Control method for low-temperature large-torque starting of brushless motor of oil pump without position sensor
By adopting a position-free sensor control method of high-frequency signal injection and extended back electromotive force in the oil pump, combined with the hysteresis smooth switching control strategy, the problem of unstable start of low speed and high torque at low temperatures is solved, and efficient low-temperature and stable operation is achieved.
Patent Information
- Application Number
- CN202510100474.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-22
- Publication Date
- 2025-05-02
AI Technical Summary
Traditional oil pumps are difficult to achieve efficient low-speed and high-torque start under low temperature conditions, and there is a problem of unstable start and loss of steps without position sensor control.
The positionless sensor control method based on high-frequency signal injection and extended back electromotive force is adopted. By using the high-frequency signal injection method and extended back electromotive force method in the low-speed and medium-high-speed domains, respectively, combined with the hysteresis smooth switching control strategy, the stable operation of the oil pump at low temperature is achieved.
It improves the start-up success rate and low-speed stable operation capability of the oil pump at low temperatures, realizes stable operation of the high and low full speed segments, reduces costs and improves control accuracy.
Smart Images

Figure CN119921615A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of oil pumps, in particular to a control method for low-temperature and high-torque starting of a brushless motor of a position sensorless oil pump. Background Art
[0002] Low-temperature engine oil has a relatively high viscosity. In order to achieve precise control of the speed closed loop and start-up at low temperatures and high torque, traditional electronic oil pumps need to add position sensors (such as Hall or magnetic encoders). Position sensors increase the cost of electronic control, and sensors are prone to failure (for example, Hall sensors are prone to failure at high temperatures of motor windings).
[0003] Existing inventions generally use position sensors or no position sensors:
[0004] 1. There is a position sensor. According to the position signal collected by the position sensor in real time, the rotor of the oil pump can be accurately known for precise control. This method has high control accuracy, but there is no cost advantage;
[0005] 2. No position sensor. Generally, position-free control is based on the back-EMF model for position estimation. However, the back-EMF is very small at low speed, which causes the model to fail. Therefore, most sensorless control uses open-loop strong drag and then switches to closed-loop control. This method has the following disadvantages: insufficient open-loop drag torque at low speed and high torque makes it easy to lose step, and motors with small back-EMF will have unstable switching when the switching speed threshold is low, resulting in startup failure. Summary of the invention
[0006] In view of the above problems, the present invention provides a control method for low-temperature and high-torque starting of a brushless motor of a position sensorless oil pump that achieves stable operation in all high and low speed ranges.
[0007] The technical solution adopted by the present invention to solve the above technical problems is: a control method for low-temperature and high-torque starting of a brushless motor of a position sensorless oil pump, comprising the following steps:
[0008] Step 1: Starting the oil pump without position sensor based on high-frequency signal injection to realize position sensorless vector control of the permanent magnet synchronous motor in the low-speed domain;
[0009] Step 2: Starting the oil pump motor without position sensor based on the extended back electromotive force to realize the position sensorless vector control of the permanent magnet synchronous motor in the medium and high speed domain;
[0010] Step 3, hysteresis-based smooth switching control strategy;
[0011] Step 3.1, confirm the low speed threshold ω1 and the medium and high speed threshold ω2, determine the threshold range of the oil pump motor speed and adopt the corresponding starting method,
[0012] According to the actual speed of the oil pump motor, the low-speed domain threshold ω1 and the medium-high speed domain threshold ω2 are confirmed. When the speed is lower than ω1, it is determined that the oil pump motor is working in the zero low-speed domain condition, and step 1 is implemented. When the speed is higher than the threshold ω2, it is determined that the oil pump motor is working in the medium-high speed domain, and step 2 is implemented.
[0013] Step 3.2, smooth switching strategy based on hysteresis control,
[0014] When the oil pump speed increases from ω1 to ω2, the high-frequency signal injection method is used when the speed is lower than ω1, and the high-frequency signal injection method is continued to be used between ω1 and ω2 until the speed exceeds ω2 and switches to the extended back-EMF method; when the speed is higher than ω2 and decreases to ω1, the extended back-EMF method is used when the speed is higher than ω2, and the extended back-EMF method is continued to be used between ω2 and ω1 until the speed is lower than ω1, and then switches to the high-frequency signal injection method.
[0015] Preferably, the step 1 is specifically as follows:
[0016] Step 1.1, construct a high-frequency mathematical model of the oil pump motor in the dq axis system, and derive a high-frequency voltage equation of the oil pump motor in the dq axis system;
[0017] The high-frequency mathematical model of the oil pump motor in the dq axis system is as follows:
[0018]
[0019] Among them, u d,q ——voltage component in dq axis system; i d,q ——current component in dq axis system;
[0020] R s ——stator resistance; L d,q ——dq axis inductance;ω——rotor electrical angle;
[0021] ——permanent magnet flux; p——differential operator;
[0022] Since the injected high-frequency signal frequency is much greater than the oil pump motor operating frequency, the winding voltage drop and back electromotive force term in the high-frequency mathematical model (1-1) are ignored, that is, R s *i d =R s *i q = 0, and Therefore, the high-frequency mathematical model of the oil pump in the dq axis system is simplified to:
[0023]
[0024] The "h" below represents "high frequency";
[0025] Step 1.2, based on the high-frequency voltage equation in the dq axis system, derive the synchronous rotating coordinate system of the observed rotor High frequency response current under the shaft system;
[0026] Establishing a coordinate system for observing synchronous rotation of the rotor And observe the relationship with the actual rotor synchronous rotating coordinate system, the rotor angle observation error is:
[0027] Actual dq axis system and observation The voltage relationship of the shaft system is as follows:
[0028]
[0029] Transforming equation (1-2) into the high-frequency current derivative equation is:
[0030]
[0031] When there is an error between the actual position and the observed rotor position, The high-frequency current response under the shaft system will change according to the injected high-frequency voltage, which can be expressed as:
[0032]
[0033] The simplification is as follows:
[0034]
[0035] Formula (1-6) is the current response under high-frequency excitation, where L is the average inductance, ΔL is the semi-differential inductance, T s is the switching cycle;
[0036] Step 1.3, high-frequency square wave signal injection, based on the high-frequency response current equation and the injected high-frequency square wave voltage signal, the relationship between the q-axis high-frequency response current and the angle error is obtained;
[0037] The high-frequency square wave voltage injection method is only used in the synchronously rotating coordinate system of the observed rotor. Axis injection of high frequency square wave voltage signal:
[0038]
[0039] Where:
[0040]
[0041] Voltage component under the shaft system; V h is the amplitude of the injected high-frequency square wave, ΔT is the period of square wave signal injection;
[0042] When high-frequency square wave is injected, the injection frequency is selected as half of the switching frequency, and the current is sampled once in each PWM cycle. of When a high-frequency voltage signal is injected into the axis, The shaft generates a high frequency response current with the same frequency and The shaft response current is theoretically 0;
[0043] Substituting equations (1-7) and (1-8) into equation (1-6), the high-frequency response current caused by the injection of high-frequency square wave voltage is expressed as:
[0044]
[0045] Where sign(·) is the sign function, indicating that the injected high-frequency voltage changes with the injection period.
[0046] From formula (1-9), we know that when the angle observation error is small enough to linearize the error signal, that is, at this time The shaft current equation is expressed as:
[0047]
[0048] Where:
[0049] Step 1.4, high frequency signal demodulation and rotor position observation;
[0050] From formula (1-10), we know that adjusting hour, That is, the rotor position converges to the actual position value,
[0051] The signal demodulation process of the high-frequency square wave injection method is to perform differential processing on the sampled q-axis current to obtain the current difference signal containing the rotor position error, modulate the amplitude to obtain the rotor position error signal, and finally estimate the rotor error signal through a phase-locked loop to obtain the converged observed rotor angular velocity. After integration, the observed rotor angle is obtained
[0052] Preferably, the step 2 is specifically as follows:
[0053] Step 2.1, constructing the extended back-EMF model, that is, constructing the voltage equation under the two-phase stationary coordinate system α-β axis system and deriving the relationship of the extended back-EMF;
[0054] The oil pump motor voltage equation in the dq coordinate system is as follows (2-1):
[0055]
[0056] By changing the coordinates of the voltage equation in the dq coordinate system, we can obtain the voltage equation in the stationary coordinate system:
[0057]
[0058] Among them, u α,β ——stator voltage component under α-β axis system; i α,β ——stator current component in α-β axis system;
[0059] L α,β ——Inductance under α-β axis; R s ——stator resistance; ω——rotor electrical angle; ——permanent magnet flux; p——differential operator; where, In order to expand the back electromotive force, equation (2-2) is simplified to the stationary coordinate system:
[0060]
[0061] in, E α,β ——The corresponding extended back electromotive force component under the α-β axis system;
[0062] Step 2.2, obtaining the estimated extended back electromotive force through the PI regulator;
[0063] Using PI regulator to achieve,
[0064]
[0065] In formula (2-4), is the observed value of the rotor angular velocity ω, for i α The observed value of for i β The observed value of For E α The observed value of For E β The observed value. Using the PI regulator to observe the back electromotive force, it is expressed as:
[0066]
[0067] In the formula, K p is the proportional gain coefficient of the PI regulator, K i is the integral gain coefficient of the PI regulator;
[0068] Step 2.3, rotor position observation;
[0069] From (2-5) we can get the estimated And the observed rotor position is obtained by inverse tangent calculation
[0070]
[0071] In the back electromotive force estimation link, compensation θ is performed based on the observations in formula (2-6) com .
[0072]
[0073] Where: is the observed angular velocity, ω c is the bandwidth of the back EMF estimation link,
[0074] The observation angle after compensation is:
[0075]
[0076] Compared with the prior art, the present invention has the following beneficial effects:
[0077] 1. The present invention can greatly improve the success rate of the oil pump startup at low temperatures and solve the problem of stable operation at low temperatures and low speeds. Under the condition of low back electromotive force at low speeds, high-frequency pulse injection is used to collect high-frequency current responses and extract position information through software algorithms. The current position information can be accurately known for precise speed closed-loop control.
[0078] 2. The present invention sets a low-speed domain threshold ω1 and a medium- and high-speed domain threshold ω2. When the oil pump speed increases from ω1 to ω2, the high-frequency signal injection method is used when the speed is lower than ω1, and the high-frequency signal injection method is continued to be used between ω1 and ω2 until the speed exceeds ω2 and switches to the extended back-electromotive force method; when the speed is higher than ω2 and decreases to ω1, the extended back-electromotive force method is used when the speed is higher than ω2, and the extended back-electromotive force method is continued to be used between ω2 and ω1 until the speed is lower than ω1, and then switches to the high-frequency signal injection method. The current oil pump speed is detected in real time during the entire operation process, thereby achieving stable operation in all high and low speed ranges. BRIEF DESCRIPTION OF THE DRAWINGS
[0079] Figure 1 It is a schematic diagram of each coordinate system of the IPMSM of the present invention;
[0080] Figure 2 It is a schematic diagram of high-frequency square wave injection of the present invention;
[0081] Figure 3It is a schematic diagram of signal demodulation of the square wave injection method of the present invention;
[0082] Figure 4 It is a schematic diagram of hysteresis control of the present invention. DETAILED DESCRIPTION
[0083] The following will be combined Figure 1-4 The present invention is described in detail. The exemplary embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.
[0084] A control method for low-temperature high-torque starting of a brushless motor of a position sensorless oil pump comprises the following steps:
[0085] Step 1: Starting the oil pump without position sensor based on high-frequency signal injection to realize position sensorless vector control of the permanent magnet synchronous motor in the low-speed domain;
[0086] The oil pump position sensorless starting method based on high-frequency signal injection mainly utilizes the anisotropy of the motor. By injecting high-frequency excitation signals into the motor, the rotor position and speed are estimated using the corresponding high-frequency current, thereby realizing position sensorless vector control of the permanent magnet synchronous motor in the low-speed domain.
[0087] Step 1.1, construct a high-frequency mathematical model of the oil pump motor in the dq axis system, and derive a high-frequency voltage equation of the oil pump motor in the dq axis system;
[0088] The mathematical model parameters of the permanent magnet synchronous motor in the ABC three-phase stationary coordinate system are coupled, which makes it difficult to control. In order to simplify the motor control, the control method of the DC motor is referred to, and the magnetic flux and torque are decoupled and controlled separately, that is, a high-frequency mathematical model is constructed in the dq axis system. The high-frequency mathematical model of the oil pump motor in the dq axis system is as follows:
[0089]
[0090] Among them, u d,q ——voltage component in dq axis system; i d,q ——current component in dq axis system;
[0091] R s ——stator resistance; L d,q ——dq axis inductance;ω——rotor electrical angle;
[0092] ——permanent magnet flux; p——differential operator;
[0093] Since the injected high-frequency signal frequency is much greater than the oil pump motor operating frequency, the winding voltage drop and back electromotive force term in the high-frequency mathematical model (1-1) can be ignored, that is, Rs *i d =R s *i q = 0, and Therefore, the high-frequency mathematical model of the oil pump in the dq axis system can be simplified as:
[0094]
[0095] The "h" below represents "high frequency";
[0096] Step 1.2, based on the high-frequency voltage equation in the dq axis system, derive the synchronous rotating coordinate system of the observed rotor High frequency response current under the shaft system;
[0097] Common high-frequency signal injection methods mainly include high-frequency rotation signal injection method, high-frequency pulse signal injection method and high-frequency square wave injection method. The high-frequency square wave injection method uses fewer filters and is simple to implement. The present invention uses this method to start the oil pump under low-speed conditions. In order to achieve high-frequency square wave injection, it is necessary to establish a synchronous rotating coordinate system for observing the rotor. And observe the relationship with the actual rotor synchronous rotating coordinate system, such as Figure 1 It can be seen that the rotor angle observation error is:
[0098] And by Figure 1 It can be seen that the actual dq axis system is consistent with the observed The voltage relationship of the shaft system is as follows:
[0099]
[0100] Transforming equation (1-2) into the high-frequency current derivative equation is:
[0101]
[0102] When there is an error between the actual position and the observed rotor position, The high-frequency current response under the shaft system will change according to the injected high-frequency voltage and can be expressed as:
[0103]
[0104] The simplification is as follows:
[0105]
[0106] Formula (1-6) is the current response under high-frequency excitation, where L is the average inductance, ΔL is the semi-differential inductance, T s is the switching cycle;
[0107] Step 1.3, high-frequency square wave signal injection, based on the high-frequency response current equation and the injected high-frequency square wave voltage signal, the relationship between the q-axis high-frequency response current and the angle error is obtained;
[0108] The high-frequency square wave voltage injection method is only used in the synchronously rotating coordinate system of the observed rotor. Axis injection of high frequency square wave voltage signal:
[0109]
[0110] Where:
[0111]
[0112] Voltage component under the shaft system; V h is the amplitude of the injected high-frequency square wave, ΔT is the period of square wave signal injection;
[0113] Figure 2 This is a schematic diagram of high-frequency square wave injection. The injection frequency is selected as half of the switching frequency, and the current is sampled once in each PWM cycle. of When a high-frequency voltage signal is injected into the axis, The shaft generates a high frequency response current with the same frequency and The shaft response current is theoretically 0;
[0114] Substituting equations (1-7) and (1-8) into equation (1-6), the high-frequency response current caused by the injection of high-frequency square wave voltage is expressed as:
[0115]
[0116] Where sign(·) is the sign function, indicating that the injected high-frequency voltage changes with the injection period.
[0117] From formula (1-9), we can see that when the angle observation error is small enough to linearize the error signal, that is, at this time The shaft current equation is expressed as:
[0118]
[0119] Where:
[0120] Step 1.4, high frequency signal demodulation and rotor position observation;
[0121] From formula (1-10), we know that adjusting hour, That is, the rotor position converges to the actual position value,
[0122] The signal demodulation process of the high-frequency square wave injection method is as follows Figure 3 As shown in the figure, the sampled q-axis current is differentially processed, that is, the high-frequency excitation current signal is After signal demodulation, the current difference signal containing the rotor position error is obtained, and the rotor position error signal is obtained by amplitude modulation. Finally, the rotor error signal is estimated through a phase-locked loop to obtain the converged observed rotor angular velocity After integration, the observed rotor angle is obtained
[0123] Step 2: Starting the oil pump motor without position sensor based on the extended back electromotive force to realize the position sensorless vector control of the permanent magnet synchronous motor in the medium and high speed domain;
[0124] After the oil pump motor is successfully started, the speed slowly increases. At this time, the motor back electromotive force is relatively high. If the high-frequency signal injection method continues to be used, higher requirements will be met. Therefore, the extended back electromotive force method is used at medium and high speeds to realize position sensorless control of the oil pump motor.
[0125] Step 2.1, the extended back-EMF model is constructed, that is, the voltage equation under the two-phase stationary coordinate system α-β axis system is constructed and the relationship of the extended back-EMF is derived.
[0126] The oil pump motor voltage equation in the dq coordinate system is as follows (2-1):
[0127]
[0128] By changing the coordinates of the voltage equation in the dq coordinate system, we can obtain the voltage equation in the stationary coordinate system:
[0129]
[0130] Among them, u α,β ——stator voltage component under α-β axis system; i α,β ——stator current component in α-β axis system;
[0131] L α,β ——Inductance under α-β axis; R s ——stator resistance; ω——rotor electrical angle; ——permanent magnet flux; p——differential operator; where, In order to expand the back electromotive force, equation (2-2) is simplified to the stationary coordinate system:
[0132]
[0133] in, E α,β ——The corresponding extended back electromotive force component under the α-β axis system;
[0134] Step 2.2, obtaining the estimated extended back electromotive force through the PI regulator;
[0135] Since the extended back-EMF cannot be directly obtained by measurement, the estimated value of the extended back-EMF can be obtained by the error between the estimated current value and the actual current value, through observers, etc., so the PI regulator is adopted to achieve it.
[0136]
[0137] In formula (2-4), is the observed value of the rotor angular velocity ω, for i α The observed value of for i β The observed value of For E α The observed value of For E β The observed value. Using the PI regulator to observe the back electromotive force, it can be expressed as:
[0138]
[0139] In the formula, K p is the proportional gain coefficient of the PI regulator, K i is the integral gain coefficient of the PI regulator;
[0140] Step 2.3, rotor position observation;
[0141] From (2-5) we can get the estimated And the observed rotor position is obtained by inverse tangent calculation
[0142]
[0143] In the back electromotive force estimation link, in order to reduce the rotor position estimation error caused by bandwidth limitation, compensation θ is performed based on the observation of formula (2-6). com .
[0144]
[0145] Where: is the observed angular velocity, ω c is the bandwidth of the back EMF estimation link,
[0146] The observation angle after compensation is:
[0147]
[0148] Step 3, hysteresis-based smooth switching control strategy;
[0149] Due to some uncertainty between the characteristics of the controller and the actual operating conditions of the oil pump motor, the detected speed may change frequently before stabilizing, or oscillate and stabilize in a small range. This situation will cause the control strategy to switch frequently, thereby increasing the amount of calculation of the system and even causing the system to fail. Figure 4 The hysteresis control schematic diagram shown in FIG. 4 shows an invalid state change zone, in which the influence of environmental fluctuations is accommodated by the hysteresis of the response, thereby achieving smooth switching of strategies.
[0150] Step 3.1, confirm the low speed threshold ω1 and the medium and high speed threshold ω2, determine the threshold range of the oil pump motor speed and adopt the corresponding starting method,
[0151] According to the actual speed of the oil pump motor, the low-speed domain threshold ω1 and the medium-high speed domain threshold ω2 are confirmed. When the speed is lower than ω1, it is determined that the oil pump motor is working in the zero low-speed domain condition, and step 1 is implemented. When the speed is higher than the threshold ω2, it is determined that the oil pump motor is working in the medium-high speed domain, and step 2 is implemented.
[0152] Step 3.2, smooth switching strategy based on hysteresis control,
[0153] When the oil pump speed increases from ω1 to ω2, the high-frequency signal injection method is used when the speed is lower than ω1, and the high-frequency signal injection method is continued to be used between ω1 and ω2 until the speed exceeds ω2 and switches to the extended back-EMF method; when the speed is higher than ω2 and decreases to ω1, the extended back-EMF method is used when the speed is higher than ω2, and the extended back-EMF method is continued to be used between ω2 and ω1 until the speed is lower than ω1, and then switches to the high-frequency signal injection method.
[0154] The technical solutions provided by the embodiments of the present invention are introduced in detail above. Specific examples are used herein to illustrate the principles and implementation methods of the embodiments of the present invention. The description of the above embodiments is only applicable to help understand the principles of the embodiments of the present invention. At the same time, for those skilled in the art, according to the embodiments of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. A control method for low-temperature high-torque starting of a brushless motor of a position sensorless oil pump, characterized in that: The following steps are involved: Step 1: Starting the oil pump without position sensor based on high-frequency signal injection to realize position sensorless vector control of the permanent magnet synchronous motor in the low-speed domain; Step 2: Starting the oil pump motor without position sensor based on the extended back electromotive force to realize the position sensorless vector control of the permanent magnet synchronous motor in the medium and high speed domain; Step 3, hysteresis-based smooth switching control strategy; Step 3.1, confirm the low speed threshold ω1 and the medium and high speed threshold ω2, determine the threshold range of the oil pump motor speed and adopt the corresponding starting method, According to the actual speed of the oil pump motor, the low-speed domain threshold ω1 and the medium-high speed domain threshold ω2 are confirmed. When the speed is lower than ω1, it is determined that the oil pump motor is working in the zero low-speed domain condition, and step 1 is implemented. When the speed is higher than the threshold ω2, it is determined that the oil pump motor is working in the medium-high speed domain, and step 2 is implemented. Step 3.2, smooth switching strategy based on hysteresis control, When the oil pump speed increases from ω1 to ω2, the high-frequency signal injection method is used when the speed is lower than ω1, and the high-frequency signal injection method is continued to be used between ω1 and ω2 until the speed exceeds ω2 and switches to the extended back-EMF method; when the speed is higher than ω2 and decreases to ω1, the extended back-EMF method is used when the speed is higher than ω2, and the extended back-EMF method is continued to be used between ω2 and ω1 until the speed is lower than ω1, and then switches to the high-frequency signal injection method.
2. The control method for low-temperature and high-torque starting of a brushless motor of a position sensorless oil pump according to claim 1, characterized in that: The step 1 specifically comprises: Step 1.1, construct a high-frequency mathematical model of the oil pump motor in the dq axis system, and derive a high-frequency voltage equation of the oil pump motor in the dq axis system; The high-frequency mathematical model of the oil pump motor in the dq axis system is as follows: Among them, u d,q ——voltage component in dq axis system; i d,q ——current component in dq axis system; R s ——stator resistance; L d,q ——dq axis inductance;ω——rotor electrical angle; ——permanent magnet flux; p——differential operator; Since the injected high-frequency signal frequency is much greater than the oil pump motor operating frequency, the winding voltage drop and back electromotive force term in the high-frequency mathematical model (1-1) are ignored, that is, R s *i d =R s *i q = 0, and Therefore, the high-frequency mathematical model of the oil pump in the dq axis system is simplified to: The "h" below represents "high frequency"; Step 1.2, based on the high-frequency voltage equation in the dq axis system, derive the synchronous rotating coordinate system of the observed rotor High frequency response current under the shaft system; Establishing a coordinate system for observing synchronous rotation of the rotor And observe the relationship with the actual rotor synchronous rotating coordinate system, the rotor angle observation error is: Actual dq axis system and observation The voltage relationship of the shaft system is as follows: Transforming equation (1-2) into the high-frequency current derivative equation is: When there is an error between the actual position and the observed rotor position, The high-frequency current response under the shaft system will change according to the injected high-frequency voltage, which can be expressed as: The simplification is as follows: Formula (1-6) is the current response under high-frequency excitation, where L is the average inductance, ΔL is the semi-differential inductance, T s is the switching cycle; Step 1.3, high-frequency square wave signal injection, based on the high-frequency response current equation and the injected high-frequency square wave voltage signal, the relationship between the q-axis high-frequency response current and the angle error is obtained; The high-frequency square wave voltage injection method is only used in the synchronously rotating coordinate system of the observed rotor. Axis injection of high frequency square wave voltage signal: Where: —— Voltage component under the shaft system; V h is the amplitude of the injected high-frequency square wave, ΔT is the period of square wave signal injection; When high-frequency square wave is injected, the injection frequency is selected as half of the switching frequency, and the current is sampled once in each PWM cycle. of When a high-frequency voltage signal is injected into the axis, The shaft generates a high frequency response current with the same frequency and The shaft response current is theoretically 0; Substituting equations (1-7) and (1-8) into equation (1-6), the high-frequency response current caused by the injection of high-frequency square wave voltage is expressed as: Where sign(·) is the sign function, indicating that the injected high-frequency voltage changes with the injection period. From formula (1-9), we know that when the angle observation error is small enough to linearize the error signal, that is, at this time The shaft current equation is expressed as: Where: Step 1.4, high frequency signal demodulation and rotor position observation; From formula (1-10), we know that adjusting hour, That is, the rotor position converges to the actual position value, The signal demodulation process of the high-frequency square wave injection method is to perform differential processing on the sampled q-axis current to obtain the current difference signal containing the rotor position error, modulate the amplitude to obtain the rotor position error signal, and finally estimate the rotor error signal through a phase-locked loop to obtain the converged observed rotor angular velocity. After integration, the observed rotor angle is obtained 3. The control method for low-temperature and high-torque starting of a brushless motor of a position sensorless oil pump according to claim 1, characterized in that: The step 2 is specifically as follows: Step 2.1, constructing the extended back-EMF model, that is, constructing the voltage equation under the two-phase stationary coordinate system α-β axis system and deriving the relationship of the extended back-EMF; The oil pump motor voltage equation in the dq coordinate system is as follows (2-1): By changing the coordinates of the voltage equation in the dq coordinate system, we can obtain the voltage equation in the stationary coordinate system: Among them, u α,β ——stator voltage component under α-β axis system; i α,β ——stator current component in α-β axis system; L α,β ——Inductance under α-β axis; R s ——stator resistance; ω——rotor electrical angle; ——permanent magnet flux; p——differential operator; where, In order to expand the back electromotive force, equation (2-2) is simplified to the stationary coordinate system: in, E α,β ——The corresponding extended back electromotive force component under the α-β axis system; Step 2.2, obtaining the estimated extended back electromotive force through the PI regulator; This is achieved using a PI regulator. In formula (2-4), is the observed value of the rotor angular velocity ω, for i α The observed value of for i β The observed value of For E α The observed value of For E β The observed value. Using the PI regulator to observe the back electromotive force, it is expressed as: In the formula, K p is the proportional gain coefficient of the PI regulator, K i is the integral gain coefficient of the PI regulator; Step 2.3, rotor position observation; From (2-5) we can get the estimated And the observed rotor position is obtained by inverse tangent calculation In the back electromotive force estimation link, compensation θ is performed based on the observations in formula (2-6) com , Where: is the observed angular velocity, ω c is the bandwidth of the back EMF estimation link, The observation angle after compensation is: