Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Behaviour-Aware Federated Forecasting Framework for Distributed Stand-Alone Wind Turbines

Bowen Li, Xiufeng Liu, Maria Sinziiana Astefanoaei

TL;DR

A two-stage federated learning framework that first clusters turbines by long-term behavioural statistics using Double Roulette Selection initialisation with recursive Auto-split refinement, and then trains cluster-specific LSTM models via FedAvg is proposed, suggesting a practical privacy-friendly solution for heterogeneous distributed turbine fleets.

Abstract

Accurate short-term wind power forecasting is essential for grid dispatch and market operations, yet centralising turbine data raises privacy, cost, and heterogeneity concerns. We propose a two-stage federated learning framework that first clusters turbines by long-term behavioural statistics using Double Roulette Selection (DRS) initialisation with recursive Auto-split refinement, and then trains cluster-specific LSTM models via FedAvg. Experiments on 400 stand-alone turbines in Denmark show that DRS-auto discovers behaviourally coherent groups and achieves competitive forecasting accuracy while preserving data locality. Behaviour-aware grouping consistently outperforms geographic partitioning and matches strong k-means++ baselines, suggesting a practical privacy-friendly solution for heterogeneous distributed turbine fleets.

A Behaviour-Aware Federated Forecasting Framework for Distributed Stand-Alone Wind Turbines

TL;DR

A two-stage federated learning framework that first clusters turbines by long-term behavioural statistics using Double Roulette Selection initialisation with recursive Auto-split refinement, and then trains cluster-specific LSTM models via FedAvg is proposed, suggesting a practical privacy-friendly solution for heterogeneous distributed turbine fleets.

Abstract

Accurate short-term wind power forecasting is essential for grid dispatch and market operations, yet centralising turbine data raises privacy, cost, and heterogeneity concerns. We propose a two-stage federated learning framework that first clusters turbines by long-term behavioural statistics using Double Roulette Selection (DRS) initialisation with recursive Auto-split refinement, and then trains cluster-specific LSTM models via FedAvg. Experiments on 400 stand-alone turbines in Denmark show that DRS-auto discovers behaviourally coherent groups and achieves competitive forecasting accuracy while preserving data locality. Behaviour-aware grouping consistently outperforms geographic partitioning and matches strong k-means++ baselines, suggesting a practical privacy-friendly solution for heterogeneous distributed turbine fleets.
Paper Structure (59 sections, 8 equations, 8 figures, 12 tables, 2 algorithms)

This paper contains 59 sections, 8 equations, 8 figures, 12 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Contributions.
  3. Background and Related Work
  4. Federated learning for wind-farm--level power forecasting
  5. Turbine-level clustering and power forecasting
  6. Overall.
  7. Problem Formulation and System Overview
  8. Scenario and client definition
  9. Short-term turbine-level forecasting task
  10. Overview of the two-stage federated framework
  11. Feature Design
  12. Turbine Behaviour Clustering Features
  13. Prediction Model Input Features
  14. Federated Clustering Framework
  15. Problem formulation
  16. ...and 44 more sections

Figures (8)

  • Figure 1: (a) 400 turbines selected via nearest-neighbour sampling. (b) Daily generation variation of 400 turbines over one day. (c) Federated clustering followed by cluster-specific federated LSTM forecasting.
  • Figure 2: Double Roulette Selection
  • Figure 3: LSTM--MLP architecture.
  • Figure 4: Comparison of behaviour- and geo-based grouping strategies.
  • Figure 5: DRS-auto clustering results.
  • ...and 3 more figures