Engineering topics

Stanfel, Paul

Publications and source records attributed to Stanfel, Paul.

4 records

DOE OSTI2021

Proof-of-concept of a reinforcement learning framework for wind farm energy capture maximization in time-varying wind

Here, we present a proof-of-concept distributed reinforcement learning framework for wind farm energy capture maximization. The algorithm we propose uses Q-Learning in a wake-delayed wind farm environment and considers time-varying, though not yet fully turbulent, wind inflow conditions. These algorithm modifications are used to create the Gradient Approximation with Reinforcement Learning and Incremental Comparison (GARLIC) framework for optimizing wind farm energy capture in time-varying conditions, which is then compared to the FLOw Redirection and Induction in Steady State (FLORIS) static lookup table wind farm controller baseline.

17 WIND ENERGY↗

DOE OSTI2021

Network Based Estimation of Wind Farm Power and Velocity Data Under Changing Wind Direction

This paper describes an estimation algorithm for velocity and power output signals in a wind farm under changing wind direction. A graph-theoretic definition describes the wind farm as a collection of nodes (turbines) and time-varying weighted edges (inter-turbine wake propagation) that change as a function of incoming wind direction. The velocity at each turbine is determined through a discrete input-output model. Changes in wind direction serve as the input and the output is defined in terms of a time-varying weighted adjacency matrix that depends on the time-delay of information propagation between turbines. These delays, which are defined in terms of the advection speed of the wind and the distance between the turbines, capture the delayed effect of wind direction changes on the inter-connectivity of the graph as the wind conditions at the farm inlet propagate through the turbine array. An event-based update framework is employed to capture time-dependent topology changes due to shifts in wind direction. Simulation results for dynamically changing wind inlet directions to a circular wind farm are compared to predictions from both the static and dynamic versions of the FLOw Redirection and Induction in Steady State (FLORIS) model. The approach is shown to enable real-time tracking of dynamic changes to wind farm power output within a framework that can be easily integrated into real-time, horizon-based, control strategies that typically do not account for wind direction changes.

distributed↗

DOE OSTI2021

Deep Reinforcement Learning for Automatic Generation Control of Wind Farms

This paper provides a model-free framework for real-time control of wind farms to accurately track a power reference signal. This problem requires tractable dynamical models for capturing the aerodynamic interaction between wind turbines and controllers that can make decisions in realtime given varying atmospheric conditions. In this paper, we propose a deep reinforcement learning framework to provide real-time yaw control of a wind farm. Modifications have been made to FLOw Redirection and Induction in Steady State (FLORIS), a modeling tool that incorporates transient wake behavior. The control problem is formulated to track a synthetic power reference signal based on historical atmospheric (wind speed and direction) information, price signals, and regulation deployment data from U.S. regional transmission operators. Results indicate that a wind farm, with this control paradigm, can achieve good tracking performance when tested with real atmospheric data.

49 EE - Wind and Water Power Program - Wind (EE-4W↗

DOE OSTI2020

A Distributed Reinforcement Learning Yaw Control Approach for Wind Farm Energy Capture Maximization: Preprint

In this paper, we present a reinforcement-learning based distributed approach to wind farm energy capture maximization using yaw control, also known as wake steering. In order to maximize the power output of a wind farm, it is often necessary for individual turbines to decrease their own power output through yaw misalignment so as to deflect their wakes away from downstream turbines. Although using model-based methods to achieve yaw misalignment is one option, a model-free method might be better suited to incorporate factors that are difficult to model or changing conditions. We propose an algorithm that adapts concepts of temporal difference reinforcement learning distributed to a multi-agent environment that allows individual turbines to act so as to optimize overall wind farm output and react to unforeseen disturbances.

controls↗