Abrasion compensation method for grinding sleeve

Abstract

The invention relates to a grinding sleeve wear compensation method which comprises the following steps: 1, in an offline preparation stage, carrying out model parameter calibration to obtain specific numerical values of a wear coefficient, a temperature correction coefficient, a feed rate correction coefficient and a viscosity correction coefficient, and storing the specific numerical values to an equipment control module; 2, in the online operation stage, real-time data collection is carried out, and collected data comprise grinding contact pressure, the rotating speed of a grinding shaft, the temperature variation, the workpiece feeding speed and the viscosity of a grinding medium; 3, according to the established actual wear predicted value model, carrying out real-time prediction calculation on the wear loss to obtain an actual wear predicted value; 4, a compensation algorithm is selected and executed, if the equipment is pressure controllable equipment, a pressure compensation algorithm is executed, and if the equipment is pressure fixed equipment, a feeding speed compensation algorithm is executed; and 5, carrying out iterative optimization on the precision of the actual wear predicted value model. The problems that a traditional grinding sleeve is large in abrasion prediction deviation and poor in compensation algorithm adaptability are solved, and the method is particularly suitable for scenes such as precise grinding of ship inertial navigation equipment and the like with extremely high machining precision requirements.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of grinding wear compensation, and specifically designs a grinding sleeve wear compensation method. BACKGROUND

[0002] In many industrial production and processing fields, grinding is a crucial process, widely used in mechanical manufacturing, optical instruments, electronic equipment and other industries, for precision machining of various materials and parts to obtain the required dimensional accuracy, surface finish and shape accuracy.

[0003] In the grinding process, the grinding sleeve, as the key component directly contacting the workpiece being ground and performing the grinding operation, its performance and state play a decisive role in the grinding quality and efficiency. Grinding sleeve wear is a key factor affecting the grinding processing quality. The existing wear prediction model only considers a single parameter (such as pressure or time), without fully integrating temperature, feed rate and other actual working condition disturbances, resulting in large prediction deviation. At the same time, the compensation algorithm is single and cannot adapt to different pressure control requirements of grinding equipment. Therefore, a multi-parameter corrected wear prediction model needs to be constructed, and a switchable compensation algorithm needs to be designed to adapt to diversified grinding scenarios. SUMMARY

[0004] The present application proposes a grinding sleeve wear compensation method to solve the problems of large prediction deviation of traditional grinding sleeve wear and poor adaptability of compensation algorithm The above-mentioned purpose of the present application is realized by the following technical scheme: A grinding sleeve wear compensation method, comprising the following steps: Step 1, in the offline preparation stage, model parameter calibration is performed to obtain the specific values of wear coefficient, temperature correction coefficient, feed rate correction coefficient and viscosity correction coefficient, which are stored in the device control module; Step 2, in the online operation stage, real-time data acquisition is performed, and the acquired data includes grinding contact pressure, grinding shaft speed, temperature change, part feed speed and grinding medium viscosity; Step 3, according to the established actual wear prediction value model, real-time wear amount prediction calculation is performed to obtain the actual wear prediction value; Step 4, compensation algorithm selection and execution, if it is a pressure controllable device, pressure compensation algorithm is executed, if it is a pressure fixed device, feed speed compensation algorithm is executed; Step 5, iteration optimization of the accuracy of the actual wear prediction value model Furthermore, in step 1, an orthogonal experimental method was used to simulate typical working conditions of inertial navigation component grinding, testing the actual grinding amount of the grinding sleeve under at least 30 different working conditions; the actual grinding amount of the grinding sleeve was measured using a high-precision measuring instrument with an accuracy of 0.1 μm. Then, the experimental data was fitted using the least squares method to obtain the wear coefficient k and the temperature correction coefficient. Feed rate correction factor Viscosity correction factor The specific value.

[0005] Furthermore, step 2 includes: first, acquiring the following parameters using the device's sensor system at a sampling frequency of 1Hz: Grinding contact pressure P, grinding shaft speed n, temperature change Workpiece feed rate f, grinding media viscosity The collected data is then filtered to remove outliers caused by electromagnetic interference; finally, it is transmitted to the computing unit of the control module.

[0006] Furthermore, step 3 includes: firstly, the control module first... Calculate the sliding speed Where D is the diameter of the grinding sleeve currently in use, preset by the operator in the system; then the grinding contact pressure P and the grinding relative sliding speed are... The single grinding time t and the surface hardness H are input into the basic model to obtain the wear amount W of the grinding sleeve. The formula of the basic model is: k is the wear coefficient; combined with the temperature change during the grinding process... Workpiece feed speed f, dynamic viscosity of grinding media Substituting into the multi-condition correction formula, the actual wear prediction value W is calculated. The actual wear prediction value model formula is as follows: , This is the temperature correction factor. This is the feed rate correction factor. This is the viscosity correction factor.

[0007] Furthermore, in step 4, the pressure compensation formula is: P_compensated = P_initial (Wmax / (Wmax-Wcorrection)), where P_initial: the initial grinding pressure preset by the process, in MPa; Wmax: the maximum allowable wear of the grinding sleeve, in μm; W_correction: the real-time predicted wear, in μm.

[0008] Furthermore, in step 4, the feed rate compensation formula is: f_compensation = f_initial (1-W correction / Wmax), where f_initial: the initial feed rate preset by the process, in mm / min.

[0009] Further, step 5 comprises: after each 10 grinding processes are completed, the actual wear W actual of the grinding sleeve is measured by using a high-precision optical profiler; and then the prediction deviation is calculated If the prediction deviation is greater than a preset threshold, the k value is re-fitted and corrected, and the model accuracy is iteratively optimized.

[0010] The present application has the advantages and positive effects that: 1. The present application establishes a prediction model with multi-working condition parameter fusion: for the first time, a grinding medium viscosity correction term is introduced into the Archard model, and the temperature and the feed amount are combined to cover the full working condition interference of the ship dummy film, and the prediction accuracy is improved by more than 30% compared with the traditional single parameter model. The deviation can be controlled within 5 microns through experimental verification.

[0011] 2. The present application adopts a double-path adaptive compensation algorithm: the compensation logic can be automatically switched according to the pressure control capability of the equipment, which is suitable for the scene where new and old grinding equipment coexist in the workshop, and the wear compensation can be realized without replacing the equipment, thereby reducing the upgrade cost.

[0012] 3. The present application sets an engineering iterative optimization mechanism: the parameter correction strategy verified by processing solves the model drift problem caused by the change of the surface state of the grinding sleeve (such as initial running-in and later aging), and ensures the stability of long-term processing, which is especially suitable for high-precision and long-period production scenes such as inertial navigation equipment. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 is the structure logic diagram of the grinding sleeve wear prediction compensation model of the present application; Figure 2 is the flowchart of the grinding sleeve wear compensation method of the present application. DETAILED DESCRIPTION

[0014] The structure of the present application will be further described below in combination with the drawings and through actual examples.

[0015] The present application solves the problems of large prediction deviation and poor adaptability of the compensation algorithm of the traditional grinding sleeve wear prediction by using the “multi-working condition parameter correction prediction model + double-path adaptive compensation algorithm”, which is especially suitable for precise grinding of ship inertial navigation equipment and other scenes with extremely high requirements on processing precision.

[0016] The present actual example explains in detail the model principle, algorithm details, implementation process and key innovation points: I. Establishing a multi-working condition parameter correction prediction model Performing in-depth analysis of the wear prediction compensation model ​The core of the model is based on the classic Archard wear model, and integrates common interference parameters in ship grinding and processing such as temperature, feed rate, and grinding media viscosity to achieve accurate prediction of "basic wear amount + working condition correction amount", which is divided into two levels. The established basic wear model for modifying the Archard model The Archard model is a classic theory for predicting mechanical wear. This invention, for the grinding sleeve processing scenario, makes an engineering correction to the "sliding speed" parameter in the model to make it more consistent with the actual equipment operation logic.

[0017] Core formula: Detailed parameter definition: W: Basic wear amount of the grinding sleeve, unit: μm, i.e. micrometer. This unit is commonly used in precision grinding of ships to measure wear accuracy. k: Wear coefficient, dimensionless, related to the compatibility of the grinding sleeve material and the workpiece material, and needs to be calibrated through orthogonal experiments in marine grinding scenarios. For example, when using a zirconia grinding sleeve to process stainless steel workpieces, the k value is typically within a certain range. interval; P: Grinding contact pressure, unit: MPa. In the grinding of marine inertial navigation components, the pressure is usually controlled at 0.5-2 MPa to avoid workpiece deformation. V: Relative sliding velocity in grinding, unit: m / s. The innovation lies in its engineering implementation. Where D is the diameter of the grinding sleeve, and the diameter of the grinding sleeve commonly used in ships is 50-200mm; n is the rotational speed of the grinding shaft, which is generally 100-500r / min and is read in real time by the equipment frequency converter; t: Single grinding processing time, unit: s, preset by the processing technology document; H: Hardness of the ground surface, in units of HV, i.e., Vickers stress, determined by the material manufacturer's parameters.

[0018] Established multi-condition correction model Traditional models only consider single parameters such as pressure and time. However, in ship grinding, fluctuations in ambient temperature, feed rate adjustments, and changes in the viscosity of grinding media (such as grinding paste) significantly affect wear patterns. This invention adds three key correction terms, forming a predictive model that covers all working conditions: Core formula: Detailed explanation of the correction items: Temperature correction item ( ): Temperature change during the grinding process, unit: Data is collected in real time by thermocouple sensors attached to the outer wall of the grinding sleeve. The ambient temperature in the workshop is typically controlled at 20±5℃. However, the local temperature may rise by 5-15 degrees Celsius during grinding. ; Temperature correction factor, unit: 1 / , calibrated through experiments; Feed rate correction term ( ): Workpiece feed speed, unit: mm / min, read by servo motor encoder; Feed rate correction factor, unit: min / mm, calibrated value is about 0.02-0.05 min / mm. The faster the feed rate, the larger the grinding contact area per unit time, and the wear amount increases proportionally.

[0019] Grinding media viscosity correction item ( ): Dynamic viscosity of grinding media (such as grinding paste or grinding fluid), in units The viscosity data is collected in real time by an online viscosity sensor. Viscosity correction factor, unit: 1 / The calibration range is clearly defined as 0.05-0.1 / The higher the viscosity, the stronger the lubrication and protection of the grinding sleeve by the medium, and the more significant the reduction in wear.

[0020] II. Dual-path adaptive compensation algorithm 1. Pressure compensation algorithm: Adaptable to pressure-controllable equipment Objective: To maintain a constant grinding efficiency (i.e., a stable amount of workpiece removed per unit time) by adjusting the grinding contact pressure, and to avoid a decrease in machining accuracy due to wear of the grinding sleeve.

[0021] Compensation formula: P_compensation = P_initial (Wmax / (Wmax-W correction)) Key parameters: P_initial: The initial grinding pressure preset by the process, in MPa; Wmax: Maximum allowable wear of the grinding sleeve, unit: μm; W correction: Real-time predicted wear amount, unit: μm, calculated by the multi-condition correction model; 2. Feed rate compensation algorithm: adapted to pressure-fixed equipment Objective: When the equipment cannot adjust the pressure, reduce the workpiece feed speed to compensate for the decrease in grinding capability caused by wear of the grinding sleeve, and ensure the final dimensional accuracy of the workpiece.

[0022] Compensation formula: f_compensation = f_initial (1-W correction / Wmax) Key parameters: f_initial: The initial feed rate preset by the process, in mm / min.

[0023] The grinding sleeve wear compensation method based on the above-mentioned "predictive model with multi-condition parameter correction + dual-path adaptive compensation algorithm" needs to combine five steps: "experimental calibration - real-time acquisition - prediction calculation - compensation execution - accuracy iteration" to ensure stable operation in actual grinding scenarios. The specific steps are as follows: Step 1: Perform model parameter calibration during the offline preparation phase. Orthogonal experimental methods were used to simulate typical working conditions of inertial navigation component grinding (such as different pressures of 0.5-2 MPa, rotation speeds of 100-500 r / min, and temperatures of 20-35°C). Feed rate: 0.1-1 mm / min; Medium viscosity: 10-50. The actual grinding amount of the grinding sleeve was tested under at least 30 different working conditions. The actual grinding amount of the grinding sleeve was measured by a high-precision measuring instrument with an accuracy of 0.1μm.

[0024] The wear coefficient k and temperature correction coefficient were obtained by fitting the experimental data using the least squares method. Feed rate correction factor Viscosity correction factor Specific values ​​(e.g., in the case of grinding inertial navigation components, k= , , , =0.08 / And store it in the device control module.

[0025] Step 2: Real-time data collection during online operation. The following parameters are collected using the equipment's sensor system at a sampling frequency of 1Hz (balancing real-time performance and data volume; 1 sampling per second is sufficient to meet the dynamic response requirements of inertial navigation component grinding): Grinding contact pressure P (measured by a pressure sensor), grinding shaft speed n (measured by an encoder), and temperature change. (Measured by thermocouple), workpiece feed speed f (measured by servo motor encoder), and grinding media viscosity. (Measured by an online viscosity sensor). The collected data is filtered to remove outliers caused by electromagnetic interference. In particular, the Kalman filter algorithm is often used in the inertial navigation component processing workshop. Finally, the filtered data is transmitted to the computing unit of the control module.

[0026] Step 3: Perform real-time wear prediction calculation The control module first based on Calculate the sliding speed Where D is the diameter of the grinding sleeve currently in use, which is preset by the operator in the system; Then P, Substituting t (processing time already consumed, recorded by the system timer) and H (hardness of the grinding sleeve, a preset parameter) into the basic model, we obtain W; Combined with the collected data f Substituting into the multi-condition correction formula, the actual wear prediction value W is calculated and corrected.

[0027] Step 4: Compensation Algorithm Selection and Execution The control module reads the device parameters (pre-entered "whether pressure adjustment is supported" flag) and automatically selects the compensation path: If it is a pressure-controllable device, such as a CNC grinding machine, then the pressure compensation algorithm is executed to calculate the P compensation and output a 4-20mA analog signal to the pressure regulating valve to adjust the grinding pressure in real time; If it is a pressure-fixed device, such as an old hydraulic press, then the feed speed compensation algorithm is executed to calculate the compensation output control signal to the feed servo motor and adjust the feed speed.

[0028] Step 5: Iterative optimization of model accuracy After every 10 grinding processes (small batch production in the workshop, 10 pieces as one inspection unit), the actual wear amount W of the grinding sleeve is measured using a high-precision optical profilometer.

[0029] Calculate prediction bias ,like (The allowable deviation threshold for precision grinding) is then used to refit and correct the k value (the wear coefficient is most affected by the surface condition of the grinding sleeve, so it is corrected first), and iteratively optimize the model accuracy to ensure long-term prediction accuracy.

[0030] Although practical examples and drawings of the invention have been disclosed for illustrative purposes, those skilled in the art will understand that various substitutions, variations and modifications are possible without departing from the spirit and scope of the invention and the appended claims. Therefore, the scope of the invention is not limited to the contents disclosed in the practical examples and drawings.

Claims

1. A method for compensating wear on a grinding sleeve, characterized in that: Includes the following steps: Step 1: In the offline preparation stage, the model parameters are calibrated to obtain the specific values ​​of wear coefficient, temperature correction coefficient, feed rate correction coefficient, and viscosity correction coefficient, and then stored in the equipment control module. Step 2, during the online operation phase, real-time data acquisition is performed. The acquired data includes: grinding contact pressure, grinding shaft speed, temperature change, workpiece feed speed, and grinding media viscosity. Step 3: Based on the established actual wear prediction model, perform real-time wear prediction calculations to obtain the actual wear prediction value; Step 4: Select and execute the compensation algorithm. If the pressure is controllable, execute the pressure compensation algorithm; if the pressure is fixed, execute the feed rate compensation algorithm. Step 5: Iteratively optimize the accuracy of the actual wear prediction model.

2. The grinding sleeve wear compensation method according to claim 1, characterized in that: In step 1, an orthogonal experimental method was used to simulate typical working conditions of inertial navigation component grinding, and the actual grinding amount of the grinding sleeve under at least 30 different working conditions was tested. The actual grinding amount of the grinding sleeve was measured using a high-precision measuring instrument with an accuracy of 0.1 μm. Then, the experimental data were fitted using the least squares method to obtain the wear coefficient k and the temperature correction coefficient. Feed rate correction factor Viscosity correction factor The specific value.

3. The grinding sleeve wear compensation method according to claim 1, characterized in that: Step 2 includes: first, using the equipment's sensor system, collecting the following parameters at a sampling frequency of 1Hz: grinding contact pressure P, grinding shaft speed n, and temperature change. Workpiece feed rate f, grinding media viscosity The collected data is then filtered to remove outliers caused by electromagnetic interference; finally, it is transmitted to the computing unit of the control module.

4. The grinding sleeve wear compensation method according to claim 1, characterized in that: Step 3 includes: First, the control module first... Calculate the sliding speed Where D is the diameter of the grinding sleeve currently in use, preset by the operator in the system; then the grinding contact pressure P and the grinding relative sliding speed are... The single grinding time t and the surface hardness H are input into the basic model to obtain the wear amount W of the grinding sleeve. The formula of the basic model is: k is the wear coefficient; combined with the temperature change during the grinding process... Workpiece feed speed f, dynamic viscosity of grinding media Substituting into the multi-condition correction formula, the actual wear prediction value W is calculated. The actual wear prediction value model formula is as follows: , This is the temperature correction factor. This is the feed rate correction factor. This is the viscosity correction factor.

5. The grinding sleeve wear compensation method according to claim 1, characterized in that: In step 4, the pressure compensation formula is: P_compensated = P_initial (Wmax / (Wmax-Wcorrection)), where P_initial: the initial grinding pressure preset by the process, in MPa; Wmax: the maximum allowable wear of the grinding sleeve, in μm; W_correction: the real-time predicted wear, in μm.

6. The grinding sleeve wear compensation method according to claim 1, characterized in that: In step 4, the feed rate compensation formula is: f_compensation = f_initial (1-W correction / Wmax), where f_initial: the initial feed rate preset by the process, in mm / min.

7. The grinding sleeve wear compensation method according to claim 1, characterized in that: Step 5 includes: after every 10 grinding cycles, measuring the actual wear amount Wactual of the grinding sleeve using a high-precision optical profilometer; and then calculating the predicted deviation. ,like Then, the k value is refitted and iteratively optimized to improve the model accuracy.

Images