FEA of Wind Turbine Tower

1
FEA of Wind Turbine Tower

• Martin Knecht

2
The Problem

• FEA modeling of a wind turbine tower.
• Analysis
• Stress
• Deflection
• Want to prevent
• Yielding
• Excessive deflection

3
Tower Construction

• Steel pipe
• Diameter 1.850 in.
• Thickness 0.225 in.
• Length 15 ft.
• Base
• Pipe screewed into 5 flange.
• Flange welded to 30 square, ¼ in. thick steel
  plate.

4
Tower Construction

• Bracket
• Houses the permanent magnet alternator (PMA)
• Six carbon fiber blades attached to PMA
• Total weight of PMA, blades and bracket 22.25 lb

5
Tower Construction

• Anchoring
• Four ¼ inch steel cables
• Breaking strength 7000 lb
• Attached 10 feet above the base of the tower (can
  not be attached higher due to blades)
• Anchored 10 feet away from the base on the roof.

6
Tower Construction

• Purpose for FEA is safety
• Constructed around expensive equipment
• Solar panels
• Weather monitoring station
• Constructed near edge of roof.

7
Loading and Conditions of Operation

• FEA modeled for high winds (60 and 100 mph)
• Constraints
• Rigid constraint at base
• Cables (ropes)
• Loads
• Weight of PMA, blades and bracket 22.25 lb
• PMA torque
• Wind drag

8
Loads

• PMA torque
• Torque due to PMA is negligible.
• With 33 mph winds it produces 7.8e-4 lb-ft
• PMA stops generating power at high wind speeds as
  a safety factor
• Not included in FEA
• Wind drag
• Determined using the relationship
• Drag ½ Cdrv2A
• air density r 1.2 kg/m3
• Wind speed v 60 and 100 mph
• Cross sectional area A 0.258 m2
• Coefficient of drag Cd 1
• Drag at 60 mph 25 lb
• Drag at 100 mph 70 lb

9
The FEA Model

• Pipe mesh
• Split into three sections in order to mesh.
• Each meshed separately
• Needed to be aware of concentrated stresses at
  pipe boundaries.
• Fixed constraint on bottom of flange
• Rope constraints to model cables

1 ft
10
The FEA Model

• Loads applied to bracket at the top of pole.
• Total body forces of 25 or 70 lb in x-direction
  to model the wind drag
• Downward total body force of 22.3 lb to model the
  weigh of the turbine
• The weight should be offset from the bracket to
  more realistically model the distribution of mass
  of the turbine

11
The FEA Model

• Meshing
• Would not mesh using h-adaptively if entire model
  was run at once.
• Each pipe meshed separately and then assembled.
• Needed smaller mesh size at pipe boundaries,
  bracket-pipe boundary and flange-pipe boundary.
• Used loads with h-adaptively to force mesh sizes
  to be smaller where desired.

12
Results

• Deformations
• 60 mph winds
• Max. displacement 2.36 in
• Location Top of pole
• 100 mph wins
• Max. displacement 6.6 in
• Location Top of pole
• Note Large deformations caused non-linear
  displacements of material.

13
Results

• Stresses at pipe-flange boundary
• 60 mph winds
• Stress 1.2e8 Pa
• 100 mph winds
• Stress 3.0e8 Pa
• Yield stress of steel 3.3e8 Pa
• At winds around 100 mph the structure will fail!

14
Improvements on Tower Model

• In order to reduce concentrated stresses and
  excessive deflections, the flange can be replaced
  by a spherical joint.
• Model was run using identical loads.

15
Improvements on Tower Model

• Deformations
• 60 mph winds
• Max. displacement 0.25 in
• Location Center of pole
• 100 mph wins
• Max. displacement 0.71 in
• Location Center of pole

16
Improvements on Tower Model

• Stresses are distributed throughout entire pole
  with the joint
• Stresses at pipe-bracket boundary
• 60 mph winds
• Stress 4.9e7 Pa
• Stresses reduced by a factor of 2.5
• 100 mph winds
• Stress 1.4e8 Pa
• Stresses reduces by a factor of 2
• Yield stress of steel 3.3e8 Pa

17
If I had more time (and computing power)

• More accurately distribute the weight of the
  turbine at the top of the tower.
• Model stresses on tower at different wind
  directions.
• Determine cable and anchor stresses.
• Consider dynamic loading modeling wind gusts.
• Determine vibrational resonances.