A three-phase permanent magnet synchronous motor full-speed domain position sensorless control method

Abstract

The application discloses a full-speed-range position sensorless control method for a three-phase permanent magnet synchronous motor and belongs to the technical field of power generation, power transformation or power distribution. In the zero-speed and low-speed stage, a constant-current variable-frequency mode is adopted to realize starting; in the medium and high-speed stage, a positionless control method is adopted to operate; and an instant switching algorithm guarantees quick and flexible switching of the low-speed constant-current variable-frequency operation and the medium and high-speed positionless control. The positionless control in the medium and high-speed stage is composed of a back electromotive force real-time calculation algorithm, back electromotive force normalization and a quadrature phase-locked loop; the response speed of the back electromotive force real-time calculation algorithm is consistent with the switching frequency, and a PI controller in the quadrature phase-locked loop realizes flexible switching of high and low speeds according to a rotating speed switching threshold. The full-speed-range positionless control algorithm only needs two motor parameters of a resistance and a cross-axis inductance, does not need a filter, has the characteristics of simplicity and reliability, good dynamic performance and low requirement on the performance of a processor, and the application range of the algorithm can cover medium and low end electric drive application scenarios.

Description

Technical Field

[0001] This invention discloses a sensorless control method for a three-phase permanent magnet synchronous motor across the entire speed range, belonging to the technical field of power generation, transformation, or distribution. Background Technology

[0002] Three-phase permanent magnet synchronous motors (PMSMs) have a simple structure and excellent speed regulation performance, making them promising for applications in many fields such as home appliances, wind power generation, electric vehicles, electric ships, and aerospace. However, PMSMs typically employ field-oriented technology for speed regulation, which is highly dependent on mechanical rotor position sensors. On the one hand, mechanical rotor position sensors are relatively expensive, limiting their application in cost-sensitive areas such as washing machines and air conditioners. On the other hand, mechanical rotor position sensors have specific requirements regarding ambient temperature, humidity, and installation methods, further restricting their practical application in most situations. Considering all these factors, eliminating rotor position sensors is crucial for the development of three-phase PMSM electric drive systems. In this context, sensorless control technology has emerged, also known simply as positionless control technology. Currently, positionless control systems have been successfully applied in the home appliance field, but the following three problems urgently need to be solved: (1) Slow speed regulation performance, because a large number of observers and filters are used to observe the rotor position, resulting in a serious degradation in speed regulation performance compared to position control. The introduction of filters often requires phase compensation; (2) Fluctuations in the observed rotor position, because with the extensive use of observers and filters, impedance matching between functional modules is mismatched, causing nonlinear oscillations; (3) Difficulty in rapid switching between high and low speed sensorless control, which is essentially due to the slow dynamic response of the rotor position observer. Therefore, in order to achieve positionless control across the entire speed range and expand the application range of three-phase permanent magnet synchronous electric drive systems, it is urgent to develop high-dynamic positionless control technology.

[0003] The existing full-speed domain positionless control technology has the following three defects: (1) In the low-speed domain, the high-frequency pulsation signal injection method is usually used, but the high-frequency pulsation signal injection method in the low-speed domain will generate high-frequency noise and electromagnetic interference, which will bring additional heat loss to the motor; (2) When transitioning from the low-speed domain to the medium-speed domain, it is necessary to comprehensively consider the effective position information obtained by the high-frequency pulsation signal injection method and the position signal observed in the medium-speed domain to obtain the given speed in the medium-speed domain. On the one hand, the multiple filters required by the high-frequency pulsation signal injection method to extract effective position information affect the dynamic response performance of the entire control method. On the other hand, the jitter problem of the sliding mode observation method commonly used in the medium-speed domain affects both the dynamic response performance of the entire control method and the smooth transition of the algorithm; (3) Dividing the speed range into three ranges: zero low speed, low speed, and medium-speed, and using different positionless algorithms for each speed range, and using the corresponding positionless algorithm after determining the range to which the initial speed of the motor belongs, the switching of positionless control algorithms in different speed ranges brings instability to the system. The coding work added by applying different positionless algorithms increases the computational burden of the digital controller, and the algorithm is complex.

[0004] The present invention aims to propose a sensorless control method for a three-phase permanent magnet synchronous motor across the entire speed range to overcome the above-mentioned defects. Summary of the Invention

[0005] The purpose of this invention is to address the limitations of the aforementioned background technology by proposing a sensorless control method for a three-phase permanent magnet synchronous motor across the entire speed range. This method abandons the approach of dividing different speed ranges and then switching to different sensorless control algorithms based on real-time speed. Instead, it first uses a ramp-start speed variable frequency drive to start the motor, and then employs a back-EMF model to achieve full-speed-range sensorless control. This achieves the goal of dynamic response sensorless control and smooth adjustment of motor speed, solving the technical problems of existing full-speed-range sensorless control technologies for permanent magnet synchronous motors, such as complex algorithms and difficulty in balancing dynamic response and steady-state accuracy.

[0006] To achieve the aforementioned objectives, this invention employs the following technical solution: It utilizes a real-time back EMF calculation algorithm, back EMF normalization, and an orthogonal phase-locked loop (PLL) considering switching timing to achieve positionless control across the entire speed domain. The real-time back EMF calculation algorithm does not include a current differential term and effectively reduces dependence on motor parameters. It achieves deadbeat-free back EMF observation without iteration, solving the problems of high computational load, slow dynamic response, and complex calculations in current back EMF observation algorithms. Furthermore, the observed back EMF values ​​are directly normalized without using filters, and the rotor speed and position are calculated using the PLL system. The PLL technology of this invention considers the switching timing between low-speed and medium-to-high-speed positionless control, enabling the motor to start from zero speed or extremely low speed and instantly switch to positionless control in the medium-to-high-speed domain.

[0007] Firstly, this invention utilizes a real-time back electromotive force calculation algorithm for a three-phase PMSM, the specific expression of which is:

[0008]

[0009] In equation (1), i α i β These are the components of the motor stator current along the α and β axes of the stationary coordinate system (α-β coordinate system); u α u β These are the components of the motor input voltage along the α and β axes of the stationary coordinate system, respectively; L q For quadrature axis inductance; R s ω is the resistance of each phase winding of the motor stator; e e is the electric angular velocity; α e β Let be an extended back electromotive force in the α-β coordinate system, and satisfy:

[0010]

[0011] In equation (2), L d For a direct-axis inductor, i d i q Let be the components of the motor stator current along the d-axis and q-axis of the dq coordinate system, respectively, where p is the number of pole pairs of the motor, and ψ is the component of the motor stator current along the d-axis and q-axis. f For permanent magnet flux linkage, θ e It represents the position information of the motor rotor.

[0012] Furthermore, the amplitude of the extended back electromotive force E is calculated. amp :

[0013]

[0014] Furthermore, regarding the extended back electromotive force e α e β Perform per-unit processing:

[0015]

[0016] Furthermore, the per-unit extended back electromotive force is input into the phase-locked loop (PLL), and the rotor information output by the PLL is used for the following calculations:

[0017]

[0018] In equation (5), The rotor position angle extracted by the phase-locked loop is due to:

[0019]

[0020] Therefore, under steady-state conditions, Δθ can be approximated as the output rotor position angle of the phase-locked loop. With the extended back electromotive force observation position angle θ e The difference between them. The phase-locked loop can achieve the observation position angle θ of the extended back electromotive force. e It exhibits zero steady-state error tracking and good steady-state performance without position control.

[0021] When the motor operates at zero speed or extremely low speed, the back electromotive force and stator current are very small, resulting in a very low signal-to-noise ratio of the useful signal. This leads to large errors in rotor position and speed detection, making the motor prone to loss of synchronization or even starting failure during startup. To solve the above problems, this invention proposes a method for smooth transition between the startup phase and the medium-to-high speed phase, which can be specifically described as follows:

[0022] During the startup phase, a given ramp speed ω is generated. start This overrides the output of the phase-locked loop's PI controller, while limiting the speed PI output amplitude to achieve low-speed constant-current start-up. Given the ramp speed ω... start It is a value that increments from 0, and after passing through the integration stage of the phase-locked loop, it yields the given changing position angle θ. start θ start It's just a virtual rotor position angle and doesn't reflect the actual rotor position during startup; its function is to generate a starting rotating magnetic field. Because the angular velocity of this rotating magnetic field is related to ω... start Equal, and ω start The speed increases from zero to low speed due to ω start The change is slow, so even if the provided θ start Even without a precise position signal, the motor can still follow ω. start start.

[0023] To enable the motor to switch from the starting stage to the medium-high speed stage, a switching signal K and a speed switching threshold ω need to be set. switch ω switch A value of 10–30 rad / s is acceptable. If the value is too small, the back electromotive force will be very small, and the estimation error of the rotor information obtained after switching will still be large, easily causing start-up failure. If the value is too large, due to θ… start At higher speeds, the error between the actual position angle and the actual position angle is large, and the motor speed has not yet reached ω. switch There is a risk of losing synchronization. When ω start <ω switch When the motor is in the starting phase, the switching signal K = 0. When ω start >ω switch When this signal is received, it indicates that the start-up phase is complete, triggering the switching signal K=1. Upon receiving this signal, the phase-locked loop system... start Instead of overriding the PI output of the phase-locked loop, ω is... startThe value is assigned to the integral initial value of the PI controller, and the PI output of the phase-locked loop is... From ω start Begin self-regulation. Output after integration It will also autonomously adjust to gradually approach the true rotor position angle; at the same time, it will further relax the output limit of the speed PI controller.

[0024] Because when the switching occurs, the output of the phase-locked loop PI changes from ω. start This allows for a smooth transition between the two modes, without any issues arising from the initial setup. With ω switch Significant differences can cause problems such as chattering during switching.

[0025] After the switch is complete, the motor enters the medium-to-high speed range, and the speed signal estimated by the phase-locked loop... It functions on two fronts: real-time calculation of back electromotive force and closed-loop speed regulation. This is achieved by changing the speed reference value ω of the speed loop. ref To achieve speed control of the motor.

[0026] Furthermore, the phase-locked loop includes:

[0027] The phase detector receives rotor position angle observation information obtained by normalizing and expanding back EMF at one input terminal and receives target rotor position angle signal at the other input terminal, and outputs the difference between the target rotor position angle and the rotor position angle obtained by normalizing and expanding back EMF.

[0028] The PI controller has its first input terminal connected to the output terminal of the phase detector, its second input terminal receiving the switching signal, and its third input terminal receiving the ramp speed signal when the switching signal indicates that the given ramp speed signal has reached the switching threshold.

[0029] A switching switch has one input terminal connected to the output terminal of a PI controller, and the other input terminal receiving a given ramp speed signal. Its control terminal receives a switching signal. When the switching signal indicates that the given ramp speed signal has not reached a switching threshold, the given ramp speed signal is transmitted to the input terminal of the integrator and the outer speed loop. When the switching signal indicates that the given ramp speed signal has reached the switching threshold, the PI controller output signal is transmitted to the input terminal of the integrator and the outer speed loop.

[0030] The integrator, whose input is connected to the output of the switching switch, outputs the target rotor position angle signal to the inner current loop and the outer speed loop.

[0031] The present invention, by adopting the above technical solution, has the following advantages:

[0032] (1) This invention combines a real-time back EMF calculation algorithm with phase-locked loop (PLL) technology, combining the advantages of both to avoid the current differential error caused by the traditional extended back EMF observation algorithm. It only requires two motor parameters, resistance and quadrature-axis inductance, effectively reducing the dependence on motor parameters. It can achieve deadbeat observation of back EMF without iteration. It directly normalizes the back EMF observation value without using a filter, which overcomes the defects of large calculation volume, slow dynamic response and complex operation of the current back EMF observation algorithm. It also solves the problem of phase delay error caused by low-pass filtering. After processing the normalized EMF through the PLL, the rotor position angle information can be obtained, which can estimate a more accurate rotor position information, thus improving the steady-state performance and dynamic performance of the positionless control system.

[0033] (2) This invention uses ramp-given speed frequency conversion starting at zero or very low speed, which has the advantage of high dynamic response compared to the high frequency signal injection method. The phase-locked loop technology used takes into account the switching timing of low speed and medium-high speed without position control, which can realize the motor starting from zero or very low speed and switching to medium-high speed domain without position control in real time. The software algorithm is simpler, greatly reducing code redundancy and complexity, and reducing the computational burden of digital controller. It has the characteristics of simple algorithm, dynamic response and steady-state accuracy, and is suitable for programming low-end digital controllers. It can effectively expand the application scope of rotorless position control technology.

[0034] (3) The flexible switching method from the starting stage to the medium-high speed stage proposed in this invention can realize the instantaneous transition from the starting stage to the medium-high speed stage. It can even set a high target speed when the motor starts, and the motor can complete the transition from starting to the target speed continuously, ensuring the speed without position control.

[0035] (4) The positionless control method proposed in this invention is applicable not only to salient pole motors but also to permanent magnet synchronous motors, and has universality in positionless control of permanent magnet synchronous motors. Attached Figure Description

[0036] Figure 1 This is a system block diagram of a three-phase PMSM without position control.

[0037] Figure 2 This is a schematic diagram of an extended back electromotive force real-time calculation algorithm used in this invention.

[0038] Figure 3 This is a schematic diagram of the phase-locked loop algorithm used in this invention.

[0039] Figure 4 This is an experimental waveform diagram of the motor starting from zero speed and accelerating to 200 rpm.

[0040] Figure 5This is an experimental waveform diagram showing the motor speed increasing from 200 rpm to 750 rpm.

[0041] Figure 6 This is an experimental waveform diagram showing the motor speed decreasing from 750 rpm to 200 rpm.

[0042] Figure 7 This is the waveform of the extended back electromotive force when the motor speed is 200 rpm.

[0043] Figure 8 This is the waveform of the extended back electromotive force when the motor speed is 750 rpm.

[0044] Figure 9 It is a waveform diagram of the motor speed starting from zero speed, increasing to 200 rpm and then to 750 rpm.

[0045] Figure 10 It is a waveform diagram of the motor speed decreasing from 750rpm to 200rpm and then to 0rpm. Detailed Implementation

[0046] To make the implementation method and technical advantages of the present invention clearer and easier to understand, the technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0047] The system block diagram of a three-phase PMSM without position control is as follows: Figure 1 As shown, the closed-loop control of this system mainly includes an outer speed loop, an inner current loop, and a phase-locked loop. Given a speed ω... ref Input to the outer speed ring, and feedback speed After the difference is calculated, the output i is obtained through the speed PI regulator. qref The instruction value, i dref The command value is set to 0. The command value for the dq axis is i. dref i qref The current is fed into the inner current loop and interacts with the feedback current i. d i q After the difference is calculated, the output control voltage u is obtained through the current PI regulator. d and u q u d and u q Based on the observed position angle The u in the α-β coordinate system is obtained through the inverse Park transform. α and u β u α and u β Using appropriate PWM techniques (such as SVPWM, space vector pulse width modulation), the variable u in the three-phase stator coordinate system is obtained. A u B u CAfter comparison with the drive circuit, a drive signal is generated to control the inverter switching transistors to turn on / off, thereby generating a three-phase current i to control the motor's motion state. A i B i C For three-phase current i A i B i C Sample and perform Clark transformation to obtain i in the α-β coordinate system α and i β On the one hand, i α and i β Based on the observed position angle After Park transformation, i in the dq coordinate system is obtained. d i q Then i d i q The feedback current is fed into the current PI regulator to achieve current closed-loop control. On the other hand, i α i β u α and u β Together, they serve as input to the extended real-time back EMF calculation algorithm, along with the observed rotational speed. Extended back electromotive force e α and e β It will be calculated in real time. The back electromotive force e will be extended. α and e β After standardization, the trigonometric function sinθ containing rotor position information is obtained. e and cosθ e sinθ e and cosθ e The data is fed into a PLL, which extracts and estimates the rotational speed. On the one hand, estimate the rotational speed Feedback will be sent to the speed PI controller to achieve closed-loop speed control. On the other hand, the speed will be estimated. The observed position angle is obtained through integration. The starting and medium-to-high speed switching system is used to control the motor's zero-speed or extremely low-speed starting and to achieve a flexible switching to the medium-to-high speed stage after starting. The estimated speed is output by the PLL. and the integrated observation position angle These are the two most critical parameters in a positionless control system, and the accuracy of their estimation directly affects the performance of the positionless control system.

[0048] The principle of the extended back electromotive force real-time calculation algorithm used in this invention is as follows: Figure 2 As shown. Taking a salient-pole PMSM as an example, the extended back EMF observation model is rewritten:

[0049]

[0050] From equation (8), it can be seen that in order to obtain the extended back electromotive force e α and e β The variable that needs to be input is u. α u β i α i β and ω e , where ω e Rotational speed that can be estimated by a phase-locked loop To obtain, that is u α and u β u output by the current PI regulator d and u q After the inverse Park transform, the expression is:

[0051]

[0052] In equation (9) Position angle observed by the phase-locked loop i α and i β The expression is obtained by sampling the three-phase current of the motor and performing a Clark transform:

[0053]

[0054] After obtaining the input variables required for the extended back electromotive force real-time calculation algorithm, according to Figure 2 The extended back EMF real-time calculation algorithm shown can obtain the observed value of the extended back EMF. and Its specific description is as follows:

[0055] The first multiplier M1 measures the motor current i in the α-β coordinate system. α and stator winding resistance R s Perform multiplication operations and output the product result to the first adder A1; the second multiplier M2 calculates the motor current i in the α-β coordinate system. β and stator winding resistance R s Perform multiplication and output the product to the third adder A3; the third multiplier M3 estimates the electrical angular velocity of the phase-locked loop. and cross-axis inductance L q Perform multiplication and output the product to the fourth multiplier M4; the fifth multiplier M5 multiplies i... αThe first adder A1 performs a multiplication operation on the product outputs of the first and fourth multipliers, and outputs the product to the third adder A3; the second adder A2 performs a summation operation on the product outputs of the first and fourth multipliers, and outputs the sum to the second adder A2; the third adder A3 performs a summation operation on the product outputs of the second and fifth multipliers, and outputs the sum to the fourth adder A4; the second adder A2 performs a multiplication operation on the summation outputs of the first adder and the voltage u in the α-β coordinate system. α Perform cumulative calculations and output the observed values ​​of the extended back electromotive force in the α-β coordinate system. The fourth accumulator sums the output of the third accumulator and the voltage u in the α-β coordinate system. β Perform cumulative calculations and output the observed values ​​of the extended back electromotive force in the α-β coordinate system.

[0056] From this algorithm, it is clear that the current differential term in the traditional extended back electromotive force is reduced by ω. e i α or ω e i β This method avoids differential errors and iterative calculations, resulting in more accurate estimation of position angle information.

[0057] The principle of the phase-locked loop algorithm used in this invention is as follows: Figure 3 As shown, its specific description is as follows:

[0058] The sixth multiplier M6 outputs to the extended back EMF observer. Take the inverse of the sum of the magnitudes of the back electromotive force and the extended back electromotive force. The purpose of performing multiplication is to standardize the extended back electromotive force (EMF), thereby converting it into a trigonometric function containing position angle information. Similarly, the seventh multiplier M7 outputs the extended back EMF observer. and the reciprocal of the magnitude of the extended back electromotive force Perform a multiplication operation and output the product as follows: The result of the eighth multiplier M8 on the output of the sixth multiplier The observation position angle output by the phase-locked loop Take the result of the cosine function Perform multiplication. The result of the ninth multiplier M9 on the output of the seventh multiplier. The observation position angle output by the phase-locked loop Take the result of the sine function Perform multiplication. The fifth adder A5 performs a subtraction operation on the outputs of the eighth and ninth multipliers, and then calculates the result. The input is fed into the proportional-integral (PI) controller of the phase-locked loop (PLL) system. The output of the PI controller is the estimated rotational speed of the PLL system. When the switching signal K=1 of the phase-locked loop system, the rotational speed is estimated. The observation position angle is output after integration. When the switching signal K=0 of the phase-locked loop system, the control system is still in the startup phase, and the starting given ramp speed ω generated inside the phase-locked loop system... start It will replace the estimation of rotational speed Entering the integration phase. When the switching signal K=1 of the phase-locked loop system, the phase-locked loop system... arrive The transfer function can be expressed as:

[0059]

[0060] In formula (11) ω n The bandwidth of the PI controller is determined. The performance of the phase-locked loop (PLL) is often affected by the dynamic performance of the speed loop and the current loop, for K... p K i The selection of K needs to be considered from the perspective of the entire control system. p If the value is too small, the estimated speed change will not keep up with the speed loop's response speed, leading to system malfunction; K p If the value of K is too large, the accuracy of the estimated rotational speed will differ significantly from the actual rotational speed, resulting in reduced steady-state performance. i Generally, a larger value is selected to ensure the reliability of position control without large speed step signals.

[0061] The zero-speed or ultra-low-speed constant current frequency conversion starting method and the flexible switching method from the starting stage to the medium-high speed stage proposed in this invention are specifically described as follows:

[0062] During startup, the phase-locked loop (PLL) switching signal K=0, outputting a limiting signal to the speed PI regulator, thereby limiting the starting current. Since the PI regulator saturates during startup, the output current remains constant. Furthermore, the starting given ramp speed ω is generated internally by the PLL system. start The increment operation can be represented as:

[0063] ω start =ω start ′+at k (12)

[0064] In the above formula, ω start ' represents the given rotational speed value obtained from the processor's previous calculation, t k Let ω be the processor's computation cycle, and 'a' be an incrementing coefficient. With a constant computation cycle, the value of 'a' determines ω. start The growth rate.

[0065] During the startup phase, the observed position angle output by the phase-locked loop... This is used to instruct the SVPWM algorithm to generate a circular rotating magnetic field in the motor, with the magnetic field rotation speed equal to ω. start / p, where p is the number of pole pairs of the motor. Since the rotational speed of the magnetic field increases slowly from 0, it can drive the motor to start following the rotating magnetic field.

[0066] Set ω start When the speed is equal to A (where A is the speed switching threshold, which can be taken as 10~30 rad / s), the phase-locked loop system enters the switching point and generates a switching signal K=1 internally. At this time, the observed position angle output by the phase-locked loop is... The estimated rotational speed ω replaces the start-up phase start The result is obtained after integration, and the speed PI regulator limit is lifted. During the startup phase, the estimated speed is affected by the very small back electromotive force and stator current. The error is relatively large, so It did not affect the control system, but as the starting speed was established, the actual speed gradually approached ω. start Extended back electromotive force estimation position angle It is also gradually approaching θ start At this time, ω start The numerical value is assigned to the integral initial value of the phase-locked loop PI controller to estimate the rotational speed. From ω start Begin self-regulation. The observation position angle obtained through integration is They also began to extend the estimation of the position angle of the back electromotive force. Smooth transition. Because at the transition point, there is... and Approaching θ start Therefore, the observation position angle It can achieve this with minimal speed fluctuations from θ start Towards Smooth transition.

[0067] To verify the practical effectiveness of this invention, an experimental platform was built based on an STM32 digital controller for verification. Figure 4 The experimental waveforms of a four-pole three-phase PMSM accelerating from 0 rpm to 200 rpm are shown. The waveforms clearly show that during startup, the observed position angle quickly becomes in phase with the actual position angle, demonstrating good reliability of the startup. The transition point from the startup phase to the medium-to-high speed phase is set at 100 rpm. The waveforms show a slight decrease in speed after the observed speed reaches 100 rpm, indicating that the speed tracked by the phase-locked loop changes from ω... start Become Afterwards, the control system still needs some time to make adjustments, but during this adjustment period, the observed speed did not fluctuate, indicating that the switching process was relatively smooth. After entering the medium-high speed stage, the observed speed quickly rose to 200 rpm.

[0068] Figure 5 and Figure 6 Experimental waveforms of a three-phase PMSM accelerating from 200 rpm to 750 rpm and decelerating from 750 rpm to 200 rpm are presented. The two graphs show that the transient process is only about 0.4 s, whether accelerating or decelerating, demonstrating the good dynamic response of the invention during speed regulation. Furthermore, the observed position angle is always in phase with the actual position angle, proving the accuracy of the observed position information and the good steady-state performance of the invention.

[0069] Figure 7 and Figure 8 The extended back electromotive force waveforms of the three-phase PMSM at 200 rpm and 750 rpm are given respectively. As can be seen from the figures... and The waveform is a relatively smooth sine wave. At an observed speed of 200 rpm, the extended back EMF amplitude is approximately 7.5V, and at 750 rpm, it is approximately 25V. Due to the limitations of the digital controller's processing performance, the sampling period of the phase-locked loop (PLL) system is set to 1 ms. At higher speeds, the number of observed position angles updated by the PLL system within one cycle is relatively smaller, resulting in a smaller stepped waveform. Shortening the PLL's sampling period would make the extended back EMF waveform smoother at high speeds.

[0070] Figure 9 and Figure 10 The diagrams show the speeds of a three-phase PMSM as it accelerates from 0 to 200 rpm and then to 750 rpm, as well as the speeds as it decelerates from 750 rpm to 200 rpm and then back to 0 rpm. The diagrams demonstrate that this invention achieves a balance between dynamic response and steady-state accuracy in full-speed-range positionless control, making it valuable for engineering applications in low- to mid-range motor controllers.

[0071] The above embodiments are merely illustrative of the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solutions based on the technical concept proposed in this invention shall fall within the scope of protection of this invention.

Claims

1. A full speed range sensorless control method for a three-phase permanent magnet synchronous motor, characterized in that, a motor is started by a given ramp speed signal, an extended motor back electromotive force is calculated in real time, and a rotor position angle observation information is obtained by normalizing the extended back electromotive force, when the given ramp speed signal does not reach a switching threshold value, the given ramp speed signal is taken as a target speed, a phase-locked loop is used to obtain a target rotor position angle according to the target speed, when the given ramp speed signal reaches the switching threshold value, the ramp speed signal is taken as an integral initial value, a phase-locked loop is used to PI regulate a difference between the target rotor position angle and the rotor position angle obtained by normalizing the extended back electromotive force, and an integral processing is performed on a target speed obtained by the PI regulation to obtain the target rotor position angle; wherein the phase-locked loop comprises: a phase detector, one input end of which receives the rotor position angle observation information obtained by normalizing the extended back electromotive force, and the other input end of which receives a target rotor position angle signal, and outputs a difference between the target rotor position angle and the rotor position angle obtained by normalizing the extended back electromotive force, a PI controller, a first input end of which is connected to an output end of the phase detector, a second input end of which receives a switching signal, and a third input end of which receives the ramp speed signal when the switching signal indicates that the given ramp speed signal reaches the switching threshold value, a switching switch, one input end of which is connected to an output end of the PI controller, the other input end of which receives the given ramp speed signal, and a control end of which receives the switching signal, and the switching switch transmits the given ramp speed signal to an input end of an integrator and a speed outer loop when the switching signal indicates that the given ramp speed signal does not reach the switching threshold value, and transmits an output signal of the PI controller to the input end of the integrator and the speed outer loop when the switching signal indicates that the given ramp speed signal reaches the switching threshold value, and the integrator, an input end of which is connected to an output end of the switching switch, and which outputs the target rotor position angle signal to a current inner loop and a speed outer loop.

2. The full speed range position sensorless control method of a three-phase permanent magnet synchronous motor according to claim 1, characterized by, The real-time calculation of the extended motor back EMF is realized by a back EMF mathematical model as follows, wherein, , are the components of the extended back EMF on the axis and axis of the stationary coordinate system respectively, , are the components of the motor stator current on the axis and axis of the stationary coordinate system respectively, , are the components of the motor input voltage on the axis and axis of the stationary coordinate system respectively, is the resistance of each phase winding of the motor stator, is the electrical angular velocity obtained according to , is the quadrature axis inductance, , is the rotor position angle obtained by normalizing the extended back EMF.

3. The full speed range position sensorless control method of a three-phase permanent magnet synchronous motor according to claim 2, characterized by, The specific expression for obtaining the rotor position angle observation information through normalized extended back-EMF is: , wherein, , is the rotor position angle observation information obtained through normalized extended back-EMF, is the amplitude of the extended back-EMF.

4. The full speed range position sensorless control method of a three-phase permanent magnet synchronous motor according to claim 3, characterized by, The difference between the target rotor position angle and the rotor position angle obtained by normalizing the extended back EMF is: wherein, is the difference between the target rotor position angle and the rotor position angle obtained by normalizing the extended back EMF, is the target rotor position angle.

5. The full speed range position sensorless control method of a three-phase permanent magnet synchronous motor according to claim 1, characterized by, The switching threshold value is 10-30 rad / s. 6.The full speed range position sensorless control method of a three-phase permanent magnet synchronous motor according to any one of claims 1 to 5, characterized in that, The specific method for obtaining the target rotor position angle according to the target speed by using the phase-locked loop is to perform integral processing on the target speed. 7.The full speed range position sensorless control method of a three-phase permanent magnet synchronous motor according to any one of claims 1 to 5, characterized in that, When the given ramp speed signal does not reach the switching threshold value, the output amplitude of the PI controller in the speed outer loop is limited.

Images