Adaptive configuration method and system for control mode of offshore wind power flexible direct current station based on typical topology

Abstract

The application discloses a method and system for adaptive configuration of offshore wind power flexible DC converter station control mode based on typical topology, and particularly relates to the technical field of adaptive DC converter station, which comprises S1: topology real-time identification, S2: electrical stress and weak point evaluation, S3: preliminary matching of control mode, S4: multi-time scale parameter optimization, S5: hardware-in-the-loop pre-verification and fine tuning, and S6: disturbance-free switching and closed-loop feedback. Through topology sensing and identification and working condition analysis and evaluation, the application matches the control mode in real time, and realizes the precise optimization of control parameters by using millisecond-level model predictive control and second-level adaptive bandwidth adjustment. In the power grid fault scene, the application significantly suppresses the DC voltage fluctuation and greatly shortens the power recovery time, thereby significantly enhancing the fault ride-through capability and dynamic quality of the system.

Description

Technical Field

[0001] This invention relates to the field of adaptive technology for DC converter stations, and more specifically, to an adaptive configuration method and system for the control mode of offshore wind power flexible DC converter stations based on typical topologies. Background Technology

[0002] Flexible DC transmission technology for offshore wind power is the main means of transmitting power from large-scale wind farms in deep-sea areas. In actual operation, the offshore wind power flexible DC system faces a complex and ever-changing operating environment: on the one hand, the intermittency and randomness of wind speed cause large fluctuations in the output of the wind farm; on the other hand, the topology of the system itself may change due to equipment maintenance, fault clearing or adjustment of operating mode.

[0003] Among existing published documents, patent publication number CN120728646A discloses a method, system, and medium for suppressing broadband oscillations in offshore wind power flexible DC grid-connected systems. This technology reconstructs the wind turbine grid-side converter into a matrix control structure by cascading weighted coefficient matrices between the inner and outer loop control loops of the converter when the stability margin is lower than a preset margin threshold. Based on impedance sensitivity iterative optimization of the matrix control structure, and simultaneously performing impedance reshaping by connecting a low-pass filter in series in the voltage and current feedback channels, broadband oscillation risk suppression is achieved. This invention achieves broadband oscillation suppression in offshore wind power flexible DC grid-connected systems through impedance modeling and matrix control strategies. However, this technology still has the following problems.

[0004] When the actual operating conditions deviate from the design conditions, especially when the topology changes, this fixed-parameter control method often fails to balance system stability and dynamic response speed, and is prone to problems such as excessive DC voltage fluctuations, power response oscillations, or even system instability.

[0005] While existing technologies mention adaptive control adjustments, most are limited to adjusting the droop coefficient at a single level or responding only to grid faults for specific disturbances. There is a lack of a comprehensive solution that can simultaneously sense topology changes and operating condition evolution, and coordinately optimize the configuration of control modes and parameters across multiple time scales. Summary of the Invention

[0006] To overcome the aforementioned shortcomings of the prior art and achieve the above objectives, the present invention provides the following technical solution: an adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology, comprising the following specific steps: S1: Real-time topology identification: Real-time acquisition of operational data of offshore wind power flexible DC system, and identification of the actual topology type of the current system based on waveform similarity matching algorithm; S2: Electrical stress and vulnerability assessment: Combining the identified topology and external environment data, extract the feature vector of the current operating condition, and conduct online assessment of the electrical stress of key equipment in the converter station to identify system vulnerabilities; S3: Preliminary control mode matching: Based on the preset "topology-operating condition-control mode" mapping relationship library, one or more candidate control modes and corresponding initial control parameters are initially matched according to the topology structure type and operating condition feature vector; S4: Multi-timescale parameter optimization: Construct a multi-timescale optimization model, and based on the candidate control modes, perform hierarchical rolling optimization on the control parameters of the current inner loop and voltage outer loop to obtain the optimized control parameters; S5: Hardware-in-the-Loop Pre-verification and Fine-tuning: Using a parallel hardware-in-the-loop simulation unit, the dynamic response of the control strategy containing the optimized control parameters is pre-verified for the next control cycle, and the optimized control parameters are fine-tuned based on the verification results to generate the final control command. S6: Disturbance-free switching and closed-loop feedback: Through disturbance-free switching technology, the final control command is sent to the converter station real-time controller for execution, and then returns to step S1 to form a closed-loop adaptive configuration.

[0007] In a preferred embodiment, identifying the actual topology type of the current system in S1 includes: pre-storing mathematical models of electrical port characteristics of various typical topologies including symmetrical unipolar structure, true bipolar structure and parallel multi-terminal structure; extracting the spectral and time-domain features of real-time voltage and current signals through FFT and wavelet packet transform, calculating the similarity with the mathematical model, and taking the topology type corresponding to the model with the highest similarity as the identification result.

[0008] In a preferred embodiment, S2 performs an online assessment of the electrical stress of key equipment in the converter station, specifically including establishing a real-time electrothermal stress model for the converter valve group, bridge arm reactor, and DC cable; and calculating the junction temperature fluctuation of IGBT, the voltage ripple of submodule capacitors, and the fluctuation rate of DC voltage based on the real-time power flow direction and amplitude, and using these as quantitative indicators to identify system vulnerabilities and stability margins.

[0009] In a preferred embodiment, the "topology-operating condition-control mode" mapping relationship library in S3 is an expert knowledge base constructed by combining offline simulation and online learning; wherein the control mode includes at least: follow-me control, network-structured control, master-slave control and droop control; the initial control parameters include droop coefficient, virtual inertia and proportional-integral coefficient of PI regulator.

[0010] In a preferred embodiment, the multi-timescale rolling optimization in S4 specifically includes: at the millisecond timescale, using a model predictive control algorithm to perform rolling optimization on the prediction time domain and control time domain parameters of the current inner loop with the goal of minimizing the current tracking error; at the second timescale, introducing a disturbance observer to perform feedforward compensation for wind speed fluctuations and external disturbances, and adaptively adjusting the control bandwidth of the voltage outer loop based on Lyapunov stability theory.

[0011] In a preferred embodiment, the parallel hardware-in-the-loop simulation unit in step S5 is a digital twin that is physically isolated from the main control system and shares the same input data stream; the dynamic response pre-verification includes: determining whether the DC voltage overshoot, power recovery time, and oscillation damping ratio of the system are all within a preset stability domain under a preset disturbance condition.

[0012] In a preferred embodiment, the seamless switching technology in step S6 is specifically a "tracking-overlaying-switching" logic: before the switching command is issued, the backup controller continuously tracks the output of the currently running controller; during the switching, the new and old control commands are weighted and overlaid through a flexible transition function; after the overlaying process is completed, the old controller is completely disconnected to achieve seamless connection of control modes.

[0013] An adaptive configuration system for the control mode of an offshore wind power flexible DC converter station based on a typical topology, comprising: the topology sensing and identification module, used to execute step S1 as described in claim 1; The working condition analysis and evaluation module is used to perform step S2 as described in claim 1; The control mode matching module has a built-in "topology-operating condition-control mode" mapping relationship library and is used to execute step S3 as described in claim 1; The parameter rolling optimization module includes a multi-timescale optimizer for performing step S4 as described in claim 1. The parallel simulation verification module includes a hardware-in-the-loop digital twin unit for performing step S5 as described in claim 1; The non-disruptive switching execution module is used to execute step S6 as described in claim 1 and send the final instruction to the converter station valve control system.

[0014] In a preferred embodiment, the parameter rolling optimization module includes: A current inner-loop predictive controller for rolling parameter optimization on a millisecond timescale; The voltage outer loop adaptive controller includes a disturbance compensation unit based on an extended state observer, which adjusts the control bandwidth on a second-scale timescale to suppress slow disturbances.

[0015] In a preferred embodiment, when the parallel simulation verification module determines that the pre-verification result does not meet the stability domain requirements, it will issue a correction instruction to the parameter rolling optimization module to trigger local re-optimization of the control parameters until the verification is passed.

[0016] The technical effects and advantages of this invention are as follows: 1. This invention achieves precise optimization of control parameters by using topology sensing and identification, operating condition analysis and evaluation, and real-time matching of control modes. It also utilizes millisecond-level model predictive control and second-level adaptive bandwidth adjustment. In power grid fault scenarios, this method significantly suppresses DC voltage fluctuations and greatly shortens power recovery time, thereby significantly enhancing the system's fault ride-through capability and dynamic quality.

[0017] 2. This invention dynamically adjusts the control parameters of each terminal through multi-timescale collaborative optimization. In situations of sudden wind speed increases, this method effectively suppresses the voltage ripple of the submodule capacitors while ensuring the power distribution accuracy and response speed of each terminal station, effectively suppressing power fluctuations and significantly improving the overall coordination and stability of the multi-terminal flexible DC system under complex operating conditions.

[0018] 3. This invention introduces a parallel simulation pre-verification and a disturbance-free switching module. During islanded mode switching, the reliability of the strategy is pre-verified through parallel hardware-in-the-loop simulation, and then a "tracking-overlay-switching" logic is used to achieve seamless connection. The frequency deviation and voltage drop during the switching process are controlled within a very small range, ensuring a smooth transition between networked mode and islanded mode, and greatly improving the safety and power supply reliability of system operation. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology, according to the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] As attached Figure 1 The adaptive configuration method for the control mode of the offshore wind power flexible DC converter station based on a typical topology, as shown, yields the following three sets of implementation examples: Example 1: Power grid fault ride-through scenario under symmetrical unipolar topology.

[0022] This embodiment focuses on the scenario where a 400MW offshore wind farm is connected to the onshore power grid via a symmetrical monopole MMC-HVDC system, and details the implementation process of each step of the adaptive configuration method.

[0023] S1: Topology Sensing and Identification. The system collects real-time data on the three-phase voltage and current on the AC side, and the voltage and power on the DC side of the converter station. The topology sensing and identification module pre-stores mathematical models of the electrical port characteristics of three typical topologies: symmetrical unipolar, true bipolar, and parallel multi-terminal. It performs FFT and wavelet packet transform on the collected electrical quantities to extract the spectral features and transient waveform time-domain features in the 0~1000Hz frequency band. Calculations show that the current waveform has a similarity of 96.7% with the mathematical model of symmetrical unipolar, and a similarity of less than 75% with other topologies. Therefore, the current topology type is identified as a symmetrical unipolar structure.

[0024] S2: Operating Condition Analysis and Assessment. Combining the identified symmetrical unipolar topology and external environmental data (current wind speed 12m / s, output power 320MW), the operating condition analysis module extracts operating condition feature vectors, including active power, reactive power, AC voltage amplitude, and frequency. Simultaneously, real-time electrothermal stress models of the converter valve group, bridge arm reactors, and DC cables are established, calculating the electrical stress indicators of key equipment: IGBT junction temperature fluctuation is 28.6℃, submodule capacitor voltage ripple coefficient is 4.2%, and DC voltage fluctuation rate is 1.8%. The assessment results show that the converter valve group is the current system vulnerability, with a moderate stability margin (margin coefficient 0.72, threshold 0.5).

[0025] S3: Control mode matching. The control mode matching module calls the "topology-operating condition-control mode" mapping relationship library built by offline simulation and online learning. Based on the symmetrical unipolar topology, current operating condition characteristics (power level, volatility), and vulnerability information, a hybrid control mode with mesh-type control as the main mode and mesh-type control as the auxiliary mode is initially matched, and the initial control parameters are given: current inner loop PI parameters Kp=0.85, Ki=125, voltage outer loop PI parameters Kp=1.2, Ki=85, and virtual inertia H=3.0s (specified).

[0026] S4: Parameter rolling optimization, the parameter rolling optimization module initiates multi-timescale optimization: Millisecond-level scale: Using a model predictive control (MPC) algorithm, with the goal of minimizing current tracking error, rolling optimization is performed on the prediction and control time domain parameters of the inner current loop. After optimization, the prediction time domain is set to 5ms and the control time domain is set to 2ms, which speeds up the current dynamic response by 15%.

[0027] On a second-level scale: A disturbance observer is introduced to detect grid voltage fluctuations. Based on Lyapunov stability theory, the outer voltage loop control bandwidth is adaptively adjusted from the initial 15Hz to 12Hz to suppress the impact of disturbances. Simultaneously, the PI parameters are fine-tuned according to real-time power changes, ultimately yielding optimized inner current loop parameters Kp=0.82 and Ki=118, and outer voltage loop parameters Kp=1.15 and Ki=79.

[0028] S5: Parallel simulation pre-verification. A parallel hardware-in-the-loop simulation unit (a digital twin physically isolated from the main control system) receives the same input data stream and performs dynamic response pre-verification for the control strategy, including optimized parameters, in the next control cycle. A disturbance condition simulating a 30% voltage drop in the grid is set, and the verification results are judged to meet the preset stability domain (DC voltage overshoot <10%, power recovery time <200ms, oscillation damping ratio >0.3). The verification results show: DC voltage overshoot 8.2%, power recovery time 185ms, and oscillation damping ratio 0.32, all meeting the requirements. No fine-tuning is needed, and the final control command is directly generated.

[0029] S6: The seamless switching module adopts a "tracking-overlay-switching" logic: Before the switching command is issued, the standby controller continuously tracks the output of the currently operating controller; during switching, the new and old control commands are weighted and overlaid within 50ms through a linear transition function; after the overlay process is completed, the old controller is completely disconnected, achieving seamless connection of control modes. The final command is sent to the converter station valve control system for execution and returns to S1 to form a closed-loop adaptive circuit.

[0030] Experimental data charts: Table 1 compares the dynamic response indicators of the system before and after adaptive configuration under power grid faults.

[0031]

[0032] Example 2: Island mode switching scenario under true bipolar topology.

[0033] This embodiment focuses on a true bipolar topology structure (600MW offshore wind power and onshore weak grid) of a flexible DC project in the Zhoushan Islands, verifying the application of the adaptive configuration method in the switching process between grid-connected mode and island mode.

[0034] S1: Topology sensing and identification involves real-time acquisition of electrical quantities at both ends of the true bipolar converter station, including DC voltage and current of pole 1 and pole 2, as well as AC side data. A waveform similarity matching algorithm is used to extract spectral and temporal features, which are then compared with a pre-stored true bipolar mathematical model. The similarity result is 98.2%, confirming that the current topology is a true bipolar structure.

[0035] S2: Operating Condition Analysis and Assessment. The operating condition analysis module detected a planned power outage maintenance command from the onshore power grid, indicating that the system is about to switch from grid-connected mode to islanded operation. Currently, the offshore wind power output is 480MW, while the islanded load is only 320MW, resulting in a power surplus of 160MW. An electrothermal stress model is established to calculate the power reverse impact stress of the DC cable and identify system vulnerabilities. Simultaneously, the IGBT junction temperature fluctuation (32.1℃) and submodule capacitor voltage ripple (3.8%) are calculated, reducing the stability margin to 0.55.

[0036] S3: Control mode matching. The mapping relationship library matches candidate modes based on the "true bipolar topology-islanding conversion condition" with network control as the main mode and droop control as the auxiliary mode. Initial parameters: virtual inertia H=3.5s, droop coefficient kd=0.05, and PI parameters preset to Kp=1.0, Ki=150 (current inner loop) and Kp=1.5, Ki=100 (voltage outer loop).

[0037] S4: Parameter rolling optimization, multi-timescale optimization module for parameter tuning: Millisecond scale: The inner current loop predictive controller aims to minimize current tracking error, optimizing the inner current loop prediction time domain to 4ms and the control time domain to 2ms, ensuring rapid voltage stabilization in islanded mode. Second scale: The outer voltage loop adaptive controller performs feedforward compensation for wind speed fluctuations based on an extended state observer, adjusting the outer voltage loop control bandwidth from 10Hz to 8Hz according to Lyapunov theory, enhancing the ability to suppress low-frequency disturbances. After optimization, the virtual inertia H is adjusted to 3.8s, and the droop coefficient kd is adjusted to 0.048.

[0038] S5: Parallel simulation pre-verification. A parallel hardware-in-the-loop simulation unit pre-verifies the islanding switching process, assuming an extreme condition of a 50MW load surge during switching. Stability domain requirements: frequency deviation <0.5Hz, voltage drop <5%, oscillation decay time <2s. Verification results: maximum frequency deviation 0.28Hz, voltage drop 3.5%, oscillation decay time 1.2s, meeting the requirements, and generating the final control command.

[0039] S6: Seamless switching execution employs a "tracking-overlay-switching" logic. A flexible S-shaped transition function completes the weighted overlay of old and new control commands within 200ms, achieving a seamless transition from networked to islanded mode. Voltage and frequency fluctuations are minimal during switching, with no impact.

[0040] Example 3: Power fluctuation suppression scenario under parallel multi-terminal topology.

[0041] This embodiment focuses on the scenario where offshore wind power is connected to three onshore power grids of different strengths (A, B, and C) via a parallel three-terminal flexible DC system (total capacity 900MW), and verifies the performance of the adaptive configuration method under the condition of sudden wind speed increase.

[0042] S1: Topology sensing and identification. The system synchronously collects electrical quantities from three converter stations. Through a multi-terminal waveform similarity matching algorithm, the similarity between the calculated and pre-stored parallel multi-terminal topology mathematical model reaches 97.5%, accurately identifying it as a parallel multi-terminal topology.

[0043] S2: Operating Condition Analysis and Assessment. The operating condition analysis module detected a sudden increase in wind speed from 10 m / s to 14 m / s, and a jump in wind power output from 550 MW to 820 MW, with an increase rate of 45 MW / s. An electrothermal stress model was established for each converter station. Calculations revealed that the capacitor voltage ripple coefficient of the C-terminal converter station submodule connected to the weak grid reached 6.8% (threshold 5.5%), identifying it as a system vulnerability. Other indicators were normal.

[0044] S3: Control mode matching. The mapping relationship library matches candidate modes for master-slave control and droop control coordination: Terminal A (strong grid) is set as the master station using constant DC voltage control, terminals B and C are set as slave stations using constant active power control, and terminal C is supplemented with droop control to enhance the support capability of weak grids. Initial parameters: The inner loop PI of the current at each terminal is set to Kp=0.8, Ki=120, the outer loop PI of the voltage is set to Kp=1.2, Ki=90, and the droop coefficient at terminal C is kd=0.06.

[0045] S4: Parameter rolling optimization, multi-timescale optimization module implements hierarchical rolling optimization: millisecond scale: model predictive control performs collaborative optimization of the inner loop parameters of the current at each end to ensure power allocation accuracy. After optimization, the prediction time domain of the inner loop of the current at each end is set to 5ms, and the control time domain is 2ms.

[0046] On a second-level scale: Based on the extended state observer, feedforward compensation is performed for wind speed disturbances, and the outer loop control bandwidth of the C-terminal voltage is dynamically adjusted, adaptively reducing it from 12Hz to 7Hz to suppress ripple. At the same time, the C-terminal droop coefficient is finely adjusted to 0.055.

[0047] S5: Parallel simulation pre-verification. A parallel hardware-in-the-loop simulation unit pre-verifies the power surge condition, requiring each terminal's performance indicators to meet the following stability domain requirements: power distribution steady-state error <2%, DC voltage fluctuation rate <2.5%, submodule capacitor voltage ripple coefficient <5.5%, power response time <150ms, and oscillation damping ratio >0.3. The verification results are shown in Table 2, and all requirements are met. After optimization, the C-terminal ripple coefficient was reduced to 4.9%, verification passed, and the final control command was generated.

[0048] S6: Seamless switching execution. The three-terminal control parameters are updated synchronously through the "tracking-overlay-switching" technology. The switching process is smooth and shock-free, and the power and voltage waveforms at each terminal are continuous and distortion-free.

[0049] Experimental data charts: Table 2 summarizes the key performance indicators of the multi-terminal system under power surge conditions:

[0050] In summary, by employing topology sensing and identification, operating condition analysis and evaluation, and real-time matching of control modes, and utilizing millisecond-level model predictive control and second-level adaptive bandwidth adjustment, precise optimization of control parameters is achieved. In power grid fault scenarios, this method significantly suppresses DC voltage fluctuations and substantially shortens power recovery time, thereby significantly enhancing the system's fault ride-through capability and dynamic performance.

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology, characterized in that: The specific steps are as follows: S1: Real-time topology identification: Real-time acquisition of operational data of offshore wind power flexible DC system, and identification of the actual topology type of the current system based on waveform similarity matching algorithm; S2: Electrical stress and vulnerability assessment: Combining the identified topology and external environment data, extract the feature vector of the current operating condition, and conduct online assessment of the electrical stress of key equipment in the converter station to identify system vulnerabilities; S3: Preliminary control mode matching: Based on the preset "topology-operating condition-control mode" mapping relationship library, one or more candidate control modes and corresponding initial control parameters are initially matched according to the topology structure type and operating condition feature vector; S4: Multi-timescale parameter optimization: Construct a multi-timescale optimization model, and based on the candidate control modes, perform hierarchical rolling optimization on the control parameters of the current inner loop and voltage outer loop to obtain the optimized control parameters; S5: Hardware-in-the-Loop Pre-verification and Fine-tuning: Using a parallel hardware-in-the-loop simulation unit, the dynamic response of the control strategy containing the optimized control parameters is pre-verified for the next control cycle, and the optimized control parameters are fine-tuned based on the verification results to generate the final control command. S6: Disturbance-free switching and closed-loop feedback: Through disturbance-free switching technology, the final control command is sent to the converter station real-time controller for execution, and then returns to step S1 to form a closed-loop adaptive configuration.

2. The adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology as described in claim 1, characterized in that: S1 identifies the actual topology type of the current system by: pre-storing mathematical models of electrical port characteristics for various typical topologies, including symmetrical unipolar structures, true bipolar structures, and parallel multi-terminal structures; extracting the spectral and time-domain features of real-time voltage and current signals through FFT and wavelet packet transform; calculating the similarity with the mathematical models; and taking the topology type corresponding to the model with the highest similarity as the identification result.

3. The adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology as described in claim 1, characterized in that: S2 performs online assessment of the electrical stress of key equipment in the converter station, specifically including establishing real-time electrothermal stress models for converter valve groups, bridge arm reactors, and DC cables; and calculating the junction temperature fluctuation of IGBTs, submodule capacitor voltage ripple, and DC voltage fluctuation rate based on real-time power flow direction and amplitude, using these as quantitative indicators to identify system vulnerabilities and stability margins.

4. The adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology as described in claim 1, characterized in that: The "topology-operating condition-control mode" mapping relationship library in S3 is an expert knowledge base built by combining offline simulation and online learning. The control modes mentioned above include at least: mesh control, network control, master-slave control, and droop control; The initial control parameters include the droop coefficient, virtual inertia, and the proportional-integral coefficient of the PI controller.

5. The adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology as described in claim 1, characterized in that: The multi-timescale rolling optimization in S4 specifically includes: at the millisecond timescale, using a model predictive control algorithm to perform rolling optimization of the prediction time domain and control time domain parameters of the current inner loop with the goal of minimizing the current tracking error; at the second timescale, introducing a disturbance observer to perform feedforward compensation for wind speed fluctuations and external disturbances, and adaptively adjusting the control bandwidth of the voltage outer loop based on Lyapunov stability theory.

6. The adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology as described in claim 1, characterized in that: The parallel hardware-in-the-loop simulation unit in step S5 is a digital twin that is physically isolated from the main control system but shares the same input data stream. The dynamic response pre-verification includes determining whether the DC voltage overshoot, power recovery time, and oscillation damping ratio of the system are all within a preset stability range under a preset disturbance condition.

7. The adaptive configuration method for the control mode of an offshore wind power flexible DC converter station based on a typical topology as described in claim 1, characterized in that: The seamless switching technology in step S6 is specifically a "tracking-overlaying-switching" logic: before the switching command is issued, the backup controller continuously tracks the output of the currently running controller; during the switching, the old and new control commands are weighted and overlaid through a flexible transition function; after the overlaying process is completed, the old controller is completely disconnected to achieve a seamless connection of control modes.

8. An adaptive configuration system for the control mode of an offshore wind power flexible DC converter station based on a typical topology, characterized in that, include: The topology sensing and identification module is used to perform step S1 as described in claim 1; The working condition analysis and evaluation module is used to perform step S2 as described in claim 1; The control mode matching module has a built-in "topology-operating condition-control mode" mapping relationship library and is used to execute step S3 as described in claim 1; The parameter rolling optimization module includes a multi-timescale optimizer for performing step S4 as described in claim 1. The parallel simulation verification module includes a hardware-in-the-loop digital twin unit for performing step S5 as described in claim 1; The non-disruptive switching execution module is used to execute step S6 as described in claim 1 and send the final instruction to the converter station valve control system.

9. The adaptive configuration system for the control mode of an offshore wind power flexible DC converter station based on a typical topology as described in claim 8, characterized in that: The parameter rolling optimization module includes: A current inner-loop predictive controller for rolling parameter optimization on a millisecond timescale; The voltage outer loop adaptive controller includes a disturbance compensation unit based on an extended state observer, which adjusts the control bandwidth on a second-scale timescale to suppress slow disturbances.

10. An adaptive configuration system for the control mode of an offshore wind power flexible DC converter station based on a typical topology, characterized in that: When the parallel simulation verification module determines that the pre-verification result does not meet the stability domain requirements, it will issue a correction instruction to the parameter rolling optimization module, triggering a local re-optimization of the control parameters until the verification is passed.

Images