Showing papers by "Jonathan W. Kimball published in 2017"

Model-based determination of closed-loop input impedance for dual active bridge converters

Abstract: This study proposes a method of determining the input impedance of a phase shift modulated dual active bridge (DAB) converter in closed-loop operation. Input impedance is an important characterization of converter behavior, particularly in regards to interactions with external sources or systems. When converter models are available, model-based determinations of input impedance are possible through the application of the Extra Element Theorem (EET). However, DAB converters are not easily modeled using standard techniques due to their high-frequency ac stage. Instead, DAB models are derived using generalized average modeling (GAM). The GAM approach allows ac power stages to be modeled accurately but creates difficulties for model-based calculations of closed-loop impedances. This study simplifies determinations of closed-loop input impedance for DAB converters by deriving standalone expressions for the driving point impedances needed to apply the EET. These expressions allow the closed-loop input impedance to be calculated for any linear controller without the derivation of a corresponding closed-loop model. The values of input impedance calculated from these expressions are validated through comparison to experimental results from hardware tests.

An Efficient Method of Determining Operating Points of Droop-Controlled Microgrids

Abstract: Traditional methods of determining steady-state power system operating points are not applicable to droop-controlled microgrids. The conditions of an islanded microgrid, including the absence of a slack bus and inherent coupling of complex power, frequency, and bus voltage, require new tools to properly analyze. In this study, a method of determining the operating points of droop-controlled microgrids is proposed. The procedure is similar in form and function to a traditional power flow analysis, but is performed in the synchronous reference frame to ensure compatibility with conventional inverter models. A quasi-Newton iterative process is used to solve the nonlinear equations pertaining to each generation unit and load bus. The method provides a structured approach to determining the consistent system-level linearization points required for large-scale microgrid studies. Grid-forming, grid-feeding, and grid-supporting generation units are supported, along with both constant impedance and constant power loads. The method is validated in hardware experiments, simulations, and comparisons to results of existing power flow algorithms. As an example of the method's potential applications, a procedure for determining droop constants that ensure equal Q sharing between generation units is constructed around the proposed method's basic functionality.

Generalized average modeling of DC subsystem in solid state transformers

Abstract: Solid state transformers (SSTs) include dc subsystems to enable plug-and-play support of dc loads, generation sources, and energy storage. Dual active bridge (DAB) converters are a suitable topology for both the primary energy conversion and load interface applications in an SST. This study considers an SST that uses DAB converters for both applications, and describes a method of generating a full model of the dc subsystem. The system-level model is constructed from generalized average models of the individual converters and therefore retains information related to both the ac and dc stages of the DABs. The size of the resulting model is limited by preserving the decoupling of key state equations in the model combination process, thereby avoiding the scalability issues involved in generalized average modeling techniques. The accuracy of the model is verified through comparisons to results from a hardware testbed. The primary application of the model is to provide a framework for small-signal stability assessment that is applicable regardless of the flow of power in the SST. While the methods used in this study are motivated by the SST application, they are also suitable to the more general case of a dc distribution system.

Analysis of a standalone microgrid stability using generic Markov jump linear systems

Abstract: Recent developments to address the stability of small power systems, such as AC & DC microgrids include the use of advanced techniques that apply to Markov Jump Linear Systems (MJLSs) The system is considered to oscillate between discrete, finite operating modes, and can be represented as a Stochastic Hybrid System (SHS) Analytical solutions in terms of the moments of the power system dynamic and algebraic states were derived in previous studies The method is validated for small systems (2 states, 2 modes) However, its scalability to larger systems remains an open question In this paper, we apply the model to a standalone microgrid (two inverters, 36 states, 2 buses and 3 modes of operation) The resulting system statistics converge to Monte Carlo simulations Future work will extend the application of the model to arbitrary large scale systems and will derive bounds on the statistics of the dynamic states

Active stabilization of line-regulating converters with constant power loads

Abstract: A control design procedure is described for line-regulating converters in dc microgrids and distribution systems with constant power loads. The controller is stabilized using the backpropagation through time algorithm with truncated state trajectories. A method of model discretization based on 4th order Runge-Kutta numerical integration is used to convert the nonlinear converter model into a representation the meets the requirements of the training algorithm. Simulation experiments are performed to demonstrate the controller performance in a variety of ideal and nonideal conditions.