control evaluation value and the target actual vibration influence intensity in the influence intensity sequence, the first proportional adjustment parameter is determined; the target actual vibration influence intensity is the last actual vibration influence intensity in the influence intensity sequence. The step of adjusting the first proportional coefficient at the current moment based on the first proportional adjustment parameter to obtain the second proportional coefficient includes: The second proportional adjustment parameter is obtained by performing an exponential function on the negative of the first proportional adjustment parameter. The second proportional adjustment parameter is multiplied by the first proportional coefficient to obtain the second proportional coefficient.

2. The end face grinding method for the stepped shaft part for clamping high-voltage switches according to claim 1, characterized in that, The control parameters also include differential coefficients; After adjusting the first proportional coefficient at the current moment based on the first proportional adjustment parameter to obtain the second proportional coefficient, the method further includes: Based on the second proportional coefficient, simulation control is performed to obtain the simulation control evaluation value corresponding to the simulated grinding force of the second proportional coefficient; Based on the simulated control evaluation value, the first differential coefficient at the current moment is adjusted to obtain the second differential coefficient.

3. The end face grinding method for the stepped shaft part for clamping high-voltage switches according to claim 2, characterized in that, The adjustment of the first differential coefficient at the current moment based on the simulated control evaluation value to obtain the second differential coefficient includes: The target fluctuation value is obtained by multiplying the difference between the simulated control evaluation value and the historical control evaluation value by the historical control evaluation value. The target fluctuation value is processed by an exponential function to obtain the differential adjustment parameter; The second differential coefficient is obtained by multiplying the differential adjustment parameter by the first differential coefficient.

4. The end face grinding method for the stepped shaft part for clamping high-voltage switches according to claim 3, characterized in that, The control parameters also include integral coefficients; After adjusting the first differential coefficient at the current moment based on the simulated control evaluation value to obtain the second differential coefficient, the process further includes: Based on the second proportional coefficient and the second differential coefficient, the first integral coefficient at the current moment is adjusted to obtain the second integral coefficient.

5. The end face grinding method for the stepped shaft part for clamping high-voltage switches according to claim 4, characterized in that, The step of adjusting the first integral coefficient at the current moment based on the second proportional coefficient and the second differential coefficient to obtain the second integral coefficient includes: The difference between the first proportional coefficient and the second proportional coefficient is divided by the first proportional coefficient to obtain the proportional adjustment degree. The difference between the first differential coefficient and the second differential coefficient is divided by the first differential coefficient to obtain the differential adjustment degree. Based on the proportional adjustment degree and the differential adjustment degree, the first integral coefficient is adjusted to obtain the second integral coefficient.

Images