Automatic adjustment method for a plug pump

Abstract

The application relates to the technical field of general control or regulation systems, and discloses an automatic regulation method for a plug taking pump operation, which comprises the following steps: synchronously acquiring a voltage vector and a current vector of a controlled execution unit; calculating a transient active power by calculating a scalar product of the voltage vector and the current vector; removing a steady power consumption component in the transient active power to extract an energy recoil pulse; generating a micro-elastic recoil energy product representing load yield information by integration; monitoring a differential gradient of the micro-elastic recoil energy product; and determining an output negative gradient duty cycle signal at a collapse moment to regulate a power slope. The application uses a vector power analysis mechanism to capture a load fragmentation signal, eliminates hydraulic hysteresis of a control loop, ensures that a negative pressure release rate of a pipeline matches a load displacement increment, and improves regulation accuracy under a load impedance step working condition.

Description

Technical Field

[0001] This invention relates to an automatic adjustment method for the operation of a thrombectomy pump, belonging to the field of general control or adjustment system technology. Background Technology

[0002] Current thrombus aspiration devices typically utilize negative pressure suction to remove occlusions within blood vessels. This field generally employs a closed-loop feedback control method based on negative pressure sensing. This method monitors the fluid state by deploying pressure sensors in the pump's piping and adjusts the output power of the drive motor according to the pressure deviation, thereby maintaining a preset aspiration intensity. While this control method can meet basic steady-state aspiration requirements, it is strictly limited by physical laws when facing transient conditions such as dense thrombus embedding or catheter adhesion to the vessel wall. Since the transmission of macroscopic fluid parameters depends on the macroscopic transport of mass and momentum, there is an unavoidable hydraulic response hysteresis within the piping. When the embedded thrombus undergoes fatigue yielding and instantaneously displaces, the mechanical impedance within the piping drops abruptly. At this time, the pressure sensor is limited by the fluid wave propagation speed and inherent integral response characteristics, causing the control unit to be unable to capture the load state in the very early stages of the impedance step. This blind spot in the control dimension causes the control unit to continuously output high-frequency drive commands, resulting in motor speed overshoot during sudden load drops and inducing negative pressure surges and flow velocity pulsations.

[0003] At the hardware improvement level, existing technologies compensate for insufficient sensing capabilities by optimizing the mechanical structure to enhance the thrombus-assisted fixation effect. For example, Chinese utility model patent CN222217895U discloses a mechanical thrombectomy device with negative pressure-assisted suction. Physical limitation is achieved by setting a clamping claw and a support plate at the right end of the suction tube. Such solutions focus on macroscopic mechanical risk avoidance, and negative pressure switching depends on the mechanical action of valve plate displacement. At the software control level, it is impossible to overcome the inherent physical limits of the fluid system. To alleviate the response lag problem, conventional improvement methods usually focus on increasing the sampling frequency of the sensor or optimizing the control parameters, attempting to shorten the control delay by enhancing the sensitivity of external feedback. Analysis shows that such linear improvement methods cannot overcome the physical limitations of fluid dynamics transmission and are difficult to distinguish the physical medium properties of increased negative pressure. When identifying the two working conditions with completely different safety requirements—catheter adsorption of the blood vessel wall and aspiration of hard thrombi—the system has discrimination ambiguity, which can easily lead to erroneous shutdown when continuous aspiration is required, or mechanical damage to the blood vessel due to continuous power output when the blood vessel wall is accidentally aspirated.

[0004] Therefore, the technical problem to be solved by this invention is to find a dynamic adjustment mechanism that is free from the constraints of macroscopic fluid transport, can distinguish the evolution of load state, and has the ability to be immune to environmental noise, starting from the underlying architecture of the control system. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: An automatic adjustment method for the operation of a thrombectomy pump, comprising the following steps: Step S1: When the control unit determines that the system is in a load extraction task, it synchronously acquires the real-time stator voltage vector U and the real-time stator current vector I of the controlled execution unit in the stationary coordinate system. Step S2: Perform scalar product operation of real-time stator voltage vector U and real-time stator current vector I at the coordinate transformation layer of the control system to analyze the transient active power that characterizes the intensity of bidirectional electromechanical energy interaction between the controlled execution unit and the external controlled load in real time. Step S3: Establish a frequency-selective filtering mechanism based on dynamic speed feedback, and use a high-pass filtering algorithm to remove the steady-state power consumption component in the transient active power that corresponds to overcoming the macroscopic load resistance of the fluid, thereby unidirectionally stripping out the energy recoil pulse that characterizes the moment of fatigue yielding of the internal structure of the controlled load from the nonlinear mechanical resonance noise of the transmission system. Step S4: Perform discretization integration on the energy recoil pulse within a preset sampling time window to generate a microelastic recoil energy product for quantifying the degree of fiber breakage and potential energy release level inside the controlled load. Step S5: Continuously monitor the change of the first-order differential gradient of the micro-elastic recoil energy product. At the instant when the first-order differential gradient satisfies the preset load collapse triggering logic, actively output a negative gradient pulse width modulation duty cycle signal that has an inverse compensation relationship with the first-order differential gradient. By setting the output power drop slope of the controlled execution unit, the pipeline negative pressure potential energy release rate and the load displacement rate calculated based on the first-order differential gradient are kept in real-time physical logic offset, thereby compensating for the control feedback blind zone caused by the inherent integral response characteristics of the pressure sensor.

[0006] Preferably, the active output of the negative gradient pulse width modulation duty cycle signal in step S5 includes: real-time calculation of the microelastic recoil energy product E. back rate of change over time ; rate of change over time The preset judgment threshold is reached and the transient active power P inst When reverse energy feedback fluctuations are generated, the power drop envelope of the controlled execution unit is set through a control algorithm, so that the attenuation slope of the negative gradient pulse width modulation duty cycle signal is related to the rate of change of time. Maintain nonlinear mapping correlations to counteract the hydraulic hysteresis of the system at the earliest stage of failure of the preceding physical interface.

[0007] Preferably, in step S2, the real-time transient active power P inst The calculation follows these rules: , where P instU represents the real-time transient active power, U represents the real-time stator voltage vector, and I represents the real-time stator current vector.

[0008] Preferably, the method for removing steady-state power consumption components in step S3 includes: obtaining the real-time rotational speed n of the controlled execution unit; constructing an adaptive notch filter whose center frequency dynamically shifts with the real-time rotational speed n; and converting the real-time transient active power P... inst An adaptive notch filter is input to filter out mechanical harmonics related to the real-time rotational speed n.

[0009] Preferably, after performing step S5, the following steps are also included: step S6, controlling the controlled execution unit to enter a resting state with a current of 0A and opening a monitoring window period; step S7, during the monitoring window period, outputting a diagnostic pulse sequence to the controlled execution unit where the electromagnetic torque is less than the static friction torque of the system.

[0010] Preferably, after step S7, the method further includes: monitoring the damped oscillation damping rate ζ generated by the electromagnetic winding of the controlled execution unit under the excitation of the diagnostic pulse sequence; comparing the damped oscillation damping rate ζ with a preset load decoupling benchmark to determine the boundary attachment state between the end of the controlled execution unit and the controlled load.

[0011] Preferably, the method further includes the following step: Step S8, determining whether the damping rate ζ of the damped oscillation reaches a preset unlocking threshold. Step S9: If the damping rate of the decaying oscillation ζ reaches the unlocking threshold... If the boundary attachment state is released, the basic driving frequency will be restored.

[0012] Preferably, the method for generating the microelastic recoil energy product in step S4 includes: extracting the envelope features of the energy recoil pulse; determining the judgment weight based on the cumulative distribution of the envelope features within a sampling period of 10ms to 50ms, so as to perform a weighted summation of the energy recoil pulse.

[0013] Preferably, setting the output power reduction slope of the controlled execution unit in step S5 includes: matching a preset pressure unloading curve library according to the evolution slope of the micro-elastic recoil energy product; controlling the operating frequency of the controlled execution unit to decay along the target curve matched in the pressure unloading curve library, so that the negative pressure release speed in the pipeline is synchronized with the spatial increment generated by the controlled load.

[0014] Preferably, the method for restoring the basic drive frequency in step S9 includes: obtaining the initial load impedance at the instant the boundary attachment state is released; calculating the initial control gain of the feedback loop based on the initial load impedance; and gradually adjusting the operating frequency to the basic drive frequency according to a preset time constant.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In the automatic adjustment of the thrombectomy pump operation, this invention overcomes the constraint of fluid dynamics transmission hysteresis on the real-time control, realizing the pre-emptive adjustment action on the physical event on the time axis. Traditional control schemes rely on macroscopic pressure or flow feedback, which are limited by the mechanical wave transport delay caused by the physical length of the pipeline, resulting in the adjustment signal lagging behind the physical moment of thrombus detachment. This invention reconstructs the motor stator into a high-frequency electroacoustic impedance observer by superimposing a high-frequency micro-perturbation signal on the basic drive frequency. This mechanism captures the surface mutation of the mechanical impedance inside the controlled pipeline by extracting the high-frequency response component in the stator current and calculating its phase angle difference. Since the interaction speed of the electrical signal is much higher than the transmission speed of the fluid pressure wave, the system can trigger an asynchronous frequency reduction command at the surface yielding stage before the macroscopic displacement of the thrombus. This leap from post-remedial to pre-intervention control eliminates the speed overshoot of the motor at the moment of load step drop and prevents the generation of destructive negative pressure surges.

[0016] 2. Eliminating ambiguity in load identification under complex operating conditions and achieving precise arbitration of vascular wall adhesion and thrombus embedding, this invention solves the industry problem that a single pressure threshold cannot distinguish load attributes by synergistically applying transient ripple variance extraction and asymmetric torque step verification. When the system detects a rigid collapse in ripple variance, it actively outputs a reduced-order pulse width modulation signal to induce a negative step in electromagnetic torque. Utilizing the physical asymmetry between vascular wall tension and thrombus rebound displacement in elastic deformation recovery time, the load nature is inverted by observing the transient ripple change trajectory at the moment of torque drop. If the ripple variance remains low, it is determined to be anatomical vascular wall adhesion; if the ripple variance jumps, it is determined to be a hard thrombus. This logic arbitration mechanism based on active detection not only ensures immediate shutdown protection in the case of false adhesion but also ensures that the suction offensive can be maintained when facing calcified thrombi, achieving simultaneous improvement in safety and surgical efficiency.

[0017] 3. To enhance control stability under electromagnetic interference and slow time-varying environments and ensure the discrimination accuracy throughout the entire surgical cycle, this invention constructs a feature component orthogonal decoupling and dynamic threshold reconstruction system based on a vector control architecture. By using coordinate transformation, the real-time stator current is decomposed into excitation current components and torque current components. The fluctuation of the excitation component is regarded as the common-mode noise floor of the system's electrical environment, and this noise floor is subtracted from the total fluctuation of the torque component to generate decoupling ripple variance. On this basis, the system automatically identifies the steady-state suction window by utilizing the trough period of the first derivative of the fundamental current, extracts the environmental baseline in real time, and superimposes it onto the initial threshold. This mechanism collaboratively eliminates high-frequency noise from the power grid, interference from equipment on the same power grid, and baseline drift caused by motor temperature rise and changes in blood viscosity, enabling the control loop to accurately lock the precursor features of thrombus yielding even in extremely harsh and variable physical environments. Attached Figure Description

[0018] Figure 1This is a flowchart illustrating the automatic adjustment process for transient power analysis and recoil energy monitoring in this invention. Figure 2 This is a diagram illustrating the electromechanical interaction logic between the thrombectomy pump control unit and the controlled load of this invention.

[0019] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0021] An automatic adjustment method for the operation of a thrombectomy pump includes the following steps: Step S1: When the control unit determines that the system is in a load extraction task, it synchronously acquires the real-time stator voltage vector U and the real-time stator current vector I of the controlled execution unit in the stationary coordinate system. Step S2: Perform scalar product operation of real-time stator voltage vector U and real-time stator current vector I at the coordinate transformation layer of the control system to analyze the transient active power that characterizes the intensity of bidirectional electromechanical energy interaction between the controlled execution unit and the external controlled load in real time. Step S3: Establish a frequency-selective filtering mechanism based on dynamic speed feedback, and use a high-pass filtering algorithm to remove the steady-state power consumption component in the transient active power that corresponds to overcoming the macroscopic load resistance of the fluid, thereby unidirectionally stripping out the energy recoil pulse that characterizes the moment of fatigue yielding of the internal structure of the controlled load from the nonlinear mechanical resonance noise of the transmission system. Step S4: Perform discretization integration on the energy recoil pulse within a preset sampling time window to generate a microelastic recoil energy product for quantifying the degree of fiber breakage and potential energy release level inside the controlled load. Step S5: Continuously monitor the change of the first-order differential gradient of the micro-elastic recoil energy product. At the instant when the first-order differential gradient satisfies the preset load collapse triggering logic, actively output a negative gradient pulse width modulation duty cycle signal that has an inverse compensation relationship with the first-order differential gradient. By setting the output power drop slope of the controlled execution unit, the pipeline negative pressure potential energy release rate and the load displacement rate calculated based on the first-order differential gradient are kept in real-time physical logic offset, thereby compensating for the control feedback blind zone caused by the inherent integral response characteristics of the pressure sensor.

[0022] Preferably, the active output of the negative gradient pulse width modulation duty cycle signal in step S5 includes: real-time calculation of the microelastic recoil energy product E. backrate of change over time ; rate of change over time The preset judgment threshold is reached and the transient active power P inst When reverse energy feedback fluctuations are generated, the power drop envelope of the controlled execution unit is set through a control algorithm, so that the attenuation slope of the negative gradient pulse width modulation duty cycle signal is related to the rate of change of time. Maintain nonlinear mapping correlations to counteract the hydraulic hysteresis of the system at the earliest stage of failure of the preceding physical interface.

[0023] Preferably, in step S2, the real-time transient active power P inst The calculation follows these rules: , where P inst U represents the real-time transient active power, U represents the real-time stator voltage vector, and I represents the real-time stator current vector.

[0024] Preferably, the method for removing steady-state power consumption components in step S3 includes: obtaining the real-time rotational speed n of the controlled execution unit; constructing an adaptive notch filter whose center frequency dynamically shifts with the real-time rotational speed n; and converting the real-time transient active power P... inst An adaptive notch filter is input to filter out mechanical harmonics related to the real-time rotational speed n.

[0025] Preferably, after performing step S5, the following steps are also included: step S6, controlling the controlled execution unit to enter a resting state with a current of 0A and opening a monitoring window period; step S7, during the monitoring window period, outputting a diagnostic pulse sequence to the controlled execution unit where the electromagnetic torque is less than the static friction torque of the system.

[0026] Preferably, after step S7, the method further includes: monitoring the damped oscillation damping rate ζ generated by the electromagnetic winding of the controlled execution unit under the excitation of the diagnostic pulse sequence; comparing the damped oscillation damping rate ζ with a preset load decoupling benchmark to determine the boundary attachment state between the end of the controlled execution unit and the controlled load.

[0027] Preferably, the method further includes the following step: Step S8, determining whether the damping rate ζ of the damped oscillation reaches a preset unlocking threshold. Step S9: If the damping rate of the decaying oscillation ζ reaches the unlocking threshold... If the boundary attachment state is released, the basic driving frequency will be restored.

[0028] Preferably, the method for generating the microelastic recoil energy product in step S4 includes: extracting the envelope features of the energy recoil pulse; determining the judgment weight based on the cumulative distribution of the envelope features within a sampling period of 10ms to 50ms, so as to perform a weighted summation of the energy recoil pulse.

[0029] Preferably, setting the output power reduction slope of the controlled execution unit in step S5 includes: matching a preset pressure unloading curve library according to the evolution slope of the micro-elastic recoil energy product; controlling the operating frequency of the controlled execution unit to decay along the target curve matched in the pressure unloading curve library, so that the negative pressure release speed in the pipeline is synchronized with the spatial increment generated by the controlled load.

[0030] Preferably, the method for restoring the basic drive frequency in step S9 includes: obtaining the initial load impedance at the instant the boundary attachment state is released; calculating the initial control gain of the feedback loop based on the initial load impedance; and gradually adjusting the operating frequency to the basic drive frequency according to a preset time constant.

[0031] Example 1: In a scenario involving thrombus removal within human blood vessels, the thrombectomy pump system faces heterogeneous thrombus impaction, with the pipeline under static negative pressure. The control unit executes an automatic adjustment method for the thrombectomy pump operation. This involves synchronously acquiring the real-time stator voltage vector U and the real-time stator current vector I of the controlled execution unit in a stationary coordinate system. At the coordinate transformation level, a scalar product operation is performed on the real-time stator voltage vector U and the real-time stator current vector I. This allows for the analysis of the transient active power P, which characterizes the intensity of the bidirectional electromechanical energy interaction between the controlled execution unit and the external controlled load. inst Among them, the real-time transient active power P inst The calculation formula is as follows: , where P inst For real-time transient active power, U is the real-time stator voltage vector, and I is the real-time stator current vector. At the instant the controlled load undergoes structural yielding, the accumulated negative pressure potential energy within the piping system is released, driving the impeller of the controlled actuator to generate instantaneous angular acceleration. The rotor speed, within microseconds, tends to exceed the synchronous speed of the stator magnetic field, generating a back electromotive force higher than the stator terminal voltage. This causes the current vector direction to shift and feed energy back to the drive circuit side, resulting in transient active power P. inst The system extracts negative energy pulse characteristics with amplitudes ranging from 0.5W to 2.0W and pulse widths from 5ms to 15ms from real-time waveform analysis. Addressing the mechanical resonance noise generated by the thrombus removal pump drive motor during suction operations, the system establishes a frequency-selective filtering mechanism based on dynamic speed feedback. By acquiring the real-time speed n of the controlled execution unit, an adaptive notch filter is constructed whose center frequency dynamically shifts with the real-time speed n, thus filtering the real-time transient active power P. inst An adaptive notch filter is input, and a high-pass filtering algorithm is used to remove transient active power P. inst The component corresponding to the steady-state power consumption component that overcomes the overall load resistance of the fluid is used to unidirectionally extract the energy recoil pulse that characterizes the yielding of the internal structure of the controlled load from the mechanical resonance noise of the transmission system.

[0032] To address the elastic potential energy release phenomenon before thrombus detachment, the system performs discretized integration on the energy recoil pulse within a preset sampling time window to generate a micro-elastic recoil energy product that quantifies the degree of fiber breakage and the potential energy release level within the controlled load. Specifically, the system extracts the envelope features of the energy recoil pulse and determines the judgment weight based on the cumulative distribution of these features within a 10ms to 50ms sampling period. The energy recoil pulses are then weighted and summed to obtain the micro-elastic recoil energy product. For the critical state where the thrombus is about to detach, the control unit monitors the micro-elastic recoil energy product E. back The first-order differential gradient change is used to calculate the microelastic recoil energy product E in real time. back rate of change over time When the rate of change over time The preset judgment threshold is reached and the transient active power P inst When reverse energy feedback fluctuations are generated, the control unit determines that the first-order differential gradient satisfies the load collapse triggering logic and actively outputs a negative gradient pulse width modulation duty cycle signal that has an inverse compensation relationship with the first-order differential gradient. By setting the output power decrease slope of the controlled execution unit, the pipeline negative pressure potential energy release rate and the load displacement rate calculated based on the first-order differential gradient are kept offset. The attenuation slope of the negative gradient pulse width modulation duty cycle signal and the time change rate maintain a nonlinear mapping relationship, thereby compensating for the control feedback delay caused by the integral response characteristics of the pressure sensor.

[0033] Example 2: In a physical experimental platform simulating the aspiration of dense thrombi within a blood vessel, a closed-loop fluid circuit was constructed using a centrifugal thrombectomy pump, simulated blood vessel tubing, and a thrombus simulator. The physical experimental platform included a permanent magnet synchronous motor drive unit with a rated speed of 15,000 rpm, and a matching current sampling circuit with a bandwidth of 2.2 kHz and a voltage sampling accuracy of 0.5%, used to acquire real-time stator voltage vector U and real-time stator current vector I within the millisecond scale of electromagnetic fluctuations. The viscosity of the simulated blood fluid used in the experiment was 3.92 mPa·s, consistent with the rheological characteristics of human blood. To evaluate the stability of the scheme under electromagnetic interference, Gaussian white noise with a signal-to-noise ratio of 25 dB and power frequency interference harmonics with a frequency of 50 Hz were superimposed in the data acquisition channel. The sampling period T... s The setting depends on the frequency of the microrecoil wave generated by thrombus yielding, which is distributed in the range of 100Hz to 500Hz; the sampling period T s The selection of the sampling period T needs to balance signal acquisition accuracy and computational overhead. When the real-time rotational speed n of the controlled execution unit increases, in order to satisfy the sampling theorem and prevent signal aliasing, the sampling period T... s The corresponding reduction; in this experiment, for the real-time speed n of 12000 rpm, the sampling period T was set. sThe time interval is 0.5 ms to ensure that the temporal resolution for capturing the energy recoil pulse is not less than 2.0 kHz; during the experiment, the sample group and the control group of this invention should have a density of 1.12 g / cm³. 3 1.34 g / cm 3 and 1.51 g / cm 3 Simulated thrombi; when the density is 1.51 g / cm³ 3 When a thrombus becomes lodged at the catheter tip, the system enters a high static negative pressure maintenance state. At this time, the control group uses a regulation method based on pipeline pressure feedback, and its pressure sensor feedback value is -65.23 kPa. When the internal fibers of the thrombus break, the control unit of the present invention generates a negative pulse with an amplitude of 1.24 W based on the transient active power resolved at the coordinate transformation layer. After removing the steady-state power consumption component of 18.52 W, the micro-elastic recoil energy generated by integral calculation is accumulated to 4.31 mJ.

[0034] Data shows that as the simulated thrombus density increased from 1.12 g / cm³, 3 Increased to 1.51 g / cm³ 3 The microelastic recoil energy product E captured by the sample group of this invention back The peak value increased linearly from 1.83 mJ to 4.31 mJ, proving a deterministic correlation between this physical index and the potential energy accumulation intensity on the load side. At the moment of complete thrombus detachment, the control group experienced a 45.2 ms response hysteresis in the pressure sensor, resulting in an instantaneous fluctuation of 2210 rpm in the motor speed and a negative pressure surge peak in the pipeline reaching 88.54 kPa. The sample group of this invention monitored the microelastic recoil energy product E. back rate of change over time Within 5.1 ms after detecting a gradient value of 0.152 J / s, a negative gradient pulse width modulation duty cycle signal is output, suppressing the peak pressure fluctuation in the pipeline to within 72.41 kPa and reducing the speed fluctuation amplitude to 312 rpm. To verify the parameter boundaries, a sampling period T is set. s This serves as a 2.0ms out-of-range control group; experimental results show that when the sampling period T... s Once the upper limit is exceeded, the microelastic recoil energy product E is insufficient because the sampling frequency cannot cover the rising edge characteristics of the energy recoil pulse. back The resolution deviation increased from 3.5% to 28.6%, the system's response time for determining thrombus detachment was delayed to 32.1ms, and the performance indicators were close to those of the control group, failing to effectively suppress negative pressure surges.

[0035] Example 3: In the initial stage of thrombus removal in human blood vessels, the control unit determines the initial technical state of the controlled execution unit and the pipeline system. Under fluid conditions where the thrombectomy pump catheter is not in contact with the thrombus, the control unit instructs the controlled execution unit to maintain a reference speed of 12,000 rpm, simultaneously acquiring the transient active power sequence within a 100 ms time window, and based on the transient active power P... inst The arithmetic mean of the sequence determines the steady-state power consumption reference P. base Simultaneously determine the transient active power P inst Variance of the sequence The threshold Th is determined based on variance. And the preset proportional coefficient k is determined, where the formula is: Where Th is the decision threshold, k is the proportionality coefficient (selected as 3.0 in this embodiment), and σ is the variance. The arithmetic square root of the value is used to filter background noise from power fluctuations generated by simulated blood pulsation. During the operation of the controlled execution unit, the center frequency f0 of the adaptive notch filter shifts in real time with the real-time rotational speed n. The formula for calculating the center frequency f0 is as follows: Where f0 is the center frequency, n is the real-time rotational speed, and Z is the number of magnetic pole pairs of the rotor of the controlled execution unit.

[0036] The control unit corrects the filter's transfer function coefficients based on the center frequency f0, and the real-time transient active power P... inst The system removes mechanical interference at a specific frequency caused by rotor eccentricity. Within a 20ms sampling period, the control unit extracts the instantaneous amplitude of the energy recoil pulse. If the instantaneous amplitude is greater than 2.0 times the variance σ, the judgment weight for the current sampling moment is set to 1.0; otherwise, the judgment weight is set to 0.2. The microelastic recoil energy product E is obtained by weighted summation of the energy recoil pulses at each sampling moment. back When the microelastic recoil energy product E back rate of change over time When the threshold Th is exceeded, the control unit determines the corresponding output power adjustment within 5.0ms; the system uses a linear mapping relationship to determine the attenuation slope S of the negative gradient pulse width modulation duty cycle signal. pwm When the rate of change over time When the speed is in the range of 0.1 J / s to 0.5 J / s, the attenuation slope S pwm The calculation formula is as follows: , among which, S pwm The attenuation slope, The constant is α, where α is the gain coefficient and β is the bias constant; the control unit is based on the attenuation slope S. pwmBy reducing the pulse width modulation duty cycle of the drive motor, the output power drop slope of the controlled execution unit is matched with the potential energy release level of the thrombus disintegration. Power unloading is completed before the pressure sensor detects the pressure step change in the pipeline, limiting the transient overshoot peak of the pipeline negative pressure to a preset safe range and protecting the blood vessel wall from mechanical adsorption damage.

[0037] Example 4: In the application environment of the catheter assembly, the automatic adjustment method of the thrombectomy pump is applied. Before starting the load aspiration task, the control unit initiates the on-site pre-calibration program. It instructs the controlled execution unit to operate in a no-load saline environment and acquires the real-time stator voltage vector U and the real-time stator current vector I. The transient active power of the controlled execution unit under zero load is calculated. The system continuously collects the transient active power for 500 sampling cycles and extracts the amplitude distribution characteristics. The judgment threshold Th is determined based on the standard deviation σ of the amplitude distribution characteristics and the preset proportional coefficient k. The specific calculation formula is as follows: Where Th is the judgment threshold, k is the proportional coefficient, which is 3.0 in this embodiment, and σ is the standard deviation of transient active power. This program realizes the baseline stripping of intrinsic mechanical friction and fluid disturbance of the hardware system, so that the triggering basis of the load judgment condition corresponds to the initial technical state of the physical entity.

[0038] During the commissioning phase after the controlled actuator and catheter tubing were adapted, the system used a parameter calibration procedure to determine the gain coefficient α and bias constant β in the power regulation logic. By introducing standard simulated blood fluids with viscosities of 3.52 mPa•s, 4.04 mPa•s, and 4.51 mPa•s, the time-varying rate of change of the microelastic recoil energy product during load collapse was measured under the condition that the controlled actuator maintained a constant reference rotational speed. The system uses a least squares algorithm to calculate the quantized values ​​of the gain coefficient α and the bias constant β based on the correlation characteristics, and stores the quantized values ​​in the memory address of the control unit to determine the attenuation slope S of the negative gradient pulse width modulation duty cycle signal. pwm At that time, a step load unloading action was introduced into simulated pipelines of different specifications and lengths, and the micro-elastic recoil energy product E was recorded. back rate of change over time The physical transmission delay t from reaching the preset threshold to the pressure sensor sensing the negative pressure overshoot fluctuation delay Select the response time constant of the controlled execution unit's output power decrease and t delay The matched values, acting as the gain coefficient α and bias constant β in a linear mapping relationship, allow the power unloading action to precede the pressure wave propagation process in the fluid medium on the time axis. This enables the controlled execution unit to respond to heterogeneous thrombus impaction conditions based on the rate of change over time. Determine the attenuation slope S of the negative gradient pulse width modulation duty cycle signal.pwm This enables the dynamic response characteristics of each module to achieve a physical closed loop in both spatial and temporal dimensions.

[0039] Example 5: In a deployment scenario adapting to different specifications of catheter assemblies with inner diameters ranging from 1.0mm to 3.0mm, the control unit determines the transfer function coefficients and the judgment threshold Th of the adaptive notch filter. Specifically, the controlled execution unit is connected to the catheter under test and placed in a simulated blood circuit with a hematocrit of 35% to 45%. The control unit instructs the controlled execution unit to run in 1000rpm increments within a speed range of 5000rpm to 15000rpm. At each speed point, the real-time transient active power sequence is acquired and its residual distribution is calculated. By adjusting the quality factor Q of the filter, the attenuation rate of the mechanical harmonic components reaches more than 40dB. The standard deviation of the intrinsic power background noise under different hardware combinations is stored in the non-volatile memory of the control unit, so that the system automatically matches the corresponding reference value according to the inner diameter specification of the catheter called during the task, and removes the power fluctuation noise caused by the difference in pipeline geometric characteristics.

[0040] To verify the stability of the controlled actuator in the face of transient high-viscosity thrombus collapse, a multi-dimensional gradient experiment was conducted to determine the gain coefficient α and bias constant β in the linear mapping relationship. The control unit acquired the negative energy feedback intensity generated by the controlled load at the moment of yielding, and correlated the negative energy feedback intensity with the rotational inertia J of the drive system to determine the value of the gain coefficient α to balance the output power decrease rate and the speed fluctuation overshoot. The time change rate of the micro-elastic recoil energy product was also considered. When the load collapse trigger logic is met, the control unit verifies the real-time transient active power P. inst Does a pulse fluctuation greater than 1.5 standard deviation occur synchronously, and after successful logic matching, does it follow the calculated attenuation slope S? pwm By reducing the pulse width modulation duty cycle, the transient fluctuation of pipeline negative pressure is maintained within 10.0% of the peak static pressure under different viscosity loads, so that the hardware response characteristics are physically aligned with the biological load state. When extracting the damping oscillation damping rate ζ, a static friction torque diagnostic pulse sequence with a frequency of 1kHz to 5kHz and a limited amplitude is output to the controlled execution unit. The residual induced current data of the stator winding after the pulse turn-off moment is collected. The corresponding damping oscillation damping rate ζ is calculated by fitting the induced current attenuation envelope characteristics using an exponential decay function. When the catheter tip adheres to the blood vessel wall, the change in the acoustic impedance at the end of the system causes a shift in the energy dissipation characteristics of the impeller and rotor micro-vibration. The damping rate value shows a quantitative increase of more than 1.5 times compared to the fluid no-load state, and the boundary adhesion state is determined.

[0041] 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.

[0042] 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 for automatically adjusting the operation of a bail pump, comprising: Includes the following steps: Step S1: When the control unit determines that the system is in a load extraction task, it synchronously acquires the real-time stator voltage vector U and the real-time stator current vector I of the controlled execution unit in the stationary coordinate system. Step S2: Perform scalar product operation of real-time stator voltage vector U and real-time stator current vector I at the coordinate transformation layer of the control system to analyze the transient active power that characterizes the intensity of bidirectional electromechanical energy interaction between the controlled execution unit and the external controlled load in real time. Step S3: Establish a frequency-selective filtering mechanism based on dynamic speed feedback, use a high-pass filtering algorithm to remove the steady-state power consumption component in the transient active power that corresponds to overcoming the macroscopic load resistance of the fluid, and unidirectionally extract the energy recoil pulse that characterizes the moment of fatigue yielding of the internal structure of the controlled load from the nonlinear mechanical resonance noise of the transmission system. Step S4: Perform discretization integration on the energy recoil pulse within a preset sampling time window to generate a microelastic recoil energy product for quantifying the degree of fiber breakage and potential energy release level inside the controlled load. Step S5: Continuously monitor the change of the first-order differential gradient of the micro-elastic recoil energy product, and at the instant when the first-order differential gradient satisfies the preset load collapse triggering logic, actively output a negative gradient pulse width modulation duty cycle signal that has an inverse compensation relationship with the first-order differential gradient. By setting the output power drop slope of the controlled execution unit, the pipeline negative pressure potential energy release rate and the load displacement rate calculated based on the first-order differential gradient are physically and logically offset in real time.

2. The automatic adjustment method for the operation of a thrombectomy pump according to claim 1, characterized in that, Step S5, which involves actively outputting the negative gradient pulse width modulation duty cycle signal, includes: real-time calculation of the microelastic recoil energy product E. back rate of change over time ; rate of change over time The preset judgment threshold is reached and the transient active power P inst When reverse energy feedback fluctuations are generated, the power drop envelope of the controlled execution unit is set through a control algorithm, so that the attenuation slope of the negative gradient pulse width modulation duty cycle signal is related to the rate of change of time. Maintain nonlinear mapping correlations to counteract the hydraulic hysteresis of the system at the earliest stage of failure of the preceding physical interface.

3. The automatic adjustment method for the operation of a thrombectomy pump according to claim 1, characterized in that, In step S2, the real-time transient active power P inst The calculation follows these rules: , where P inst U represents the real-time transient active power, U represents the real-time stator voltage vector, and I represents the real-time stator current vector.

4. The automatic adjustment method for the operation of a thrombectomy pump according to claim 1, characterized in that, The method for eliminating the steady power consumption component in step S3 comprises: acquiring the real-time rotating speed n of the controlled execution unit; constructing an adaptive notch filter with the center frequency dynamically migrating with the real-time rotating speed n; inputting the real-time transient active power P inst The adaptive notch filter is inputted to filter out the mechanical harmonics related to the real-time rotating speed n.

5. The automatic adjustment method for the operation of a thrombectomy pump according to claim 1, characterized in that, After executing step S5, the following steps are also included: Step S6, controlling the controlled execution unit to enter a resting state with a current of 0A and opening a monitoring window period; Step S7, during the monitoring window period, outputting a diagnostic pulse sequence to the controlled execution unit where the electromagnetic torque is less than the static friction torque of the system.

6. The automatic adjustment method for the operation of a thrombectomy pump according to claim 5, characterized in that, Step S7 is followed by: monitoring the damped oscillation damping rate ζ generated by the electromagnetic winding of the controlled execution unit under the excitation of the diagnostic pulse sequence; comparing the damped oscillation damping rate ζ with the preset load decoupling benchmark to determine the boundary attachment state between the end of the controlled execution unit and the controlled load.

7. The automatic adjustment method for the operation of a thrombectomy pump according to claim 6, characterized in that, It also includes the following steps: Step S8: Determine whether the damping rate ζ of the damped oscillation reaches the preset unlocking threshold. Step S9: If the damping rate of the decaying oscillation ζ reaches the unlocking threshold... If the boundary attachment state is released, the basic driving frequency will be restored.

8. The automatic adjustment method for the operation of a thrombectomy pump according to claim 1, characterized in that, The method for generating the microelastic recoil energy product in step S4 includes: extracting the envelope features of the energy recoil pulse; determining the judgment weight based on the cumulative distribution of the envelope features within a sampling period of 10ms to 50ms, and then performing a weighted summation of the energy recoil pulse.

9. The automatic adjustment method for the operation of a thrombectomy pump according to claim 1, characterized in that, The step S5 involves setting the output power reduction slope of the controlled execution unit, which includes: matching the preset pressure unloading curve library according to the evolution slope of the micro-elastic recoil energy product; controlling the operating frequency of the controlled execution unit to decay along the target curve matched in the pressure unloading curve library, so that the negative pressure release speed in the pipeline is synchronized with the spatial increment generated by the controlled load.

10. The automatic adjustment method for the operation of a thrombectomy pump according to claim 7, characterized in that, The method for restoring the basic drive frequency in step S9 includes: obtaining the initial load impedance at the instant the boundary attachment state is released; calculating the initial control gain of the feedback loop based on the initial load impedance; and gradually adjusting the operating frequency to the basic drive frequency according to a preset time constant.

Images