A dynamic predictive control method of permanent magnet synchronous motor driving system

Abstract

The application discloses a kind of dynamic prediction control methods of permanent magnet synchronous motor drive system, for grid-connected motor side converter and net side converter, the control method is by obtaining actual DC bus voltage reference, then by dynamic reference module generation net side converter net side active power reference, then by evaluating predefined cost function to obtain optimal voltage vector.The application generates active power reference in net side converter using dynamic parameter module, without additional PI controller, and since the cost function includes power constraints, it ensures that the actual DC bus voltage reference remains within a safe range, avoiding excessive charging or discharging current, and also avoiding performance degradation due to the common integral saturation problem in PI control loop.

Description

TECHNICAL FIELD

[0001] The present application relates to power electronics, more particularly to a dynamic predictive control method of permanent magnet synchronous motor drive system. BACKGROUND

[0002] In the prior art, PI-MPC is one of the common solutions to control the back-to-back (BTB) converter-fed PMSG system, which generates active power reference with an additional PI controller for the internal power controller to track; however, the control method requires an additional independent PI controller to form a PI control loop to provide power set value for the internal MPC controller, but the integral saturation problem commonly exists in the PI control loop, which easily leads to the performance degradation of the back-to-back converter.

[0003] In view of this, the present application has carried out in-depth research on the basis, and thus the present case is produced. SUMMARY

[0004] The purpose of the present application is to provide a dynamic predictive control method of permanent magnet synchronous motor drive system, which can limit the maximum power level without additional PI control loop, and avoid performance degradation due to the integral saturation problem commonly existing in the PI control loop.

[0005] To achieve the above purpose, the solution of the present application is:

[0006] A dynamic predictive control method of permanent magnet synchronous motor drive system, the permanent magnet synchronous motor drive system comprising a motor-side converter and a grid-side converter connected in parallel, the control method comprising the following steps:

[0007] S1, obtaining the actual DC bus voltage reference at each moment The expression of the actual DC bus voltage reference is:

[0008] (1),

[0009] In the formula, represents the DC bus voltage reference, represents the step number to reach , and represents the actual value of the DC bus voltage at time k, which is obtained by sampling with a voltage sensor;

[0010] S2, to reach the actual DC bus voltage reference at the next moment, a dynamic reference module is used to generate the grid-side active power reference of the grid-side converter, and the expression of the grid-side active power reference is:

[0011] (2) ,

[0012] Where, represents the d-axis component of the grid-side voltage in the grid-side converter, represents the resistance value in the grid-side converter, Indicates the power consumed by the grid-side resistor in the grid-side converter; is the torque reference; is the actual speed, obtained by sampling the speed sensor; is the bus capacitance parameter, Ts is the sampling period; Represents motor and DC link capacitors Total power;

[0013] Solve and get the grid-side active power reference The expression is:

[0014] (3);

[0015] S3. Select the optimal voltage vector by evaluating a predefined cost function , the cost function is:

[0016] ,

[0017] Where, is the active power at time k+2, is the reactive power weight coefficient, is the reactive power reference, is the reactive power at time k+2, is the bus voltage weight coefficient, is the bus voltage at time k+2; and denote the constraints of active power and reactive power respectively;

[0018] The calculation formula is: , where is the sampling period, is the busbar capacitance value, is the motor side switch signal at time k+1, is the grid-side switch signal at time k+1; is the motor side current at time k+1, is the grid-side current at time k+1;

[0019] S4, obtained through step S3 and , represents the motor side cost function and represents an improved grid-side cost function, and then outputs the estimated motor-side switching state at time k+1 to the motor-side converter and the estimated grid-side switching state at time k+1 to the grid-side converter

[0020] In step S2, the grid-side active power reference is added to , and the grid-side active power reference is expressed as:

[0021] .

[0022] In step S1, the actual DC bus voltage reference at each moment is calculated by using a dynamic reference module prediction range , which is expressed as,

[0023] ,

[0024] wherein represents a reference prediction range used in a steady state, represents an adjustment coefficient, represents a set of positive integers, represents the DC bus voltage at time k.

[0025] After using the above structure, the present application has the following beneficial effects:

[0026] 1. The present application ensures that the actual DC bus voltage reference is kept within a safe range by evaluating a predefined cost function, since the maximum power constraint is also included in the cost function, and overlarge charging or discharging currents are avoided, effectively preventing significant overshoot of the DC link voltage and avoiding performance degradation due to the integral saturation problem commonly found in PI control loops; at the same time, since the active power and the DC link voltage are used to generate a proper reference, no additional voltage control loop is needed.

[0027] 2. The present application uses a dynamic reference module prediction range to calculate the actual DC bus voltage reference at each moment, thereby reducing the dynamic response time by reducing the prediction range and ensuring the control performance of the steady-state DC bus voltage. BRIEF DESCRIPTION OF DRAWINGS

[0028] Fig. 1 is a back-to-back permanent magnet synchronous motor drive system;

[0029] Fig. 2 is a control principle diagram of a traditional PI-MPC;

[0030] Fig. 3 is a control principle diagram of the dynamic prediction control method of the present application. DETAILED DESCRIPTION

[0031] In order to further explain the technical solutions of the present application, the present application is described in detail below through specific embodiments.

[0032] The present application provides a dynamic prediction control method of a permanent magnet synchronous motor driving system, as shown in the figure, the permanent magnet synchronous motor driving system adopts a conventional back-to-back permanent magnet synchronous motor driving system, the driving system includes a permanent magnet synchronous motor (PMSG), a motor side converter (MSC) and a grid side converter (GSC), the motor side converter and the grid side converter are connected in parallel in a conventional manner, and the permanent magnet synchronous motor is connected to the motor side converter (MSC) in a conventional manner. Figs. 1-3

[0033] The dynamic prediction control method provided by the embodiment is based on the above driving system, and the control method includes the following steps:

[0034] S1, obtaining an actual DC bus voltage reference at each moment The expression of the actual DC bus voltage reference is,

[0035] (1),

[0036] In the formula, represents the DC bus voltage reference; represents the number of steps to reach the DC bus reference voltage , which is calculated by the following formula; is the DC bus voltage value at time k, which can be obtained by sampling with a voltage sensor; is the DC bus voltage reference, which is artificially set.

[0037] S2, generating a grid side active power of the grid side converter by using a dynamic reference module The dynamic reference module is:

[0038] (2),

[0039] In the formula, represents the d-axis component of the grid side voltage in the grid side converter, represents the resistance value in the grid side converter, represents the power consumed by the grid side resistance in the grid side converter, represents the motor speed at time k, is the PMSM torque reference, which is artificially set; is the DC bus capacitance value, is the sampling frequency. ​

[0040] wherein, , is the power of the motor (PMSG), is the power of the DC-link capacitor , in other words, represents the total power of the motor (PMSG) and the DC-link capacitor .

[0041] Preferably, to eliminate the steady-state error, the above equation (2) is added with to compensate for the error caused by the switching power loss and conduction power loss, so that the expression of the dynamic reference module, i.e. the grid-side active power is:

[0042] .

[0043] Solving the above equation, the dynamic reference module is:

[0044] .

[0045] S3, the optimal voltage vector is selected by evaluating a predefined cost function, the expression of which is:

[0046] ,

[0047] wherein, represents the estimated grid-side active power at k+2, represents the estimated grid-side reactive power at k+2, represents the grid-side reactive power reference, is the DC bus voltage at k+2, is the reactive power weight coefficient, is the DC bus voltage weight coefficient, is the reactive power reference, which is artificially set; is the bus voltage at k+2; and represent the active power and reactive power constraints, respectively; wherein each instantaneous power reference above is obtained from the filtered DC bus voltage reference.

[0048] Further, the calculation formula of the above is: wherein, is the sampling period, is the bus capacitance value, is the motor-side switching signal at k+1, is the grid-side switching signal at k+1. ​is the motor-side current at k+1 time, is the grid-side current at k+1 time.

[0049] Further, through the above step S3, the and , motor-side cost function and improved grid-side cost function are obtained, and then the same way as the conventional PI-MPC control method is used to output the estimated motor-side switch state at k+1 time to the motor-side converter and the estimated grid-side switch state at k+1 time to the grid-side converter.

[0050] It should be noted that, as Figs. 2-3 indicated, the setting and / or obtaining of other parameters of the present application, in addition to the above-mentioned improvements of the present application, are all prior art, that is, the PI-MPC control method used in the permanent magnet synchronous motor drive system can be obtained, for example, in the above step S4, and obtaining, and again for example is the motor rotor angle, obtained by a position sensor; is the motor rotor speed, obtained by a speed sensor; is the DC bus voltage value at k time, obtained by voltage sensor sampling; is the motor-side switch signal at k time, that is, the calculated optimal switch signal at k+1 time.

[0051] As a preferred mode, in step S1, a dynamic reference prediction range is used to calculate the actual DC bus voltage reference at each time,

[0052] ,

[0053] In the formula, indicates the reference prediction range used in steady state, indicates the adjustment coefficient, indicates a set of positive integers, indicates the DC bus voltage at k time.

[0054] In this way, since the DC bus voltage tracking performance is largely dependent on the above step , and the step is a trade-off between dynamic response time and steady-state DC bus voltage ripple, therefore, in order to improve the dynamic performance of the present application, the dynamic prediction step is used to calculate the actual DC bus voltage reference at each time, so as to reduce the dynamic response time by reducing the prediction step, to ensure the control performance of the steady-state DC bus voltage.

[0055] The application is a dynamic prediction control method of a permanent magnet synchronous motor driving system, which only needs to evaluate a predefined cost function, and the maximum power constraint is included in the cost function, so that the reference voltage value is ensured to be kept in a safe range and to avoid excessive charging or discharging current, effectively prevent significant overshoot of the DC circuit voltage, and ensure that the DC link capacitor is not damaged due to voltage overshoot, and the service life is improved.

[0056] Further, the dynamic prediction control method of the application generates appropriate reference through active power and DC link voltage, that is, adjusts the DC bus voltage by following the specified trajectory of the voltage reference and the power reference, without an additional voltage control loop (i.e., a PI control loop), avoids the performance degradation caused by the integral saturation problem commonly existing in the conventional voltage control loop (i.e., the PI control loop), and ensures the control performance of the driving system, i.e., the unit grid-side power factor, low current harmonics, and stable current bus voltage.

[0057] The above is only the preferred embodiment of the present application, and any equivalent changes and modifications made within the scope of the claims of the present application shall belong to the scope of the claims of the present application.​

Claims

1. A dynamic predictive control method for a permanent magnet synchronous motor drive system, wherein the permanent magnet synchronous motor drive system includes a motor-side converter and a grid-side converter connected to the grid, characterized in that: The control method comprises the following steps: S1. Get the actual DC bus voltage reference at each moment , the actual DC bus voltage reference The expression is: （1）， Where, Indicates the DC bus voltage reference, Indicates reaching The number of steps, Indicates the actual value of the DC bus voltage at time k, obtained by sampling the voltage sensor; S2, to achieve the actual DC bus voltage reference at the next moment , a dynamic reference module is used to generate the grid-side active power reference of the grid-side converter , the grid side active power reference The expression is: （2） ， Where, represents the d-axis component of the grid-side voltage in the grid-side converter, represents the resistance value in the grid-side converter, Indicates the power consumed by the grid-side resistor in the grid-side converter; is the torque reference; is the actual speed, obtained by sampling the speed sensor; is the busbar capacitance parameter, Ts is the sampling period; Represents motor and DC link capacitors Total power; Solve and get the grid-side active power reference The expression is: （3）； S3. Select the optimal voltage vector by evaluating a predefined cost function , the cost function is: ， Where, is the active power at time k+2, is the reactive power weight coefficient, is the reactive power reference, is the reactive power at time k+2, is the bus voltage weight coefficient, is the bus voltage at time k+2; and denote the constraints of active power and reactive power respectively; The calculation formula is: , where is the sampling period, is the busbar capacitance value, is the motor side switch signal at time k+1, is the grid-side switch signal at time k+1; is the motor side current at time k+1, is the grid-side current at time k+1; S4, obtained through step S3 and , represents the motor side cost function and It represents the improved grid-side cost function, and then outputs the estimated motor-side switch state at time k+1 to the machine-side converter and the estimated grid-side switch state at time k+1 to the grid-side converter respectively.

2. The dynamic predictive control method for a permanent magnet synchronous motor drive system according to claim 1, characterized in that: In step S2, the grid-side active power reference Add , the grid side active power reference The expression is: 。 3. The dynamic predictive control method for a permanent magnet synchronous motor drive system according to claim 1, characterized in that: In step S1, the dynamic reference module prediction range is used to calculate the actual DC bus voltage reference at each moment. , the expression is, ， Where, represents the reference prediction range used in steady state, represents the adjustment coefficient, represents the set of positive integers, represents the DC bus voltage at time k.

Images