Impedance modelling and stability analysis of power grid-connected inverter-based resources

Dai, Shuailong (2024). Impedance modelling and stability analysis of power grid-connected inverter-based resources. University of Birmingham. Ph.D.

Preview

Dai2024PhD .pdf
Text
Available under License All rights reserved.
Download (8MB) | Preview

Abstract

Modern power systems are transitioning from synchronous generators dominated power systems to inverter-based resources (IBRs) dominated power systems. This change leads to low inertia and low fault current power systems, which tends to cause power system oscillations. Traditional modelling methods have the difficulty to capture the dynamic response of complex control systems. In addition, detailed white box models of inverters are generally unavailable due to commercial intellectual property and other commercial reasons. Therefore, impedance-based stability analysis has various applications for such black-box models.
This thesis aims at the development of methods and tools for the enhancement of the stability of power systems with a high proportion of renewable energy. This is to be achieved by development of a number of methods and tools, which include impedance modelling methods using measurements, impedance-based stability analysis method and tool with fast preliminary screening and detailed impedance-based stability theory, novel auxiliary phase-locked loop (PLL) control method to enhance to enlarge the stable boundary, stability enhancement by impedance tuning method using perturbation and observation, a data-driven automation method for clustering and aggregation of multiple operating point impedance models, and a data-driven automation method using a Bayesian optimisation method to determine stability boundary.
The two impedance-based modelling methods are implemented to demonstrate the characteristics of the interconnected power system. The impedance modelling method derived from state space equations is presented and the impedance modelling method based on measurements is also presented. Then they are compared, which indicates that the impedance model based on measurements can match the impedance model derived from state-space equations very well. Furthermore, several important concerns for practical applications are addressed and appropriate guidance is provided to obtain a highly accurate model via case study.
Based on the impedance model of the interconnected system, the impedance-based stability analysis is widely adopted to understand the system oscillation phenomenon normally caused by control interaction. A two-step screening method including multi-infeed impact factor (MIIF) and short circuit ratio (SCR) index are proposed in this thesis to identify the potential threats. If a risk is identified, then secondly a full-scale analysis is automated using an automation tool as a scripting tool. Hence, the systematic method, including preliminary screening and impedance-based stability theory, is proposed to trade off effort and accuracy.
Since the stability analysis is presented, the further actions on stability enhancement could be implemented. The information obtained from the state-space model is validated with simulations, revealing how the control parameters affect the impedance transfer function and eigenvalues. A novel auxiliary phase-locked loop (PLL) control method is proposed to enlarge the stable boundary of the operating points by impedance reshaping methods.
For the black box model, it would be a bit challenge to enhance the stability for inverter cause the inner control topology and parameters are unknown to the engineer. Therefore, the impedance tuning method using perturbation and observation is proposed for black-box model to enhance the stability. The changes in eigenvalue explain the effect of variable disturbance on stability. When the real part of eigenvalues decreases, the changes on parameters should be retained to enhance the stability.
The measurement-based impedance model is applicable for specific operating point and it could be different under multiple operating points. Giving tens of thousands of models for a power grid with multiple operating points and making these models available to equipment manufacturers and wind farmer developers/operators would be impractical. Such large amounts of data sets should be clustered and aggregated into a limited number of impedance models. Hence a data-driven automation method for clustering and aggregation of multiple operating point impedance models is proposed. A clustering method based on discrete Fréchet distance (DFD) is proposed to allocate impedance data to different groups. An aggregation method with data shift is proposed to reduce the data size and generate the aggregated transfer function. A data-driven toolbox for stability analysis is developed to make above process be automated and improve system efficiency.
The method of traversing all operating points to determine the system stability boundary is time-consuming, so a fast method for optimizing the stability operating range is needed. Data-driven automation method using a Bayesian optimisation method to determine stability boundary is proposed. Compared with grid search method, this data-driven method selects the next operating point to test based on the experience of previous operating points. The proposed Bayesian optimization method reduce the required number of frequency scan to find stable operating boundaries, especially with a large number of operating points. The user-friendly interface of the toolbox has been developed for engineers to use in a broader range of applications.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):