A position sensorless control method for brushless DC motor based on model predictive control

Abstract

The present invention discloses a brushless DC motor position sensorless control method based on model predictive control. Firstly, the rotor position angle and speed information of the motor are obtained by constructing a brushless DC motor model reference adaptive system, and then the q-axis current reference value i is obtained by using the speed outer loop PI regulator. q ref ; Then, the d-q axis component i of the stator current at time (k+1) is predicted online according to the prediction model d (k+1), i q (k+1); further, a value function is constructed using the reference value and predicted value of the stator current d‑q axis component at time (k+1), and the optimal voltage vector is selected by minimizing the value function, thereby obtaining the optimal switching state of the inverter; the brushless DC motor control method provided by the present invention adopts a finite control set model predictive control strategy, does not need to adjust the current inner loop parameters, can obtain better dynamic and steady-state performance, and realizes position sensorless operation, thereby improving system reliability and reducing hardware costs.

Description

Technical Field

[0001] The invention relates to a position sensorless control method for a brushless DC motor based on model predictive control, and belongs to the field of motor drive and control. Background Art

[0002] DC motors have many advantages such as good speed regulation performance, high operating efficiency, and simple control. They have been dominant in the field of electrical transmission for a long period of time before the 1970s. However, the presence of its mechanical contact commutator and brushes brings fatal weaknesses such as sparks, noise, electromagnetic interference, and short life, resulting in poor operating reliability and troublesome maintenance of DC motors. In addition, the high manufacturing cost greatly limits its application range. With the development of power electronics technology, microelectronics technology, and computer technology, especially the progress of vector control technology, high-performance AC induction motor variable frequency speed regulation systems have been widely used. AC speed regulation systems are gradually replacing DC speed regulation systems and occupying a dominant position in the field of motion control. It is well known that although the performance of AC servo drive systems is comparable to that of DC servo systems, they require complex coordinate transformations and are quite complex to control. Therefore, it has become an important task for many scientific and technological workers to find a new type of motor that has the advantages of simple DC motor control, good speed regulation performance, and high operating efficiency, as well as the advantages of asynchronous motors such as solid structure, no brushes, and easy maintenance. With the development of power electronic devices, inverter control technology and new permanent magnet materials, the brushless direct current motor (BLDC), which uses electronic commutation and combines the advantages of the above-mentioned DC motor and AC asynchronous motor, has been widely used since its introduction.

[0003] At present, brushless DC motors have technical advantages that cannot be replaced by other motors, including: low noise, low electromagnetic interference, high efficiency, high torque density, stable operation, long life, high reliability, small size, low power consumption, easy maintenance, simple control, low price, etc. Therefore, brushless DC motors are replacing other types of motors in many occasions, such as DC motors used in optical drives, hard disks and disks in computer systems, DC motors that need to meet the requirements of low noise, high temperature resistance, and the ability to withstand certain vibrations and shocks in communication equipment, industrial automation systems, aerospace, weapons and equipment, and medical equipment have gradually been replaced by precision permanent magnet brushless DC motors. Asynchronous motors used in air conditioners, refrigerators, and washing machines in household appliances, and AC asynchronous motors used to drive electric vehicles are also being replaced by brushless DC motors.

[0004] In the sine wave control strategy of brushless DC motor, the software algorithm is more complicated than square wave control, but it can obtain smoother torque output performance. This control method requires knowing the precise rotor position, so some sensors coaxially connected to the motor should be installed on the motor, such as switch-type Hall sensors, photoelectric encoders, rotary transformers, etc. to detect the rotor position, making the spatial coordinate system transformation possible. The position sensor is also the most vulnerable part of the permanent magnet brushless DC motor system, which will bring a series of problems, such as increasing system cost and complexity, and being unsuitable in harsh environments. Summary of the invention

[0005] Purpose of the invention: In view of the above-mentioned prior art, a position sensorless control method for a brushless DC motor based on model predictive control is proposed. The rotor position signal and speed information of the motor are obtained through a model reference adaptive system, avoiding the installation of traditional mechanical sensors. The model predictive control strategy is adopted to avoid the influence of the traditional vector control current inner loop and parameter setting on the accuracy of rotor position angle estimation, thereby promoting the practical application of model predictive control in motor control systems.

[0006] Technical solution: A brushless DC motor position sensorless control method based on model predictive control, comprising the following steps:

[0007] Step 1, determining an adjustable model and a reference model of a model reference adaptive system;

[0008] Step 2: Design a reference adaptive law of the model reference adaptive system to obtain rotor position and speed information;

[0009] Step 3: Obtain the q-axis current reference value i according to the speed outer loop PI controller q ref ;

[0010] The q-axis current reference value i is obtained by using the speed outer loop PI controller q ref , and give the d-axis current reference value i d ref =0;

[0011] Step 4: Calculate the predicted dq axis current value i at time (k+1) d (k+1), i q (k+1);

[0012] Use the current sensor to obtain the three-phase stator current i of the motor at time k abc (k), and combined with the rotor position information obtained by the model reference adaptive system to i abc (k) coordinate transformation is performed to obtain the stator current dq axis component i at time k d (k), iq (k), calculate the predicted value i of the stator current dq axis component at time (k+1) according to the prediction model and combined with the current equation d (k+1), i q (k+1);

[0013] Step 5: Select the optimal voltage vector u through rolling optimization of the value function opt ;

[0014] A value function is constructed according to the reference value and predicted value of the dq-axis component of the stator current at time (k+1). The voltage vector of the brushless DC motor is brought into the value function for rolling optimization, and the optimal voltage vector is selected to obtain the optimal switching state of the inverter.

[0015] Furthermore, the method of determining the adjustable model and the reference model of the model reference adaptive system in step 1 is as follows:

[0016] According to the brushless DC motor current equation (1), and define i d '=i d +ψ f / L s ,i q '=i q ,u d '=u d +R s ·ψ f / L s ,u q '=u q , we get the reference model of the model reference adaptive system, as shown in formula (2). We express formula (2) with estimated values ​​to get the adjustable model (3):

[0017]

[0018]

[0019]

[0020] Where u is voltage; i is current; ψ is flux; the two subscripts dq represent the direct axis and quadrature axis coordinate components of the stator respectively; ω e is the rotor angular velocity; R s is the stator winding resistance; L s is the equivalent synchronous inductance of the stator winding; ψ s and ψ f are the stator flux vector and the rotor flux vector respectively; is the estimated value of dq axis current,

[0021] Furthermore, the step 2 designs a reference adaptive law of the model reference adaptive system to obtain rotor position and speed information, as follows:

[0022] Popov's hyperstability theory is used to design the reference adaptive law (4), thereby obtaining the estimated rotor angular velocity Using formula (5) Integrate to get the estimated rotor position angle By using the reference adaptive law, the model reference adaptive system can make the estimated value gradually approach the reference value;

[0023]

[0024]

[0025] Where k p and k i They are the proportional gain and integral gain of the PI regulator respectively; τ is the time constant.

[0026] Furthermore, in step 3, the q-axis current reference value i is obtained according to the speed outer loop PI controller. q ref The method is as follows:

[0027] The reference speed n ref The difference between the actual speed n obtained by the model reference adaptive system and the actual speed n n Input the speed outer loop PI controller, and obtain the q-axis current reference value i according to formula (6): q ref ;

[0028]

[0029] In the formula, k p and k i They are the proportional gain and integral gain of the voltage regulator respectively, and s represents a complex variable.

[0030] Furthermore, the step 4 calculates the predicted dq axis current value i at time (k+1) d (k+1), i q The (k+1) method is as follows:

[0031] Use the current sensor to obtain the three-phase stator current i of the motor at time k abc (k), and combined with the rotor position information obtained by the model reference adaptive system to i abc (k) coordinate transformation is performed to obtain the stator current dq axis component i at time k d (k), i q(k), and then the current differential equation is discretized according to the first-order Euler equation shown in formula (7), and the dq axis current prediction value i at time (k+1) is obtained according to formula (8): d (k+1), i q (k+1);

[0032]

[0033]

[0034] Where y(k+1) and y(k) represent the states at time k+1 and time k respectively; T s is the sampling period of the system; u d (k) and u q (k) is the dq-axis component of the stator voltage at time k.

[0035] Furthermore, the step 5 selects the optimal voltage vector u by rolling optimization of the value function opt The method is as follows:

[0036] First, let the output i of formula (6) at time k be q ref =i q ref (k+1), and define i d ref (k+1)=0, the stator current dq axis component amplitude reference value i at time (k+1) d ref (k+1), i q ref (k+1) and predicted value i d (k+1), i q (k+1) Input cost function (9), and substitute the basic voltage vector u of the brushless DC motor in turn i , calculate the value function output g under different i i ; Then use formula (10) to obtain the minimized value function output g min ; The final output satisfies g min The basic voltage vector u opt (u opt The corresponding switch state is S x (min), min = 1, 2, ..., 8);

[0037]

[0038] g min =min{g i},i=1,2,...,8 (10).

[0039] Beneficial effects: The present invention is based on a brushless DC motor powered by a two-level inverter, and adopts a model reference adaptive method to obtain the rotor position signal and speed information of the motor, avoiding the installation of traditional mechanical sensors, realizing position sensorless operation, reducing the space volume, and increasing the reliability of the system operation under specific environments; and adopts a finite control set model predictive control, avoiding the complex parameter setting process of the traditional vector control current inner loop, and effectively improving the dynamic and steady-state performance of BLDC. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of a brushless DC motor position sensorless control method based on model predictive control of the present invention;

[0041] Figure 2 It is a flow chart of the method for controlling a brushless DC motor without a position sensor based on model predictive control of the present invention;

[0042] Figure 3 A curve showing a change in speed estimation value and actual value of a position sensorless control method for a brushless DC motor based on model predictive control provided by the present invention;

[0043] Figure 4 A change curve between the estimated value and the actual value of the rotor position in the brushless DC motor position sensorless control method based on model predictive control provided by the present invention;

[0044] Figure 5 The present invention provides a curve showing the variation of the rotor position estimation error of the brushless DC motor position sensorless control method based on model predictive control. DETAILED DESCRIPTION

[0045] The present invention will be further explained below in conjunction with the accompanying drawings.

[0046] A brushless DC motor position sensorless control method based on model predictive control is based on a system such as Figure 1 As shown, it includes a speed outer loop PI regulator 1, a value function module 2, an inverter 3, a brushless DC motor module 4, a coordinate transformation module 5, a model predictive control module 6, and a model reference adaptive system module 7;

[0047] The speed outer loop PI regulator 1 collects the reference speed n refAfter processing the difference between the actual speed n obtained by the model reference adaptive system, the data is transmitted to the value function module 2 and the model reference adaptive system module 7 respectively. The coordinate transformation module 5 collects the three-phase stator current of the brushless DC motor module 4 through the current sensor, and also collects the rotor position angle estimation value processed by the model reference adaptive system module 7. After processing by the coordinate transformation module 5, the data is processed by the model prediction control module 6 and then transmitted to the value function module 2. After processing the received data, the value function module 2 controls the operation of the brushless DC motor module 4 by controlling the inverter 3.

[0048] like Figure 2 As shown, the method comprises the following steps:

[0049] Step 1: Determine the adjustable model and reference model of the model reference adaptive system:

[0050] According to the brushless DC motor current equation (1), and define i d '=i d +ψ f / L s ,i q '=i q ,u d '=u d +R s ·ψ f / L s ,u q '=u q , we get the reference model of the model reference adaptive system, as shown in formula (2). We express formula (2) with estimated values ​​to get the adjustable model (3):

[0051]

[0052]

[0053]

[0054] Where u is voltage; i is current; ψ is flux; the two subscripts dq represent the direct axis and quadrature axis coordinate components of the stator respectively; ω e is the rotor angular velocity; R s is the stator winding resistance; L s is the equivalent synchronous inductance of the stator winding; ψ s and ψ f are the stator flux vector and the rotor flux vector respectively; is the estimated value of dq axis current,

[0055] Step 2: Design the reference adaptive law of the model reference adaptive system to obtain the rotor position and speed information:

[0056] Popov's hyperstability theory is used to design the reference adaptive law (4), thereby obtaining the estimated rotor angular velocity Using formula (5) Integrate to get the estimated rotor position angle By using the reference adaptive law, the model reference adaptive system can make the estimated value gradually approach the reference value;

[0057]

[0058]

[0059] Where k p and k i They are the proportional gain and integral gain of the PI regulator respectively; τ is the time constant.

[0060] Step 3: Obtain the q-axis current reference value i according to the speed outer loop PI controller q ref :

[0061] The reference speed n ref The difference between the actual speed n obtained by the model reference adaptive system and the actual speed n n Input the speed outer loop PI controller, and obtain the q-axis current reference value i according to formula (6): q ref ;

[0062]

[0063] In the formula, k p and k i They are the proportional gain and integral gain of the voltage regulator respectively, and s represents a complex variable.

[0064] Step 4: Calculate the predicted dq axis current value i at time (k+1) d (k+1), i q (k+1):

[0065] Use the current sensor to obtain the three-phase stator current i of the motor at time k abc (k), combined with the rotor position information obtained in step 2, the coordinate transformation of the three-phase stator current is performed to obtain the dq-axis component i of the stator current at time k d (k) and i q (k), after discretizing the current differential equation according to the first-order Euler equation shown in formula (7), the dq axis current prediction value i at time (k+1) is obtained according to formula (8): d(k+1), i q (k+1);

[0066]

[0067]

[0068] Where y(k+1) and y(k) represent the states at time k+1 and time k respectively; T s is the sampling period of the system; u d (k) and u q (k) is the dq-axis component of the stator voltage at time k.

[0069] Step 5: Select the optimal voltage vector u through cost function rolling optimization opt :

[0070] First, let the output i of formula (6) at time k be q ref =i q ref (k+1), and define i d ref (k+1)=0; set the stator current dq axis component amplitude reference value i at time (k+1) d ref (k+1), i q ref (k+1) and predicted value i d (k+1), i q (k+1) Input cost function (9), and substitute the basic voltage vector u of the brushless DC motor in turn i , calculate the value function output g under different i i ; Then use formula (10) to obtain the minimized value function output g min ; The final output satisfies g min The basic voltage vector u opt (u opt The corresponding inverter switch state is S x (min), min = 1, 2,…, 8).

[0071]

[0072] g min =min{g i},i=1,2,...,8 (10)

[0073] The simulation results of the brushless DC motor position sensorless control system based on model predictive control are as follows: Figure 3 , Figure 4 , Figure 5 shown. Figure 3This is the change curve of the estimated value and actual value of the speed of the brushless DC motor without position sensor control based on model predictive control. Figure 4 This is the variation curve of the estimated value and actual value of the rotor position of the brushless DC motor without position sensor control based on model predictive control. Figure 5 The curve of the rotor position estimation error of the brushless DC motor position sensorless control based on model predictive control is shown in Figure 2. From the simulation results, it can be seen that when the motor speed increases from zero to the reference speed of 600r / min, there is a large error between the speed estimation value and the actual value, but as the speed increases and stabilizes, the speed estimation error gradually decreases, and the rotor position estimation error also gradually decreases. It can be shown that the brushless DC motor position sensorless control technology based on model predictive control can meet the needs of actual motor control performance.

[0074] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for controlling a brushless DC motor without position sensor based on model predictive control, characterized in that: The steps include: Step 1, determining an adjustable model and a reference model of a model reference adaptive system; The method of determining the adjustable model and the reference model of the model reference adaptive system in step 1 is as follows: According to the brushless DC motor current equation (1), and define i d '=i d +ψ f / L s ,i q '=i q ,u d '=u d +R s ·ψ f / L s ,u q '=u q , we get the reference model of the model reference adaptive system, as shown in formula (2). We express formula (2) with estimated values ​​to get the adjustable model (3): Where u is voltage; i is current; ψ is flux; the two subscripts dq represent the direct axis and quadrature axis coordinate components of the stator respectively; ω e is the rotor angular velocity; R s is the stator winding resistance; L s is the equivalent synchronous inductance of the stator winding; ψ s and ψ f are the stator flux vector and the rotor flux vector respectively; is the estimated value of dq axis current, Step 2: Design a reference adaptive law of the model reference adaptive system to obtain rotor position and speed information; The step 2 designs the reference adaptive law of the model reference adaptive system, and then obtains the rotor position and speed information, as follows: Popov's hyperstability theory is used to design the reference adaptive law (4), thereby obtaining the estimated rotor angular velocity Using formula (5) Integrate to get the estimated rotor position angle Using the reference adaptive law, the model reference adaptive system can make the estimated value gradually approach the reference value; Where k p and k i are the proportional gain and integral gain of the PI regulator respectively; τ is the time constant; Step 3: Obtain the q-axis current reference value i according to the speed outer loop PI controller q ref ; The q-axis current reference value i is obtained by using the speed outer loop PI controller q ref , and give the d-axis current reference value i d ref =0; In step 3, the q-axis current reference value i is obtained according to the speed outer loop PI controller. q ref The method is as follows: The reference speed n ref The difference between the actual speed n obtained by the model reference adaptive system and the actual speed n n Input the speed outer loop PI controller, and obtain the q-axis current reference value i according to formula (6): q ref ; In the formula, k p and k i are the proportional gain and integral gain of the voltage regulator respectively, and s represents a complex variable; Step 4: Calculate the predicted dq axis current value i at time (k+1) d (k+1), i q (k+1); Use the current sensor to obtain the three-phase stator current i of the motor at time k abc (k), and combined with the rotor position information obtained by the model reference adaptive system to i abc (k) coordinate transformation is performed to obtain the stator current dq axis component i at time k d (k), i q (k), calculate the predicted value i of the stator current dq axis component at time (k+1) according to the prediction model and combined with the current equation d (k+1), i q (k+1); The step 4 calculates the predicted dq axis current value i at time (k+1) d (k+1), i q The (k+1) method is as follows: Use the current sensor to obtain the three-phase stator current i of the motor at time k abc (k), and combined with the rotor position information obtained by the model reference adaptive system to i abc (k) coordinate transformation is performed to obtain the stator current dq axis component i at time k d (k), i q (k), and then the current differential equation is discretized according to the first-order Euler equation shown in formula (7), and the dq axis current prediction value i at time (k+1) is obtained according to formula (8): d (k+1), i q (k+1); Where y(k+1) and y(k) represent the states at time k+1 and time k respectively; T s is the sampling period of the system; u d (k) and u q (k) is the dq-axis component of the stator voltage at time k; Step 5: Select the optimal voltage vector u through rolling optimization of the value function opt ; The value function is constructed according to the reference value and predicted value of the stator current dq axis component at the time (k+1), the voltage vector of the brushless DC motor is brought into the value function for rolling optimization, the optimal voltage vector is selected, and then the optimal switching state of the inverter is obtained; the step 5 selects the optimal voltage vector u through the rolling optimization of the value function opt The method is as follows: First, let the output i of formula (6) at time k be q ref =i q ref (k+1), and define i d ref (k+1)=0, the stator current dq axis component amplitude reference value i at time (k+1) d ref (k+1), i q ref (k+1) and predicted value i d (k+1), i q (k+1) Input cost function (9), and substitute the basic voltage vector u of the brushless DC motor in turn i , calculate the value function output g under different i i ; Then use formula (10) to obtain the minimized value function output g min ; The final output satisfies g min The basic voltage vector u opt (u opt The corresponding switch state is S x (min), min = 1, 2, ..., 8); g min =min{g i },i=1,2,...,8(10).

Images