Smart Grid Optimization

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

AI-enhanced power grid optimization can reduce emissions that are the equivalent of removing 6.5 million (U.S.) gas-powered mid-size passenger vehicles from the road for a year. “AI” is a much broader term than what most people think of—it’s not all LLMs! When it comes to reducing energy waste and operational power grid emissions, AI can help by dispatching generation assets more optimally, reducing losses, congestion, and cost. In our paper, which will be presented at the NeurIPS 2025 Workshop "Tackling Climate Change with Machine Learning," we analyze the operational emissions associated with training CANOS, Google DeepMind’s graph neural network for solving AC Optimal Power Flow (OPF) on a 10,000-bus power system. We then estimate how emissions and energy use would change if these dispatch solutions were used to determine generator (power plant) dispatch decisions, instead of the status-quo linear approximations used in many power markets to set generator output. Especially compared to training something as complex as an LLM, training these GNNs—which have a focused task (learning OPF solutions)—“pays back” all energy and emissions costs associated with the model's training within a single hour. At a country-wide scale, operating the grid more efficiently using these models is approximately equivalent to removing 6.5 million (U.S.) gas-powered mid-size passenger vehicles from the road for a year. Of course, a full analysis would require a lifecycle carbon assessment of training these GNNs. And we'd have to run the actual power grid models themselves across ISOs, not just a 10,000 bus synthetic grid. Additionally, we'd need to model other grid components and concepts like ancillary services, self-schedulers, and more. But even if we’re off by, say, a HUNDRED times, the conclusion is still clear: using a GNN approximation for dispatch can reduce energy use and emissions relative to DC OPF-based approximations. (Even if we're off by the training emissions by a *thousand* times, this holds true.) If you’re at NeurIPS in San Diego this year, please come chat with me at the session if you’re interested in this work! Read more here: https://lnkd.in/g9aqhXpy And stop saying "AI" when you actually mean LLMs. :)

Grid stability and security are becoming data + control problems. Utilities and large energy operators are already using Artificial Intelligence (AI) to move from reactive alarms to predictive, resilient, and cyber-aware operations—especially as renewables increase volatility. Here’s where Machine Learning (ML) and Deep Learning (DL) deliver real impact: ✅ Anomaly Detection: clustering + autoencoders to flag abnormal grid states and potential cyber events ✅ Fault Detection & Classification: Decision Trees, Random Forests, Support Vector Machine (SVM) models using voltage/current/frequency features ✅ Predictive Maintenance: Remaining Useful Life (RUL) forecasting to reduce unplanned outages (breakers, transformers, lines) ✅ Voltage Stability: Recurrent Neural Network (RNN) + Long Short-Term Memory (LSTM) models to anticipate instability and corrective actions ✅ Cybersecurity: Intrusion Detection System (IDS) + Anomaly Detection System (ADS) using supervised and unsupervised Machine Learning (ML) ✅ Optimal Power Flow (OPF): faster optimization with Machine Learning (ML) surrogates + Linear Programming (LP), Quadratic Programming (QP), Interior Point Method (IPM) constraint handling ✅ Forecasting: Autoregressive Integrated Moving Average (ARIMA) + Seasonal Autoregressive Integrated Moving Average (SARIMA) for load and generation inputs ✅ Uncertainty: Monte Carlo simulation + stochastic programming for renewables and market variability ✅ Autonomous control (next wave): Reinforcement Learning (RL) + Multi-Agent Reinforcement Learning (MARL), plus Federated Learning for privacy-preserving training What’s your biggest grid pain right now: false alarms, asset failures, voltage events, congestion, or cybersecurity? #ArtificialIntelligence #MachineLearning #DeepLearning #PowerSystems #GridReliability #Cybersecurity #PredictiveMaintenance #EnergyTransition

I am pleased to share that our paper, “EV Charging and V2G Operation for Distribution System VPP including Model Predictive Control”, has been published in IEEE Access and available in IEEE Xplore. The paper presents a comprehensive framework for developing smart‑grid Virtual Power Plants (VPPs) that integrate electric vehicle (EV) charging and vehicle‑to‑grid (V2G) operation using industry communication standards such as OCPP, IEC 15118, and IEC 61851. Leveraging a Python‑based OpenDSS VPP platform with thousands of experimental smart‑meter profiles, the IEEE 123‑node test feeder, and a large set of survey‑based EV models, the study proposes a model‑predictive control (MPC) methodology that incorporates power‑flow calculations to minimize utility costs and maintain near‑constant substation loading. The work introduces the concepts of “EV hourly and daily homogeneity” to evaluate the impact of natural‑disaster‑type synchronized charging events, identifying adverse effects beginning at 20% and 50% homogeneity and showing that extreme overlap can increase system losses by up to 1.45 MWh. Results demonstrate up to 26% cost reduction during VPP operation under California retail assumptions and, in a multi‑objective formulation, show that coordinated VPP control can simultaneously reduce system losses by 45% and flatten substation load within a 5% tolerance, highlighting the broader operational benefits of intelligent EV‑centric VPP management. Congratulations to all my co‑authors Rosemary Alden, Malcolm McCulloch, and Dan M. Ionel. 📄 Link to Open Access publication: https://lnkd.in/gSCaiKCv

Everyone talks about how slow it is to build new transmission lines. Less noticed is how much capacity is being freed — right now — on the wires we already have. Three families of “grid-enhancing technologies” (GETs) are scaling fast: (1) advanced reconductoring with modern high-performance conductors that can double capacity within existing rights-of-way; (2) dynamic and ambient-adjusted line ratings (DLR/AAR) that raise safe operating limits based on real weather, not worst-case assumptions; and (3) power-flow control, topology optimization, and other software tools that route power away from bottlenecks to under-used lines. Together, these are connecting more renewables, cutting curtailment and congestion, and buying precious time while big new lines are planned and built. GETs complement — not replace — new transmission. They reduce congestion and keep projects moving while long-lead lines, HVDC backbones, and interregional upgrades work through siting and permitting. Bottom line: We don’t need to wait a decade for every gigawatt of grid capacity. Sensors, software, and smarter wires are quietly turning today’s network into tomorrow’s — doubling capacity on key spans, adding double-digit ratings on windy days, and routing power around bottlenecks. It’s pragmatic, portfolio-based progress that’s already cutting congestion and connecting clean energy at scale. #gridenhancingtechnologies #get #reconductoring #dlr #aar #sensors #topologyoptimization #congestion #bottlenecks #hvdc #energytransition https://lnkd.in/eawe5mkm

The grid isn’t ready, not because we lack power, but because we can’t control it fast enough ⚡ A few shifts are now impossible to ignore: • Interconnection queues are forcing a flexibility‑first mindset • AI‑driven data center load is locking up supply years ahead • DERs, storage, and VPPs aren’t niche anymore, they’re core infrastructure • Utilities leaning on software to manage volatility, not just build capacity It’s no longer about how much power you have; it’s about when you can deliver it and how intelligently you move it. That’s why we’re seeing real momentum behind: - VPPs and DER orchestration - Demand response at scale - Co‑located storage + intelligent dispatch - Real‑time grid optimization - Interconnection Studies - DERMS & Forecasting Software The winners won’t just generate power - They’ll control it. Examples we’re watching: • EnergyHub - operating one of North America’s largest cross‑DER VPP platforms with millions of devices under management. • PowerFlex + WeaveGrid — partnering to orchestrate EV charging and DER capacity for utilities. How are you seeing these shifts play out, especially in ERCOT and other constrained markets? #GridFlexibility #EnergySoftware #SmartGrid #DERs #VirtualPowerPlants #EnergyInnovation #FutureOfEnergy