An Adaptive Control Method for Interface Properties Based on Phase Change Heat Transfer Mechanism

Abstract

This invention relates to the field of industrial automatic control system device manufacturing technology, and discloses an interface property adaptive control method based on phase change heat transfer mechanism. The method includes: synchronously sampling the impedance parameters and bus current parameters of the heat transfer interface and storing them in a register; calculating the liquid phase ratio characteristic value and feedforward flow compensation parameters; dividing the system into intervals using a state machine; matching a stepped compensation gradient when the upper limit is reached and the feedforward exceeds the limit to generate a voltage command and drive the stress adjustment unit to adjust the mechanical output; and superimposing a depolarization pulse current to neutralize residual charge when the temperature change rate exceeds a threshold. This invention combines impedance characteristics and feedforward parameters to penetrate the latent heat blind zone, eliminate the phase lag effect caused by physical time delay, and achieve synchronization between command issuance and property characteristics, effectively improving the system's adjustment accuracy and stability.

Description

Technical Field

[0001] This invention belongs to the field of industrial automatic control system device manufacturing technology, and particularly relates to an interface property adaptive control method based on phase change heat transfer mechanism. Background Technology

[0002] Currently, with the increasing manufacturing density of power electronic switching devices and the integration density of control components, active heat dissipation control for high heat flux density components is a core component to ensure the long-term stable operation of the control and regulation system. Traditional control loops use a fixed contact thermal resistance model, adjusting the fan speed or coolant flow rate by collecting feedback signals from the external cold end temperature to maintain the junction temperature within a safe range. As a new type of heat dissipation carrier, the heat transfer medium absorbs the latent heat generated by transient high power loads. The solid-liquid two-phase properties at the heat transfer interface exhibit changing characteristics with the evolution of the phase change temperature plateau. The acoustic resistance change characteristics inside the medium layer change before the external temperature field, directly corresponding to the real-time evolution of the two-phase property distribution.

[0003] Under conditions of rapid, high-power load changes, the latent heat absorption process of the heat transfer medium during solid-liquid melting leads to a stable resting temperature plateau at the heat dissipation interface. This creates a blind spot in the closed-loop feedback loop, preventing it from tracking the transient evolution of the interface's contact properties until the latent heat is exhausted and the medium is completely liquefied, causing a step-like abrupt change in the interface's equivalent thermal resistance. Due to the inherent physical phase deviation between the heat transfer time constant and the control action, the traditional cold-end feedback control system exhibits a significant hysteresis in its adjustment signals. Conventional improvements such as increasing control gain or widening the cooling channels are limited by the control cabinet's space and hardware modification costs, and are prone to triggering high-frequency adjustments at points of abrupt change in heat transfer properties. Overshooting is a problem. For example, Chinese invention patent CN117352469A discloses a high overload capacity press-fit IGBT power module based on phase change material. It integrates phase change materials with different melting points by opening a central filling groove and a side filling groove in the phase change copper layer to reduce thermal stress. However, this solution is a passive heat dissipation mechanism that relies on fixed static hardware physical boundaries. When facing high-frequency and heavy load conditions, the statically configured passive phase change material lacks active adjustment means. The transient temperature difference in the internal heat transfer process and the low thermal conductivity of the material lead to response lag. Moreover, the preset fixed melting point boundary line cannot adapt to the wide-frequency variable load requirements. Under extreme conditions, it is easy to trigger a sudden jump in heat transfer stress and film boiling due to latent heat capacity overload saturation. The control output exhibits continuous self-excited oscillation at the nodes of severe alternating load disturbance.

[0004] Therefore, traditional control paths cannot keep up with the nonlinear transient changes in phase change properties, resulting in control time lag and command overshoot. The technical problem to be solved by this invention is how to construct a feedforward closed-loop control topology based on the nonlinear lag characteristics of the property evolution at the heat dissipation interface, adaptively control the transient heat transfer stress state on both sides of the heat dissipation interface, and achieve online control of heat transfer properties. Summary of the Invention

[0005] This invention aims to solve the problems of control lag and command overshoot caused by the nonlinear evolution of heat transfer properties in existing control paths.

[0006] In this technical solution, an adaptive control method for interface properties based on a phase change heat transfer mechanism includes the following steps:

[0007] Step S101: Synchronously collect the acoustic characteristic impedance parameters and main bus current parameters of the heat transfer interface, and store the acoustic characteristic impedance parameters and main bus current parameters as state feature vectors into the timing data register.

[0008] Step S102: Retrieve the state feature vector in the timing data register, calculate and output the liquid phase ratio feature value that characterizes the melting ratio of the heat transfer medium based on the monotonic correspondence between the acoustic characteristic impedance parameter and the pre-stored pure solid reference impedance parameter and pure liquid reference impedance parameter, and perform timing difference calculation on the main bus current parameter to generate feedforward flow compensation parameter.

[0009] Step S103: The range of liquid phase ratio characteristic value is divided into three non-overlapping control intervals by using a discrete state machine. When the liquid phase ratio characteristic value is within the phase change control interval, the nominal reference control voltage is maintained. When the liquid phase ratio characteristic value reaches the upper limit critical value of the phase change control interval and the rate of change of the feedforward flow compensation parameter exceeds the preset sudden change threshold, a step-by-step compensation gradient is matched according to the absolute span of the liquid phase ratio characteristic value deviating from the upper limit critical value, a differential control voltage command is generated, and the interface stress adjustment unit is driven to adjust the transient mechanical output of the heat transfer interface.

[0010] Step S104: Monitor the temperature change rate parameter of the heat transfer interface. When the temperature change rate parameter exceeds the safe temperature change threshold, superimpose a depolarization reverse electric field pulse current on the differential control voltage command to neutralize the physical residual polarization charge inside the interface stress adjustment unit.

[0011] Preferably, step S101 includes the following sub-steps: step S1011, synchronously sampling the acoustic characteristic impedance parameters of the heat transfer interface and the main bus current parameters using a sampling period of 20μs; step S1012, combining the synchronously sampled acoustic characteristic impedance parameters and main bus current parameters into a multi-dimensional flow data stream, and continuously transferring them to the timing data register according to the acquisition time.

[0012] Preferably, step S103 includes the following sub-steps: Step S1031, in the discrete state machine, the value range of the liquid phase ratio characteristic value is subdivided into three control intervals. When the liquid phase ratio characteristic value is in the range of 0.35 to 0.55, it is taken as the phase change control interval and the current nominal reference of the control voltage is maintained; Step S1032, when the liquid phase ratio characteristic value reaches the upper limit critical value of 0.55 and the rate of change of the feedforward flow compensation parameter exceeds the preset mutation threshold, the absolute span of the liquid phase ratio characteristic value deviating from 0.55 is determined; Step S1033, according to the absolute span, a stepped step compensation gradient is dynamically matched, and the stepped step compensation gradient is accumulated on the nominal reference of the control voltage to update the differential control voltage command.

[0013] Preferably, in step S1033, generating a differential control voltage command and driving the interface stress adjustment unit to adjust the transient mechanical output of the heat transfer interface includes: the control variation judgment module outputs an updated differential control voltage command to the interface stress adjustment unit, driving the interface stress adjustment unit to change the output displacement, and adjusting the micro-contact resistance in real time by adjusting the degree of micro-contact compression on the heat transfer interface side.

[0014] Preferably, step S104 includes the following sub-steps: Step S1041, the surface physical temperature of the heat transfer interface is monitored in real time by the temperature parameter sampling unit, and the first derivative of the surface physical temperature with respect to time is calculated to generate a temperature change rate parameter; Step S1042, it is determined whether the temperature change rate parameter exceeds the safe temperature change threshold of 5°C / s, and when the temperature change rate parameter continues to exceed 5°C / s, the depolarization control law is activated; Step S1043, a depolarization reverse electric field pulse current with an amplitude of 15% of the nominal reference of the control voltage is superimposed on the differential control voltage command to eliminate the charge accumulation generated inside the interface stress adjustment unit.

[0015] Preferably, the adaptive control method further includes the following automatic calibration parallel steps: Step S105, continuously compare the main bus current parameter in the timing data register with the preset no-load current threshold, and when the main bus current parameter is lower than the preset no-load current threshold for 5 seconds and the control loop is in a resting ready state, issue a reference calibration test command; Step S106, the control interface stress adjustment unit applies a standard step pressure to the heat transfer interface and reads the current acoustic characteristic impedance parameter as a transient calibration reference value; Step S107, replace the original stored pure solid-state reference impedance parameter with the transient calibration reference value.

[0016] Preferably, the adaptive control method further includes the following parallel safety protection steps: Step S108, performing an over-limit comparison on the liquid phase ratio characteristic value to determine whether the liquid phase ratio characteristic value exceeds the safe phase change melting threshold of 0.85; Step S109, when it is determined that the liquid phase ratio characteristic value exceeds 0.85 for three consecutive control cycles, and the feedforward flow compensation parameter is higher than the critical over-limit parameter, i.e., the preset steady-state feedforward upper limit voltage value of 5.5V stored in the read-only memory of the central control unit to characterize the power electronic switching device at the extreme heavy load safety edge, triggering the overheating melting mechanism; Step S110, issuing a safety cut-off command to forcibly shut down the main bus power supply circuit of the system.

[0017] Preferably, before calculating the timing difference of the main bus current parameters in step S102, the high-frequency spike noise in the sampled main bus current parameters and acoustic characteristic impedance parameters is filtered out using a low-pass filter unit to smooth out the transient differential oscillation of the differential control voltage command.

[0018] Preferably, the interface property adaptive control method based on the phase change heat transfer mechanism further includes, at the boundary of each control cycle, re-storing the updated acoustic characteristic impedance parameters and main bus current parameters as initial feedback variables into the timing data register, and performing long-cycle closed-loop adaptive control of heat transfer properties when transient heat flow transitions occur at the heat transfer interface through the interactive iteration of acoustic characteristic impedance parameters and main bus current parameters.

[0019] Compared with existing technologies, the interface property adaptive control method based on phase change heat transfer mechanism of the present invention has the following advantages:

[0020] 1. In the adaptive control of interface properties in phase change heat transfer mechanism, by coordinating the acoustic characteristic impedance parameters of the heat dissipation interface and the bus current parameters characterizing the load change trend, the control system changes the hysteresis loop that relies on cold end temperature feedback, so that the adjustment start point is moved forward to the two-phase evolution stage of the heat transfer medium; when the component is overloaded, the feedforward quantity extracted by the time-series differential operation and the decoupled liquid phase ratio characteristic value are correlated in time, so that the error input of the controller crosses the temperature stagnation zone caused by the latent heat of phase change, and eliminates the command phase lag caused by the melting plateau period of the medium; through the feedback control of multi-dimensional state vector flow, the high-frequency oscillation of the control voltage command is avoided when the closed-loop system generates a step-like change in the interface thermal resistance.

[0021] 2. The control arbitration module introduces an event-driven discrete state machine decision law to divide the internal phase parameters into non-overlapping control intervals. When the liquid phase ratio characteristic value is in the phase change control region, the system maintains the current control voltage reference. When the parameter reaches the upper limit boundary and the feedforward variable shows a positive transition, the feedforward correction operator dynamically matches the step-wise compensation gradient based on the absolute span of the parameter deviation from the maintenance window. This interval division and the synergy of the adaptive mapping matrix enable the control adjustment unit to adjust the force field output of the piezoelectric drive in advance according to the evolution trend of the two phases of the medium. By adjusting the contact pressure, the micro-squeezing state of the interface is changed, realizing the adaptive adjustment closed loop of heat transfer properties.

[0022] 3. To address the risks of physical charge retention in the actuator under high-frequency dynamic alternating loads and degradation of the heat transfer medium after long-term cycling, a secondary deviation compensation circuit and an online calibration procedure are added to the control loop. When the timing data register detects that the continuous temperature change rate of the heat transfer interface exceeds the preset a priori boundary, the system automatically triggers the depolarization control program, superimposing a transient depolarization pulse on the differential control voltage command to eliminate the response hysteresis caused by residual polarization charge inside the drive component. At the same time, when the bus current is continuously below the no-load threshold in a low-load resting state, the control and regulation unit automatically issues a reference test command to recalibrate the solid-state reference impedance parameters by applying a standard step pressure. Attached Figure Description

[0023] Figure 1 This is a flowchart of the steps of the interface property adaptive control method of the present invention;

[0024] Figure 2 This is the state transition diagram of the adaptive control of interface properties according to the present invention. Detailed Implementation

[0025] 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.

[0026] An adaptive control method for interface properties based on a phase change heat transfer mechanism includes the following steps:

[0027] Step S101: Synchronously collect the acoustic characteristic impedance parameters and main bus current parameters of the heat transfer interface, and store the acoustic characteristic impedance parameters and main bus current parameters as state feature vectors into the timing data register.

[0028] Step S102: Retrieve the state feature vector in the timing data register, calculate and output the liquid phase ratio feature value that characterizes the melting ratio of the heat transfer medium based on the monotonic correspondence between the acoustic characteristic impedance parameter and the pre-stored pure solid reference impedance parameter and pure liquid reference impedance parameter, and perform timing difference calculation on the main bus current parameter to generate feedforward flow compensation parameter.

[0029] Step S103: The range of liquid phase ratio characteristic value is divided into three non-overlapping control intervals by using a discrete state machine. When the liquid phase ratio characteristic value is within the phase change control interval, the nominal reference control voltage is maintained. When the liquid phase ratio characteristic value reaches the upper limit critical value of the phase change control interval and the rate of change of the feedforward flow compensation parameter exceeds the preset sudden change threshold, a step-by-step compensation gradient is matched according to the absolute span of the liquid phase ratio characteristic value deviating from the upper limit critical value, a differential control voltage command is generated, and the interface stress adjustment unit is driven to adjust the transient mechanical output of the heat transfer interface.

[0030] Step S104: Monitor the temperature change rate parameter of the heat transfer interface. When the temperature change rate parameter exceeds the safe temperature change threshold, superimpose a depolarization reverse electric field pulse current on the differential control voltage command to neutralize the physical residual polarization charge inside the interface stress adjustment unit.

[0031] Preferably, step S101 includes the following sub-steps: step S1011, synchronously sampling the acoustic characteristic impedance parameters of the heat transfer interface and the main bus current parameters using a sampling period of 20μs; step S1012, combining the synchronously sampled acoustic characteristic impedance parameters and main bus current parameters into a multi-dimensional flow data stream, and continuously transferring them to the timing data register according to the acquisition time.

[0032] Preferably, step S103 includes the following sub-steps: Step S1031, in the discrete state machine, the value range of the liquid phase ratio characteristic value is subdivided into three control intervals. When the liquid phase ratio characteristic value is in the range of 0.35 to 0.55, it is taken as the phase change control interval and the current nominal reference of the control voltage is maintained; Step S1032, when the liquid phase ratio characteristic value reaches the upper limit critical value of 0.55 and the rate of change of the feedforward flow compensation parameter exceeds the preset mutation threshold, the absolute span of the liquid phase ratio characteristic value deviating from 0.55 is determined; Step S1033, according to the absolute span, a stepped step compensation gradient is dynamically matched, and the stepped step compensation gradient is accumulated on the nominal reference of the control voltage to update the differential control voltage command.

[0033] Preferably, in step S1033, generating a differential control voltage command and driving the interface stress adjustment unit to adjust the transient mechanical output of the heat transfer interface includes: the control variation judgment module outputs an updated differential control voltage command to the interface stress adjustment unit, driving the interface stress adjustment unit to change the output displacement, and adjusting the micro-contact resistance in real time by adjusting the degree of micro-contact compression on the heat transfer interface side.

[0034] Preferably, step S104 includes the following sub-steps: Step S1041, the surface physical temperature of the heat transfer interface is monitored in real time by the temperature parameter sampling unit, and the first derivative of the surface physical temperature with respect to time is calculated to generate a temperature change rate parameter; Step S1042, it is determined whether the temperature change rate parameter exceeds the safe temperature change threshold of 5°C / s, and when the temperature change rate parameter continues to exceed 5°C / s, the depolarization control law is activated; Step S1043, a depolarization reverse electric field pulse current with an amplitude of 15% of the nominal reference of the control voltage is superimposed on the differential control voltage command to eliminate the charge accumulation generated inside the interface stress adjustment unit.

[0035] Preferably, the adaptive control method further includes the following automatic calibration parallel steps: Step S105, continuously compare the main bus current parameter in the timing data register with the preset no-load current threshold, and when the main bus current parameter is lower than the preset no-load current threshold for 5 seconds and the control loop is in a resting ready state, issue a reference calibration test command; Step S106, the control interface stress adjustment unit applies a standard step pressure to the heat transfer interface and reads the current acoustic characteristic impedance parameter as a transient calibration reference value; Step S107, replace the original stored pure solid-state reference impedance parameter with the transient calibration reference value.

[0036] Preferably, the adaptive control method further includes the following parallel safety protection steps: Step S108, performing an over-limit comparison on the liquid phase ratio characteristic value to determine whether the liquid phase ratio characteristic value exceeds the safe phase change melting threshold of 0.85; Step S109, when it is determined that the liquid phase ratio characteristic value exceeds 0.85 for three consecutive control cycles, and the feedforward flow compensation parameter is higher than the critical over-limit parameter, i.e., the preset steady-state feedforward upper limit voltage value of 5.5V stored in the read-only memory of the central control unit to characterize the power electronic switching device at the extreme heavy load safety edge, triggering the overheating melting mechanism; Step S110, issuing a safety cut-off command to forcibly shut down the main bus power supply circuit of the system.

[0037] Preferably, before calculating the timing difference of the main bus current parameters in step S102, the high-frequency spike noise in the sampled main bus current parameters and acoustic characteristic impedance parameters is filtered out using a low-pass filter unit to smooth out the transient differential oscillation of the differential control voltage command.

[0038] Preferably, the interface property adaptive control method based on the phase change heat transfer mechanism further includes, at the boundary of each control cycle, re-storing the updated acoustic characteristic impedance parameters and main bus current parameters as initial feedback variables into the timing data register, and performing long-cycle closed-loop adaptive control of heat transfer properties when transient heat flow transitions occur at the heat transfer interface through the interactive iteration of acoustic characteristic impedance parameters and main bus current parameters.

[0039] Example 1: In the closed-loop regulation of an industrial automatic control system device, when the local transient heat load on the high-power, high-heat-flux-density control component side exhibits a nonlinear step increase due to the high-frequency heavy-load switching of power electronic switching devices, the heat transfer medium inside the heat dissipation interface undergoes a phase change from solid to liquid. This phase change process has a latent heat absorption stage, causing the temperature rise rate of the heat dissipation interface to present a stable resting plateau during this period. This causes the traditional control and regulation system that relies on the negative feedback of the external cold end temperature to generate a control time lag dead zone. Consequently, the temperature error input signal of the controller cannot reflect the nonlinear hysteresis characteristics of the internal contact properties of the heat transfer interface as the phase change occurs in real time. This results in a phase lag and time constant mismatch between the control command and the actual property response of the interface. Until the latent heat is exhausted and the heat transfer medium is completely liquefied, causing a step-like change in the equivalent thermal resistance of the heat dissipation interface, the control and regulation system cannot respond to the sudden heat flow in advance, resulting in overshoot and continuous high-frequency self-excited oscillation in the output command. As a result, thermal stress concentration and temperature exceedance occur inside the component, ultimately triggering thermal breakdown protection and unplanned shutdown failure.

[0040] The status acquisition module synchronously and discretely samples the acoustic characteristic impedance parameters of the heat transfer interface. and main bus current parameters The acquired acoustic characteristic impedance parameters and main bus current parameters The state feature vector is stored in the timing data register in real time. The central control unit retrieves the state feature vector from the timing data register and, based on the acoustic characteristic impedance parameters... Relative to pre-stored pure solid-state reference impedance parameters and pure liquid reference impedance parameters The monotonic correspondence is established by retrieving the acoustic characteristic impedance parameter value of the current cycle from the central control unit and performing a two-step subtraction and one-step division data logic operation. First, the acoustic characteristic impedance parameter value of the current cycle is subtracted from the pure solid-state reference impedance parameter value to obtain the solid-phase impedance difference component. Then, the pure liquid-state reference impedance parameter value is subtracted from the pure solid-state reference impedance parameter value to obtain the total phase change impedance span component. Finally, the solid-phase impedance difference component is used as the numerator and the total phase change impedance span component is used as the denominator for division. This allows for precise quantitative output of the volume melting ratio of the medium at the microscopic level as it transforms from solid to liquid within a discrete value range of 0 to 1. This eliminates nonlinear mapping ambiguity and establishes an absolute monotonically decreasing numerical correspondence chain to calculate and output the liquid phase ratio characteristic value representing the melting ratio of the heat transfer medium. The characteristic value of the liquid phase ratio The linear calculation formula is as follows: ,in, This is a dimensionless characteristic value of the liquid phase fraction. The characteristic impedance reference value for a pure solid phase change medium. The acoustic characteristic impedance parameter of the heat transfer interface. The characteristic impedance reference value is for a pure liquid phase change medium, and the central control unit controls the main bus current parameters. Calculate the time-series difference to generate feedforward flow compensation parameters characterizing the load abrupt change trend. .

[0041] The control arbitration module uses a discrete state machine to process the liquid phase ratio characteristic value. The value range is divided into three non-overlapping control intervals, at the liquid phase fraction characteristic value. When the core sustaining range is between 0.35 and 0.55, the control and regulation unit maintains the current nominal control voltage reference, and when the liquid phase ratio characteristic value... Crossing the critical upper limit threshold of 0.55 and the feedforward flow compensation parameter When the rate of change exceeds the preset mutation threshold, the feedforward flow compensation parameter With liquid phase ratio characteristic value Timing-based coordinated constraint control is performed, with the control arbitration module determining the liquid phase ratio characteristic value. The absolute span of the deviation from the core maintenance range is dynamically matched with a preset stepped compensation gradient, and the stepped compensation gradient is accumulated on the nominal reference of the control voltage to update the differential control voltage command. The control command output unit outputs the updated differential control voltage command. The interface stress adjustment unit is driven to change the output displacement to adjust the transient mechanical output of the heat transfer interface. By changing the degree of local contact compression on the heat transfer interface side, the local contact resistance is dynamically adjusted, and the melting point equilibrium line of the phase change medium is adaptively intervened, so as to complete the advance adjustment of the control path before the temperature of the heat dissipation interface changes abruptly.

[0042] Under this control regulation, which utilizes state machine interval determination and feedforward flow compensation parameters in a coordinated manner, the sensing end of the closed-loop regulation system overcomes the inherent heat transfer resistance of the heat dissipation interface. The control starting point of the control system is advanced from the lagging temperature transition of the heat conduction surface to the evolution stage of the solid-liquid two-phase morphology of the medium. This eliminates the phase lag effect caused by the internal heat transfer delay on the closed-loop regulation loop. The response time delay of the control loop to transient heat load is 12ms, which stabilizes the solid-liquid mixed state of the phase change heat transfer medium within the maintenance range of the phase change core with a large equivalent specific heat capacity. This avoids the high-frequency oscillation and overshoot phenomenon of commands caused by the step-like abrupt change in physical properties of traditional control systems. The overshoot amplitude of the control components inside the high-power whole machine control cabinet is reduced accordingly, and the frequency of component structure manufacturing failures caused by alternating thermal stress is also reduced. This achieves the closed-loop regulation and long-term operation stability of the industrial automatic control system under transient thermal shock conditions.

[0043] Example 2: In the thermal control test of industrial automatic control system device manufacturing, for the transient thermal shock condition caused by high-frequency heavy-load switching of high-power power electronic switching devices, a thermal control test bench containing a high-power semiconductor switching array is used to provide transient heat transfer with a power density range of 100W / cm² to 800W / cm², in order to reproduce the nonlinear step growth characteristics of transient thermal load inside the component; the functional specifications required for the data acquisition physical path of the state acquisition module to meet the following performance boundaries, wherein the acoustic characteristic impedance parameters used to measure both sides of the heat transfer interface are... The acoustic sensor's measurement range covers 1.0 × 10⁻⁶. 6 kilograms per square meter per second up to 5.0 × 10⁻⁶ 6 The physical resolution is 1000 kg / m² / s, and the dynamic response bandwidth is greater than or equal to 100 kHz. When determining the sampling period of the state acquisition module, this sampling period parameter is established as the discrete sampling interval. The main technical factors affecting the value of this parameter include the rate of change of transient overload current and the speed of sound wave propagation within the heat transfer medium layer. The technical trade-off lies in achieving a balance between capturing the real-time changes in transient heat transfer morphology at the interface and reducing the data processing load on the central control unit. The function decision rule it follows stipulates that when the electrical switching frequency of the heating element reaches 50kHz, in order to prevent signal aliasing of thermal dynamic characteristics and accurately capture the starting point of the phase change latent heat control blind zone according to the Nyquist sampling criterion, the sampling period... The value tends towards the lower limit of its range, thus determining a specific example value of 20 μs under this particular typical operating condition.

[0044] In the specific data acquisition physical path, the acoustic sensor includes a piezoelectric ultrasonic transducer array fixed at the bottom of the heat transfer interface. This array consists of two sets of oppositely arranged transmitting and receiving chips. Under the control of the central control unit, the active acoustic signal generator continuously applies a 5 MHz high-frequency pulse excitation electrical signal to the transmitting chip. The transmitting chip is excited to generate a high-frequency ultrasonic stress wave that penetrates vertically through the heat transfer medium layer. When the ultrasonic stress wave reaches the solid-liquid phase change interface, due to the phonon scattering and acoustic impedance difference between the solid and liquid media, the sound wave undergoes partial reflection and transmission at the interface. The transmitted sound wave signal is captured by the receiving chip on the opposite side. The receiving chip converts the received sound pressure signal into a weak high-frequency electrical signal and sends it to the signal modulation and demodulation control logic of the state acquisition module. In the front-end circuit, it is amplified by 20 dB by a low-noise amplifier, and the amplitude is extracted and digitally sampled by an envelope detector and a high-speed analog-to-digital converter. Finally, the sound wave reflection coefficient and transmission coefficient are calculated by a high-speed integrator, thereby decoupling in real time and outputting the acoustic characteristic impedance parameters that characterize the current physical resistance change characteristics of the interface.

[0045] When the piezoelectric ultrasonic transducer array connected to the status acquisition module receives a high-frequency ultrasonic stress wave, it outputs a raw voltage signal. The central control unit extracts the maximum amplitude of the envelope of the raw voltage signal as the sound pressure characterization quantity. Based on the acoustic quantitative relationship that the reflection coefficient is equal to the difference between the interface impedance and the intrinsic impedance of the transducer divided by the sum of the two, the voltage variation value of the sound pressure characterization quantity is converted into the acoustic impedance variation value. When the heat transfer interface is in static assembly and the medium is in a pure solid state, the first calibration voltage value is latched. When the temperature control source is controlled to completely liquefy the medium, the second calibration voltage value is latched. The central control unit generates the absolute physical quantity data of the current acoustic characteristic impedance parameter based on the linear extrapolation relationship between the first calibration voltage value and the second calibration voltage value.

[0046] The status acquisition module synchronously and discretely samples the acoustic characteristic impedance parameters of the heat transfer interface. and main bus current parameters The acquired acoustic characteristic impedance parameters and main bus current parameters The state feature vector is stored in the timing data register in real time. Then, the central control unit retrieves the state feature vector from the timing data register and, based on the acoustic characteristic impedance parameters... Relative to pre-stored pure solid-state reference impedance parameters and pure liquid reference impedance parameters The monotonic correspondence is used to calculate and output the liquid phase ratio characteristic value representing the melting ratio of the heat transfer medium. The characteristic value of the liquid phase ratio The linear calculation formula is as follows: ,in, The dimensionless liquid phase fraction characteristic value, subscript Indicates liquid state. The characteristic impedance reference value for a pure solid phase change medium, subscript Indicates solid state. The acoustic characteristic impedance parameter of the heat transfer interface, subscript Interface, The characteristic impedance reference value is for a pure liquid phase change medium, and the central control unit controls the main bus current parameters. Calculate the time-series difference to generate feedforward flow compensation parameters characterizing the load abrupt change trend. It is 4.5V / ms.

[0047] Under the sample group operating conditions corresponding to the lower limit critical value, due to the coexistence of external interference, the original acoustic characteristic impedance parameters acquired by the state acquisition module are... At 3.15×10 6 kilograms per square meter per second up to 3.22 × 10⁻⁶ 6 Random fluctuations occur between kilograms per square meter per second, and the original main bus current parameters... It exhibits a non-ideal step transition morphology from 10A to 150A; the central control unit retrieves the aforementioned noisy raw characteristic data stream, and its internal synchronous differential calibration loop suppresses power frequency interference harmonics, decoupling the output intermediate characteristic liquid phase ratio eigenvalue. The acoustic characteristic impedance parameter remained stable at 0.35 under the corresponding normal operating conditions of the sample group, after processing by the calibration circuit. The central control unit drives the output of the median liquid phase ratio characteristic value. The value is 0.45, which corresponds to the feedforward flow compensation parameter. Maintained at 0.2V / ms; under the sample conditions corresponding to the upper limit critical value, the strong heat load input drives the heat transfer medium close to the melting boundary, and the calibrated acoustic characteristic impedance parameters... Make the calculated intermediate liquid phase ratio characteristic value Reaching the absolute upper limit of 0.55, the accompanying current surge leads to a change in the calculated feedforward current compensation parameter. A high rate of change transition occurs, reaching 8.2V / ms.

[0048] The control arbitration module receives the intermediate calculated characteristic parameters of the above discrete sample groups and calls the discrete state machine decision rules. When the liquid phase ratio characteristic value... When the normal median value is 0.45, the control command output unit maintains a stable nominal reference output of 5.0V for the control voltage. At this time, the heat transfer medium is in a relative equilibrium state with two phases coexisting. When the system encounters the sample condition of the upper limit critical value, the liquid phase ratio characteristic value... When the critical upper limit threshold of 0.55 is reached, the feedforward flow compensation parameter... With liquid phase ratio characteristic value The trigger time-dimensional collaborative constraint interlock, the control arbitration module matches the output amplitude of a stepped compensation gradient of 2.5V in real time according to the span of its deviation window, and accumulates the stepped compensation gradient on the nominal reference of the control voltage to update the differential control voltage command. The control command output unit outputs the updated differential control voltage command. The interface stress adjustment unit is driven to change the output displacement to adjust the transient mechanical output of the heat transfer interface. By changing the degree of local contact compression on the heat transfer interface side, the local contact resistance is dynamically adjusted, and the melting point equilibrium line of the phase change medium is adaptively intervened, so as to complete the advance adjustment of the control path before the temperature of the heat dissipation interface changes abruptly.

[0049] To verify the rationality of the numerical envelope boundary optimization and characterize the performance inflection point caused by nonlinear effects, the experiment subjected the heating component to overload conditions exceeding the specified range. The data indicated that when transient overheating leads to excessive melting of the heat transfer medium and the liquidus ratio characteristic value... When the threshold value exceeds 0.70, the equivalent thermal resistance of the heat dissipation interface undergoes a step-like nonlinear abrupt change, resulting in a film boiling saturation effect within the heat transfer medium. This causes an overall decrease in heat dissipation efficiency of 45%, and the internal temperature of the control component rapidly increases at a rate of 15℃ / s. The appearance of this performance degradation inflection point confirms the scientific necessity of the upper threshold value of 0.55 as a physical boundary to prevent complete medium loss and failure. Correspondingly, when the control condition drops below the lower limit boundary and the liquid phase ratio characteristic value... When the voltage drops to 0.20, the latent heat absorption stage cannot be triggered within the medium, and adjustments to the control voltage command cannot interfere with the phase change melting point line. This contradictory negative effect confirms that the 0.35 to 0.55 range is a working window that balances the system's transient shock response and stable heat transfer. Comparison of experimental data covering the complete intensity gradient of the aforementioned problem confirms that using an interface property adaptive control method based on the phase change heat transfer mechanism allows the control system to adaptively adjust the transient stress state on both sides of the heat transfer interface to match the thermal shock load. Compared to not using acoustic... The control group system, which interlocks the characteristic impedance parameters with the main bus current parameters, exhibits a command feedback lag of up to 85ms and a local temperature overshoot of 24°C when faced with the same sudden change in heat flow. In contrast, the prototype system of this invention reduces the response delay of the control loop to 12ms and controls the transient overshoot temperature rise of the control components to within 3.5°C. The data, in its overall logic, echoes the technical feature of penetrating the latent heat sensing blind zone through state vector feedback control. The eigenvalue control method has engineering feasibility and industrial practical value in the manufacturing conditions of industrial automatic control system devices.

[0050] Example 3: This example combines Figures 1 to 2 This paper describes an adaptive control method for interface properties based on a phase change heat transfer mechanism, such as... Figure 1 As shown, step S101 synchronously collects the acoustic characteristic impedance parameters and main bus current parameters of the heat transfer interface and stores them in the timing data register as state feature vectors. Step S102 calculates the output liquid phase ratio characteristic value according to the monotonic correspondence between the acoustic characteristic impedance parameters and the pure solid / liquid reference, and performs timing differential calculation on the main bus current parameters to generate feedforward flow compensation parameters. Step S103 divides the control interval through a discrete state machine. When the liquid phase ratio reaches the upper limit and the feedforward flow exceeds the limit, a differential control voltage command is generated according to the deviation from the absolute span and a stepped compensation gradient is matched to drive the stress adjustment unit to adjust the transient mechanical output. Step S104 monitors the temperature change rate parameter of the heat transfer interface. When it exceeds the safety threshold, a depolarized reverse electric field pulse current is superimposed on the differential control voltage command to neutralize the physical residual polarization charge inside the interface stress adjustment unit.

[0051] like Figure 2As shown, starting from the resting ready state, when the closed-loop adaptive control is activated, it enters the phase change control state. In the phase change control state, when the liquid phase ratio reaches the upper limit and the feedforward flow exceeds the limit, the mechanical output is adjusted by differential voltage to switch to the stepped compensation state, and it can return to the phase change control state. In the phase change control state, when the temperature change rate exceeds the safety threshold, it switches to the depolarization control state by neutralizing the physical residual polarization charge, and it can return to the phase change control state. In the phase change control state, when the overheat fuse mechanism is triggered, it switches to the overheat fuse state. When the current is lower than the no-load current threshold, it switches from the resting ready state to the reference calibration test state, and returns to the resting ready state after performing a replacement overlay pure solid-state reference.

[0052] Example 4: In the ultra-high frequency power electronic converter heat dissipation monitoring of industrial automatic control system devices, when the system faces dynamic stress impact conditions such as short-circuit overload protection of semiconductor switching transistors or extreme alternating power cycles, the local temperature change rate on both sides of the heat transfer interface instantaneously exceeds the preset critical temperature change threshold. Due to the piezoelectric and pyroelectric lattice distortion effects between the heat transfer medium and the sensor contact surface, the residual polarization charge at the interface coupling point accumulates in a step-like manner. This accumulation of polarization charge generates local electrostatic field blockage, thereby interfering with the electrical signal transmission impedance of the acoustic sensor inside the state acquisition module, resulting in discrete acquisition of acoustic characteristic impedance parameters. Non-thermodynamic drift occurs; this sensor zero-point drift caused by charge physical polarization prevents the central control unit from accurately decoupling the liquid fraction characteristic value that reflects the true liquefaction ratio. This causes the closed-loop control system to issue incorrect differential control voltage commands when the heat transfer boundary properties change abruptly, ultimately leading to a control dead zone in the overall regulating circuit and hardware thermal overload failure.

[0053] To quantify the degree of interference of residual polarization charge accumulated at the heat dissipation interface due to piezoelectric lattice distortion on impedance sampling, this technical solution, based on Gauss's law for electrostatic fields and the principle of charge conservation, defines a depolarization charge neutralization parameter for quantitative analysis by calculating the ratio of the residual polarization charge obtained by integrating the signal loop to the preset single-pulse load charge. Its linear calculation formula is as follows: ,in, The parameter for depolarization charge neutralization. This represents the residual polarization charge on the heat dissipation interface side. To determine the reference charge amount for a single non-periodic pulse in the depolarization pulse current, the state acquisition module automatically activates the charge neutralization control program when it detects that the local temperature change rate on the heat transfer interface side exceeds the critical temperature change threshold of 50℃ / s. This is in response to the situation where heavy load stress occurs on the heat dissipation interface side. The residual polarization charge of Coulomb and the reference charge are set as follows: During coulomb calculation, the central control unit substitutes the residual polarization charge and the reference charge into the linear calculation formula to decouple and generate a depolarization charge neutralization parameter with a value of 3.0. The control arbitration module retrieves the depolarization charge neutralization parameter from the timing data register. Based on its discrete scalar value, the corresponding pulse flow control instruction is retrieved from the register, and the depolarization charge neutralization parameter is adjusted. Within a charge accumulation range of 2.5 or greater, the control command output unit invokes the built-in active phase-shift calibration law. Specifically, through the timing control firmware of the central control unit, a set of high-low level inverted time-division pulse offset actions is executed during each microsecond-level ultrasonic signal pulse acquisition interval window. Specifically, at the 1μs point after the ultrasonic transducer completes the reception of the previous cycle's acoustic pulse and the output pin level returns to zero, a high-impedance isolation protection is applied to the transducer pin by the external inverter half-bridge topology drive circuit. Then, immediately following, during the 2μs to 4μs interval of this acquisition interval, the pin voltage is pulled down to -0.75V in the negative direction, forming a constant reverse static DC electric field with a duration of 2μs and an amplitude of 15% of the nominal reference voltage. This forces the stray positive charges bound to the electrode surface due to distortion inside the piezoelectric crystal to undergo direct physical recombination and neutralization with the injected negative charges. This increases the transmission frequency of the depolarized pulse current output to the signal transmission circuit from 5kHz to 20kHz, and at the acoustic characteristic impedance parameter... During the signal acquisition intervals, negative alternating charges are directionally injected into the signal electrodes of the acoustic sensor to neutralize the residual charges adhering to the electrode interfaces, until the residual polarization charge is fed back by current integration. The impedance is reduced below the differential threshold, thereby eliminating the sensor zero-point drift caused by residual charge accumulation and restoring the original electrical balance state of impedance sampling.

[0054] To ensure that no logic deadlock or force field control misalignment occurs when the depolarization reverse electric field pulse current and differential control voltage act simultaneously on the same physical interface stress adjustment unit within the same cycle, this invention implements a time-division multiplexing alternating control timing at the hardware and algorithm levels. Within each complete 20-microsecond control cycle, the first 18 microseconds are allocated to the force field control period. During this time, the central control unit closes the depolarization control path and only outputs the differential control voltage to the interface stress adjustment unit, driving it to complete precise output displacement and output quantity. In the last 2 microseconds of the control cycle, the system automatically switches to the charge offset period. At this time, the force field control output level is forcibly locked at the high impedance maintenance state of the current sampling point. The depolarization control path is instantaneously activated and directionally injects a reverse electric field pulse current with an amplitude of 15% of the nominal reference. Because the duration of this depolarization pulse current is extremely short and the frequency is extremely high, its energy only acts to neutralize the spatial residual polarization charge accumulated inside the piezoelectric lattice. Within the mechanical inertia time constant scale of the piezoelectric ceramic actuator, it will not cause substantial step fluctuations in the output displacement, thus ensuring complete... The transient interference to the mechanical force field is completely avoided. The drive circuit of the interface stress adjustment unit adopts a half-bridge inverter topology with dual-channel high-speed isolated gate drivers. At the end of the force field control period of the first 18 microseconds within the 20 microsecond control cycle, the central control unit generates a 50 nanosecond hardware dead time to cut off the forward drive path, so that the level state of the interface stress adjustment unit is maintained at high impedance. In the last 2 microsecond charge offset period, the depolarization control path is turned on, and a reverse electric field with an amplitude of 15% of the nominal reference control voltage is injected. Taking advantage of the physical characteristic that the time span of the charge offset period is less than the mechanical response delay of the interface stress adjustment unit, the reverse electric field energy is used to neutralize the residual polarization charge in space and maintain the constant state of mechanical output displacement. Regarding the aforementioned temperature change rate determination, 5°C per second is the basic safe temperature change threshold for triggering the normal depolarization control law, while 50°C per second corresponds to the secondary emergency specific criterion for activating the active phase shift calibration law and the step increase of the transmission frequency under extreme impact. The two operate in a layered and nested manner in the control logic, together forming a complete closed-loop protection system.

[0055] Under this adaptive neutralization control law based on the phase change heat transfer mechanism and deep coupling of depolarized pulse current, the automatic control system eliminates the measurement bias caused by thermoelectric coupling charge under extreme temperature stress conditions. The acoustic characteristic impedance parameters output by the state acquisition module... The measurement error was stabilized and reduced to within 0.5%, thus ensuring the central control unit's accuracy in determining the liquid phase ratio characteristic value. and feedforward flow compensation parameters Precise closed-loop decoupling, updating the issued differential control voltage command. Maintaining real-time synchronization with the actual evolution of interface properties, the core temperature overshoot amplitude of the control component does not exceed 2.1℃ when subjected to an extreme step transient thermal load impact of 800W / cm². This eliminates the divergence of closed-loop control signals and adjustment lag caused by the accumulation of polarization charge in the sensor. The closed-loop control stability and long-cycle operation life of the entire cabinet power electronic control device are improved in harsh thermoelectric coupling environments.

[0056] Example 5: When the system faces the calibration conditions of the heat transfer medium after replacement, the control terminal applies a set clamping force to the heat dissipation interface before the closed loop is started. The central control unit collects the initial impedance on the heat dissipation interface side when the heat transfer medium is in a solid state at 25°C, and uses the initial impedance as the pure solid-state reference impedance parameter. Write the data to the register. The temperature control source drives the heat transfer medium to heat up to 85°C, the liquid state. The acquisition module then acquires the impedance again and uses this value as the pure liquid reference impedance parameter. Write to the register to establish the intrinsic characteristic baseline for the current deployment environment.

[0057] The central control unit retrieves the pure solid-state reference impedance parameters from the register. With pure liquid reference impedance parameters The impedance calibration factor is calculated based on the difference between the two. Impedance calibration factor The calculation formula is as follows: ,in, For impedance calibration factor, The characteristic impedance reference value for a pure solid phase change medium. The characteristic impedance reference value of a pure liquid phase change medium is used to control the arbitration module in determining the impedance calibration factor. Lock the register when it is within the feature window, and update the impedance calibration factor. It is used directly as a product factor in the liquid phase ratio characteristic value. In the decoupling calculation, the initial static bias caused by material batch differences is eliminated, so that the subsequent output differential control voltage command is more effective. Corresponding to the actual phase transition stage of physical properties, the temperature step response deviation inside the heat dissipation component steadily converges within 0.3℃ during operation testing.

[0058] Example 6: In scenarios involving the replacement of high-power power electronic switching devices in different batches or the initialization of the deployment environment for control components, the thermal control test bench provides a set temperature field to establish a solid-liquid two-phase property calibration path at the heat dissipation interface. This allows for the construction of a control reference and step gradient matrix for online adaptive adjustment. During the initial system assembly stage, the temperature parameter sampling unit monitors the surface physical temperature of the heat transfer interface. When the surface physical temperature is maintained at 25°C and the heat transfer medium is in a solid state, the state acquisition module collects the acoustic characteristic impedance parameters of the heat transfer interface. The collected acoustic characteristic impedance parameters are then used as the pure solid-state reference impedance parameters. Stored in a high-speed timing data register, containing pure solid-state reference impedance parameters. The calibrated value is 15,500,000 kg / m²·s. Based on this, the temperature control source drives the heat transfer medium to heat up. When the surface physical temperature is maintained at 85°C and the heat transfer medium is in a liquid state, the state acquisition module collects the acoustic characteristic impedance parameters of the heat transfer interface and uses the acoustic characteristic impedance parameters collected at this time as the pure liquid reference impedance parameters. Write the high-speed timing data register, which contains the pure liquid reference impedance parameters. The calibrated value is 7,500,000 kg / m²·s.

[0059] The central control unit retrieves the pure solid-state reference impedance parameters from the high-speed timing data register. With pure liquid reference impedance parameters The impedance calibration factor is calculated based on the difference between the two. Impedance calibration factor The calculation formula is as follows: ,in, For impedance calibration factor, The characteristic impedance reference value for a pure solid phase change medium. Using the characteristic impedance reference value of a pure liquid phase change medium, the central control unit substitutes the value into the calculation formula to calculate an impedance calibration factor of 0.000000125 m²·s / kg. Acoustic characteristic impedance parameters used to characterize the heat dissipation interface side Its impedance parameter as it decreases with melting of the medium is compared to that of a pure solid reference. The extent to which the difference widens makes the difference related to the impedance calibration factor. The product accurately corresponds to the liquid phase fraction characteristic value. The upward trend is used to eliminate the initial static bias caused by material batch differences.

[0060] To determine the interval boundaries of the adaptive control law, the central control unit configures three calibration sample cases to align with the boundary points of the control window, based on the acoustic characteristic impedance parameters. Under the first calibration sample group operating condition of 12,700,000 kg / m²·s, the central control unit uses pure solid-state reference impedance parameters. Acoustic characteristic impedance parameters and impedance calibration factor The characteristic value of the liquid phase fraction was calculated to be 0.35. Using this as the lower limit of the core sustaining range, the nominal control voltage reference maintains the current steady-state output of 5.0V, while the acoustic characteristic impedance parameter... Under the operating condition of the second calibration sample group with a flow rate of 11,900,000 kg / m²·s, the central control unit calculated a liquid phase fraction characteristic value of 0.45. Using the normal median value as the core maintenance range, at which point the phase change medium is in a state of equilibrium with solid and liquid coexistence and the nominal reference control voltage is maintained at 5.0V, the acoustic characteristic impedance parameter... Under the operating condition of the third calibration sample group with a flow rate of 11,100,000 kg / m²·s, the central control unit calculated a liquid phase fraction characteristic value of 0.55. The core maintenance interval is defined as the upper limit of the critical value. At this point, the heat transfer medium touches the edge of the latent heat absorption platform and the equivalent thermal resistance of the interface on the heat dissipation boundary shows a trend of nonlinear abrupt change. The aforementioned core maintenance interval is limited to the range of 0.35 to 0.55 and has a tight thermodynamic boundary support. When the liquid phase ratio characteristic value is lower than 0.35, the solid lattice inside the heat transfer medium still occupies an absolute dominant position and has not yet fully formed a continuous multidimensional macroscopic heat conduction two-phase network. The system cannot make full use of the phase change latent heat absorption effect of the large equivalent specific heat capacity. When the liquid phase ratio characteristic value exceeds 0.55, the heat transfer medium melts excessively and exhibits obvious fluid shear thinning behavior. The sensitivity of mechanical stress adjustment to local contact resistance drops sharply by more than 50%, which can easily cause medium loss and lead to deterioration of heat dissipation efficiency.

[0061] Therefore, maintaining the range between 0.35 and 0.55 ensures high transient shock resistance and long-term thermal control stability. Furthermore, a depolarization reverse electric field pulse current with an amplitude of 15% of the nominal control voltage reference is superimposed. This 15% proportionality is derived from the experimentally measured residual polarization charge of the piezoelectric material. Standardized coulomb counter measurements show that, under extreme transient thermal shock conditions, the average residual physical polarization charge enriched within the interface stress adjustment unit due to piezoelectric lattice distortion is 1.5 x 10⁻⁹ coulombs. If the depolarization pulse current amplitude is lower than the nominal reference... If the amplitude is 15%, the electric field strength is insufficient to neutralize the residual charge in the deep lattice, and the zero-point drift of the sensor cannot be eliminated. If the amplitude is higher than 15% of the nominal reference, the strong reverse electric field will induce a nonlinear reverse mechanical stress jump in the drive component, destroying the control accuracy of the output displacement. Precisely locking the amplitude coefficient at 15% can effectively achieve a self-consistent closed loop of electrical balance and mechanical output stability. To balance signal noise suppression and response delay, the control arbitration module determines the preset mutation threshold of the discrete state machine decision law through graded gradient testing. This is done when the system is switched without load and the main bus current parameter is... Feedforward flow compensation parameters generated by time-differential calculation under operating conditions with high-frequency electromagnetic noise. A fluctuation component of 0.5V / ms to 1.5V / ms is generated under no-load fluctuation. If the preset sudden change threshold is set below 1.5V / ms, the control loop will experience misjudgments and oscillations in the control voltage command caused by the fluctuation component during the resting phase. If the preset sudden change threshold is set above 5.0V / ms, the feedforward current compensation parameters will be affected under high-current nonlinear step load conditions. Since the phase transition melting point line cannot be changed, the control arbitration module determines the preset mutation threshold to be 3.0V / ms through data aggregation and performance trade-offs. This preset mutation threshold is used as the quantitative comparison rule for the discrete state machine decision law, so that the system can suppress electromagnetic interference while controlling the feedforward response delay to less than 12ms. The aforementioned preset mutation threshold of 3.0V / ms has clear engineering boundaries and physical significance. In actual industrial environments, when the heating component switches between no-load and no-load or when the normal operating voltage fluctuates, due to the inherent power frequency noise and random electromagnetic coupling interference of the power supply circuit, the parasitic fluctuation rate of the feedforward flow compensation parameter in the timing differential operation is usually in the range of 0.5V / ms to 1.5V / ms.

[0062] Therefore, setting the physical lower limit of the preset mutation threshold to 3.0 volts per millisecond can improve the anti-interference margin of the control system and ensure that the system will not generate self-excited oscillation and misjudgment of control voltage commands due to high-frequency spike electromagnetic noise during the resting phase. When high-power power electronic switching devices encounter heavy load switching or sudden overload thermal shock, the real transient jump of the main bus current will cause the differential change rate of the feedforward current compensation parameter to quickly exceed 5.0 volts per millisecond. Limiting the upper limit of this mutation threshold to 3.0 volts per millisecond can ensure that the feedforward response loop is 100% activated in the very early stage of overload, thereby completing the advance output of differential voltage commands within a time delay window of less than 12 milliseconds, effectively blocking the temperature overshoot caused by the heat transfer blind zone. The discrete state machine decision law uses the built-in adaptive mapping matrix to determine the liquid phase ratio characteristic value. absolute span of deviation from the upper limit threshold Determine the voltage command change amount, absolute span The calculation formula is as follows: ,in, For absolute span, The characteristic value of the liquid phase ratio is the liquid phase ratio under the operating conditions of the third calibration sample group. Reaching 0.55 and feedforward flow compensation parameters When the rate of change exceeds the preset mutation threshold of 3.0V / ms, the control arbitration module initiates the adaptive mapping matrix. If the calculated absolute span... The adaptive mapping matrix outputs a stepped compensation gradient of 2.5V with a value of 0.00. The control command output unit accumulates this stepped compensation gradient on the nominal control voltage reference of 5.0V, generates and outputs a differential control voltage command with an amplitude of 7.5V to the interface stress adjustment unit. If the high-power overload condition causes the liquid phase fraction characteristic value to decrease... The absolute span was increased to 0.60 and calculated. When the value is 0.05, the adaptive mapping matrix matches the output value of a stepped compensation gradient of 12.5V, and the differential control voltage command generated by the control command output unit is accumulated. The voltage is correspondingly increased to 17.5V to drive the interface stress adjustment unit to increase the contact pressure of the heat transfer interface and reduce the contact thermal resistance, thereby controlling the temperature rise trend of the heat dissipation interface. Under commissioning conditions, the main bus current parameter is made to jump to 80% of the rated maximum current value. When the liquid phase ratio characteristic value exceeds the upper limit critical value of the phase change control range and the feedforward flow compensation parameter changes abruptly, the control loop increases the differential control voltage command in increments of 0.5V and monitors the changing trend of the temperature change rate parameter. When the first derivative of the surface physical temperature with respect to time changes from positive to negative or zero, the total accumulated voltage change is recorded and input into the control lookup table as the step-like compensation gradient corresponding to the absolute span. The quantitative writing of the discrete gradient data of each level in the adaptive mapping matrix is ​​completed. Furthermore, the underlying physical mechanism of the aforementioned adaptive intervention on the melting point equilibrium line of the phase change medium does not refer to changing the intrinsic macroscopic thermodynamic melting point of the heat transfer medium itself, but rather to controlling the phase change evolution trend by regulating the dynamic thermal equilibrium state at the micro level. When the interface stress adjustment unit increases the macroscopic mechanical contact pressure of the heat transfer interface, the microscopic protrusions on both sides of the heat transfer interface undergo local elastoplastic deformation, which rapidly expands the actual effective microscopic contact area within a scale of 1 square millimeter. The microscopic extrusion contact resistance then drops sharply. This change in the microscopic contact state directly improves the transient convection and conduction combined heat transfer of the heat transfer interface.

[0063] Under the influence of a significant increase in heat transfer rate, the local temperature gradient and thermal boundary layer thickness within the phase change medium layer are reconstructed, causing spatial displacement of the phase change front, the interface between the solid and liquid phases where their properties dynamically evolve. Since the local heat flux and latent heat absorption rate reach a new dynamic equilibrium at the microscopic level, this macroscopically manifests as successful intervention in the evolution of the solid-liquid phase change. This achieves advance adjustment and smooth transition of the control path before a substantial jump in the junction temperature outside the heating element. During the initial power supply phase before the adaptive control loop operates, the central control unit deploys a pre-condition verification procedure to handle initial environmental anomalies, and the timing data acquisition module reads the acoustic characteristic impedance parameters of the heat transfer interface. The acoustic characteristic impedance parameters read Exceeding the reference constraint range of 14,000,000 kg / m²·s to 16,000,000 kg / m²·s or the main bus current parameter When the current exceeds the preset no-load current threshold, the central control unit determines that there is mechanical contact peeling at the heat dissipation interface or a fault in the sensor's electrical path, and controls the system to send a differential control voltage command. It forcibly locks to a zero-level state, sends a status fault code to the overall management terminal, and refuses to start the closed-loop adaptive control process, thereby avoiding control divergence caused by abnormal initial conditions.

[0064] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit of this application and the scope of protection of this invention, and all of these forms are within the protection scope of this application.

Claims

1. An adaptive control method for interface properties based on a phase change heat transfer mechanism, characterized in that, Includes the following steps: Step S101: Synchronously collect the acoustic characteristic impedance parameters and main bus current parameters of the heat transfer interface, and store the acoustic characteristic impedance parameters and main bus current parameters as state feature vectors into the timing data register. Step S102: Retrieve the state feature vector in the timing data register, calculate and output the liquid phase ratio feature value that characterizes the melting ratio of the heat transfer medium based on the monotonic correspondence between the acoustic characteristic impedance parameter and the pre-stored pure solid reference impedance parameter and pure liquid reference impedance parameter, and perform timing difference calculation on the main bus current parameter to generate feedforward flow compensation parameter. Step S103: The range of liquid phase ratio characteristic value is divided into three non-overlapping control intervals by using a discrete state machine. When the liquid phase ratio characteristic value is within the phase change control interval, the nominal reference control voltage is maintained. When the liquid phase ratio characteristic value reaches the upper limit critical value of the phase change control interval and the rate of change of the feedforward flow compensation parameter exceeds the preset sudden change threshold, a step-by-step compensation gradient is matched according to the absolute span of the liquid phase ratio characteristic value deviating from the upper limit critical value, a differential control voltage command is generated, and the interface stress adjustment unit is driven to adjust the transient mechanical output of the heat transfer interface. Step S104: Monitor the temperature change rate parameter of the heat transfer interface. When the temperature change rate parameter exceeds the safe temperature change threshold, superimpose a depolarization reverse electric field pulse current on the differential control voltage command to neutralize the physical residual polarization charge inside the interface stress adjustment unit.

2. The interface property adaptive control method based on phase change heat transfer mechanism according to claim 1, characterized in that, Step S101 includes the following sub-steps: Step S1011, synchronously sample the acoustic characteristic impedance parameters of the heat transfer interface and the main bus current parameters using a sampling period of 20μs; Step S1012, combine the synchronously sampled acoustic characteristic impedance parameters and main bus current parameters into a multi-dimensional flow data stream, and continuously transfer them to the timing data register according to the acquisition time.

3. The adaptive control method for interface properties based on a phase change heat transfer mechanism according to claim 1, characterized in that, Step S103 includes the following sub-steps: Step S1031, in the discrete state machine, the value range of the liquid phase ratio characteristic value is subdivided into three control intervals. When the liquid phase ratio characteristic value is in the interval of 0.35 to 0.55, it is taken as the phase change control interval and the current nominal reference of the control voltage is maintained; Step S1032, when the liquid phase ratio characteristic value reaches the upper limit critical value of 0.55 and the rate of change of the feedforward flow compensation parameter exceeds the preset mutation threshold, the absolute span of the liquid phase ratio characteristic value deviating from 0.55 is determined; Step S1033, according to the absolute span, a stepped compensation gradient is dynamically matched, and the stepped compensation gradient is accumulated on the nominal reference of the control voltage to update the differential control voltage command.

4. The interface property adaptive control method based on phase change heat transfer mechanism according to claim 3, characterized in that, The step S1033, which generates a differential control voltage command and drives the interface stress adjustment unit to adjust the transient mechanical output of the heat transfer interface, includes: the control change judgment module outputs an updated differential control voltage command to the interface stress adjustment unit, drives the interface stress adjustment unit to change the output displacement, and adjusts the micro-contact resistance in real time by adjusting the degree of micro-contact compression on the heat transfer interface side.

5. The interface property adaptive control method based on phase change heat transfer mechanism according to claim 1, characterized in that, Step S104 includes the following sub-steps: Step S1041, the surface physical temperature of the heat transfer interface is monitored in real time by the temperature parameter sampling unit, and the first derivative of the surface physical temperature with respect to time is calculated to generate the temperature change rate parameter; Step S1042, it is determined whether the temperature change rate parameter exceeds the safe temperature change threshold of 5°C / s, and when the temperature change rate parameter continues to exceed 5°C / s, the depolarization control law is activated; Step S1043, a depolarization reverse electric field pulse current with an amplitude of 15% of the nominal reference of the control voltage is superimposed on the differential control voltage command to eliminate the charge accumulation generated inside the interface stress adjustment unit.

6. The interface property adaptive control method based on phase change heat transfer mechanism according to claim 1, characterized in that, The adaptive control method also includes the following automatic calibration parallel steps: Step S105, continuously compare the main bus current parameter in the timing data register with the preset no-load current threshold. When the main bus current parameter is lower than the preset no-load current threshold for 5 seconds and the control loop is in a resting ready state, issue a reference calibration test command; Step S106, the control interface stress adjustment unit applies a standard step pressure to the heat transfer interface and reads the current acoustic characteristic impedance parameter as a transient calibration reference value; Step S107, replace the original stored pure solid-state reference impedance parameter with the transient calibration reference value.

7. The interface property adaptive control method based on phase change heat transfer mechanism according to claim 1, characterized in that, The adaptive control method also includes the following safety protection parallel steps: Step S108, compare the liquid phase ratio characteristic value with the limit and determine whether the liquid phase ratio characteristic value exceeds the safe phase change melting threshold of 0.85; Step S109, when it is determined that the liquid phase ratio characteristic value exceeds 0.85 for three consecutive control cycles, and the feedforward flow compensation parameter is higher than the critical limit parameter, that is, the steady-state feedforward upper limit voltage value of 5.5V, which is preset and stored in the read-only memory of the central control unit and is used to characterize the power electronic switching device at the extreme heavy load safety edge, the overheating melting mechanism is triggered; Step S110, issue a safety cut-off command to forcibly shut down the main bus power supply circuit of the system.

8. The adaptive control method for interface properties based on phase change heat transfer mechanism according to claim 1, characterized in that, Before calculating the timing difference of the main bus current parameters in step S102, the high-frequency spike noise in the sampled main bus current parameters and acoustic characteristic impedance parameters is filtered out by using a low-pass filter unit to smooth out the transient differential oscillation of the differential control voltage command.

9. The adaptive control method for interface properties based on a phase change heat transfer mechanism according to claim 1, characterized in that, The interface property adaptive control method based on phase change heat transfer mechanism also includes, at the boundary of each control cycle, re-storing the updated acoustic characteristic impedance parameters and main bus current parameters as initial feedback variables into the timing data register, and through the interactive iteration of acoustic characteristic impedance parameters and main bus current parameters, running long-cycle closed-loop adaptive regulation of heat transfer properties when transient heat flux transition occurs at the heat transfer interface.

Images