Open-Source High-Fidelity Aggregate Composite Load Models of Emerging Load Behaviors for Large-Scale Analysis
This project is to develop a set of regional-level, scalable, open-source load models and tools, including large-scale aggregate load protection, advanced load composition data, and next-generation load model data tool (LMDT).
POC: Pavel Etingov, Yuan Liu
Background
Inaccurate, incomplete or insufficient load models can significantly skew study results, leading to conclusions that deviate from reality, faulty system designs and decision making. Poor modeling can also have severe consequences such as power outage or cascading faults and system failures. The number of events known as Fault Induced Delayed Voltage Recovery (FIDVR) has been increasing in Southern California, Tennessee, Kansas and Florida since the late 1980s and early 1990s. The existing load models still cannot provide a true representation of the dynamic load response, especially because of: a) the increasing proliferation of single-phase motor loads such as residential air conditioners (A/Cs), b) the insufficient resolution of existing protective device models for equipment and buildings. As a result, PNNL proposes to answer the need for a fundamentally different, new class of load models and next generation of load data tools.
The schematic of composite load model is depicted by Figure 1.
