Brushless motor control circuit for high temperature dual control water distributor

Abstract

The present application relates to the field of intelligent control, and specifically discloses a brushless motor control circuit for a high-temperature dual-control water distributor, which includes: an MCU control circuit, a gate drive circuit, a current detection circuit, a three-phase H-bridge circuit and a high-temperature DC brushless motor; wherein, the current detection circuit is electrically connected to the three-phase H-bridge circuit to collect real-time current, and inputs the real-time current to the MCU control circuit, and the MCU control circuit determines the rotor position based on the real-time current, and outputs a PWM control signal based on the rotor position; wherein, the PWM control signal output by the MCU control circuit drives the high-temperature DC brushless motor through the gate drive circuit and the three-phase H-bridge circuit.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically, to a brushless motor control circuit for a high-temperature dual-control water distributor. Background Art

[0002] In the field of oil extraction, intelligent stratified water injection technology plays a vital role in improving crude oil recovery rates. The performance of the high-temperature dual-control water distributor, as the key core equipment of this technology, directly determines the effect of stratified water injection. This type of water distributor usually works in harsh environments such as high temperature and high pressure underground. It is necessary to precisely control the opening of the internal water nozzle to adjust the water injection volume of each oil layer. This places extremely high demands on the reliability and accuracy of its drive motor and its control circuit. Brushless DC motors are ideal for such applications due to their high efficiency, long life, and ease of precise control. Therefore, the development of a brushless motor control circuit for high-temperature dual-control water distributors that can adapt to high-temperature environments, achieve precise control, and ensure long-term stable operation is of great significance to improving the overall performance of intelligent water injection equipment and ensuring efficient oil field extraction.

[0003] Traditional brushless motor control schemes typically rely on Hall-effect sensors to detect rotor position. However, Hall-effect sensors significantly degrade reliability in high-temperature environments (such as 150°C or higher underground), becoming prone to failure and increasing system complexity and cost. Therefore, sensorless control technology—which estimates rotor position by detecting electrical signals (such as back-EMF and phase current) during motor operation rather than relying on physical position sensors—has become a research hotspot for high-temperature applications. However, existing sensorless control methods based on back-EMF often suffer from inaccurate rotor position estimation during motor startup, low-speed operation, or sudden load changes due to weak or difficult-to-detect back-EMF signals. This can affect control performance and even lead to startup failure. In particular, in high-temperature dual-control water distributor applications, where precise control of start-up, shutdown, and operation is required, accurately acquiring rotor position information within a wide dynamic range and under harsh operating conditions is a core challenge facing sensorless control technology.

[0004] Therefore, an optimized brushless motor control circuit for a high-temperature dual-control water distributor is desired. Summary of the Invention

[0005] In order to solve the above-mentioned technical problems, the present application is proposed. The embodiment of the present application provides a brushless motor control circuit for a high-temperature dual-control water distributor, which does not rely on traditional physical position sensors, but collects motor current signals (including A and B phase currents) and DC bus voltage signals in real time through a current detection circuit, extracts key features from these signals, and further performs complex response reasoning based on implicit analysis of signal features on the extracted signal features to estimate the electrical angle of the motor rotor. Compared with traditional sensorless control methods that rely on physical sensors or are based on back electromotive force, this technical solution can help improve the accuracy of rotor position estimation and the overall robustness of the system in high-temperature harsh environments, especially in challenging working conditions such as motor starting and low-speed operation, thereby showing superior performance, thereby ensuring precise control and long-term stable operation of the high-temperature dual-control water distributor.

[0006] According to one aspect of the present application, a brushless motor control circuit for a high-temperature dual-control water distributor is provided, which includes: an MCU control circuit, a gate drive circuit, a current detection circuit, a three-phase H-bridge circuit and a high-temperature DC brushless motor; wherein, the current detection circuit is electrically connected to the three-phase H-bridge circuit to collect real-time current, and inputs the real-time current to the MCU control circuit, the MCU control circuit determines the rotor position based on the real-time current, and outputs a PWM control signal based on the rotor position; wherein, the PWM control signal output by the MCU control circuit drives the high-temperature DC brushless motor through the gate drive circuit and the three-phase H-bridge circuit.

[0007] In the above-mentioned brushless motor control circuit for the high-temperature dual-control water distributor, during the rotor alignment stage, the MCU control circuit controls the gate drive circuit and the three-phase H-bridge circuit to apply a fixed DC current to phases A and B of the high-temperature DC brushless motor.

[0008] In the above-mentioned brushless motor control circuit for the high-temperature dual-control water distributor, in the open-loop acceleration stage, the MCU control circuit applies a PWM signal to the gate drive circuit and the three-phase H-bridge circuit to generate a continuously accelerating magnetic field in the stator of the high-temperature DC brushless motor, wherein the rotor of the high-temperature DC brushless motor gradually increases its speed from zero to the minimum rate under the action of the continuously accelerating magnetic field.

[0009] In the above-mentioned brushless motor control circuit for high-temperature dual-control water distributor, in the closed-loop control stage, the MCU control circuit collects three-phase current in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase current, and then outputs a PWM signal based on the rotor electrical angle.

[0010] In the above-mentioned brushless motor control circuit for high-temperature dual-control water distributor, the MCU control circuit collects three-phase current in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase current, including the steps of: obtaining an A-phase current analog signal and a B-phase current analog signal; obtaining a DC bus voltage signal; extracting current signal features from the A-phase current analog signal and the B-phase current analog signal to obtain an A-phase current analog signal feature coding vector and a B-phase current analog signal feature coding vector; extracting voltage signal features from the DC bus voltage signal to obtain a DC bus voltage signal feature coding vector; based on the current analog signal joint coding features between the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector and the DC bus voltage signal feature coding vector, performing response inference based on implicit analysis of signal features to obtain a voltage-current response inference coding vector; and feature decoding the voltage-current response inference coding vector to obtain an estimated value of the rotor electrical angle.

[0011] Compared with the existing technology, the brushless motor control circuit for high-temperature dual-control water distributors provided in this application does not rely on traditional physical position sensors. Instead, it uses a current detection circuit to collect motor current signals (including A and B phase currents) and DC bus voltage signals in real time, extract key features from these signals, and further perform complex response reasoning based on implicit analysis of signal features on the extracted signal features to estimate the electrical angle of the motor rotor. Compared with traditional sensorless control methods that rely on physical sensors or back electromotive force, this technical solution can help improve the accuracy of rotor position estimation and the overall robustness of the system in harsh high-temperature environments, especially in challenging working conditions such as motor starting and low-speed operation, thereby ensuring precise control and long-term stable operation of high-temperature dual-control water distributors.

[0012] Furthermore, response inference based on implicit analysis of signal features can understand and quantify the complex interactions between current and voltage characteristics, overcoming the limitations of traditional methods under complex operating conditions and signal interference. Specifically, the input feature vectors (i.e., the joint encoding vector of the AB phase current analog signal features and the DC bus voltage signal feature encoding vector) are ordered to focus on the intrinsic strength of the features. Equal-granularity feature segmentation is then performed to enable detailed analysis of local features. Subsequently, the core transfer response inference unit performs deep interaction modeling on the local feature segments of each pair of AB phase current analog signal and DC bus voltage signal, capturing their mutual influence, response patterns, or information transfer mechanisms. This unit retains rich details of the local voltage-current signal interactions in a matrix format. Finally, a sequence transfer and aggregation module integrates this local voltage-current signal interaction information to capture dependencies and overall trends across different feature strength ranges. The ultimate goal is to generate a concise and information-rich voltage-current response inference encoding vector. This encoding vector characterizes the dynamic response relationship between the applied voltage and the generated current in a highly generalized manner that is highly discriminative of rotor position. By deeply exploring and understanding the complex coupling relationship between voltage and current, the control system can more accurately identify the subtle electrical signal characteristics caused by rotor position changes, and maintain high estimation reliability even under harsh operating conditions such as high temperature, high noise, motor parameter drift, or severe load fluctuations. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0014] Figure 1 Schematic diagram of a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application;

[0015] Figure 2 This is a schematic diagram of an MCU control circuit for a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application;

[0016] Figure 3 1 is a schematic diagram of a gate drive circuit of a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application;

[0017] Figure 4 This is a schematic diagram of a current detection circuit of a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application;

[0018] Figure 5 This is a schematic diagram of a 3-phase H-bridge circuit and a high-temperature brushless DC motor for a high-temperature dual-control water distributor according to an embodiment of the present application;

[0019] Figure 6 This is a flow chart of the MCU control circuit of the brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application, which collects three-phase currents in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase currents;

[0020] Figure 7 This is a data flow diagram of the MCU control circuit of the brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application, in which the MCU control circuit collects three-phase current in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase current;

[0021] Figure 8 A flowchart for performing response inference based on implicit analysis of signal characteristics to obtain a voltage-current response inference coding vector based on a joint current analog signal coding feature between the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector, and the DC bus voltage signal feature coding vector, for a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application;

[0022] Figure 9 A flowchart of performing response inference based on implicit analysis of signal characteristics on the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector of the brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application to obtain the voltage-current response inference coding vector. DETAILED DESCRIPTION

[0023] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0024] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0025] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0026] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0027] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0028] In the technical solution of the present application, a brushless motor control circuit for a high-temperature dual-control water distributor is proposed. Figure 1 FIG. 1 is a schematic diagram showing the principle of a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application. Figure 1 As shown, the brushless motor control circuit for the high-temperature dual-control water distributor includes: an MCU control circuit, a gate drive circuit, a current detection circuit, a 3-phase H-bridge circuit and a high-temperature DC brushless motor; wherein, the current detection circuit is electrically connected to the 3-phase H-bridge circuit to collect real-time current, and inputs the real-time current to the MCU control circuit, and the MCU control circuit determines the rotor position based on the real-time current, and outputs a PWM control signal based on the rotor position; wherein, the PWM control signal output by the MCU control circuit drives the high-temperature DC brushless motor through the gate drive circuit and the 3-phase H-bridge circuit.

[0029] Figure 2 FIG. 1 is a schematic diagram of an MCU control circuit for a brushless motor control circuit of a high-temperature dual-control water distributor according to an embodiment of the present application. Figure 2 As shown, the MCU control circuit includes: an external crystal oscillator circuit, an external power supply filter circuit, a program download and debug interface circuit, and a chip dsPIC33CK256MP306T-H / PT; the chip dsPIC33CK256MP306T-H / PT also includes an ADC analog-to-digital acquisition unit, a PWM pulse width modulation unit, an SPI communication unit, a UART communication unit, and a GPIO port configuration unit.

[0030] Figure 2Pins 36, 37, 39, and 38 of the MCU chip dsPIC33CK256MP306T-H / PT are SPI communication module pins, which are the clock signal SCK, data input SDI, data output SDO, and chip select signal SPI_CS, respectively. This communication interface is connected to the main control circuit of the high-temperature dual-control water distributor in this solution, completing data communication with the instrument main control board, realizing the adjustment of the water tap opening and the transmission of data such as the opening and motor current. Figure 2 The capacitors C12 and C17 and the crystal oscillator X1 together form a crystal oscillator circuit, which is connected to the 28th and 30th pins of the MCU chip dsPIC33CK256MP306T-H / PT. Figure 2 The capacitors C3, C4 and C5 together form the filter circuit of the ADC power supply.

[0031] Figure 3 FIG. 1 is a schematic diagram of a gate drive circuit of a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application. Figure 2 and Figure 3 As shown, Figure 2 Pins 1, 2, 63, 64, 61, and 62 of the MCU chip dsPIC33CK256MP306T-H / PT are the PWM signals output by the MCU chip. Pins 1 and 2 output the high-end and low-end PWM control signals of phase A respectively, pins 63 and 64 output the high-end and low-end PWM control signals of phase B respectively, and pins 61 and 62 output the high-end and low-end PWM control signals of phase C respectively; these three PWM signals are respectively connected to the Figure 3 The 6, 7, 4, 5 and 3, 2 pins of the middle gate control chip are connected.

[0032] like Figure 2 and Figure 3 As shown, Figure 2 The 5th pin of the MCU chip dsPIC33CK256MP306T-H / PT is used as the enable output signal and Figure 3 The MCP8022T-3315H / NHXVAO enable input pin 8 is connected.

[0033] like Figure 2 and Figure 3 As shown, Figure 2 The 3rd pin of the MCU chip dsPIC33CK256MP306T-H / PT is used as the wake-up output signal. Figure 3 The MCP8022T-3315H / NHXVAO wake-up input pin 20 is connected.

[0034] like Figure 2 and Figure 3 As shown, Figure 3 The fault output pin 9 of MCP8022T-3315H / NHXVAO is connected to Figure 2 The fault input pin 4 of the MCU chip dsPIC33CK256MP306T-H / PT is connected to the MCU chip dsPIC33CK256MP306T-H / PT.

[0035] like Figure 2 and Figure 3 As shown, Figure 2 Pin 34 of the MCU chip dsPIC33CK256MP306T-H / PT is used as the transmit pin of the UART communication module and is connected to the left pin of R0. The right pin of R0 is then connected to pin 35 of the MCU chip dsPIC33CK256MP306T-H / PT (the receive pin of the UART communication module). Pin 35 of the MCU chip dsPIC33CK256MP306T-H / PT is connected to Figure 3 The communication pin 1 of MCP8022T-3315H / NHXVAO is connected, so the communication between the MCU chip dsPIC33CK256MP306T-H / PT and the gate control chip MCP8022T-3315H / NHXVAO is half-duplex, and sending and receiving data are performed alternately.

[0036] like Figure 2 and Figure 3 As shown, Figure 3 Pins 17, 14, and 11 of MCP8022T-3315H / NHXVAO are the output pins of the built-in operational amplifier of the chip, respectively. Figure 2 The 33, 17, and 18 pins of the MCU chip dsPIC33CK256MP306T-H / PT are connected. These three pins are input channels 2, 3, and 4 of the internal ADC analog-to-digital conversion module of the MCU chip dsPIC33CK256MP306T-H / PT.

[0037] Figure 4 FIG. 1 is a schematic diagram of a current detection circuit of a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application. Figure 3 and Figure 4 As shown, Figure 3 Pins 17, 14, and 11 of MCP8022T-3315H / NHXVAO are the signal output pins of the built-in operational amplifier of the chip. Figure 4 The right end pins of R21 and C25 are connected together. Figure 4 The right end pins of R22 and C26 are connected together. Figure 4 The right end pins of R31 and C33 are connected together.

[0038] like Figure 3 and Figure 4 As shown, Figure 3Pins 19 and 18 of MCP8022T-3315H / NHXVAO are the signal input pins of the built-in operational amplifier of the chip, and pin 19 is connected to Figure 4 The right end pin of the middle resistor R14, pin 18 is connected to Figure 4 The right pin of R18.

[0039] like Figure 3 and Figure 4 As shown, Figure 3 Pins 16 and 15 of MCP8022T-3315H / NHXVAO are the signal input pins of the built-in operational amplifier of the chip, and pin 16 is connected to Figure 4 The right end pin of the middle resistor R16, pin 15 is connected to Figure 4 The right pin of R20.

[0040] like Figure 3 and Figure 4 As shown, Figure 3 Pins 13 and 12 of MCP8022T-3315H / NHXVAO are the signal input pins of the built-in operational amplifier of the chip, and pin 13 is connected to Figure 4 The right end pin of the middle resistor R25, pin 12 is connected to Figure 4 The right pin of R30.

[0041] like Figure 3 As shown in FIG, diodes D1, D2, D3, D4, capacitors C1, C2, C9, C10, C11, resistors R5, R6, R7 and the bootstrap module inside the gate control chip MCP8022T-3315H / NHXVAO together form a bootstrap circuit so that the chip MCP8022T-3315H / NHXVAO can drive the high-end N-channel MOS tube.

[0042] Figure 5 The schematic diagram of the 3-phase H-bridge circuit and high-temperature brushless DC motor for high-temperature dual-control water distributor according to the embodiment of the present application is shown in FIG. Figure 3 and Figure 5 As shown, Figure 3 The 29, 26, and 23 pins of the middle gate control chip MCP8022T-3315H / NHXVAO are Figure 5 The gate drive output signal of the mid- and high-end MOS tube, where pin 29 is connected to the Figure 5 Pin 1 (gate) and pin 26 of the MOS tube Q11 are connected to the Figure 5 Pin 1 (gate) and pin 23 of the MOS tube Q21 are connected to the Figure 5 Pin 1 (gate) of the middle MOS tube Q31.

[0043] like Figure 3 and Figure 5 As shown, Figure 3 Pins 33, 32, and 31 of the middle gate control chip MCP8022T-3315H / NHXVAO are Figure 5 The gate drive output signal of the low-end MOS tube, where pin 33 is connected to the Figure 5 Pin 1 (gate) and pin 32 of the MOS tube Q12 are connected to the Figure 5 Pin 1 (gate) and pin 31 of the MOS tube Q22 are connected to the Figure 5 Pin 1 (gate) of the middle MOS tube Q32.

[0044] like Figure 3 and Figure 5 As shown, Figure 3 The 28, 25, and 22 pins of the middle gate control chip MCP8022T-3315H / NHXVAO are Figure 5 The midpoint bias signal of the three-phase H-bridge circuit A, B, and C. Pin 28 is connected to the Figure 5 Pin 3 (source) of MOS tube Q11 and pin 2 (drain) of Q12, and pin 25 are connected to Figure 5 Pin 3 (source) of MOS tube Q21 and pin 2 (drain) of Q22, and pin 22 are connected to Figure 5 Pin 3 (source) of the MOS tube Q31 and pin 2 (drain) of Q32.

[0045] like Figure 4 and Figure 5 As shown, Figure 4 Resistors R11, R13, R14, R17, R18, R21, capacitors C21, C22, C25 and Figure 5 R26 in the circuit forms the motor A phase current acquisition and amplification circuit; Figure 4 Resistors R12, R15, R16, R19, R20, R27, capacitors C23, C24, C26 and Figure 5 R27 in the circuit forms the motor B phase current acquisition and amplification circuit; Figure 4 Resistors R23, R24, R25, R29, R30, R31, capacitors C31, C32, C33 and Figure 5 R26 in the circuit constitutes the motor C phase current acquisition and amplification circuit.

[0046] like Figure 5As shown, MOS tubes Q11, Q12 and R26 form the A-phase bridge arm circuit of the three-phase H-bridge circuit; MOS tubes Q21, Q22 and R27 form the B-phase bridge arm circuit of the three-phase H-bridge circuit; MOS tubes Q31, Q32 and R28 form the C-phase bridge arm circuit of the three-phase H-bridge circuit; M1 is a high-temperature DC three-phase brushless motor, and its three-phase windings are respectively connected to the midpoint of the three-phase H-bridge circuit.

[0047] Specifically, Figure 1 The hardware circuit operation of a brushless motor control circuit for a high-temperature dual-control water distributor is described as follows: The dsPIC33CK256MP306T-H / PT MCU, serving as the core of the control circuit, outputs six PWM control signals of corresponding frequencies according to a fixed software algorithm and timing sequence. These PWM signals are then fed to the gate driver chip MCP8022T-3315H / NHXVAO. Ultimately, after optimization of the chip's internal logic circuitry, they are output to the gates of the MOS transistors in the three-phase H-bridge circuit to drive the motor. Simultaneously, a three-phase current sampling circuit collects the real-time current of each bridge arm and inputs it to the ADC built into the dsPIC33CK256MP306T-H / PT MCU. Based on the currents of each phase, a software algorithm is used to determine the rotor position. Based on this rotor position, PWM signals are then output for the corresponding bridge arm to complete the motor commutation.

[0048] like Figure 1As shown, this brushless motor control circuit for a high-temperature dual-control water distributor utilizes FOC (Field Oriented Control) to drive and control the brushless DC motor through hardware circuitry and software algorithms. Field-oriented control requires precise rotor position information, but this solution employs a sensorless brushless motor drive approach, lacking a rotor position sensor. This requires other methods for determining rotor position. This solution employs a mathematical model algorithm to determine rotor position. This involves real-time acquisition of the three-phase currents of the H-bridge circuit and a series of mathematical operations, such as the Clarke transform, the Park transform, and their inverse transforms. The primary challenge in sensorless control of a brushless DC motor lies in determining rotor position during startup. The underlying software of this solution primarily operates in three phases: During the rotor alignment phase, the MCU control circuit controls the gate drive circuit and the three-phase H-bridge circuit to apply a fixed DC current to phases A and B of the high-temperature brushless DC motor. During the open-loop acceleration phase, the MCU control circuit applies PWM signals to the gate drive circuit and the three-phase H-bridge circuit to generate a continuously accelerating magnetic field in the stator of the high-temperature brushless DC motor. Under the influence of the continuously accelerating magnetic field, the rotor of the high-temperature brushless DC motor gradually increases its speed from zero to a minimum speed. During the closed-loop control phase, the MCU control circuit collects three-phase currents in real time through the current detection circuit, estimates the rotor electrical angle based on the three-phase currents, and then outputs a PWM signal based on the rotor electrical angle.

[0049] Accordingly, in order to overcome the limitations of traditional control methods and adapt to the stringent requirements of high-temperature dual-control water distributors for motor control, this technical solution proposes real-time acquisition of three-phase current through a current detection circuit and estimation of the rotor electrical angle based on the three-phase current. The current signal is used to estimate the rotor position because the current signal runs through the entire process of motor operation. Even at low speed or startup, the current signal still exists and can be measured. Compared with the back-electromotive force method, it has a wider effective operating range, and components such as current sampling resistors are more stable and reliable at high temperatures than Hall sensors. Furthermore, in order to improve the accuracy and robustness of rotor position estimation based solely on current signals, especially under signal noise interference, motor parameter changes, or complex operating conditions, this solution introduces more complex signal processing and intelligent reasoning mechanisms. Specifically, by extracting the features of the A and B phase current analog signals and the DC bus voltage signal, and performing response reasoning based on implicit analysis of these signal encoding features, it is possible to more deeply explore the complex nonlinear mapping relationship between the voltage and current signals, thereby more accurately inverting the rotor electrical angle. Compared with traditional filtering and simple mathematical model calculations, this advanced signal processing and inference method can better adapt to the dynamic changes and noise interference of the signal, and extract deep features that are more strongly correlated with the rotor position, thereby significantly improving the rotor position estimation accuracy of the brushless motor sensorless control in harsh environments such as high temperature, as well as the overall control performance and reliability of the system.

[0050] Figure 6 This is a flow chart of the MCU control circuit of the brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application, which collects three-phase current in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase current. Figure 7 This is a data flow diagram of the MCU control circuit of the brushless motor control circuit for the high-temperature dual-control water distributor according to the embodiment of the present application, which collects three-phase current in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase current. Figure 6 and Figure 7As shown, according to the embodiment of the present application, the MCU control circuit of the brushless motor control circuit for the high-temperature dual-control water distributor collects three-phase current in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase current, including the steps of: S100, obtaining the A-phase current analog signal and the B-phase current analog signal; S200, obtaining the DC bus voltage signal; S300, extracting the current signal features from the A-phase current analog signal and the B-phase current analog signal to obtain the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector; S400, extracting the voltage signal features from the DC bus voltage signal to obtain the DC bus voltage signal feature coding vector; S500, based on the current analog signal joint coding features between the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector and the DC bus voltage signal feature coding vector, performing response inference based on implicit analysis of signal features to obtain a voltage-current response inference coding vector; S600, feature decoding the voltage-current response inference coding vector to obtain an estimated value of the rotor electrical angle.

[0051] Specifically, in steps S100 and S200, analog signals for phase A and phase B current are acquired, along with a DC bus voltage signal. It should be understood that without the use of physical position sensors (such as Hall sensors, which are particularly unreliable in high-temperature environments), the motor's real-time operating status, including its rotor position and speed, must be determined by analyzing and inferring the motor's own electrical signals. The three-phase current (the third phase can be inferred by measuring any two phases, typically measuring the currents of phases A and B) directly reflects the electromagnetic effect of applied voltage in the motor windings and the modulation of current by the back EMF generated by the motor's rotation. It is one of the most direct and fully measurable signals containing rotor position information, and is particularly easier to acquire and process than the back EMF signal during startup and low speeds. Furthermore, the DC bus voltage is the energy source driving the three-phase H-bridge, and its value changes directly affect the voltage level actually applied to the motor windings via PWM control. Therefore, acquiring the DC bus voltage signal provides essential reference information for accurately interpreting current response and compensating for the impact of power supply fluctuations on control effectiveness. Comprehensive analysis of the dynamic characteristics and interrelationships of these real-time voltage and current signals is the fundamental basis for high-precision rotor electrical angle estimation, especially in applications that need to overcome high-temperature environmental interference and achieve stable control in the full speed range.

[0052] Specifically, in one example of the present application, the steps for acquiring analog current signals for phase A and phase B, and obtaining a DC bus voltage signal, are as follows: First, current sensing elements are placed at appropriate locations within the three-phase H-bridge circuit to sample the current flowing through the phase A and phase B windings. Common current sensing elements include low-resistance sampling resistors (shuntresistors) or Hall-effect-based current sensors. For example, a sampling resistor can be connected in series with the circuit path connected to the lower transistor of the bridge arm (usually the ground terminal). When current flows through the resistor, a voltage drop proportional to the current is generated across it. Next, because the voltage drop across the current sampling resistor is often very small and may be affected by common-mode voltage or noise, the acquired analog current signal requires signal conditioning. This typically involves amplifying the weak voltage signal using a differential amplifier or a dedicated current sampling amplifier, and performing filtering (such as low-pass filtering) to suppress high-frequency switching noise and external interference, ultimately conditioning the signal to the input voltage range and signal quality requirements of the MCU's internal ADC (analog-to-digital converter). To obtain the DC bus voltage signal, a resistor divider circuit is typically used to proportionally attenuate the higher DC bus voltage to within the MCU's ADC input range. Appropriate filtering may also be required to eliminate the effects of power supply ripple or transient fluctuations on the voltage measurement.

[0053] The conditioned and filtered analog signals of phase A current, phase B current, and DC bus voltage are then input into the ADC channels of the MCU control circuit. The MCU samples these analog signals periodically, typically in synchronization with the PWM switching signal, according to a preset sampling strategy, converting the analog quantities into digital quantities. The choice of sampling moment is crucial; switching transients should be avoided, and sampling should be performed at a time when the current waveform is stable and representative. Finally, the MCU obtains the digitized values ​​of phase A current, phase B current, and DC bus voltage. These values ​​serve as the raw digital inputs for subsequent signal feature extraction and response inference based on implicit analysis of signal features, thereby initiating the rotor electrical angle estimation process. Through this series of standardized acquisition and processing processes, the control system can reliably and real-time acquire the key electrical signal information required to drive the brushless motor.

[0054] Specifically, in step S300, current signal features are extracted from the A-phase current analog signal and the B-phase current analog signal to obtain a phase A current analog signal feature encoding vector and a phase B current analog signal feature encoding vector. It should be understood that in a brushless motor control circuit for a high-temperature dual-control water distributor, although the original current analog signal contains information about the motor's operating status, this information is often mixed and inefficient to directly utilize. In particular, under harsh operating conditions such as high temperature, high dynamics, and noise, the signal quality is further degraded. Therefore, in the technical solution of the present application, current signal features are further extracted from the A-phase current analog signal and the B-phase current analog signal to obtain a phase A current analog signal feature encoding vector and a phase B current analog signal feature encoding vector. In particular, in a specific example of the present application, a signal feature extractor based on a dilated convolutional neural network model can be used to extract current signal features. This allows key information in the original current analog signal that is strongly correlated with rotor position and insensitive to interference to be expressed in a more compact and robust form, eliminating redundancy and noise components to provide high-quality input for subsequent processing. That is, feature extraction of current signals aims to convert the original analog signal, which may contain noise and redundant information, into a structured digital representation with higher information density that is easier for machine learning models to understand and process.

[0055] Specifically, in step S400, voltage signal features are extracted from the DC bus voltage signal to obtain a DC bus voltage signal feature encoding vector. It should be understood that since the DC bus voltage signal is not only the direct source of energy for the drive motor, its own fluctuations and characteristics also indirectly reflect the motor load condition, power supply quality, and the interaction relationship with the current. The original bus voltage signal may contain power supply noise, ripple, and dynamic fluctuations caused by load changes. Directly using these unprocessed signals is difficult to accurately reveal the deep correlation between them and the rotor position, and may even introduce interference. Therefore, voltage signal features are further extracted from the DC bus voltage signal to obtain a DC bus voltage signal feature encoding vector. In particular, in a specific example of the present application, a signal feature extractor based on a void convolutional neural network model can be used to extract voltage signal features, thereby extracting key information valuable for rotor position estimation from the original bus voltage signal, filtering out irrelevant noise and redundant components, and forming a more stable and representative signal representation. The purpose of this step is to convert the raw bus voltage signal into a structured feature encoding vector for effective fusion and processing with the current feature encoding vectors extracted from the A and B phase currents. By combining the voltage and current features, the complex, nonlinear dynamic relationship between applied voltage and current response under varying power and load conditions can be more comprehensively learned and understood, leading to a more accurate inversion of the motor rotor's real-time electrical angle.

[0056] Specifically, in step S500, based on the joint coding features of the current analog signal between the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector and the DC bus voltage signal feature coding vector, response inference based on implicit analysis of signal features is performed to obtain a voltage-current response inference coding vector. Figure 8 A flowchart of performing response reasoning based on implicit analysis of signal characteristics to obtain a voltage-current response reasoning coding vector based on the current analog signal joint coding characteristics between the A-phase current analog signal characteristic coding vector and the B-phase current analog signal characteristic coding vector and the DC bus voltage signal characteristic coding vector for a brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application. Figure 8 As shown, according to the brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application, step S500 includes: S510, combining the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector to obtain an AB-phase current analog signal feature joint coding vector as the current analog signal joint coding feature; S520, performing response inference based on implicit signal feature analysis on the AB-phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector to obtain the voltage-current response inference coding vector.

[0057] Specifically, in step S510, the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector are combined to obtain an AB-phase current analog signal feature joint coding vector as the current analog signal joint coding feature. It should be understood that although analyzing a single-phase current feature alone can extract certain motor operation information, since the three-phase current of a brushless motor (usually the third phase is inferred by measuring two-phase currents) is inherently interrelated and coupled, the single-phase feature cannot fully reflect the electromagnetic state of the motor and the relative position of the rotor in the stator magnetic field. Therefore, in the technical solution of the present application, the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector are combined to obtain an AB-phase current analog signal feature joint coding vector. In particular, the characteristic coding vectors of the A-phase current analog signal and the B-phase current analog signal can be cascaded to combine the characteristic coding vectors of the A-phase and B-phase currents, which can fully capture the dynamic linkage characteristics, amplitude relationship and phase relationship between the two-phase currents. These comprehensive information is crucial for accurately judging the real-time electromagnetic position and motion state of the rotor, especially in sensorless control, where the current signal is the main basis for indirectly inferring the rotor position.

[0058] Specifically, in step S520, response inference based on implicit signal feature analysis is performed on the AB phase current analog signal feature joint encoding vector and the DC bus voltage signal feature encoding vector to obtain the voltage-current response inference encoding vector. It should be understood that, during actual motor operation, the relationship between the joint encoding representation of the AB phase current analog signal features and the DC bus voltage signal feature encoding vector is not a simple linear correspondence. Instead, it is influenced by the combined influence of motor parameters (e.g., inductance, resistance), load variations, power supply fluctuations, and environmental factors such as high temperature, exhibiting highly complex, nonlinear dynamic characteristics. Traditional analysis methods often have difficulty capturing these deep, implicit feature interactions and response patterns. Therefore, in order to deeply mine implicit information strongly correlated with rotor position from the joint encoding representation of the AB phase current analog signal features and the DC bus voltage signal feature encoding vector, and to overcome the limitations of traditional methods under complex operating conditions and signal interference, in the technical solution of the present application, response inference based on implicit signal feature analysis is performed on the joint encoding representation of the AB phase current analog signal features and the DC bus voltage signal feature encoding vector to obtain a voltage-current response inference encoding vector. Response inference based on implicit analysis of signal features enables understanding and quantification of the complex interactions between current and voltage features. Specifically, the input feature vectors (i.e., the joint encoding vector of the AB phase current analog signal features and the DC bus voltage signal features) are ordered to focus on the intrinsic strength of the features. Equal-granularity feature segmentation is then performed to enable detailed analysis of local features. Subsequently, the core transfer response inference unit performs deep interaction modeling on the local feature segments of each pair of AB phase current analog signal and DC bus voltage signal, capturing their mutual influence, response patterns, or information transfer mechanisms. This unit retains rich details of the local voltage-current signal interactions in a matrix format. Finally, the sequence transfer and aggregation module integrates this local voltage-current signal interaction information to capture dependencies and overall trends across different feature strength ranges. The ultimate goal is to generate a concise and information-rich voltage-current response inference encoding vector. This encoding vector characterizes the dynamic response relationship between the applied voltage and the resulting current in a highly generalized manner that is highly discriminative of rotor position. By deeply exploring and understanding the complex coupling relationship between voltage and current, the control system can more accurately identify the subtle electrical signal characteristics caused by rotor position changes, and maintain high estimation reliability even under harsh operating conditions such as high temperature, high noise, motor parameter drift, or severe load fluctuations.

[0059] Figure 9The flowchart of the brushless motor control circuit for high-temperature dual-control water distributor according to the embodiment of the present application is to perform response reasoning based on implicit analysis of signal characteristics on the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector to obtain the voltage-current response reasoning coding vector. Figure 9 As shown, according to the brushless motor control circuit for a high-temperature dual-control water distributor according to an embodiment of the present application, step S520 includes: S521, ordering and equally granularly dividing the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector to obtain a sequence of AB phase current analog signal local feature ordered coding vectors and a sequence of DC bus voltage signal local feature ordered coding vectors; S522, inputting each group of corresponding AB phase current analog signal local feature ordered coding vectors and DC bus voltage signal local feature ordered coding vectors in the sequence of AB phase current analog signal local feature ordered coding vectors and the sequence of DC bus voltage signal local feature ordered coding vectors into a transfer response inference unit to obtain a sequence of voltage-current local transfer response coding matrices; S523, performing transfer response inference sequence transmission on the sequence of voltage-current local transfer response coding matrices to obtain the voltage-current response inference coding vector.

[0060] Accordingly, according to an embodiment of the present application, step S521 includes: performing ordered arrangement of the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector based on the size of the eigenvalue to obtain the AB phase current analog signal feature ordered arrangement coding vector and the DC bus voltage signal feature ordered arrangement coding vector; performing equal-granularity feature segmentation on the AB phase current analog signal feature ordered arrangement coding vector and the DC bus voltage signal feature ordered arrangement coding vector to obtain a sequence of the AB phase current analog signal local feature ordered coding vectors and a sequence of the DC bus voltage signal local feature ordered coding vectors.

[0061] More specifically, the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector are ordered based on the eigenvalue size to obtain the AB phase current analog signal feature ordered arrangement coding vector and the DC bus voltage signal feature ordered arrangement coding vector, which can be expressed as follows: ;in, and represent the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector respectively, Indicates sorting of vector elements. and They respectively represent the ordered arrangement coding vectors of the AB phase current analog signal characteristics and the ordered arrangement coding vectors of the DC bus voltage signal characteristics.

[0062] It should be understood that while the features extracted from the raw current and voltage signals contain information about the motor's operation, the order of these features within the vector can be arbitrary, dependent on the implementation of the feature extraction algorithm or the designer's subjective settings, and not necessarily related to the intrinsic importance or strength of the features themselves. However, in response inference based on implicit analysis of signal features, the complex relationship between the applied motor voltage and the current response is often more profoundly reflected in the relative strength and distribution of different electrical features, rather than their original positions within the input vector. Especially when capturing complex nonlinear interactions between features, a normalized representation based on numerical ordering can help the network learn and generalize more effectively. Therefore, by sorting vector elements based on the magnitude of their eigenvalues, the arbitrariness introduced by the original order can be eliminated, allowing subsequent processing to focus on the intrinsic strength and relative magnitude of the features, thereby improving understanding of the signal's essential properties. The purpose of this step is to provide a standardized input representation that is somewhat invariant or equivariant to the input order. By arranging the AB phase current analog signal feature joint encoding vector and the DC bus voltage signal feature encoding vector in an ordered manner according to their numerical values, even if the input signal sampling or feature extraction process differs slightly, resulting in a different order of the original features, as long as the feature values ​​remain consistent, the ordered vectors obtained after sorting will be identical or similar. This normalization process ensures that subsequent feature segmentation and local interaction modeling can always be performed on feature intervals of similar strength or importance, thereby simplifying the network's learning task and improving the model's generalization and robustness.

[0063] More specifically, the ordered arrangement coding vectors of the AB phase current analog signal features and the ordered arrangement coding vectors of the DC bus voltage signal features are subjected to equal granularity feature segmentation to obtain a sequence of the AB phase current analog signal local feature ordered coding vectors and a sequence of the DC bus voltage signal local feature ordered coding vectors, which can be expressed as follows: ;in, represents the feature segmentation function, They represent the first, second, and third vectors in the sequence of ordered encoding vectors of the local characteristics of the AB phase current analog signal. and The ordered encoding vector of the local characteristics of the AB phase current analog signal, They represent the first, second, and third vectors in the sequence of ordered encoding vectors of the local characteristics of the DC bus voltage signal. and The ordered encoding vector of the local characteristics of the DC bus voltage signal, for and The voltage-current local transfer response encoding matrix between is matrix multiplication.

[0064] It should be understood that even after feature extraction and ordered arrangement, the ordered arrangement encoding vectors of the AB phase current analog signal features and the ordered arrangement encoding vectors of the DC bus voltage signal features may still contain a large number of dimensions. Directly performing global interaction modeling on the entire high-dimensional vectors is computationally intensive and difficult to capture the fine-grained correlations between features. The electrical behavior of motors often exhibits localized coupling relationships. For example, a current feature of a specific intensity may be closely correlated with a voltage variation pattern within a specific range. This correlation is easily diluted or obscured in the high-dimensional global space. Therefore, by performing an equal-granularity feature splitting operation, the ordered feature vector can be decomposed into a series of continuous sub-vector segments (local features) of equal size (i.e., equal granularity). This transforms the complex global interaction problem into a local interaction analysis between a series of simpler and easier-to-model AB phase current analog signal features and DC bus voltage signal features. The purpose of this step is to provide input data for the subsequent transfer response inference unit, enabling it to capture the local interaction patterns between the AB phase current features and the DC bus voltage features at a finer granularity. By dividing the ordered vector into local feature segments of equal granularity, the subsequent transfer response inference unit can focus on the mutual influence and response relationship between the AB phase current and DC bus voltage features within a similar feature strength range. The granularity of the segmentation, as a key hyperparameter, determines the level of analysis refinement. Smaller granularity can capture more subtle local correlations, while larger granularity focuses on more macroscopic local trends. This localized analysis strategy enables the model to more effectively learn the specific response patterns of current and voltage signals within different feature strength ranges, such as the association between weak current features and low voltage changes, or the association between strong current features and high voltage transients.

[0065] More specifically, in step S522, each corresponding set of the ordered coded vectors of the local characteristics of the AB phase current analog signals and the ordered coded vectors of the local characteristics of the DC bus voltage signals in the sequence of the ordered coded vectors of the local characteristics of the AB phase current analog signals and the sequence of the ordered coded vectors of the local characteristics of the DC bus voltage signals is input into the transfer response inference unit to obtain a sequence of voltage-current local transfer response coding matrices, which can be expressed as follows: ;in, for and The voltage-current local transfer response encoding matrix between is matrix multiplication, for activation function, is the trainable modulation weight matrix.

[0066] It should be understood that simply segmenting the ordered permutation encoding vectors of the AB phase current analog signal features and the ordered permutation encoding vectors of the DC bus voltage signal features into local segments is insufficient to reveal the complex interactions between them. During brushless motor operation, the current response and the applied voltage may exhibit highly nonlinear, dynamically changing coupling relationships across different feature intensity ranges and local detail levels. This relationship cannot be fully captured through simple linear combinations or global comparisons. Therefore, a dedicated mechanism, namely a "transfer response inference unit," is required to deeply model the interactions between each pair of local feature segments extracted from the current and voltage feature sequences, representing specific intensity ranges, thereby precisely capturing their mutual influence, response patterns, or information transfer mechanisms. This is achieved by deeply modeling and quantizing the complex, nonlinear interactions between each pair of corresponding local current feature vectors and local voltage feature vectors. The "transfer response inference unit," a carefully designed neural network module, leverages the powerful fitting capabilities of deep learning to quantify and encode various possible interaction patterns between the ordered local feature encoding vectors of the AB phase current analog signals and the ordered local feature encoding vectors of the DC bus voltage signal within a specific eigenvalue range, including their local similarities, differences, alignments, conditional dependencies, and co-activation or inhibition. In this way, a voltage-current local transfer response encoding matrix is ​​generated for each pair of input local feature vectors. This matrix is ​​not a simple scalar score, but rather aims to encapsulate rich and structured interaction information within the corresponding local region. For example, it captures the local correlation strength at the element-level of voltage and current, preserving more interaction details and providing high-quality local insights for a comprehensive understanding of the overall interaction. This fine-grained interaction modeling enables the model to more accurately identify the true mapping relationship between voltage and current under different operating conditions (such as startup, low speed, load variation, and high temperature effects), thereby improving the accuracy and robustness of rotor electrical angle estimation.

[0067] More specifically, in step S523, the sequence of the voltage-current local transfer response coding matrices is subjected to transfer response inference sequence transfer to obtain the voltage-current response inference coding vector, which is expressed as: ;in, Indicates flattening the matrix into vector processing, is the voltage-current local transfer response encoding vector, for of norm, is the value of the natural exponential function with the natural constant e as the base, To transfer the response inference weighted operation, for encoder, An encoding vector is inferred for the voltage-current response.

[0068] It should be understood that although the previous step uses the transfer response inference unit to deeply model the interaction between each pair of local feature segments of the AB phase current analog signal and the DC bus voltage signal and generate a sequence of local transfer response encoding matrices, which detail the interactions between voltage and current features in different local regions, the overall voltage-current response of the motor is not a simple collection of isolated local interactions. Local interactions in regions of different feature strengths are interdependent and influence each other, and their evolution (as ranked by feature strength level) implies global, deep-level response patterns. Therefore, a mechanism is required to integrate the local interaction information distributed across different feature strength intervals and critically consider the sequential relationships between them to form a comprehensive understanding of the overall interaction. A discrete sequence of local matrices alone is insufficient to fully characterize this global dynamic. By performing transfer response inference sequence propagation, it is possible to learn and capture how the local interaction patterns between current and voltage signals evolve with changes in feature strength, discovering dependencies and global context across different feature strength intervals. This process aims to meaningfully integrate and refine the rich current-voltage local interaction information, ultimately outputting a concise and information-rich voltage-current response inference encoding vector. This vector is a highly abstract and compact representation of the overall, deep-level interaction response characteristics between the two original input feature vectors (the combined current and voltage features). It aims to capture the overall pattern of current response caused by voltage application and how this pattern is affected by various factors (including rotor position). Compared to relying solely on raw features or simple aggregation methods, this vector, obtained through fine-grained local interaction modeling and global sequence aggregation, can more accurately and stably reflect electrical state information that is strongly correlated with the electrical angle of the motor rotor.

[0069] Preferably, it can be seen that the segmentation granularity of the characteristic segmentation function, while affecting the local partition range of the voltage-current local transfer response coding matrix, will also directly affect the expression of the coupling relationship between the corresponding local characteristic ordered coding vector of the AB phase current analog signal and the local characteristic ordered coding vector of the DC bus voltage signal.

[0070] Since the voltage-current local transfer response encoding matrix is ​​the transfer response between two local regions, it actually uses its row vector as a benchmark to express the spatial measure of the transfer response space, and the row vector dimension is also the representation of the local region size mentioned above, if the local region size, that is, the row vector length If it is introduced as a local size strength constraint, then the low-dimensional representation metric of the voltage-current local transfer response encoding matrix, namely the F norm It should follow a Poisson-like relationship: ; that is, the row vector length The number of low-rank spatial measures acting on the local region transfer response space as a local size strength constraint is Second-rate, represents the F norm of the matrix, express The factorial of . From this, we can solve the parameter .

[0071] Thus, the transfer-response interaction is introduced by and the mean expectation is In the case of a Poisson-like process, and Describing the edge connection representation of the spatial measure, the transfer response association probability between the two local regions is further determined as: ;in, represents the two-norm of the vector, is the transfer response association probability.

[0072] Then, the transfer response association probability To adjust the size of the local area Make iterative adjustments: That is, under the strict guarantee that the expected degree is In this case, the regularization of the overall spatial metric is determined by the regularization constraint of the mean expected connection probability fluctuation of each row, so that the structured coupling information in the voltage-current local spatial measure can avoid local overfitting and enhance the global expression ability of the sequence of the voltage-current local transfer response encoding matrix.

[0073] Specifically, in step S600, the voltage-current response inference coding vector is feature-decoded to obtain an estimated value of the rotor electrical angle. That is to say, although the voltage-current response inference coding vector is highly condensed and rich in information, it is itself an abstract high-dimensional feature expression, which contains the implicit association between complex electrical characteristics and rotor position, but has not yet been intuitively mapped into actual angle information that can be used for control. Therefore, in the technical solution of the present application, the voltage-current response inference coding vector is further feature-decoded to obtain an estimated value of the rotor electrical angle, so as to meet the control system's demand for real-time and accurate grasp of the rotor position. Through effective feature decoding, the system can accurately restore the real-time position of the rotor in the stator magnetic field of the motor, provide an accurate basis for the subsequent PWM control signal generation, and thus achieve efficient closed-loop control of the brushless DC motor.

[0074] Specifically, in an embodiment of the present application, the voltage-current response inference encoding vector is feature decoded to obtain an estimated value of the rotor electrical angle, including: passing the voltage-current response inference encoding vector through a decoder-based rotor electrical angle controller to obtain an estimated decoded value of the rotor electrical angle.

[0075] Furthermore, after obtaining the estimated decoded value of the rotor electrical angle, the estimated decoded value of the rotor electrical angle, obtained through implicit signal feature analysis and feature decoding, is first used as the core feedback signal. The control system then enters a hierarchical control loop, consisting of an outer-loop position or speed controller and an inner-loop current controller. The outer-loop controller (e.g., a PID controller) receives a target position or speed command from a host computer or a preset motion profile and compares it with the rotor speed estimated through integration (e.g., by taking the time derivative of the angle estimate or obtained through a specific observer) or a direct position estimate. This generates a torque command for the motor, which is then converted into a reference current command in the dq coordinate system. Specifically, this command generates a q-axis current command for torque control. The d-axis current command for flux control is typically set to zero to achieve maximum efficiency (below base speed).

[0076] The inner-loop current controller is then responsible for accurately tracking these reference current commands. The current controller first compares the actual d-axis and q-axis currents, acquired in real time and Park-transformed in the previous step, with the reference current commands to generate current errors. These two current error signals are input to independent current regulators (typically PI controllers), which output reference voltage commands in the dq coordinate system. Subsequently, an inverse Park transform is performed to convert the voltage commands in the dq coordinate system into actual voltages that can be applied to the motor phase windings. Using the previously decoded rotor electrical angle estimate, the reference voltage commands are converted to voltage commands in the stationary coordinate system through an inverse Park transform. Finally, the stationary voltage commands are input to the space vector pulse width modulation (SVPWM) module. Based on the input voltage commands, the SVPWM algorithm calculates the duty cycle and timing of the PWM pulse signals that drive each power switch in the three-phase H-bridge circuit. After being amplified and isolated by the gate drive circuit, these PWM signals are applied to the power switches of the three-phase H-bridge, generating a synthetic vector voltage that precisely drives the three-phase windings of the high-temperature brushless DC motor. This voltage enables the actual phase current to quickly track the output commands of the current controller, thereby achieving precise control of the motor's torque, speed, or position. The entire closed-loop control process is executed in real time, continuously adjusting the drive voltage based on the latest estimated rotor electrical angle, ensuring stable and efficient motor operation under various operating conditions.

[0077] In summary, according to the embodiment of the present application, a brushless motor control circuit for a high-temperature dual-control water distributor is described. It does not rely on traditional physical position sensors, but instead uses a current detection circuit to collect motor current signals (including A and B phase currents) and DC bus voltage signals in real time, extract key features from these signals, and further perform complex response reasoning on the extracted signal features based on implicit analysis of signal features to estimate the electrical angle of the motor rotor. Compared to traditional sensorless control methods that rely on physical sensors or back electromotive force, this technical solution can help improve the accuracy of rotor position estimation and the overall robustness of the system in harsh high-temperature environments, especially in challenging working conditions such as motor starting and low-speed operation, thereby ensuring precise control and long-term stable operation of the high-temperature dual-control water distributor.

[0078] While various embodiments of the present disclosure have been described above, the above descriptions are illustrative, non-exhaustive, and not intended to be limiting of the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A brushless motor control circuit for a high-temperature dual-control water distributor, characterized in that: include: MCU control circuit, gate drive circuit, current detection circuit, 3-phase H-bridge circuit and high-temperature DC brushless motor; The current detection circuit is electrically connected to the three-phase H-bridge circuit to collect real-time current, and inputs the real-time current to the MCU control circuit. The MCU control circuit determines the rotor position based on the real-time current and outputs a PWM control signal based on the rotor position. The PWM control signal output by the MCU control circuit drives the high-temperature DC brushless motor through the gate drive circuit and the three-phase H-bridge circuit; In the closed-loop control stage, the MCU control circuit collects three-phase currents in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase currents, and then outputs a PWM signal based on the rotor electrical angle; The MCU control circuit collects three-phase currents in real time through the current detection circuit and estimates the rotor electrical angle based on the three-phase currents, including: Obtaining the A-phase current analog signal and the B-phase current analog signal; Obtain DC bus voltage signal; Extracting current signal features from the A-phase current analog signal and the B-phase current analog signal to obtain an A-phase current analog signal feature coding vector and a B-phase current analog signal feature coding vector; Extracting voltage signal features from the DC bus voltage signal to obtain a DC bus voltage signal feature encoding vector; Based on the current analog signal joint coding feature between the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector and the DC bus voltage signal feature coding vector, performing response reasoning based on implicit analysis of signal features to obtain a voltage-current response reasoning coding vector; Performing feature decoding on the voltage-current response inference encoding vector to obtain an estimated value of the rotor electrical angle; Based on the joint coding feature of the current analog signal between the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector and the DC bus voltage signal feature coding vector, response reasoning based on implicit analysis of signal features is performed to obtain a voltage-current response reasoning coding vector, including: Combining the A-phase current analog signal feature coding vector and the B-phase current analog signal feature coding vector to obtain an AB-phase current analog signal feature joint coding vector as the current analog signal joint coding feature; Response reasoning based on implicit analysis of signal characteristics is performed on the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector to obtain the voltage-current response reasoning coding vector.

2. The brushless motor control circuit for a high-temperature dual-control water distributor according to claim 1, characterized in that: During the rotor alignment phase, the MCU control circuit controls the gate drive circuit and the three-phase H-bridge circuit to apply a fixed DC current to phases A and B of the high-temperature brushless DC motor.

3. The brushless motor control circuit for a high-temperature dual-control water distributor according to claim 2, characterized in that: During the open-loop acceleration phase, the MCU control circuit applies a PWM signal to the gate drive circuit and the three-phase H-bridge circuit to generate a continuously accelerating magnetic field in the stator of the high-temperature DC brushless motor, wherein the rotor of the high-temperature DC brushless motor gradually increases its speed from zero to a minimum rate under the action of the continuously accelerating magnetic field.

4. The brushless motor control circuit for a high-temperature dual-control water distributor according to claim 1, characterized in that: Performing response reasoning based on implicit analysis of signal characteristics on the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector to obtain the voltage-current response reasoning coding vector, including: The AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector are arranged in an ordered manner and divided into equal granularity to obtain a sequence of AB phase current analog signal local feature ordered coding vectors and a sequence of DC bus voltage signal local feature ordered coding vectors; Inputting each corresponding group of the AB phase current analog signal local feature ordered code vectors and the DC bus voltage signal local feature ordered code vectors in the sequence of the AB phase current analog signal local feature ordered code vectors and the sequence of the DC bus voltage signal local feature ordered code vectors into a transfer response inference unit to obtain a sequence of voltage-current local transfer response code matrices; The voltage-current local transfer response coding matrix sequence is subjected to transfer response inference sequence transfer to obtain the voltage-current response inference coding vector.

5. The brushless motor control circuit for a high-temperature dual-control water distributor according to claim 4, characterized in that: The AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector are arranged in an ordered manner and divided into equal granularity to obtain a sequence of AB phase current analog signal local feature ordered coding vectors and a sequence of DC bus voltage signal local feature ordered coding vectors, including: Performing an ordered arrangement based on the eigenvalues ​​of the AB phase current analog signal feature joint coding vector and the DC bus voltage signal feature coding vector to obtain an AB phase current analog signal feature ordered arrangement coding vector and a DC bus voltage signal feature ordered arrangement coding vector; The ordered arrangement coding vectors of the AB phase current analog signal features and the ordered arrangement coding vectors of the DC bus voltage signal features are subjected to equal-granularity feature segmentation to obtain a sequence of the AB phase current analog signal local feature ordered coding vectors and a sequence of the DC bus voltage signal local feature ordered coding vectors.

6. The brushless motor control circuit for a high-temperature dual-control water distributor according to claim 1, characterized in that: Feature decoding is performed on the voltage-current response inference encoding vector to obtain an estimated value of the rotor electrical angle, including: passing the voltage-current response inference encoding vector through a decoder-based rotor electrical angle controller to obtain an estimated decoded value of the rotor electrical angle.

Images