Dieser Leitfaden ist derzeit auf Englisch verfügbar.
Technology / Technical guide
Wind Turbine Fatigue Cycles: Range, Mean & Counting
A changing load history contains more information than its largest value. Fatigue descriptions distinguish cycle range, mean and occurrence count. Counting those cycles is one step; relating them to material damage is another.
A cycle spectrum is a description, not a lifetime by itself.
The loading quantity, counting method and material model must remain separate. Duration and model assumptions matter when extending a recorded history beyond its own boundary.
01 / Cycle measures
Range and amplitude differ by a factor of two.
For a simple cycle with maximum σ_max and minimum σ_min, range is σ_max − σ_min, amplitude is half that range, and mean is half their sum.
For a simple cycle with maximum σ_max and minimum σ_min, range is σ_max − σ_min, amplitude is half that range, and mean is half their sum. State the sign and stress component used.
A hypothetical cycle from −4 to +12 MPa has range 16 MPa, amplitude 8 MPa and mean 4 MPa. These are invented values; they establish no blade stress or allowable fatigue condition.
02 / Irregular history
Reversals do not all describe equal cycles.
An irregular sequence can contain nested excursions of different sizes.
An irregular sequence can contain nested excursions of different sizes. Rainflow counting provides a defined way to describe cycles by range, mean and count rather than treating each sampled peak as an identical event.
MathWorks documents these distinct output quantities. Sampling, filtering and treatment of incomplete histories affect the represented record. A count of cycles must not be confused with a count of data samples or rotor revolutions.
03 / Material relation
A load spectrum needs a compatible damage model.
A stress-life relation links a specified stress condition to a represented number of cycles under stated material and test conditions.
A stress-life relation links a specified stress condition to a represented number of cycles under stated material and test conditions. Direction, mean stress, environment and the definition of failure can influence that relation.
A curve for one material or loading type is not a universal rotor-life curve. A force spectrum also requires a justified transformation before it can be treated as a local material-stress spectrum.
04 / Damage sum
A linear index contains strong assumptions.
In a simple Palmgren–Miner model, D = Σn_i/N_i.
In a simple Palmgren–Miner model, D = Σn_i/N_i. Here n_i is counted occurrence in a group and N_i the corresponding modelled cycle capacity under its specified condition. This is a linear cumulative model, not a direct measurement of remaining life.
For an invented two-group example, 50/500 plus 100/2,000 gives D = 0.15. The denominators are hypothetical. That number does not establish 85% remaining life or authorise operation of a component.
05 / History effects
The count does not retain every causal feature.
A cycle histogram compresses a time history.
A cycle histogram compresses a time history. Some sequence information, changing states and interactions can be lost in that compression. If the material’s response depends on those features, a linear grouped sum may not represent the full behaviour.
Likewise, projecting a short history into years adds assumptions about future conditions and occurrence. The observation duration and representativeness are distinct from the mathematical counting operation.
06 / Scope
Composite fatigue needs its relevant description.
Different material directions and failure mechanisms can require different fatigue interpretations.
Different material directions and failure mechanisms can require different fatigue interpretations. An apparently modest peak relative to a static strength is not a complete assessment of repeated loading.
No MARSTON fatigue curve, damage threshold or lifespan is supplied here. The stress guide explains the quantity boundary; this page describes cycles and model logic without a test or service procedure.
Continue the review with Wind Blade Composite Stress, Strain & Directionality.
Continue the review with Wind Blade Mode Shapes & Natural Frequencies.
Continue the review with Wind Turbine Tower Disturbance & Cyclic Blade Loading.
Extend the cycle-counting review with mean stress correction in fatigue.
General background: MathWorks: Cycle counts, ranges and means · MathWorks: Stress-life and cumulative-damage model concepts. These references explain industry concepts; they are separate from the applicable MARSTON product and project documents.
Key concepts and conditions.
Read each quantity with its reference, units and stated conditions. A general relation does not establish a model-specific value.
| Concept / quantity | Meaning / relation | Conditions / limits |
|---|---|---|
| Cycle range | σ_max − σ_min | Not the same as amplitude. |
| Amplitude | Range/2 | Retain the sign/component convention. |
| Mean | (σ_max + σ_min)/2 | Different from cycle magnitude alone. |
| Rainflow count | Occurrence by range/mean | Not sample count or automatic rotor-revolution count. |
| D = Σn_i/N_i | Linear cumulative model index | Not a measured percentage of life remaining. |
Engineering questions.
Is range equal to amplitude?
No. For the stated simple cycle, amplitude is half the range.
Does each rotor revolution count as one fatigue cycle?
Not universally; the load history and component quantity define the cycles.
Can a peak value describe fatigue?
No. Range, mean, counts and material response also matter.
Does D = 0.15 prove 85% remaining life?
No. It is an illustrative model index with hypothetical denominators.
Can another material’s curve determine blade life?
Not without an applicable material and loading basis.
Does rainflow alone predict failure?
No. It counts a defined history; damage interpretation requires separate assumptions.
Technical enquiry
Discuss the technical question.
A changing load history contains more information than its largest value. Share the model and applicable operating or design information for an engineering discussion.