Reactive Power Compensation Strategy of the Electric Vehicle Connected to the Distribution Network in the Limit State Considering Voltage Constraint

Zhiyuan Lin, Fei Tang, Caiyang Yu, Haibo Li, Lei Zhong, Xinyu Wang, Huipeng DengSustainability 2023 · DOI

TL;DR

This paper proposes a novel strategy to manage voltage stability in distribution networks with increasing electric vehicle (EV) charging loads. By combining a holomorphic embedding method for voltage analysis with electrical distance-based zoning and spectral clustering, the approach identifies critical buses for reactive power compensation, significantly improving voltage stability compared to existing methods.

What Problem Does This Paper Address?

The widespread adoption of electric vehicles (EVs) presents a significant challenge to the stability of existing power distribution networks. The high and often uncoordinated charging demands of EVs can lead to voltage fluctuations and even voltage collapse, especially when the network is already operating near its limits. Traditional methods for reactive power compensation, which often rely on continuous power flow and may not consider the entire system's voltage profile, struggle to effectively address these dynamic and localized voltage issues. This paper aims to bridge this gap by developing a more sophisticated approach to identify and mitigate voltage stability problems caused by EV charging.

Methods

The paper introduces a multi-faceted approach:

  • EV Load Modeling: A model is developed to estimate the charging power of EVs at a given bus, considering factors like battery characteristics and charging modes.
  • Holomorphic Embedding Method (HEM) for Voltage Stability Analysis: This core technique embeds a "load margin index" (s-factor) into the power flow equations. By analyzing the rational function representing voltage and using Padé approximation, the method can predict voltage collapse points by examining the distribution of zeros and poles. This avoids iterative power flow calculations and offers a more direct stability assessment.
  • Electrical Distance and Spectral Clustering for Zoning: To manage the complexity of the power system, the network is divided into voltage control zones. Electrical distance, derived from the system's impedance matrix, quantifies the coupling between buses. Spectral clustering and k-means algorithms are then used to partition the network into regions based on this electrical distance, aiming for weak inter-regional coupling and strong intra-regional coupling.
  • Reactive Power Compensation Strategy: Within each zone, the bus with the weakest voltage stability margin (identified via HEM) is targeted for reactive power compensation. This localized approach aims for efficient resource allocation.

Datasets

The study utilizes the IEEE 33-bus system for simulation and validation. This is a standard benchmark system commonly used in power system analysis, available within the MATPOWER software package.

Key Insights and Ideas

  • HEM as a powerful tool for direct voltage stability assessment: The paper demonstrates that HEM, combined with Padé approximation, can efficiently predict voltage stability limits without iterative power flow solutions, offering a significant advantage in speed and visualization.
  • Electrical distance as a superior metric for network partitioning: Using electrical impedance-based distance provides a more realistic representation of power system coupling than traditional network topology metrics, leading to more effective voltage control zoning.
  • Targeted compensation is more effective: Instead of broad compensation, focusing reactive power support on the most vulnerable buses within specific zones yields superior voltage stability improvements.
  • EV charging is a critical factor for future grid stability: The paper underscores the growing importance of addressing EV charging impacts on distribution network voltage stability.

Key Results

The proposed reactive power compensation strategy, which targets the bus with the weakest voltage stability margin within each electrically-defined zone, showed significant improvements:

  • Compared to an uncompensated scheme, the proposed method improved the voltage stability margin by 1.626121813 times (Scheme 3 vs. Scheme 1 in the paper's comparison).
  • Compared to a coupling-degree-based compensation scheme, the proposed method improved the voltage stability margin by 1.160494345 times (Scheme 3 vs. Scheme 2 in the paper's comparison).
  • The method successfully partitioned the IEEE 33-bus system into four voltage control zones using spectral clustering and electrical distance.

Limitations and Future Work

  • Load Model Simplification: The paper acknowledges that the EV load model could be further improved to account for more complex charging behaviors and scenarios.
  • Network Scope: The current study focuses on distribution networks. Future work could extend this methodology to transmission networks.
  • Integrated Optimization: The authors suggest exploring the optimization of integrated reactive power compensation across both transmission and distribution networks.
  • More Complex Scenarios: Future research could consider more complex grid conditions, such as dynamic load fluctuations and various EV penetration levels.

Summary updated

Abstract

To solve the voltage stability problem of electric vehicles connected to the distribution network in the limit state, a reactive power compensation strategy based on the holomorphic embedding method and electrical distance is proposed. Firstly, the load model of the electric vehicle charging station is constructed, and the limit of the charging power of the electric vehicle connected to a certain bus is obtained. Then, the power flow embedding equation of the power system is constructed by the holomorphic embedding method, and the analytical expression of the voltage rational function is introduced based on the Padé approximation algorithm. The voltage collapse point is solved by the distribution of zeros and poles of the rational function. Then, a method of reactive power and voltage control partition based on electrical distance is proposed. According to the principle of weak regional coupling and strong interval coupling, the power system is divided into several regions by spectral clustering and a k-means clustering algorithm. The order of the voltage stability margin value s is obtained by connecting the limit charging power to each bus of the power system. In this paper, the reactive power compensation strategy proposes to add reactive power compensation devices to the buses with the weakest voltage stability margin in different zones. Finally, compared with other reactive power compensation strategies 1 and 2, the reactive power compensation strategy provided in this paper is increased by 1.626121813 and 1.160494345 times, respectively. The superiority of this method is verified by simulation.
BibTeX
@inproceedings{lin2023reactive,
  author = {Zhiyuan Lin and Fei Tang and Caiyang Yu and Haibo Li and Lei Zhong and Xinyu Wang and Huipeng Deng},
  title = {{Reactive Power Compensation Strategy of the Electric Vehicle Connected to the Distribution Network in the Limit State Considering Voltage Constraint}},
  booktitle = {Sustainability},
  year = {2023},
  doi = {10.3390/su15118634},
}
Figure