(482c) Data-Driven Optimal Design of Combined Heat and Power for Residential Neighborhoods

Data-Driven Optimal Design of Combined Heat and Power for Residential Neighborhoods

Abigail D. Ondeck, Michael Baldea, and Thomas F. Edgar

McKetta Department of Chemical Engineering

The University of Texas at Austin, 1 University Station C0400, Austin, TX 78712

email: aondeck@utexas.edu

Combined heat and power (CHP) facilities are a very promising path to reducing CO₂ emissions and increasing efficiency in the power generation sector, especially when combined with residential solar photovoltaic (PV) power generation. CHP facilities rely on natural gas, a cleaner fuel than coal, to generate electricity, as well as hot and chilled water to provide heating and cooling. The ability to supply these essential residential utilities in an efficient way on a medium to large scale opens the path for combining district cooling, heating and power generation, and suggests that CHP plants are an appealing choice as an integrated utility supplier for the neighborhood of the future. Yet, there are currently no CHP plants that serve exclusively residential neighborhoods, and published works exploring this possibility are scarce.

In our previous work [1], we demonstrated that a CHP plant with PV integration can in principle meet the demands of a residential neighborhood during the summer months.  However, a CHP plant must be optimally sized to maximize efficiency in producing electricity, heating, and cooling, and to lower the capital and marginal costs. High generation efficiency and low costs are typically achieved by producing the minimum amount of thermal energy necessary to meet thermal loads, while not limiting the amount of electricity generated. The majority of the approaches described in the literature for optimal CHP plant sizing (e.g. load-duration curve, multi-criteria sizing) thus assume that a connection to the grid is available to act as a profitable outlet for any electricity generated in excess.

In the paper, we describe a novel strategy for the optimal sizing and integration of a CHP plant as a utility producer for a residential neighborhood operating in island (i.e., grid-disconnected) mode, and the potential for using photovoltaics and energy storage, implemented in a centralized fashion, to alleviate fluctuations in residential demand. Utilizing data collected by Pecan Street Research Inc., a non-profit smart grid demonstration project headquartered at The University of Texas at Austin,  residential heating, cooling, and electricity demand are analyzed and evaluated. These data are then used to create a time-resolved energy profile describing residential energy use, which is subsequently serves as a basis for optimizing  the size of the CHP plant so that all utility demands are met, while minimizing the capital and marginal costs. Based on the optimally sized CHP plant, we compute an optimal operating schedule while accounting for PV electricity generation and energy storage.

References:

[1] Ondeck, A.D., Baldea, M., Edgar, T.F. (2015). Data-Driven Modeling and Optimal Operation of District-level Combined Heat and Power and Photovoltaic Power Generation System. Applied Energy, submitted.