Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CAMEO:A Co-design Architecture for Multi-objective Energy System Optimization

Rounak Meyur, Tonya Martin, Sumit Purohit

TL;DR

This paper addresses the need for scalable, multi-objective co-design in energy systems by introducing CAMEO, a modular, cloud-scale architecture that enables design-space exploration across heterogeneous models and data. It combines a declarative problem language, standardized interfaces, and a Nextflow-based workflow to run multiple optimization formulations in parallel, supported by containerization for portability. The authors demonstrate two formulations on an offshore wind-farm with energy storage, illustrating how Formulation A uses random scenario sets and Formulation B uses scenario trees to capture uncertainty, with analysis of computational overhead and results. The work advances practical co-design by delivering a reusable, scalable framework that accelerates design, validation, and decision-support for complex energy systems, with future directions including broader module libraries and user interfaces.

Abstract

Co-design plays a pivotal role in energy system planning as it allows for the holistic optimization of interconnected components, fostering efficiency, resilience, and sustainability by addressing complex interdependencies and trade-offs within the system. This leads to reduced operational costs and improved financial performance through optimized system design, resource allocation, and system-wide synergies. In addition, system planners must consider multiple probable scenarios to plan for potential variations in operating conditions, uncertainties, and future demands, ensuring robust and adaptable solutions that can effectively address the needs and challenges of various systems. This research introduces Co-design Architecture for Multi-objective Energy System Optimization (CAMEO), which facilitates design space exploration of the co-design problem via a modular and automated workflow system, enhancing flexibility and accelerating the design and validation cycles. The cloud-scale automation provides a user-friendly interface and enable energy system modelers to efficiently explore diverse design alternatives. CAMEO aims to revolutionize energy system optimization by developing next-generation design assistant with improved scalability, usability, and automation, thereby enabling the development of optimized energy systems with greater ease and speed.

CAMEO:A Co-design Architecture for Multi-objective Energy System Optimization

TL;DR

This paper addresses the need for scalable, multi-objective co-design in energy systems by introducing CAMEO, a modular, cloud-scale architecture that enables design-space exploration across heterogeneous models and data. It combines a declarative problem language, standardized interfaces, and a Nextflow-based workflow to run multiple optimization formulations in parallel, supported by containerization for portability. The authors demonstrate two formulations on an offshore wind-farm with energy storage, illustrating how Formulation A uses random scenario sets and Formulation B uses scenario trees to capture uncertainty, with analysis of computational overhead and results. The work advances practical co-design by delivering a reusable, scalable framework that accelerates design, validation, and decision-support for complex energy systems, with future directions including broader module libraries and user interfaces.

Abstract

Co-design plays a pivotal role in energy system planning as it allows for the holistic optimization of interconnected components, fostering efficiency, resilience, and sustainability by addressing complex interdependencies and trade-offs within the system. This leads to reduced operational costs and improved financial performance through optimized system design, resource allocation, and system-wide synergies. In addition, system planners must consider multiple probable scenarios to plan for potential variations in operating conditions, uncertainties, and future demands, ensuring robust and adaptable solutions that can effectively address the needs and challenges of various systems. This research introduces Co-design Architecture for Multi-objective Energy System Optimization (CAMEO), which facilitates design space exploration of the co-design problem via a modular and automated workflow system, enhancing flexibility and accelerating the design and validation cycles. The cloud-scale automation provides a user-friendly interface and enable energy system modelers to efficiently explore diverse design alternatives. CAMEO aims to revolutionize energy system optimization by developing next-generation design assistant with improved scalability, usability, and automation, thereby enabling the development of optimized energy systems with greater ease and speed.
Paper Structure (12 sections, 8 figures, 1 table)

This paper contains 12 sections, 8 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Works
  3. Architecture and Methodology
  4. Characterizing approaches
  5. Minimum descriptive language
  6. Co-design workflow system
  7. Use Case Demonstration
  8. Co-design optimization problem
  9. Nextflow process definitions
  10. Results from Formulation A.
  11. Results from Formulation B.
  12. Conclusion

Figures (8)

  • Figure 1: CAMEO Architecture: Modular Design to support multiple instances of co-design workflows
  • Figure 2: CAMEO use case: optimal energy storage sizing problem needs to be executed for the various battery configurations and wind farm locations.
  • Figure 3: Summary of the Nextflow workflow for Formulation A constructed using the different processes.
  • Figure 4: Optimal battery size for 5 different wind farm locations and 2 different battery chemistry. An optimization problem (based on Formulation A) is solved for different battery duration and ratings, which alters the initial installation investment of the batteries.
  • Figure 5: Statistics of performance metrics for Formulation A showing the job duration, CPU usage, and memory allocated for the different Nextflow processes in the workflow.
  • ...and 3 more figures