Switched reluctance wind turbine sub-region model predictive control method

Abstract

The application discloses a kind of switch reluctance wind turbine subarea model prediction control methods.The method is according to switch reluctance generator rotor position and is divided into excitation current building area and torque tracking area, before stator, rotor alignment position, using current tracking model prediction control, after stator, rotor alignment position, using torque tracking model prediction control.Reference current is calculated in excitation current building area according to given torque and speed online, and torque feedforward compensation mechanism is introduced, and positive torque generated in excitation current building process is compensated.The application can realize reference current adaptive adjustment with speed and torque demand, improve interphase torque handover continuity, reduce torque ripple, and improve switch reluctance wind turbine wide-speed domain operation performance.

Description

Technical Field

[0001] This invention belongs to the field of wind power generation system control technology, specifically relating to a sectional model predictive control method for switched reluctance wind turbine generators. Background Technology

[0002] With the development of wind power technology, wind turbine generators need to maintain high operating efficiency and good dynamic performance over a wide speed range and under varying wind speed conditions. Compared with traditional doubly-fed induction generators and permanent magnet synchronous generators, switched reluctance generators (SRGs) have advantages such as simple structure, no rotor windings and permanent magnets, high reliability, and especially the ability to maintain high efficiency over a wide speed range and start at low wind speeds, making them promising for applications in the wind power field. However, the non-sinusoidal phase current and high saturation of the core magnetic circuit are their main operating characteristics, and their inherent torque pulsation problem can affect the stability of system operation.

[0003] Existing SRG control methods mainly include angle position control, current chopping control, voltage PWM control, direct torque control, and model predictive control. Among them, model predictive control can predict the future current or torque under different switching states based on the motor model and select the optimal switching state through a cost function, thus it has been gradually applied to the control of AC motors, including switched reluctance motors. However, existing SRG model predictive control methods usually use a uniform control objective throughout the entire conduction range, failing to fully consider the differences in electromagnetic characteristics before and after stator-rotor alignment. In addition, the reference current in the excitation current-building region of existing methods often relies on empirically given or fixed parameter settings, failing to simultaneously address the current build-up and torque output requirements under different operating conditions, thereby affecting the operating performance of switched reluctance wind turbine generators over a wide speed range. Furthermore, during the excitation current-building process before stator-rotor alignment, the establishment of phase winding current may generate a positive torque opposite to the direction of the generated torque. Existing control methods typically lack effective compensation for this positive torque, resulting in significant torque fluctuations in switched reluctance wind turbine generators during variable speed operation and commutation. Summary of the Invention

[0004] This invention addresses the limitations of switched reluctance generator control technology in the prior art by proposing a regional model predictive control method for switched reluctance wind turbine generators. This method improves the continuity of phase-to-phase torque transfer, reduces torque pulsation, and enhances the operating performance of switched reluctance wind turbine generators under wide speed range and variable wind speed conditions.

[0005] This invention includes the following steps:

[0006] The system collects the rotor position, speed, phase current, and DC bus voltage of the switched reluctance generator. Based on the rotor position, the control process is divided into an excitation current-building region and a torque tracking region. In the excitation current-building region, the reference current is calculated online based on the given torque and speed, and the control tracking reference current is predicted based on the current model. In the torque tracking region, the control tracking reference torque is predicted based on the torque model. The torque feedforward compensation is determined based on the positive torque generated during the excitation current-building process, and the torque setpoint in the torque tracking region is corrected.

[0007] The excitation current-building region is the area between the turn-on angle and the alignment position of the stator teeth and rotor teeth, and the torque tracking region is the area between the alignment position of the stator teeth and rotor teeth and the turn-off angle.

[0008] The excitation current-building region model predictive control mainly includes: online calculation of the reference current in the excitation current-building region; and current tracking using current model predictive control.

[0009] The method for calculating the reference current in the excitation current-building region is as follows:

[0010] Calculate the reference current for the excitation build-up region based on the torque requirement of the wind turbine's maximum power point tracking (MPPT). Align the position. arrive ( The change in current at the position (where the current overlaps between two adjacent phases by an angle) can be approximated as a straight line within the interval [missing information]. Choose any angle According to the torque distribution function The torque at position is:

[0011] (1)

[0012] in, for Torque at that point for The torque distribution function value at that location. The target torque is given based on the MPPT requirements.

[0013] The current at the location is:

[0014] (2)

[0015] but

[0016] (3)

[0017] In the formula, To align the position current, for Position current.

[0018] From (3), we can obtain

[0019] (4)

[0020] When the switched reluctance wind turbine is operating at variable speed, at the cut-in wind speed Based on this, when the wind speed increases, the reference current in the build-up area should be appropriately limited:

[0021] (5)

[0022] In the formula, To determine the optimal rotor angular velocity of the SRG at the cut-off wind speed, The optimal rotor angular velocity of the SRG is given under different wind speeds.

[0023] The objective function for predictive control of the excitation flow-building region model is:

[0024] (6)

[0025] In the formula, Predict the current for each phase. This is the reference current for each phase.

[0026] Candidate voltage set The optimal control input can be expressed as:

[0027] (7)

[0028] The optimal switching state of the power converter can be obtained by selecting the voltage of each phase, and then applied to the power converter to achieve dynamic tracking of the reference value of the current in the current-building region.

[0029] The objective function for predictive control in the torque tracking region model is:

[0030] (8)

[0031] In the formula, For each phase, predict the torque. This is the reference torque for each phase.

[0032] The torque feedforward compensation amount is obtained by looking up a table based on the predicted current and location of the current-building zone:

[0033] (9)

[0034] In the formula, For each phase, predict the torque. It is a two-dimensional static torque table.

[0035] The method for correcting the torque setpoint in the torque tracking region by the torque feedforward compensation is as follows:

[0036] (10)

[0037] This is the reference torque after current phase compensation. Predicting torque for the next phase flow region. The reference torque before current phase compensation. This indicates the rotor position for the next phase.

[0038] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: This invention proposes a zoned torque model predictive control method suitable for switched reluctance wind turbine generators. Different control objectives are constructed in the excitation current-building region and the torque tracking region. Addressing the problem that the reference current in the excitation current-building region relies on empirical settings and is difficult to adapt to speed changes, an online design method for the reference current is proposed, enabling the current reference to adaptively adjust according to speed and target torque. Simultaneously, to address the problem that the positive torque generated in the current-building region causes abrupt changes in total torque, a torque feedforward compensation mechanism is introduced to compensate for the positive torque in the current-building region, thereby improving the smoothness of the total electromagnetic torque and reducing torque ripple. Attached Figure Description

[0039] Figure 1 This is a block diagram of the predictive control of the sectional model of the switched reluctance wind turbine generator group in this invention;

[0040] Figure 2 for and Relationship curve diagram;

[0041] Figure 3 This is a comparison diagram of electromagnetic torque under different control methods when the wind speed is 4 m / s, according to an embodiment of the present invention.

[0042] Figure 4 This is a graph showing the changes in wind energy utilization coefficient and rotational speed under a step change in wind speed according to an embodiment of the present invention;

[0043] Figure 5 This is a phase current curve under a step change in wind speed according to an embodiment of the present invention. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0045] Example 1

[0046] In one embodiment, a regional model predictive control method for switched reluctance wind turbine generators is implemented, comprising the following steps:

[0047] The controller collects the rotor position, rotor angular velocity, phase current, and DC bus voltage of the SRG in each sampling period;

[0048] Control area division. Let... For opening angle, For stator and rotor alignment position, For the shut-off angle, The rotor position angle is calculated over one rotor cycle. At that time, the current phase is in the excitation current-building region; when At that time, the current phase is in the torque tracking region.

[0049] The reference current for the excitation build-up region is generated online by the controller based on the target torque, speed, torque distribution function, and electromagnetic torque characteristics of the switched reluctance generator provided by the MPPT requirements.

[0050] Alignment position arrive ( The change in current at the position (where the current overlaps between two adjacent phases by an angle) can be approximated as a straight line within the interval [missing information]. Choose any angle According to the torque distribution function The torque at position is:

[0051] (1)

[0052] in, for Torque at that point for The torque distribution function value at that location. The target torque is given based on the MPPT requirements.

[0053] The current at the location is:

[0054] (2)

[0055] but

[0056] (3)

[0057] In the formula, To align the position current, for Position current.

[0058] From (3), we can obtain

[0059] (4)

[0060] When the switched reluctance wind turbine is operating at variable speed, at the cut-in wind speed Based on this, when the wind speed increases, the reference current in the build-up area should be appropriately limited:

[0061] (5)

[0062] In the formula, To determine the optimal rotor angular velocity of the SRG at the cut-off wind speed, The optimal rotor angular velocity of the SRG is given under different wind speeds.

[0063] Current model predictive control in the excitation current-building region. In the excitation current-building region, the controller predicts the phase current in the next sampling cycle based on the current prediction model for different candidate switching states.

[0064] (6)

[0065] In the formula, The sampling period is The phase current at the current sampling time. To predict the phase current at the next sampling time, The phase voltage corresponding to the candidate switch state. For phase resistance, The rotor angular velocity at the previous sampling time. It is a magnetic flux.

[0066] Construct the objective function:

[0067] (7)

[0068] In the formula, Predict the current for each phase. This is the reference current for each phase.

[0069] Candidate voltage set The optimal control input can be expressed as:

[0070] (8)

[0071] The optimal switching state of the power converter can be obtained by selecting the voltage of each phase, and then applied to the power converter to achieve dynamic tracking of the reference value of the current in the current-building region.

[0072] Torque tracking region model predictive control. Construct the objective function:

[0073] (9)

[0074] In the formula, For each phase, predict the torque. This is the reference torque for each phase.

[0075] Candidate voltage set The optimal control input can be expressed as:

[0076] (10)

[0077] The optimal switching state of the power converter can be obtained by selecting the voltage of each phase. This state is then applied to the power converter to achieve online optimal control of the torque tracking region.

[0078] Torque feedforward compensation. To reduce the impact of the positive torque in the current-building region on the total electromagnetic torque, this embodiment introduces a torque feedforward compensation mechanism. The torque feedforward compensation amount is obtained by looking up a table based on the predicted current and location of the current-building region:

[0079] (11)

[0080] In the formula, For each phase, predict the torque. It is a two-dimensional static torque table, and the instantaneous torque for any position and current can be obtained by two-dimensional interpolation.

[0081] Feedforward compensation at rotor position This process is performed in real time, primarily involving the superposition of the instantaneous positive torque generated by the subsequent excitation phase onto the reference torque of the generator phase. The method for correcting the torque setpoint in the torque tracking region is as follows:

[0082] (12)

[0083] This is the reference torque after current phase compensation. Predicting torque for the next phase flow region. This is the reference torque before compensation for the current phase.

[0084] Example 2

[0085] Another embodiment of the present invention is a switched reluctance wind power generation system, wherein the partition model predictive control method of the switched reluctance wind turbine generator set adopts the method described in Embodiment 1.

[0086] A simulation model of a three-phase 6 / 4 switched reluctance wind turbine generator system was built on the MATLAB / Simulink platform. The entire simulation implementation case includes the wind turbine, speed-increasing gearbox, SRG, auxiliary DC power supply, asymmetrical half-bridge power converter, bus capacitor, resistive load, etc.

[0087] The wind turbine drives the SRG after being accelerated by a gearbox. The gearbox transmission ratio is j=6, and the cut-in wind speed is v. cut-in=4m / s. When the system starts working, the SRG works in electric mode and drives the generator set to accelerate together with the wind turbine. After the generator set reaches the speed at the cut-in wind speed, it switches to power generation operation and enters the MPPT control process.

[0088] In this embodiment of the invention, the wind turbine rotor radius R = 1.6m, and the wind energy utilization coefficient is... for:

[0089] (1)

[0090] In the formula, For the tip speed ratio, and Relationship curves as follows Figure 2 As shown.

[0091] The SRG power converter adopts an asymmetric half-bridge topology, self-excited mode, and initial auxiliary DC power supply U. s =120V, bus capacitance C=4500 The moment of inertia of the mechanical system is J = 0.0017. The coefficient of viscous friction, D, is 0.000813. The SRG uses the position where the stator salient pole and the rotor groove center coincide as the reference zero point. During the starting phase, the phase winding opening angle is... Set as Cut-off angle Set as After startup, switch to generator operation, and adjust the phase winding opening angle accordingly. Set as Cut-off angle Set as Stator and rotor alignment positions Overlap angle in torque distribution function Simulation results are presented for conventional model predictive control, partitioned model predictive control, and partitioned model predictive control with torque feedforward compensation.

[0092] The control effect and analysis of the regional model predictive control method for switched reluctance wind turbine generators are as follows:

[0093] Figure 3 The figure shows a comparison of electromagnetic torque under different control methods when the wind speed is 4 m / s according to an embodiment of the present invention. Simulation results show that the partitioned model predictive control method using torque feedforward compensation can significantly reduce torque ripple. Figure 4 This is a graph showing the wind energy utilization coefficient and rotational speed variation under a step change in wind speed according to an embodiment of the present invention. Simulation results show that the SRG can track the optimal angular velocity command under both step increases and decreases in wind speed, and the wind energy utilization coefficient is maintained at C. pmax=0.438, achieving MPPT control of wind turbine generator sets with good accuracy. Figure 5 The figure shows the phase current curve under a step change in wind speed according to an embodiment of the present invention. Simulation results show that when the wind speed increases from 5 m / s to 6 m / s, the calculated reference current before alignment decreases from 3.92 A to 3.27 A. The reference current in the excitation current-building zone can be adaptively adjusted according to wind speed and torque.

[0094] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A predictive control method for a sectional model of a switched reluctance wind turbine generator, characterized in that, Includes the following steps: Collect rotor position, speed, phase current, and DC bus voltage of the switched reluctance generator; Based on the rotor position, the control region of the current phase winding is determined, and the control process of the current phase winding is divided into the excitation current building region and the torque tracking region. When the current phase winding is in the excitation current building region, the reference current of the excitation current building region is calculated online according to the given torque requirement and the speed, and the optimal switching state that makes the predicted phase current track the reference current of the excitation current building region is selected based on the current model predictive control. When the current phase winding is in the torque tracking region, select the optimal switching state that makes the predicted electromagnetic torque track the torque setpoint. The torque feedforward compensation amount is determined based on the positive torque generated during the excitation current building process, and the torque feedforward compensation amount is used to correct the torque setpoint in the torque tracking zone.

2. The method for predictive control of a switched reluctance wind turbine generator group based on a regional model according to claim 1, characterized in that, The excitation current-building region is the open-circuit angle. To the position where the stator teeth and rotor teeth are aligned The torque tracking region is the area between the stator teeth and rotor teeth alignment position and the shut-off angle.

3. The method for predictive control of a switched reluctance wind turbine generator group based on a regional model according to claim 2, characterized in that, The excitation current-building region model predictive control mainly includes: Online calculation of reference current in the excitation build-up region; Current tracking is achieved using current model predictive control.

4. The method for predictive control of a switched reluctance wind turbine generator group based on a regional model according to claim 3, characterized in that, The method for calculating the reference current in the excitation current-building region is as follows: Calculate the reference current for the excitation build-up region based on the torque requirement of the wind turbine's maximum power point tracking (MPPT). Align the position. arrive ( The change in current at the position (where the current overlaps between two adjacent phases by an angle) can be approximated as a straight line within the interval [missing information]. Choose any angle According to the torque distribution function The torque at position is: (1) in, for Torque at that point for The torque distribution function value at that location. The target torque is given based on the MPPT requirements. Current at location for: (2) but (3) In the formula, To align the position current, for Position current. From (3), we can obtain (4) When the switched reluctance wind turbine is operating at variable speed, at the cut-in wind speed Based on this, when the wind speed increases, the reference current in the flow-forming zone should be appropriately limited. (5) In the formula, To determine the optimal rotor angular velocity of the SRG at the cut-off wind speed, The optimal rotor angular velocity of the SRG is given under different wind speeds.

5. The method for predictive control of a switched reluctance wind turbine generator group based on a regional model according to claim 4, characterized in that, Construct the objective function: (6) In the formula, Predict the current for each phase. This is the reference current for each phase. Candidate voltage set Optimal control input It can be represented as: (7) The optimal switching state of the power converter can be obtained by selecting the voltage of each phase, and then applied to the power converter to achieve dynamic tracking of the reference value of the current in the current-building region.

6. The method for predictive control of a switched reluctance wind turbine generator group based on a regional model according to claim 1, characterized in that, The torque tracking model predictive control objective function in the torque tracking region is: (8) In the formula, For each phase, predict the torque. This is the reference torque for each phase.

7. The method for predictive control of a switched reluctance wind turbine generator group based on a regional model according to claim 1, characterized in that, The torque feedforward compensation amount is obtained by looking up a table based on the predicted current and location of the current-building zone: (9) In the formula, For each phase, predict the torque. It is a two-dimensional static torque table, and the instantaneous torque for any position and current can be obtained by two-dimensional interpolation. Feedforward compensation at rotor position The compensation is performed in real time, mainly by superimposing the reference torque of the generator phase onto the instantaneous positive torque generated by the subsequent excitation phase. The compensation strategy is as follows: (10) This is the reference torque after current phase compensation. Predicting torque for the next phase flow region. The reference torque before current phase compensation. This indicates the rotor position for the next phase.

Images