Managing Renewable Energy Resources Using Equity-Market Risk Tools - the Efficient Frontiers
Authors:
Haim Grebel,
Divya Vikas,
Jim Shi
Abstract:
The energy market, and specifically the renewable sector carries volatility and risks, similar to the financial market. Here, we leverage on a well-established, return-risk approach, commonly used by equity portfolio-managers and apply it to energy resources. We visualize the relationship between the resources' costs and their risks in terms of efficient frontiers. We apply this analysis to public…
▽ More
The energy market, and specifically the renewable sector carries volatility and risks, similar to the financial market. Here, we leverage on a well-established, return-risk approach, commonly used by equity portfolio-managers and apply it to energy resources. We visualize the relationship between the resources' costs and their risks in terms of efficient frontiers. We apply this analysis to publically available data for various US regions: Central, Eastern and Western coasts. Since risk management is contingent on costs, this approach sheds useful light in assessing dynamic pricing in modern electrical grids. By integrating geographical and temporal dimensions into our research, we aim at providing more nuanced and context-specific recommendations for energy resource allocation. This approach may help decision-makers in the renewable energy sector to make informed choices that account for regional variations, climatic conditions, and long-term performance trends.
△ Less
Submitted 6 May, 2024;
originally announced May 2024.
Realizing Neural Decoder at the Edge with Ensembled BNN
Authors:
Devannagari Vikas,
Nancy Nayak,
Sheetal Kalyani
Abstract:
In this work, we propose extreme compression techniques like binarization, ternarization for Neural Decoders such as TurboAE. These methods reduce memory and computation by a factor of 64 with a performance better than the quantized (with 1-bit or 2-bits) Neural Decoders. However, because of the limited representation capability of the Binary and Ternary networks, the performance is not as good as…
▽ More
In this work, we propose extreme compression techniques like binarization, ternarization for Neural Decoders such as TurboAE. These methods reduce memory and computation by a factor of 64 with a performance better than the quantized (with 1-bit or 2-bits) Neural Decoders. However, because of the limited representation capability of the Binary and Ternary networks, the performance is not as good as the real-valued decoder. To fill this gap, we further propose to ensemble 4 such weak performers to deploy in the edge to achieve a performance similar to the real-valued network. These ensemble decoders give 16 and 64 times saving in memory and computation respectively and help to achieve performance similar to real-valued TurboAE.
△ Less
Submitted 18 June, 2021;
originally announced June 2021.
