Tong W. Wind Power Generation and Wind Turbine Design

Подождите немного. Документ загружается.

PART III

DESIGN OF WIND TURBINE COMPONENTS

This page intentionally left blank

CHAPTER 13

Blade materials, testing methods

and structural design

Bent F. Sørensen , John W. Holmes , Povl Brøndsted & Kim Branner

Risø National Laboratory for Sustainable Energy

The Technical University of Denmark.

A major trend in wind energy is the development of larger wind turbines for

offshore wind farms. Since access to offshore wind turbines is difﬁ cult and costly,

it is of great importance that they operate safely and reliable. The wind turbine

rotor blades, which are the largest rotating component of a wind turbine, are

designed for an expected lifetime of 20 years. During this period of time, the

blades will be subjected to varying loads. Large wind turbine blades are made of

composite materials and can develop a number of interacting failure modes. High

structural reliability can be achieved by designing the blades against the develop-

ment of these failure modes. This chapter provides an overview of experimental

and modeling tools for the design of wind turbine blades, with particular emphasis

on evolution and interaction of various failure modes. This involves knowledge of

materials, testing methods and structural design.

1 Introduction

Structural design of wind turbine blades for horizontal axis wind turbines is a

complicated process that requires know-how of materials, modeling and testing

methods. A wind turbine blade must be designed against undesired aero-elastic

phenomena and failures for a great variety of aerodynamic load cases and environ-

mental conditions. Thus, the design process involves a number of different areas,

such as knowledge of the external loads originating from wind and gravity and

knowledge of the performance, the strength and the endurance of the full structure

and of the basic materials used. The goal of the design process is to ensure that

the wind turbine blade will function safely for its design life. The design lifetime of

modern wind turbines is normally thought to be 20 years, and the corresponding

418 Wind Power Generation and Wind Turbine Design

number of rotations (blade-tower passings) is of the order 10

to 10

, which is

approximately two orders of magnitude higher than the load cycles experienced

by composite materials used in other highly loaded structural applications such as

helicopter blades.

The main trends in the development of wind turbine blades are towards longer

and optimized blades; this is particularly the case for offshore wind turbines. The

weight of a large wind turbine blade also increases the loads on the rotor input

shaft and bearings as well as the wind turbine tower and mechanisms used to con-

trol yaw and pitch of the blades. Weight savings is therefore of great importance,

and signiﬁ cant efforts are devoted by wind turbine companies in the selection of

materials. To ensure that the blades can meet the required design life, the materials

must have high stiffness, be fatigue resistant, and be damage tolerant.

As for other low-weight-driven designs, the material considerations thus involve

the speciﬁ c stiffness (i.e. the stiffness divided by density), speciﬁ c strength

(strength divided by density) and speciﬁ c fatigue limit (fatigue limit stress divided

by density).

This chapter provides an overview of the interconnection between the blade

design process, material properties, materials testing and sub-component testing

and full-scale blade testing. As shown in Fig. 1 , the design of modern wind turbine

blades involves an understanding of material behavior and failure modes at many

length scales. This design process requires close collaboration between engineers

involved in modeling of aerodynamic loads, structural analysis and composite

materials and technicians responsible for the manufacturing process, quality con-

trol and on-site inspection and monitoring of blades. This chapter starts with a

review of manufacturing processing (Section 2), followed by a description of full-

scale blade testing including some results for blades tested to failure (Section 3)

that leads to a classiﬁ cation of common failure modes (Section 4). The material

properties that control the development of the various failure modes are presented

in Section 5. Section 6 describes experimental methods for determination of these

properties. Various examples of the use of the modern design methods that make

use of these strength-controlling material properties are given in Section 7. Finally,

Section 8 contains a discussion and closure.

2 Blade manufacture

The design of the wind turbine blade is a compromise between aerodynamic

and structural considerations. Aerodynamic considerations usually dominate the

design for the outer two-thirds of the blade, while structural considerations are

more important for the design of the inner one third of the blade.

2.1 Loads on wind turbine rotor blades

The rotor blade is loaded in a combination of ﬂ apwise and edgewise loads. Basi-

cally the blades are exposed to three different load sources. One is the wind load

that through the lift and drag on the aerodynamic proﬁ le loads the blade primarily

Blade Materials, Testing Methods and Structural Design 419

in bending ﬂ apwise. The second load source is the gravity varying edgewise from

tension/compression in leading edge and compression/tension in trailing edge. This

is the main reason for the edgewise fatigue bending of the blade. Finally, the blades

are exposed to centrifugal forces during the rotation. However, these longitudinal

loads are relatively low and often not taken into account in the design. Furthermore,

the design loads are divided into static loads and cyclic loads. International design

recommendations (e.g. IEC 61400-1 [ 1 ]) specify both types of loads. Moreover, the

blades will be subjected to a wide range of environmental conditions.

2.2 Blade construction

Modern wind turbine blades are structurally advanced constructions utilizing

composite laminates, sandwich core materials, gelcoat ﬁ lms and adhesive joints.

Matrix

Fiber

t ~ 100-500 m

h ~ 1-50 mm

H ~ 50-100 mm

L ~ 10-60 m

d ~ 5-150 m

R 10-50 nm

Face (laminate)

Core

Roughness scale

Lamina scale

Microscale

Laminate scale

Wing scale

~ 0,5-10 nm

Molecular scale

Sandwich scale

Figure 1: Modern blade design requires an understanding of how materials and

structures behave at various length scales, ranging from the molecular

scale (e.g. interfacial adhesion in adhesive joints) to the blade scale (e.g.

the dynamic coupling of the blade and tower). Changes made at any

length scale will affect the blade reliability. For instance, an increase in

the bond strength of the ﬁ ber/matrix interface will increase the materi-

als strength at the lamina scale, potentially leading to an increase in the

overall blade strength.

420 Wind Power Generation and Wind Turbine Design

Although there are a variety of wind turbine designs (reﬂ ecting different manufacturing

processes, material selection and design philosophy), the functionality of wind

turbine blades from a structural viewpoint can be understood by considering the

blade as a load-carrying beam (spar) enclosed by a shell. The primary purpose

of the shell is to give the blade an aerodynamic shape, creating the aerodynamic

forces that make the wind turbine blade to rotate and thus extracts energy from the

wind to make electrical power. The aerodynamic forces are transmitted to the wind

turbine hub through a load-carrying beam within the blade. The load-carrying

beam can be made as a box girder, sometimes called the main spar, or as laminates

in the aeroshell supported by webs. Figure 2 shows a sketch of the cross section of

a typical wind turbine blade. Steel bolts are present at the root, where the blade is

to be attached to the hub of a wind turbine.

Most blade manufacturing techniques involves making the blades in several

parts that are eventually joined by adhesive bonds. Figure 3 shows two common

design approaches. The one design involves the use of load-carrying laminates in

the aeroshells and webs for providing shear stiffness and buckling resistance. The

other design constitutes of two aeroshells bonded to a load-carrying box girder.

Sandwich structures are used extensively in the aeroshells and are also often used

in the webs (or correspondingly, in the sides of the box girder).

Leading edge

Main spar

(load carrying box)

Upwind side

Downwind side

To w a r d s ti p

Trailing edge

Aerodynamic

shell

Adhesive joint

Adhesive layer

Flange

(load-carrying laminate - tension)

Web

(sandwich)

Adhesive joint

Face

Sandwich panel

Core

Adhesive layer

Flange

(load-carrying laminate - compression)

Figure 2: Terminology and deﬁ nitions of coordinate system and structural parts of

a wind turbine blade [ 2 ].

Blade Materials, Testing Methods and Structural Design 421

2.3 Materials

Materials used for wind turbine blades must be low density and possess high

strength, fatigue resistance and damage tolerance. Large parts of the blades are

made of composite materials, i.e. materials that consist of more constituents, e.g.

long aligned ﬁ bers embedded in a continuous phase called the matrix material.

The shells that deﬁ ne the aerodynamic blade proﬁ le are typically constructed

using polymer matrix composites (PMCs) (e.g. glass ﬁ ber reinforced polyester)

and sandwich structures consisting of PMC face sheets and lightweight closed-cell

polymer foam or end-grain balsa wood cores. The box girder is constructed using

glass ﬁ ber composites or carbon ﬁ ber composites.

Concerning ﬁ bers , the majority of large turbine blades use E-glass laminates for

the aeroshells. Because blade mass is a critical design consideration, more expen-

sive carbon ﬁ ber composites are also being increasingly utilized. In addition to

much lower density (a factor of 2/3 lower than glass ﬁ bers), carbon ﬁ bers offer

several advantages for blade applications, including a much higher elastic modu-

lus, strength and fatigue life. Carbon ﬁ ber laminates and hybrid carbon/glass ﬁ ber

laminates are currently being utilized by some blade manufacturers. For the aeroshell,

carbon ﬁ bers can be used for selective reinforcement but offer great potential for

innovative blade designs. Although carbon ﬁ bers have attractive modulus, density

(a)

(b)

Figure 3: Two common design concepts of wind turbine blades manufactured as

two half-shells joined by adhesive bonds: (a) integrating the load-carrying

laminates in the aeroshell, which is supported by shear webs, and (b)

use of a box girder which is glued to the inside of the aeroshells.

422 Wind Power Generation and Wind Turbine Design

and strength, there are some drawbacks beyond higher cost. The compressive

strength can be lower than that of glass ﬁ ber composites. In addition, the compres-

sive strength of carbon ﬁ ber composites is very sensitive to the presence of manu-

facturing-related defects such as ﬁ ber misalignment and ﬁ ber waviness [ 3– 5 ]. For

certain regions of the blade, such as the aeroshell and internal sandwich structures,

wood or other natural materials which typically have a much lower density than

glass ﬁ ber composites can be used. For example, Vestas UK utilizes birch wood

which is laminated with carbon ﬁ bers to produce the aeroshell of large (40 m and

longer) wind turbine blades. Because of their low mass, these blades can be pro-

duced in thinner proﬁ les which further reduce rotating mass. Other materials,

including bamboo are also being developed as green alternatives for use in wind

turbine aeroshells and for sandwich structures within the blade [ 6 ].

The composite matrix is also an important consideration for wind turbine blades.

Currently, polyester is the most common choice as the matrix for glass ﬁ ber lami-

nates, but epoxy resins and vinylester are also used because of their superior

mechanical properties. Each matrix material has a set of cost, manufacturing and

mechanical behavior tradeoffs that must be considered. For example, in compari-

son with vinylester, the use of epoxy resins can increase the compressive strength

of glass ﬁ ber laminates by as much as 10–15%. For carbon ﬁ ber composites, epoxy

is the most common matrix material. Current research efforts are aimed at the

development of matrix materials with improved toughness, cure proﬁ les and the

development of improved low-viscosity infusion grades for permeating thick ﬁ ber

stacks. From an environmental viewpoint, the use of thermosetting polymers like

polyester and epoxy poses many problems and additional attention is being given

to the use of thermoplastics such as polybutylene teraphthalate (PBT). A key envi-

ronmental advantage of thermoplastics is their potential to be recycled at the end

of the turbine life cycle.

Since most wind turbine blades are bonded together, the adhesives used to join

blade segments will have a direct inﬂ uence on the reliability of the blade. The

adhesives utilized in blades are primarily epoxy, polyurethane and methacrylate-

based adhesives. The quality and mechanical properties (e.g. strength, creep,

defect tolerance) of blade adhesives are critical since they are used to bond very

large areas of the blade aeroshells and box girder and are subjected to complex

cyclic loading histories. Moreover, adhesives must maintain adequate bond

strength over a wide range of environmental conditions. As discussed later, the

defect tolerance and fracture resistance of adhesive joints is determined through

fracture mechanics mechanics-based testing.

As noted earlier, sandwich structures are widely used in the blade aeroshell

and shear webs in some spar designs. The core material is low-density materials,

primarily balsa wood and polymer foams.

Finally, the surface of wind turbine blades are painted with gelcoats to protect

the composite materials from damage originating from UV-radiation and to limit

the environmental exposure to the blade, e.g. humidity which may change

(decrease) the mechanical properties of the composite materials. The gelcoats

should maintain their adhesion to the underlying surface despite large changes in

Blade Materials, Testing Methods and Structural Design 423

humidity and temperature and should be sufﬁ ciently wear resistant to last for

20 years.

2.4 Processing methods

Wind turbine blades can be manufactured using procedures similar to that used for

composite aircraft structures and composite boat hulls; namely lamination by the

use of pre-pregs, hand lay-up and vacuum assisted resin transfer molding.

Both the manufacturing approaches shown in Fig. 3 involve the manufacturing

of the aeroshells as the ﬁ rst step. Since the aeroshells do not carry much load

themselves, they are primarily made of lightweight sandwich structures. The

ﬁ bers (or pre-pregs) and sandwich materials are usually placed by hand in an

open mould. The mould is closed before curing, so that fumes can be removed

during the processing. Next, in the second step, the moulds are opened and the

webs or box girder is placed in between the aeroshells. In the third step, the

moulds with the two half-shells are closed and the two half-shells and webs (or

box girder) are bonded together. Such processing methods have several advan-

tages. First, the quality of the aeroshells and webs (or box girder) can be con-

trolled before the ﬁ nal assemblage. Secondly, the adhesive layers allow a relatively

large dimensional tolerance, as the adhesive layer thickness can range from a

millimeter to several centimeters. This is particularly useful with increasing sizes

of the blade components.

3 Testing of wind turbine blades

3.1 Purpose

Full-scale testing ( Fig. 4 ) is mandatory for certiﬁ cation of large wind turbine

blades. The basic purpose of these blade tests is to demonstrate that the blade type

has the prescribed reliability with reference to speciﬁ c limit states with a reason-

able level of certainty. According to Det Norske Veritas (DNV) [ 7 ], a limit state is

a deﬁ ned as a state beyond which the structure no longer satisﬁ es the requirements.

The following categories of limit states are of relevance for structures: ultimate

limit state (ULS), fatigue limit state (FLS), and serviceability limit state (SLS).

The blade should be manufactured according to a certain set of speciﬁ cations in

order to ensure that the test blade is representative of the whole series of blades.

In other words, the purpose of the blade tests is to verify that the speciﬁ ed limit

states are not reached and that the type of blade possesses the projected strength

and lifetime.

Normally, the full-scale tests used for certiﬁ cation are performed on a very lim-

ited number of samples; only one or two blades of a given design are tested so that

no statistical distribution of production blade strength can be obtained. Therefore,

although the tests do give information valid for the blade type, they cannot replace

either a rigorous design process or the use of a quality control system for blade

production.

424 Wind Power Generation and Wind Turbine Design

Additionally, tests can be used to determine blade properties in order to validate

some vital design assumptions used as inputs for the design load calculations.

Finally, full-scale tests give valuable information to the designers on how the

structure behaves in the test situation and which structural details that are important

and should be included in the structural models for design. Especially, valuable

information is obtained if the blade is tested to failure.

3.2 Certiﬁ cation tests (static and cyclic)

According to DNV [ 7 ], it is required that the test program for a blade type shall be

composed of at least the following tests in this order:

Mass, centre of gravity, stiffness distribution and natural frequencies •

Static tests •

Fatigue load tests •

Post fatigue static tests •

Figure 4: Full-scale testing: a 61.5-m long wind turbine blade subjected to static

test (ﬂ apwise direction) [Courtesy of LM Glasﬁ ber A/S].