A bridge arm current model prediction network control method applied to a marine wind power DRU-MMC system

Abstract

The application discloses a bridge arm current model prediction network construction control method applied to an offshore wind power DRU-MMC system. The bridge arm current prediction model of the MMC is used to organically integrate the voltage control outer ring and the current control inner ring of the traditional network construction type control strategy, control of capacitor voltage, circulating current, AC current, AC voltage and power and other targets of the MMC is converted into regulation of the bridge arm current, the logic structure of the controller is greatly simplified, nesting of the controller is reduced, a complex setting process of multiple control parameters is effectively avoided, a problem that control effects among multiple targets in the traditional strategy cannot be considered is solved, coupling phenomena among multiple traditional controllers under a weak power grid are greatly eliminated, and therefore, a stable AC power grid is built for the offshore wind power station. The application has the advantages of simple control logic, fast response speed and wide power grid parameter adaptation range, can effectively improve operation performance of a far-sea wind power conversion platform, and has high application value in actual engineering.

Description

Technical Field

[0001] This invention belongs to the field of power system transmission and distribution technology, specifically relating to a bridge arm current model prediction grid control method applied to offshore wind power DRU-MMC systems. Background Technology

[0002] As offshore wind power development gradually extends into deep-sea areas, the stable, efficient, and economical long-distance transmission of large-capacity electrical energy has become a focus of widespread academic attention. As the mainstream engineering solution currently unrestricted by transmission distance, high-voltage flexible DC transmission systems based on modular multilevel converters (MMCs) still face significant challenges in terms of the size, weight, and construction cost of offshore converter platforms. With the increase in transmission capacity and voltage levels, the surge in the number of submodules further deteriorates the economics of offshore converter platforms. To further improve the economics of flexible DC transmission systems, the academic community combines diode rectifier units (DRUs), which offer large transmission capacity, low cost, high reliability, and small size, with MMCs through series or parallel connections to form DRU-MMC transmission systems. This allows the DRUs to handle most of the DC voltage drop or transmit most of the active power, thereby reducing the selection requirements for MMCs and lowering the overall size, weight, and construction cost of offshore converter platforms. Simultaneously, the bidirectional power transmission and grid control advantages of MMCs can be fully utilized to achieve the construction of offshore AC grids and black start of offshore wind farms without modifying the grid structure of offshore wind turbines. Compared to the control of a pure MMC system, the control of a DRU-MMC system is more complex, with more control objectives, and it has higher requirements for stability, robustness, and response speed.

[0003] Currently, network-based control strategies proposed in engineering and academia still focus on decoupled control based on proportional-integral (PI) and proportional-resonant (PR) controllers, mainly consisting of a three-layer nested structure from the power control loop to the AC voltage control loop, and then to the current control inner loop. However, such methods suffer from problems such as numerous and complex control parameters, prominent multi-controller coupling, difficulty in achieving multi-objective control effects, and poor robustness to changes in grid parameters. Furthermore, their response speed cannot meet the needs of practical engineering. Summary of the Invention

[0004] The purpose of this invention is to provide a bridge arm current model prediction grid control method for offshore wind power DRU-MMC systems. Compared with traditional control strategies, it greatly simplifies the logic structure and control parameters, avoids multi-layer nesting of controllers, and solves the problems of complex tuning of multiple control parameters, prominent coupling of multiple controllers, inability to achieve multiple target control effects, and poor robustness to the power grid environment in traditional control strategies.

[0005] A grid-based control method for arm current model prediction in offshore wind power DRU-MMC systems is characterized by: the grid-based arm current model prediction control technology being applicable to offshore converter platforms formed by parallel or series connections of DRU and MMC type topologies; the grid-based arm current model prediction control method being applicable to offshore converter platforms formed by parallel or series connections of diode rectifier units (DRU) and modular multilevel converters (MMC) type topologies; the control method being based on a single-step prediction model of MMC arm current, including active power control and reactive power compensation, arm current reference value calculation, AC voltage reference value calculation, arm current model calculation, and submodule commissioning quantity prediction steps; the active power control and reactive power compensation steps can be based on active power command P. m_cmd Real-time reactive power q on the grid side m Generate three-phase AC current command value i j_cmd j=a,b,c; the calculation steps for the bridge arm current reference value can be combined with P m_cmd and MMC DC side voltage command value U m_dc Obtain the upper and lower bridge arm current reference signals i jp_cmd and i jn_cmd The steps for calculating the AC voltage reference value can construct a stable AC voltage reference signal u. j_cmd The steps for calculating the bridge arm current model and predicting the number of submodules to be put into operation can be based on each instruction i. jp_cmd i jn_cmd u j_cmd And the sampled values, for the number n of bridge arm deployment sub-modules jp_pre n jn_pre The prediction is performed, and the prediction result is directly fed into the capacitor voltage sorting strategy to generate the driving signal for MMC.

[0006] Furthermore, in the active power control and reactive power compensation steps, the three-phase AC current command value i j_cmd Commands i from the d-axis and q-axis of alternating current d_cmd i q_cmd Obtained after inverse Parker transformation; the i d_cmd According to the active instruction P m_cmd Direct calculation, no controller adjustment required; the i q_cmd The real-time reactive power q on the grid side, after filtering out secondary fluctuations, is controlled by a PI controller. m It is obtained through regulation.

[0007] Furthermore, the reference signal for the bridge arm current is respectively transmitted through i jp_cmd =P m_cmd / U m_dc / 3+0.5i j_cmd and i jn_cmd =P m_cmd / U m_dc / 3-0.5i j_cmd Obtained through calculation.

[0008] Furthermore, the AC voltage reference value u j_cmd byu d_cmd and u q_cmd After inverse Parker transformation, the required AC voltage phase θ is obtained by integration from a given frequency ω. d_cmd and u q_cmd These are the reference values ​​for the d-axis and q-axis components of the AC voltage, respectively.

[0009] Furthermore, in the steps of calculating the bridge arm current model and predicting the number of submodules to be put into operation, the number of submodules to be put into operation is determined by n. jp_pre =round{[0.5u m_dc -u j_cmd -R arm i jp -L arm (i jp_cmd -i jp ) / T s ] / u C_ave_jp} and n jn_pre =round{[0.5u m_dc +u j_cmd -R arm i jp -L arm (i jn_cmd -i jn ) / T s ] / u C_ave_jn} is calculated to obtain, where u m_dc R is the sampled value of the DC side voltage of the MMC. arm and L arm These are the bridge arm equivalent resistance and bridge arm inductance of the MMC, respectively. jp and i jn These are the sampled values ​​of the upper and lower arm currents of phase j of the MMC, respectively, T s For the system control cycle, u C_ave_jp and u C_ave_jn These are the average values ​​of the instantaneous voltages of the capacitors in the upper and lower bridge arm submodules, respectively.

[0010] This invention utilizes the arm current prediction model of the MMC (Multi-Mechanical Control Controller) to organically integrate the voltage control outer loop and current control inner loop of the traditional grid-type control strategy. It transforms the control of targets such as capacitor voltage, circulating current, AC current, AC voltage, and power of the MMC into the regulation of the arm current. This not only significantly simplifies the controller's logic structure and reduces controller nesting but also effectively avoids the complex tuning process of multiple control parameters. It solves the problem of the inability to simultaneously achieve control effects across multiple objectives in traditional strategies and greatly eliminates the coupling phenomenon between multiple traditional controllers in weak grid conditions, thereby constructing a stable AC grid for offshore wind turbines. Compared with traditional strategies, this invention has the advantages of simple control logic, fast response speed, and wide adaptability to grid parameters, effectively improving the operational performance of offshore wind power converter platforms and possessing high application value in practical engineering.

[0011] Based on the above technical solution, the present invention has the following beneficial technical effects:

[0012] 1. For offshore wind power DRU-MMC systems, this invention features a simple control structure, no multi-layered controller nesting, few control parameters, and a simple tuning process.

[0013] 2. This invention can effectively balance the control effects between multiple objectives, weaken or eliminate the coupling phenomenon between multiple controllers, and has the advantage of fast response speed.

[0014] 3. This invention has strong robustness to changes in the parameters of the offshore AC power grid. Attached Figure Description

[0015] Figure 1(a) and Figure 1(b) are schematic diagrams of the topology of the offshore wind power DRU-MMC series system and the DRU-MMC parallel system, respectively.

[0016] Figure 2(a) and Figure 2(b) are the overall control block diagram and the control block diagram of the active power control and reactive power compensation steps of the bridge arm current model prediction network control method provided by the present invention, respectively.

[0017] Figures 3(a), 3(b), 3(c), 3(d), 3(e), 3(f), 3(g), and 3(h) show the steady-state and transient response characteristics of the DRU-MMC system for each AC side power, offshore AC grid voltage, offshore AC grid current, MMC AC current, DRU AC current, each DC side voltage, MMC A-phase bridge arm current and circulating current, and MMC A-phase submodule capacitor voltage, respectively. Detailed Implementation

[0018] To describe the present invention in more detail, the technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] This implementation scheme is applied to the offshore wind power DRU-MMC system, proposing a bridge arm current model prediction grid control method. This method can be used for offshore converter platforms formed by parallel or series connections of DRU and MMC type topologies. The topology diagrams of the DRU-MMC series system and the DRU-MMC parallel system are shown in Figure 1(a) and Figure 1(b), respectively. The proposed bridge arm current model prediction grid control method organically integrates the voltage control outer loop and current control inner loop of the traditional grid control strategy using the bridge arm current prediction model of the MMC. Its control block diagram is shown in Figure 2(a), including active power control and reactive power compensation, bridge arm current reference value calculation, AC voltage reference value calculation, bridge arm current model calculation, and submodule commissioning quantity prediction steps. The control block diagram of the active power control and reactive power compensation link is shown in Figure 2(b). The active power control and reactive power compensation steps are based on the active power command P... m_cmd Real-time reactive power q on the grid side m Generate three-phase AC current command value i j_cmd (j=a, b,c); the calculation steps for the bridge arm current reference value can be combined with P m_cmd and MMC DC side voltage command value U m_dc Obtain the upper and lower bridge arm current reference signals i jp_cmd and i jn_cmd The AC voltage reference value calculation step constructs a stable AC voltage reference signal u. j_cmd The steps for calculating the bridge arm current model and predicting the number of submodules to be put into operation can be based on each instruction i. jp_cmd i jn_cmd u j_cmd And the sampled values, for the number n of bridge arm deployment sub-modules jp_pre n jn_pre The prediction is performed, and the result is directly fed into the capacitor voltage sorting strategy to generate the driving signal for the MMC.

[0020] Three-phase AC current command value i j_cmd Commands i from the d-axis and q-axis of alternating current d_cmd i q_cmd It is obtained after inverse Parker transformation. Where i d_cmd According to the active power command P m_cmd Direct calculation, no controller adjustment required, i q_cmd The real-time reactive power q on the grid side, after filtering out secondary fluctuations, is controlled by a PI controller. m The bridge arm current reference value i is obtained through adjustment. jp_cmd and i jn_cmd Can be accessed via i jp_cmd =P m_cmd / U m_dc / 3+0.5i j_cmd and i jn_cmd =Pm_cmd / U m_dc / 3-0.5i j_cmd Calculated AC voltage reference value u j_cmd can be derived from u d_cmd and u q_cmd After inverse Parker transformation, the required AC voltage phase θ is obtained by integration from a given frequency ω. d_cmd and u q_cmd These are the reference values ​​for the d-axis and q-axis components of the AC voltage, respectively.

[0021] Ultimately, the number of submodules deployed in each bridge arm is determined by n. jp_pre =round{[0.5u m_dc -u j_cmd -R arm i jp -L arm (i jp_cmd -i jp ) / T s ] / u C_ave_jp} and n jn_pre =round{[0.5u m_dc +u j_cmd -R arm i jp -L arm (i jn_cmd -i jn ) / T s ] / u C_ave_jn The prediction calculation is obtained, where u m_dc R is the sampled value of the DC side voltage of the MMC. arm and L arm These are the bridge arm equivalent resistance and bridge arm inductance of the MMC, respectively. jp and i jn These are the sampled values ​​of the upper and lower arm currents of phase j of the MMC, respectively, T s For the system control cycle, u C_ave_jp and u C_ave_jn These are the average values ​​of the instantaneous voltages of the capacitors in the upper and lower bridge arm submodules, respectively.

[0022] To verify the feasibility of the bridge arm current model prediction network control method proposed in this invention, an electromagnetic transient simulation model of the offshore wind power DRU-MMC series system was built in MATLAB / Simulink, taking the DRU-MMC series system topology as an example. The main system parameters are shown in Table 1.

[0023] Under the proposed bridge arm current model prediction grid control method, the active power transmitted by the offshore wind power DRU-MMC series system decreases from 1.0 pu to 0.7 pu at t=1.0s. The simulation results of the steady-state and transient response characteristics of the system during this process are shown in Figures 3(a), 3(b), 3(c), 3(d), 3(e), 3(f), 3(g), and 3(h).

[0024]

[0025] As can be seen from Figures 3(a), 3(b), 3(c), 3(d), 3(e), 3(f), 3(g), and 3(h), the bridge arm current model prediction network control method proposed in this invention can control the DRU-MMC system to operate stably under rated operating conditions and transient processes with varying transmission power. The MMC transmits only a very small portion of active power and can dynamically compensate for the reactive power required by the DRU in real time, enabling the system to operate stably under unity power factor conditions. Simultaneously, during transient processes, the MMC can still maintain the stability of the offshore AC grid, DC side voltage, and submodule capacitor voltage.

[0026] The above specific embodiments are used to explain and illustrate the present invention, and are merely implementation examples of the present invention, not to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made to the present invention within the scope of protection of the claims of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for network control based on bridge arm current model prediction in offshore wind power DRU-MMC systems, characterized in that: The network-type arm current model predictive control method is used for offshore converter platforms formed by parallel or series connection of diode rectifier units (DRUs) and modular multilevel converters (MMCs). The control method is based on a single-step predictive model of the MMC arm current and includes active power control and reactive power compensation, arm current reference value calculation, AC voltage reference value calculation, arm current model calculation, and submodule deployment prediction steps. The active power control and reactive power compensation steps can be implemented according to the active power command P. m_cmd Real-time reactive power q on the grid side m Generate three-phase AC current command value i j_cmd j=a, b, c; the calculation steps for the bridge arm current reference value can be combined with P m_cmd and MMC DC side voltage command value U m_dc Obtain the upper and lower bridge arm current reference signals i jp_cmd and i jn_cmd ; The steps for calculating the AC voltage reference value can construct a stable AC voltage reference signal u. j_cmd The steps for calculating the bridge arm current model and predicting the number of submodules to be put into operation can be based on each instruction i. jp_cmd i jn_cmd u j_cmd And the sampled values, for the number n of bridge arm deployment sub-modules jp_pre n jn_pre The prediction is performed, and the prediction result is directly fed into the capacitor voltage sorting strategy to generate the driving signal for MMC.

2. The bridge arm current model prediction network control method according to claim 1, characterized in that: In the active power control and reactive power compensation steps, the three-phase AC current command value i j_cmd Commands i from the d-axis and q-axis of alternating current d_cmd i q_cmd Obtained after inverse Parker transformation; the i d_cmd According to the active instruction P m_cmd Direct calculation, no controller adjustment required; the i q_cmd The real-time reactive power q on the grid side, after filtering out secondary fluctuations, is controlled by a PI controller. m It is obtained through regulation.

3. The bridge arm current model prediction network control method according to claim 1, characterized in that: The bridge arm current reference signal is respectively passed through i jp_cmd =P m_cmd / U m_dc / 3+0.5i j_cmd and i jn_cmd =P m_cmd / U m_dc / 3-0.5i j_cmd Obtained through calculation.

4. The bridge arm current model prediction network control method according to claim 1, characterized in that: The AC voltage reference value u j_cmd byu d_cmd and u q_cmd After inverse Parker transformation, the required AC voltage phase θ is obtained by integration from a given frequency ω. d_cmd and u q_cmd These are the reference values ​​for the d-axis and q-axis components of the AC voltage, respectively.

5. The bridge arm current model prediction network control method according to claim 1, characterized in that: In the steps of calculating the bridge arm current model and predicting the number of submodules to be put into operation, the number of submodules to be put into operation is respectively determined by n. jp_pre =round{[0.5u m_dc -u j_cmd -R arm i jp -L arm (i jp_cmd -i jp ) / T s ] / u C_ave_jp } and n jn_pre =round{[0.5u m_dc +u j_cmd -R arm i jp -L arm (i jn_cmd -i jn ) / T s ] / u C_ave_jn } is calculated to obtain, where u m_dc R is the sampled value of the DC side voltage of the MMC. arm and L arm These are the bridge arm equivalent resistance and bridge arm inductance of the MMC, respectively. jp and i jn These are the sampled values ​​of the upper and lower arm currents of phase j of the MMC, respectively, T s For the system control cycle, u C_ave_jp and u C_ave_jn These are the average values ​​of the instantaneous voltages of the capacitors in the upper and lower bridge arm submodules, respectively.

Images