A brushless fuel pump system

Abstract

The present application relates to the technical field of oil pump, in particular to a brushless fuel pump system, comprising a microcontroller, the microcontroller acquires real-time phase current of the motor, bus voltage of the system and fluid temperature; based on the bus voltage and the fluid temperature, a dynamic threshold is constructed; the instantaneous change rate of the real-time phase current is extracted, and its absolute value is compared with the dynamic threshold, when the absolute value continuously exceeds the dynamic threshold, it is determined that the cavitation phenomenon occurs; when the cavitation phenomenon occurs, the current cruise mode is cut off, and the gas cavity protection mode is switched to execute the exhaust operation; during the operation of the gas cavity protection mode, based on the fluctuation characteristics of the real-time phase current, the variance value reflecting the rotor load fluctuation degree is continuously calculated; when it is confirmed that the variance value meets the preset recovery condition, the operation of switching back to the cruise mode from the gas cavity protection mode is executed to restore the high-pressure oil supply state; the running reliability and stability of the brushless fuel pump under complex working conditions are improved.

Description

Technical Field

[0001] This invention relates to the field of oil pump technology, and more specifically to a brushless fuel pump system. Background Technology

[0002] Brushless fuel pump systems are widely used in the fuel supply systems of internal combustion engine vehicles and hybrid vehicles. They use a microcontroller to directionally control a permanent magnet motor to drive the impeller to rotate at high speed, creating a pressure difference within the pump chamber, thereby achieving the technical effect of accurately outputting high-pressure fuel.

[0003] In the prior art, since the controller of an oil pump is generally encapsulated inside the pump body, the heat generated during its operation is difficult to dissipate to the external environment through an effective heat conduction path, which can easily lead to overheating of the electronic components inside the controller, affecting its operational stability and service life.

[0004] Chinese patent CN116733646B discloses a brushless motor fuel pump with a built-in controller, comprising a control unit, a brushless motor, and a fuel pump connected in sequence. The control unit includes an outlet cap, a controller, a thermal pad, a mounting bracket, and a metal heat sink. The mounting bracket is fixedly installed on the upper end of the brushless motor, and a flow chamber for fuel outflow is provided between the mounting bracket and the upper end of the brushless motor. The flow chamber, the brushless motor, and the fuel pump are connected in sequence. The outlet cap is fixedly connected to the upper end of the mounting bracket and has an outlet port communicating with the flow chamber. The mounting bracket and the outlet cap are sealed together to form a sealed cavity for mounting the controller. The controller, the thermal pad, and the metal heat sink are connected in sequence from top to bottom. The lower part of the metal heat sink is sealed and installed in a mounting hole provided in the mounting bracket, and the metal heat sink passes through the mounting hole and is exposed in the flow chamber.

[0005] While the above solutions can achieve heat dissipation for the controller, there are still corresponding defects in the actual operation of the fuel pump. For example, when the engine operating conditions change suddenly, the fuel level in the tank is too low, or there is a momentary blockage in the fuel supply line, the inlet of the fuel pump is prone to "cavitation" due to a sudden drop in pressure. The generation and collapse of a large number of bubbles will not only cause hydraulic shock, noise and cavitation, but more importantly, it will severely disturb the density of the medium flowing through the impeller and the torque of the motor load. For systems that rely on precise rotor position feedback for magnetic field orientation control, such sudden changes in load and medium characteristics will directly cause the rotor position estimation algorithm based on back electromotive force or current observation to become inaccurate, and the controller will be unable to maintain the directional lock on the rotor magnetic field, that is, the "lack of lock" phenomenon will occur.

[0006] To address disturbances caused by cavitation, existing technologies typically employ a logic-based judgment mechanism that sets a single static current threshold. This mechanism identifies cavitation and triggers degradation protection by monitoring the rate of change of phase current. However, the conditions for cavitation are closely related to the saturated vapor pressure of the fuel. In winter and summer, and when using pure gasoline and ethanol gasoline, the viscosity, density, and vaporization critical point of the fuel are completely different. At the same time, the system's bus voltage also fluctuates. Fixed logic judgment rules are prone to misjudgment or missed judgment under extreme high temperatures or low pressures. If the threshold is set too sensitively, normal voltage fluctuations or fluid pulsations may trigger unwarranted degradation, leading to insufficient fuel supply. If the threshold is set too sluggishly, the system may not respond in time when actual cavitation occurs, still resulting in a "lack of lock-up." Summary of the Invention

[0007] The purpose of this invention is to provide a brushless fuel pump system to solve the technical problems mentioned in the background section.

[0008] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a brushless fuel pump system, including a microcontroller, which executes the following control logic: Obtain the real-time phase current of the motor, the bus voltage of the system, and the fluid temperature; Based on bus voltage and fluid temperature, synchronous compensation calculations are performed on preset basic thresholds to construct dynamic thresholds for identifying cavitation boundaries. Extract the instantaneous rate of change of the real-time phase current and compare its absolute value with the dynamic threshold in real time. When the absolute value continues to exceed the dynamic threshold, the system is determined to enter the state of sudden load unloading caused by cavitation. Based on the determination of the sudden unloading state of the load, the current cruise mode is cut off and the preset cavitation protection mode is switched to perform the exhaust operation. During operation in cavitation protection mode, the variance value used to reflect the degree of rotor load fluctuation is continuously calculated based on the fluctuation characteristics of real-time phase current. When the variance value is confirmed to meet the preset recovery conditions, the operation of smoothly switching back from cavitation protection mode to cruise mode is performed to restore the high-pressure fuel supply state.

[0009] Preferably, the step of constructing a dynamic threshold for identifying the cavitation boundary specifically includes: The voltage compensation coefficient corresponding to the bus voltage and the temperature compensation coefficient corresponding to the fluid temperature are obtained respectively; the voltage compensation coefficient and the temperature compensation coefficient are obtained in real time through a one-dimensional linear lookup table method pre-stored in the microcontroller. The current dynamic threshold is obtained by correlating the basic threshold with the voltage compensation coefficient and the temperature compensation coefficient respectively. The temperature compensation coefficient is set to decrease monotonically as the fluid temperature increases, so as to reduce the dynamic threshold and improve the cavitation detection sensitivity when the fuel is at a high temperature and close to the vaporization critical point; the voltage compensation coefficient is set to decrease synchronously with the bus voltage drop ratio, so as to eliminate the interference caused by normal voltage fluctuations on the current change rate.

[0010] Preferably, the step of cutting off the current cruise mode further includes freezing the internal operating state of the microcontroller: Upon determining that the system has entered a state of sudden load unloading, a freeze command is issued simultaneously. Based on the freeze command, the position observer's state update is forcibly stopped, and the error integral accumulation process of the speed regulator and current regulator is simultaneously paused to prevent the observer's calculation from diverging and the regulator from entering saturation due to sudden load changes.

[0011] Preferably, the step of switching to a preset cavitation protection mode to perform the exhaust operation specifically includes: In cavitation protection mode, the microcontroller continuously applies a low-frequency drive waveform to the stator coil of the motor, forcing the motor rotor to switch from high-speed closed-loop operation to low-speed synchronous operation. By reducing the rotational speed, the centrifugal force on the fluid in the impeller channel is reduced, allowing the fuel vapor and bubbles generated by cavitation to be smoothly discharged through the exhaust gap or channel inside the pump body by relying on their own buoyancy.

[0012] Preferably, the step of calculating the variance value includes the following specific algorithm: A sliding window of fixed time length is set inside the microcontroller, and the real-time instantaneous value of the phase current of all historical sampling points within the window is obtained in each calculation cycle; Calculate the average of all instantaneous current values ​​within the current sliding window; Calculate the squared difference between the instantaneous current value and the average value within the window, and then take the average of all the squared values ​​as the variance value of the current cycle.

[0013] Preferably, the evaluation logic for confirming that the variance value meets the preset recovery conditions is as follows: Real-time comparison of continuously calculated variance values ​​with pre-stored recovery thresholds; When the impeller channel is still in a gas-liquid mixed state, the load fluctuates violently, and the variance value will remain in the range above the recovery threshold. When the continuously calculated variance value has dropped below the recovery threshold, and the state below the threshold remains stable for more than the preset judgment time, the microcontroller determines that the air bubbles in the pump chamber have been emptied and the liquid fuel has re-established a stable load, and outputs a trigger signal that meets the recovery conditions.

[0014] Preferably, the operation of smoothly switching back from cavitation protection mode to cruise mode specifically includes angle forced alignment control: After the microcontroller receives the recovery trigger signal, it reads the real-time electrical angle currently output by the drive module in cavitation protection mode; The real-time electrical angle is used as the initial angle value and directly written into the position observer to achieve forced alignment of the control system angle, thereby eliminating the angle deviation caused by the instant of mode switching.

[0015] Preferably, the smooth switching operation further includes torque relief control: After completing the forced angle alignment, the microcontroller unfroze the position observer, speed regulator, and current regulator, and reactivated the cruise mode. At the moment of reactivation, the microcontroller sets the torque current command output by the current regulator to ramp control mode. Based on the preset ramp time, the control torque current command rises smoothly at a set slope to the target value that matches the current vehicle's required flow rate, in order to suppress the large current surge and torque fluctuations generated during mode switching.

[0016] Preferably, the microcontroller is also equipped with safe operating time monitoring logic to limit the time of exhaust operation: The total time the microcontroller will maintain in cavitation protection mode to perform the venting operation will be strictly limited to a safe time window; When the fuel pump is venting at low speed within the safety window, the system relies on the accumulator system at the front end of the pipeline to release the reserve fuel to maintain the system's base oil pressure, ensuring that the fuel supply is not interrupted during the cavitation venting phase when the pump body is temporarily unable to supply fuel at high pressure, thus avoiding engine shutdown.

[0017] Preferably, the current fluid temperature is obtained by acquiring the signal from a thermistor installed in the fuel line.

[0018] The technical effects and advantages of this invention are as follows: This invention monitors cavitation in real time using dynamic thresholds and makes multi-faceted compensation adjustments based on voltage, temperature, and fuel type, significantly improving the accuracy and environmental adaptability of cavitation identification. When cavitation is detected, the system immediately switches to a low-speed open-loop exhaust mode and freezes the observer and regulator integrals to prevent control lock-up and calculation divergence caused by sudden load changes. During the exhaust process, the load recovery status is dynamically evaluated based on the current variance, and an adaptive safety time window is used to manage the exhaust duration, ensuring sufficient exhaust without timeout. Furthermore, through forced angle alignment and torque current ramp control, a smooth transition from exhaust mode to high-pressure cruise mode is achieved, ensuring the stability of fuel pump supply. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2 This is a schematic diagram of the control flow of the present invention; Figure 3 This is a schematic diagram of the control of the cavitation protection mode of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1 Reference Figures 1 to 3 As shown, this embodiment provides a brushless fuel pump system, including a pump body, an impeller, a permanent magnet synchronous motor, an inverter bridge drive circuit, and a microcontroller.

[0022] The pump body serves as the chamber through which fuel flows, forming a pressurized space for the fuel in conjunction with the impeller. The impeller is mounted on the output shaft of the permanent magnet synchronous motor. When the motor rotates, the impeller rotates at high speed, performing work on the fuel, thereby establishing the required pressure difference between the pump's inlet and outlet to achieve fuel delivery. This part is existing technology and will not be elaborated upon here.

[0023] The permanent magnet synchronous motor serves as the power source, with its rotor coaxially connected to the impeller, driving the impeller to rotate at high speed via electromagnetic torque. Three-phase current lines are drawn from the motor's stator windings. The motor employs a sensorless design, meaning no internal position sensors are used to detect the rotor angle; instead, a sensorless closed-loop control algorithm is employed for control.

[0024] The inverter bridge drive circuit serves as the power interface connecting the microcontroller and the permanent magnet synchronous motor. It receives pulse width modulation signals from the microcontroller and converts the DC bus voltage into three-phase AC power with controllable amplitude and frequency, thereby driving the permanent magnet synchronous motor to operate.

[0025] The microcontroller, through its built-in analog-to-digital converter interface, receives in real-time instantaneous values ​​of the motor's three-phase current from the current sampling circuit, DC bus voltage values ​​from the voltage sampling circuit, and fuel temperature signals from temperature sensors installed in the fuel lines, such as negative temperature coefficient thermistors. The microcontroller has a pre-stored control program to execute a cavitation anti-lock-off cooperative control method, specifically: Obtain the real-time phase current of the motor, the bus voltage of the system, and the fluid temperature; Based on bus voltage and fluid temperature, a synchronous compensation calculation is performed on a preset basic threshold to construct a dynamic threshold for identifying the cavitation boundary. Extract the instantaneous rate of change of the phase current in real time and compare its absolute value with the dynamic threshold in real time. When the absolute value continues to exceed the dynamic threshold, the system is determined to enter the state of sudden load unloading caused by cavitation.

[0026] Specifically, the microcontroller uses its built-in analog-to-digital converter to simultaneously acquire the instantaneous values ​​of the three-phase current from the current sampling circuit and the DC bus voltage from the voltage sampling circuit at a preset sampling period, such as 100 microseconds.

[0027] Meanwhile, the microcontroller reads the resistance value of the thermistor installed in the fuel line through the temperature sensor interface; the microcontroller has a table pre-stored inside recording the corresponding resistance values ​​of this type of thermistor at different temperatures, i.e., the resistance-temperature correspondence table. By querying the resistance-temperature correspondence table, the real-time reading resistance value can be converted into the corresponding fluid temperature value.

[0028] Within each control cycle, after acquiring the original three-phase current, the microcontroller first performs Clark transformation and Park transformation on the acquired three-phase current to obtain the direct-axis current and quadrature-axis current in the synchronous rotating coordinate system. The quadrature-axis current can reflect the electromagnetic torque of the motor, and its rate of change is a sensitive indicator of the load torque change. Specifically, when the load driven by the motor, i.e. the fluid resistance of the fuel pump, suddenly changes, the quadrature-axis current will immediately change rapidly.

[0029] The microcontroller calculates the difference between the quadrature-axis current value of the current cycle and the quadrature-axis current value of the previous cycle, and then divides it by the sampling period to obtain an approximate instantaneous rate of change.

[0030] It should be noted that the Clark transformation and Park transformation involved in this embodiment are common processing steps in the field of motor control. The Clark transformation converts the current value from the three-phase stationary coordinate system into the current value in the two-phase stationary coordinate system. Then, the Park transformation is performed to further convert the current value in the two-phase stationary coordinate system into the two-phase rotating coordinate system that rotates synchronously with the magnetic field of the motor rotor. After these two transformations, the direct-axis current and quadrature-axis current are obtained. The above are conventional operations in the prior art and will not be described in detail here.

[0031] To accurately determine whether the instantaneous rate of change has reached the level of cavitation, a dynamic threshold is constructed within the microcontroller. The dynamic threshold consists of a base threshold, a voltage compensation coefficient, and a temperature compensation coefficient. The base threshold is an empirical value that can be determined through numerous experiments under standard laboratory conditions, such as a nominal voltage of 12V and a room temperature of 25℃.

[0032] The voltage compensation coefficient is obtained by consulting a preset voltage compensation coefficient table inside the microcontroller, which uses the real-time acquired bus voltage as input. When the actual detected bus voltage is equal to the nominal voltage, the voltage compensation coefficient is usually defined as 1, meaning that no voltage compensation is needed. When the actual voltage is lower than the nominal voltage, the system considers the power supply conditions to be worse, and the same load change may cause more significant current fluctuations. To avoid misjudgment, the sensitivity of cavitation detection needs to be adjusted. At this time, the voltage compensation coefficient obtained from the table will decrease linearly with the proportion of voltage drop, thereby reducing the dynamic judgment threshold and making the detection more sensitive. Conversely, when the actual voltage is higher than the nominal voltage, the compensation coefficient is usually maintained at 1 or a fixed value, indicating that the system considers the power supply to be sufficient and there is no need to adjust the threshold due to the voltage increase.

[0033] It should be noted that the nominal voltage is the DC power supply voltage that the system is expected to use or is rated to use under normal operating conditions; when constructing the voltage compensation coefficient table, this standard reference voltage value is used as the benchmark; in the automotive environment, the nominal voltage is often set to the typical voltage of the vehicle battery, i.e., 12V.

[0034] The temperature compensation coefficient is obtained by querying a temperature compensation coefficient table preset inside the microcontroller. The temperature compensation coefficient table takes the real-time collected fluid temperature as input. Specifically, the higher the fuel temperature, the easier it is to vaporize and generate bubbles. Therefore, in order to improve the sensitivity of cavitation detection under high temperature conditions so as to detect it in the early stage of bubble generation, the temperature compensation coefficient Kt decreases monotonically with the increase of temperature.

[0035] The dynamic threshold used for comparison in each control cycle is ultimately determined by the product of the base threshold, the voltage compensation coefficient, and the temperature compensation coefficient.

[0036] Among them, based on the identification of two key external variables that affect the sensitivity of cavitation, namely voltage and temperature, the judgment criteria can be adapted to environmental changes through a lookup table compensation method, avoiding the judgment failure that may occur when the fixed threshold is used in winter and summer, or when the voltage is high and low.

[0037] Finally, the microcontroller compares the absolute value of the cross-axis current change rate calculated in real time with the calculated dynamic threshold. When the absolute value of the cross-axis current change rate exceeds the current dynamic threshold in multiple control cycles, the microcontroller determines that the system has entered a state of sudden load unloading caused by cavitation.

[0038] It should be noted that in this embodiment, the system is only confirmed to have entered the load sudden unloading state when the absolute value of the cross-axis current change rate calculated in multiple consecutive control cycles, such as three control cycles, exceeds the current dynamic threshold. This ensures the reliability of the state determination and avoids triggering unnecessary protection actions due to individual interference pulses.

[0039] Furthermore, based on the determination of the sudden unloading state of the load, the current cruise mode is cut off and the preset cavitation protection mode is switched to perform the exhaust operation. Upon determining that the system has entered a state of sudden load unloading, a freeze command is issued simultaneously. Based on the freeze command, the position observer's state update is forcibly stopped, and the error integral accumulation process of the speed regulator and current regulator is simultaneously paused to prevent the observer's calculation from diverging and the regulator from entering saturation due to sudden load changes.

[0040] Specifically, when cavitation is detected, the microcontroller triggers a mode switching command, first cutting off the closed-loop control circuit of the currently running sensorless magnetic field orientation control algorithm, that is, cutting off the cruise mode operation.

[0041] Specifically, the microcontroller stops executing the rotor position and speed estimation program based on the position observer to avoid erroneous estimation results caused by severe distortion of the motor current and voltage signals due to the sudden load change caused by cavitation. At the same time, it stops outputting new torque current commands to the current regulator based on the deviation between the target speed and the actual estimated speed. Instead, the microcontroller calls the preset cavitation protection mode control program from the program memory to perform the exhaust operation.

[0042] While the microcontroller triggers the mode switching command, it simultaneously issues a freeze command. This freeze command is applied to the position observer responsible for estimating the position, the speed regulator responsible for maintaining stable rotation speed, and the current regulator responsible for controlling the magnitude of the motor current, including the direct-axis current regulator and the quadrature-axis current regulator.

[0043] For the position observer, the freeze command causes it to stop receiving new current and voltage samples for state updates, and its internal estimation results are kept at the value at the moment of freezing, that is, the last moment before cavitation occurs, which is relatively reliable.

[0044] It should be noted that in this embodiment, the position observer runs inside the microcontroller to replace the function of a traditional position sensor. Specifically, in the sensorless control system of a brushless motor, the position observer continuously analyzes the voltage and current signals during motor operation and uses a preset mathematical model, such as a sliding mode observer, to perform real-time calculations, thereby estimating the real-time position angle and rotational speed of the rotor.

[0045] For units like speed regulators and current regulators, which typically employ proportional-integral (PI) control, the freeze signal applies to the integrator, stopping it from accumulating the error signal. The integrator's output value remains at the value at the freeze moment. The proportional portion of the regulator is usually unaffected, but because the entire control mode has been switched, its input error has changed after the mode switch.

[0046] Furthermore, in cavitation protection mode, the microcontroller continuously applies a low-frequency drive waveform to the stator coil of the motor, forcing the motor rotor to switch from high-speed closed-loop operation to low-speed synchronous operation. By reducing the rotational speed, the centrifugal force on the fluid in the impeller channel is greatly reduced, allowing the fuel vapor and bubbles generated by cavitation to be smoothly discharged through the exhaust gap or channel inside the pump body by relying on their own buoyancy.

[0047] Specifically, when the microcontroller confirms cavitation based on the aforementioned judgment logic and switches to cavitation protection mode, the microcontroller switches the operating mode of the inverter bridge drive circuit from space vector pulse width modulation (SVM) to a simple voltage / frequency ratio control mode. Specifically, in normal cruise mode, the microcontroller uses SVM to generate drive signals. This technology requires real-time calculation of the rotor's position and speed to synthesize a stable rotating magnetic field, thereby achieving high-speed and precise torque control of the motor. However, under conditions of load abrupt changes and signal distortion caused by cavitation, continuing to rely on this complex closed-loop control requiring precise feedback is neither feasible nor safe. Therefore, the microcontroller switches the drive mode to a simple and reliable open-loop control method, namely the voltage / frequency ratio control mode. In this mode, the control target is no longer precise speed or torque, but rather providing the motor with a stable, low-speed rotating magnetic field, enabling it to operate smoothly at a preset low speed.

[0048] The microcontroller has a pre-stored low-frequency, constant-amplitude three-phase voltage sine wave meter. The frequency of the drive waveform is set in a low range, such as 5Hz to 10Hz. The voltage amplitude is preset according to the set frequency and motor parameters to ensure that sufficient torque can be generated to maintain the low-speed synchronous rotation of the motor, without causing excessive current due to excessive voltage.

[0049] The microcontroller directly generates six PWM signals to drive the six switching transistors of the inverter bridge based on the low-frequency drive waveform, thereby applying a low-speed rotating magnetic field to the three-phase windings of the motor stator. Since the control is open-loop at this time and does not rely on any position feedback, the speed of this rotating magnetic field is entirely determined by the sine wave frequency output by the microcontroller. The permanent magnet rotor in the motor will follow this magnetic field and rotate at the same low speed, with the speed decreasing from thousands of revolutions per minute in cruise mode to hundreds of revolutions per minute or even lower.

[0050] It should be noted that the permanent magnet synchronous motor in the fuel pump rotates on three-phase alternating current, while the vehicle power supply provides direct current. In this embodiment, the inverter bridge is an electronic conversion device that converts DC to AC, converting the DC power from the vehicle power supply into three-phase alternating current to drive the motor, according to the instructions of the microcontroller.

[0051] The inverter bridge circuit consists of three symmetrical branches, corresponding to the U, V, and W phase windings of the motor. Each phase branch consists of two power switching transistors connected in series, namely the upper bridge arm switch and the lower bridge arm switch. There are a total of six switching transistors in the three groups. The six PWM signals output by the microcontroller control the on and off of the six switching transistors respectively. By setting the on and off sequence of the six switches in a very short time, a three-phase AC voltage waveform with controllable frequency, amplitude, and phase can be synthesized at the three terminals of the motor. That is, the AC current that needs to drive the motor to rotate is simulated by digital switching.

[0052] In cruise mode, the microcontroller calculates the switching timing of the six transistors using space vector pulse width modulation to generate a high-speed, smoothly rotating magnetic field that drives the motor to pump oil efficiently. When switching to cavitation protection mode, the microcontroller generates the switching timing according to a preset low-frequency sine wave pattern, causing the six transistors to operate and ultimately applying a fixed low-frequency AC voltage to the motor, forcing the motor to rotate at low speed to achieve the exhaust function.

[0053] Because the centrifugal force exerted by the impeller on the fluid is proportional to the square of the rotational speed, the centrifugal force weakens sharply when the rotational speed decreases from high to low. In a cavitation state, the impeller flow channel is filled with a mixture of air bubbles and fuel. At high speed, the strong centrifugal force throws all substances, including air bubbles, toward the outer edge of the impeller, making it difficult for the air bubbles to aggregate and float. As the rotational speed decreases, the centrifugal force exerted on the air bubbles and pushing them outwards becomes significantly reduced. Under the combined effect of their own buoyancy and slight fluid disturbance, the air bubbles gradually float to the surface and are discharged from the pump chamber through the exhaust channel inherent in the pump body structure design, where they are redissolved or enter the fuel tank, thus eliminating cavitation and restoring the flow channel.

[0054] Furthermore, during operation in cavitation protection mode, the variance value used to reflect the degree of rotor load fluctuation is continuously calculated based on the fluctuation characteristics of real-time phase current. When the variance value is confirmed to meet the preset recovery conditions, the operation of smoothly switching back from cavitation protection mode to cruise mode is performed to restore the high-pressure fuel supply state.

[0055] Specifically, after the microcontroller controls the motor to enter the voltage / frequency ratio control mode, that is, after entering the cavitation protection mode, the load fluctuation will be directly reflected in the current of the drive motor. Therefore, in order to accurately determine whether the bubbles in the pump chamber have been discharged and whether the stable liquid load has been re-established, the microcontroller continuously collects the phase current of the motor through its analog-to-digital converter.

[0056] To quantify the degree of load fluctuation, the microcontroller maintains a sliding time window of fixed length. In each control cycle, the latest current sample value is added to the window, and the oldest value is removed, thereby updating the data sequence within the window.

[0057] It should be noted that in this embodiment, the length of the sliding time window corresponds to 50 milliseconds of sampling data; assuming that the microcontroller samples the current once every millisecond, then the window stores the data obtained from the most recent fifty samples.

[0058] For N instantaneous current values ​​within the current sliding window, the microcontroller first calculates their arithmetic mean; the mean represents the overall central level of the current during this time period. Then, for each current sample value within the window, the square of its deviation from this mean is calculated. Squaring aims to eliminate the mutual cancellation of positive and negative deviations and amplify and quantify any fluctuations deviating from the mean. Finally, the squared differences calculated for all sampled values ​​are summed and divided by the total number of sample points to obtain their mean, i.e., the variance. The magnitude of the variance directly reflects the overall degree to which the current signal deviates from its average level within this window period. Specifically, if the current is stable and each value is close to the mean, the squared difference will be small, and the variance will be small; if the current fluctuates drastically, with some values ​​far above the mean and some far below the mean, the squared difference will be large, and the variance will also be large.

[0059] When a large number of bubbles remain in the impeller flow channel, creating an unstable state of gas and liquid mixture, the random generation, movement, and collapse of these bubbles easily lead to violent and irregular fluctuations in the resistance experienced by the impeller, i.e., the mechanical load on the motor. These violent load fluctuations force large-amplitude, high-frequency pulsations in the motor current, resulting in a large calculated current variance. As low-frequency rotation continues, the bubbles are gradually expelled from the pump chamber through gaps or exhaust channels under buoyancy, and the flow channel is gradually refilled with liquid fuel. When the flow channel is essentially filled with uniform liquid, the load on the impeller tends to stabilize, and the fluctuation amplitude of the drive current significantly decreases. At this point, the current variance calculated based on the latest window data will show a significant decreasing trend.

[0060] It is important to note that during pump assembly, there is a tiny gap between the impeller and the inner wall of the pump casing, which serves as an exhaust channel. During normal high-pressure pumping, the fluid flows tightly against the channel wall under high-speed centrifugal force, so the exhaust effect of this gap is not significant. However, when the motor enters low-speed exhaust mode, the fluid velocity and pressure decrease significantly. Because the density of the air bubbles mixed in the fuel is much smaller than that of the liquid, they will naturally rise under buoyancy, allowing them to move along the gap towards areas with lower pump chamber pressure or those equipped with gas-liquid separation structures. Setting a gap to expel air bubbles is existing technology and will not be elaborated further here.

[0061] The microcontroller compares the continuously calculated variance value with a preset recovery threshold. This recovery threshold is calibrated experimentally and represents the current variance level when the load basically recovers to stability. When the variance value is monitored to be lower than the recovery threshold for M consecutive calculation cycles, the microcontroller determines that the air bubbles in the pump chamber have been basically emptied, the liquid fuel has re-established a stable hydraulic load, and the system meets the conditions for exiting the cavitation protection mode.

[0062] Furthermore, after the microcontroller receives the recovery trigger signal, it reads the real-time electrical angle currently output by the drive module in cavitation protection mode; The real-time electrical angle is used as the initial angle value and directly written into the position observer to achieve forced alignment of the control system angle, thereby eliminating the angle deviation that may occur during mode switching.

[0063] After completing the forced angle alignment, the microcontroller unfroze the position observer, speed regulator, and current regulator, and reactivated the cruise mode. At the moment of reactivation, the microcontroller sets the torque current command output by the current regulator to ramp control mode. Based on the preset ramp time, the control torque current command rises smoothly at a set slope to the target value that matches the current vehicle's required flow rate, in order to suppress the large current surge and torque fluctuation that may occur during mode switching.

[0064] Specifically, when the system meets the conditions for exiting the cavitation protection mode, the microcontroller first performs the synchronization operation of the control angle. In the cavitation protection mode, the drive waveform is a three-phase sinusoidal voltage generated according to a preset frequency and phase, and its electrical angle is a known and continuously changing quantity. When the microcontroller issues a switching command, it reads the real-time electrical angle value output by the voltage / frequency ratio control module at this moment. Subsequently, the microcontroller assigns this electrical angle value to the frozen position observer, overwriting the estimated electrical angle value stored inside it. That is, it forces the angle estimated by the observer to be aligned with the electrical angle of the motor being driven at the moment, eliminating the initial deviation that may be caused by the angle asynchrony during the observation freezing period.

[0065] After the forced angle alignment is completed, the microcontroller immediately unfreezes the various control modules. Specifically, the position observer is reactivated and, starting from the new assigned angle value, resumes receiving real-time current and voltage sampling signals, and calculates and updates the estimated position and speed of the rotor accordingly. At the same time, the integrators inside the speed regulator and current regulator also resume operation, restarting the accumulation calculation of speed and current errors, thereby restoring the dynamic performance of the closed-loop regulation.

[0066] While resuming cruise mode, the microcontroller sets the torque current command output by the current regulator to ramp control mode. Specifically, in cavitation protection mode, the motor only maintains low-speed rotation, requiring very little torque, and the corresponding quadrature-axis current command value is also at a low level. However, after switching back to cruise mode, sufficient torque needs to be provided immediately to drive the impeller to build high voltage to meet the vehicle's fuel demand. If the current command value is directly increased from the previous low level to the high target value required for high voltage, a huge current surge and torque fluctuation will occur.

[0067] To avoid this problem, the microcontroller adopts a ramp control strategy. It records the command value at the moment of switching as the starting point, and then controls the command value to gradually increase the current command at a constant slope according to a preset ramp time until the target value is reached after the power ramp time that the device can tolerate. During this process, the current regulator tracks this smoothly rising command, so that the motor torque increases smoothly, the speed increases smoothly, and finally the normal high-pressure oil supply state is seamlessly transitioned.

[0068] Furthermore, the microcontroller is also equipped with safe operating time monitoring logic to limit the time of exhaust operations: The total time the microcontroller will maintain in cavitation protection mode to perform the venting operation will be strictly limited to a safe time window; When the fuel pump performs low-speed exhaust within this safety window, the system relies on the accumulator system at the front end of the pipeline to release reserve fuel to maintain the system's base fuel pressure.

[0069] Specifically, the microcontroller has an internal timer to monitor the duration of the cavitation protection mode. The timer starts counting from the moment the cavitation protection mode is entered. The microcontroller has a preset maximum allowable safe time window. During the operation of the cavitation protection mode, the microcontroller continuously compares the current duration with the maximum allowable safe time.

[0070] Modern fuel systems typically have an accumulator system, or accumulator, in the line downstream of the fuel pump and upstream of the engine injectors. The accumulator stores fuel at a certain pressure when the fuel pump is working normally. When the fuel pump is temporarily unable to provide high-pressure fuel due to exhaust operation, the fuel stored in the accumulator can be released to maintain the minimum pressure required for engine injection for a short time, preventing the engine from stalling due to momentary fuel cut-off.

[0071] Therefore, as long as the exhaust operation is completed within the maximum permissible safe time and the cruise mode is successfully switched back to restore fuel supply, the normal operation of the engine will not be affected. If the restoration conditions are not met after the exhaust time reaches the maximum permissible safe time, it indicates that the cavitation is extremely serious or the exhaust structure of the pump is abnormal. At this time, the microcontroller will forcibly exit the cavitation protection mode and attempt to restore the cruise mode, while triggering the fault diagnosis flag to prompt the system to perform further checks, but prioritizing ensuring that the fuel supply is not interrupted.

[0072] Example 2 While the above embodiments can prevent lock-up loss by switching modes when cavitation occurs, the conditions for cavitation are closely related to the saturated vapor pressure of the fuel. The viscosity, density, and vaporization critical point of the fuel are completely different in winter and summer, and when using pure gasoline and ethanol gasoline. At the same time, the bus voltage of the system will also fluctuate. Fixed logic judgment rules are prone to misjudgment or omission under extreme high temperature or low pressure. In view of this, this embodiment is further improved based on Embodiment 1.

[0073] Reference Figures 1 to 3 As shown, based on the bus voltage and fluid temperature, a synchronous compensation calculation is performed on the preset basic threshold to construct a dynamic threshold for identifying the cavitation boundary. Obtain the voltage compensation coefficient corresponding to the bus voltage and the temperature compensation coefficient corresponding to the fluid temperature respectively; The voltage compensation coefficient and temperature compensation coefficient are obtained in real time using a one-dimensional linear lookup table method pre-stored in the microcontroller.

[0074] Specifically, in Example 1, a one-dimensional linear lookup table method was used to independently compensate for voltage and temperature. In this example, the influence of voltage and temperature on fuel pump cavitation was further considered, and fuel characteristic parameters, such as fuel type identifiers, were introduced as a third-dimensional compensation factor.

[0075] In practice, the microcontroller has a three-dimensional lookup table pre-stored inside. Its inputs are bus voltage, fluid temperature, and fuel type identifier. The fuel type identifier can be obtained through the vehicle bus or set by the user. The output is a comprehensive compensation coefficient.

[0076] The construction of the three-dimensional lookup table is based on a large amount of test data. The test requires running the fuel pump with different types of fuel under different voltage and temperature conditions, and recording the current change rate characteristic value corresponding to the system starting to exhibit cavitation under each combination. By analyzing this data, the optimal detection threshold for various fuels under different voltage and temperature combinations is calibrated and normalized to a compensation coefficient relative to the base threshold. For example, for ethanol gasoline, due to its high latent heat of vaporization, it may be more difficult to vaporize at high temperatures. Therefore, its temperature compensation coefficient may decay more slowly than that of pure gasoline in the high-temperature region.

[0077] In each control cycle, the microcontroller synchronously collects the current bus voltage and fluid temperature, and obtains the currently set fuel type identifier; then, it uses these three sets of data as coordinates to query the internally stored three-dimensional lookup table, thereby obtaining a comprehensive compensation coefficient that matches the current operating conditions. Furthermore, the lookup table method is upgraded to piecewise linear interpolation or quadratic interpolation to improve the parameter accuracy between table index points; finally, the dynamic threshold used to determine whether cavitation has occurred in real time is obtained by multiplying the basic threshold and the comprehensive compensation coefficient, thereby improving the accuracy of cavitation identification.

[0078] Furthermore, a sliding window of fixed time length is set inside the microcontroller, and the real-time instantaneous values ​​of the phase current of all historical sampling points within the window are obtained in each calculation cycle; Calculate the average of all instantaneous current values ​​within the current sliding window; Calculate the squared difference between the instantaneous current value and the average value within the window, and then take the average of all the squared values ​​as the variance value of the current cycle. Real-time comparison of continuously calculated variance values ​​with pre-stored recovery thresholds; When the impeller channel is still in a gas-liquid mixed state, the load fluctuates violently, and the variance value will remain in the range above the recovery threshold. When the continuously calculated variance value has dropped below the recovery threshold, and this state below the threshold has been stably maintained for more than the preset judgment time, the microcontroller determines that the air bubbles in the pump chamber have been basically emptied.

[0079] Specifically, in Implementation 1, a sliding window with a fixed time length is used. In this implementation, the length of the sliding window is dynamically adjusted according to the actual electrical frequency of the motor in cavitation protection mode. Specifically, when the motor enters cavitation protection mode, the microcontroller will obtain the current voltage frequency value of the driving motor in real time. This frequency determines the time required for the motor current to complete one complete periodic change, i.e., one electrical cycle. The microcontroller then calculates the duration of this electrical cycle and sets the length of the sliding window to the time of K electrical cycles. This ensures that each variance calculation is always based on current data of an integer number of electrical cycles, avoiding the introduction of additional harmonic fluctuations due to the asynchronous relationship between the window length and the electrical cycle. This makes the variance value more purely reflect load fluctuations and improves the accuracy of judging whether the bubble has been emptied.

[0080] Example 1 used a fixed recovery threshold, while in this example, it is dynamically set based on the initial variance value at the moment of entering cavitation protection mode. The microcontroller records the variance value of the first calculation cycle when switching to protection mode, i.e., the initial variance value; the magnitude of this initial value directly reflects the severity of the load disturbance at that time, i.e., the severity of cavitation; subsequently, the system sets the recovery determination threshold to a small proportion of this initial value, such as one-tenth.

[0081] At the same time, the time required to determine that the load has recovered to a stable state is no longer fixed, but is related to the rate at which the initial variance value decreases; for example, the system may require the current variance value to decrease from a high initial level and remain stable below a low threshold for a period of time that is several electrical cycles long before determining recovery.

[0082] This dynamic threshold and judgment duration mechanism allows the recovery conditions to adapt to the severity of the current cavitation event; the larger the initial variance, the more severe the disturbance, which may require stricter recovery criteria or a slightly longer observation time; if the initial disturbance is slight, the recovery conditions are relatively lenient.

[0083] Furthermore, the total time the microcontroller will maintain in cavitation protection mode to perform the venting operation will be strictly limited to a safe time window.

[0084] Specifically, while Implementation 1 uses a fixed safety time window, this implementation upgrades it to an adaptive safety window based on historical cavitation event statistics.

[0085] In practice, the microcontroller's internal non-volatile memory has a recording area, which is used to record the actual exhaust duration of the most recent successful triggering of cavitation protection and the eventual smooth recovery.

[0086] When it is necessary to set a safe time window for the next possible cavitation event, the microcontroller calculates the average and standard deviation of these historical cavitation times.

[0087] The adaptive safety time window for this and subsequent cavitation protection modes is set as follows: Safety time window = average time + safety factor × standard deviation, where the safety factor is a preset number greater than zero; this is to set the safety time window within a reasonable range.

[0088] At the same time, the system still sets an absolute time limit as a final guarantee. The calculated adaptive safety window value must never exceed this absolute time limit, thus ensuring that there is a final time constraint in any extreme case.

[0089] Through the above operations, when the system is in a condition that is prone to cavitation but usually exhausts air quickly, the safety window will be shorter, which is conducive to quickly attempting recovery; when the system is in a condition that exhausts air slowly, the safety window will be automatically extended, giving more sufficient time to exhaust air and reducing the risk of forced recovery due to insufficient time.

[0090] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A brushless fuel pump system, characterized in that, Includes a microcontroller that executes the following control logic: Obtain the real-time phase current of the motor, the bus voltage of the system, and the fluid temperature; Based on bus voltage and fluid temperature, synchronous compensation calculations are performed on preset basic thresholds to construct dynamic thresholds for identifying cavitation boundaries. Extract the instantaneous rate of change of the real-time phase current and compare its absolute value with the dynamic threshold in real time. When the absolute value continues to exceed the dynamic threshold, the system is determined to enter the state of sudden load unloading caused by cavitation. Based on the determination of the sudden unloading state of the load, the current cruise mode is cut off and the preset cavitation protection mode is switched to perform the exhaust operation. During operation in cavitation protection mode, the variance value used to reflect the degree of rotor load fluctuation is continuously calculated based on the fluctuation characteristics of real-time phase current. When the variance value is confirmed to meet the preset recovery conditions, the operation of smoothly switching back from cavitation protection mode to cruise mode is performed to restore the high-pressure fuel supply state. The step of cutting off the current cruise mode further includes freezing the internal computing state of the microcontroller: Upon determining that the system has entered a state of sudden load unloading, a freeze command is issued simultaneously. Based on the freeze command, the position observer's state update is forcibly stopped, and the error integral accumulation process of the speed regulator and current regulator is simultaneously paused to prevent the observer's calculation from diverging and the regulator from entering saturation due to sudden load changes.

2. The brushless fuel pump system according to claim 1, characterized in that, The steps for constructing a dynamic threshold for identifying cavitation boundaries specifically include: The voltage compensation coefficient corresponding to the bus voltage and the temperature compensation coefficient corresponding to the fluid temperature are obtained respectively; the voltage compensation coefficient and the temperature compensation coefficient are obtained in real time through a one-dimensional linear lookup table method pre-stored in the microcontroller. The current dynamic threshold is obtained by correlating the basic threshold with the voltage compensation coefficient and the temperature compensation coefficient respectively. The temperature compensation coefficient is set to decrease monotonically as the fluid temperature increases, so as to reduce the dynamic threshold and improve the cavitation detection sensitivity when the fuel is at a high temperature and close to the vaporization critical point; the voltage compensation coefficient is set to decrease synchronously with the bus voltage drop ratio, so as to eliminate the interference caused by normal voltage fluctuations on the current change rate.

3. The brushless fuel pump system according to claim 1, characterized in that, The step of switching to the preset cavitation protection mode to perform the exhaust operation specifically includes: In cavitation protection mode, the microcontroller continuously applies a low-frequency drive waveform to the stator coil of the motor, forcing the motor rotor to switch from high-speed closed-loop operation to low-speed synchronous operation. By reducing the rotational speed, the centrifugal force on the fluid in the impeller channel is reduced, allowing the fuel vapor and bubbles generated by cavitation to be smoothly discharged through the exhaust gap or channel inside the pump body by relying on their own buoyancy.

4. The brushless fuel pump system according to claim 1, characterized in that, The steps for calculating the variance, and the specific algorithm, include: A sliding window of fixed time length is set inside the microcontroller, and the real-time instantaneous value of the phase current of all historical sampling points within the window is obtained in each calculation cycle; Calculate the average of all instantaneous current values ​​within the current sliding window; Calculate the squared difference between the instantaneous current value and the average value within the window, and then take the average of all the squared values ​​as the variance value of the current cycle.

5. The brushless fuel pump system according to claim 4, characterized in that, The evaluation logic for confirming that the variance value meets the preset recovery conditions is as follows: Real-time comparison of continuously calculated variance values ​​with pre-stored recovery thresholds; When the impeller channel is still in a gas-liquid mixed state, the load fluctuates violently, and the variance value will remain in the range above the recovery threshold. When the continuously calculated variance value has dropped below the recovery threshold, and the state below the threshold remains stable for more than the preset judgment time, the microcontroller determines that the air bubbles in the pump chamber have been emptied and the liquid fuel has re-established a stable load, and outputs a trigger signal that meets the recovery conditions.

6. The brushless fuel pump system according to claim 1, characterized in that, The smooth switching from cavitation protection mode back to cruise mode specifically includes angle forced alignment control: After the microcontroller receives the recovery trigger signal, it reads the real-time electrical angle currently output by the drive module in cavitation protection mode; The real-time electrical angle is used as the initial angle value and directly written into the position observer to achieve forced alignment of the control system angle, thereby eliminating the angle deviation caused by the instant of mode switching.

7. The brushless fuel pump system according to claim 6, characterized in that, Smooth switching operation also includes torque relief control: After completing the forced angle alignment, the microcontroller unfroze the position observer, speed regulator, and current regulator, and reactivated the cruise mode. At the moment of reactivation, the microcontroller sets the torque current command output by the current regulator to ramp control mode. Based on the preset ramp time, the control torque current command rises smoothly at a set slope to the target value that matches the current vehicle's required flow rate, in order to suppress the large current surge and torque fluctuations generated during mode switching.

8. The brushless fuel pump system according to claim 7, characterized in that, The microcontroller also has internal safe runtime monitoring logic to limit the time of exhaust operations: The total time the microcontroller will maintain in cavitation protection mode to perform the venting operation will be strictly limited to a safe time window; When the fuel pump is venting at low speed within the safety window, the system relies on the accumulator system at the front end of the pipeline to release the reserve fuel to maintain the system's base oil pressure, ensuring that the fuel supply is not interrupted during the cavitation venting phase when the pump body is temporarily unable to supply fuel at high pressure, thus avoiding engine shutdown.

9. The brushless fuel pump system according to claim 1, characterized in that, The current fluid temperature is obtained by collecting the signal from the thermistor installed in the fuel line.

Images