Burton T. (et. al.) Wind energy Handbook

Подождите немного. Документ загружается.

7.10.1 Slab foundations

Slab foundations are chosen when competent material exists within a few metres of

the surface. The overturning moment is resisted by an eccentric reaction to the

weight of the turbine, tower, foundation and overburden (allowing for buoyancy, if

the water table can rise above the base of the slab). The eccentricity of the reaction,

and hence the magnitude of the restoring moment, is limited by the load carrying

capacity of the sub-strata, which determines the width of the area at the edge of the

slab required to carry the gravity loads. Brinch Hansen (1970) provides straightfor-

ward rules for calculating the slab bearing capacity under these conditions, based

on the simplifying assumption of uniform loading over the loaded area. However,

if the substrata behave elastically, tilting of the slab foundation is likely to result in

a linear distribution of bearing stress over the loaded area, so an alternative

approach is to base the design on the maximum rather than the average value. The

GL rules additionally require that positive bearing stress exists over the whole

width of the foundation when the turbine is opera ting, which limits the maximum

overturning moment under these conditions to WB/6, where W is the gravity load

and B is the slab width. This requirement can add signiﬁcantly to the required

foundation size.

Four alternative slab foundation arrangements are shown in Figure 7.44. Figure

7.44(a) shows a slab of uniform thickness, with its upper surface just above ground

level, which is chosen when bedrock is near the ground surface. The main

reinforcement consists of top and bottom mats to resist slab bending and the slab is

made thick enough for shear reinforcement not to be required. The second variant

shown in Figure 7.44(b) is a slab surmounted by a pedestal. This is used when the

(a)

(b)

(c)

(d)

Figure 7.44 (a) Plain Slab; (b) Stub and Pedestal; (c) Stub Tower Embedded in Tapered Slab;

(d) Slab Held Down by Rock Anchors

FOUNDATIONS 465

bedrock is at a greater depth than the slab thickness required to resist the slab

bending moments and shear loads. The gravity load on the substrata is increased

by virtue of the overburden, so the overall slab plan dimensions can be reduced

somewhat.

The third variant, shown in Figure 7.44(c) is similar to the second, but embodies

two possible modiﬁcations which can be applied independently: replacement of the

pedestal by a stub tower embedded in the slab, and introduction of a tapering slab

depth. The stub tower has to be perforated near the top of the slab to allow radial

top face reinforcement to pass through it, and reinforcement to resist punching

shear loads from the tower stub bottom ﬂange must be incorporated. Tapering the

slab depth has the merit of saving material, but is difﬁcult to execute.

Rock anchors eliminate the need to add weight to a gravity foundation for

counterbalance purposes, and thus enable the foundation size to be signiﬁcantly

reduced, provided bearing capacities are sufﬁciently high (see Figure 7.44(d)).

Specialist contr actors are needed for rock anchor installation, so they only ﬁnd

occasional use.

The ideal shape of gravity foundation in plan is a circle, but in view of the

complications of providing circular formwork, an octagonal shape is usually chosen

instead. Sometimes slabs are square in plan to simplify the shuttering and reinforce-

ment further.

7.10.2 Multi-pile foundations

In weaker ground, a piled foundation often makes mo re efﬁcient use of materials

than a slab. Figure 7.45(a) illustrates a foundation consisting of a pile cap resting on

eight cylindrical piles arranged in a circle. Overturning is resisted by both pile

vertical and lateral loads, the latter being generated by moments applied to the

head of each pile. Consequently the reinforcement must be arranged to provide full

(a)

(b)

(c)

Figure 7.45 (a) Pile Group and Cap; (b) Solid Mono-pile; (c) Hollow Mono-pile

466

COMPONENT DESIGN

moment continuity between the piles and the pile cap. Holes for the piles can be

auger drilled and the piles cast in situ after the positioning of the reinforcement

cage.

7.10.3 Concrete mono-pile foundations

A concrete mono-pile foundation consists of a single large diameter concrete

cylinder, which resists overturning by mobilizing soil lateral loads alone (see Figure

7.45(b)). These lateral loads can be calculated conservatively for sand by using

either simple Rankine theory for passive pressures on retaining walls, which

ignores soil/wall friction, or Coulomb theory, which includes it. However, in the

case of a mono-pile, friction on the sides of the soil wedge notionally displaced

when the pile begins to tilt provides further resistance, and this is accounted for in

the solution due to Brinch Hansen (1961).

This type of foundation is an attractive option when the water table is low and

the soil properties enable a deep hole to be excavated from above without the sides

caving in. However, while simple, the concept is relatively expensive in terms of

materials. The hollow cylinder variant illustrated in Figure 7.45(c) uses materials

much less extravagantly by replacing the concrete in the body of the cylinder,

which has no structural role to play, with ﬁll.

Figure 7.46 Piled Foundation for Steel Lattice Tower

FOUNDATIONS 467

7.10.4 Foundations for steel lattice towers

The legs of steel lattice towers are relatively widely spaced, and lend themselves to

separate fo undations. Bored cast in situ piles are commonly used (see Figure 7.46).

The mechanism for resisting overturning is simply uplift and downthrust on the

piles, but the piles must also be designed for the bending moments induced by the

horizontal she ar load. Pile uplift is resisted by friction on the surface of the piles,

which depends on both the soil/pile friction angle and the lateral soil pre ssure.

Considerable uncertainty surrounds the magnitude of these quantities, so Eurocode

7 recommends the use of pile testing to establish pile capacity.

The angle sections forming the base of the tower legs are cast in place when the

concrete for the piles is poured. A framework is assembled in advance, incorporat-

ing the leg base sections, so that the legs can be set at the correct spacing and

inclination before concreting.

References

American Gear Manufacturers Association/American Wind Energy Association, (1996).

‘Recommended practices for design and speciﬁcation of gearboxes for wind turbine generator

systems’. AGMA/AWEA 921-A97.

American National Standards Institute/American Gear Manufacturers Association, (1995).

‘Fundamental rating factors and calculation methods for involute spur and helical gear teeth’.

ANSI/AGMA 2001-C95.

Anderson, C. G. et al., (1993). ‘Yaw system loads of HAWTS’. ETSU Report No W/42/00195/

REP. Energy Technology Support Unit, Harwell, UK.

Anderson, C. G., Heerkes, H. and Yemm, R., (1998). ‘Prevention of edgewise vibration on

large stall-regulated blades’. Proceedings of the BWEA Conference, pp 95–102.

Barbero, E. J., (1998). Introduction to composite materials design. Taylor and Francis, Philadel-

phia, USA.

Bond, I. P. and Ansell, M. P., (1998). ‘Fatigue properties of jointed wood composites. Part I:

Statistical analysis, fatigue master curves and constant life diagrams’. J. Mat. Sci., 33,

2751–2762, and ‘Part II: Life prediction: analysis for variable amplitude loading’. J. Mat.

Sci., 33, 4121–4129.

Bonﬁeld, P. W. and Ansell, M. P., (1991). ‘Fatigue properties of wood in tension, compression

and shear’. J. Mat. Sci., 26, 4765–4773.

Bonﬁeld, P. W. et al., (1992). ‘Fatigue testing of wood composites for aerogenerator blades.

Part VII: Alternative wood species and joints’. Proceedings of the BWEA Conference,

pp 243–249.

Brinch Hansen, J., (1961). ‘The ultimate resistance of rigid piles against transverse forces’.

Danish Geotechnical Institute Report No. 12.

Brinch Hansen, J., (1970). ‘A revised and extended formula for bearing capacity’. Danish

Geotechnical Institute Bulletin No. 28.

British Standards Institution, (1986). ‘BS436: Spur and helical gears – Part 3 Method for

calculation of contact and root bending stress limitations for metallic involute gears’.

Corbet, D. C., (1991). ‘Investigation of materials and manufacturing methods for wind turbine

blades’. ETSU Report No. W/44/00261. Energy Technology Support Unit, Harwell, UK.

Corbet, D. C., Brown, C. and Jamieson, P., (1993). ‘The selection and cost of brakes for

468

COMPONENT DESIGN

horizontal axis stall-regulated wind turbines’. ETSU Report No. WN 6065. Energy Technol-

ogy Support Unit, Harwell, UK.

Echtermeyer, A. T., Hayman, E. and Ronold, K. O., (1996). ‘Comparison of fatigue curves for

glass composite laminates.’ Design of Composite Structures against Fatigue, (Ed. R. M.

Mayer). Mechanical Engineering Publications, Bury St Edmunds, UK.

European Convention for Constructional Steelwork, (1988). Recommendations on buckling of

shells.

Freris, L. L. (ed) (1990). Wind energy conversion systems, Prentice Hall.

Fuglsang, P. L. and Madsen, H. A., (1995). ‘A design study of a 1 MW stall-regulated

rotor’. Riso National Laboratory Report No. R-799. Riso National Laboratory, Roskilde,

Denmark.

Hancock, M. and Bond, I. P., (1995). ‘The new generation of wood composite wind turbine

rotor blades – design and veriﬁcation’. Proceedings of the BWEA Conference, pp 47–52.

Hancock, M. Sonderby, O. and Schubert, M. A., (1997). ‘Design, development and testing of a

31 m wood composite stall regulated blade for serial production’. Proceedings of the EWE

Conference, pp 206–212.

Heier, S., (1998). Grid integration of wind energy conversion systems. John Wiley & Sons,

Chichester, UK.

Hindmarsh, J., (1984). Electrical machines and their application. Pergamon Press, Oxford, UK.

ck, M., (1983). ‘Calculation of S/N curves for Steel, Cast Steel and Cast Iron – Synthetic

S/N curves’. Verein Deutsher Eisenhu

ttenleute Report No. ABF 11. Verlag Stahleisen,

sseldorf, Germany.

Jamieson, P. and Brown, C. J., (1992). ‘The optimization of stall-regulated rotor design’.

Proceedings of the BWEA Conference, pp 79–84.

Jones, R. and Smith, G. A., (1993). ‘High quality mains power from variable-speed wind

turbines’. IEE Conference, Renewable Energy – Clean Power 2001.

Joose, P. A. and van Delft, D. R. V., (1996). ‘Has fatigue become a wearisome subject? –

Overview of 12 years’ of materials research in the Netherlands’. Proceedings of the EUWEC

Conference, pp 902–906.

Krause, P. C., (1986). Analysis of electric machinery. McGraw-Hill, New York, USA.

Mayer, R. M., (1996). Design of composite structures against fatigue. Mechanical Engineering

Publications, Bury St Edmunds, UK.

McPherson, G., (1990). An introduction to electrical machines and transformers. John Wiley &

Sons, New York, USA.

Mohan, N., Undeland, T. M. and Williams, W. P., (1995). Power electronics, converters

application and design, Second Edition. John Wiley & Sons, New York, USA.

Petersen, J. T. et al., (1998). ‘Prediction of dynamic loads and induced vibrations in stall’. Riso

National Laboratory Report No. R-1045. Riso National Laboratory, Roskilde, Denmark.

Pena, R., Clare, J. C. and Asher, G. M., (1996). ‘Doubly fed induction generator using back-to-

back PWM converters and its application to variable speed wind-energy generators’. IEE

Proceedings Electric Power Applications, 143, pp 231–241.

Schmidt, H. and Neuper, M., (1997). ‘Zum elastostatischen Tragverhalten exzentrisch gezo-

gener L-Sto

ße mit vorgespannten Scrauben’ (‘On the elastostatic behaviour of an eccen-

trically tensioned L-joint with prestressed bolts’), Stahlbau, 66, 163–168.

Schmidt, H., Winterstetter, T. A. and Kramer, M., (1999). ‘Non-linear elastic behaviour of

imperfect, eccentrically tensioned L-ﬂange ring joints with prestressed bolts as basis for

fatigue design’. Proceedings of the European Conference On Computational Mechanics .

Thomsen, K., (1998). ‘The statistical variation of wind turbine fatigue loads.’ Riso National

Laboratory Report No. R-1063. Riso National Laboratory, Roskilde, Denmark.

Timoshenko, S. P. and Gere, J. M., (1961). Theory of Elastic Stability, Second Edition. McGraw-

Hill, New York, USA.

REFERENCES 469

Van Delft, D. R. V., de Winkel, G. D. and Joose, P. A., (1996). ‘Fatigue behaviour of ﬁbreglass

wind turbine blade material under variable amplitude loading’. Proceedings of the EUWEC

Conference, pp 914–918.

Verein Deutscher Ingenieure, (1986/1988). ‘VDI 2230 Part 1: Systematic calculation of high

duty bolted joints – Joints with one cylindrical bolt’.

Warneke, O., (1984). ‘Use of a double-fed induction machine in the Growian large wind

energy converter’. Siemens Power Engineering, VI, 1, pp 56–59.

Wilson, R. A., (1990). ‘Implementation and optimization of mechanical brakes and safety

systems’. Proceedings of a DEn/BWEA Workshop on ‘Mechanical systems for wind turbines’.

470

COMPONENT DESIGN

The Controller

In the most general sense, the wind turbine control system consists of a number of

sensors, a number of actuators, and a system consisting of hardware and software

which processes the input signals from the sensors and generates output signals for

the actuators. The sensors might include, for example:

• an anemometer,

• a wind vane,

• at least one rotor speed sensor,

• an electrical power sensor,

• a pitch position sensor,

• various limit switches,

• vibration sensors,

• temperature and oil level ind icators,

• hydraulic pressure sensors,

• operator switches, push buttons, etc.

The actuators might include a hydraulic or electric pitch actuator, sometimes a

generator torque controller, generator contactors, switches for activating shaft

brakes, yaw motors, etc.

The system that processes the inputs to generate outputs usually consists of a

computer or microprocessor-based controller which carries out the normal control

functions needed to operate the turbine, supplemented by a highly reliable hard-

wired safety system. The safety syst em must be capable of overriding the normal

controller in order to bring the turbine to a safe state if a serious problem occurs.

8.1 Functions of the Wind-turbine Controller

8.1.1 Supervisory control

Supervisory control can be considered as the means whereby the turbine is brought

from one operational state to another. The operational states might, for example,

include:

• stand-by, when the turbine is available to run if external conditions permit,

• start-up,

• power production,

• shutdown, and

• stopped with fault.

It is possible to envisage other states, or it may be useful to further subdivide some

of these states. As well as deciding when to initiate a switch from one state to

another, the supervisory controller will carry out the sequence control required. As

an example, the sequence control for start-up of a ﬁxed-speed pitch-regulated wind

turbine might consist of the following steps:

• power-up the pitch actuator;

• release the shaft brake;

• ramp the pitch position demand at a ﬁxed rate to some starting pitch;

• wait until the rotor speed exceeds a certain small value;

• engage the closed loop pitch control of speed;

• ramp the speed demand up to synchronous speed;

• wait until the speed has been cl ose to the target speed for a speciﬁed time;

• close the generator contactors;

• engage the closed loop pitch control of power; and

• ramp the power demand up to the rated level.

The supervisory controller must check that each stage is successfully completed before

moving on to the next. If any stage is not completed within a certain time, or if any

faults are detected, the supervisory controller should change to shut-down mode.

8.1.2 Closed-loop control

The closed-loop controller is usually a software-based system that automatically

adjusts the operational state of the turbine in orde r to keep it on some pre-deﬁned

operating curve or characteristic. Some examples of such control loops are:

472 THE CONTROLLER

• control of blade pitch in order to regulate the power output of the turbine to the

rated level in above-rated wind speeds;

• control of blade pitch in order to follow a predetermined speed ramp during

start-up or shut-down of the turbine;

• control of generator torque in order to regulate the rotational speed of a variable-

speed turbine;

• control of yaw motors in order to minimize the yaw tracking error.

Some of these control loops may require very fast responses in order to prevent

the turbine wanderi ng far from its correct operating cu rve. Such controllers may

need to be designed very carefully if good performance is to be achieved without

detrimental effects on other aspects of the turbine’s operation. Others, such as yaw

control, are typically rather slow acting, and careful design is then much less

critical.

This chapter examines the main issues behind closed-loop controller design, and

presents some of the techniques that can be used to effect a successful design.

8.1.3 The safety system

It is helpful to consider the safety system as quite distinct from the main or ‘normal’

control system of the turbine. Its function is to bring the turbine to a safe condition

in the event of a serious or potentially serious problem. This usually means bringing

the turbine to rest with the brakes applied.

The normal wind-turbine supervisory controller should be capable of starting

and stopping the turbine safely in all foreseeable ‘normal’ conditions, including

extreme winds, loss of the electrical network, and most fault conditions which are

detected by the controller. The safety system acts as a back-up to the main control

system, and takes over if the main system appears to be failing to do this. It may

also be activated by an operator-controlled emergency stop button.

Thus the safety system must be independent from the main control system as far

as possible, and must be designed to be fail-safe and highly reliable. Rather than

utilizing any form of computer or micro-processor based logic, the safety system

would normally consist of a hard-wired fail-safe circuit linking a number of

normally open relay contacts that are held closed when all is healthy. Then if any

one of those contacts is lost, the safety system trips, causing the appropriate fail -safe

actions to operate. This might include disconnecting all electrical systems from the

supply, allowing fail-safe pitching to the feather position, and allowing the spring-

applied shaft brake to come on.

The safety system might, for example, be tripped by any one of the following:

• rotor overspeed, i.e., reaching the hardware overspeed limit – this is set higher

than the software overspeed limit which would cause the normal supervisory

FUNCTIONS OF THE WIND-TURBINE CONTROLLER 473

controller to initiate a shut-down (see Figure 8.1 for typical arrangement of rotor

speed sensing equipment on low-speed shaft);

• vibration sensor trip, which might indicate that a major structural failure has

occurred;

• controller watchdog timer expired: the controller should have a watchdog tim er

which it resets every controller timestep – if it is not reset within this time, this

indicates that the controller is faulty and the safety system should shut down the

turbine;

• emergency stop button pressed by an operator;

• other faults indicating that the main controller might not be able to control the

turbine.

Figure 8.1 Low-speed Shaft Speed Sensing System. (Three proximity sensors mounted on a

bracket attached to the front of the (integrated) gearbox register the passage of the teeth on

the shaft circumference, and provide independent speed signals for the control and safety

systems. The ﬂange onto which the hub is bolted is immediately to the left of the teeth).

474

THE CONTROLLER