An automated control system and method for turning an agricultural drive shaft

Abstract

The application relates to the technical field of agricultural machinery accessory machining and manufacturing, and discloses an automatic control system and method for turning processing of an agricultural transmission shaft, which comprises the following steps: synchronously acquiring cutting force and motion parameters to calculate instantaneous mechanical energy input power; combining a tool tip coordinate and a main shaft phase angle to solve transient equivalent radial stiffness; introducing centrifugal force generated by eccentric mass at the end of a workpiece to analyze transient radial retreat displacement and construct a dimensionless damage energy dissipation ratio; updating particle swarm optimization parameters according to the ratio to calculate optimal angular velocity and a feed speed; and finally converting the predicted residual retreat displacement into active interference compensation displacement, and converting the active interference compensation displacement into control voltage and torque instructions to perform accurate servo control. The application effectively suppresses dynamic tool deflection in the machining of a variable cross-section shaft, and improves turning precision.

Description

Technical Field

[0001] This invention relates to the field of agricultural machinery parts processing and manufacturing technology, and more specifically, to an automated control system and method for turning agricultural drive shafts. Background Technology

[0002] Agricultural machinery drive shafts (such as power take-off shafts) are typically equipped with rigid joints at both ends, characterized by large mass and significant eccentricity. The middle section, to achieve sliding contact, often employs an extremely thin-walled, non-axisymmetric shape (such as a lemon-shaped or star-shaped tube). In traditional turning processes, automated control systems mostly rely on pre-set constant feed rates or fixed proportional-integral-derivative (PID) control parameters. However, when the tool is turning an irregularly shaped thin-walled tube section while the spindle rotates at high speed, the moment of inertia of the workpiece's middle section will pulsate at high frequency with the rotation angle, causing its bending resistance to exhibit a non-linear collapse in the spatial axial direction. Simultaneously, the asymmetrical joints at both ends will generate strong rotational centrifugal forces, further weakening the system's dynamic compressive stiffness. Traditional static control methods cannot perceive and dynamically respond to this multi-state superposition of time-varying tool deflection phenomena in real time. This not only easily leads to deviations in the tool's trajectory from expectations but also easily leaves chatter ripples of varying depths on the workpiece surface, restricting the machining accuracy and production yield of the drive shaft assembly surfaces. Summary of the Invention

[0003] This invention provides an automated control system and method for turning agricultural drive shafts, which solves the technical problems mentioned in the background art.

[0004] Firstly, an automated control method for turning agricultural drive shafts is provided, applied to machining equipment comprising a cutting tool, a force measuring module, a spindle encoder, a servo piezoelectric tool holder, and a spindle servo motor, including: Simultaneously acquire the cutting force parameters and motion parameters of the cutting interface, and calculate the instantaneous mechanical energy input power of the system; Extract the current absolute axial coordinate of the tool and the instantaneous absolute phase angle of the spindle rotation, and calculate the transient equivalent radial stiffness; The equivalent eccentric mass at the end of the workpiece is obtained, and the high-speed rotating centrifugal force generated by it is superimposed with the instantaneous cutting force and divided by the dynamic attenuation stiffness to analyze the transient radial yield displacement. Calculate the strain potential energy generated by the transient radial yield displacement, divide it by the total mechanical work done based on the input power integral within one rotation cycle of the main shaft, and generate the destructive energy dissipation ratio; The optimization parameters of the preset particle swarm algorithm are updated using the dissipation ratio, and the optimal machining angular velocity and optimal feed rate that minimize the ratio are calculated. Substitute the optimal machining angular velocity and the optimal feed rate into the yield prediction model to calculate the residual yield displacement and invert it to convert it into an active interference compensation displacement. The compensated displacement and the optimal machining angular velocity are converted into the piezoelectric control voltage input to the servo piezoelectric tool holder and the target rigid torque command input to the spindle servo motor.

[0005] Secondly, an automated control method for turning agricultural drive shafts, applied in any of the automated control systems described in the claims, includes: The data acquisition module synchronously acquires the cutting force parameters and motion parameters of the cutting interface and calculates the instantaneous mechanical energy input power of the system. The transient equivalent radial stiffness calculation module extracts the current absolute axial coordinate of the tool and the instantaneous absolute phase angle of the spindle rotation, and calculates the transient equivalent radial stiffness. The transient radial retraction displacement calculation module obtains the equivalent eccentric mass at the end of the workpiece, and then superimposes the high-speed rotating centrifugal force generated by it with the instantaneous cutting force and divides it by the dynamic attenuation stiffness to analyze the transient radial retraction displacement. The destructive energy dissipation ratio calculation module calculates the strain potential energy generated by the transient radial yield displacement, divides it by the total mechanical work done based on the input power integral within one rotation cycle of the main shaft, and generates the destructive energy dissipation ratio. The particle swarm optimization module updates the optimization parameters of the preset particle swarm algorithm using the dissipation ratio, and calculates the optimal machining angular velocity and the optimal feed rate that minimize the ratio. The active interference compensation displacement calculation module substitutes the optimal machining angular velocity and the optimal feed rate into the yield prediction model to calculate the residual yield displacement and inverts it to convert it into active interference compensation displacement. The industrial control module converts the compensated displacement and the optimal machining angular velocity into the piezoelectric control voltage input to the servo piezoelectric tool holder and the target rigid torque command input to the spindle servo motor.

[0006] The beneficial effects of this invention are as follows: by acquiring the mechanical energy input power in real time and reconstructing the spatiotemporal equivalent radial stiffness of the non-axisymmetric body, the dynamic yielding displacement under the action of forced centrifugal force is separated; the dimensionless energy dissipation ratio constructed in this way realizes the adaptive optimization of machining parameters and active servo interference, effectively avoiding the frequency-varying vibration caused by the time-varying distortion of bending capacity when the irregular shaft is rotating at high speed, suppressing the machining ripples on the mating surface, and improving the turning surface quality of thin-walled irregular shafts and the system operation stability. Attached Figure Description

[0007] Figure 1This is a flowchart of an automated control method for turning agricultural drive shafts according to the present invention. Detailed Implementation

[0008] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0009] Example 1: This example applies to turning equipment that includes a cutting tool, force measurement module, spindle encoder, servo piezoelectric tool holder, spindle servo motor, axial position encoder, displacement feedback module, and industrial controller. All calculations and parameter acquisitions are performed using the International System of Units (SI), and all physical quantities adhere to SI standards. The minimum interpolation period of the industrial controller is 100 microseconds. The sampling frequency of all acquisition modules is synchronized with the interpolation period. Hardware trigger latching is used to achieve clock synchronization of multi-source data, with a synchronization error not exceeding 10% of the interpolation period.

[0010] like Figure 1 As shown, an automated control method for turning agricultural drive shafts is applied to machining equipment comprising a cutting tool, a force measuring module, a spindle encoder, a servo piezoelectric tool holder, and a spindle servo motor, including: Simultaneously acquire the cutting force parameters and motion parameters of the cutting interface, and calculate the instantaneous mechanical energy input power of the system; Extract the current absolute axial coordinate of the tool and the instantaneous absolute phase angle of the spindle rotation, and calculate the transient equivalent radial stiffness; The equivalent eccentric mass at the end of the workpiece is obtained, and the high-speed rotating centrifugal force generated by it is superimposed with the instantaneous cutting force and divided by the dynamic attenuation stiffness to analyze the transient radial yield displacement. Calculate the strain potential energy generated by the transient radial yield displacement, divide it by the total mechanical work done based on the input power integral within one rotation cycle of the main shaft, and generate the destructive energy dissipation ratio; The optimization parameters of the preset particle swarm algorithm are updated using the dissipation ratio, and the optimal machining angular velocity and optimal feed rate that minimize the ratio are calculated. Substitute the optimal machining angular velocity and the optimal feed rate into the yield prediction model to calculate the residual yield displacement and invert it to convert it into an active interference compensation displacement. The compensated displacement and the optimal machining angular velocity are converted into the piezoelectric control voltage input to the servo piezoelectric tool holder and the target rigid torque command input to the spindle servo motor.

[0011] Simultaneously acquire the cutting force parameters and motion parameters of the cutting interface, and calculate the instantaneous mechanical energy input power of the system.

[0012] Multi-source raw data is synchronously acquired through a force measurement module and a spindle encoder, completing data preprocessing and clock alignment. The force measurement module uses a piezoelectric three-dimensional cutting force measuring device installed at the front end of the servo piezoelectric tool holder. The sampling frequency is consistent with the interpolation frequency of the industrial controller. The raw acquired signal is first processed by an 8th-order Butterworth low-pass filter with the filter cutoff frequency set to 1 / 10 of the sampling frequency to eliminate high-frequency vibration interference and noise, and simultaneously performs real-time correction of signal zero-point drift. The zero-point correction uses the arithmetic mean of 1000 sets of data acquired under no-load conditions before machining as the zero-point reference. The spindle encoder is an incremental photoelectric encoder installed at the output end of the spindle servo motor. The encoder has no less than 2000 lines and calculates the transient angular velocity of the spindle in real time using the M / T method, while simultaneously calculating the absolute phase angle of the spindle rotation by accumulating the number of pulses. The axial position encoder is installed at the Z-axis feed end of the machine tool to acquire the current absolute axial coordinate of the tool in real time. All acquired data are latched and acquired simultaneously through hardware trigger signals. The trigger signals are uniformly issued by the interpolation clock of the industrial controller to ensure that the timestamps of all data are perfectly aligned. For packet loss or out-of-range data that occurs during the acquisition process, the linear interpolation results of the first 3 sets of valid data are used to complete the data. The number of consecutive data points completed in a single instance shall not exceed 3. If more than 3 consecutive abnormal data points are detected, a processing pause warning will be triggered.

[0013] Define and extract cutting force and motion parameters, clarifying the physical meaning and value specifications of each parameter. The transient radial back-cutting force is denoted as... The unit is Newtons (N), which is the instantaneous force exerted on the tool during cutting the workpiece at time t, opposite to the radial feed direction. It is directly obtained from the pre-processed radial channel data of the force measurement module, and its value range does not exceed the rated range of the force measurement module. Radial feed rate, denoted as... The unit is meters per second (m / s), which is the instantaneous feed rate of the tool along the radial direction of the workpiece at time t. It is calculated from the position loop feedback data of the servo piezoelectric tool holder through first-order differential calculation. The differential time window is a single interpolation cycle, and its value range does not exceed the rated maximum feed rate of the servo piezoelectric tool holder. The main cutting force is denoted as... The unit is Newtons (N), which is the instantaneous force exerted on the tool during cutting the workpiece at time t, acting in the same direction as the spindle rotation tangentially. It is directly obtained from the pre-processed tangential channel data of the force measurement module, and its value range does not exceed the rated range of the force measurement module. The spindle transient angular velocity is denoted as... The unit is radians per second (r / s), which is the instantaneous angular velocity of the workpiece rotating with the spindle at time t. It is calculated from the pulse signal of the spindle encoder using the M / T method. The calculation time window is consistent with the interpolation period, and the value range does not exceed the rated maximum angular velocity of the spindle servo motor. The instantaneous cutting outer diameter is denoted as... The unit is meters, which is the instantaneous outer diameter of the workpiece corresponding to the current cutting position of the tool at time t. It is obtained by linear interpolation matching between the current absolute axial coordinate of the tool and the preset machining contour digital model of the workpiece. The machining contour digital model contains the cross-sectional outer diameter data of the workpiece across its entire axial length. The interpolation step size is consistent with the interpolation period. The cutting force parameters include the transient radial back cutting force and the main cutting force, and the motion parameters include the radial feed rate and the transient angular velocity of the spindle.

[0014] The radial mechanical power is calculated based on the transient radial back-cutting force and radial feed rate. Radial mechanical power is the instantaneous power consumed by the tool to overcome the radial cutting force during radial feed, denoted as... The unit is watts, and the calculation formula is: When the radial feed rate is 0, the radial mechanical power is directly taken as 0 to avoid invalid calculations.

[0015] Calculate the tangential mechanical power based on the main cutting force, the spindle's transient angular velocity, and the instantaneous cutting outer diameter. The tangential mechanical power is the instantaneous power consumed by the spindle in rotating the workpiece to overcome the main cutting force, denoted as […]. The unit is watts, where half of the instantaneous cutting outer diameter is the instantaneous radius of rotation at the cutting position, calculated using the following formula: When the transient angular velocity of the spindle is 0, the tangential mechanical power is directly taken as 0 to avoid invalid calculations.

[0016] The instantaneous mechanical energy input power of the system is calculated based on the radial and tangential mechanical work power. The instantaneous mechanical energy input power is the total instantaneous mechanical power input to the machining system from the external environment during the cutting process, denoted as […]. The unit is watts. All parameters involved in the calculation are synchronously collected data at the same timestamp to ensure the timing consistency of the power calculation. The calculation formula is: In the formula, t is the independent variable of time, in seconds. When the spindle is in the start-stop phase, and the instantaneous angular velocity and radial feed rate of the spindle are both 0, the instantaneous mechanical energy input power of the system is directly taken as 0.

[0017] Extract the current absolute axial coordinate of the tool and the instantaneous absolute phase angle of the spindle rotation, and calculate the transient equivalent radial stiffness.

[0018] Obtain the basic material and structural parameters required for stiffness calculations, and clarify the physical meaning and acquisition method of each parameter. Elastic modulus of steel, The unit is Pascal (GPa), which is the Young's modulus of the steel used in the workpiece. It is obtained from the national standard mechanical property parameters corresponding to the workpiece material grade. The value range for commonly used structural steels is 190 GPa to 210 GPa. For 40Cr alloy structural steel commonly used in agricultural drive shafts, the benchmark value is 206 GPa. The total clamping length is denoted as... The unit is meters, representing the total axial length between the two clamping points of the workpiece. It is measured from the actual installation position of the machine tool chuck clamping end face and the tailstock center support end face, with a measurement accuracy of not less than 0.01 millimeters. The absolute axial coordinate is denoted as... The unit is meters, which is the absolute axial position of the tool tip relative to the workpiece chuck clamping end reference surface at time t. It is obtained from real-time feedback data of the machine tool's Z-axis axial position encoder, and the value ranges from 0 to... ,when Less than 0 or greater than When the tool is determined to be outside the workpiece region, stiffness calculation is paused, and the theoretical stiffness limit value of the solid end is output. The maximum principal moment of inertia of a non-axisymmetric body is denoted as... The value, expressed in meters to the fourth power, is the maximum of the two orthogonal principal moments of inertia of the workpiece's non-axisymmetric cross-section corresponding to the tool's current axial position at time t. It is obtained from the cross-sectional geometry at the corresponding axial position in the workpiece's machining contour digital model, calculated using the formula for calculating the moment of inertia of the cross-section. For workpieces with variable cross-sections, this parameter is a continuous function of the absolute axial coordinates and updates in real time with the tool's axial position. The minimum principal moment of inertia of a non-axisymmetric shape is denoted as... The value, expressed in meters to the fourth power, is the minimum of the two orthogonal principal moments of inertia of the workpiece's non-axisymmetric cross-section corresponding to the current axial position of the tool at time t. It is obtained from the cross-sectional geometry at the corresponding axial position in the workpiece's machining contour digital model, calculated using the formula for calculating the moment of inertia. For workpieces with variable cross-sections, this parameter is a continuous function of the absolute axial coordinates and is updated in real-time with the tool's axial position. For thin-walled irregular cross-sections, corrections for shear deformation and cross-sectional warping effects must be considered. The correction coefficient is obtained from the formula for calculating the mechanical properties of thin-walled cross-sections. The corrected principal moment of inertia is the original calculated value multiplied by the correction coefficient, which ranges from 0.8 to 1.0 and is determined by the ratio of the cross-section's wall thickness to its outer diameter.

[0019] Based on the elastic modulus of steel, the total clamping length, and the absolute axial coordinates, a spatially continuous attenuation term is calculated. This term characterizes the variation of the workpiece's bending stiffness at different axial positions with respect to the tool position. It is derived from a bending stiffness model based on a beam with fixed ends and is denoted as... The unit is Newtons per meter cubed, and the calculation formula is: When the tool is at the midpoint of the axial direction, that is When the spatial continuous attenuation term reaches its minimum value, the workpiece bending stiffness is at its lowest; when the tool approaches the clamping positions at both ends, i.e. Approaching 0 or At this time, the spatial continuous decay term increases rapidly, corresponding to a rapid increase in the bending stiffness of the workpiece. When Less than 0.01 meters or greater When the distance is meters, the spatial continuous decay term takes the fixed calculated value at the two clamping positions to avoid calculation divergence caused by the denominator approaching 0.

[0020] Calculate the instantaneous absolute phase angle of rotation based on the spindle's transient angular velocity. The instantaneous absolute phase angle of rotation is the cumulative angular displacement of the workpiece as it rotates from its initial machining zero position, denoted as [missing information]. The unit is radians, used to characterize the real-time rotational orientation of the workpiece's non-axisymmetric cross-section. The machining starting zero position is determined by the zero-point signal calibration of the spindle encoder, and the calculation formula is: In the formula The time integration variable is a local variable, measured in seconds, and the integration interval is from the start of machining at time zero to the current time t. For spindle speed-changing conditions, the integration process employs a variable-step trapezoidal numerical integration method, with the integration step size consistent with the interpolation cycle to ensure the real-time performance and accuracy of the phase angle calculation. When the spindle is stationary, i.e. At that time, the instantaneous rotation of the absolute phase angle maintains its current value.

[0021] Based on the maximum principal moment of inertia, minimum principal moment of inertia, and instantaneous absolute phase angle of rotation, the time-parameterized rotational term is calculated. The time-parameterized rotational term characterizes the periodic variation of the moment of inertia of the non-axisymmetric section with the workpiece rotation angle; its frequency is twice the frequency of the principal spindle rotation, denoted as […]. The unit is the fourth power of meters, and the calculation formula is: When the workpiece has an axisymmetric circular cross section The time-excitation rotation term is a fixed value and does not change with the rotation angle, which conforms to the mechanical properties of an axisymmetric section.

[0022] The transient equivalent radial stiffness is calculated based on the spatially continuous decay term and the time-parametric rotation term. The transient equivalent radial stiffness is the equivalent radial bending stiffness of the workpiece at the current tool position and current rotation angle, denoted as […]. The unit is Newtons per meter (N / m). It integrates the stiffness attenuation due to axial position and the stiffness pulsation effect due to rotational angle, while also considering the changes in cross-sectional properties of the variable cross-section workpiece. The calculation formula is: When the tool is in the area outside the workpiece, the transient equivalent radial stiffness is directly taken as the theoretical stiffness limit value of the solid end; when the calculation result is less than 0, it is determined that the system is close to the critical state of instability, and the transient equivalent radial stiffness is taken as the preset minimum stiffness threshold. The minimum stiffness threshold is set to 1000 N / m to avoid divergence caused by negative stiffness in subsequent calculations.

[0023] Obtain the equivalent eccentric mass at the end of the workpiece and analyze the transient radial yield displacement.

[0024] Obtain the basic cutting and structural parameters required for yield displacement calculation, and clarify the physical meaning and acquisition method of each parameter. The intrinsic specific cutting energy of the material is denoted as... The unit is joules per cubic meter (J / m³), which is the energy required to remove material from a unit volume of workpiece by cutting. It is obtained through cutting performance tests on the workpiece material, or by referring to recommended values ​​for the corresponding material in metal cutting process manuals. For 40Cr alloy structural steel commonly used in agricultural drive shafts, the benchmark value is 2.5 × 10⁻⁶. 9 The parameter, joules per cubic meter, can be corrected through cutting force calibration tests based on actual cutting conditions. The correction factor ranges from 0.8 to 1.2. The instantaneous depth of cut is denoted as... The unit is meters, which is the radial cutting depth of the tool along the workpiece at time t. It is obtained by matching the preset cutting depth in the machining program with the current axial position of the tool, and its value does not exceed the maximum cutting depth of the machine tool. Eccentricity is denoted as... The unit is meters. It represents the perpendicular distance between the center of mass of the equivalent eccentric mass at the workpiece end and the axis of rotation of the workpiece. It is obtained through dynamic balancing tests after the workpiece is clamped. The rotational speed during the dynamic balancing test is not lower than the highest rotational speed during machining, and the test accuracy is not lower than G1 level balancing accuracy. The equivalent eccentric mass at the end is denoted as... The unit is kilograms. It represents the equivalent eccentric mass concentrated at the ends of the workpiece, representing the asymmetric structure at both clamping points. The equivalent reference position is the clamping end faces of the workpiece. It is calculated using the mass properties of the workpiece's 3D model. The distributed eccentric mass of the asymmetric structure at both ends is combined into a single-end concentrated equivalent eccentric mass using the rigid body mechanics equivalence principle. Alternatively, it can be obtained from dynamic balance test data. The calculation formula is... ,in The remaining imbalance obtained from the dynamic balancing test is expressed in kilogram-meters.

[0025] The instantaneous feed per revolution is calculated based on the radial feed rate and the instantaneous angular velocity of the spindle. The instantaneous feed per revolution is the radial feed displacement of the tool per revolution of the spindle, denoted as... The unit is meters per revolution, and the calculation formula is: In the formula The value is pi, taken as 3.141592653589793. When the instantaneous angular velocity of the spindle is less than 1 radian per second, the instantaneous feed per revolution is set to a preset minimum feed threshold to avoid calculation divergence caused by the denominator approaching 0. The minimum feed threshold is set to 1 × 10⁻⁶. -6 Rice per revolution.

[0026] Based on the intrinsic cutting energy of the material, the instantaneous depth of cut, and the instantaneous feed per revolution, the basic cutting resistance is calculated. The basic cutting resistance is the instantaneous cutting reaction force of the workpiece material on the tool during the cutting process, denoted as... The unit is cow, and the calculation formula is: When the tool does not cut into the workpiece, the basic cutting resistance is taken as 0.

[0027] Based on the equivalent eccentric mass at the end, eccentricity, transient angular velocity of the spindle, and instantaneous absolute phase angle of rotation, the high-speed rotating centrifugal force is calculated. This high-speed rotating centrifugal force is the radial alternating centrifugal force generated by the eccentric mass rotating at high speed with the spindle, denoted as... The unit is Newtons (N). The direction of the force is collinear with the radial feed direction of the tool. The coordinate system is consistent with the cutting force coordinate system. The calculation formula is: When the spindle is stationary, that is At that time, the centrifugal force during high-speed rotation is directly taken as 0.

[0028] Based on the transient equivalent radial stiffness, the equivalent eccentric mass at the end, and the transient angular velocity of the spindle, the dynamic attenuation stiffness is calculated. The dynamic attenuation stiffness is the workpiece's equivalent radial stiffness after considering the stiffness softening effect caused by rotational centrifugal force, denoted as... The unit is Newtons per meter (N / m), and the calculation formula is: To avoid negative stiffness issues caused by system instability, when the calculated result is less than the preset minimum stiffness threshold of 1000 N / m, the dynamic attenuation stiffness is directly taken as the minimum stiffness threshold to ensure the stability of subsequent calculations.

[0029] Based on fundamental cutting resistance, high-speed rotating centrifugal force, and dynamically attenuated stiffness, the transient radial yield displacement is analyzed, and closed-loop coupled iterative correction is performed. The transient radial yield displacement is the instantaneous radial elastic deformation displacement of the workpiece relative to the tool under the combined action of cutting force and centrifugal force, denoted as... The unit is meters. This displacement directly causes the tool deflection error during the machining process. The calculation formula is: Since the yield displacement directly causes a change in the actual depth of cut, and the change in the actual depth of cut in turn alters the cutting force and the yield displacement, a convergent stable value needs to be obtained through closed-loop coupled iterative calculation. The iteration rules are as follows: Initial iteration step k=0, take the initial back cut depth. Calculate the initial retraction displacement based on the preset values ​​of the machining program. ; In the (k+1)th iteration, update the actual back-cutting depth: ; Based on the updated actual depth of cut, calculate the yield displacement at the (k+1)th iteration step. ; The iterative convergence condition is The iteration terminates when either condition is met (e.g., the depth of cut is measured in meters) or the maximum number of iterations (20) is reached, and the converged transient radial retraction displacement is output. If the actual depth of cut is less than 0 during the iteration process, it is determined that the tool has completely disengaged from the workpiece, and the transient radial retraction displacement is directly set to 0.

[0030] Calculate the strain potential energy generated by the transient radial yield displacement and generate the destructive energy dissipation ratio.

[0031] Based on transient equivalent radial stiffness and transient radial yield displacement, strain potential energy is calculated. Strain potential energy is the elastic potential energy stored in the workpiece when it undergoes radial elastic yield deformation, denoted as . The unit is joules. This portion of energy represents meaningless energy dissipation during processing, which can cause workpiece vibration and processing errors. The calculation formula is: When the transient radial yield displacement is 0, the strain potential energy is directly taken as 0.

[0032] Based on the instantaneous mechanical energy input power of the system, the total mechanical work done by the spindle in one rotation cycle is calculated. Total mechanical work is the total mechanical energy input to the machining system from the external environment during one complete rotation cycle of the spindle, denoted as […]. The unit is joules, and the calculation formula is: In the formula For time integration, the local variable is in seconds. Let t be the rotation period of the spindle at time t, in seconds, and the calculation formula is: .

[0033] For spindle speed change conditions, when the spindle angular velocity changes by more than 5% within the integral interval, a variable-period integration method is used. The integral interval is adjusted to the actual time interval corresponding to the previous complete spindle rotation cycle, ensuring a perfect match between the integral interval and the actual spindle rotation cycle. For spindle start-stop phases, when the spindle rotation cycle is greater than 1 second, the total mechanical work is taken as the integral result of the input power within the most recent second. When the spindle transient angular velocity is 0, the total mechanical work is taken as the preset minimum energy threshold of 1×10⁻⁶. -3 To avoid computational divergence caused by the denominator approaching zero, the integration operation employs the trapezoidal numerical integration method, with the integration step size consistent with the interpolation period.

[0034] Based on strain potential energy and total mechanical work, a destructive energy dissipation ratio is generated. The destructive energy dissipation ratio is a dimensionless parameter, denoted as . The value is greater than 0, representing the proportion of ineffective energy dissipation caused by workpiece yielding deformation during processing to the total input energy. The larger this proportion, the more severe the vibration and tool deflection error of the processing system, and the worse the processing stability. The calculation formula is: When the strain potential energy is 0, the destructive energy dissipation ratio is directly set to 0; when the calculation result is greater than 1, the system is judged to be in a strong flutter state, and the destructive energy dissipation ratio is set to the maximum value of 1 to ensure the stability of the parameter value range.

[0035] The optimal machining angular velocity and optimal feed rate are calculated by updating the particle swarm optimization parameters using the dissipation ratio.

[0036] The parameter space definition, dimensionless rules, and initialization settings of the particle swarm optimization (PSO) algorithm are clearly defined to ensure the dimensional homogeneity of vector operations. The optimization variables of the PSO algorithm are two independent physical parameters: the spindle machining angular velocity and the radial feed rate. Each parameter undergoes independent dimensionless processing to ensure dimensional consistency in the iterative calculations for each parameter. The baseline value for the spindle machining angular velocity is the rated maximum angular velocity of the machine tool spindle, denoted as [missing value]. The unit is radians per second, and the dimensionless principal axis angular velocity parameter is denoted as . The calculation formula is: The value ranges from [0.05, 1.0], corresponding to an actual angular velocity range of 5% to 100% of the rated maximum angular velocity. The reference value for the radial feed rate is the rated maximum feed rate of the servo piezoelectric tool holder, denoted as... The unit is meters per second, and the dimensionless feed rate parameter is denoted as... The calculation formula is: The value range is [0.01, 1.0], corresponding to an actual feed rate range of 1% to 100% of the rated maximum feed rate. The particle swarm's position vector is a two-dimensional vector, with each element corresponding to the dimensionless principal axis angular velocity parameter and the dimensionless feed rate parameter, respectively. Both elements are dimensionless, ensuring the homogeneity of dimensions in vector addition and subtraction operations. The particle swarm's step size vector is a two-dimensional vector with the same dimension as the position vector, with each element corresponding to the iteration step size of the two dimensionless parameters, and the dimensions perfectly match the position vector. The algorithm initialization settings are as follows: the population size is set to 30 particles, the maximum number of iterations is set to 100, the initial position vector is randomly generated using a uniform distribution within the parameter value range, the initial step size vector is set to 0.1 times the width of the parameter value range, the individual optimal position vector is initialized to the initial position vector of the particle, and the global optimal position vector is initialized to the initial position vector of the particle with the best fitness value in the population. The first natural random number... With the second natural random number The values ​​are uniformly distributed random numbers in the range [0,1], and are regenerated in each iteration. The two are independent of each other.

[0037] Obtain the basic parameters required for the optimization calculation, and clarify the physical meaning and acquisition method of each parameter. The theoretical stiffness limit of the solid end is denoted as... The unit is Newtons per meter (N / m), which is the theoretical maximum radial stiffness of the solid clamping section at the end of the workpiece. It is calculated based on the bending stiffness model of a beam with fixed supports at both ends. The calculation formula is as follows: ,in Let be the moment of inertia of the solid circular section at the end of the workpiece, expressed in meters to the fourth power, calculated from the diameter of the solid section at the end. The optimal position vector for the individual is denoted as . Let be a two-dimensional dimensionless vector, representing the position vector of the i-th particle in the iteration history when it achieves the minimum destructive energy dissipation ratio. Each particle independently maintains its own individual optimal position vector, which is updated after each iteration. The global optimal position vector is denoted as . is a two-dimensional dimensionless vector, representing the position vector of the entire particle swarm that achieves the minimum destructive energy dissipation ratio throughout its iteration history. The entire particle swarm maintains a unique globally optimal position vector, which is updated after each iteration. The current parameter position vector is denoted as . Let be a two-dimensional dimensionless vector, representing the position vector of the i-th particle in the k-th iteration, with the superscript k indicating the iteration number. The original optimization step size vector is denoted as . , is a two-dimensional dimensionless vector, and is the iteration step size vector corresponding to the i-th particle in the k-th iteration.

[0038] Based on the transient equivalent radial stiffness and the theoretical stiffness limit of the solid end, the spatial degradation ratio is calculated. The spatial degradation ratio is a dimensionless parameter, denoted as . The value ranges from [0,1], representing the degree of degradation of the workpiece stiffness at the current tool position relative to the theoretical limit stiffness. The larger the value, the more severe the workpiece stiffness degradation. The calculation formula is: When the calculation result is less than 0, the spatial degradation ratio is directly set to 0; when the calculation result is greater than 1, the spatial degradation ratio is directly set to 1.

[0039] Based on the equivalent eccentric mass at the end, the transient angular velocity of the spindle, and the transient equivalent radial stiffness, the centrifugal softening breakdown ratio is calculated. The centrifugal softening breakdown ratio is a dimensionless parameter, denoted as . The value is greater than 0, representing the degree of workpiece stiffness softening caused by centrifugal force. The larger the value, the more significant the centrifugal softening effect, and the more prone the machining system is to instability. The calculation formula is: When the calculation result is greater than 1, the centrifugal softening breakdown ratio is directly set to 1 to ensure the stability of the parameter value range.

[0040] Define a fitness function to clarify the optimization objective and constraints. The fitness function aims to minimize the proportion of destructive energy dissipation, while incorporating a penalty term based on machine tool hardware constraints to ensure that the optimization result remains within the allowable range of the machine tool hardware. The fitness value of the i-th particle in the k-th iteration is denoted as... The calculation formula is: In the formula The destructive energy dissipation ratio for the i-th particle under the corresponding parameters. The penalty term, which is dimensionless, is used to constrain the particle's position vector. It is set to 1000 when the particle's position vector exceeds the parameter range, and 0 otherwise. The optimization objective is to minimize the fitness value; a smaller fitness value indicates better processing parameters for the particle.

[0041] The optimization step size vector is updated based on the proportion of destructive energy dissipation, spatial degradation, and centrifugal softening breakdown. The updated optimization step size vector is the particle step size of the (k+1)th iteration, denoted as... , is a two-dimensional dimensionless vector, and its calculation formula is: The physical meanings of the three terms in the formula are as follows: The first term is the adaptive inertia maintenance term, which is the part of the particle velocity update that inherits the iteration step size of the previous generation. It is used to maintain the iteration inertia of the particles. Its weight is adaptively adjusted according to the proportion of destructive energy dissipation. When the dissipation proportion is large, the inertia weight increases, which accelerates the global search capability of the algorithm. The second term is the individual cognition update term, which is the velocity component that the particle learns from its own historical best solution. It is used to guide the particle to converge to its own historical best position. Its weight is adaptively adjusted according to the degree of workpiece stiffness degradation. When the stiffness degradation is severe, the individual learning weight increases, which improves the algorithm's adaptability to local working conditions. The third term is the group risk avoidance update term, which is the velocity component that the particle learns from the group's global best solution. It is used to guide the particle to converge to the group's best position. Its weight is adaptively adjusted according to the intensity of the centrifugal softening effect. When the centrifugal softening effect is significant, the group learning weight increases, which improves the algorithm's ability to avoid the risk of system instability. After the step size vector is updated, the step size needs to be limited. The maximum step size of each element is limited to 0.2 times the width of the parameter value range to avoid the algorithm diverging due to an excessively large step size.

[0042] Update the particle position vector to complete the iterative update of individual optimality and global optimality. The formula for updating the position vector of the i-th particle in the (k+1)-th iteration is: After the position vector is updated, boundary constraints are applied to each element. The dimensionless principal axis angular velocity parameter is limited to the range of [0.05, 1.0], and the dimensionless feed velocity parameter is limited to the range of [0.01, 1.0]. Elements outside these ranges are truncated to boundary values. Based on the updated position vector, the actual processing parameters and fitness value corresponding to each particle are calculated, completing the updates for individual and global optima: 1. For the i-th particle, if the fitness value of the current iteration is less than the fitness value corresponding to the individual optimal position vector, then the individual optimal position vector is updated to the position vector of the current iteration; 2. For the entire particle swarm, if the minimum fitness value in the current iteration is less than the fitness value corresponding to the global optimal position vector, then the global optimal position vector is updated to the position vector of the particle with the minimum fitness value in the current iteration.

[0043] The iteration convergence check outputs the optimal machining angular velocity and optimal feed rate. The iteration terminates when either of the following conditions is met: 1. The number of iterations reaches the preset maximum of 100 iterations; 2. The fitness value corresponding to the global optimal position vector does not decrease for 20 consecutive iterations, with a decrease threshold set to 1×10⁻⁶. -63. The proportion of destructive energy dissipation corresponding to the global optimal position vector is less than the preset qualified threshold of 0.01, that is, the proportion of ineffective energy dissipation is less than 1%. When any termination condition is met, the iteration terminates, and the dimensionless parameters corresponding to the global optimal position vector are denormalized to obtain the optimal machining angular velocity and the optimal feed rate. The optimal machining angular velocity is denoted as... The unit is radians per second, and the calculation formula is: In the formula This represents the dimensionless principal axis angular velocity parameter corresponding to the globally optimal position vector. The optimal feed rate is denoted as... The unit is meters per second, and the calculation formula is: In the formula This represents the dimensionless feed rate parameter corresponding to the globally optimal position vector. The algorithm iteration process is executed in a background thread of the industrial controller, with the execution cycle synchronized with the machining interpolation cycle. After each iteration, the optimal machining parameters are updated to ensure real-time optimization of the machining parameters.

[0044] Substituting the optimal machining angular velocity and optimal feed rate into the yield prediction model, it is transformed into active interference compensation displacement.

[0045] Based on the optimal machining angular velocity and optimal feed rate, the optimal feed per revolution is calculated. The optimal feed per revolution is the radial feed displacement of the tool per revolution of the spindle under optimal machining parameters, denoted as . The unit is meters per revolution, and the calculation formula is: The optimal cutting resistance is calculated based on the intrinsic cutting energy of the material, the instantaneous depth of cut, and the optimal feed per revolution. The optimal cutting resistance is the cutting reaction force of the workpiece material on the tool under optimal machining parameters, denoted as […]. The unit is cow, and the calculation formula is: The optimal forced centrifugal force is calculated based on the equivalent eccentric mass at the end, the eccentricity, the optimal machining angular velocity, and the instantaneous absolute phase angle of rotation. The optimal forced centrifugal force is the radial alternating centrifugal force generated by the rotation of the eccentric mass at the end under optimal machining parameters, denoted as […]. The unit is cow, and the calculation formula is: The optimal centrifugal decay stiffness is calculated based on the transient equivalent radial stiffness, the equivalent end eccentric mass, and the optimal machining angular velocity. The optimal centrifugal decay stiffness is the workpiece's equivalent radial stiffness considering the centrifugal softening effect under optimal machining parameters, denoted as […]. The unit is Newtons per meter (N / m), and the calculation formula is: When the calculated result is less than the preset minimum stiffness threshold of 1000 N / m, the optimal centrifugal decay stiffness is directly taken as the minimum stiffness threshold.

[0046] Based on the optimal cutting resistance, optimal forced centrifugal force, and optimal centrifugal decay stiffness, the residual yield displacement is calculated using a yield prediction model. This model predicts the radial yield displacement that the workpiece will still generate under optimal machining parameters; this residual yield displacement is denoted as... The unit is meters, and the calculation formula is: The calculation of residual yield displacement also adopts the closed-loop coupled iterative correction method, and the iteration rule is consistent with the iteration rule of transient radial yield displacement to ensure the accuracy of the calculation results.

[0047] Based on the residual retraction displacement, an active interference compensation displacement is generated. This active interference compensation displacement is a radial pre-compensation displacement of the tool used to counteract the workpiece retraction displacement, denoted as... The unit is meters. This displacement is achieved by controlling the tool to advance towards the workpiece's rotation center axis in advance, thus offsetting the machining error caused by the workpiece's retraction. The calculation formula is: To avoid overcutting caused by excessive compensation, the maximum absolute value of the active interference compensation displacement is limited to a preset maximum compensation threshold, which is set to 0.1 mm. Compensation values ​​exceeding this range are directly truncated to the threshold.

[0048] The compensated displacement and optimal machining angular velocity are converted into the piezoelectric control voltage of the servo piezoelectric tool holder and the target rigid torque command of the spindle servo motor.

[0049] Obtain the hardware parameters required for control command calculations, and clarify the physical meaning and acquisition method of each parameter. The piezoelectric conversion constant is denoted as... The unit is meter per volt (mV), which is the piezoelectric strain constant of the piezoelectric ceramic. It characterizes the ratio of the strain generated by the piezoelectric ceramic under the action of an electric field to the electric field strength. It is obtained from the model parameters of the piezoelectric ceramic used in the servo piezoelectric tool holder. The value range of commonly used piezoelectric ceramics is 200 × 10⁻⁶ mV. -12 600×10 meters per volt -12 Meters per volt. The number of piezoelectric layers, denoted as , a positive integer, represents the total number of piezoelectric ceramic laminations within the servo piezoelectric tool holder, obtained from the hardware structural parameters of the servo piezoelectric tool holder, with a commonly used value range of 50 to 200 layers. The no-load moment of inertia is denoted as... The unit is kilogram-square meter (kg / m²), which is the moment of inertia of a machine tool spindle system in a workpiece-free state. It is obtained from the machine tool spindle's factory parameters or through moment of inertia testing. Clock interpolation in microseconds is denoted as . The unit is microseconds, which is the servo interpolation cycle of the industrial controller. It is set by the interpolation clock frequency of the machine tool CNC system, and the commonly used value range is 100 microseconds to 1000 microseconds.

[0050] Based on the active interference compensation displacement, the piezoelectric conversion constant, and the number of piezoelectric layers, the piezoelectric control voltage is calculated. The piezoelectric control voltage is the control voltage signal that drives the servo piezoelectric tool holder to generate the corresponding compensation displacement, denoted as […]. The unit is volt, and the calculation formula is: To protect the piezoelectric ceramic hardware, the piezoelectric control voltage needs to be limited. The voltage range is restricted to the rated operating voltage range of the servo piezoelectric tool holder. Voltage values ​​exceeding the range are directly truncated as boundary values. At the same time, a linear correction for the hysteresis characteristics of the piezoelectric ceramic is added. The correction coefficient is obtained from the factory calibration parameters of the servo piezoelectric tool holder.

[0051] The combined moment of inertia is calculated based on the unloaded moment of inertia, the equivalent eccentric mass at the end, and the eccentricity. The combined moment of inertia is the total moment of inertia of the spindle system after the workpiece is clamped, denoted as... The unit is kilograms per square meter, and the calculation formula is: The Newtonian torque is calculated based on the optimal machining angular velocity, the spindle transient angular velocity, the combined moment of inertia, and the clock interpolation in microseconds. The angular velocity increment is the difference between the optimal machining angular velocity and the current spindle transient angular velocity, denoted as . The unit is radians per second, and the calculation formula is: The Newtonian moment of inertia is the moment of inertia required for the spindle system to complete the change of angular velocity within the interpolation cycle, denoted as . The unit is Newton-meter (Nm), and the calculation formula is: In the formula It converts the interpolation period unit from microseconds to seconds to ensure dimensional consistency.

[0052] The working resistance torque is calculated based on the system's instantaneous mechanical energy input power and the spindle's transient angular velocity. The working resistance torque is the spindle output torque required to overcome the cutting force, denoted as... The unit is Newton-meter (Nm), and the calculation formula is: When the transient angular velocity of the main shaft is less than 1 radian per second, the torque of the resistance force is directly taken as 0 to avoid calculation divergence caused by the denominator approaching 0.

[0053] Based on Newtonian inertial torque and work resistance torque, the target rigid torque command for the spindle servo motor is calculated. The target rigid torque command is the total torque control signal sent to the spindle servo motor, denoted as... The unit is Newton-meter (Nm), which combines the inertial torque required for spindle speed change and the resistance torque required for cutting. The calculation formula is: To protect the spindle servo motor hardware, the target rigid torque command needs to be limited. The torque range is restricted to the rated peak torque range of the spindle servo motor. Torque values ​​exceeding the range are directly truncated to boundary values. At the same time, a torque change rate limit is added to avoid spindle speed fluctuations caused by sudden torque changes.

[0054] The execution timing rules for the entire processing flow are as follows: In the pre-processing preparation stage, workpiece clamping, dynamic balance testing, import of the digital model of the machining contour, hardware parameter calibration, zero-point calibration of the acquisition module, and initialization settings of the particle swarm algorithm are completed. After the processing starts, in each interpolation cycle, the synchronous acquisition and preprocessing of multi-source data are completed first, and the instantaneous mechanical energy input power of the system is calculated. Simultaneous calculation of transient equivalent radial stiffness, transient radial yield displacement, and destructive energy dissipation ratio; The background thread synchronously executes the iterative optimization of the particle swarm optimization algorithm, updating the optimal machining angular velocity and optimal feed rate in each interpolation cycle; The active interference compensation displacement is calculated based on the optimal machining parameters, converted into piezoelectric control voltage and spindle torque commands, and synchronously sent to the servo piezoelectric tool holder and spindle servo motor for execution. After execution, the actual output displacement of the tool post is acquired through the displacement feedback module, and the actual spindle angular velocity is acquired through the spindle encoder to complete the closed-loop feedback correction and adjust the control command for the next interpolation cycle.

[0055] The exception handling rules are as follows: When the collected data shows anomalies for more than 10 interpolation cycles consecutively, the processing is paused and an early warning signal is output. When the destructive energy dissipation ratio exceeds 0.5 for more than 20 interpolation cycles, it is determined that the system has strong vibration, and the spindle speed and feed rate are automatically reduced until the dissipation ratio falls back to the qualified range. When the dynamic attenuation stiffness approaches the minimum stiffness threshold, the maximum spindle speed is automatically limited to prevent system instability. When the piezoelectric control voltage or spindle torque command remains in a limited state for more than 5 interpolation cycles, the machining parameters are adjusted to reduce the cutting load.

[0056] Example 2: An automated control method for turning agricultural drive shafts, applied in any of the automated control systems described in this example, comprising: The data acquisition module synchronously acquires the cutting force parameters and motion parameters of the cutting interface and calculates the instantaneous mechanical energy input power of the system. The transient equivalent radial stiffness calculation module extracts the current absolute axial coordinate of the tool and the instantaneous absolute phase angle of the spindle rotation, and calculates the transient equivalent radial stiffness. The transient radial retraction displacement calculation module obtains the equivalent eccentric mass at the end of the workpiece, and then superimposes the high-speed rotating centrifugal force generated by it with the instantaneous cutting force and divides it by the dynamic attenuation stiffness to analyze the transient radial retraction displacement. The destructive energy dissipation ratio calculation module calculates the strain potential energy generated by the transient radial yield displacement, divides it by the total mechanical work done based on the input power integral within one rotation cycle of the main shaft, and generates the destructive energy dissipation ratio. The particle swarm optimization module updates the optimization parameters of the preset particle swarm algorithm using the dissipation ratio, and calculates the optimal machining angular velocity and the optimal feed rate that minimize the ratio. The active interference compensation displacement calculation module substitutes the optimal machining angular velocity and the optimal feed rate into the yield prediction model to calculate the residual yield displacement and inverts it to convert it into active interference compensation displacement. The industrial control module converts the compensated displacement and the optimal machining angular velocity into the piezoelectric control voltage input to the servo piezoelectric tool holder and the target rigid torque command input to the spindle servo motor.

[0057] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. An automated control method for turning agricultural drive shafts, applied to machining equipment comprising a cutting tool, a force measuring module, a spindle encoder, a servo piezoelectric tool holder, and a spindle servo motor, characterized in that, include: Simultaneously acquire the cutting force parameters and motion parameters of the cutting interface, and calculate the instantaneous mechanical energy input power of the system; Extract the current absolute axial coordinate of the tool and the instantaneous absolute phase angle of the spindle rotation, and calculate the transient equivalent radial stiffness; The equivalent eccentric mass at the end of the workpiece is obtained, and the high-speed rotating centrifugal force generated by it is superimposed with the instantaneous cutting force and divided by the dynamic attenuation stiffness to analyze the transient radial yield displacement. Calculate the strain potential energy generated by the transient radial yield displacement, divide it by the total mechanical work done based on the input power integral within one rotation cycle of the main shaft, and generate the destructive energy dissipation ratio; The optimization parameters of the preset particle swarm algorithm are updated using the dissipation ratio, and the optimal machining angular velocity and optimal feed rate that minimize the ratio are calculated. Substitute the optimal machining angular velocity and the optimal feed rate into the yield prediction model to calculate the residual yield displacement and invert it to convert it into an active interference compensation displacement. The compensated displacement and the optimal machining angular velocity are converted into the piezoelectric control voltage input to the servo piezoelectric tool holder and the target rigid torque command input to the spindle servo motor.

2. The automated control method for turning agricultural drive shafts according to claim 1, characterized in that, Simultaneously acquire cutting force and motion parameters at the cutting interface, and calculate the instantaneous mechanical energy input power of the system, including: The transient radial back-cutting force, radial feed rate, main cutting force, spindle transient angular velocity, and instantaneous cutting outer diameter are synchronously acquired by the force measuring module and the spindle encoder. The cutting force parameters include the transient radial back-cutting force and the main cutting force, and the motion parameters include the radial feed rate and the spindle transient angular velocity. Multiplying the transient radial back-cutting force by the radial feed rate yields the radial mechanical power. The tangential mechanical power is obtained by continuously multiplying the main cutting force, the instantaneous angular velocity of the spindle, and half of the instantaneous cutting outer diameter. The instantaneous mechanical energy input power of the system is obtained by adding the radial mechanical power and the tangential mechanical power.

3. The automated control method for turning agricultural drive shafts according to claim 2, characterized in that, Extract the current absolute axial coordinate of the tool and the instantaneous absolute phase angle of the spindle rotation, and calculate the transient equivalent radial stiffness, including: Obtain the elastic modulus of steel, the total clamping length, and the maximum and minimum principal moments of inertia of the non-axisymmetric shape; Multiply the elastic modulus of the steel by three times and the total clamping length, then divide by the product of the square of the absolute axial coordinate and the square of the difference between the total clamping length and the absolute axial coordinate, to obtain the spatial continuous attenuation term. The instantaneous absolute phase angle of rotation is obtained by integrating the transient angular velocity of the main shaft from time zero to the current time. Divide the sum of the maximum principal moment of inertia and the minimum principal moment of inertia by two, add the difference between the maximum principal moment of inertia and the minimum principal moment of inertia divided by two, and multiply by twice the cosine value of the instantaneous absolute phase angle of rotation to obtain the time-parameter rotation term. Multiplying the spatial continuous decay term by the time-parameterized rotation term yields the transient equivalent radial stiffness.

4. The automated control method for turning agricultural drive shafts according to claim 3, characterized in that, Obtain the equivalent eccentric mass at the end of the workpiece, and then superimpose the high-speed rotating centrifugal force generated by it with the instantaneous cutting force, dividing by the dynamic attenuation stiffness to analyze the transient radial yield displacement, including: Obtain the material's intrinsic cutting energy, instantaneous depth of cut, and eccentricity; Multiplying twice the value of pi by the radial feed rate and then dividing by the instantaneous angular velocity of the spindle, we obtain the instantaneous feed per revolution. Multiply the intrinsic cutting energy of the material, the instantaneous depth of cut, and the instantaneous feed per revolution to obtain the basic cutting resistance as the instantaneous cutting force; The high-speed rotating centrifugal force is obtained by continuously multiplying the equivalent eccentric mass at the end, the eccentricity, the square of the transient angular velocity of the main shaft, and the cosine of the instantaneous absolute phase angle of rotation. The dynamic attenuation stiffness is obtained by subtracting the product of the end equivalent eccentric mass and the square of the transient angular velocity of the main shaft from the transient equivalent radial stiffness. After adding the basic cutting resistance and the high-speed rotating centrifugal force, and dividing by the dynamic attenuation stiffness, the transient radial yield displacement is obtained.

5. The automated control method for turning agricultural drive shafts according to claim 4, characterized in that, Calculate the strain potential energy generated by the transient radial yield displacement, divide it by the total mechanical work done based on the input power integral within one rotation cycle of the main shaft, and generate the destructive energy dissipation ratio, including: The strain potential energy is calculated by continuously multiplying half of the transient equivalent radial stiffness by the square of the transient radial relief displacement. Using the current moment as the upper limit of integration and the quotient of the current moment minus twice the pi divided by the transient angular velocity of the main shaft as the lower limit of integration, the instantaneous mechanical energy input power of the system is integrated over time to obtain the total mechanical work done. The destructive energy dissipation ratio is generated by dividing the strain potential energy by the total mechanical work done.

6. The automated control method for turning agricultural drive shafts according to claim 5, characterized in that, The optimization parameters of the preset particle swarm optimization algorithm are updated using the dissipation ratio, and the optimal machining angular velocity and optimal feed rate that minimize this ratio are calculated, including: Obtain the theoretical stiffness limit of the solid end, the individual optimal position vector, the global optimal position vector, the current parameter position vector, the original optimization step size vector, and the first and second natural random numbers; Divide the transient equivalent radial stiffness by the theoretical stiffness limit of the solid end, and subtract the ratio from the value to obtain the spatial degradation ratio. Multiply the equivalent eccentric mass at the end by the square of the transient angular velocity of the main shaft, and then divide by the transient equivalent radial stiffness to obtain the centrifugal softening breakdown ratio. The adaptive inertia preservation term is obtained by multiplying the destructive energy dissipation ratio with the original optimization step size vector, which is the optimization parameter. The individual cognitive update term is obtained by continuously multiplying the spatial degradation ratio, the first natural random number, and the difference between the individual's optimal position vector and the current parameter position vector; The group risk avoidance update term is obtained by continuously multiplying the centrifugal softening and breakdown ratio, the second natural random number, and the difference between the global optimal position vector and the current parameter position vector; The adaptive inertia retention term, the individual cognition update term, and the group risk avoidance update term are added to obtain the updated optimization step length vector. The vector is iterated in the parameter space to output the extreme angular velocity and extreme feed rate that minimize the proportion of destructive energy dissipation. These are used as the optimal processing angular velocity and the optimal feed rate, respectively.

7. The automated control method for turning agricultural drive shafts according to claim 6, characterized in that, Substituting the optimal machining angular velocity and the optimal feed rate into the retraction prediction model to calculate the residual retraction displacement, and then inverting it, converts it into an active interference compensation displacement, including: The optimal feed per revolution is obtained by dividing the product of twice the value of pi and the optimal feed rate by the optimal machining angular velocity. The optimal cutting resistance is obtained by multiplying the intrinsic cutting energy of the material, the instantaneous depth of cut, and the optimal feed per revolution. The optimal forced centrifugal force is obtained by continuously multiplying the equivalent eccentric mass at the end, the eccentricity, the square of the optimal machining angular velocity, and the cosine of the instantaneous absolute phase angle of rotation. The optimal centrifugal decay stiffness is obtained by subtracting the product of the end equivalent eccentric mass and the square of the optimal machining angular velocity from the transient equivalent radial stiffness. The sum of the optimal cutting resistance and the optimal forced centrifugal force is divided by the optimal centrifugal decay stiffness as the yield prediction model, and the residual yield displacement is calculated. Taking the opposite of the residual yield displacement, the active interference compensation displacement that presses against the rotation center axis is generated.

8. The automated control method for turning agricultural drive shafts according to claim 7, characterized in that, Converting the compensated displacement and the optimal machining angular velocity into the piezoelectric control voltage input to the servo piezoelectric tool holder and the target rigid torque command input to the spindle servo motor includes: Obtain the piezoelectric conversion constant, the number of piezoelectric layers, the no-load rotational inertia, and the clock interpolation microseconds; The piezoelectric control voltage input to the servo piezoelectric tool holder is obtained by multiplying the piezoelectric conversion constant and the number of piezoelectric layers by removing the active interference compensation bit; Multiply the equivalent eccentric mass at the end by the square of the eccentricity, and add the unloaded moment of inertia to obtain the combined moment of inertia. The angular velocity increment is obtained by subtracting the spindle transient angular velocity from the optimal machining angular velocity. The Newtonian moment of inertia is obtained by multiplying the overall moment of inertia by the angular velocity increment and dividing by the clock interpolation microseconds. The instantaneous mechanical energy input power of the system is divided by the transient angular velocity of the main shaft to obtain the work resistance torque; The Newtonian inertial torque is added to the work resistance torque to obtain the target rigid torque command input to the spindle servo motor.

9. An automated control method for turning agricultural drive shafts, applied in the automated control system for turning agricultural drive shafts as described in any one of claims 1-8, characterized in that, include: The data acquisition module synchronously acquires the cutting force parameters and motion parameters of the cutting interface and calculates the instantaneous mechanical energy input power of the system. The transient equivalent radial stiffness calculation module extracts the current absolute axial coordinate of the tool and the instantaneous absolute phase angle of the spindle rotation, and calculates the transient equivalent radial stiffness. The transient radial retraction displacement calculation module obtains the equivalent eccentric mass at the end of the workpiece, and then superimposes the high-speed rotating centrifugal force generated by it with the instantaneous cutting force and divides it by the dynamic attenuation stiffness to analyze the transient radial retraction displacement. The destructive energy dissipation ratio calculation module calculates the strain potential energy generated by the transient radial yield displacement, divides it by the total mechanical work done based on the input power integral within one rotation cycle of the main shaft, and generates the destructive energy dissipation ratio. The particle swarm optimization module updates the optimization parameters of the preset particle swarm algorithm using the dissipation ratio, and calculates the optimal machining angular velocity and the optimal feed rate that minimize the ratio. The active interference compensation displacement calculation module substitutes the optimal machining angular velocity and the optimal feed rate into the yield prediction model to calculate the residual yield displacement and inverts it to convert it into active interference compensation displacement. The industrial control module converts the compensated displacement and the optimal machining angular velocity into the piezoelectric control voltage input to the servo piezoelectric tool holder and the target rigid torque command input to the spindle servo motor.

Images