Improving Energy System Performance with Near-Optimal Solutions

Optimizing Energy Networks for a Sustainable Future My recent advancement in energy systems modeling—a high-performance Energy Network Optimization Model, built in #Julia using #JuMP and #HiGHS. This model integrates fossil generation, renewable sources, and battery storage to provide cost-effective, environmentally compliant, and highly reliable energy dispatch strategies. Key Highlights: High-Performance Optimization with Julia & JuMP: - Implemented using JuMP, a powerful algebraic modeling language for optimization. - Solved using HiGHS, an industry-leading solver known for its speed and efficiency in handling large-scale linear programming problems. - Julia’s computational speed and efficient memory handling make this model scalable for real-time market applications. Cost Minimization & Operational Efficiency: - The objective function minimizes total operational costs, balancing generation, start-up, and battery operation expenses for optimal market performance. Renewable Energy Integration & Curtailment Management: - The model maximizes clean energy penetration while effectively managing renewable curtailment to mitigate intermittency. Advanced Battery Storage Dynamics: - Explicit constraints model charging, discharging, and storage efficiency losses, enhancing grid flexibility. Emission Compliance: - Enforces emission cap constraints, ensuring regulatory compliance and supporting sustainability targets. Reliability Through Operational Constraints: - Incorporates demand balance, unit commitment, ramp rate limits, and spinning reserve requirements to maintain grid stability and resilience against unexpected demand fluctuations. Market Advantages: The model leverages mixed integer programming (MIP) for global optimality, ensuring transparent, scalable, and real-time deployable decision-making. Julia + JuMP dramatically improves computational efficiency, making it ideal for real-world energy markets, utility operators, and policymakers seeking cost savings and carbon reductions. Full project access, including source code, CI/CD pipelines, and detailed documentation, is available on my GitHub upon request: https://lnkd.in/eDC7VVHS Looking forward to engaging with industry experts on how this model can be adapted, extended, and applied in real-world energy systems. Let’s push the boundaries of smart, sustainable energy optimization! #EnergyOptimization #JuliaLang #JuMP #CleanEnergy #Sustainability #LinearProgramming #EnergyMarkets #SmartGrid #Innovation

Various advanced #control sequences for #chiller plants with water-side economizers (WSE) have been proposed in literature, with limited evaluation and #optimization. Maximizing #energy savings by selecting different sequences and parameters based on plant configuration, load, and climate is crucial. Our new paper addresses this gap by developing near-optimal advanced control sequences for chiller plants with WSEs. This work is the final product of our ASHRAE #Research Project 1661, sponsored by ASHRAE TC 4.7, 1.4, and 7.5. In this study, we categorized advanced control sequences into condenser water, chilled water, and hybrid controls. Representative sequences from each category were identified and the plant was optimized for 504 scenarios using various combinations with dynamic #Modelica models. The recommended near-optimal sequences showcased energy consumption reductions of up to 15% relative to the baseline, depending on configuration, load profile, and climate. I extend my gratitude to my current and former lab members (Cary Faulkner, Julia Ho, CHENGNAN SHI, Chengliang Fan, Nasim Ildiri) for their contributions, as well as the ASHRAE PMS members (Keith Cockerham, Timothy McDowell, Jeff Stain, and Li Song) for their valuable advice during the project. Special thanks to Penn State University for sponsoring the open access of our paper at https://lnkd.in/dBXms3Ut. #datacenter #architecturalengineering #hvac

Very happy to share our latest research on the optimal design of distributed multi-energy systems, where we investigate the tradeoffs between #affordability, #sustainability, and #reliability. 🌍 🧑💻 We build upon the state of the art by modeling conversion and storage technologies under both normal and failure conditions, and we identify optimal system designs to meet energy needs while minimizing carbon emissions and energy not supplied. ⚖️ The case study of a Swiss neighborhood shows that - already today - local multi-energy systems can reduce costs by 15%, while lower emissions by 64%, and increasing reliability by 80% compared to conventional designs. 🔥 Heating is key. Optimal tradeoffs are achieved by integrating heat pumps and thermal storage, with a focus on reducing system failures and maintaining energy supply during grid interruptions. 🔋 Interestingly, whereas combined heat and power (CHP) is preferred for looser emission targets, solar PV and batteries are better suited for strict emission reductions. This highlights the importance of tailoring energy systems to specific goals, especially in low-emission grids. ✨ The work was a great and very fun research effort led by Arvind Srinivasan and supported by Giovanni Sansavini at ETH Zürich. You can find the work open-source here and in the comments below. ⬇️ #EnergySystems #Optimization #SystemDesign #EnergyCommunities

I am SO excited to announce that an idea I have had in mind for ages has just been published in Energy Conversion and Management! 🎉 Big kuddos to Mohammad Hossein Nejati Amiri ! Our new paper introduces an Explainable Deep Reinforcement Learning (XDRL) framework for resilient and battery-aware microgrid control. Operating microgrids during extreme weather is a massive challenge, and while AI can help, its "black-box" nature makes it hard to trust for safety-critical systems. Here is how we tackled that: - Transparent AI: We integrated SHAP and LIME to explain exactly why the agent makes specific charging, discharging, and load-allocation decisions. - Balanced Optimization: The model jointly optimizes a priority-weighted Resilience Index (keeping critical loads running) with a life-cycle-aware battery term (improving the life of the system). - Tested on Real Extremes: We validated the policy using real-world data from Cyclone Laila in Kothapatnam, India. - High Performance, Low Cost: The policy achieves a resilience score just 0.33% below a traditional Model Predictive Control (MPC) benchmark, while increasing expected battery life by about 5%. Best of all, online inference is roughly 5800x cheaper computationally than solving MPC at each step! ⚡It opens up edge-ready, near-optimal options ! Read the full open-access paper here: https://lnkd.in/ewvFD-p6, and our next step is to implement it in real-world conditions in our smart sip project! #Microgrids #ReinforcementLearning #ExplainableAI #XAI #RenewableEnergy #EnergyManagement

This study introduces an innovative optimization and energy management system designed for a network of interconnected microgrids featuring intermittent non-polluting generators, renewable resources, battery storage, and diesel generators. The interconnected cluster, operating off-grid, leverages community microgrids to enhance power performance through mutual and bidirectional power exchange. By integrating non-polluting generators, battery storage, and power exchange, the reliance on diesel generators is minimized, leading to reduced operational costs and fuel consumption within the cluster. To prevent simultaneous bidirectional power exchange between microgrids, a bidirectional power exchange mechanism is proposed. The optimization and energy management processes take into account the transmission distance, conducting case studies for varying levels of renewable energy penetration and demand response across hourly and day-ahead operations. The study's outcomes demonstrate that the proposed methodology presents optimal solutions for efficiently operating the cluster while ensuring effective power exchange at minimal operational costs and fuel consumption. The research findings reveal that the optimized interconnected hybrid microgrids significantly decrease daily operational costs and fuel consumption by 6.74% and 4.33%, respectively, compared to hybrid microgrids lacking power exchange. Furthermore, these interconnected microgrids exhibit a substantial improvement compared to isolated microgrids solely reliant on renewable energy and diesel generators, with reductions of 24.44% in operational costs and 54.30% in fuel consumption.

Getting caught up on great work done by my colleagues on the Open Energy Outlook team, including Aditya Sinha, Aranya Venkatesh, Katie Jordan, Ph.D., Cameron Wade, Hadi Eshraghi, Anderson Rodrigo De Queiroz, Paulina Jaramillo, and Jeremiah Johnson. They recently published a paper examining near-optimal, net-zero energy futures in the United States using an open source model and dataset (Temoa).   What do the authors mean by “near-optimal”? A least-cost energy system provides a useful reference point, but alternative model solutions may offer desirable characteristics. The authors apply modeling-to-generate-alternatives (MGA) to identify near-optimal solutions—configurations that cost slightly more than the least-cost option but feature different energy technology mixes. These alternative mixes highlight diverse pathways to achieving energy system goals.   What did the authors do in this paper? The authors applied MGA to identify 1,100 net-zero energy system configurations that were within 1% of the least cost solution. To make the results more tractable, the authors applied a clustering algorithm to the full set of results and identified six similar near-optimal decarbonization strategies, as shown in the figure below.   What did they find? The black line represents a constraint requiring emissions to reach net-zero by 2050. Despite the shared net-zero target, residual emissions associated with hard-to-decarbonize sectors are a consistent feature across the clustered solutions. To offset those residual emissions, carbon removal technologies (producing negative emissions) must be deployed. The near-optimal results suggest that carbon removal options can vary widely and offset the remaining sector-specific emissions in different ways. For example, one cluster relies heavily on direct air capture (“High DAC”), which offsets higher industrial sector emissions. By contrast, another cluster (“High Elec”) relies on clean electricity and end-use electrification, which requires significantly less carbon management.   Their analysis offers a nuanced view of the choices and trade-offs involved in achieving net-zero emissions in the United States. The results show that even a 1% budget increase can lead to vastly different fuel and technology pathways. Future policies can shape decarbonization efforts based on preferred system characteristics. The paper was published open access in Nature Communications: https://lnkd.in/eMcNiXhZ