Pulsed Lasers in Lidar Application: Models for Optimizing Wind Turbine Performance

Pulsed Lasers in Lidar Application: Models for Optimizing Wind Turbine Performance

ABSTRACT

Since the appearance of the first wind turbines at the end of the nineteenth century, wind energy has been considered to be a renewable energy source for not just developed countries but also developing countries as well. Thus, since 2004, there has been a steady rise in wind energy production worldwide, with substantial actively installed capacity in Africa. However, due to the fact that wind turbines are highly dynamic systems that are excited by stochastic loads from the wind, variations in this disturbance usually medium-term (-changes during the space of a few hours or minutes cause variations in power output which must be accepted by the system to which the turbine is connected-) and short-term (typically wind gusts which will introduce cyclic loadings which must be absorbed by the wind turbine with high susceptibility to fatigue damage) negatively impacts heavily on Levelized Cost Of Energy (LCOE) of wind energy (i.e., the average cost per unit of energy over the lifetime of the turbine, including capital costs, operations and maintenance costs, and all other relevant expenses) [1].

As a result, there has been slow progress in the growth and development of more powerful turbines which have some significant advantages such as less visual impact to local people, and projects with more profitability. While traditional wind turbine control design utilize feedback control algorithms such as Artificial Neural Network (AAN) algorithms to address this challenge, this has often proved ineffective because they are only able to react to impacts of wind changes on the turbine dynamics after these impacts have already occurred [2].

Consequently, as a promising alternative, Light Detection and Ranging (Lidar) allows preview information about the approaching wind to be used to improve wind turbine control including blade pitch, generator torque, and yaw direction, thereby optimizing operational performance of the wind turbine through increase in energy yield, while keeping structural loads low [3]. Therefore, it is our goal in this thesis to carry out a thorough exposition of modeling associated with this trend. We will first focus on lidar system modeling with particular emphasis on the laser device which is the primary component of the lidar systems. Then we will explore wind and wind turbine modeling through aero-elastic simulations, and then wind field reconstructions with correlations between Lidar systems and Wind turbines [4]. We will end with an insight into what is to be expected with regards to the lidar scanning pattern and consequently the entire lidar-wind turbine models and simulations when pulsed Lasers operating in the pico and/or femtosecond regime are used in the laser system as against the traditional nanoseconds pulsed lasers.

TABLE OF CONTENTS

Abstract i
Acknowledgments iii
Dedication v
Table of Contents vii
List of Tables viii
List of Figures x
1 Introduction 1
1.1 Background and Motivation . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 The Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 The Aim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.6 Scope of Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.7 Limitation of Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2 Literature Review 8
2.1 Wind Turbines and Wind Modeling . . . . . . . . . . . . . . . . . . . 8
2.1.1 The Physics of Wind Energy . . . . . . . . . . . . . . . . . . . 8
2.1.2 What is a Wind Turbine? . . . . . . . . . . . . . . . . . . . . 13
2.1.3 Wind Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 LIDAR and Lidar Modeling . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.1 Lidar Operating Principle . . . . . . . . . . . . . . . . . . . . 19
2.2.2 Lidar Components . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2.3 Lidar Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3 Correlation between Wind Systems and LiDAR Systems . . . . . . . 24
3 Models and Methodology 26
3.1 Classical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.1.1 Structural Mechanics . . . . . . . . . . . . . . . . . . . . . . . 26
3.1.2 Modal shape functions and Principle of Virtual Work . . . . . 26
3.1.3 Cyclic Loading . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2 Fluid Mechanical Models . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.2.1 Rotational Effects . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.2 Forces in the Rotating Frame of Reference . . . . . . . . . . . 36
3.2.3 Boundary Layer Assumptions . . . . . . . . . . . . . . . . . . 36
3.2.4 Attached Flow on a Rotating Blade . . . . . . . . . . . . . . . 37
3.3 Quantum Mechanical Model . . . . . . . . . . . . . . . . . . . . . . . 38
3.3.1 Quantum Scattering . . . . . . . . . . . . . . . . . . . . . . . 39
3.3.2 Lidar Equation . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.3.3 Elastic-Backscattered Lidar . . . . . . . . . . . . . . . . . . . 46
3.4 Correlation models and algorithm for field reconstruction . . . . . . . 48
3.5 Wind Fields and Wind Evolution Models . . . . . . . . . . . . . . . . 50
3.5.1 The Great Plains-Low Level Jet Wind Field . . . . . . . . . . 51
3.5.2 Exponential Wind Evolution Model . . . . . . . . . . . . . . . 53
3.5.3 LES Stable Boundary Layer Wind Field . . . . . . . . . . . . 54
4 Results and Discussions 57
4.1 Lidar Measurement Coherence . . . . . . . . . . . . . . . . . . . . . . 57
4.2 Components of Measurement Coherence . . . . . . . . . . . . . . . . 60
4.3 Lidar Measurements of Evolving Wind Fields . . . . . . . . . . . . . 61
4.4 Measurements Using the Exponential Wind Evolution Model . . . . . 63
5 Conclusions 67
5.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.2 Perspectives (Future Outlook) . . . . . . . . . . . . . . . . . . . . . . 68
Bibliography 69

CHAPTER ONE

INTRODUCTION

1.1 Background and Motivation

“The fuel in the earth will be exhausted in a thousand or more years, and its mineral wealth, but man will find substitutes for these in the winds, the waves, the sun’s heat, and so forth.” True to those words of John Burroughs uttered over a century ago, man is really finding viable renewable energy alternatives in the wind. In fact, since the appearance of the first wind turbines at the end of the nineteenth century, wind energy has been considered to be a renewable energy source for not just developed countries but also developing countries as well. Thus, there has been a steady rise in wind energy production worldwide since 2004, with substantial actively installed capacity in Africa. In fact, in [5], it was noted that wind energy is the “fastest growing installed alternative-energy production”, with at least 20% of United States energy expected to be supplied by offshore and onshore wind farms by 2030.

Renewable energies constitute excellent solutions to both the increase of energy consumption and environment problems. Among these energies, wind energy is very interesting. Wind energy is the subject of advanced research. In the development of wind turbine, the design of its different structures is very important. It will ensure: the robustness of the system, the energy efficiency, the optimal cost and the high reliability. The use of advanced control technology and new technology products allows bringing the wind energy conversion system in its optimal operating mode. Different strategies of control can be applied on generators, systems relating to blades, etc. in order to extract maximal power from the wind [6].

To achieve this ambitious goal, the cost of wind energy must be able to compete favorably with the cost of traditional fossil fuels. This implies that the Levelized Cost Of Energy (LCOE) of wind energy defined as the average cost per unit of energy over the lifetime of the wind turbine, including capital costs, operations and maintenance costs, and all other relevant expenses [3], usually expressed in $/kWh, must be significantly reduced. A typical conventional approach to this will be to increase wind energy production, and so the trend within the last three decades has been the development and deployment of larger and more powerful wind turbines.

As at 2017, according to [7], Vestas V-164 rated at 9.5 MW is the most powerful wind turbine. The rationale behind this, no doubt is primarily to harness more wind resources and increase profitability, though some secondary reasons such as aesthetics, less visual impact to local people and environmental issues cannot be neglected. This approach in itself has not been without daunting challenges. To understand these challenges, consider briefly the design and operation of a wind turbine.

A cutaway-section of the dominant horizontal axis three-blade wind turbine is shown in figure 1.1 below:

Figure 1.1: A cut-away section of a horizontal axis three-blade wind turbine From the diagram above, the turbine consists of three main parts namely the blades, the nacelle and the tower. The blades which are fastened to the nacelle by means of the rotor are subjected to rotational motion by the wind. This rotational motion is amplified by the gearbox, high speed and low speed shafts in the nacelle which then rotates the generator (which is also in the nacelle), and the generator in turn produces Direct Current. The whole structure is supported by the tower.

This simple but powerful picture underscores the fact that wind turbines are highly dynamic systems that are excited by stochastic loads from the wind. These winds are generally not predictable or dependable as variations in this disturbance often occur. These variations can be medium-term that is changed during the space of a few hours or minutes cause variations in power output which must be accepted by the system to which the turbine is connected, or short-term – typically wind gusts which will introduce cyclic loadings which must be absorbed by the wind turbine with high susceptibility to fatigue damage. As shown above, these variations are managed by control systems on the nacelle such as the pitch system for the blades and the brake for generator control, as well as the yaw motor fitted in the tower.

Sadly, these controls rely on traditional feedback mechanism based on algorithms such as Artificial Neural Network (AAN) algorithms, which are only able to react to impacts of wind changes on the turbine dynamics after these impacts have already occurred, thereby impacting negatively on LCOE. The situation is best illustrated with an individual riding a bicycle with a blind fold, who only reacts when he/she must have crashed into something or had an impact.

Given this scenario, a natural question follows: What if the individual could ride the bicycle with his eyes open, thereby seeing the obstacle before hand and consequently avoiding the impact or crash? Hence, as a promising alternative, Light Detection and Ranging (Lidar) allows preview information about the approaching wind to be used to improve wind turbine control including blade pitch, generator torque, and yaw direction, thereby optimizing operational performance of the wind turbine through increase in energy yield, while keeping structural loads low [1].

1.2 The Problem

As stated clearly in the abstract as well as in the analysis of a typical wind turbine parts in the introduction above, the problem that light detecting and ranging devices hope to address, at least theoretically in the models and in systems that has begun to implement same is the erratic behavior of the approaching wind in front of the turbine and the feedback method of collecting data for optimization by the traditional models. As Eric Simley, Holger Fürst, Florian Haizmann and David Schlipf, in the article “Optimizing Lidars for Wind Turbine Control Applications–Results from the IEA Wind Task 32 Workshop” Remote Sens. 2018, 10, 863 noted clearly, these approach has proved to be ineffective in addressing the problem of control and design in the wind energy development. In their very words, due to the fact that wind turbines are highly dynamic systems that are excited by stochastic loads from the wind, variations in this disturbance usually medium-term (-changes during the space of a few hours or minutes cause variations in power output which must be accepted by the system to which the turbine is connected-) and short-term (typically wind gusts which will introduce cyclic loadings which must be absorbed by the wind turbine with high susceptibility to fatigue damage) negatively impacts heavily on Levelized Cost Of Energy (LCOE) of wind energy (i.e., the average cost per unit of energy over the lifetime of the turbine, including capital costs, operations and maintenance costs, and all other relevant expenses) [1]. As a result, there has been slow progress in the growth and development of more powerful turbines which have some significant advantages such as less visual impact to local people, and projects with more profitability. While traditional wind turbine control design utilize feedback control algorithms such as Artificial Neural Network (AAN) algorithms to address this challenge, this has often proved ineffective because they are only able to react to impacts of wind changes on the turbine dynamics after these impacts have already occurred [2]. Consequently, as a promising alternative, Light Detection and Ranging (Lidar) allows preview information about the approaching wind to be used to improve wind turbine control including blade pitch, generator torque, and yaw direction, thereby optimizing operational performance of the wind turbine through
increase in energy yield, while keeping structural loads low [1].

1.3 The Aim

The general purpose or the overall goal of this research is to carry out an exhaustive exposition of modeling associated with the Lidar systems as it applies to wind turbines particularly with the pulsed laser systems as its main components. As seen from the literature review in chapter two, this is not a novel area for environments where wind turbine technology is already a household term. It is one of those technological applications aimed at improving the performance of the wind turbine over time.

1.4 Objectives

To achieve the set goal/aim of this research work, we shall employ the methodology outlined in the third chapter of the project. Consequently, we shall therefore approach the discussions on Lidar systems as it applies to wind turbine technology by laying the physical foundations upon which the technology rests. We will traditionally start as expected, the classical mechanics approach by seeing the turbine as a rigid body in motion. This will entail an analysis of the structural mechanics and the cyclic loading of wind turbines.

Another physical approach to the analysis of the wind turbine is the fluid mechanical approach. In this regard, we shall go back to the model established by Navier-stokes equation for compressible Newtonian flow and the aero-elastic simulation.

In the main, the approaching wind before a turbine is essentially a compressible fluid hence the application of the relevant Navier-stokes equations. As is typical of the foundational mechanics of any physical systems, the discussions of the wind energy will obviously be incomplete without a quantum mechanical approach. The physics of the laser technology which is an essential part of the Lidar system is only clarified by the basic principles of quantum mechanics. The dynamic equation we shall employ here is the time dependent Schrodinger wave equation. The TDSE will enable us to elucidate the principles, the major components of the pulsed lasers operating at the nanoseconds regime for lidar systems. This will help to establish the correlation models and algorithms for the field reconstruction in view of formulating new insights into the laser scan patterns in the picoseconds and the femtoseconds
regime.

Thus, we will explore wind and wind turbine modeling through aero-elastic simulations, and then wind field reconstructions with correlations between Lidar systems and Wind turbines [4]. We will end with an insight into what is to be expected with regards to the lidar scanning pattern and consequently the entire lidar-wind turbine models and simulations when pulsed Lasers operating in the pico and/or femtosecond regime are used in the laser system as against the traditional nanoseconds pulsed lasers.

1.5 Significance

This thesis is expected to significantly contribute to the theoretical understanding of the correlation between wind turbines and lidar systems by providing invaluable insights into overcoming the barriers preventing the widespread use of Lidars for wind turbine control strategies for overcoming those barriers, and ideas for maximizing the effectiveness of Lidars for control applications. The significance of this research follows the main purpose of the International Energy Agency (IEA) wind task workshop

32 that was held in Boston, MA, USA in July 2016 [8]. Thus, the significance of this research follows the analysis of Eric Simley et al [1] who argued that: The workshop, ‘optimising Lidar designs for wind energy applications’ was held to identify Lidar system properties that are desirable for wind turbine control applications and help foster the widespread application of Lidar-assisted control (LAC).

Through multidisciplinary approach which is the modern trend of LAC, researches of this type will join the myriads of standard literatures in Journals to further overcome the barriers to the use of Lidar for wind turbine control such as optimization of lidar scan patterns by minimizing the error between the measurement and rotor effective wind speed. In addition, frequency domain methods for directly calculating measurement error using a stochastic wind field model. This process is applied to the optimization of several continuous waves and pulsed Doppler Lidar scan patterns. Also, the research intends to contribute to the design process for a Lidar-assisted pitch controller for rotor speed regulation. Again, using measurements from an optimized scan pattern shows that the rotor speed regulation obtained after optimizing the LAC scenario through time domain simulations matches the performance predicted by the theoretical frequency domain model [9].

1.6 Scope of Work

We will first focus on lidar system modeling with particular emphasis on the laser device which is the primary component of the lidar systems. Then we will explore wind and wind turbine modeling through aero-elastic simulations, and then wind field reconstructions with correlations between Lidar systems and Wind turbines [4].

Many of the literature in the Lidar applications to wind turbine technology as we shall see in chapter three focus on the results of the scan pattern of the pulsed lasers in the nanoseconds regime. Consequently, in the spirit of experimental extrapolation, we shall attempt theoretically and in the python codes to simulate lidar systems for Pico and Femto seconds pulsed lasers. We will end with an insight into what is to be expected with regards to the lidar scanning pattern and consequently the entire lidar-wind turbine models and simulations when pulsed Lasers operating in the pico and/or femtosecond regime are used in the laser system as against the traditional nanoseconds pulsed lasers.

1.7 Limitation of Work

This work will focus on three-bladed horizontal axis wind turbines operating in average wind speed areas. We will not attempt to consider other types of wind turbines neither will we attempt to consider wind turbines operating in low wind speed areas.

The available time and space are the necessary conditions that compel this limitation to the research work. To achieve results that are essentially measurable and susceptible to scientific tests using modern data analysis means, it is necessary to focus only on a particular aspect of the wind technology while leaving a huge corpus to further research and development. This is the trend we hope to approach the research work as it will be evident in the subsequent chapters of the work.