Search NASA⌕ Search
Home/Archive/DOE OSTI

DOE OSTI · 2373075

A deep learning-based battery sizing optimization tool for hybridizing generation plants

Lin, Yingqian·Li, Binghui·Singh, Vivek Kumar·Mosier, Thomas M.·Kim, Sangwook·Tanim, Tanvir R.·Griffel, L. Michael·Alam, S.M. Shafiul·Balliet, Hill·Mahalik, Matthew R.·Kwon, Jonghwan
View original source ↗Read source document ↗Export metadata ↓

Abstract

Hybrid generation and energy storage systems offer the ability to increase flexibility of the combined asset. This flexibility can be used to increase provision of services already provided by the generation asset, such as timing sale of electricity to the energy market during high price periods, and also enable provision of additional services, such as ancillary services or contribute to resource adequacy. From a generation asset owner perspective, the decision to hybridize includes selecting an energy storage system that, among other factors, maximizes financial performance of the energy storage investment. Yet, existing tools to optimize energy storage sizing are either too rudimentary (i.e., based on “rules of thumb”) or too complex to implement (i.e., require specialized engineering and software knowledge and a high-performance computer to run). This work presents a novel deep learning-based battery sizing optimization tool that is designed to help generation asset owners easily assess preliminary sizing considerations for potential battery investments to hybridize their generation facility. The tool uses deep learning to predict revenue over a broad search space of potential battery sizes, estimates capital and operating costs (including accounting for battery degradation), and computes financial performance of each potential battery system investment, recommending a system with maximum financial performance. The tool is tested and validated for hydropower assets. Finally, this tool will help a greater cross-section of industry consider investments in battery systems, increasing their revenue and helping them compete in rapidly evolving electrify markets.

Explore related subjects

13 HYDRO ENERGY25 ENERGY STORAGEBattery sizing optimizationDeep learningEnergy StorageFinancial performance optimizationHydropowerRenewable energy

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Lin, Yingqian, Li, Binghui, Singh, Vivek Kumar, Mosier, Thomas M., Kim, Sangwook, Tanim, Tanvir R., Griffel, L. Michael, Alam, S.M. Shafiul, Balliet, Hill, Mahalik, Matthew R., Kwon, Jonghwan. 2024-01-03. A deep learning-based battery sizing optimization tool for hybridizing generation plants. https://doi.org/10.1016/j.renene.2023.119911

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

DOE OSTI2026

Validation of the NLR Pumped Storage Hydropower Cost Model

The National Laboratory of the Rockies (NLR) first released its pumped storage hydropower (PSH) cost model in 2023 as the most detailed bottom-up PSH cost model available to the public. It is available both as a spreadsheet and an interactive web tool, enabling users with a variety of PSH interests to transparently characterize costs of alternative PSH sites and designs. The PSH cost model cannot replace detailed site-level studies and design, but it is important to validate it against other industry PSH cost estimates. The initial model methodology report validated the cost model for a single proposed site, the Eagle Mountain Project in California. This slide deck documents an expanded validation exercise using cost data from six other sites: Goldendale (Washington), Seminoe (Wyoming), Gordon Butte (Montana), Swan Lake (Oregon), White Pine (Oregon), and Lewis Ridge (Kentucky). It compares itemized costs from Federal Energy Regulatory Commission (FERC) applications and other reported costs with NLR PSH cost model outputs after customizing inputs for each site. The validation exercise finds that the NLR model's conservative indirect cost assumptions often drive overall cost overestimation, with direct cost comparisons typically agreeing more closely. All cost model estimates are well within an Association for the Advancement of Cost Engineering (AACE) Class 5 estimation range (-50% to +100%), with five within the AACE Class 4 range (-30% to +50%) and four being within 15%. This result is considered reasonable performance for a parametric model applied at a preliminary design stage.

13 HYDRO ENERGY↗
DOE OSTI2026

Pumped Storage Hydropower Operation & Maintenance Cost Estimation

The National Laboratory of the Rockies (NLR) develops and hosts a pumped storage hydropower (PSH) cost model that is the most detailed bottom-up PSH cost model available to the public. It is available both as a spreadsheet and an interactive web tool, enabling users with a variety of PSH interests to transparently characterize costs of alternative PSH sites and designs. The NLR PSH cost model was designed originally to consider only upfront capital costs only. This slide deck describes methodology to expand the cost model to include operations and maintenance (OM) costs. OM costs are characterized as five distinct components with unique sources and methods for cost estimation. By combining methods for each of these components into a cumulative OM cost estimate, these methods allow a more complete estimation of total OM costs that agrees with existing literature values. The methods are scalable and transparent, allowing them to be readily to applied to any prospective PSH facility for a representative preliminary OM cost estimate in advance of detailed site-specific engineering and other studies.

13 HYDRO ENERGY↗
DOE OSTI2026

Staying Current: A Community Readiness Framework for Marine Energy Applied to River Current Energy in Alaska

Marine energy-including wave, tidal, and river current energy-can provide a local energy source for rural and remote communities. Marine energy has the potential to bolster self-sufficiency and create economic opportunities while preserving ecological integrity. Communities may be interested in deploying, testing, and advancing these early-stage technologies to meet their needs. However, limited capacity, workforce constraints, and other barriers can challenge development. To better understand a community's interest in and preparedness for marine energy, we developed a suite of 150 'metrics of readiness.' Organized across seven categories-technical, social, environmental, strategic, governance, economic, financial-and 29 subcategories, the metrics provide a holistic perspective beyond the technical aspects of an energy device. We conducted a desktop application of the metrics of readiness for Igiugig, Alaska. The metrics were applied retrospectively for two points in time: before (2009) and after (2018) in-stream testing of a river current energy device. By documenting changes among categories and subcategories of the metrics, our results show the evolving nature of community readiness for river current energy. They also illustrate how our interdisciplinary framework captures the investment in environmental effects research and commitment to strategic planning that occurred in Igiugig. In future applications, we envision the framework could be used to foster public engagement in marine energy, collaborate with communities in project development, shape capacity building activities, prioritize investments, and inform research needs. While our study focuses on enabling river current energy in Alaska, the metrics of readiness have the potential to inform implementation of other renewable technologies with communities in new geographies.

13 HYDRO ENERGY↗