Dissertation - Open Access
Doctor of Philosophy (PhD)
Electrical Engineering and Computer Science
Timothy M. Hansen
The primary aim of this dissertation is to provide synthetic residential load models with granular level information on the customers having information about the appliances that constitute each individual residential customer through time. The synthetic load model is capable of being widely utilized by the power system research community since only publicly available data is utilized for its generation. This gives researcher’s access to how the synthetic load was made and also how accurate the model is in representing real power system regions. As the title of the dissertation suggests, the synthetic residential load models are intended for smart city energy management studies. Smart city energy management studies have the ability to control tens of thousands of electricity customers in a coordinated manner to enact system-wide electric load changes. Such load changes have the potential to reduce congestion (i.e. stress on power system components) and peak demand (i.e. the need for peaking generation), among other benefits. For smart city energy management studies to have the capability of evaluating how their strategies would impact the actual power system, datasets that accurately characterize the system load are required that also contain individual loads of all buildings in a given area. Currently, such data is publicly unavailable due to privacy concerns. This dissertation’s synthetic residential load model combines a top down and bottom up approach for modeling individual residential customers and their individual electric assets, each possessing their own characteristics, using time-varying queueing models. The aggregation of all customer loads created by the queueing models represents a known city-sized load curve to be used in smart city energy management studies. The dissertation presents three queueing residential load models that make use of only publicly available data to alleviate privacy concerns. The proposed approach is mainly driven by the aggregated distribution companies load. An open-source Python tool to allow researchers to generate residential load data for their studies is also provided. The simulation results comparing the three queueing synthetic load models consider the ComEd region (utility company from Chicago, IL) to demonstrate the model’s characteristics, impact of the choice of model parameters, and scalability performance of the Python tool. The developed residential synthetic queueing load models are utilized to create the Midwest 240-Node distribution test case system, which generates appliance-level synthetic residential load for 1,120 homes for the Iowa State distribution system test case with 193 load nodes over three feeders. The Midwest 240-Node is a real distribution system from the Midwest region of the U.S. with real one-year smart meter data at the hourly aggregated node level resolution for 2017 available in an OpenDSS model. The synthetic residential queueing load model generated for the Midwest 240-Node one-year date has a mean absolute percentage error of 2.5828% in relation to the real smart meter data. The Midwest 240-Node distribution system OpenDSS model was converted to GridLAB-D to enable smart grid and transactive energy studies. The percentage of maximum error observed on voltage magnitude from the OpenDSS to GridLAB-D model is below 0.0009%. The GridLAB-D model and the generated synthetic residential load is made publicly available. The Midwest 240-Node real distribution system with the synthetic residential load that follows the real data from smart meters is intended to be a distributed energy active consumer test system network. The focus of the developed synthetic residential load models is smart city energy management studies; however, they can be utilized in many power systems studies to evaluate economic and technical impacts of distributed energy resources. For example, this dissertation also presents the utilization of the synthetic models for a PV rich low voltage network. The main component of the smart grid is demand response. Demand response, or energy management, utilizes commonly passive load in to active power system resources. Residential demand response, when aggregated, is capable of performing system-wide changes that enable its participation in the power system markets. This dissertation developed residential synthetic models to enable the standardization of approaches and allow different approaches to be compared under the same environment.
Number of Pages
South Dakota State University
In Copyright - Educational Use Permitted
Bereta dos Reis, Fernando, "Data Driven Synthetic Load Modeling for Smart City Energy Management Studies" (2020). Electronic Theses and Dissertations. 4073.
Available for download on Friday, August 20, 2021