Visible to the public Synthesizing Energy Consumption Data Using a Mixture Density Network Integrated with Long Short Term Memory

TitleSynthesizing Energy Consumption Data Using a Mixture Density Network Integrated with Long Short Term Memory
Publication TypeConference Paper
Year of Publication2019
AuthorsSarochar, J., Acharya, I., Riggs, H., Sundararajan, A., Wei, L., Olowu, T., Sarwat, A. I.
Conference Name2019 IEEE Green Technologies Conference(GreenTech)
Keywordscentralized data analytics, Collaboration, commercial consumer energy consumption, Communication networks, composability, Data analysis, data dissemination, Data models, data privacy, data storage, data synthesis, energy consumption, fully synthetic energy consumption data, Human Behavior, human behavioral impacts, information dissemination, load patterns, Logic gates, long short-term memory network, LSTM, MDN, Metrics, mixture density network, multiple critical infrastructures, policy-based governance, power engineering computing, power grid, Predictive models, privacy, privacy concerns, Probability density function, pubcrawl, residential consumer energy consumption, residential smart meters, resilience, Resiliency, Scalability, security, smart cities, smart grid consumer privacy, smart meters, smart power grids, Standards
AbstractSmart cities comprise multiple critical infrastructures, two of which are the power grid and communication networks, backed by centralized data analytics and storage. To effectively model the interdependencies between these infrastructures and enable a greater understanding of how communities respond to and impact them, large amounts of varied, real-world data on residential and commercial consumer energy consumption, load patterns, and associated human behavioral impacts are required. The dissemination of such data to the research communities is, however, largely restricted because of security and privacy concerns. This paper creates an opportunity for the development and dissemination of synthetic energy consumption data which is inherently anonymous but holds similarities to the properties of real data. This paper explores a framework using mixture density network (MDN) model integrated with a multi-layered Long Short-Term Memory (LSTM) network which shows promise in this area of research. The model is trained using an initial sample recorded from residential smart meters in the state of Florida, and is used to generate fully synthetic energy consumption data. The synthesized data will be made publicly available for interested users.
Citation Keysarochar_synthesizing_2019