Date of Award

Summer 2022

Document Type

Open Access Dissertation


Electrical Engineering

First Advisor

Herbert L. Ginn III


The concept of power electronics building blocks (PEBB) has driven advancements in highly modularized converter systems with many identical subsystems. PEBBs are distributed subsets of converter systems and thus require communication with a control system for their coordination. For this type of system, the communication latency with hard deterministic deadlines is the driving attribute of communication system requirements. However, inherent communication requirements for PEBB-based converter systems also provide opportunities for coordination of energy flow.

Leveraging developments in Gigabit serial communication channels, a control and communication platform architecture for distributed control schemes based on the 2D-Torus communication network topology was developed for building-block-based power converter systems. The control platform architecture allows integrated control actions between PEBB control nodes to support energy coordinating operations. In addition to the platform architecture, a control method was developed that takes advantage of its communication speed.

Predictive control methods utilize an internal model of the system to provide very fast regulation, which is suitable in this work since each PEBB is well defined. A distributed control architecture utilizing a predictive control scheme was developed to maintain fast regulation of voltage and current in a distributed manner within a time frame that can take advantage of the low latency provided by the 2-D Torus communication network.

Distributed control requires state information from other control nodes for overall coordination. A strategy to minimize data communication was developed to scale the distributed predictive control across the most extensive PEBB-based system. Direct communication of low-level sensor measurements and control data between PEBBs in the network would increase data communication, resulting in data bottlenecks as system sizes are scaled up. Thus, a partitioning method was developed to reduce data transmission as much as possible by developing a real-time model-informed framework requiring only partial or limited knowledge of the system parameters. The proposed design relies on a multi-loop predictive controller that uses an observer's estimated current and voltage states as the feedback values. The observer is based on a real-time model distributed utilizing a co-simulation method to partition the model such that each PEBB has a minimum sub-set consisting of that PEBB’s circuit elements.

The PEBB control platform architecture and distributed predictive control framework developed in this dissertation allows distributed PEBB control nodes to rapidly coordinate and respond to multiple energy flow requirements with data exchanges as the core.