An Electric Spring Control Strategy Based on Finite Control Set-Model Predictive Control

Tao Zhang, Hao Qiang, Zheng Zheng, Chuang LuJournal Européen des Systèmes Automatisés 2020 · DOI

TL;DR

This paper introduces a novel control strategy for Electric Springs (ES) using Finite Control Set-Model Predictive Control (FCS-MPC). This approach effectively stabilizes critical load voltages in the face of renewable energy fluctuations and nonlinear loads, outperforming traditional methods by directly controlling the ES without the need for Pulse Width Modulation (PWM).

What Problem Does This Paper Address?

The increasing integration of renewable energy sources (RES) into power grids, while beneficial, introduces significant challenges related to voltage fluctuations and power quality. Electric Springs (ES) have emerged as a promising solution for mitigating these issues by actively compensating for voltage variations on the load side. However, traditional ES control methods, such as Proportional-Integral (PI) controllers, struggle to cope with the complexities of modern power systems, including the proliferation of nonlinear loads and unpredictable disturbances. These limitations lead to difficulties in achieving precise voltage regulation and require extensive parameter tuning. The paper aims to address this gap by proposing a more robust and intelligent control strategy for ES that can handle these challenges effectively.

Methods

The core of this paper is the application of Finite Control Set-Model Predictive Control (FCS-MPC) to an Electric Spring system. The FCS-MPC approach offers several advantages:

  • Model-Based Prediction: It utilizes a predictive model of the load-side circuit to forecast future system behavior. This allows for proactive control actions rather than reactive responses.
  • Finite Control Set: Unlike traditional MPC that requires a continuous control signal, FCS-MPC operates on a finite set of possible control actions (in this case, the switching states of the inverter). This simplifies the control problem and eliminates the need for Pulse Width Modulation (PWM).
  • Optimization-Based Control: The FCS-MPC algorithm aims to minimize a performance indicator function, which quantifies the error between the predicted ES output voltage and the desired reference voltage. This optimization process directly determines the optimal switching sequence.
  • Direct Control: The FCS-MPC directly controls the switching states of the inverter bridge, bypassing the PWM generation stage. This simplifies the control architecture and can improve response times.
  • Constraint Handling: FCS-MPC inherently handles system constraints, such as voltage and current limits, which is crucial for robust operation.

The authors first establish a load-side circuit prediction model and then design the FCS-MPC controller. The control strategy involves generating a reference voltage for critical loads (CLs), predicting the ES output voltage, and then selecting the optimal switching state to minimize the prediction error.

Datasets

Not discussed in the paper. The evaluation relies on simulations and physical experiments using a dSPACE platform.

Key Insights and Ideas

  • FCS-MPC for ES is a powerful combination: Applying FCS-MPC to ES offers a significant improvement in voltage regulation capabilities, especially in dynamic and uncertain environments.
  • Eliminating PWM is a key benefit: By directly controlling the switching states, the FCS-MPC approach simplifies the ES control system and avoids the complexities and potential limitations associated with PWM.
  • Robustness to disturbances: The FCS-MPC strategy demonstrates strong performance in maintaining stable critical load voltages even when faced with significant supply-side voltage fluctuations and nonlinear loads.
  • Directly addressing nonlinearities: The predictive nature of MPC allows it to inherently handle nonlinearities in the system, which is a major advantage over linear controllers.
  • Feasible for real-time implementation: The paper highlights that the computational demands of the proposed FCS-MPC are manageable for real-time implementation on platforms like dSPACE, making it practical for actual deployment.

Key Results

The paper presents simulation and experimental results demonstrating the effectiveness of the proposed FCS-MPC control strategy for Electric Springs.

  • Voltage Stabilization: The ES, controlled by FCS-MPC, effectively stabilized critical load voltages (Ucl) at approximately 220V and 30V, respectively, during both voltage drops and rises in the supply-side voltage. In contrast, systems without the ES experienced significant voltage fluctuations.
  • Fast Response: The system exhibited a quick response to voltage disturbances, with Ucl stabilizing rapidly after initial minor fluctuations.
  • Robustness: The FCS-MPC controlled ES maintained stable Ucl even with a 10% variation in supply-side voltage.
  • No Modulation Link: The FCS-MPC directly generates control signals without requiring a PWM modulation link, simplifying the system.
  • Simulation and Experimental Validation: The proposed strategy was validated through both MATLAB/Simulink simulations and physical experiments using a dSPACE platform, showing consistent and positive results.

Limitations and Future Work

  • Computational Load: While the paper states the computational load is manageable, the complexity of MPC can still be a limiting factor for very high switching frequencies or extremely complex systems. The paper mentions that the calculation time of their system fully meets the requirements of the dSPACE platform, implying it's not a bottleneck for their specific setup.
  • Model Accuracy: The performance of MPC is dependent on the accuracy of the system model. Any inaccuracies in the load-side circuit prediction model could impact the control performance.
  • Future Work: The paper suggests that the FCS-MPC approach provides a good reference for the engineering application of ES. Further research could explore its application in more complex multi-phase systems or investigate adaptive modeling techniques to further enhance robustness against unmodeled dynamics. The paper also implicitly suggests that further optimization of the performance indicator function could be explored.

Summary updated

Abstract

As a novel voltage control device, electric spring (ES) can effectively suppress the voltage fluctuations across critical loads (CLs), and solve the various problems with electrical quality induced by the grid access of renewable energy resources (RES). However, the traditional controllers for the ES system can no longer meet the control requirements, as the environment is complicated by the growing number of load-side nonlinear loads and uncertain disturbances. To solve the problem, this paper proposes a control system based on finite control set-model predictive control (FCS-MPC), and applies it to the ES. Firstly, a load-side circuit prediction model was established and analyzed. Next, a control system was designed based on FCS-MPC. Finally, the proposed system was proved feasible and effective through MATLAB/Simulink simulation and dSPACE physical experiment. The results show that the proposed FCS-MPC system can directly control the ES, easily handle system constraints, achieve robust dynamic and static performance, eliminating the need for pulse width modulation (PWM).
BibTeX
@article{zhang2020electric,
  author = {Tao Zhang and Hao Qiang and Zheng Zheng and Chuang Lu},
  title = {{An Electric Spring Control Strategy Based on Finite Control Set-Model Predictive Control}},
  journal = {Journal Européen des Systèmes Automatisés},
  year = {2020},
  doi = {10.18280/jesa.530403},
}
Figure