Solid State Transformer with Predictive Digital Twin

The modern electric power grid is facing increasing stress due to a fundamental shift in supply and demand technologies. These increased stresses require the grid modernize and shift from analog systems to systems with increased data streams and digital controls. One such technology that can be utilized to accomplish these goals is the solid-state transformer (SST). The SST provides isolation between low and high-voltage connections by electronics-based power conversion, with a high frequency transformer. In addition to providing the same functionality as a legacy transformer, the SST is able to control the active and reactive power automatically providing reactive power compensation, power quality improvements, and current and voltage regulation at a reduced size, weight, and power losses. However, one concern that may arise with using SST over legacy systems is increased reliability challenges in which factors such as use case and operational conditions can lead to unpredictable behavior. Cornerstone Research Group Inc. (CRG) proposes to design, develop, and demonstrate a predictive digital twin for its SST hardware. Around the SST hardware architecture and simulation models, CRG will integrate the dynamic data-driven framework developed by the team to provide real-time prognostic health monitoring and dynamic decision-making suggestions to the electrical grid controller and maintenance crew. In Phase I, CRG will integrate its SST technology with an outer-loop, data-driven methodology for prognostic health monitoring with real-time onboard decision making to develop a predictive digital twin of the system. The team will leverage its predictive digital twin framework: we will build a physics-based offline dataset using SST simulation models and subsequently use standard reduced-order modeling techniques to build computationally efficient online models that, when integrated with a machine learning framework, will enable dynamic, real-time prognostic health monitoring. To limit the risk to the program CRG will integrate limited degraded state models into the digital twin model and demonstrate the model via a simulation environment. CRG will initially introduce degraded SiC MOSFET models into the system-level model. MOSFET failures represent 21% of total electronic converter failure modes and provide a plethora of available physic-based models for both the healthy state and degraded health states. CRG’s goal is to leverage this SBIR program, and previous development on CRG’s existing 72kW SST hardware to develop a digital twin with a predictive health monitoring framework. This approach will meet the near-term needs of improving the stability of U.S. electrical grid, while providing a developmental platform to generate the algorithms that reduce the maintenance and replacement costs