A spindle dynamic stiffness test analysis method and device

Abstract

The present application relates to the field of main shaft dynamic stiffness test, especially to a main shaft dynamic stiffness test analysis method and device. The method comprises the following steps: performing multi-working condition operation excitation test on the main shaft, and collecting main shaft test response parameters of multiple working conditions based on a multi-dimensional high-precision sensor array; performing dynamic noise cleaning processing and adaptive filtering optimization on the main shaft test response parameters to generate adaptive filtering optimized response parameters; performing dynamic test deviation prediction on the adaptive filtering optimized response parameters, and then performing multiple rounds of deviation iteration compensation optimization to generate dynamic deviation optimized parameters; performing transient dynamic stiffness peak value calculation on the dynamic deviation optimized parameters, and then performing long-term running dynamic stiffness change trend analysis to construct dynamic stiffness characteristic curves under multiple working conditions. The present application realizes accurate and efficient mechanical dynamic stiffness change characteristic test analysis.

Description

Technical Field

[0001] This invention relates to the field of spindle dynamic stiffness testing, and more particularly to a spindle dynamic stiffness testing and analysis method and apparatus. Background Technology

[0002] With the continuous development of high-precision machining technology, the spindle, as a key component in CNC machine tools, precision machining equipment, and other industrial equipment, directly affects the machining accuracy and production efficiency of products. Under high-load, high-speed rotation conditions, the dynamic stiffness of the spindle (i.e., its stiffness performance under dynamic loads) has become one of the important indicators for measuring its performance. The level of dynamic stiffness not only determines the vibration and deformation characteristics of the spindle during machining, but also directly relates to machining quality, equipment lifespan, and production efficiency.

[0003] Unlike traditional static stiffness testing methods, dynamic stiffness testing needs to be conducted under high-speed spindle rotation and dynamic loads. Considering the complex behavior and operating states of the spindle under different working conditions, the requirements for testing methods and techniques become more complex. Especially under high speed, high load, and complex operating conditions, the dynamic performance of the spindle is easily affected by temperature, vibration, mechanical wear, and other external factors, making it particularly difficult to accurately assess the dynamic stiffness of the spindle.

[0004] Currently, traditional methods for testing the dynamic stiffness of spindles mostly rely on simple sensor measurements, such as accelerometers and displacement sensors. These methods typically provide only single-dimensional data and cannot comprehensively reflect the dynamic performance of the spindle under different operating conditions. Furthermore, existing testing technologies often exhibit significant errors when faced with dynamic changes and external disturbances, making it difficult to effectively guarantee the accuracy and reliability of the test results. Especially during long-term operation and in complex environments, the health status and dynamic stiffness changes of the spindle require continuous monitoring, but traditional methods often lack real-time performance and accuracy, failing to effectively capture the dynamic characteristics of the spindle under different loads, speeds, temperatures, and other conditions.

[0005] With the continuous development of intelligent manufacturing, precision machining, and automation technologies, the demand for spindle dynamic stiffness testing is becoming increasingly complex and diverse. Especially driven by emerging technologies such as the Industrial Internet, the Internet of Things, big data, and artificial intelligence, traditional testing methods alone can no longer meet the requirements for high precision, real-time performance, and multi-dimensional analysis. Therefore, there is an urgent need for an intelligent spindle dynamic stiffness testing and analysis method based on advanced algorithms to achieve accurate measurement, anomaly detection, and predictive maintenance of spindle dynamic performance. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention proposes a method and apparatus for testing and analyzing the dynamic stiffness of a spindle, thereby resolving at least one of the aforementioned technical problems.

[0007] To achieve the above objectives, this invention provides a method for testing and analyzing the dynamic stiffness of a spindle. The spindle is placed on a dynamic loading test platform, which is equipped with a multi-dimensional high-precision sensor array. The method includes the following steps: Step S1: Perform multi-condition operation excitation test on the spindle, and collect spindle test response parameters for multiple conditions based on a multi-dimensional high-precision sensor array; Step S2: Perform dynamic noise cleaning and adaptive filtering optimization on the spindle test response parameters to generate adaptive filtering optimized response parameters; Step S3: Perform dynamic test deviation prediction on the adaptive filter optimization response parameters, and then perform multiple rounds of deviation iterative compensation optimization to generate dynamic deviation optimization parameters; Step S4: Calculate the peak value of transient dynamic stiffness for the dynamic deviation optimization parameters, then analyze the long-term dynamic stiffness change trend, and construct dynamic stiffness characteristic curves under multiple working conditions; Step S5: Acquire thermal imaging images of multi-condition operation excitation test, and perform quantitative correlation analysis based on the dynamic stiffness characteristic curve to obtain the dynamic stiffness influence law of thermal expansion. Step S6: Based on the dynamic stiffness influence law, predict the change in thermal expansion stiffness, and perform thermal expansion effect stripping processing to generate a spindle dynamic stiffness characteristic report.

[0008] This invention simulates the spindle's working state under various operating conditions by conducting multi-condition operational excitation tests. These conditions include different loads, speeds, and accelerations. This step allows for comprehensive acquisition of the spindle's response data under multiple operating conditions, providing rich data support for subsequent analysis. Utilizing a multi-dimensional, high-precision sensor array (such as accelerometers, displacement sensors, and strain gauges) to acquire test response parameters enables precise capture of the spindle's dynamic behavior in different directions and at different frequencies. The multi-dimensional configuration of the sensors improves the comprehensiveness and accuracy of data acquisition, making the analysis results more reliable. During spindle testing, external environmental factors (such as vibration, temperature fluctuations, and electromagnetic interference) may introduce noise signals. Dynamic noise cleaning effectively filters out these environmental noises, retaining key test response data and improving data quality. Adaptive filtering technology automatically adjusts filter parameters based on data characteristics, thereby optimizing signal quality. Unlike traditional filtering methods with fixed parameters, adaptive filtering can flexibly adjust according to changing conditions in actual testing, effectively removing high-frequency noise and low-frequency interference, making the data more accurate and reliable. During testing, sensor accuracy limitations or equipment errors may lead to deviations in test data. By dynamically predicting deviations in the response parameters optimized through adaptive filtering, these errors can be identified in advance, allowing for appropriate compensation measures. This process reduces systematic errors in testing and improves data accuracy. Employing a multi-round iterative deviation compensation optimization method, deviations can be gradually reduced through repeated optimization and adjustment, ultimately yielding more accurate test results. The iterative optimization process can gradually eliminate error sources and improve the reliability of the final test data. The peak transient dynamic stiffness is a crucial indicator of the spindle's operating condition. By accurately calculating the peak dynamic stiffness, we can effectively identify the spindle's stiffness performance under instantaneous loads and detect potential anomalies, such as resonance or excessively low local stiffness. Long-term analysis of dynamic stiffness variation trends under multiple operating conditions can identify performance changes in the spindle over long-term use. This helps predict the spindle's fatigue life, optimize usage plans, and provide a basis for subsequent maintenance and fault prediction. During high-speed spindle operation, temperature changes cause thermal expansion, which in turn affects dynamic stiffness. By acquiring thermal imaging images, the temperature distribution of the spindle can be monitored in real time, revealing the impact of temperature changes on spindle stiffness. Through quantitative correlation analysis between thermal imaging data and dynamic stiffness characteristic curves, the specific impact of temperature changes on spindle stiffness can be accurately calculated. This step helps to understand the influence of thermal expansion on spindle performance and provides data support for further optimization of spindle design and operating conditions. Based on the dynamic stiffness influence of thermal expansion, the dynamic stiffness changes caused by temperature variations under different operating conditions can be predicted. This prediction can help engineers perform temperature control and structural optimization during the design phase to reduce the negative impact of thermal expansion on spindle performance.By stripping away the thermal expansion effect, accurate characteristics of mechanical dynamic stiffness variation are obtained. This report provides crucial reference for long-term spindle performance evaluation, design optimization, and failure prediction. The detailed analysis results in the report help users understand the spindle's health status, service life, and maintenance requirements.

[0009] This specification provides a spindle dynamic stiffness testing and analysis apparatus for performing the spindle dynamic stiffness testing and analysis method described above, comprising: The multi-condition testing module is used to perform multi-condition operation excitation tests on the spindle and collect spindle test response parameters for multiple conditions based on a multi-dimensional high-precision sensor array. The adaptive filtering module is used to perform dynamic noise cleaning and adaptive filtering optimization on the spindle test response parameters, and generate adaptive filtering optimized response parameters. The deviation compensation module is used to dynamically test and predict the deviation of the adaptive filtering optimization response parameters, and then perform multiple rounds of deviation iterative compensation optimization to generate dynamic deviation optimization parameters. The dynamic stiffness characteristic module is used to calculate the peak value of transient dynamic stiffness for dynamic deviation optimization parameters, and then to analyze the long-term dynamic stiffness change trend to construct dynamic stiffness characteristic curves under multiple working conditions. The thermal expansion correlation module is used to acquire thermal imaging images of multi-condition operation excitation tests and perform quantitative correlation analysis based on the dynamic stiffness characteristic curve to obtain the dynamic stiffness influence law of thermal expansion. The thermal expansion stripping module is used to predict the change in thermal expansion stiffness based on the dynamic stiffness influence law, and to perform thermal expansion effect stripping processing to generate a spindle dynamic stiffness characteristic report.

[0010] This invention utilizes multi-condition testing to cover the spindle's performance under various loads and speeds, providing abundant experimental data to facilitate a comprehensive understanding of the spindle's dynamic performance. A high-precision sensor array ensures accurate acquisition of spindle response data under different operating conditions, reducing signal noise and errors and improving test accuracy. This multi-dimensional data provides a foundation for subsequent analyses (such as dynamic filtering and dynamic stiffness analysis), contributing to more accurate performance evaluation. Dynamic noise removal effectively identifies and removes unnecessary signals such as environmental noise and mechanical interference, improving signal quality. An adaptive filtering algorithm automatically adjusts filtering parameters based on signal variation characteristics under different operating conditions, ensuring optimal signal processing under varying conditions and reducing unnecessary signal loss. The filtered and optimized response data is more accurate, providing high-quality input data for subsequent dynamic deviation prediction and dynamic stiffness analysis. Real-time prediction and compensation for potential deviations during testing continuously optimizes data and reduces error accumulation. Multi-round iterative deviation compensation ensures data quality in each round of testing, significantly enhancing the reliability of spindle test results. This process helps eliminate test errors caused by environmental fluctuations and equipment instability, resulting in more accurate and reliable final test results. Transient dynamic stiffness peak calculations can assess the spindle's stiffness performance under start-up, shutdown, and other rapidly changing operating conditions, helping to identify performance bottlenecks. Long-term dynamic stiffness trend analysis reveals potential stiffness decay or variation patterns during long-term operation, providing valuable information for equipment maintenance and optimization. Dynamic stiffness characteristic curves visually display spindle stiffness changes under different operating conditions, helping users fully understand the spindle's dynamic performance under various load conditions. Thermal imaging images reflect real-time spindle temperature changes, revealing the impact of temperature fluctuations on dynamic stiffness. By combining temperature data with dynamic stiffness characteristics, the influence of thermal expansion on spindle dynamic stiffness can be quantified, further improving the accuracy of dynamic stiffness testing. Clearly defining the impact of thermal expansion on dynamic stiffness provides a basis for spindle temperature control design and optimizes the spindle's thermal management system. By stripping away the thermal expansion effect, pure mechanical dynamic stiffness characteristics are obtained, ensuring results are unaffected by temperature fluctuations. Based on the dynamic stiffness characteristic report, accurate predictions of spindle dynamic stiffness changes under different temperatures and operating conditions can be made, providing important reference for spindle design, commissioning, and maintenance. The report provides a clear trend of dynamic stiffness variation, the effects of thermal expansion, and optimization suggestions to help make more reasonable decisions on equipment commissioning and optimization. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the steps of the spindle dynamic stiffness testing and analysis method of the present invention; Figure 2 This is a detailed flowchart illustrating the implementation steps of step S1. Figure 3 This is a detailed flowchart illustrating the implementation steps of step S2; Figure 4 This is a flowchart illustrating the detailed implementation steps of step S3. Detailed Implementation

[0012] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0013] This application provides a method and apparatus for testing and analyzing the dynamic stiffness of a spindle. The execution entities of the method and apparatus include, but are not limited to, mechanical equipment, data processing platforms, cloud server nodes, and network upload devices that can be considered general computing nodes in this application. The data processing platform includes, but is not limited to, at least one of an audio / image management system, an information management system, and a cloud-based data management system.

[0014] Please see Figures 1 to 4 This invention provides a method for testing and analyzing the dynamic stiffness of a spindle, which includes the following steps: Step S1: Perform multi-condition operation excitation test on the spindle, and collect spindle test response parameters for multiple conditions based on a multi-dimensional high-precision sensor array; Step S2: Perform dynamic noise cleaning and adaptive filtering optimization on the spindle test response parameters to generate adaptive filtering optimized response parameters; Step S3: Perform dynamic test deviation prediction on the adaptive filter optimization response parameters, and then perform multiple rounds of deviation iterative compensation optimization to generate dynamic deviation optimization parameters; Step S4: Calculate the peak value of transient dynamic stiffness for the dynamic deviation optimization parameters, then analyze the long-term dynamic stiffness change trend, and construct dynamic stiffness characteristic curves under multiple working conditions; Step S5: Acquire thermal imaging images of multi-condition operation excitation test, and perform quantitative correlation analysis based on the dynamic stiffness characteristic curve to obtain the dynamic stiffness influence law of thermal expansion. Step S6: Based on the dynamic stiffness influence law, predict the change in thermal expansion stiffness, and perform thermal expansion effect stripping processing to generate a spindle dynamic stiffness characteristic report.

[0015] This invention simulates the spindle's working state under various operating conditions by conducting multi-condition operational excitation tests. These conditions include different loads, speeds, and accelerations. This step allows for comprehensive acquisition of the spindle's response data under multiple operating conditions, providing rich data support for subsequent analysis. Utilizing a multi-dimensional, high-precision sensor array (such as accelerometers, displacement sensors, and strain gauges) to acquire test response parameters enables precise capture of the spindle's dynamic behavior in different directions and at different frequencies. The multi-dimensional configuration of the sensors improves the comprehensiveness and accuracy of data acquisition, making the analysis results more reliable. During spindle testing, external environmental factors (such as vibration, temperature fluctuations, and electromagnetic interference) may introduce noise signals. Dynamic noise cleaning effectively filters out these environmental noises, retaining key test response data and improving data quality. Adaptive filtering technology automatically adjusts filter parameters based on data characteristics, thereby optimizing signal quality. Unlike traditional filtering methods with fixed parameters, adaptive filtering can flexibly adjust according to changing conditions in actual testing, effectively removing high-frequency noise and low-frequency interference, making the data more accurate and reliable. During testing, sensor accuracy limitations or equipment errors may lead to deviations in test data. By dynamically predicting deviations in the response parameters optimized through adaptive filtering, these errors can be identified in advance, allowing for appropriate compensation measures. This process reduces systematic errors in testing and improves data accuracy. Employing a multi-round iterative deviation compensation optimization method, deviations can be gradually reduced through repeated optimization and adjustment, ultimately yielding more accurate test results. The iterative optimization process can gradually eliminate error sources and improve the reliability of the final test data. The peak transient dynamic stiffness is a crucial indicator of the spindle's operating condition. By accurately calculating the peak dynamic stiffness, we can effectively identify the spindle's stiffness performance under instantaneous loads and detect potential anomalies, such as resonance or excessively low local stiffness. Long-term analysis of dynamic stiffness variation trends under multiple operating conditions can identify performance changes in the spindle over long-term use. This helps predict the spindle's fatigue life, optimize usage plans, and provide a basis for subsequent maintenance and fault prediction. During high-speed spindle operation, temperature changes cause thermal expansion, which in turn affects dynamic stiffness. By acquiring thermal imaging images, the temperature distribution of the spindle can be monitored in real time, revealing the impact of temperature changes on spindle stiffness. Through quantitative correlation analysis between thermal imaging data and dynamic stiffness characteristic curves, the specific impact of temperature changes on spindle stiffness can be accurately calculated. This step helps to understand the influence of thermal expansion on spindle performance and provides data support for further optimization of spindle design and operating conditions. Based on the dynamic stiffness influence of thermal expansion, the dynamic stiffness changes caused by temperature variations under different operating conditions can be predicted. This prediction can help engineers perform temperature control and structural optimization during the design phase to reduce the negative impact of thermal expansion on spindle performance.By exploiting the thermal expansion effect, accurate characteristics of mechanical dynamic stiffness variation are obtained. This report provides crucial reference for long-term spindle performance evaluation, design optimization, and failure prediction. The detailed analysis results in the report help users understand the spindle's health status, service life, and maintenance requirements.

[0016] In the embodiments of the present invention, see Figure 1 The diagram below illustrates the steps of a spindle dynamic stiffness testing and analysis method according to the present invention. In this example, the steps of the spindle dynamic stiffness testing and analysis method include: Step S1: Perform multi-condition operation excitation test on the spindle, and collect spindle test response parameters for multiple conditions based on a multi-dimensional high-precision sensor array; In this embodiment, test preparation is required before conducting multi-condition operation excitation tests. This includes equipment selection, configuration, and site layout. Suitable test equipment, such as high-precision sensor arrays, should be selected. These sensors should be able to monitor the dynamic response parameters of the spindle in real time. Commonly used sensors include accelerometers, displacement sensors, and temperature sensors. A three-dimensional accelerometer array is configured to capture vibration data in the XYZ directions. Ensure that the sensitivity, frequency range, and measurement range of each sensor are suitable for the spindle's operating conditions. For accelerometers, the sensitivity should be between 0.1 g and 100 g, and the frequency range should cover the spindle's operating frequency (e.g., 0-5 kHz). Furthermore, the sensor installation positions should be rationally arranged according to the spindle design drawings to ensure coverage of all key parts of the spindle. After test preparation, a multi-condition test plan needs to be developed. Different operating conditions are set according to the actual operating conditions of the spindle, including factors such as load, speed, and temperature. Common operating condition settings include: Operating Condition A: High load (100% load), high speed (3000 RPM), normal ambient temperature (25°C).

[0017] Operating Condition B: Medium load (75% load), medium speed (2000 RPM), temperature rise environment (45°C).

[0018] Operating condition C: Low load (50% load), low speed (1500 RPM), low temperature environment (15°C).

[0019] The settings for each operating condition should be rationally arranged according to the application scenario and design requirements of the spindle. Furthermore, an excitation strategy should be designed, such as using a motor or vibration exciter to apply different load conditions to the spindle to simulate the dynamic response under actual operating conditions. During the test, the spindle is started and excited to run. A pre-configured high-precision sensor array is used to collect the dynamic response data of the spindle in real time. The data acquisition system should have a high-frequency sampling capability to ensure that subtle changes in the spindle under different operating conditions are captured. The data acquisition frequency is set to 1000 Hz to ensure that sufficient data points are recorded under each operating condition. Under operating condition A, the main test response parameters monitored include acceleration, displacement, and temperature. The accelerometer provides real-time feedback of the vibration acceleration signal, the displacement sensor records the radial and axial displacement of the spindle, and the temperature sensor monitors changes in the spindle surface temperature. After data acquisition, all monitored data should be stored promptly and preliminarily processed. Data analysis software (such as MATLAB or LabVIEW) is used to organize and filter the acquired data, removing noise and outliers. Preliminary data processing can be accomplished by setting thresholds. A minimum threshold is set for the vibration acceleration signal, filtering out noise signals below this value. Additionally, a test response parameter report is generated for each operating condition, recording the operating condition settings, acquisition time, sensor data, and their statistical characteristics (such as mean, standard deviation, etc.). This data will provide the foundation for subsequent dynamic analysis and dynamic stiffness testing.

[0020] Step S2: Perform dynamic noise cleaning and adaptive filtering optimization on the spindle test response parameters to generate adaptive filtering optimized response parameters; In this embodiment, response parameter data is collected from spindle testing. This data typically includes vibration signals, temperature changes, load conditions, etc., ensuring data integrity and consistency. Data should be recorded in time-series format for subsequent analysis. Vibration signal data is collected in the format of "timestamp, vibration acceleration," such as "2023-04-01 10:00:00, 0.5g," ensuring the dataset covers multiple operating conditions. Before data analysis, signal cleaning is performed to remove noise and outliers. A threshold can be set to identify and remove signal values ​​outside the normal range. The standard deviation method is typically used to define outliers. If the mean of the vibration signal is 0.5 g and the standard deviation is 0.1 g, the threshold is set to mean ± 2 * standard deviation, i.e., 0.3 g to 0.7 g. Any value outside this range is considered noise and must be removed. Time-domain analysis methods, such as the sliding window technique, are used to detect noise in the signal. By analyzing signal fluctuations, high-energy, short-duration noise components are identified. These components typically manifest as high-amplitude, short-duration signals. An energy threshold is set; when the instantaneous amplitude of vibration acceleration exceeds 1 g, it is marked as a noise range for subsequent processing. For identified noise components, low-pass or median filters are used for removal. Low-pass filters effectively eliminate high-frequency noise, while median filters are suitable for removing impulse noise. A median filter is applied to process the vibration signal, with a window size of 5 data points to smooth the signal and remove noise. The processed signal should be smoother, reducing the impact of sudden noise. The noise-cleaned signal data is recorded in a database for subsequent analysis and comparison. The cleaning effect is evaluated by comparing the signal waveforms before and after cleaning to observe the noise reduction. The recording format is "timestamp, optimized vibration acceleration," such as "2023-04-01 10:00:00, 0.45 g," and waveforms before and after cleaning are generated for comparison. A suitable adaptive filtering algorithm, such as Least Mean Squares (LMS) or Recursive Least Squares (RLS), is selected for dynamic signal optimization. These algorithms can dynamically adjust filter parameters according to changes in the input signal. The LMS algorithm was selected, with a learning rate of 0.01 to ensure rapid adaptation to signal changes. During adaptive filtering, the filter parameters were continuously adjusted to minimize the error between the output signal and the reference signal. The reference signal could be the cleaned signal or the desired signal. The error was gradually reduced by iteratively updating the filter weights. If the mean square error (MSE) between the filter output signal and the actual vibration signal gradually decreased, it indicated that the filtering effect was gradually optimized. The dynamic response parameters after adaptive filtering were saved, and relevant information was recorded for later use. This ensured that the optimized signal accurately reflected the dynamic performance of the spindle.The recording format is "timestamp, adaptive filtering optimization response parameters", such as "2023-04-01 10:00:00, optimized vibration acceleration 0.42 g", and an optimized waveform is generated for visualization. The signal waveforms before and after noise removal, as well as before and after optimization, are compared to evaluate the overall removal and filtering effect. Signal stability and noise reduction are checked. The signal graphs before and after optimization are compared to observe changes in signal fluctuation amplitude, ensuring that the optimized signal is smoother and the fluctuation range is reduced. Performance evaluation metrics such as signal-to-noise ratio (SNR) are calculated to quantify signal clarity and stability. An improved SNR generally indicates improved signal quality. If the SNR before removal is 5 dB and increases to 15 dB after removal, it indicates a significant improvement in signal quality.

[0021] Step S3: Perform dynamic test deviation prediction on the adaptive filter optimization response parameters, and then perform multiple rounds of deviation iterative compensation optimization to generate dynamic deviation optimization parameters; In this embodiment, data is extracted from the adaptive filtering optimized response parameters obtained in the previous stage to ensure data integrity and consistency. This data should include dynamic response signals under different operating conditions, such as vibration acceleration, rotational speed, and load status. Optimized vibration signals are collected under operating conditions A and B, recorded in the format "timestamp, operating condition type, optimized vibration acceleration," such as "2023-04-01 10:00:00, operating condition A, 0.42 g." A suitable dynamic test deviation prediction model is selected, such as linear regression, support vector machine (SVM), or neural network. These models can predict potential deviations based on known response parameters. A linear regression model is used, fitting vibration acceleration as the independent variable and the actual measured dynamic stiffness as the dependent variable to establish the model. The selected model is trained using historical data to ensure that it accurately captures the relationship between dynamic response parameters and actual test results. The model's performance is evaluated through cross-validation, and model parameters are optimized to improve prediction accuracy. If the root mean square error (RMSE) on the validation set after model training is 0.02 g, the model's predictive ability is good. The trained model is used to predict the dynamic deviation of the current adaptive filtering optimized response parameters, calculating the possible deviation values. This process must ensure that the prediction results reflect the performance changes of the spindle under different operating conditions. If the prediction result shows a dynamic deviation of 0.03 g under operating condition A, it is recorded as "Operating Condition A, predicted dynamic deviation value 0.03 g". A suitable compensation optimization algorithm, such as iterative least squares (ILS) or adaptive filtering algorithm, is selected to iteratively compensate for the dynamic deviation. This algorithm can dynamically adjust according to the predicted deviation value. The number of iterations is set to 5, and the learning rate is 0.01 to ensure that each iteration can effectively reduce the deviation. In each iteration, the difference between the current response parameter and the predicted deviation is calculated, and the optimization parameters are adjusted according to this difference. The updated parameters should gradually approach the true value, reducing the dynamic deviation. If the deviation is found to be 0.03 g in the first iteration, the optimization parameters are adjusted to "current value - 0.01 g" in the next iteration to gradually compensate. The dynamic deviation optimization parameters after each iteration are recorded in the database to ensure the traceability of each compensation process. The record should include the deviation value and optimized parameters for each iteration. The record format is "Working Condition Type, Iteration Round, Optimized Dynamic Deviation", such as "Working Condition A, Round 1, Optimized Dynamic Deviation 0.02 g". During the iteration process, the convergence of the deviation needs to be continuously checked to determine whether to continue iteration. A convergence threshold is usually set; iteration stops when the deviation falls below this threshold. If the deviation drops to 0.005 g in the fifth iteration, it is recorded as "Working Condition A, Iteration Ended, Final Dynamic Deviation 0.005 g". Based on the results of multiple iterations of compensation optimization, the final optimized dynamic deviation parameters are generated. These parameters will be used for subsequent dynamic stiffness testing and analysis to ensure the accuracy and reliability of the test results.The final record is "Operating Condition A, Dynamic Deviation Optimization Parameter 0.005 g". The process and results of dynamic deviation optimization are compiled into a report, detailing the parameter changes and prediction effects for each iteration. Ensure the report is clear for subsequent analysis and decision-making. The report should include information such as "Operating Condition Type, Iteration Round, and Final Dynamic Deviation" to facilitate team members' understanding of the optimization process and results.

[0022] Step S4: Calculate the peak value of transient dynamic stiffness for the dynamic deviation optimization parameters, then analyze the long-term dynamic stiffness change trend, and construct dynamic stiffness characteristic curves under multiple working conditions; In this embodiment, data is extracted from the dynamic deviation optimization parameters collected in step S3. This data should include the optimized vibration response signal, as well as the corresponding load and rotational speed information, for dynamic stiffness calculation. The optimized vibration acceleration under condition A is recorded as 0.42 g, the load as 100 N, and the rotational speed as 3000 RPM. Dynamic stiffness is typically defined as the ratio of applied force to generated displacement, and can be calculated using the formula K = F / x, where K is the dynamic stiffness, F is the applied force, and x is the corresponding displacement. To calculate the peak transient dynamic stiffness, analysis should be performed at the peak point of the vibration signal. If the peak displacement of the vibration signal under condition A is 0.002 m and the applied force is 100 N, then the transient dynamic stiffness is calculated as K = 100 N / 0.002 m = 50,000 N / m. By analyzing the time-series data of the vibration signal, the peak point of the transient signal is identified. Signal processing techniques, such as peak detection algorithms, can be used in this process to ensure accurate peak location. The peak value in the vibration signal is extracted using a peak detection algorithm and recorded as "Condition A, transient dynamic stiffness peak 50,000 N / m", ensuring that peak values ​​for all conditions are recorded. Dynamic stiffness measurement data for each condition is collected during long-term operation. This data should cover different time periods and conditions for trend analysis. Dynamic stiffness data for Condition A and Condition B at different time points are recorded, forming a dataset containing time, condition type, and dynamic stiffness value, such as "timestamp, condition type, dynamic stiffness value". A suitable time series analysis method, such as moving average or linear regression, is selected to assess the long-term trend of dynamic stiffness. Moving average can smooth the data and help identify trends. A window size of 3 is set, and the moving average of dynamic stiffness is calculated to observe the stability and trend of the data. The collected data is analyzed to generate a dynamic stiffness trend graph. Statistical analysis software (such as Excel or MATLAB) can be used to plot the curve of dynamic stiffness changing over time to visually display the trend. The generated trend chart shows that the dynamic stiffness of operating condition A is initially 50,000 N / m, then gradually decreases to 48,000 N / m, recording the trend change. Long-term dynamic stiffness change trends are recorded in a database to ensure the traceability of trend data for each operating condition. An analysis report is generated, displaying key data and trend charts of dynamic stiffness changes. The record format is "Operating Condition Type, Time Period, Dynamic Stiffness Change," such as "Operating Condition A, Q1 2023, Dynamic Stiffness decreased from 50,000 N / m to 48,000 N / m." The dynamic stiffness characteristic curve is a graph showing the relationship between dynamic stiffness values ​​and different operating conditions (such as load, speed, and temperature). These curves help analyze the dynamic performance of the spindle under different working conditions. Different operating condition parameters are set, and the relationship between dynamic stiffness and load is plotted to ensure that the curves reflect the true dynamic performance characteristics. Based on the collected dynamic stiffness data and operating condition parameters, dynamic stiffness characteristic curves are plotted using data visualization tools.Ensure the curve clearly shows the change in dynamic stiffness under different operating conditions. The generated characteristic curve shows the trend of dynamic stiffness decreasing with increasing load under operating condition A, and is recorded as "Dynamic Stiffness Characteristic Curve for Operating Condition A".

[0023] Step S5: Acquire thermal imaging images of multi-condition operation excitation test, and perform quantitative correlation analysis based on the dynamic stiffness characteristic curve to obtain the dynamic stiffness influence law of thermal expansion. In this embodiment, a high-resolution thermal imaging camera is prepared and calibrated as necessary to ensure accurate temperature measurement capabilities. During calibration, a standard temperature object (such as a blackbody radiation source) can be used to verify the camera's temperature measurement accuracy. During calibration, a standard object with a temperature of 100°C is selected to ensure that the error between the temperature displayed by the thermal imaging camera and the actual temperature does not exceed ±1°C. The spindle is subjected to excitation tests under different operating conditions, while thermal imaging images are acquired in real time using the thermal imaging camera. The tests should cover various operating conditions, such as different loads, speeds, and running times, to comprehensively evaluate the temperature distribution. Operating condition A is set as high load (100% load) and high speed (3000 RPM), and operating condition B is set as low load (50% load) and low speed (1500 RPM). Thermal imaging images are acquired under each operating condition. The acquired thermal imaging images are classified and stored according to timestamps and operating condition types for subsequent analysis. Each image should contain spindle temperature distribution information and record the corresponding time and operating conditions. The record format is "timestamp, operating condition type, image file name", such as "2023-04-01 10:00:00, operating condition A, thermal_image_A1.jpg". Extract the dynamic stiffness characteristic curve data from step S4, ensuring the data covers the relationship between dynamic stiffness values ​​and temperature changes under different operating conditions. If necessary, match the dynamic stiffness data with the corresponding thermal imaging images. Record the dynamic stiffness value as 50,000 N / m and temperature as 75°C under operating condition A, and the dynamic stiffness value as 48,000 N / m and temperature as 65°C under operating condition B. Select a suitable quantitative analysis method, such as linear regression or multiple regression analysis, to establish a model of the relationship between dynamic stiffness and temperature. Through analysis, the influence of thermal expansion on dynamic stiffness can be identified. Set up a linear regression model, using dynamic stiffness value as the dependent variable and temperature change as the independent variable for fitting analysis. Use the collected dynamic stiffness data and temperature data to build the model, and calculate the regression coefficients and correlation indices. Ensure the model can reflect the degree of influence of temperature changes on dynamic stiffness. Analysis revealed that dynamic stiffness decreases by approximately 200 N / m for every 1°C increase in temperature. The correlation coefficient R² was recorded to assess the model's fit. Analysis results were stored in a database to ensure the traceability of dynamic stiffness influence for each operating condition. An analysis report was generated, showcasing the quantitative relationship between dynamic stiffness and temperature and its influence. The recording format was "Operating Condition Type, Temperature, Dynamic Stiffness Variation," such as "Operating Condition A, Temperature 75°C, Dynamic Stiffness decreases by 200 N / m with increasing temperature." Based on the quantitative correlation analysis results, the influence of thermal expansion on dynamic stiffness was elucidated. This ensured the analysis provided a theoretical basis for subsequent design and optimization. The analysis results indicate that the dynamic stiffness of the spindle gradually decreases with increasing temperature, a phenomenon primarily caused by the thermal expansion of the material.

[0024] Step S6: Based on the dynamic stiffness influence law, predict the change in thermal expansion stiffness, and perform thermal expansion effect stripping processing to generate a spindle dynamic stiffness characteristic report.

[0025] In this embodiment, data on the influence of dynamic stiffness are collected based on previous analysis results. This data includes the relationship between dynamic stiffness values ​​and temperature changes, ensuring data completeness and accuracy for predicting changes in thermal expansion stiffness. The trend of dynamic stiffness changing with temperature under operating condition A is recorded, with a dynamic stiffness of 50,000 N / m and a temperature of 75°C, and a decrease of 200 N / m for every 1°C increase. Using the previously established linear regression model, the dynamic stiffness change under different temperature conditions is predicted. A temperature range is set, for example, from ambient temperature (25°C) to operating temperature (85°C), and the dynamic stiffness change within this range is calculated. The predicted dynamic stiffness value at each temperature is calculated using the formula K = K0 - ΔK × ΔT, where K0 is the initial dynamic stiffness, ΔK is the change in dynamic stiffness for every 1°C increase, and ΔT is the temperature change. If the dynamic stiffness is 50,000 N / m at 25°C, then the predicted dynamic stiffness at 85°C is K = 50,000 N / m - 200 N / m × (85°C - 25°C) = 44,000 N / m. Record the prediction results in the database to ensure that the dynamic stiffness change under each temperature condition is traceable. Generate a preliminary analysis report showing the predicted data of dynamic stiffness changing with temperature. The record format is "Temperature, Predicted Dynamic Stiffness", such as "85°C, Dynamic Stiffness 44,000 N / m". Select an appropriate stripping method to isolate the thermal expansion effect from the dynamic stiffness characteristics. Commonly used methods include the finite difference method or regression analysis to ensure accurate identification of the impact of thermal expansion on dynamic stiffness. Set the stripping threshold to the amount of dynamic stiffness change caused by thermal expansion (e.g., a decrease of 200 N / m for every 1°C increase) and adjust it in the dynamic stiffness characteristic curve. For each operating condition, the dynamic stiffness data is stripped to remove the thermal expansion effect caused by temperature changes. The stripped dynamic stiffness value is then calculated to obtain more accurate mechanical dynamic stiffness characteristics. If the dynamic stiffness of operating condition A at 85°C is 44,000 N / m, after stripping the thermal expansion effect, it is recorded as "Dynamic stiffness after stripping: 42,000 N / m". Based on all previous analysis results, a report is written. The analysis process and results of each step are recorded in detail, including the trend of dynamic stiffness change with temperature, the results of stripping the thermal expansion effect, and its impact on mechanical performance. The conclusion states: "By stripping the thermal expansion effect, the mechanical dynamic stiffness characteristics of the spindle under high-temperature conditions are more accurately evaluated, providing an important basis for subsequent design optimization."

[0026] In this embodiment, see Figure 2The diagram below illustrates the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Obtain the preset spindle operation monitoring log; Multi-condition operating parameters are defined for the spindle operation monitoring log to obtain operating test parameters for multiple conditions; Calculate the duration of multiple historical peak operating times based on the spindle operation monitoring logs; The extreme condition test time is calculated based on the duration of the multiple historical peak operating times. Based on the operating test parameters and extreme operating test time under multiple working conditions, the spindle is subjected to multi-working condition operating excitation test through a dynamic loading test platform, and the spindle test response parameters under multiple working conditions are collected based on a multi-dimensional high-precision sensor array.

[0027] In this embodiment, the source of the spindle operation monitoring log is confirmed. These logs are typically generated by the spindle's built-in sensors and data acquisition system, recording key parameters such as the spindle's operating status, load, speed, and temperature. Ensuring the reliability and integrity of the data source is fundamental to the analysis. The data is confirmed to be stored in an SQL database or CSV file, and accessibility is ensured. The spindle operation monitoring log is extracted from the confirmed data source, and the recorded content should include timestamps, speeds, bearing temperatures, load conditions, etc. The extracted data should cover the spindle's operating cycle for subsequent analysis. Hourly monitoring data from the past 30 days is extracted, recorded in the format of "timestamp, speed, load, temperature". The extracted monitoring logs are preprocessed, including data cleaning and formatting, to ensure data consistency and integrity. Missing and outlier values ​​are checked during processing and imputed or removed accordingly. If speed data is missing at a certain point in time, it can be imputed using the average of previous and subsequent time points to ensure the accuracy of the analysis. Based on the data in the spindle operation monitoring log, multiple operating parameters are defined. These operating conditions should consider different load, speed, and temperature conditions to comprehensively evaluate the spindle's performance. Define operating condition A as high load (100% load) and high speed (3000 RPM), operating condition B as medium load (75% load) and medium speed (2000 RPM), and operating condition C as low load (50% load) and low speed (1000 RPM). For each defined operating condition, collect relevant operating test parameters, including speed, load, and ambient temperature. These parameters will be used for subsequent testing and analysis. The test parameters for operating condition A might be "speed: 3000 RPM, load: 100%", while the parameters for operating condition B might be "speed: 2000 RPM, load: 75%". Record the operating test parameters for multiple defined operating conditions in a database for later reference and comparison. Ensure that the parameters for each operating condition fully describe its operating status. The record format is "operating condition type, speed, load, temperature", such as "operating condition A, 3000 RPM, 100%, 25°C". Determine the method for calculating the historical peak operating duration. Peak duration typically refers to the duration during which the spindle reaches its maximum operating state under a specific load or speed. The analysis should consider the spindle's duty cycle and its peak frequency. Define "peak duration" as the period during which the speed or load exceeds a certain threshold (e.g., 90% load). Analyze the extracted monitoring logs to calculate the historical peak duration of the spindle under different operating conditions. A sliding window technique can be used to identify peaks within a defined time period. If, in the past 30 days of monitoring data, the spindle reached 3000 RPM in 10 cycles under operating condition A, with durations of 5 minutes, 10 minutes, etc., then record the duration of each peak. Organize and record the calculated historical peak durations to ensure data integrity and traceability. Each record should include the operating condition type and the corresponding peak duration.The record format is "Operating Condition Type, Peak Duration", such as "Operating Condition A, Peak Duration: 15 minutes". Define the test time for the extreme operating condition based on historical peak operating durations and the spindle's operating characteristics. The extreme operating condition typically refers to testing under the most demanding conditions to evaluate the spindle's ultimate performance. Set the extreme operating condition to be tested at maximum load (120%) and maximum speed (3500 RPM). Calculate the test time under the extreme operating condition using the historical peak duration calculation results, combined with the spindle's material properties and fatigue limit. A fatigue life prediction model can be used to predict the safe operating time under the extreme operating condition. If the historical peak duration is 15 minutes, the test time under the extreme operating condition can be estimated at 10 minutes to ensure a safety margin. Record the calculated extreme operating condition test time in the database for reference and use in subsequent tests. The record should include the definition of the extreme operating condition and the corresponding test time. The record format is "Operating Condition Type, Extreme Test Time", such as "Extreme Operating Condition, Test Time: 10 minutes". Configure a dynamic loading test platform to simulate the spindle's operating state under different working conditions. The test platform should have real-time control and data acquisition capabilities, and be able to dynamically adjust the load and speed. Ensure that the load device of the test platform can be quickly adjusted to the set 100% or 120% load, while also having the ability to precisely control the speed. Based on the previously defined multiple working condition parameters, excite the spindle through the dynamic loading test platform. Record the spindle's operating response and performance data under each working condition. Test under working condition A, recording the spindle's operating noise, vibration, temperature changes, and other response parameters to ensure the accuracy of data acquisition. Utilize a multi-dimensional high-precision sensor array to collect the spindle's test response parameters under different working conditions. These sensors should be installed at key locations on the spindle to obtain comprehensive performance data. Deploy accelerometers, temperature sensors, and vibration sensors to monitor the spindle's dynamic response in real time during testing. Record the collected multi-working-condition response parameters in real time during testing, and perform data analysis after testing. The analysis should focus on indicators such as the spindle's dynamic stiffness, stability, and fatigue performance. Generate test data reports, including key performance indicators such as the vibration amplitude, frequency response, and temperature changes of the spindle under various operating conditions.

[0028] In this embodiment, the specific steps for defining multi-condition operating parameters on the spindle operation monitoring log to obtain operating test parameters for multiple conditions are as follows: Perform routine operation analysis on the spindle operation monitoring log to identify different operation scenarios; Calculate the load frequency and amplitude for the different operating scenarios; Based on the load frequency and amplitude, perform multi-condition load characteristic analysis to obtain the load characteristics of each scenario; Calculate the natural frequencies and modal characteristics for the different operating scenarios; Based on the inherent frequency and modal characteristics, the main modal characteristics are mined to obtain the constant frequency resonant mode for each scene; Based on the load characteristics of each operating condition and the constant frequency resonant mode of each operating condition, the operating parameters of multiple operating conditions are defined, thereby obtaining the operating test parameters of multiple operating conditions.

[0029] In this embodiment, the spindle operation monitoring logs undergo data preprocessing, including data cleaning, standardization, and formatting. This ensures the consistency, integrity, and availability of the monitoring data, removing noise and outliers to facilitate subsequent analysis. A unified timestamp format is used to handle missing and outliers, ensuring the completeness of records at each time point, such as "timestamp, speed, load, temperature." Based on the preprocessed data, different operating scenarios are defined. These scenarios should be categorized according to the spindle's operating status, load conditions, and speed. Common scenarios include normal operation, high load, low load, and overload. Scenario A is defined as normal operation (speed 2000 RPM, load 80%), Scenario B as high load (speed 2500 RPM, load 100%), and Scenario C as low load (speed 1500 RPM, load 50%). Cluster analysis (such as K-means clustering or DBSCAN) is used to analyze the operating data, automatically identifying different operating scenarios. By analyzing changes in parameters such as load, speed, and temperature, the spindle's operating state pattern is identified. The data was divided into three categories using K-means clustering, corresponding to normal operation, high load, and low load scenarios, ensuring that the characteristics of each scenario were distinct. Load signals were extracted from the monitoring data and analyzed in both the time and frequency domains. The load frequency and amplitude for each identified operating scenario were calculated to comprehensively understand the load characteristics of the spindle under different operating conditions. Fourier transform analysis was used to analyze the load signal for each scenario, extracting spectral data and analyzing frequency components. For each operating scenario, the load frequency and amplitude were calculated. Frequency can be obtained using FFT (Fast Fourier Transform), while amplitude represents the maximum or average load value. In scenario A, the dominant frequency of the load signal is 2 Hz, and the amplitude is 50 N, recorded as "Scenario A: Frequency 2 Hz, Amplitude 50 N". Appropriate feature extraction methods were selected to analyze the load characteristics of each operating scenario. Statistical feature extraction methods, such as mean, standard deviation, kurtosis, and skewness, can be used. The mean and standard deviation of the load signal for each scenario were calculated to assess the stability and volatility of the load. For each scenario, its load characteristics are calculated, including mean, maximum, minimum, and standard deviation, to comprehensively assess load variations. In scenario B, the calculated load mean is 90 N, and the standard deviation is 10 N, recorded as "Scenario B: Mean 90 N, Standard Deviation 10 N". The load characteristic results for each scenario are recorded in a database to ensure traceability and ease of subsequent analysis. The record format is "Scenario Type, Load Characteristics", such as "Scenario B, Mean 90 N, Standard Deviation 10 N". For each operating scenario, modal analysis is prepared to calculate the natural frequencies of the spindle under different operating conditions. Modal analysis generally employs experimental modal analysis (EMA) or finite element analysis (FEA) methods. Experimental conditions are set, and excitation is applied to the spindle using a vibrator, monitoring its response. Through modal analysis, the natural frequencies of each scenario are identified. Spectral analysis is used to determine the main natural frequencies of the spindle under different operating conditions.In scenario A, the first natural frequency of the spindle is calculated to be 15 Hz, recorded as "Scenario A: Natural Frequency 15 Hz". Besides the natural frequency, modal damping ratio and modal shape, among other characteristics, need to be calculated to comprehensively understand the spindle's dynamic performance. The modal damping ratio in scenario A is recorded as 0.05, and the modal shape as "First Modal Shape Description". The natural frequencies and modal characteristics of each scenario are integrated to identify the constant frequency resonant mode. These modal characteristics will help understand the spindle's dynamic behavior under different operating conditions. If the modal characteristics of the natural frequency 15 Hz are consistent across multiple scenarios, this frequency can be identified as the constant resonant mode. The constant frequency resonant mode for each scenario is recorded, and its impact on spindle performance is summarized. If 15 Hz is the resonant frequency, its impact on spindle stability needs to be considered. The recording format is "Scenario Type, Constant Frequency Resonant Mode", such as "Scenario A, 15 Hz". Based on the load characteristics and constant frequency resonant mode of each scenario, multi-condition operating test parameters are defined. These parameters will be used for subsequent performance testing and evaluation. The operating test parameters for condition A are set as "speed 2000 RPM, load 80%, resonant frequency 15 Hz". The defined multi-condition operating test parameters are recorded in the database to ensure the accuracy and traceability of each parameter. Simultaneously, verification is performed to ensure the rationality of the parameters. The recording format is "condition type, speed, load, resonant frequency", such as "condition A, 2000 RPM, 80%, 15Hz". A comprehensive analysis is performed based on all recorded operating parameters to provide a basis for subsequent spindle dynamic stiffness testing. It is ensured that the definition of each parameter reflects the actual operating conditions.

[0030] In this embodiment, see Figure 3 The diagram below illustrates the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: The spindle test response parameters are decomposed into multiple time-frequency parameters to obtain the test response time-frequency spectrum. Non-vibrational component features were identified from the time-frequency spectrum of the test response, and environmental noise points were extracted. The signal-to-noise ratio of the environmental noise points is calculated, and dynamic noise cleaning is performed to obtain noise optimization test response parameters. Dynamic response parameter change analysis is performed on the noise optimization test response parameters to generate dynamic response change characteristics; Adaptive filtering optimization is performed based on the dynamic response change characteristics to generate adaptive filtering optimization response parameters.

[0031] In this embodiment, test response parameters of the spindle are collected, including vibration signals, temperature changes, and load conditions. These data need to be cleaned and standardized to ensure their integrity and consistency. Missing and outlier values ​​must be removed during the cleaning process. Vibration signals recorded during the spindle test are extracted, ensuring the data format is "timestamp, vibration acceleration," and converted to appropriate units (e.g., g). A suitable time-frequency analysis method, such as Short-Time Fourier Transform (STFT) or Wavelet Transform (WT), is selected to convert the time-domain signal into a time-frequency domain signal. These methods can effectively capture the instantaneous frequency changes of the signal. Wavelet transform is used to analyze the spindle's vibration signal, decomposing it into different frequency components for subsequent feature analysis. The selected time-frequency decomposition method is applied to analyze the spindle's test response data, generating a time-frequency spectrum. This spectrum should show the frequency components of the signal and their changes over different time periods. The generated time-frequency spectrum shows that the main frequency components of a certain test phase are 50 Hz and 150 Hz, indicating the duration and amplitude changes of these frequencies. Signal processing techniques, such as threshold-based detection methods or machine learning algorithms, are employed to identify non-vibrational components in the time-frequency spectrum. These components typically manifest as high-energy, low-frequency noise signals. An energy threshold is set; when the energy of a frequency component exceeds this threshold, it is identified as a noise component. Based on the identified non-vibrational components, corresponding environmental noise points are extracted. These noise points typically appear as spikes or outliers at specific frequencies in the time-frequency spectrum. A noise point identified in the time-frequency spectrum might correspond to a signal with a frequency of 60 Hz, which is usually power supply frequency noise. For each extracted noise point, its signal-to-noise ratio (SNR) is calculated. SNR is an important indicator for evaluating signal quality; it assesses signal clarity by comparing the power of the signal and noise. If the power of the vibration signal is 0.8 g², and the noise power is 0.2 g², then the SNR is calculated as 10 * log10(0.8 / 0.2) = 10 dB. Based on the calculated SNR, a dynamic noise cleaning algorithm is applied to process the raw test response data. Common noise removal methods include low-pass filtering, high-pass filtering, or adaptive filtering. If the signal-to-noise ratio at a noise point is below 5 dB, that point is considered noise, and a low-pass filter is used to remove that frequency component. The test response parameters after noise removal are recorded in a database for subsequent analysis and comparison. Each cleaned parameter is traceable. The recording format is "timestamp, optimized vibration acceleration," such as "2023-04-01 12:00:00, vibration acceleration 0.3 g." Dynamic response analysis is performed on the noise-optimized test response parameters to extract variation characteristics. This can be achieved by calculating statistical characteristics such as the mean, variance, and peak value of the parameters. The variation of the optimized vibration acceleration within a certain time window is analyzed, and its mean and standard deviation are calculated to evaluate the dynamic performance of the spindle.Time series analysis methods, such as moving average or exponential smoothing, are used to analyze the trends in dynamic response parameters. This helps identify performance changes in the spindle under different operating conditions. If the vibration acceleration gradually increases under high load conditions, it indicates that the spindle's dynamic performance may be declining. A suitable adaptive filtering algorithm, such as the Least Mean Square (LMS) algorithm or Kalman filtering, is selected to optimize the dynamic response parameters. Adaptive filtering dynamically adjusts filter parameters based on changes in the input signal. Using the LMS algorithm, a learning rate and target signal are set to optimize the dynamic response parameters. During adaptive filtering, filter parameters are continuously adjusted to minimize the error between the output signal and the target signal. This ensures a clearer dynamic response signal. As the input signal changes, the filter weights are dynamically adjusted to filter out unwanted noise components. The adaptively filtered dynamic response parameters are recorded in a database to ensure traceability and accuracy. These optimized parameters will be used for subsequent analysis and evaluation. The recording format is "timestamp, adaptive filtering optimized response parameters", such as "2023-04-01 12:00:00, optimized vibration acceleration 0.2 g".

[0032] In this embodiment, see Figure 4 The diagram below illustrates the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: The frequency of each sensor signal is calculated by optimizing the response parameters of the adaptive filtering to obtain the sensor signal frequency; The sampling frequency of each sensor in a multi-dimensional high-precision sensor array is calculated to generate the expected sampling signal frequency for each sensor. Based on the expected sampling signal frequency, perform sampling signal distortion analysis on the sensor signal frequency to identify sensor sampling signal distortion information; The environmental noise points are analyzed for time-series variation characteristics to generate noise time-series variation features; Dynamic test deviation prediction is performed based on the distortion information of sensor sampling signals and the temporal variation characteristics of noise, so as to obtain the dynamic test deviation value; Based on the dynamic test deviation value, the adaptive filtering optimization response parameters are optimized through multiple rounds of deviation iteration compensation to generate dynamic deviation optimization parameters.

[0033] In this embodiment, sensor signal data is extracted from the adaptively filtered and optimized response parameters. This data should include the dynamic response of the spindle under different operating conditions to ensure data integrity and consistency. Vibration acceleration, temperature changes, and other signals recorded during spindle testing are extracted, formatted as "timestamp, signal value". A suitable spectral analysis method, such as Fast Fourier Transform (FFT), is selected to calculate the frequency components of the sensor signals. FFT can convert time-domain signals into frequency-domain signals, thereby identifying the main frequency components in the signal. FFT is applied to analyze the vibration signal to obtain a spectrum, identifying the main frequency components and their amplitudes. The main frequency of each sensor signal is calculated through spectral analysis, and this frequency information is recorded. The identified frequency components will be used for subsequent analysis and comparison. If the main frequency of a sensor signal is 100 Hz, it is recorded as "Sensor A, signal frequency 100 Hz". Signals are extracted one by one from the multi-dimensional high-precision sensor array to calculate the sampling frequency of each sensor. It is ensured that the signal of each sensor can be accurately sampled within the same time window. The sampling time for each sensor during the test is set to 1 second, and the sampling data of each sensor is recorded. Calculate the sampling frequency of each sensor, typically the number of samples per second. If a sensor collects 100 data points in 1 second, its sampling frequency is 100 Hz. If sensor B records 120 data points in 1 second, the calculated sampling frequency is 120 Hz. Record the calculated expected sampling signal frequency for each sensor in a database to ensure the traceability of each sensor's parameters. The recording format is "Sensor type, expected sampling frequency," such as "Sensor B, 120Hz." Select a suitable signal distortion analysis method, such as phase difference analysis or amplitude distortion analysis, to identify distortion information in the sensor's sampled signals. Use phase difference analysis to detect phase changes in the signal at different time points and identify potential distortions. Perform distortion analysis on the signal of each sensor, calculating key parameters such as phase difference and amplitude changes to evaluate the signal quality and accuracy. If the phase difference of sensor C exceeds a set threshold (e.g., 5°), record that the sensor signal has distortion. Based on the environmental noise points extracted in the previous steps, perform noise temporal variation characteristic analysis. Identify the temporal variation characteristics of noise signals to understand their impact on test response. Extract noise signal data from different time periods and analyze their frequency and amplitude variations. Perform time-series analysis on the extracted noise signals, calculating statistical characteristics such as mean, variance, and peak value to assess noise variation. Use the sliding window method to analyze the dynamic characteristics of the noise signals. If the mean of the noise signal is 60 dB and the variance is 5 dB within a certain time period, it is recorded as "Noise Time-Series Characteristics, Mean 60 dB, Variance 5 dB".The analyzed noise temporal variation characteristics are recorded in a database to ensure the traceability of each feature and provide a basis for subsequent dynamic test deviation prediction. The recording format is "noise feature, mean, variance", such as "noise feature, mean 60 dB, variance 5 dB". A suitable dynamic test deviation prediction model, such as a linear regression model or a machine learning algorithm (such as a support vector machine), is selected to combine sensor sampling signal distortion information and noise temporal variation characteristics for prediction. The linear regression model is used to predict the dynamic test deviation value by taking the sensor signal distortion characteristics and noise variation characteristics as input. The selected prediction model is trained using historical test data to ensure that the model can accurately capture the impact of signal distortion and noise variation on the dynamic test results. The model performance is evaluated through cross-validation. If the root mean square error (RMSE) of the model on the test set is lower than a set threshold (such as 0.1), the model is considered to perform well. The trained model is applied to predict the current sensor signal distortion information and noise temporal variation characteristics to calculate the dynamic test deviation value. If the predicted dynamic test deviation value is 0.05, it is recorded as "dynamic test deviation value 0.05". Choose a suitable compensation optimization method, such as iterative least squares or adaptive filtering, to perform iterative compensation for the deviation of the adaptive filtering optimized response parameters. Use iterative least squares to adjust the optimization parameters, ensuring that the deviation is reduced in each iteration. Based on the calculated dynamic test deviation value, perform multiple rounds of deviation iterative compensation optimization. Each iteration should continuously update the optimization parameters to gradually reduce the dynamic test deviation. If the deviation decreases to 0.02 after the first iteration, record the optimization parameters and continue iterating until the deviation reaches an acceptable range (e.g., less than 0.01). Record the dynamic response parameters after multiple rounds of deviation iterative compensation optimization in a database to ensure traceability and ease of subsequent use. The record format is "timestamp, dynamic deviation optimization parameter", such as "2023-04-01 12:00:00, optimized value 0.98".

[0034] In this embodiment, step S4 includes the following steps: The dynamic deviation optimization parameters are analyzed in multiple dimensions for each working condition to extract the vibration amplitude, frequency response and displacement phase difference of the spindle. The vibration amplitude, frequency response, and displacement phase difference are analyzed under multiple working conditions to generate the dynamic stiffness response characteristics for each working condition. The transient dynamic stiffness peak value is calculated for the dynamic stiffness response characteristics, and the transient dynamic stiffness peak value for each working condition is extracted. Based on the dynamic stiffness response characteristics, a long-term dynamic stiffness variation trend analysis is performed to generate long-term dynamic stiffness trend characteristics. Based on the transient dynamic stiffness peak value and long-term dynamic stiffness trend characteristics of each working condition, the comprehensive characteristics of time-series dynamic stiffness are fitted to construct dynamic stiffness characteristic curves under multiple working conditions.

[0035] In this embodiment, test data for each operating condition is extracted from the dynamic deviation optimization parameters, including vibration amplitude, frequency response, and displacement phase difference. The integrity and consistency of the data are ensured for subsequent analysis. The vibration amplitude (e.g., 0.5 g), frequency response (e.g., 50 Hz), and displacement phase difference (e.g., 30°) of the spindle under different operating conditions are recorded, ensuring the data format is "operating condition type, vibration amplitude, frequency response, displacement phase difference". For each operating condition, an individual analysis is performed, calculating the statistical characteristics of the vibration amplitude, frequency response, and displacement phase difference. This is achieved by comparing test data under different operating conditions, ensuring that the extracted features clearly reflect the dynamic performance of the spindle. Under operating condition A, the vibration amplitude is 0.5 g, the frequency response is 50 Hz, and the displacement phase difference is 30°; these characteristics are recorded to form a dataset. The extracted vibration amplitude, frequency response, and displacement phase difference for each operating condition are recorded in a database for subsequent analysis and comparison. Each record is ensured to accurately reflect the operating condition characteristics. The recording format is "Operating Condition Type, Vibration Amplitude, Frequency Response, Displacement Phase Difference", such as "Operating Condition A, 0.5g, 50 Hz, 30°". Dynamic stiffness refers to the stiffness characteristics of the spindle under vibration, usually defined as the ratio of vibration amplitude to applied force. Based on the extracted vibration amplitude and frequency response, the dynamic stiffness response characteristics for each operating condition are calculated. If the applied force is 100 N and the vibration amplitude is 0.5 g under operating condition A, the dynamic stiffness can be calculated as K = F / x = 100 N / 0.5 g. The test data for each operating condition are analyzed to calculate the corresponding dynamic stiffness response characteristics. This includes combining the vibration amplitude and frequency response to form a dynamic stiffness characteristic curve. If the dynamic stiffness of operating condition A is 2000 N / m, it is recorded as "Operating Condition A, Dynamic Stiffness 2000 N / m". The calculated dynamic stiffness response characteristics are recorded in the database to ensure that the characteristics of each operating condition are traceable. This provides the basic data for subsequent calculation of transient dynamic stiffness peak values. The recording format is "Working Condition Type, Dynamic Stiffness Response Characteristics", such as "Working Condition A, 2000 N / m". The transient dynamic stiffness peak value refers to the maximum stiffness characteristic exhibited by the spindle under instantaneous dynamic load. By analyzing the dynamic stiffness response characteristics, the transient peak value for each working condition is identified. In the dynamic stiffness curve under Working Condition A, the highest point is identified, which is the transient dynamic stiffness peak value.

[0036] Use data analysis tools (such as MATLAB or Python) to perform peak detection on the dynamic stiffness response characteristics to determine the peak transient dynamic stiffness under each operating condition. By finding the local maximum value of the dynamic stiffness curve, if the peak transient dynamic stiffness under operating condition A is 2500 N / m, record it as "Operating Condition A, Peak Transient Dynamic Stiffness 2500 N / m". Collect dynamic stiffness data of the spindle during long-term operation. Ensure the data covers long-term series under different operating conditions for trend analysis. Collect weekly dynamic stiffness measurement data over the past few months to form a time-series dataset. Select a suitable time-series analysis method, such as the moving average or linear regression, to analyze the long-term trend of dynamic stiffness. Ensure the method can capture the gradual changes in dynamic stiffness. Apply the moving average method to calculate the weekly average dynamic stiffness to observe its trend. Record the analyzed long-term dynamic stiffness trend characteristics in a database, ensuring the traceability of each characteristic to provide a basis for subsequent dynamic stiffness characteristic fitting. The recording format is "Time, Average Dynamic Stiffness", such as "2023-04-01, Dynamic Stiffness 2400 N / m". Combining the transient dynamic stiffness peak value and long-term dynamic stiffness trend characteristics for each working condition, a comprehensive time-series dynamic stiffness characteristic is defined. This ensures that the fitted model can reflect the dynamic characteristics of each working condition.

[0037] The comprehensive characteristic of time-series dynamic stiffness is defined as a "weighted average of the peak transient dynamic stiffness and the long-term trend." A suitable fitting model, such as a multinomial regression model or an exponential smoothing model, is selected to describe the changes in time-series dynamic stiffness. The model is ensured to accurately reflect the dynamic stiffness characteristics under different operating conditions. The dynamic stiffness data is fitted using a multinomial regression model to obtain dynamic stiffness characteristic curves. The fitted dynamic stiffness characteristic curves are recorded in a database, ensuring that the characteristics of each operating condition are traceable, and visualization charts are generated for subsequent analysis. The record format is "Operating Condition Type, Dynamic Stiffness Characteristic Curve," such as "Operating Condition A, Dynamic Stiffness Characteristic Curve."

[0038] In this embodiment, step S5 includes the following steps: Real-time acquisition of thermal imaging images from multiple operating conditions and excitation tests using a thermal imaging camera; The thermal imaging image is subjected to principal axis temperature distribution analysis to generate principal axis temperature distribution data; Multi-time-point temperature distribution fluctuation evolution of the spindle temperature distribution data is performed to construct a real-time temperature distribution fluctuation field; Based on the real-time temperature distribution fluctuation field, a quantitative correlation analysis of thermal expansion and dynamic stiffness changes was conducted on the dynamic stiffness characteristic curves under multiple working conditions to obtain the influence law of dynamic stiffness on thermal expansion.

[0039] In this embodiment, a thermal imaging camera is prepared and calibrated as necessary to ensure accurate temperature measurement capabilities. During calibration, a standard object with a known temperature should be used to ensure accurate readings from the thermal imaging camera. Calibration is performed using a water bath or hot block to ensure the temperature displayed by the camera matches the actual temperature, with the error controlled within ±1°C. The spindle is subjected to excitation tests under different operating conditions, and images acquired by the thermal imaging camera during operation are recorded. Excitation tests can include different loads, speeds, and temperature conditions to comprehensively evaluate the spindle's performance. Operating condition A is set as high load (100% load) and high speed (3000 RPM), and operating condition B is set as low load (50% load) and low speed (1500 RPM). Thermal imaging images are acquired under each operating condition. The images acquired by the thermal imaging camera are categorized and stored according to timestamps and operating condition types. The images should include the temperature distribution of the spindle under different operating conditions for subsequent analysis. The record format is "timestamp, operating condition type, image file name", such as "2023-04-01 10:00:00, operating condition A, image1.jpg". Select suitable image processing software (such as MATLAB or ImageJ) to process the thermal imaging images and extract the temperature distribution data of the main axis. These software programs typically have temperature measurement and image analysis capabilities, enabling the extraction of temperature values ​​from the thermal image. Use MATLAB's image processing toolbox to load the thermal imaging images and extract the corresponding temperature information. Perform temperature distribution analysis on each thermal imaging image to generate corresponding temperature distribution data. The data should include the temperature values ​​of each part of the main axis and their corresponding coordinates for subsequent fluctuation evolution analysis. Extract the temperature values ​​of the main axis from the image and generate a dataset in the format "coordinates, temperature value", such as "(x1, y1), 75°C". Record the generated main axis temperature distribution data in a database to ensure that the temperature data for each operating condition is traceable. The record should include the time of temperature distribution, operating condition type, and specific temperature value. Collect temperature distribution data collected at different time points to construct a time series dataset. These data should cover the temperature changes of the spindle under different operating conditions. Record the temperature distribution data of operating condition A at different time points to form a time-series dataset containing temperature values ​​from multiple tests. Use the collected time-series data to perform fluctuation evolution analysis on the spindle's temperature distribution. Statistical analysis methods, such as moving average or standard deviation, can be used to evaluate the temperature fluctuations at different time points. Calculate the temperature mean and standard deviation at each time point to generate a fluctuation field reflecting the temperature change trend of the spindle during operation. Extract the dynamic stiffness characteristic curve data for each operating condition from previous dynamic stiffness tests. The dynamic stiffness characteristic curve should include the relationship between dynamic stiffness values ​​and temperature under different loads and speeds. Record the dynamic stiffness characteristic curve for operating condition A, containing a dataset of "temperature, dynamic stiffness value", such as "75°C, 2000 N / m".Choose appropriate statistical analysis methods, such as regression analysis or correlation analysis, to establish a quantitative relationship between temperature and dynamic stiffness. Analyze the impact of temperature changes on dynamic stiffness to identify the influence of thermal expansion on dynamic stiffness. Use linear regression analysis to reveal the relationship between temperature changes and dynamic stiffness changes. Record the results of the quantitative correlation analysis in a database to ensure the traceability of the analysis results. Generate an analysis report showing the influence of temperature changes on dynamic stiffness for future research reference. The recording format is "Temperature, Dynamic Stiffness Influence Law," such as "For every 1°C increase in temperature, dynamic stiffness decreases by 50 N / m."

[0040] In this embodiment, step S6 includes the following steps: Obtain the spindle design drawing and perform 3D structural analysis to obtain the 3D structural features of the spindle; Identify the material properties of different components based on the spindle design drawings; Based on the material properties and the three-dimensional structural features of the main shaft, a three-dimensional point cloud model is constructed to build a three-dimensional geometric model of the main shaft. Based on the real-time temperature distribution fluctuation field, the three-dimensional geometric model of the principal axis is dynamically mapped to construct a real-time temperature distribution three-dimensional model. The dimensional changes under thermal expansion are calculated on the real-time temperature distribution three-dimensional model to extract the thermal expansion dimensional change value of each component; Based on the thermal expansion dimensional change values, the temperature gradient thermal stress distribution is mined to construct the temperature gradient thermal stress distribution characteristics. Based on the dynamic stiffness influence law, the thermal expansion stiffness change is predicted according to the temperature gradient thermal stress distribution characteristics, so as to obtain the predicted value of pure thermal expansion stiffness change. The dynamic stiffness characteristic curve is stripped of thermal expansion effect based on the predicted value of pure thermal expansion stiffness change, thereby obtaining accurate mechanical dynamic stiffness change characteristics. Based on the precise mechanical dynamic stiffness variation characteristics, a dynamic stiffness variation coupling analysis is performed to generate a spindle dynamic stiffness characteristic report.

[0041] In this embodiment, CAD design drawings of the spindle are obtained from the design department. These drawings should detail the spindle's geometry, the location and dimensions of each component, ensuring data integrity and accuracy. The obtained design drawings should include the spindle body, bearing housings, connecting components, etc., and the recording format is "design drawing file name, acquisition date," such as "main_spindle_design.dwg, 2023-04-01." The design drawings are imported using 3D modeling software (such as SolidWorks or CATIA) and 3D structural analysis is performed. During the analysis, the geometric features of the spindle need to be extracted, including length, diameter, curvature, etc. Through software analysis, the total length of the spindle is determined to be 500 mm, and the maximum diameter is 50 mm. The specific dimensions of each component are recorded. The analysis results are recorded in a database to ensure that the 3D structural features of each component are traceable. The record should include the component name, dimensions, and geometric features. The record format is "component name, length, diameter," such as "spindle body, 500 mm, 50 mm." Based on the spindle design drawings, the material properties of different components are identified. Common material properties include density, elastic modulus, and coefficient of thermal expansion, which affect the dynamic performance and thermal behavior of the spindle. If the main material of the spindle is alloy steel, the elastic modulus can be set to 210 GPa, and the coefficient of thermal expansion to 11.5 × 10⁻⁶ / °C. Use a material database (such as MatWeb or ASM Material Data) to query the detailed properties of the required materials to ensure data accuracy and reliability. When querying the material properties of alloy steel, the record format should be "Material Name, Elastic Modulus, Coefficient of Thermal Expansion," such as "Alloy Steel, 210 GPa, 11.5 × 10⁻⁶ / °C." Record the identified material properties in the database to ensure that the material properties of each component are traceable, facilitating subsequent modeling and analysis. The record format should be "Component Name, Material Type, Material Property," such as "Spindle, Alloy Steel, Elastic Modulus 210 GPa."

[0042] Based on the 3D structural features and material properties of the main axis, a 3D point cloud model is created. Point cloud data of the main axis can be generated using a laser scanner or 3D modeling software, ensuring the accuracy of the point cloud data. Scanning the main axis with a laser scanner yields point cloud data containing hundreds of thousands of data points, forming a 3D geometric model. The collected point cloud data is then imported into 3D modeling software for point cloud processing and modeling, generating the 3D geometric model of the main axis. During the modeling process, it is crucial to ensure that the model's details and dimensions match the design. Figure 1 The 3D model of the main axis generated by the modeling software should be consistent with the design. Figure 1This process involves collecting detailed data on the temperature distribution fluctuation field obtained from previous analyses. This data should reflect the temperature changes of the spindle under different operating conditions for dynamic mapping. Temperature distribution data at different time points should be recorded to form a dataset in the format of "time, temperature value". The real-time temperature distribution data should be combined with the spindle's 3D geometric model for dynamic mapping. Finite element analysis software (such as ANSYS or COMSOL) can be used to map the temperature data onto the 3D model. The 3D model should be loaded into the software, and the temperature data applied to the model surface to generate a real-time temperature distribution 3D model. The generated real-time temperature distribution 3D model should be saved and visualized for subsequent analysis and understanding. Ensure the model can intuitively reflect the temperature distribution. The recording format should be "real-time temperature model filename, generation date", such as "temperature_distribution_model.stl, 2023-04-01".

[0043] Based on the coefficient of thermal expansion of the material, calculate the dimensional changes of each component of the spindle at different temperatures. A commonly used calculation method is the linear expansion formula: ΔL = L0 × α × ΔT, where ΔL is the dimensional change, L0 is the original length, α is the coefficient of thermal expansion, and ΔT is the temperature change. If a component has an original length of 100 mm, a coefficient of thermal expansion of 11.5 × 10⁻⁶ / °C, and a temperature change of 50°C, then the dimensional change is ΔL = 100 mm × 11.5 × 10⁻⁶ × 50 = 0.0575 mm. Apply thermal expansion calculations to each component of the spindle and extract the dimensional change values ​​for each component. Ensure the accuracy and consistency of the calculation results. Record the dimensional change of each component in the format "Component Name, Dimensional Change of Thermal Expansion", such as "Spindle, 0.0575 mm". Based on the dimensional change values ​​of thermal expansion, calculate the thermal stress distribution of the spindle under the temperature gradient. The thermal stress formula can be used: σ = E × α × ΔT, where σ is the thermal stress and E is the elastic modulus. If the elastic modulus of a component is 210 GPa, the coefficient of thermal expansion is 11.5 × 10⁻⁶ / °C, and the temperature change is 50°C, then the thermal stress is σ = 210 GPa × 11.5 × 10⁻⁶ × 50 = 121.25 MPa. Using the calculated thermal stress values, the temperature gradient thermal stress distribution characteristics of each component of the spindle are constructed. This ensures that the analysis reflects stress changes under different working conditions. The thermal stress distribution of each component is recorded in the format "Component Name, Thermal Stress Value", such as "Shaft, 121.25 MPa". The thermal stress distribution characteristics are recorded in a database to ensure that the thermal stress data of each component is traceable and to provide a basis for subsequent stiffness change prediction. The recording format is "Component Name, Thermal Stress Characteristics", such as "Shaft, Thermal Stress 121.25 MPa". Based on the previous dynamic stiffness characteristic curve and the characteristics of thermal stress distribution under temperature gradient, analyze the influence of thermal expansion on dynamic stiffness. Ensure that the analysis results reflect the effect of temperature changes on dynamic stiffness. If a trend of decreasing dynamic stiffness with increasing temperature is observed, record it as "dynamic stiffness gradually decreases with increasing temperature".

[0044] A predictive model for changes in thermal expansion stiffness is established, typically using linear or multinomial regression methods. Temperature and thermal stress are used as independent variables, and dynamic stiffness as the dependent variable for fitting. Multinomial regression analysis is used to derive the relationship curve between dynamic stiffness and temperature, and the predicted values ​​are recorded. The predicted changes in thermal expansion stiffness are recorded in a database for subsequent analysis and optimization of the dynamic stiffness characteristic curve. The recording format is "Temperature, Predicted value of thermal expansion stiffness change," such as "75°C, dynamic stiffness decreases by 100 N / m." A suitable stripping method is selected to remove the thermal expansion effect from the dynamic stiffness characteristic curve. The difference method or regression analysis is typically used to subtract the influence of thermal expansion stiffness change from the dynamic stiffness curve. If the dynamic stiffness characteristic curve at a specific temperature is 2000 N / m, and the thermal expansion effect is 100 N / m, then the stripped dynamic stiffness is 1900 N / m. The dynamic stiffness characteristic curve under each operating condition is stripped to ensure that the obtained curve reflects the true mechanical dynamic stiffness change characteristics. Record the stripping results for each working condition, formatted as "Working Condition Type, Stripped Dynamic Stiffness," such as "Working Condition A, 1900 N / m." Record the stripped dynamic stiffness characteristics in a database for subsequent analysis and report generation. The recording format is "Working Condition Type, Precise Mechanical Dynamic Stiffness Variation Characteristics," such as "Working Condition A, Precise Dynamic Stiffness 1900 N / m." Select a suitable coupled analytical method, combining thermal expansion effects and dynamic stiffness variations for in-depth analysis. Finite element analysis or other numerical simulation methods can be used to evaluate the impact of dynamic stiffness variations on the overall spindle performance. Set up a finite element analysis model, considering the coupled effects of temperature and thermal stress on dynamic stiffness. Perform coupled analysis, generate a report on the dynamic stiffness variation characteristics of the spindle, and record the analysis results and key data. Ensure the results accurately reflect the dynamic performance of the spindle under different working conditions. Record the dynamic stiffness variation value, thermal stress characteristics, and related influencing factors for each working condition, forming a detailed analysis report. Organize the generated dynamic stiffness characteristic reports for subsequent review and decision-making. Ensure the report is detailed and the data is accurate, providing a reference for subsequent optimization. The report should include information such as "operating condition type, dynamic stiffness variation characteristics, and thermal stress distribution," ensuring completeness.

[0045] In this embodiment, a spindle dynamic stiffness testing and analysis device is provided for performing the spindle dynamic stiffness testing and analysis method described above, including: The multi-condition testing module is used to perform multi-condition operation excitation tests on the spindle and collect spindle test response parameters for multiple conditions based on a multi-dimensional high-precision sensor array. The adaptive filtering module is used to perform dynamic noise cleaning and adaptive filtering optimization on the spindle test response parameters, and generate adaptive filtering optimized response parameters. The deviation compensation module is used to dynamically test and predict the deviation of the adaptive filtering optimization response parameters, and then perform multiple rounds of deviation iterative compensation optimization to generate dynamic deviation optimization parameters. The dynamic stiffness characteristic module is used to calculate the peak value of transient dynamic stiffness for dynamic deviation optimization parameters, and then to analyze the long-term dynamic stiffness change trend to construct dynamic stiffness characteristic curves under multiple working conditions. The thermal expansion correlation module is used to acquire thermal imaging images of multi-condition operation excitation tests and perform quantitative correlation analysis based on the dynamic stiffness characteristic curve to obtain the dynamic stiffness influence law of thermal expansion. The thermal expansion stripping module is used to predict the change in thermal expansion stiffness based on the dynamic stiffness influence law, and to perform thermal expansion effect stripping processing to generate a spindle dynamic stiffness characteristic report.

[0046] This invention utilizes multi-condition testing to cover the spindle's performance under various loads and speeds, providing abundant experimental data to facilitate a comprehensive understanding of the spindle's dynamic performance. A high-precision sensor array ensures accurate acquisition of spindle response data under different operating conditions, reducing signal noise and errors and improving test accuracy. This multi-dimensional data provides a foundation for subsequent analyses (such as dynamic filtering and dynamic stiffness analysis), contributing to more accurate performance evaluation. Dynamic noise removal effectively identifies and removes unnecessary signals such as environmental noise and mechanical interference, improving signal quality. An adaptive filtering algorithm automatically adjusts filtering parameters based on signal variation characteristics under different operating conditions, ensuring optimal signal processing under varying conditions and reducing unnecessary signal loss. The filtered and optimized response data is more accurate, providing high-quality input data for subsequent dynamic deviation prediction and dynamic stiffness analysis. Real-time prediction and compensation for potential deviations during testing continuously optimizes data and reduces error accumulation. Multi-round iterative deviation compensation ensures data quality in each round of testing, significantly enhancing the reliability of spindle test results. This process helps eliminate test errors caused by environmental fluctuations and equipment instability, resulting in more accurate and reliable final test results. Transient dynamic stiffness peak calculations can assess the spindle's stiffness performance under start-up, shutdown, and other rapidly changing operating conditions, helping to identify performance bottlenecks. Long-term dynamic stiffness trend analysis reveals potential stiffness decay or variation patterns during long-term operation, providing valuable information for equipment maintenance and optimization. Dynamic stiffness characteristic curves visually display spindle stiffness changes under different operating conditions, helping users fully understand the spindle's dynamic performance under various load conditions. Thermal imaging images reflect real-time spindle temperature changes, revealing the impact of temperature fluctuations on dynamic stiffness. By combining temperature data with dynamic stiffness characteristics, the influence of thermal expansion on spindle dynamic stiffness can be quantified, further improving the accuracy of dynamic stiffness testing. Clearly defining the impact of thermal expansion on dynamic stiffness provides a basis for spindle temperature control design and optimizes the spindle's thermal management system. By stripping away the thermal expansion effect, pure mechanical dynamic stiffness characteristics are obtained, ensuring results are unaffected by temperature fluctuations. Based on the dynamic stiffness characteristic report, accurate predictions of spindle dynamic stiffness changes under different temperatures and operating conditions can be made, providing important reference for spindle design, commissioning, and maintenance. The report provides a clear trend of dynamic stiffness variation, the effects of thermal expansion, and optimization suggestions to help make more reasonable decisions on equipment commissioning and optimization.

[0047] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0048] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for testing and analyzing the dynamic stiffness of a spindle, characterized in that, The spindle is placed on a dynamic loading test platform, which is equipped with a multi-dimensional high-precision sensor array, and includes the following steps: Step S1: Perform multi-condition operation excitation test on the spindle, and collect spindle test response parameters for multiple conditions based on a multi-dimensional high-precision sensor array; Step S2: Perform dynamic noise cleaning and adaptive filtering optimization on the spindle test response parameters to generate adaptive filtering optimized response parameters; Step S3: Perform dynamic test deviation prediction on the adaptive filter optimization response parameters, and then perform multiple rounds of deviation iterative compensation optimization to generate dynamic deviation optimization parameters; Step S4: Calculate the peak value of transient dynamic stiffness for the dynamic deviation optimization parameters, then analyze the long-term dynamic stiffness change trend, and construct dynamic stiffness characteristic curves under multiple working conditions; Step S5: Acquire thermal imaging images of multi-condition operation excitation test, and perform quantitative correlation analysis based on the dynamic stiffness characteristic curve to obtain the dynamic stiffness influence law of thermal expansion. Step S6: Based on the dynamic stiffness influence law, predict the change in thermal expansion stiffness, and perform thermal expansion effect stripping processing to generate a spindle dynamic stiffness characteristic report.

2. The spindle dynamic stiffness testing and analysis method according to claim 1, characterized in that, The specific steps of step S1 are as follows: Obtain the preset spindle operation monitoring log; Multi-condition operating parameters are defined for the spindle operation monitoring log to obtain operating test parameters for multiple conditions; Calculate the duration of multiple historical peak operating times based on the spindle operation monitoring logs; The extreme condition test time is calculated based on the duration of the multiple historical peak operating times. Based on the operating test parameters and extreme operating test time under multiple working conditions, the spindle is subjected to multi-working condition operating excitation test through a dynamic loading test platform, and the spindle test response parameters under multiple working conditions are collected based on a multi-dimensional high-precision sensor array.

3. The spindle dynamic stiffness testing and analysis method according to claim 2, characterized in that, The specific steps for defining multi-condition operating parameters in the spindle operation monitoring log to obtain operating test parameters for multiple conditions are as follows: Perform routine operation analysis on the spindle operation monitoring log to identify different operation scenarios; Calculate the load frequency and amplitude for the different operating scenarios; Based on the load frequency and amplitude, perform multi-condition load characteristic analysis to obtain the load characteristics of each scenario; Calculate the natural frequencies and modal characteristics for the different operating scenarios; Based on the inherent frequency and modal characteristics, the main modal characteristics are mined to obtain the constant frequency resonant mode for each scene; Based on the load characteristics of each operating condition and the constant frequency resonant mode of each operating condition, the operating parameters of multiple operating conditions are defined, thereby obtaining the operating test parameters of multiple operating conditions.

4. The spindle dynamic stiffness testing and analysis method according to claim 1, characterized in that, The specific steps of step S2 are as follows: The spindle test response parameters are decomposed into multiple time-frequency parameters to obtain the test response time-frequency spectrum. Non-vibrational component features were identified from the time-frequency spectrum of the test response, and environmental noise points were extracted. The signal-to-noise ratio of the environmental noise points is calculated, and dynamic noise cleaning is performed to obtain noise optimization test response parameters. Dynamic response parameter change analysis is performed on the noise optimization test response parameters to generate dynamic response change characteristics; Adaptive filtering optimization is performed based on the dynamic response change characteristics to generate adaptive filtering optimization response parameters.

5. The spindle dynamic stiffness testing and analysis method according to claim 4, characterized in that, Step S3 is as follows: The frequency of each sensor signal is calculated by optimizing the response parameters of the adaptive filtering to obtain the sensor signal frequency; The sampling frequency of each sensor in a multi-dimensional high-precision sensor array is calculated to generate the expected sampling signal frequency for each sensor. Based on the expected sampling signal frequency, perform sampling signal distortion analysis on the sensor signal frequency to identify sensor sampling signal distortion information; The environmental noise points are analyzed for time-series variation characteristics to generate noise time-series variation features; Dynamic test deviation prediction is performed based on the distortion information of sensor sampling signals and the temporal variation characteristics of noise, so as to obtain the dynamic test deviation value; Based on the dynamic test deviation value, the adaptive filtering optimization response parameters are optimized through multiple rounds of deviation iteration compensation to generate dynamic deviation optimization parameters.

6. The spindle dynamic stiffness testing and analysis method according to claim 1, characterized in that, The specific steps of step S4 are as follows: The dynamic deviation optimization parameters are analyzed in multiple dimensions for each working condition to extract the vibration amplitude, frequency response and displacement phase difference of the spindle. The vibration amplitude, frequency response, and displacement phase difference are analyzed under multiple working conditions to generate the dynamic stiffness response characteristics for each working condition. The transient dynamic stiffness peak value is calculated for the dynamic stiffness response characteristics, and the transient dynamic stiffness peak value for each working condition is extracted. Based on the dynamic stiffness response characteristics, a long-term dynamic stiffness variation trend analysis is performed to generate long-term dynamic stiffness trend characteristics. Based on the transient dynamic stiffness peak value and long-term dynamic stiffness trend characteristics of each working condition, the comprehensive characteristics of time-series dynamic stiffness are fitted to construct dynamic stiffness characteristic curves under multiple working conditions.

7. The spindle dynamic stiffness testing and analysis method according to claim 1, characterized in that, The specific steps of step S5 are as follows: Real-time acquisition of thermal imaging images from multiple operating conditions and excitation tests using a thermal imaging camera; The thermal imaging image is subjected to principal axis temperature distribution analysis to generate principal axis temperature distribution data; Multi-time point temperature distribution fluctuation evolution is performed on the spindle temperature distribution data to construct a real-time temperature distribution fluctuation field; Based on the real-time temperature distribution fluctuation field, a quantitative correlation analysis of thermal expansion and dynamic stiffness changes was conducted on the dynamic stiffness characteristic curves under multiple working conditions to obtain the influence law of dynamic stiffness on thermal expansion.

8. The spindle dynamic stiffness testing and analysis method according to claim 1, characterized in that, The specific steps of step S6 are as follows: Obtain the spindle design drawing and perform 3D structural analysis to obtain the 3D structural features of the spindle; Identify the material properties of different components based on the spindle design drawings; Based on the material properties and the three-dimensional structural features of the main shaft, a three-dimensional point cloud model is constructed to build a three-dimensional geometric model of the main shaft. Based on the real-time temperature distribution fluctuation field, the three-dimensional geometric model of the principal axis is dynamically mapped to construct a real-time temperature distribution three-dimensional model. The dimensional changes under thermal expansion are calculated on the real-time temperature distribution three-dimensional model to extract the thermal expansion dimensional change value of each component; Based on the thermal expansion dimensional change values, the temperature gradient thermal stress distribution is mined to construct the temperature gradient thermal stress distribution characteristics. Based on the dynamic stiffness influence law, the thermal expansion stiffness change is predicted according to the temperature gradient thermal stress distribution characteristics, so as to obtain the predicted value of pure thermal expansion stiffness change. The dynamic stiffness characteristic curve is stripped of thermal expansion effect based on the predicted value of pure thermal expansion stiffness change, thereby obtaining accurate mechanical dynamic stiffness change characteristics. Based on the precise mechanical dynamic stiffness variation characteristics, a dynamic stiffness variation coupling analysis is performed to generate a spindle dynamic stiffness characteristic report.

9. A spindle dynamic stiffness testing and analysis system, characterized in that, For performing the method as described in claim 1, comprising: The multi-condition testing module is used to perform multi-condition operation excitation tests on the spindle and collect spindle test response parameters for multiple conditions based on a multi-dimensional high-precision sensor array. The adaptive filtering module is used to perform dynamic noise cleaning and adaptive filtering optimization on the spindle test response parameters, and generate adaptive filtering optimized response parameters. The deviation compensation module is used to dynamically test and predict the deviation of the adaptive filtering optimization response parameters, and then perform multiple rounds of deviation iterative compensation optimization to generate dynamic deviation optimization parameters. The dynamic stiffness characteristic module is used to calculate the peak value of transient dynamic stiffness for dynamic deviation optimization parameters, and then to analyze the long-term dynamic stiffness change trend to construct dynamic stiffness characteristic curves under multiple working conditions. The thermal expansion correlation module is used to acquire thermal imaging images of multi-condition operation excitation tests and perform quantitative correlation analysis based on the dynamic stiffness characteristic curve to obtain the dynamic stiffness influence law of thermal expansion. The thermal expansion stripping module is used to predict the change in thermal expansion stiffness based on the dynamic stiffness influence law, and to perform thermal expansion effect stripping processing to generate a spindle dynamic stiffness characteristic report.

Images