Wind Farm Control Research Articles

Distributionally robust model predictive control for wind farms

In this paper, we develop a distributionally robust model predictive control framework for the control of wind farms with the goal of power tracking and mechanical stress reduction of the individual wind turbines. We introduce an ARMA model to predict the turbulent wind speed, where we merely assume that the residuals are sub-Gaussian noise with statistics contained in an moment-based ambiguity set. We employ a recently developed distributionally model predictive control scheme to ensure constraint satisfaction and recursive feasibility of the control algorithm. The effectiveness of the approach is demonstrated on a practical example of five wind turbines in a row.

Coordinated control of wind farm and supercapacitor energy storage system for dynamic performance reinforcement of multi-area power systems

Coordinated control of wind farm and supercapacitor energy storage system for dynamic performance reinforcement of multi-area power systems

Coordinated Power Reserve Control of Wind Farm for Frequency Regulation

Coordinated Power Reserve Control of Wind Farm for Frequency Regulation

Data-Driven Multi-Resolution Probabilistic Energy and Reserve Bidding of Wind Power

The current wind farm control schemes qualify wind power producers (WPPs) to provide balancing services in complement to energy in modern electricity markets. Accordingly, WPPs are responsible for real-time deviations in both energy and reserve market floors, which are settled at different time scales. WPPs should adjust their output to cope with fast wind variations, which are critical in the balancing stage. In this paper, we devise a reliable high-temporal-resolution day-ahead bidding framework for WPPs considering the ultra-short-term wind stochasticity. To that end, the model for the bidding strategy is enriched with a probabilistic constraint controlling the confidence level on reserve bids to enhance the reliability of the offered capacity. Additionally, an original Auxiliary Classifier Wasserstein Generative Adversarial Network (ACWGAN) is proposed to generate high-temporal-resolution wind speed scenarios to be embedded into the bidding framework. The numerical results firstly confirm the superiority of the proposed ACWGAN over the other GAN-based alternatives. For instance, ACWGAN can reach 30% higher classification accuracy compared to conditional Wasserstein GAN. Then, the effectiveness of the proposed data-driven method over its single-resolution counterpart and other scenario representation methods is verified regarding the minimization of the negative impact of wind variability on WPPs' profit and reliability of offered reserve bids.

Koopman Model Predictive Control for Wind Farm Yield Optimization with Combined Thrust and Yaw Control

Two novel approaches to data-driven wind farm control via Koopman model predictive control are presented, both combining thrust and yaw control for yield optimization and power reference tracking. The Koopman framework is used to build prediction models to predict wake effects of upwind on downwind turbines. This paper extends previous work by using yaw in addition to thrust control. The test case is a wind farm consisting of two turbines and wind with constant speed and direction parallel to the main axis of the farm. In closed-loop simulation, the two Koopman model predictive control designs reduce the tracking error considerably with regards to a previously published baseline controller, which used solely axial induction control. It is also demonstrated that this can be achieved with relatively small yaw angles, avoiding mechanical loads acting on turbines operating misaligned to the wind, making this a promising approach for further investigations in 3D medium and high fidelity simulation environments.

A CFD‐based analysis of dynamic induction techniques for wind farm control applications

SummaryRecently, dynamic induction control is gaining the interest of the wind energy community as a promising strategy to increase the overall wind farm power production. Such a technique is based on a dynamic variation of the upstream rotor thrust, generated through a suitable blade pitch motion, to promote a faster wake recovery. Notwithstanding some promising results already published, the knowledge of the physical mechanism, connecting dynamic induction to the increased in‐wake velocity, was not yet exploited to enhance control effectiveness. This paper, through a computational fluid dynamics procedure based on large eddy simulations coupled with actuator line models, provides a description of the working principles of this control from a fluid dynamics standpoint. The analyses show that the faster recovery is strictly connected to the ability of the blade tip vortices to roll up and sucking energy from the outer flow. Exploiting such knowledge, a novel control strategy, which improves the vortex roll up mechanism, is proposed and analyzed. The new control proved more effective than standard techniques especially for very low turbine spacing.

Reliability assessment of cyber-physical power systems considering the impact of predicted cyber vulnerabilities

This paper presents a reliability assessment technique for cyber-physical power systems (CPPSs) that incorporates cybersecurity issues by considering non-normal random variables with non-linear dependencies. Our model uses bidirectional encoder representations from transformers (BERTs) to predict the severity of cyber vulnerabilities by analyzing their textual description. This practical model allows the engineer to manually enter text that describes threats to predict the probability of cyber vulnerability. We use a Bayesian attack graph to simulate the attack paths and a Markov model to demonstrate the consequences of cyber attacks on the CPPS. Furthermore, a hybrid approximate-analytical approach is employed to assess the overall reliability of the CPPS when considering the load and wind generation uncertainties, given their non-linear dependencies. The proposed method is implemented on the modified IEEE Reliability Test System. The results demonstrate that the reliability indices for a simultaneous cyber attack on corporate networks and wind farm control centers are more significant than those obtained for other attack scenarios. The proposed prediction method is highly accurate, and the predicted vulnerability scores align with those provided by the National Vulnerability Database. The results are compared with sequential Monte Carlo simulation, subset simulation, and cross-entropy-based importance sampling methods. The comparison shows that the proposed method outperforms other methods in terms of convergence speed and accuracy.

Wind inflow observation from load harmonics via neural networks: A simulation and field study

In this work, feed-forward neural networks are used to estimate the inflow of a wind turbine in real time during operation. A previously published method achieved this capability by correlating the turbine response to the wind inflow via a piecewise-linear model. Specifically, the existing formulation used the once-per-revolution harmonics of both in- and out-of-plane blade bending moments to estimate four wind parameters: the vertical and horizontal shears, and the vertical and horizontal misalignment angles. The novel method presented here builds on the same concept, but replaces the linear model with neural networks. Compared to the previous approach, the neural-based one is simpler to tune and use, because it decouples the wind parameters, which can be observed each one independently from the others. The proposed formulation is tested using both aeroservoelastic simulations and field data obtained on a 3.5 MW machine. In the field, independent measurements of the inflow are provided by a nearby hub-tall met mast, providing data for the training and verification of the wind observer. Notwithstanding some limitations of the experimental setup, results indicate that the neural networks produce estimates of the shears and yaw misalignment angle with accuracies compatible with typical use cases in wind turbine and wind farm control. Additionally, the neural-based observer appears to slightly outperform the original formulation in the lower wind speed range, where stronger nonlinearities in the load-wind mapping can occur.

Wind farm control for wake-loss compensation, thrust balancing and load-limiting of turbines

As renewable energy sources such as wind farms become dominant, new challenges emerge for operating and controlling them. Traditionally, wind farm control aims to dispatch power set-points to individual turbines to maximize energy extraction and, thus, their usage as assets. Yet, grid balance and frequency support are fundamental in presence of high renewable penetration and volatility of energy prices and demand. This requires a paradigm change, moving from power maximization to revenue maximization. In this paper, three active power control strategies pushing this shift of paradigm are investigated, namely: wake-loss compensation, thrust balancing, and load-limiting control.The findings of large eddy simulations of a reference wind farm show that wake-loss compensation indeed improves the power generation on waked wind farms, but at the price of increased structural loads on certain turbines. The addition of a thrust balancing can equalize the stresses of individual turbines and their wear in the long term, while still attaining the required power output at the farm level. Furthermore, load-limiting controllers could potentially aid by allowing maintenance to be scheduled in a single time window, thus reducing operation and maintenance costs.

Neural Networks for Improving Wind Power Efficiency: A Review

The demand for wind energy harvesting has grown significantly to mitigate the global challenges of climate change, energy security, and zero carbon emissions. Various methods to maximize wind power efficiency have been proposed. Notably, neural networks have shown large potential in improving wind power efficiency. In this paper, we provide a review of attempts to maximize wind power efficiency using neural networks. A total of three neural-network-based strategies are covered: (i) neural-network-based turbine control, (ii) neural-network-based wind farm control, and (iii) neural-network-based wind turbine blade design. In the first topic, we introduce neural networks that control the yaw of wind turbines based on wind prediction. Second, we discuss neural networks for improving the energy efficiency of wind farms. Last, we review neural networks to design turbine blades with superior aerodynamic performances.

Wind farm flow control: prospects and challenges

Abstract. Wind farm control has been a topic of research for more than two decades. It has been identified as a core component of grand challenges in wind energy science to support accelerated wind energy deployment and to transition to a clean and sustainable energy system for the 21st century. The prospect of collective control of wind turbines in an array, to increase energy extraction, reduce structural loads, improve the balance of systems, reduce operation and maintenance costs, etc. has inspired many researchers over the years to propose innovative ideas and solutions. However, practical demonstration and commercialization of some of the more advanced concepts has been limited by a wide range of challenges, which include the complex physics of turbulent flows in wind farms and the atmosphere, uncertainties related to predicting structural load and failure statistics, and the highly multi-disciplinary nature of the overall design optimization problem, among others. In the current work, we aim at providing a comprehensive overview of the state of the art and outstanding challenges, thus identifying the key research areas that could further enable commercial uptake and success of wind farm control solutions. To this end, we have structured the discussion on challenges and opportunities into four main areas: (1) insight in control flow physics, (2) algorithms and AI, (3) validation and industry implementation, and (4) integrating control with system design (co-design).

Ensemble-Based Flow Field Estimation Using the Dynamic Wind Farm Model FLORIDyn

Wind farm control methods allow for a more flexible use of wind power plants over the baseline operation. They can be used to increase the power generated, to track a reference power signal or to reduce structural loads on a farm-wide level. Model-based control strategies have the advantage that prior knowledge can be included, for instance by simulating the current flow field state into the near future to take adequate control actions. This state needs to describe the real system as accurately as possible. This paper discusses what state estimation methods are suitable for wind farm flow field estimation and how they can be applied to the dynamic engineering model FLORIDyn. In particular, we derive an Ensemble Kalman Filter framework which can identify heterogeneous and changing wind speeds and wind directions across a wind farm. It does so based on the power generated by the turbines and wind direction measurements at the turbine locations. Next to the states, this framework quantifies uncertainty for the resulting state estimates. We also highlight challenges that arise when ensemble methods are applied to particle-based flow field simulations. The development of a flow field estimation framework for dynamic low-fidelity wind farm models is an essential step toward real-time dynamic model-based closed-loop wind farm control.

Model Predictive Control for Wind Farm Power Tracking With Deep Learning-Based Reduced Order Modeling

Dynamic power control of wind farms (WFs) is necessary to provide automatic generation control (AGC) services for the power system. However, cooperative WF control for AGC remains a great challenge because of the nonlinear and high-dimensional nature of the wake flow dynamics. To address this challenge, this article proposes a model predictive control (MPC) framework with deep learning-based reduced-order modeling (ROM). Two novel neural network architectures are designed, which successfully formulate a WF ROM capturing the dominant wake steering dynamics in a computationally efficient manner. Compared to physical models, the data-driven ROM reduces the number of model states by orders of magnitude. Then, a novel WF AGC framework embedding the derived WF ROM is proposed. Thrust coefficient and yaw steering are both employed to optimize WF power tracking performance. Compared to prior WF AGC controllers, the dynamic yaw actuation is first optimized for AGC considering the wake steering dynamics. Case studies validate the effectiveness of the deep learning-based WF ROM at capturing the wake traveling dynamics. The WF controllers were stress-tested under time-varying inflow directions. The proposed MPC can react to different wind directions and generates higher-quality control performance than existing alternatives with extended trackable AGC range and better dynamic power tracking performance.

Wind plant controls

The development of operational strategies for wind farms as an integrated plant system to achieve a variety of goals from elevating power production to reducing maintenance needs has generated a great deal of interest in recent years. Achieving these operational goals requires an estimate of the energy available and the wind conditions affecting each turbine. The importance of the aerodynamic interaction of wind turbines with the dynamic atmospheric resource means that wakes (the momentum deficit due to power extraction) and their interactions through the farm have the largest influence on the available energy. Predicting the influence of wakes and their interactions, therefore, form the basis of wind farm control strategies to reduce power production losses, track a power signal, mitigate structural loading, or balance the wear and tear on wind turbines to decrease operation and maintenance costs. The articles in the “Advances in Wind Plant Controls: Strategies, Implementation, and Validation” Special Topic in the Journal of Renewable and Sustainable Energy describe the further development and evaluation of wake models and new approaches to wake steering that exploit advances in sensing or estimation to improve control performance.

The revised FLORIDyn model: implementation of heterogeneous flow and the Gaussian wake

Abstract. In this paper, a new version of the FLOw Redirection and Induction Dynamics (FLORIDyn) model is presented. The new model uses the three-dimensional parametric Gaussian FLORIS model and can provide dynamic wind farm simulations at a low computational cost under heterogeneous and changing wind conditions. Both FLORIS and FLORIDyn are parametric models which can be used to simulate wind farms, evaluate controller performance and can serve as a control-oriented model. One central element in which they differ is in their representation of flow dynamics: FLORIS neglects these and provides a computationally very cheap approximation of the mean wind farm flow. FLORIDyn defines a framework which utilizes this low computational cost of FLORIS to simulate basic wake dynamics. This is achieved by creating so-called observation points (OPs) at each time step at the rotor plane which inherit the turbine state. In this work, we develop the initial FLORIDyn framework further considering multiple aspects. The underlying FLORIS wake model is replaced by a Gaussian wake model. The distribution and characteristics of the OPs are adapted to account for the new parametric model but also to take complex flow conditions into account. To achieve this, a mathematical approach is developed to combine the parametric model and the changing, heterogeneous world conditions and link them with each OP. We also present a computationally lightweight wind field model to allow for a simulation environment in which heterogeneous flow conditions are possible. FLORIDyn is compared to Simulator for Offshore Wind Farm Applications (SOWFA) simulations in three- and nine-turbine cases under static and changing environmental conditions. The results show a good agreement with the timing of the impact of upstream state changes on downstream turbines. They also show a good agreement in terms of how wakes are displaced by wind direction changes and when the resulting velocity deficit is experienced by downstream turbines. A good fit of the mean generated power is ensured by the underlying FLORIS model. In the three-turbine case, FLORIDyn simulates 4 s simulation time in 24.49 ms computational time. The resulting new FLORIDyn model proves to be a computationally attractive and capable tool for model-based dynamic wind farm control.

Towards multi-fidelity deep learning of wind turbine wakes

Engineering wake models that accurately predict wake in a computationally efficient manner are very important for tasks such as layout optimization and control of wind farms. In this paper, we explore an application of deep learning (DL) to learn the wake model from hierarchies of physics-based approaches ranging from analytical models to an approximate form of the Reynolds-averaged Navier–Stokes equations. We first illustrate the application of principal component analysis to obtain a lower-dimensional representation that allows a computationally tractable training and deployment of DL models. Then, the DL model is trained to learn the mapping from input parameter space to the principal components, which are then used to reconstruct the three-dimensional flow field. Additionally, we investigate a composite framework consisting of two neural networks to learn the correlation between low- and high-fidelity data with Gauss and curl models treated as proxies for low- and high-fidelity models, respectively. The prediction from both DL models matches well with the high-fidelity data with a maximum relative percentage error for the kinetic energy flux of <1%. This work opens up possibilities for data-efficient construction of surrogate models for wake prediction that can be used to study the influence of wind speed and yaw angles on wind farm power production.

Adaptive Response Method for Communication Failures of Hierarchical Reactive Power Control in Wind Farms

Advanced controllers have been widely introduced in power system-related research to alleviate the uncertainties of renewable sources. The performance of the controllers should be verified in a realistic environment before the controllers are used in practical applications. This study focuses on voltage regulation of wind farms in voltage drop cases where communication failures occur and discusses modifications done on the online wind farm management system. The modified management system is designed to respond to a voltage drop when the outer reference signal from the upper-level controller is absent due to technical problems, especially communication errors. Such a situation has not been considered in most of the management systems in the literature. The predictable situations in real fields have been further verified under the hardware-in-the-loop simulation environment RTDS <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TM</sup> . During communication failure, the proposed method can respond automatically to voltage recovery, along with a designed objective function. A comparative analysis of actual control modes was also conducted to prove the relevance and effectiveness of the proposed method.

New engineering wake model for wind farm applications

Engineering wake model is widely used in today's wind farm design because of its fast performance. In the literature, there were good developments in the track of engineering wake models. However, its accuracy is still not ideal. In this paper, two novel analytical wake models (one uses a Gaussian shape and the other uses a cosine shape) are developed to predict the wind velocity distribution in the wake region of a wind turbine by applying the momentum and mass conservation laws. Next, the two models are validated using published experimental data and large-eddy simulation (LES) data, and also compared to the analytical wake model of Ishihara and Qian which is considered to be the state-of-the-art engineering wake model in the literature. In all considered cases, the relative errors between the new models and experimental or LES data are much lower than those between the Ishihara and Qian model and experimental or LES data. From the comparisons, it is also seen that the new model with a Gaussian shape is slightly better in far-field wake prediction in most cases while the new model with a cosine shape is slightly better in a few cases. Because of their good performance, these two models are recommended in wind farm design including both wind farm layout optimization and wind farm control.

A physics-inspired neural network model for short-term wind power prediction considering wake effects

Accurate short-term wind power prediction plays an essential role in the wind farm control and the dispatch of wind energy into the power system. Incorporating physical factors that have a major impact on wind farm power generation into machine learning algorithms has always been an important way to improve prediction accuracy. Overlooked in the literature, however, is the influence of wind turbines wakes in improving model predictions. In this work, a physics-inspired neural network model for short-term wind power prediction is developed considering wake effects. Different from traditional neural network models, part of the nodes in the proposed model are determined by the analytical wake model, which enhances the statistical prediction model physically. In this way, the model can be well adapted to the wake effects in the wind farm. Verifications in the actual wind farm case illustrate that there is a good agreement with the prediction results and measured data. Compared with traditional models, the wind power prediction performance of the proposed model has improved by more than 20% in terms of RMSE. Based on this work, we recommend that the wake effect should be considered in the short-term wind power prediction model, which is of great benefit to improving its accuracy.

Two-stage model predictive control for a hydrogen-based storage system paired to a wind farm towards green hydrogen production for fuel cell electric vehicles

This study proposes a multi-level model predictive control (MPC) for a grid-connected wind farm paired to a hydrogen-based storage system (HESS) to produce hydrogen as a fuel for commercial road vehicles while meeting electric and contractual loads at the same time. In particular, the integrated system (wind farm + HESS) should comply with the “fuel production” use case as per the IEA-HIA report, where the hydrogen production for fuel cell electric vehicles (FCEVs) has the highest unconditional priority among all the objectives. Based on models adopting mixed-integer constraints and dynamics, the problem of external hydrogen consumer requests, optimal load demand tracking, and electricity market participation is solved at different timescales to achieve a long-term plan based on forecasts that then are adjusted at real-time. The developed controller will be deployed onto the management platform of the HESS which is paired to a wind farm established in North Norway within the EU funded project HAEOLUS. Numerical analysis shows that the proposed controller efficiently manages the integrated system and commits the equipment so as to comply with the requirements of the addressed scenario. The operating costs of the devices are reduced by 5%, which corresponds to roughly 300 commutations saved per year for devices.