Skip to content
#federated learning Open access

Grid-constrained peer-to-peer energy trading: enforcement-mechanism selection, personalized federated forecasting and auditable energy-quantity settlement

Oct 2026 · Applied Energy · 47 references
Smart Grid Energy Management

Abstract

Decentralized energy markets can coordinate distributed generation, flexible demand and storage, but feeder feasibility, household forecasting and smart-contract settlement are usually assessed separately. This study develops a grid-constrained Decentralized Energy Marketplace (DEM) linking household commitments, bilateral matching, feeder-capacity screening, personalized federated forecasting and auditable energy-quantity settlement in a module-coupled, open-loop configuration with no feedback from settlement outcomes to forecasting or bidding. The principal finding concerns the enforcement mechanism applied when a feeder limit binds. Across 45 paired scenarios of measured Ausgrid households under a 1.5 kW branch-transfer stress cap, whole-trade removal retains 74.9% of the price-only submitted-price surplus, whereas marginal re-clearing retains 93.1%, a paired gain of 0 . 0 2 1 9 ± 0 . 0 0 6 0 EUR per interval ( 𝑡 = 7 . 1 2 ); this ordering holds in eight seasonal and weekday/weekend windows spanning the measured year, with the summer case-study week as the largest-gap case. AC power flow on 4320 accepted schedules shows no voltage or thermal violations; the first demand-side limit occurs at 1.7 times peak demand. In a 30-client non-IID benchmark, plain FedAvg underperforms local training for 87% of client–seed pairs, a known federated-learning effect demonstrated on measured energy data; per-client fine-tuning removes most of the penalty, and forecast-generated commitments materially change deviation-adjusted outcomes. On-chain replay exposed an energy-conservation defect missed by the initial tests; after correction, all 116 tradable intervals (1,325 trades) reconciled within integer rounding and a randomized differential suite matched an integer-exact reference on two EVM clients. Efficient constraint handling, client-level personalization and realistic multi-party settlement testing are central to credible blockchain-enabled local energy trading.

View source

Similar papers

#machine learning Review Open access Oct 2014

Software development in startup companies: A systematic mapping study

The results indicate that software engineering work practices are chosen opportunistically, adapted and configured to provide value under the constrains imposed by the startup context.

Nicolò Paternoster, Carmine Giardino, M. Unterkalmsteiner et al. · 394 citations · ⚡54
#machine learning Review Open access Jun 2014

Why Early-Stage Software Startups Fail: A Behavioral Framework

This state-of-practice investigation was performed using a literature review followed by a multiple-case study approach and presents how inconsistency between managerial strategies and execution can lead to failure by means of a behavioral framework.

Carmine Giardino, Xiaofeng Wang, P. Abrahamsson · 175 citations · ⚡19
#machine learning Review Open access Oct 2016

“Failures” to be celebrated: an analysis of major pivots of software startups

This study conducts a case survey study based on the secondary data of the major pivots happened in 49 software startups, and demonstrates that customer need pivot is the most common among all pivot types.

Sohaib Shahid Bajwa, Xiaofeng Wang, Anh Nguyen-Duc et al. · 127 citations · ⚡15
#machine learning Review Open access May 2016

Key Challenges in Software Startups Across Life Cycle Stages

It is found that what perceived as biggest challenges by software startups do vary across different life cycle stages, even though its significance decreases when the learning focuses of the startups move from problem to solution and their products mature.

Xiaofeng Wang, Henry Edison, Sohaib Shahid Bajwa et al. · 62 citations · ⚡6

Related blog posts

MIT News · Artificial Intelligence Oct 7, 2026

Discovering the value of humanistic inquiry

Students in MIT’s Concourse program delve deeply into the human condition, debate challenging questions, and learn to develop judgment about issues that can’t be quantified.

Microsoft Research Blog Oct 7, 2026

Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses

Training AI agents with reinforcement learning can be challenging because their tools, context, and decision-making are managed by complex frameworks. Agent Lightning connects existing agents to RL training, making it easier to improve them without rebuilding them. The post Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.