A bidirectional, disturbance-free switching method and system for doubly-fed generator unit grid connection and grid construction control

Abstract

This invention discloses a bidirectional, disturbance-free switching method and system for doubly-fed induction generator (DFIG) grid-connection control, relating to the field of power grid control technology. The method includes: 1) determining the dq-axis orientation mode and providing an orientation flag; 2) switching the generator-side grid-connection control model between dq-axis orientation modes using the orientation flag; and selecting the root grid or the network-connection mode using a switching command; 3) setting enable triggering, pre-synchronization, switch switching, and PI integral initial value reset stages in the control loops of grid-connection control and network-connection control. Upon receiving a switching command, the control loop is enabled as needed, the network-connection pre-synchronization loop is pre-synchronized, and the discrimination switch is switched, triggering the reset of all PI integral initial values ​​in the control loop and connecting to the switched control loop. This invention solves the problems of existing technologies, such as the lack of consideration for orientation consistency leading to drastic switching fluctuations, excessive integral calculations, and poor pre-synchronization accuracy resulting in significant impacts during switching.

Description

Technical Field

[0001] This invention relates to the field of power grid technology, specifically to a bidirectional, disturbance-free switching method and system for doubly fed generator units to connect with and connect to the grid. Background Technology

[0002] With the vigorous development of new energy sources such as wind and solar power, the proportion of new energy power electronic converters in the power system is constantly increasing. The power grid has transformed from a system dominated by synchronous machines to a "dual-high" power system with a high proportion of new energy sources and high power electronic integration. This has led to a gradual decrease in the system's short-circuit ratio and a significant increase in the characteristics of a weak power grid. To ensure stable system operation, there is an urgent need for grid-type control modifications to wind power grid-connected systems.

[0003] Doubly fed induction generators (DFIGs) are crucial power generation and grid connection sources in power systems. Their operating modes must be dynamically adjusted according to grid characteristics: in strong grid environments, grid-following control is used to achieve power generation and grid connection; however, when the grid exhibits weak grid characteristics, it is necessary to switch from grid-following control to grid connection control to provide the system with the necessary frequency and voltage support. Therefore, the bidirectional online seamless switching between grid-following control and grid connection control becomes a core element in ensuring stable system operation.

[0004] Currently, the switching between grid-following (GFL) and grid-forming (GFM) modes of doubly-fed induction generator (DFIG) wind turbines mostly adopts a switching discrimination method, such as the technical solution disclosed by Li Zhenbo in "Smooth Switching Control Strategy for Dual-Mode Operation of Voltage Source and Current Source of Doubly-Followed Wind Turbines" published in the 22nd issue of "Electrical Drive" in 2022. However, this switching method has significant drawbacks: First, it requires the simultaneous operation of two control strategies, namely grid-following and grid-forming, which consumes a large amount of controller computing resources due to excessive integral calculations, and may even require additional dedicated controllers to meet the computing requirements; Second, when receiving the switching command, the pre-synchronization loop and the control loop start the discrimination access simultaneously, resulting in a short pre-synchronization time, which cannot accurately achieve pre-synchronization and is prone to causing large shocks during the switching process; Third, it does not fully consider the orientation consistency problem between grid-forming control and grid-following control. If there is a difference in the orientation methods of the two, it may cause violent switching fluctuations, and in severe cases, it may even lead to direct system instability. Summary of the Invention

[0005] To address the shortcomings and defects of the existing technologies, this invention proposes a bidirectional, disturbance-free switching method and system for doubly-fed generator (DFIG) grid-connection control. Considering the directional consistency between grid-connection and grid-connection control, this invention achieves disturbance-free switching by setting enable triggering, pre-synchronization, switching, and PI integral initial value reset steps in the key loops of grid-connection and grid-connection. This effectively avoids simultaneous operation of the grid-connection and grid-connection loops, reduces the waste of integrator computing resources on the controller, and allows for grid-connection modification without significant changes to the original grid-connection strategy. This solves the problems of existing technologies, such as the lack of consideration for directional consistency leading to drastic switching fluctuations, excessive integral calculations, and poor pre-synchronization accuracy causing significant impacts during switching.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for bidirectional, disturbance-free switching of a doubly-fed generator unit to the grid and its network control includes the following steps:

[0008] S1: Determine the dq axis orientation mode of the doubly fed generator unit during current operation and provide the orientation flag;

[0009] S2: Construct two sets of machine-side grid construction control models under different dq-axis orientations. By adding a discrimination switch to the outer loop of the grid voltage and combining it with the orientation flag, the machine-side grid construction control model can be switched between dq-axis orientation modes. By adding discrimination switches to the active power loop and the inner current loop of the grid and combining them with the switching command, the root grid or the grid can be selected.

[0010] S3: In the control loop of network control and network construction control, enable triggering, pre-synchronization, switch switching and PI integral initial value reset are set. When a switching command is received, the control loop is enabled as needed, the network construction pre-synchronization loop is pre-synchronized and the discrimination switch is switched, triggering the reset of all PI integral initial values ​​in the control loop and connecting to the control loop after the switch.

[0011] S4: Exit enable as needed and complete the switch.

[0012] In S2, the constructed machine-side grid control model includes a grid active power loop control model, a grid reactive power loop control model, a grid voltage outer loop control model, and a current inner loop control model. The grid voltage outer loop control model is constructed by combining the output results of the grid active power loop control model and the grid reactive power loop control model. The current inner loop control model is also constructed by combining the output results of the grid active power loop control model and the grid voltage outer loop control model. The output results of the current inner loop control model are used to drive the rotor-side converter switching action after SPWM or SVPWM modulation to achieve closed-loop control.

[0013] In S2, the active power loop is configured to use virtual inertia control, droop control, and pre-synchronization control of the grid formation angle. The constructed grid formation active power loop control model is as follows:

[0014] (1)

[0015] In equation (1), the input is the active power setpoint. P ref Active output value P meas and active droop power P droop ,in, P ref Issued by human intervention or scheduling. P meas The power was calculated using three-phase instantaneous power. P droop Through the sagging coefficient in sagging control K droop_p It is calculated by multiplying the deviation of the network angular frequency ω and the rated angular frequency ω0; J The output is the virtual inertia coefficient; the netting angle is the output. θ GFM It is obtained by integrating the angular frequency ω of the mesh; K ps This is the pre-synchronization ratio coefficient. θ PLL Stator / grid voltage locking angle, This indicates that the high-level signal in this section is enabled.

[0016] In S2, the constructed network reactive power loop control model is as follows:

[0017] (2)

[0018] In equation (2), the inputs are the reactive power setpoints. Q ref reactive power output value Q meas and voltage rating U 0, the output is the voltage reference value of the outer loop of the grid voltage. E ref , K droop_q The droop coefficient is... K p_q and K i_q These are the proportional and integral parameters of the PI controller in the reactive power loop of the network.

[0019] In S2, the constructed grid voltage outer loop control model is as follows:

[0020] (3)

[0021] In equation (3), the inputs are the voltage setpoints of the stator d-axis. U sd_ref and feedback value U sd_meas and the voltage setpoint of the stator q-axis. U sq_ref and feedback value U sq_meas When using d-axis orientation, take: U sd_ref =E ref , U sq_ref = 0; When using q-axis orientation, take: U sd_ref = 0, U sq_ref =-E ref The output is the current setpoint for the rotor's d-axis. I rd_ref and the current setpoint of the rotor q-axis I rq_ref ; K p_ud , K i_ud and K p_uq , K i_uq These are the proportional and integral parameters of the PI controller in the outer loop of the stator dq shaft network voltage; feedback value. U sd_meas , U sq_meas The grid-connection angle is based on the output of the grid-connected active power loop control model using stator voltage. θ GFM The coordinates are transformed to obtain the result.

[0022] In S2, the constructed current inner loop control model is as follows:

[0023] (4)

[0024] In equation (4), the input is the current setpoint of the rotor d-axis. I rd_ref and feedback value I rd_meas and the current setpoint for the rotor q-axis. I rq_ref and feedback value I rq_meas The output is the voltage setpoint for the rotor's d-axis. Urd_ref and the voltage setpoint of the rotor q-axis U rq_ref ; K p_Id , K i_Id and K p_Iq , K i_Iq These are the proportional and integral parameters of the PI controller in the d-axis current inner loop and the q-axis current inner loop, respectively; feedback value. I rd_meas , I rq_meas Both use the rotor current and the grid-connected active power loop control model output to determine the grid-connected angle. θ GFM The results were obtained through calculation.

[0025] In S1, the dq axis orientation mode is determined based on the per-unit voltage values ​​of the doubly fed generator stator dq axis. The orientation flag of the dq axis orientation mode is set as DR. If Ud=1 and Uq=0, the doubly fed generator is currently d-axis oriented and is marked as DR=1; if Ud=0 and Uq=-1, the doubly fed generator is currently q-axis oriented and is marked as DR=0.

[0026] In S2, the switching between dq-axis orientation modes of the machine-side network control model is achieved by adding a discrimination switch to the outer loop of the network voltage and combining it with the orientation flag.

[0027] Two dual-input, single-output discrimination switches, denoted as KG1 and KG2, are set at the input of the outer loop of the grid voltage. The input of discrimination switch KG1 is the voltage reference value output by the grid reactive power loop control model. E ref And 0, the input of the discrimination switch KG2 is the voltage reference value output by the grid reactive power loop control model. E ref The negative values ​​and 0;

[0028] The directional flag bit DR is used as the discrimination signal, and KG1=KG2=DR. The input switching of the grid reactive power loop control model is realized according to the DR value.

[0029] In S2, the selection of the root network or the network structure by adding a discrimination switch to the active power loop and the inner current loop of the network structure, combined with the switching command, refers to:

[0030] A dual-input single-output discrimination switch, denoted as KM1, is set at the output end of the grid active power loop. Two dual-input single-output discrimination switches, denoted as KM2 and KM3, are set at the dq-axis current setpoint of the inner current loop. The inputs of discrimination switch KM1 are the grid phase-locked loop angle and the output angle of the grid active power loop, respectively. The inputs of discrimination switch KM2 are the output of the PI controller of the grid-connected active power loop and the rotor d-axis current setpoint output of the grid voltage outer loop control model, respectively. I rd_ref The inputs to the discrimination switch KM3 are the output of the grid-controlled reactive power loop PI controller and the rotor q-axis current setpoint output of the grid voltage outer loop control model. I rd_ref ;

[0031] The discrimination flag KM=0 indicates the grid-following mode, and the discrimination flag KM=1 indicates the grid-building mode. KM1=KM2=KM3=KM. When a switching command is received, the discrimination flag KM is given. The input selection of the current inner loop control model is switched according to the KM value to realize the bidirectional switching between grid-following and grid-building modes.

[0032] In S3, setting enable triggers in the control loops of grid-connected control and grid-building control, and enabling the control loops on demand, refers to setting enable switches K in the grid-connected power loop respectively. enable,1 1. Set an enable switch K in the active power loop of the network. enable,2 The same enable switch K is set in the reactive power loop and the outer voltage loop of the grid. enable,3 1. Set the enable switch K in the phase-locked loop. enable,4 And set enable switch K in the pre-synchronization loop of the network. enable,5 Upon receiving a switching command, the enable switch is enabled as needed according to the switching command.

[0033] In S3, PI integral initial value reset refers to resetting the initial PI integral values ​​of the PI controllers in the grid-connected power loop, grid-connected reactive power loop, and grid-connected voltage outer loop respectively after receiving a trigger signal at any rising or falling edge; whereby the reset values ​​of the PI controller in the grid-connected power loop are the current feedback values ​​of the rotor d-axis. I rd_meas Current feedback value of rotor q-axis I rq_meas When the reactive power loop of the grid uses direct PI control or droop control, the reset value of the PI controller is 0. The reset values ​​of the PI controller in the outer voltage loop of the grid are the current feedback values ​​of the rotor d-axis. I rd_meas Current feedback value of rotor q-axis I rq_meas Furthermore, the rotor dq axis current reset values ​​are all based on real-time acquisition values ​​plus a delay of one sampling period, denoted as 1 / Z.

[0034] When a switchover command to switch from the root network to the network structure is received, the specific switchover process is as follows:

[0035] S1.1: Determine the orientation mode of the dq axis based on the collected per-unit voltage values ​​of the doubly fed motor stator dq axis, and give the orientation flag bit DR value;

[0036] S1.2: Enable the active power loop in the network to enable K enable,2= 1. Enable network pre-synchronization ring to enable K enable,5 =1, after pre-synchronization is complete, make K =1, enable,5 =0 Exit pre-synchronization;

[0037] S1.3: Enables the grid reactive power loop and the grid voltage outer loop to enable K enable,3 =1, control the discrimination switches KG1 and KG2 according to the orientation flag bit DR value;

[0038] S1.4: According to the switching instruction, KM=1, triggering the reset of the initial values ​​of all PI integrals in the control loop and connecting them to the network control loop;

[0039] S1.5: Make K enable,1 =0 and K enable,4 =0, which enables the grid power loop and the grid phase-locked loop to be deactivated, completing the switching;

[0040] Among them, in K enable,5 During the period =1, when the difference between the mesh angle and the follow-mesh angle is less than 0.05 degrees, K is executed. enable,5 After =0, switch to S1.3. When the difference between the netting angle and the following netting angle is greater than 0.05 degrees, continue to run pre-synchronization until the difference between the netting angle and the following netting angle is less than 0.05 degrees.

[0041] Then, execute S1.4 and S1.5 in sequence and determine whether the system power fluctuates. If the fluctuation value is less than 0.05 degrees and the power output is stable within 10 fundamental frequency cycles, the switching is completed. If the power fluctuation is too large and the fluctuation is still greater than 0.1 pu within 20 fundamental frequency cycles or if a limit is triggered, the switching fails and the unit shuts down for protection.

[0042] When a switch command to switch from the network structure to the root network is received, the specific switchover process is as follows:

[0043] S2.1: Enable the K-loop lock-in. enable,4 =1, lock the stator voltage to obtain the grid angle;

[0044] S2.2: Calculate the deviation between the output frequency of the phase-locked loop and the rated frequency. When the deviation is less than 0.03Hz, proceed to S.3.

[0045] S2.3: Enable network pre-synchronization loop to enable K enable,5=1, after pre-synchronization is complete, make K =1, enable,5 =0 Exit pre-synchronization;

[0046] S2.4: Enable grid power loop to enable K enable,1 =1, control the discrimination switches KG1 and KG2 according to the orientation flag bit DR value;

[0047] S2.5: According to the switching command, KM=0, triggering the reset of the initial values ​​of all PI integrals in the control loop and connecting to the network control loop;

[0048] S2.6: Make K enable,2= 0 and K enable,3= 0, which enables and deactivates the active power loop, reactive power loop, and outer voltage loop of the grid, completing the switching;

[0049] Among them, in K enable,5 During period =1, when the difference between the meshing angle and the follow-me-off angle is less than 0.05 degrees, K is executed. enable,5 After =0, switch to S2.4. When the difference between the netting angle and the following netting angle is greater than 0.05 degrees, continue to run pre-synchronization until the difference between the netting angle and the following netting angle is less than 0.05 degrees.

[0050] Then, execute S2.5 and S2.6 in sequence and determine whether the system power fluctuates. If the fluctuation value is less than 0.05 degrees and the power output is stable within 10 fundamental frequency cycles, the switching is completed. If the power fluctuation is too large and the fluctuation is still greater than 0.1 pu within 20 fundamental frequency cycles or if a limit is triggered, the switching fails and the unit shuts down for protection.

[0051] A doubly fed generator unit grid-connection control bidirectional disturbance-free switching system includes:

[0052] The orientation flag confirmation module is used to determine the dq axis orientation mode of the doubly fed generator unit during current operation and to provide the orientation flag bit;

[0053] The model building module is used to build two sets of machine-side network control models with different dq-axis orientations;

[0054] The switching selection module is used to switch the machine-side grid control model between dq-axis orientation modes by adding a discrimination switch to the outer loop of the grid voltage and combining it with the orientation flag; it is also used to select the root grid or the grid by adding discrimination switches to the grid active power loop and the inner loop of the grid and combining them with the switching command.

[0055] The processing module is used to set the enable trigger, pre-synchronization, switch switching and PI integral initial value reset links in the control loop of network-following control and network-building control. After receiving the switching command, it enables the control loop as needed, pre-synchronizes the network-building pre-synchronization loop and switches the discrimination switch, triggers the reset of the initial PI integral values ​​in the control loop and connects to the control loop after the switch.

[0056] The exit module is used to exit the enable function as needed and complete the switch.

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0058] 1. Based on the consideration of the orientation consistency between network construction control and network following control, this invention achieves seamless switching between network following and network construction by setting enable triggering, pre-synchronization, switch switching and PI integral initial value reset in the key loops of network following and network construction. This can effectively avoid the simultaneous operation of network following and network construction loops, reduce the waste of controller computing resources by the integrator, and network construction can be modified without making major changes to the original network following strategy.

[0059] It is important to note that conventional methods typically involve simultaneous pre-synchronization and switching. However, this approach can lead to significant switching fluctuations due to large angle differences caused by incomplete pre-synchronization. Other methods involve setting a fixed pre-synchronization period, exiting regardless of completion at the set time, which also results in large angle differences. This invention, however, utilizes specifically designed enable triggering, pre-synchronization, switching, and PI integral initial value reset mechanisms to switch based on pre-synchronization deviations. Pre-synchronization is considered complete only when the minimum set deviation is met, allowing subsequent actions to proceed. Therefore, this invention offers advantages such as consistent orientation before and after switching, low computational resource consumption, high pre-synchronization accuracy, minimal impact during switching, and prevention of severe switching fluctuations or instability.

[0060] 2. This invention also designs a doubly fed generator switching scheme based on the dq axis orientation, and the two schemes can be selected online. It has better adaptability to the already mature and operating grid-connected doubly fed generators, and the grid transformation can be carried out without making major changes to the original grid system.

[0061] 3. The present invention sets up enable triggering, pre-synchronization, switch switching and PI integral initial value reset logic control in the key loops of network following and network construction, which can effectively avoid the simultaneous operation of network following and network construction loops and reduce the waste of controller computing resources by the integrator.

[0062] 4. This invention designs a timing logic for bidirectional switching between grid connection and network construction. Switching threshold judgment and shutdown protection are set between switching steps to achieve bidirectional uninterrupted switching between grid connection and network construction, avoiding large power fluctuations or instability during switching. Attached Figure Description

[0063] Figure 1 This is a flowchart of the switching process of the present invention;

[0064] Figure 2 A flowchart illustrating the switching process from grid connection to grid construction for doubly-fed generator units;

[0065] Figure 3 A flowchart illustrating the switching process from the doubly fed generator unit's main grid to its feeder grid;

[0066] Figure 4 This is a system block diagram of the present invention;

[0067] Figure 5 This is a diagram showing the test results of the network switching to the network structure according to the present invention;

[0068] Figure 6 This is a test result diagram of the switching of the mesh structure to the root mesh in this invention;

[0069] Figure 7 The test results for switching from mesh to mesh without considering the dq axis orientation are shown in the figure.

[0070] Figure 8 The test results for switching from mesh construction to root mesh without considering the dq axis orientation method are shown in the figure.

[0071] Figure 9 The test results diagram shows the results of switching from the existing network to the new network without considering whether pre-synchronization was completed. Detailed Implementation

[0072] The present invention will be further described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0073] Example 1

[0074] This embodiment provides a bidirectional, disturbance-free switching method for doubly-fed induction generator (DFIG) grid-connection control. This method is applicable to various DFIG converter applications, including DFIG wind turbines, DFIG flywheel energy storage units, and DFIG pumped storage units. Figure 1 As shown, it includes the following steps:

[0075] S1: Determine the dq axis orientation mode of the doubly fed generator unit during current operation and provide the orientation flag.

[0076] Specifically, this step determines the dq axis orientation mode based on the per-unit voltage values ​​of the doubly-fed generator stator dq axis. The orientation flag for the dq axis orientation mode is set to DR. If Ud=1 and Uq=0, the doubly-fed generator is currently d-axis oriented and marked as DR=1; if Ud=0 and Uq=-1, the doubly-fed generator is currently q-axis oriented and marked as DR=0.

[0077] It should be noted that the values ​​of Ud and Uq are not actually exact integers; they deviate from integers to a certain extent. This solution obtains them by rounding to the nearest integer.

[0078] S2: Construct two sets of machine-side grid construction control models under different dq-axis orientations. By adding a discrimination switch to the outer loop of the grid voltage and combining it with the orientation flag, the machine-side grid construction control model can be switched between dq-axis orientation modes. By adding discrimination switches to the active power loop and the inner current loop of the grid construction and combining them with the switching command, the root grid or the grid can be selected.

[0079] In this step, the constructed grid-side control model includes a grid active power loop control model, a grid reactive power loop control model, a grid voltage outer loop control model, and a current inner loop control model. The grid voltage outer loop control model is constructed by combining the outputs of the grid active power loop control model and the grid reactive power loop control model. The current inner loop control model is also constructed by combining the outputs of the grid active power loop control model and the grid voltage outer loop control model. The output of the current inner loop control model is used to drive the rotor-side converter switching action after SPWM or SVPWM modulation, thus achieving closed-loop control.

[0080] The active power loop is configured to employ virtual inertia control, droop control, and pre-synchronization control of the grid formation angle. The constructed active power loop control model is as follows:

[0081] (1)

[0082] In equation (1), the input is the active power setpoint. P ref Active output value P meas and active droop power P droop ,in, P ref Issued by human intervention or scheduling. P meas The power was calculated using three-phase instantaneous power. P droop Through the sagging coefficient in sagging control K droop_p It is calculated by multiplying the deviation of the network angular frequency ω and the rated angular frequency ω0; J The output is the virtual inertia coefficient; the netting angle is the output. θ GFM It is obtained by integrating the angular frequency ω of the mesh; K ps This is the pre-synchronization ratio coefficient. θ PLLStator / grid voltage locking angle, This indicates that the high-level signal in this section is enabled.

[0083] The reactive power network can be controlled by direct PI control or droop control. The constructed reactive power loop control model is as follows:

[0084] (2)

[0085] In equation (2), the inputs are the reactive power setpoints. Q ref reactive power output value Q meas and voltage rating U 0, the output is the voltage reference value of the outer loop of the grid voltage. E ref , K droop_q The droop coefficient is... K p_q and K i_q These are the proportional and integral parameters of the PI controller in the reactive power loop of the network.

[0086] Among them, the above-mentioned active power output value P meas and reactive power output value Q meas It can be obtained by collecting the three-phase voltage and current and then calculating the three-phase instantaneous power. This calculation method is a conventional existing technology and will not be elaborated further.

[0087] The constructed grid voltage outer loop control model is as follows:

[0088] (3)

[0089] In equation (3), the inputs are the voltage setpoints of the stator d-axis. U sd_ref and feedback value U sd_meas and the voltage setpoint of the stator q-axis. U sq_ref and feedback value U sq_meas When using d-axis orientation, take: U sd_ref =E ref , U sq_ref = 0; When using q-axis orientation, take: U sd_ref = 0, U sq_ref =-Eref The output is the current setpoint for the rotor's d-axis. I rd_ref and the current setpoint of the rotor q-axis I rq_ref ; K p_ud , K i_ud and K p_uq , K i_uq These are the proportional and integral parameters of the PI controller in the outer loop of the stator dq shaft network voltage; feedback value. U sd_meas , U sq_meas The grid-connection angle is based on the output of the grid-connected active power loop control model using stator voltage. θ GFM The coordinates are transformed to obtain the result.

[0090] The constructed current inner loop control model is as follows:

[0091] (4)

[0092] In equation (4), the input is the current setpoint of the rotor d-axis. I rd_ref and feedback value I rd_meas and the current setpoint for the rotor q-axis. I rq_ref and feedback value I rq_meas The output is the voltage setpoint for the rotor's d-axis. U rd_ref and the voltage setpoint of the rotor q-axis U rq_ref ; K p_Id , K i_Id and K p_Iq , K i_Iq These are the proportional and integral parameters of the PI controller in the d-axis current inner loop and the q-axis current inner loop, respectively; feedback value. I rd_meas , I rq_meas Both use the rotor current and the grid-connected active power loop control model output to determine the grid-connected angle. θ GFM The results were obtained through calculation.

[0093] Those skilled in the art will understand that the grid-side converter uses conventional DC voltage and reactive power PI control, which will not be elaborated here.

[0094] In this step, the switching between dq-axis orientation modes of the machine-side grid control model is achieved by adding a discrimination switch to the outer loop of the grid voltage and combining it with the orientation flag.

[0095] like Figure 1 As shown, two dual-input, single-output discrimination switches, denoted as KG1 and KG2, are set at the input of the outer loop of the grid voltage. The input of discrimination switch KG1 is the voltage reference value output by the grid reactive power loop control model. E ref And 0, the input of the discrimination switch KG2 is the voltage reference value output by the grid reactive power loop control model. E ref The negative value and 0.

[0096] The orientation flag DR is used as the discrimination signal, and KG1=KG2=DR. The input switching of the grid reactive power loop control model is realized based on the DR value. DR=1 indicates that the control loop adopts d-axis orientation, and DR=0 indicates that the control loop adopts q-axis orientation. The inputs of switch KG1 are the grid reactive power loop output voltage reference values. E ref (KG1=1) and 0 (KG1=0), the output is the stator d-axis voltage setpoint. U sd_ref The inputs of switch KG2 are respectively - E ref (KG2=0) and 0 (KG2=1), the output is the stator q-axis voltage setpoint. U sq_ref The stator d-axis voltage setpoint U sd_ref and stator q-axis voltage setpoint U sq_ref This serves as the input to the outer loop model of the grid voltage.

[0097] In this step, the selection of the root network or the network structure by adding a discrimination switch to the active power loop and the inner current loop of the network, combined with the switching command, refers to:

[0098] like Figure 1 As shown, a dual-input single-output discrimination switch is set at the output end of the active power loop of the grid and denoted as KM1. Two dual-input single-output discrimination switches are set at the dq axis current setting point of the inner current loop and denoted as KM2 and KM3.

[0099] The discrimination flag KM=0 indicates grid-following mode, and KM=1 indicates grid-building mode. The inputs to discrimination switch KM1 are the grid phase-locked loop angle (KM1=0) and the output angle of the grid-building active power loop (KM1=1), respectively. The inputs to discrimination switch KM2 are the output of the grid-following control active power loop PI controller (KM2=0) and the rotor d-axis current setpoint output from the grid-building voltage outer loop control model, respectively.I rd_ref (KM2=1), the inputs to the discrimination switch KM3 are the output of the grid-controlled reactive power loop PI controller (KM3=0) and the rotor q-axis current setpoint output from the grid voltage outer loop control model. I rd_ref (KM3=1). The trigger signal only sets the KM flag bit. The KM1-KM3 flag bits are the same as the KM flag bits, i.e., KM1=KM2=KM3=KM. When the switching command arrives, the discrimination identifier KM is given. The input selection of the current inner loop control model is switched according to the KM value to realize bidirectional switching between grid-following and grid-building. In addition, the KM flag bit also serves as the input of the trigger signal for the integral reset described below.

[0100] S3: In the control loops of network-following control and network-building control, enable triggering, pre-synchronization, switch switching, and PI integral initial value reset are set. When a switching command is received, the control loop is enabled as needed, the network-building pre-synchronization loop is pre-synchronized, and the discrimination switch is switched. This triggers the reset of all PI integral initial values ​​in the control loop and connects to the switched control loop.

[0101] In this step, such as Figure 1 As shown, setting enable triggers in the control loops of grid-connected control and grid-building control, and enabling the control loops on demand, refers to setting enable switches K in the grid-connected power loop respectively. enable,1 1. Set an enable switch K in the active power loop of the network. enable,2 The same enable switch K is set in the reactive power loop and the outer voltage loop of the grid. enable,3 1. Set the enable switch K in the phase-locked loop. enable,4 And set enable switch K in the pre-synchronization loop of the network. enable,5 Upon receiving a switching command, the enable switch is enabled as needed according to the switching command.

[0102] In this step, resetting the initial PI integral value means resetting the initial PI integral values ​​of the PI controllers in the grid-connected power loop, the grid-connected reactive power loop, and the grid-connected voltage outer loop respectively after receiving a trigger signal at any rising or falling edge; wherein, the reset value of the PI controller in the grid-connected power loop is the current feedback value of the rotor d-axis. I rd_meas Current feedback value of rotor q-axis I rq_meas When the reactive power loop of the grid uses direct PI control or droop control, the reset value of the PI controller is 0. The reset values ​​of the PI controller in the outer voltage loop of the grid are the current feedback values ​​of the rotor d-axis. I rd_meas Current feedback value of rotor q-axis I rq_measFurthermore, the rotor dq axis current reset values ​​are all based on real-time acquisition values ​​plus a delay of one sampling period, denoted as 1 / Z, where Z represents the period in the frequency domain.

[0103] S4: Exit enable as needed and complete the switch.

[0104] Based on the above, the switching between the root network and the structural network will now be explained in detail.

[0105] like Figure 2 As shown, when a doubly fed generator unit is in grid-fed control mode and needs to switch to grid-connected control mode, after receiving the switching command from the root grid to the grid-connected mode, the specific switching process is as follows:

[0106] S1.1: Determine the dq-axis orientation method based on the collected per-unit voltage values ​​of the doubly-fed motor stator dq-axis, and provide the orientation flag DR value. When U d =1、 U q When =0, DR=1; when U d =0、 U q When DR = -1, DR = 0.

[0107] S1.2: Enable the active power loop in the network to enable K enable,2= 1. Enable network pre-synchronization ring to enable K enable,5 =1 to begin pre-synchronization, and after pre-synchronization is complete, set K to... enable,5 =0 Exit pre-synchronization.

[0108] S1.3: Enables the grid reactive power loop and the grid voltage outer loop to enable K enable,3 =1, based on the orientation flag DR value, control the discrimination switches KG1 and KG2 to realize the selection of the network orientation mode.

[0109] S1.4: According to the switching instruction, KM=1, triggering the reset of the initial values ​​of all PI integrals in the control loop and connecting them to the network control loop.

[0110] S1.5: Make K enable,1 =0 and K enable,4 =0, which enables and deactivates the grid power loop and grid phase-locked loop, completing the switching.

[0111] Among them, in K enable,5 During the period =1, when the difference between the mesh angle and the follow-mesh angle is less than 0.05 degrees, K is executed. enable,5 After =0, switch to S1.3. When the difference between the netting angle and the following netting angle is greater than 0.05 degrees, continue to run pre-synchronization until the difference between the netting angle and the following netting angle is less than 0.05 degrees.

[0112] Then, execute S1.4 and S1.5 in sequence. After S1.4 and S1.5 are completed, check whether the system power fluctuates. If the fluctuation value is less than 0.05 degrees and the power output is stable within 10 fundamental frequency cycles, the switching is completed. If the power fluctuation is too large and the fluctuation is still greater than 0.1 pu within 20 fundamental frequency cycles or if a limit is triggered, the switching fails and the unit shuts down for protection.

[0113] like Figure 3 As shown, when the doubly fed generator unit is in grid-connected control mode and needs to switch to root grid control mode, after receiving the switching command to switch from grid-connected to root grid, the specific switching process is as follows:

[0114] S2.1: By default, the orientation method consistent with the network control has been adopted before network construction and operation, or the orientation method determined according to S1.1 is adopted, enabling the network phase-locked loop to make K enable,4 =1, lock the stator voltage to obtain the grid angle.

[0115] S2.2: Calculate the deviation between the output frequency of the phase-locked loop and the rated frequency. When the deviation is less than 0.03Hz, complete the locking and proceed to S.3. When the deviation is greater than 0.03Hz, continue running S2.1 and S2.2 until the deviation is less than 0.03Hz.

[0116] S2.3: Enable network pre-synchronization loop to enable K enable,5 =1 to begin pre-synchronization, and after pre-synchronization is complete, set K to... enable,5 =0 Exit pre-synchronization.

[0117] S2.4: Enable grid power loop to enable K enable,1 =1, based on the directional flag bit DR value, control the discrimination switches KG1 and KG2.

[0118] S2.5: According to the switching command, KM=0, triggering the reset of the initial values ​​of all PI integrals in the control loop and connecting them to the network control loop.

[0119] S2.6: Make K enable,2= 0 and K enable,3= 0 enables the grid active power loop, grid reactive power loop, and grid voltage outer loop to be deactivated, completing the switching.

[0120] Among them, in K enable,5 During period =1, when the difference between the meshing angle and the follow-me-off angle is less than 0.05 degrees, K is executed. enable,5 After =0, switch to S2.4. When the difference between the netting angle and the following netting angle is greater than 0.05 degrees, continue to run pre-synchronization until the difference between the netting angle and the following netting angle is less than 0.05 degrees.

[0121] Then, execute S2.5 and S2.6 in sequence. After S2.5 and S2.6 are completed, check whether the system power fluctuates. If the fluctuation value is less than 0.05 degrees and the power output is stable within 10 fundamental frequency cycles, the switching is completed. If the power fluctuation is too large and the fluctuation is still greater than 0.1 pu within 20 fundamental frequency cycles or if a limit is triggered, the switching fails and the unit shuts down for protection.

[0122] Example 2

[0123] This embodiment also provides a bidirectional, disturbance-free switching system for doubly-fed generator unit grid-connection control, such as... Figure 4 As shown, it includes:

[0124] The orientation flag confirmation module is used to determine the dq axis orientation mode of the doubly fed generator unit during current operation and to provide the orientation flag bit.

[0125] The model building module is used to build two sets of machine-side network control models with different dq-axis orientations.

[0126] The switching selection module is used to switch the machine-side grid control model between dq-axis orientation modes by adding a discrimination switch to the outer loop of the grid voltage and combining it with the orientation flag; it is also used to select the root grid or the grid by adding discrimination switches to the active power loop and the inner current loop of the grid and combining them with the switching command.

[0127] The processing module is used to set enable triggering, pre-synchronization, switch switching and PI integral initial value reset links in the control loop of network-following control and network-building control. After receiving the switching command, it enables the control loop as needed, pre-synchronizes the network-building pre-synchronization loop and switches the discrimination switch, triggers the reset of the initial PI integral values ​​of all PI loops in the control loop and connects to the control loop after the switch.

[0128] The exit module is used to exit the enable function as needed and complete the switch.

[0129] In detail, based on the same innovative concept, the specific implementation process of this embodiment can be found in the implementation of the method in Embodiment 1, and the repeated parts will not be described again.

[0130] Experimental Example

[0131] This experimental example tests the method provided by the present invention. Figure 5 and Figure 6 The test results diagram between the wire mesh and the wire mesh is shown. Figure 5 and Figure 6 It can be seen that after adopting the method of the present invention, the fluctuation during bidirectional switching is minimal and does not affect the safe and stable operation of the system. Figure 7 and Figure 8 The image shows test results without considering the dq axis orientation when switching between the root mesh and the mesh structure. Figure 7 and Figure 8 It is known that when the orientation angles of the following network and the network structure are inconsistent, large power fluctuations and limiting occur before and after the handover, seriously threatening system stability. Figure 9 The figure shows the test results of switching from the root network to the structural network without considering whether pre-synchronization was completed. As can be seen from the figure, significant power fluctuations occur during the switching process, which also seriously threatens system stability. Therefore, this invention can achieve seamless switching between the root network and the structural network, and the switching process is smooth and reliable, demonstrating greater innovation.

Claims

1. A method for bidirectional, disturbance-free switching of doubly-fed generator unit's grid-connection control, characterized in that... Includes the following steps: S1: Determine the dq axis orientation mode of the doubly fed generator unit during current operation and provide the orientation flag; S2: Construct two sets of machine-side grid construction control models under different dq-axis orientations. By adding a discrimination switch to the outer loop of the grid construction voltage and combining it with the orientation flag, the machine-side grid construction control model can be switched between dq-axis orientation modes. By adding discrimination switches to the active power loop and the inner current loop of the grid construction and combining them with the switching command, the selection of grid following or grid construction can be realized. S3: In the control loop of network control and network construction control, enable triggering, pre-synchronization, switch switching and PI integral initial value reset are set. When a switching command is received, the control loop is enabled as needed, the network construction pre-synchronization loop is pre-synchronized and the discrimination switch is switched, triggering the reset of all PI integral initial values ​​in the control loop and connecting to the control loop after the switch. S4: Exit enable as needed and complete the switch.

2. The bidirectional, disturbance-free switching method for doubly-fed generator unit grid-connection control according to claim 1, characterized in that: In S2, the constructed machine-side grid control model includes a grid active power loop control model, a grid reactive power loop control model, a grid voltage outer loop control model, and a current inner loop control model. The grid voltage outer loop control model is constructed by combining the output results of the grid active power loop control model and the grid reactive power loop control model. The current inner loop control model is also constructed by combining the output results of the grid active power loop control model and the grid voltage outer loop control model. The output results of the current inner loop control model are used to drive the rotor-side converter switching action after SPWM or SVPWM modulation to achieve closed-loop control.

3. The bidirectional, disturbance-free switching method for doubly-fed generator unit grid-connection control according to claim 2, characterized in that: In S2, the active power loop is configured to use virtual inertia control, droop control, and pre-synchronization control of the grid formation angle. The constructed grid formation active power loop control model is as follows: （1） In equation (1), the input is the active power setpoint. P ref Active output value P meas and active droop power P droop ,in, P ref Issued by human intervention or scheduling. P meas The power was calculated using three-phase instantaneous power. P droop Through the sagging coefficient in sagging control K droop_p It is calculated by multiplying the deviation of the network angular frequency ω and the rated angular frequency ω0; J The output is the virtual inertia coefficient; the netting angle is the output. θ GFM It is obtained by integrating the angular frequency ω of the mesh; K ps This is the pre-synchronization ratio coefficient. θ PLL Stator / grid voltage locking angle, This indicates that the high-level signal in this section is enabled. The constructed reactive power loop control model is as follows: （2） In equation (2), the inputs are the reactive power setpoints. Q ref reactive power output value Q meas and voltage rating U 0, the output is the voltage reference value of the outer loop of the grid voltage. E ref , K droop_q The droop coefficient is... K p_q and K i_q These are the proportional and integral parameters of the PI controller in the reactive power loop of the network; The constructed grid voltage outer loop control model is as follows: （3） In equation (3), the inputs are the voltage setpoints of the stator d-axis. U sd_ref and feedback value U sd_meas and the voltage setpoint of the stator q-axis. U sq_ref and feedback value U sq_meas When using d-axis orientation, take: U sd_ref =E ref , U sq_ref = 0; When using q-axis orientation, take: U sd_ref = 0, U sq_ref =-E ref The output is the current setpoint for the rotor's d-axis. I rd_ref and the current setpoint of the rotor q-axis I rq_ref ; K p_ud , K i_ud and K p_uq , K i_uq These are the proportional and integral parameters of the PI controller in the outer loop of the stator dq shaft network voltage; feedback value. U sd_meas , U sq_meas The grid-connection angle is based on the output of the grid-connected active power loop control model using stator voltage. θ GFM The coordinates are transformed to obtain the result. The constructed current inner loop control model is as follows: （4） In equation (4), the input is the current setpoint of the rotor d-axis. I rd_ref and feedback value I rd_meas and the current setpoint for the rotor q-axis. I rq_ref and feedback value I rq_meas The output is the voltage setpoint for the rotor's d-axis. U rd_ref and the voltage setpoint of the rotor q-axis U rq_ref ; K p_Id , K i_Id and K p_Iq , K i_Iq These are the proportional and integral parameters of the PI controller in the d-axis current inner loop and the q-axis current inner loop, respectively; feedback value. I rd_meas , I rq_meas Both use the rotor current and the grid-connected active power loop control model output to determine the grid-connected angle. θ GFM The results were obtained through calculation.

4. The bidirectional, disturbance-free switching method for doubly-fed generator unit grid-connection control according to claim 3, characterized in that: In S1, the dq axis orientation mode is determined based on the per-unit voltage values ​​of the doubly fed generator stator dq axis. The orientation flag of the dq axis orientation mode is set as DR. If Ud=1 and Uq=0, the doubly fed generator is currently d-axis oriented and is marked as DR=1; if Ud=0 and Uq=-1, the doubly fed generator is currently q-axis oriented and is marked as DR=0.

5. The bidirectional, disturbance-free switching method for doubly-fed generator unit grid-connection control according to claim 4, characterized in that: In S2, the switching between dq-axis orientation modes of the machine-side network control model is achieved by adding a discrimination switch to the outer loop of the network voltage and combining it with the orientation flag. Two dual-input, single-output discrimination switches, denoted as KG1 and KG2, are set at the input of the outer loop of the grid voltage. The input of discrimination switch KG1 is the voltage reference value output by the grid reactive power loop control model. E ref And 0, the input of the discrimination switch KG2 is the voltage reference value output by the grid reactive power loop control model. E ref Negative values ​​and 0; The directional flag bit DR is used as the discrimination signal, and KG1=KG2=DR. The input switching of the grid reactive power loop control model is realized according to the DR value.

6. A bidirectional, disturbance-free switching method for doubly-fed generator unit grid-connection control according to claim 2 or 3, characterized in that: In S2, the selection of grid connection or grid integration is achieved by adding a discrimination switch to the grid active power loop and the inner current loop, combined with switching commands. A dual-input single-output discrimination switch, denoted as KM1, is set at the output end of the grid active power loop. Two dual-input single-output discrimination switches, denoted as KM2 and KM3, are set at the dq-axis current setpoint of the inner current loop. The inputs of discrimination switch KM1 are the grid phase-locked loop angle and the output angle of the grid active power loop, respectively. The inputs of discrimination switch KM2 are the output of the PI controller of the grid-connected active power loop and the rotor d-axis current setpoint output of the grid voltage outer loop control model, respectively. I rd_ref The inputs to the discrimination switch KM3 are the output of the grid-controlled reactive power loop PI controller and the rotor q-axis current setpoint output of the grid voltage outer loop control model. I rd_ref ; The discrimination flag KM=0 indicates the grid-following mode, and the discrimination flag KM=1 indicates the grid-building mode. KM1=KM2=KM3=KM. When a switching command is received, the discrimination flag KM is given. The input selection of the current inner loop control model is switched according to the KM value to realize the bidirectional switching between grid-following and grid-building modes.

7. A method for bidirectional, disturbance-free switching of doubly-fed generator unit grid-connection control according to claim 2 or 3, characterized in that: In S3, setting enable triggers in the control loops of grid-connected control and grid-building control, and enabling the control loops on demand, refers to setting enable switches K in the grid-connected power loop respectively. enable,1 1. Set an enable switch K in the active power loop of the network. enable,2 The same enable switch K is set in the reactive power loop and the outer voltage loop of the grid. enable,3 1. Set the enable switch K in the phase-locked loop. enable,4 And set enable switch K in the pre-synchronization loop of the network. enable,5 Upon receiving a switching command, the enable switch is enabled as needed according to the switching command.

8. The bidirectional, disturbance-free switching method for doubly-fed generator unit grid-connection control according to claim 7, characterized in that: In S3, PI integral initial value reset refers to resetting the initial PI integral values ​​of the PI controllers in the grid-connected power loop, grid-connected reactive power loop, and grid-connected voltage outer loop respectively after receiving a trigger signal at any rising or falling edge; whereby the reset values ​​of the PI controller in the grid-connected power loop are the current feedback values ​​of the rotor d-axis. I rd_meas Current feedback value of rotor q-axis I rq_meas When the reactive power loop of the grid uses direct PI control or droop control, the reset value of the PI controller is 0. The reset values ​​of the PI controller in the outer voltage loop of the grid are the current feedback values ​​of the rotor d-axis. I rd_meas Current feedback value of rotor q-axis I rq_meas Furthermore, the rotor dq axis current reset values ​​are all based on real-time acquisition values ​​plus a delay of one sampling period, denoted as 1 / Z.

9. A method for bidirectional, disturbance-free switching of doubly-fed generator unit grid-connection control according to claim 8, characterized in that: When a handover command to switch from the network to the network structure is received, the specific handover process is as follows: S1.1: Determine the orientation mode of the dq axis based on the collected per-unit voltage values ​​of the doubly fed motor stator dq axis, and give the orientation flag bit DR value; S1.2: Enable the active power loop in the network to enable K enable,2= 1. Enable network pre-synchronization ring to enable K enable,5 =1, after pre-synchronization is complete, make K =1, enable,5 =0 Exit pre-synchronization; S1.3: Enables the grid reactive power loop and the grid voltage outer loop to enable K enable,3 =1, control the discrimination switches KG1 and KG2 according to the orientation flag bit DR value; S1.4: According to the switching instruction, KM=1, triggering the reset of the initial values ​​of all PI integrals in the control loop and connecting them to the network control loop; S1.5: Make K enable,1 =0 and K enable,4 =0, which enables the grid power loop and the grid phase-locked loop to be deactivated, completing the switching; Among them, in K enable,5 During the period =1, when the difference between the mesh angle and the follow-mesh angle is less than 0.05 degrees, K is executed. enable,5 After =0, switch to S1.

3. When the difference between the netting angle and the following netting angle is greater than 0.05 degrees, continue to run pre-synchronization until the difference between the netting angle and the following netting angle is less than 0.05 degrees. Then, execute S1.4 and S1.5 in sequence and determine whether the system power fluctuates. If the fluctuation value is less than 0.05 degrees and the power output is stable within 10 fundamental frequency cycles, the switching is completed. If the power fluctuation is too large and the fluctuation is still greater than 0.1 pu within 20 fundamental frequency cycles or if a limit is triggered, the switching fails and the unit shuts down for protection.

10. A bidirectional, disturbance-free switching method for doubly-fed generator unit grid-connection control according to claim 8, characterized in that: When a switchover command to switch from network architecture to root network is received, the specific switchover process is as follows: S2.1: Enable the K-loop lock-in. enable,4 =1, lock the stator voltage to obtain the grid angle; S2.2: Calculate the deviation between the output frequency of the phase-locked loop and the rated frequency. When the deviation is less than 0.03Hz, proceed to S.

3. S2.3: Enable network pre-synchronization loop to enable K enable,5 =1, after pre-synchronization is complete, make K =1, enable,5 =0 Exit pre-synchronization; S2.4: Enable grid power loop to enable K enable,1 =1, control the discrimination switches KG1 and KG2 according to the orientation flag bit DR value; S2.5: According to the switching command, KM=0, triggering the reset of the initial values ​​of all PI integrals in the control loop and connecting to the network control loop; S2.6: Make K enable,2= 0 and K enable,3= 0, which enables and deactivates the active power loop, reactive power loop, and outer voltage loop of the grid, completing the switching; Among them, in K enable,5 During period =1, when the difference between the meshing angle and the follow-me-off angle is less than 0.05 degrees, K is executed. enable,5 After =0, switch to S2.

4. When the difference between the netting angle and the following netting angle is greater than 0.05 degrees, continue to run pre-synchronization until the difference between the netting angle and the following netting angle is less than 0.05 degrees. Then, execute S2.5 and S2.6 in sequence and determine whether the system power fluctuates. If the fluctuation value is less than 0.05 degrees and the power output is stable within 10 fundamental frequency cycles, the switching is completed. If the power fluctuation is too large and the fluctuation is still greater than 0.1 pu within 20 fundamental frequency cycles or if a limit is triggered, the switching fails and the unit shuts down for protection.

11. A bidirectional, disturbance-free switching system for doubly-fed generator unit grid-connection control, characterized in that... include: The orientation flag confirmation module is used to determine the dq axis orientation mode of the doubly fed generator unit during current operation and to provide the orientation flag bit; The model building module is used to build two sets of machine-side network control models with different dq-axis orientations; The switching selection module is used to switch the machine-side network control model between dq-axis orientation modes by adding a discrimination switch to the outer loop of the network voltage and combining it with the orientation flag. This is used to select between following the grid or forming the grid by adding a discrimination switch to the active power loop and the inner current loop of the grid and combining it with switching commands. The processing module is used to set the enable trigger, pre-synchronization, switch switching and PI integral initial value reset links in the control loop of network-following control and network-building control. After receiving the switching command, it enables the control loop as needed, pre-synchronizes the network-building pre-synchronization loop and switches the discrimination switch, triggers the reset of the initial PI integral values ​​in the control loop and connects to the control loop after the switch. The exit module is used to exit the enable function as needed and complete the switch.

Images