Torque adjustment control method of magnetic stirrer based on sample characteristics

Abstract

The invention relates to the technical field of motor control, and discloses a torque adjustment control method of a magnetic stirrer based on sample characteristics, which comprises the following steps: acquiring an armature three-phase current and a rotor position signal, and converting the armature three-phase current and the rotor position signal into a quadrature-axis actually measured current and a direct-axis actually measured current by utilizing coordinate transformation logic; extracting a magnetic declination estimated value reflecting the phase lag degree of the stirrer based on an electromagnetic coupling model; calculating a magnetic declination change rate according to the magnetic declination estimated value of the continuous sampling period; a transient compensation current is generated by adopting a nonlinear compensation operator and combining a real-time synchronous torque coefficient; according to the invention, the transient characteristics of the magnetic declination are utilized to realize the real-time sensing of the load disturbance, the magnetic coupling slippage caused by the sudden change of the characteristics of the sample is effectively inhibited, and the stability of the magnetic declination is improved. And the ineffective temperature rise of the motor is reduced while the system stability is guaranteed.

Description

Technical Field

[0001] This invention relates to a torque adjustment control method for a magnetic stirrer based on sample characteristics, belonging to the field of electric motor control technology. Background Technology

[0002] Currently, maintaining the stability of the stirring process by establishing a closed-loop speed or torque regulation system is the mainstream practice in the industry. Pulse width modulation technology is usually used to adjust the stator current to match the resistance torque generated by the rotation of the stir bar in the fluid. In actual biochemical synthesis or polymerization reactions, the rotating magnetic field of the motor stator and the external magnetic stir bar are not rigidly connected, but are connected by a non-contact elastic transmission link formed by a magnetic coupler. As the sample viscosity evolves nonlinearly during the reaction process, the magnetic declination between the inner and outer magnetic rotors continues to expand with the increase of load. This hysteresis of the dynamic response caused by the elastic characteristics of the magnetic field makes the traditional control method based on rotor position or average torque current face physical constraints when dealing with complex load changes.

[0003] Under high-load conditions of magnetic stirring, the extent of magnetic declination expansion determines the instability boundary of the magnetic coupling link. When the sample viscosity increases abruptly or non-Newtonian fluid characteristics appear, conventional proportional-integral controllers often lack the ability to detect the magnetoelastic potential energy accumulation process, causing the timing of compensation current injection to lag behind the evolution of physical slip. To prevent magnetic coupling slippage and skipping phenomena, the industry practice is to reserve a large torque margin or implement a shutdown-restart strategy after detecting a step-out signal. This static strategy, which sacrifices electromagnetic conversion efficiency, not only causes the motor to operate in a low-efficiency region and generates unnecessary armature temperature rise, but also induces system collapse at the reaction critical point due to the inability to offset magnetic declination fluctuations in real time. Currently, there are many directions for improvement. Limited by optimizing the physical layout of the magnetic transmission structure, the load resistance is improved by enhancing the hardware coupling strength of the magnetic circuit, but the real-time capture and closed-loop countermeasure of magnetic declination fluctuations at the control level are ignored. For example, Chinese invention patent with publication number CN111135758A discloses a high-torque magnetic stirrer, which provides a large mechanical torque by setting a transmission disk at the bottom of the cylinder and using the edge of the transmission disk to mesh with the drive motor. This solution solves the problem of power source output under heavy load conditions, but it is still essentially a static mechanical transmission method. It lacks a dynamic adaptive adjustment mechanism to deal with the nonlinear evolution of sample viscosity with the reaction process, especially when sudden magnetic slippage caused by non-Newtonian fluid characteristics occurs, and cannot eliminate the inherent hysteresis instability of the magnetic coupling system.

[0004] Therefore, the technical problem to be solved by this invention is how to extract the micro-harmonic characteristics in the motor operating parameters to reconstruct the dynamic stiffness factor of the magnetic coupler, and based on this, establish an armature magnetic field adjustment mechanism that can adaptively compensate for magnetic declination deviation and suppress viscoelastic oscillation. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A torque adjustment control method for a magnetic stirrer based on sample characteristics, comprising the following steps:

[0006] Step S1: Obtain the stator three-phase current and rotor position signal of the permanent magnet synchronous motor, and use coordinate transformation logic to map the stator three-phase current to the rotating coordinate system to obtain the quadrature axis measured current and the direct axis measured current.

[0007] Step S2: Based on the measured current of the quadrature axis, the measured current of the direct axis, and the rotor position signal, the magnetic declination estimate, which characterizes the lag of the stirrer magnetic field relative to the driving magnetic field, is extracted using a preset electromagnetic coupling model.

[0008] Step S3: Obtain the estimated value of magnetic declination within the continuous sampling period. By calculating the ratio of the difference of the estimated value of magnetic declination at adjacent sampling times to the sampling period, the rate of change of magnetic declination, which characterizes the transient evolution rate of the magnetic coupling system, is obtained.

[0009] Step S4: Based on the rate of change of magnetic declination, a transient compensation current is generated by a preset nonlinear compensation operator. The proportional coefficient of the nonlinear compensation operator is adjusted online based on the real-time synchronous torque coefficient of the permanent magnet synchronous motor. The real-time synchronous torque coefficient is pre-calibrated based on the magnetic circuit saturation state of the measured current on the direct axis.

[0010] Step S5: The transient compensation current is superimposed on the quadrature axis current command value to output the corrected quadrature axis control vector to the pulse width modulation module. The electromagnetic compensation torque is generated in the opposite direction to the magnetic declination rate by the instantaneous fine adjustment of the armature magnetic field rotation angular velocity.

[0011] Preferably, step S3 specifically includes: extracting the estimated value of magnetic declination at the current sampling time. and the estimated magnetic declination at the previous sampling time. Based on the sampling period T, the rate of change of magnetic declination is calculated using the following formula. : Among them, the rate of change of magnetic declination The transient compensation current injection delay is less than 50 μs, which is the elastic load disturbance characteristic generated by the stir bar during the stirring process due to the non-Newtonian fluid properties of the sample.

[0012] Preferably, after step S2, the method further includes: extracting transient current response data of the stator circuit within the zero vector segment generated by the pulse width modulation module; performing online identification of stator resistance parameters based on the transient current response data, and calculating the winding temperature drift correction coefficient of the permanent magnet synchronous motor; and updating the parameters of the electromagnetic coupling model using the winding temperature drift correction coefficient to eliminate magnetic declination estimation error.

[0013] Preferably, before step S1, the method further includes: superimposing a high-frequency sinusoidal perturbation signal on the basic speed command of the permanent magnet synchronous motor; extracting the response characteristic component in the cross-axis measured current that has the same frequency as the high-frequency sinusoidal perturbation signal; and linearly decoupling the mechanical friction torque and the sample viscous torque based on the amplitude dependence of the response characteristic component.

[0014] Preferably, the frequency range of the high-frequency sinusoidal perturbation signal is set to 100Hz to 500Hz.

[0015] Preferably, in step S4, the output logic of the nonlinear compensation operator is limited to: when the absolute value of the rate of change of magnetic declination exceeds the preset out-of-step critical threshold, the amplitude of the transient compensation current is increased.

[0016] Preferably, the method further includes the following steps: calculating the frequency distribution characteristics of the characteristic ripple component in the cross-axis measured current; comparing the frequency distribution characteristics with a preset sample phase transition model; and outputting an early warning signal characterizing a sudden change in sample viscosity when the evolution slope of the frequency distribution characteristics reaches the trigger condition.

[0017] Preferably, while outputting the warning signal, the maximum value of the rate of change of magnetic declination is locked, and the dynamic response bandwidth of the nonlinear compensation operator is reset based on the maximum value.

[0018] Preferably, in step S5, the adjustment range of the electromagnetic compensation torque is limited to between 0.8 and 1.2 of the rated torque of the permanent magnet synchronous motor.

[0019] Preferably, step S5 is executed by the hardware multiplier inside the processor to ensure that the calculation frequency of the transient compensation current is consistent with the switching frequency of the pulse width modulation module.

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] 1. In the torque regulation control of sample characteristics, by extracting the characteristic ripple component related to the magnetic declination of the magnetic coupler in the quadrature shaft current component of the drive motor, a real-time mapping relationship between the armature magnetic field of the motor and the physical displacement of the external magnetic rotor is established. When the sample viscosity evolution causes load torque fluctuations, electromagnetic compensation opposite to the magnetoelastic slip trend is generated by adjusting the phase lead angle of the voltage space vector. This active counterbalancing mechanism based on the elastic characteristics of the magnetic field enables the magnetic stirring system to maintain the dynamic balance of the inner and outer magnetic rotors when facing high viscosity step loads, avoiding the derailment of the stirrer caused by phase mismatch of the power transmission link.

[0022] 2. By utilizing the sinusoidal perturbation signal superimposed on the basic speed command of the drive motor and the extraction of the same-frequency response characteristics in the quadrature-axis current component, physical-level decoupling of mechanical contact friction and fluid viscous resistance is achieved. By identifying the response differences of different resistance sources to frequency perturbations, the zero-point offset of the load torque characteristic model is corrected in real time, ensuring that the torque compensation command accurately applies to the rheological characteristics of the sample. This eliminates the systematic interference caused by the change of positive pressure at the bottom of the stir bar and container on the current feedback, and improves the qualitative identification accuracy of the system for key phase transition nodes of the sample.

[0023] 3. By utilizing the zero vector segment of the pulse width modulation period to perform transient current detection, online identification of changes in motor stator parameters and chain updates of model parameters are achieved. Through real-time tracking of electromagnetic parameter temperature drift, the load torque estimation deviation caused by motor heating is eliminated, maintaining the logical consistency of the mapping relationship between current vector and physical resistance torque throughout the entire experimental cycle. This self-calibration mechanism, which utilizes the drive time slot, ensures that the drive motor operates within the high-efficiency range of electromagnetic conversion while reducing the ineffective armature temperature rise caused by excessive torque margin settings. Attached Figure Description

[0024] Figure 1 This is the main control flowchart for the adaptive torque adjustment of the magnetic stirrer of the present invention;

[0025] Figure 2 This is a functional logic architecture diagram of the magnetic stirrer torque adjustment control system of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments; it should be understood that the following embodiments are only used to explain the present invention and are not intended to limit the scope of protection of the present invention.

[0027] This invention discloses a torque regulation control method for a magnetic stirrer based on sample characteristics. It utilizes the electromagnetic parameter changes of a permanent magnet synchronous motor to invert the mechanical state evolution of an external magnetic coupler. Based on current vector analysis in a synchronous rotating coordinate system, this method extracts the ripple characteristics of the quadrature-axis current component, which includes sample viscosity and viscoelastic information. By constructing a magnetic circuit elastic feedback closed loop, a compensating torque is generated to offset magnetic declination fluctuations. The system mainly consists of a phase current acquisition module, a coordinate transformation module, a magnetic coupling state observation module, a transient compensation current generation module, and a space vector pulse width modulation module. These modules interact through real-time data streams, achieving controlled regulation of transient energy exchange between the external stirrer and the driving magnetic field at the electromagnetic level. During the operation of the magnetic stirrer's drive motor, the stator three-phase current fluctuates with the load torque. To extract characteristic parameters directly related to the stirrer load, the system employs a current vector analysis procedure. A sampling resistor acquires the three-phase stator current signal of the motor. The coordinate transformation module calculates the measured quadrature-axis current in the synchronous rotating coordinate system based on the acquired three-phase stator current and rotor position signal. Measured current along the direct axis The time-varying alternating phase current is converted into a rotating coordinate system parameter with DC properties, where the cross-axis measured current is... The DC average value characterizes the steady-state output torque of the motor, while the measured current superimposed on the quadrature axis... The micro-harmonic components on the surface carry the dynamic characteristics of the magnetic coupler; during the initial operation phase of the system, a pre-calibration procedure based on frequency-controlled scanning is executed, driving the permanent magnet synchronous motor to perform no-load operation in the speed range of 500rpm to 1500rpm, while simultaneously using the current sampling loop to obtain the measured quadrature-axis current. Discrete sequences are analyzed, and the characteristic ripple energy density at a frequency twice the rotor frequency is extracted using Fast Fourier Transform. Combined with the rotor position signal, the initial phase deviation of the stirrer under zero-load conditions is calculated. Numerical fitting of the initial phase deviation at different speed points is performed using the least squares method to obtain a dedicated synchronous torque coefficient mapping table for the stirrer unit. The actual stirring operation process is then used to extract the current operating conditions. The energy increment obtained by subtracting the initial phase deviation value from the characteristic ripple energy density is input as an independent variable into the electromagnetic coupling model. This model has a pre-set numerical mapping table containing 16 sets of discrete feature points. The energy density value ranges from 0.05 mJ to 3.20 mJ, divided in 0.20 mJ increments. This corresponds to outputting magnetic declination estimates ranging from 5.0 to 45.0 degrees. Intermediate data within the table are supplemented using a linear interpolation algorithm. The final output is a magnetic declination estimate characterizing the lag of the stirrer's magnetic field relative to the driving magnetic field. .

[0028] The phase difference between the inner and outer magnetic rotors, i.e., magnetic declination, generated during power transmission by the magnetic coupler introduces fluctuations of a specific frequency into the motor's back electromotive force. Since existing technologies typically ignore this magnetoelastic characteristic, they cannot effectively offset the evolution of physical slip when the sample viscosity changes abruptly. This solution employs a magnetic declination estimation procedure based on characteristic ripple to estimate the cross-axis measured current. Characteristic ripple components related to the magnetic declination of the magnetic coupler are extracted. Using a pre-defined electromagnetic coupling model, an estimate of the magnetic declination, representing the lag of the stirrer's magnetic field relative to the driving magnetic field, is extracted based on the energy distribution of the characteristic ripple components. This procedure utilizes the high-frequency component of the motor stator current as a virtual sensor to achieve online identification of the displacement state of physically isolated loads. Addressing transient shocks caused by non-Newtonian fluid characteristics commonly found in biochemical reactions, where average current feedback alone is insufficient to suppress torsional oscillations in magnetically coupled links, a transient damping procedure based on the magnetic declination evolution rate is implemented. The control unit acquires the estimated magnetic declination value within a continuous sampling period. and the estimated magnetic declination at the previous sampling time. The rate of change of magnetic declination, which characterizes the transient evolution rate of the magnetically coupled system, can be calculated using the following formula. : ,in, This represents the rate of change of magnetic declination. This is the estimated value of magnetic declination at the current sampling time. The magnetic declination is the estimated value at the previous sampling time, and T is the sampling period. In this embodiment, the sampling period T is set to 50 μs, based on the rate of change of magnetic declination. Transient compensation current is generated using a nonlinear compensation operator. The transient compensation current It is inversely mapped to the trend of magnetic declination and is used to generate an electromagnetic damping term inside the motor to suppress mechanical oscillations.

[0029] The processor acquires the magnetic declination estimate within a continuous sampling period. and the estimated magnetic declination at the previous sampling time The rate of change of magnetic declination, which characterizes the transient evolution rate of the magnetically coupled system, was calculated. According to the rate of change of magnetic declination The absolute value determines the piecewise gain coefficient. ,when The absolute value is less than the preset out-of-step threshold. At that time, set To maintain the steady-state response of the current loop using the first proportional coefficient, when The absolute value reaches or exceeds the aforementioned out-of-step threshold. At that time, set Switching to a second proportional coefficient to enhance dynamic damping, where the second proportional coefficient is greater than the first, transient compensation current... Amplitude from and The product is determined, and the result is... After being capped and superimposed onto the quadrature-axis current command value, the above multiplication and superposition logic is executed in a single instruction cycle by the processor's internal hardware multiplier, ensuring that the transient compensation current injection delay is less than 50μs at the physical level, generating a value similar to the magnetic declination rate. Electromagnetic compensation torques in opposite directions; transient compensation currents The quadrature-axis current command value is superimposed on the output, and the corrected quadrature-axis control vector is output to the pulse width modulation module. Based on the input control vector, the rotational angular velocity of the armature magnetic field is finely adjusted by changing the duty cycle of the inverter power transistors, generating a rate of change of magnetic declination. The opposing electromagnetic compensation torque is calculated and output by the hardware processor within a control time slot of less than 50μs after receiving current feedback, using an internal hardware multiplier. This adjustment method does not rely on adding mechanical damping components, but rather utilizes the rapid change of electromagnetic energy gradient to absorb the transient oscillating kinetic energy of the magnetic coupler. The hardware processor executes a real-time voltage vector phase injection procedure, and the control unit obtains the transient compensation current. Then, the arctangent function is used to convert it into the rotational angular velocity correction Δω of the voltage space vector. The pulse width modulation module adjusts the timing of the trigger pulses of the three-phase inverter according to the rotational angular velocity correction Δω. When the magnetic declination estimate is... It shows an increasing trend and its rate of change When the value is positive, the system reduces the duration of the effective voltage vector in the current sector, causing the instantaneous rotational speed of the armature magnetic field to lag behind the mechanical speed of the rotor. This allows the induced electromagnetic force between the armature magnetic field and the stirrer magnetic field to generate a damping term that suppresses oscillation, thus achieving closed-loop hedging against transient mechanical energy fluctuations in the magnetic coupling system.

[0030] In the high-viscosity operating range, the mechanical frictional resistance generated between the stir bar and the bottom of the container is equal to the fluid viscous resistance. The high degree of overlap in the components leads to deviations in the system's judgment of the sample's phase transition point. To address this challenge, the system employs a frequency domain signal injection detection procedure, superimposing a high-frequency sinusoidal perturbation signal onto the motor's base speed command. The frequency range of this signal is set from 100Hz to 500Hz, and the measured cross-axis current is acquired simultaneously. The response characteristic component, which is in the same frequency as the high-frequency sinusoidal perturbation signal, is linearly decoupled from the mechanical frictional resistance torque and the sample viscous resistance torque based on the dependence of torque response amplitude and phase at different frequencies. The system utilizes the identified mechanical frictional resistance component to update the zero-point offset of the load torque model in real time, ensuring that the transient compensation current accurately applies to the rheological properties of the sample. During long-term operation, the motor experiences temperature rise, leading to increased stator resistance. The drift of the torque constant causes errors in magnetic declination estimation. To eliminate temperature field interference, the system utilizes online parameter identification during the zero-vector segment of the pulse width modulation period. During the zero-vector time slot of the inverter, transient current response data of the stator circuit is extracted. The calculation unit identifies the online change in stator resistance based on this transient response and calculates the winding temperature drift correction coefficient for the permanent magnet synchronous motor. This coefficient is used to correct the electromagnetic parameters in the electromagnetic coupling model in real time, compensating for the magnetic declination estimation deviation caused by thermal effects. This procedure ensures the constancy of the mapping ratio between the current vector and the physical resistance torque. To address magnetic saturation conditions... Under nonlinear distortion, the system performs inductor sensitivity monitoring, superimposes a high-frequency perturbation voltage on the control vector, and detects the instantaneous current response generated by the motor to determine the inductor sensitivity parameter. When the inductor sensitivity parameter is lower than the preset magnetic circuit saturation threshold, the system performs amplitude limiting processing on the current compensation amount of the output torque and simultaneously reduces the target speed of the motor. In addition, by calculating the spectral energy distribution entropy of the characteristic ripple component, the viscoelastic evolution characteristics of the sample are identified, and a virtual damping amount is injected into the current loop. This mechanism improves the mechanical stability of the magnetic stirrer under extreme heavy load conditions by limiting ineffective heat loss and suppressing low-frequency oscillation.

[0031] Example 1: In the high-concentration epoxy resin polymerization reaction, the fluid viscosity increases from 500 mPa·s to 15000 mPa·s within 300 seconds as the degree of polymerization increases. Furthermore, the system exhibits non-Newtonian shear-thinning characteristics, causing a sudden jump in the nonlinear resistance torque experienced by the external stir bar during rotation. This results in the stir bar's magnetic field lag angle relative to the driving magnetic field approaching the physical boundary of the synchronous torque, creating a risk of instantaneous slippage between the stir bar and the internal permanent magnet synchronous motor magnetic field. The system extracts the cross-axis measured current in the synchronous rotating coordinate system through a coordinate transformation module. Under this operating condition, due to the change in load viscous damping, the measured current across the quadrature axis... A set of characteristic ripples with a frequency twice that of the motor rotor frequency is generated in the high-frequency components. This ripple carries the dynamic distortion information of the magnetic declination of the magnetic coupler. To verify the stability of this technical solution under extreme variable loads, the control unit uses the magnetic coupling state observation module to extract the energy density of this characteristic ripple and calculates the estimated value of the magnetic declination, which characterizes the degree of phase lag in magnetic field coupling. The magnetic declination is used as input to the transient damping protocol. The system locks the magnetic declination at two adjacent moments within a sampling period T of 50 μs, and extracts the rate of change of magnetic declination using differential operations. The calculation formula is as follows: ,in, This represents the rate of change of magnetic declination. This is the estimated value of magnetic declination at the current sampling time. is the estimated magnetic declination at the previous sampling time, and T is the sampling period.

[0032] Magnetic declination estimate Real-time acquisition for transient compensation current The generation provides the necessary technical prerequisites, including transient compensation current. By applying a reverse action to the quadrature-axis control loop, the motor's ability to suppress external magnetic field disturbances is enhanced, forming a collaborative adjustment mechanism based on the sensing of magnetic field elastic potential energy. This allows the system to compensate for the physical hysteresis caused by the viscoelasticity of the sample by microsecond-level phase substitution of the driving magnetic field angular velocity without increasing the motor's rated power. It resolves the technical dilemma between stable torque output and response speed during high-viscosity stirring. The system no longer relies on a static torque margin to prevent slippage, but instead uses a dynamic electromagnetic damping term to adaptively counteract load impacts. This transforms the previously experience-based torque protection into deterministic control based on the slope of magnetic declination evolution. At the moment when the epoxy resin polymerization reaction reaches its viscosity peak of 15000 mPa·s, the system monitors the rate of change of magnetic declination. Operator injection that triggers nonlinear compensation, transient compensation current The amplitude is adjusted to 1.1 times the rated current according to the magnetic circuit saturation state, suppressing the physical oscillation of the stir bar, eliminating the skipping phenomenon, and maintaining continuous and stable operation during the stirring process under viscosity change conditions.

[0033] Example 2: In a magnetic stirring test system driven by a permanent magnet synchronous motor, the test environment simulates the electromagnetic interference conditions of an actual industrial site. The sampling frequency of the current sensor is set to 20kHz, the resolution to 12bit, and the sampling accuracy controlled within 0.05%, used to capture the measured quadrature-axis current. The system detects the microscopic ripple characteristics and balances the real-time performance of the signal with the computational load of the processor by setting the sampling period T. Based on the physical bandwidth determination rules of the magnetic coupling system, when the monitored characteristic ripple spectrum coverage is within 1kHz, a sampling period T of 50μs is selected to avoid signal aliasing. A circular buffer with a depth of 256 sampling points is created in memory. Every time 128 new sampling data points are updated in the buffer, a Fast Fourier Transform (FFT) is triggered, satisfying a 50% time window overlap rate. This value provides at least 20 sampling points per characteristic peak at 3000rpm, ensuring the accuracy of the magnetic declination estimation. The discretization accuracy.

[0034] To verify the reliability of this technical solution under electromagnetic conditions, the experimental group superimposed 20dB of Gaussian white noise and 50Hz power frequency interference harmonics in the acquisition channel, and set up three sample groups for comparison. The sample group of this invention adopted the complete technical solution, while the control group adopted conventional vector control. The partially missing control group only retained magnetic declination observation and turned off the transient damping procedure. During the dynamic process of the sample viscosity stepping from 500 mPa·s to 10000 mPa·s, the cross-axis measured current of the control group was... The system exhibits periodic low-frequency oscillations with an amplitude reaching 15.6% of the rated current, and the magnetic declination suddenly increases to over 90 degrees at the moment of slippage, causing system shutdown. The present invention's sample group obtains the estimated magnetic declination value by extracting the characteristic ripple component. Its key intermediate data shows the rate of change of magnetic declination. The viscosity reaches 48.5 rad / s at the viscosity step, triggering the transient compensation current. Generation, transient compensation current A suppressive torque is generated in the armature magnetic field, causing the cross-axis measured current to... The fluctuation amplitude decreased from 15.6% to 2.4%, and the measured slip critical torque increased by 40.2% compared with the control group.

[0035] The experiment further investigated the response of the sample group of the present invention under gradient changes in sample properties. Four intensity levels were set with sample viscosities of 1000 mPa·s, 5000 mPa·s, 10000 mPa·s, and 15000 mPa·s. The monitoring data showed the rate of change of magnetic declination. The peak value exhibits a regular distribution with viscosity grade, corresponding to the generated transient compensation current. The measured current along the direct axis was measured with amplitudes of 0.22A, 1.13A, 2.78A, and 4.25A, respectively, under extreme conditions where the viscosity exceeded 18000 mPa·s. The resulting deep saturation of the magnetic circuit caused the inductor sensitivity parameter to drop to 0.32mH, triggering the system's built-in physical limiting logic and transient compensation current. Locked at 1.1 times the rated current, it no longer varies with the rate of change of magnetic declination. The target speed was automatically reduced from 1000 rpm to 200 rpm to maintain magnetic field coupling. The experimental results showed that by mapping the ripple energy of the armature current to the dynamic evolution characteristics of the magnetic declination, the system suppressed the transient instability of the magnetic coupling system. This process proved that by injecting a virtual damping term into the electromagnetic control loop, the technical problem between stable output and dynamic response in high viscosity stirring scenarios was solved, and deterministic regulation based on the sensing of magnetic field elastic potential energy was achieved.

[0036] Example 3: For silicone oil samples with non-Newtonian fluid properties, the system acquires torque compensation parameters. The test platform includes a floating-point processing core with a computing frequency of 100MHz and a current feedback register with 12-bit precision. Transient instability suppression is performed at the electromagnetic control level. Under no-load conditions on the motor, the control unit adjusts the voltage vector to ensure the measured direct-axis current... The current is increased from 0A to 10A in 0.5A increments, and the synchronous torque value of the motor under each steady-state current condition is recorded simultaneously using a torque sensor. Calculate the synchronous torque value Corresponding cross-axis measured current The ratio is used to establish a linear mapping table for the real-time synchronous torque coefficient; the control unit acquires the measured quadrature-axis current with a sampling period T of 50μs. The discrete sequence is smoothed using a Hanning window weighted smoothing method with a sliding window containing 256 sampling points. The sum of squares of harmonic amplitudes at frequencies twice the rotor frequency is extracted and defined as the energy density value of the characteristic ripple component. Magnetic declination estimation is triggered when the energy density value exceeds the ambient background noise by 10 dB. The solution program.

[0037] During the operation phase where the viscosity of the silicone oil sample changes nonlinearly with shear rate, the hardware processor superimposes a sinusoidal perturbation signal with a frequency of 1kHz and an amplitude of 5V onto the d-axis control voltage. By analyzing the response amplitude of the stator current, the inductor sensitivity parameter is obtained. When the sample viscosity gradient increases to 20000 mPa·s, causing the magnetic circuit to tend to saturate, the inductor sensitivity parameter decreases from the initial 1.25mH to 0.85mH. Based on this decreasing gradient, the system determines that it is currently in the critical operating condition of the magnetic circuit saturation threshold. At this time, the control unit calculates the spectral energy distribution entropy of the characteristic ripple component, normalizes the spectral energy within the sliding window according to the frequency range, and obtains a quantitative index reflecting the degree of frequency distribution disorder. When the evolution slope of this index exceeds... When the sample is determined to have entered the viscoelastic phase transition stage, the control unit locks the rate of change of magnetic declination. The maximum value, and the transient compensation current The amplitude gain was reduced by 15% to suppress control oscillations in the magnetic saturation region. Through the above calibration and algorithm execution procedures, the system achieved a millisecond-level response to viscosity steps in silicone oil samples within the physical safety boundary. The magnetic coupling slippage critical point measured in the experiment shifted by 35.8% towards heavy-load conditions compared to the conventional scheme without compensation, achieving an optimized balance between stable output and system heat loss under high viscosity stirring conditions.

[0038] Example 4: During the on-site commissioning of the new magnetic stirrer unit, to ensure that the electromagnetic coupling model matches the physical response characteristics of the physical entity, the control unit executes a pre-calibration procedure for the magnetic coupling parameters based on controlled self-testing. In the initial state where the reaction vessel is not loaded with samples, the permanent magnet synchronous motor is driven to perform a frequency step scan within the speed range of 500 rpm to 1500 rpm, while simultaneously using a sampling chip to acquire the cross-axis measured current. The fundamental component amplitude is calculated, and the natural phase deviation of the stirrer's magnetic field under zero-load conditions is calculated in conjunction with the rotor position signal. The hardware processor uses the least squares method to perform numerical fitting on the phase difference at different speed points, calculates and fills the synchronous torque coefficient mapping table dedicated to this stirrer unit, and thus determines the magnetic coupling stiffness factor that reflects the magnetic field energy transfer capability. .

[0039] When faced with initial impedance differences arising from different experimental environments, the control unit executes an electromagnetic baseline calibration procedure based on the current response envelope. While the motor is in a quasi-stationary state, a small-amplitude DC detection excitation is injected using a pulse width modulation module to extract the transient response characteristics of the stator phase current and calculate the stator resistance of the current circuit. In addition to the additional voltage drop losses caused by physical wiring, the obtained calibration data is stored in the weighted addressing space of the nonlinear compensation operator as the basis for generating transient compensation current. The zero-point correction offset makes the proportional gain The rate of change of magnetic declination is generated across the entire speed range. The reverse suppressive torque enables the control system to adaptively counteract the inherent magnetic characteristics of different physical batches of magnetic rotors.

[0040] Example 5: In the deployment of a large-scale industrial 100L magnetic stirring system, due to disturbances caused by the physical circuit length and terminal contact resistance, the system performs pre-calibration of the loop impedance through controlled excitation. Before the motor enters the running state, the control unit injects a DC detection pulse with an amplitude of 0.15 times the rated current of the motor using the inverter bridge arm. The duration of this DC detection pulse is set to 200ms. The hardware processor collects the stator terminal voltage and response current during this period and calculates the comprehensive stator resistance including line losses. The obtained values ​​are written into the address space of the back electromotive force compensation term in the electromagnetic coupling model as estimates of magnetic declination. The baseline correction amount enables the control system to adaptively offset system deviations caused by differences in line impedance under different physical deployment environments in load characteristic identification; during the operation of stirring polymer melt samples, in order to identify transient transitions in fluid properties, the system executes a function based on spectral energy distribution entropy. The program control unit acquires 512-point Fast Fourier Transform spectrum data processed by the Hanning window and calculates the energy proportion of each frequency point. Using the formula A quantitative index reflecting the dispersion of the current ripple spectrum is obtained. When the system detects that the rate of change of this quantitative index exceeds a certain threshold for three consecutive sampling cycles within a 500ms sampling time window, the index is considered valid. At that time, the hardware controller determines that the sample has entered the critical region of non-Newtonian fluid phase transition and simultaneously increases the proportional gain of the nonlinear compensation operator. Switch to sensitivity mode to enable transient compensation current. The amplitude gain coefficient was increased to 1.2 times the initial calibration value, enabling dynamic sensing and torque compensation of the magnetic field hysteresis angle evolution under heavy load viscous conditions; among which, The entropy of the spectral energy distribution. Here, k represents the energy percentage at a given frequency point, and k is the frequency index. For proportional gain, This is the transient compensation current.

[0041] In the stability stress test for high-intensity magnetic saturation conditions, the system executed an operator adaptive switching procedure based on the inductor sensitivity gradient, and the test sample group monitored the direct-axis measured current. When the current exceeds 0.8 times the rated value and the inductance sensitivity parameter drops to 0.72mH, the hardware controller determines that the internal magnetic circuit of the motor has entered a deep saturation state. At this time, the system extracts the measured quadrature-axis current. Spectral energy distribution entropy of characteristic ripple components The value was monitored to be stable around 2.45. Based on the preset saturation correction model, the control unit reduced the real-time synchronous torque coefficient from the initial 1.0 to 0.82, and simultaneously adjusted the proportional gain. The value of was compressed from 1.2 to 0.75. The measured data showed that the step response overshoot of the current loop decreased from 12.4% to 3.6%. This process confirmed that the nonlinear oscillation caused by magnetic saturation could be suppressed by dynamically adjusting the bandwidth of the control operator.

[0042] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method of controlling the torque of a magnetic stirrer based on the characteristics of a sample, characterized by, The method comprises the following steps: Step S1, obtaining the stator three-phase current of the permanent magnet synchronous motor and the rotor position signal, and mapping the stator three-phase current to the rotating coordinate system by using the coordinate transformation logic to obtain the measured quadrature axis current and the measured direct axis current; Step S2, based on the measured quadrature axis current, the measured direct axis current and the rotor position signal, using a preset electromagnetic coupling model to extract a magnetic bias angle estimation value representing the lag degree of the stirrer magnetic field relative to the driving magnetic field; Step S3, obtaining the magnetic bias angle estimation value in the continuous sampling period, and obtaining the magnetic bias angle change rate representing the transient evolution rate of the magnetic coupling system by calculating the difference between the magnetic bias angle estimation values at adjacent sampling times and the sampling period; Step S4, according to the magnetic bias angle change rate, generating a transient compensation current through a preset nonlinear compensation operator, the proportional coefficient of the nonlinear compensation operator is adjusted online according to the real-time synchronous torque coefficient of the permanent magnet synchronous motor, and the real-time synchronous torque coefficient is pre-calibrated according to the magnetic circuit saturation state of the measured direct axis current; Step S5, superimposing the transient compensation current on the quadrature axis current command value to output the corrected quadrature axis control vector to the pulse width modulation module, and using the instantaneous fine adjustment of the armature magnetic field rotation angular velocity to generate an electromagnetic compensation torque opposite to the direction of the magnetic bias angle change rate.

2. The method according to claim 1, wherein the magnetic stirrer is controlled based on the torque adjustment according to the characteristics of the sample. Step S3 specifically comprises: extracting the magnetic declination estimation value at the current sampling moment and the magnetic declination estimation value at the previous sampling moment ; based on the sampling period T, calculating the magnetic declination change rate : ; wherein the magnetic declination change rate is the elastic load disturbance characteristic of the stirrer in the stirring process affected by the sample non-Newtonian fluid characteristics, and the injection delay of the transient compensation current is less than 50 μs.

3. The method according to claim 1, wherein the magnetic stirrer is characterized by a torque adjustment control based on sample characteristics. After step S2, the method further comprises: in the zero vector segment generated by the pulse width modulation module, extracting the transient current response data of the stator circuit; performing online identification of the stator resistance parameter based on the transient current response data, calculating the winding temperature drift correction coefficient of the permanent magnet synchronous motor; updating the parameters of the electromagnetic coupling model using the winding temperature drift correction coefficient to eliminate the magnetic bias angle estimation error.

4. The method of claim 1, wherein the magnetic stirrer is controlled based on the torque adjustment of the sample characteristics. Before step S1, the method further comprises: superimposing a high-frequency sinusoidal perturbation signal on the basic speed instruction of the permanent magnet synchronous motor; extracting the response characteristic component in the measured quadrature axis current which is the same frequency as the high-frequency sinusoidal perturbation signal; based on the amplitude dependence of the response characteristic component, linearly decoupling the mechanical friction torque and the sample viscous torque.

5. The method of claim 4, wherein the torque adjustment control is based on a characteristic of the sample. The frequency range of the high-frequency sinusoidal perturbation signal is set to 100Hz to 500Hz.

6. The method of claim 1, wherein the magnetic stirrer is controlled based on the torque adjustment of the sample characteristics. In step S4, the output logic of the nonlinear compensation operator is limited to: when the absolute value of the magnetic bias angle change rate exceeds a preset out-of-step critical threshold, the amplitude of the transient compensation current is increased.

7. The method of claim 1, wherein the torque adjustment is based on a characteristic of the sample. The method further comprises the following steps: Calculate the frequency distribution characteristics of the characteristic ripple component in the measured quadrature axis current; compare the frequency distribution characteristics with a preset sample phase change model; when the evolution slope of the frequency distribution characteristics reaches the trigger condition, output a warning signal representing a sudden change in the sample viscosity.

8. The method according to claim 7, wherein the torque adjustment control is based on the characteristics of the sample. At the same time of outputting the warning signal, the maximum value of the magnetic bias angle change rate is locked.

9. The method of claim 1, wherein the torque adjustment control is based on a characteristic of the sample. In step S5, the adjustment range of the electromagnetic compensation torque is limited to 0.8 to 1.2 times the rated torque of the permanent magnet synchronous motor.

Images