Extended state observer-based predictive control for soft open point

Zhengqi Wang, Peng Lu, Haoyu Zhou, Li ChenFrontiers in Energy Research 2023 · DOI

Abstract

This paper proposes an extended state observer-based ultra-local model-free three-vector predictive control method for Soft Open Point (SOP). First, the Ultra-Local Model-Free Predictive Control (ULMFPC) method is proposed to improve the robustness of the system, which only uses the input and output of the outer-loop, and any other parameters are not involved. Second, considering parameter perturbations and external disturbances in the SOP system, an expansion state observer (ESO) is established to observe the SOP system’s total perturbations and the perturbations are compensated in real-time to improve the system. Third, to solve the problem of significant current harmonics in traditional model predictive control (MPC), a three-vector MPC method (TV-MPC) is adopted to reduce the total harmonic distortion rate (THD) of the current. Finally, it is verified by simulation that the proposed method can effectively reduce the current harmonics of the SOP system, rate value setting time, and improve the dynamic performance effectively. When perturbations occur in the system, the proposed method can improve the anti-interference and robustness of the system.

The Challenge of Modern Distribution Networks

As renewable energy sources like solar and wind flood the distribution network, power imbalances and voltage instability become major headaches for grid operators. Soft Open Points (SOPs)—essentially power electronic devices that replace traditional mechanical switches—are the solution. They allow for flexible, fast regulation of power flow between different parts of the grid.

Model of a two-port Soft Open Point (SOP) system.

However, SOPs are complex, non-linear systems. Traditional control methods like Proportional-Integral (PI) controllers or standard Model Predictive Control (MPC) are "model-dependent." This means they rely on precise knowledge of system parameters like inductance ($L$) and capacitance ($C$). In the real world, these values drift as components age or heat up. Furthermore, standard MPC often produces significant current harmonics (electrical "noise"), which degrades power quality.

A Multi-Layered Control Architecture

The authors propose a multi-layered control architecture to decouple the system from its physical parameters and clean up its output. This approach focuses on the individual Voltage Source Converters (VSCs) that make up the SOP.

Detailed model of the Voltage Source Converter (VSC1).

The control strategy consists of three main components:

  1. Ultra-Local Model-Free Predictive Control (ULMFPC): Used for the outer voltage loop, this method treats the system as a "black box." It only looks at input and output data, ignoring specific physical parameters. This makes the controller inherently robust against component aging.
  2. Extended State Observer (ESO): This acts as a real-time "watchdog." It observes the difference between the predicted and actual state of the system, bundling all unknown disturbances and parameter mismatches into a single "total perturbation" term. It then compensates for these errors in real-time.
  3. Three-Vector MPC (TV-MPC): Traditional MPC picks one voltage vector per switching period, which is a blunt instrument that causes "jitter" in the current. TV-MPC combines three vectors (one zero vector and two non-zero vectors) in a single period. This finer-grained control significantly reduces current ripples.

Performance Evaluation

The study validated this approach using a high-fidelity MATLAB/Simulink simulation of a two-port SOP system. The system parameters used for the simulation (e.g., 650V DC-side voltage, 5,000 $\mu$F capacitance) were chosen to reflect realistic operating conditions.

The proposed ULMFPC + ESO + TV-MPC method was compared against traditional PI and standard MPC setups with the following results:

  • Power Quality: The Total Harmonic Distortion (THD) of the current was slashed from 0.54% (traditional MPC) to just 0.09% (proposed TV-MPC) on the inverter side.
  • Dynamic Response: When tracking the rated DC voltage, the proposed method reached the target in 0.029 seconds, nearly twice as fast as the traditional PI method (0.057 seconds).
  • Robustness: During a simulated load jump (dropping from 100A to 50A), the proposed controller showed significantly less "jitter" and faster recovery than traditional PI controllers, which suffered from large oscillations.

Conclusion and Future Directions

By shifting from a complex physical model to a simplified "ultra-local" mathematical model, the authors demonstrated that precise parameter knowledge is not a prerequisite for high-performance control. The integration of an ESO allows the system to handle "one-beat delays" and external disturbances effectively.

While the simulation results are compelling, future research will focus on hardware validation using physical SOP prototypes to account for sensor noise and hardware-level latencies. Additionally, scaling this model-free approach to multi-port systems and integrating energy storage systems remain key areas for exploration.

Summary updated

Abstract

This paper proposes an extended state observer-based ultra-local model-free three-vector predictive control method for Soft Open Point (SOP). First, the Ultra-Local Model-Free Predictive Control (ULMFPC) method is proposed to improve the robustness of the system, which only uses the input and output of the outer-loop, and any other parameters are not involved. Second, considering parameter perturbations and external disturbances in the SOP system, an expansion state observer (ESO) is established to observe the SOP system’s total perturbations and the perturbations are compensated in real-time to improve the system. Third, to solve the problem of significant current harmonics in traditional model predictive control (MPC), a three-vector MPC method (TV-MPC) is adopted to reduce the total harmonic distortion rate (THD) of the current. Finally, it is verified by simulation that the proposed method can effectively reduce the current harmonics of the SOP system, rate value setting time, and improve the dynamic performance effectively. When perturbations occur in the system, the proposed method can improve the anti-interference and robustness of the system.
BibTeX
@inproceedings{wang2023extended,
  author = {Zhengqi Wang and Peng Lu and Haoyu Zhou and Li Chen},
  title = {{Extended state observer-based predictive control for soft open point}},
  booktitle = {Frontiers in Energy Research},
  year = {2023},
  doi = {10.3389/fenrg.2023.1089258},
}
Figure