Rainflow

rainflow is a C++ implementation of the ASTM E1049-85 rain flow cycle counting algorithm for fatigue analysis. It is based on Python library rainflow

The implementation consists of a single header rainflow.h and source file rainflow.cpp and has zero dependencies.

Usage

Function RainFlow::count_cycles returns a map of ranges and the corresponding number of cycles.

It takes two arguments:

• RainFlow::Series series - an input vector of samples to process,
• double binSize - (optional) specifies the width of each cycle-counting bin

Returns a RainFlow::Counts map.

Example:

#include "rainflow.h"

int main()
{
    RainFlow::Series series { -2, 1, -3, 5, -1, 3, -4, 4, -2 };
    auto counts = RainFlow::count_cycles( series, 2.0 );
    /* counts:
        { 2, 0.0 },
        { 4, 2.0 },
        { 6, 0.5 },
        { 8, 1.0 },
        { 10, 0.5 }
    */
}

It is possible to obtain full information about each cycle using an extract_cycles function.

It takes a single argument:

• RainFlow::Series series - an input vector of samples to process

Returns a RainFlow::Cycles vector of RainFlow::Cycle structs. RainFlow::Cycle struct has fields: range, mean, count, start_index and end_index.

Example:

#include "rainflow.h"

int main()
{
    RainFlow::Series series { -2, 1, -3, 5, -1, 3, -4, 4, -2 };
    auto cycles = RainFlow::extract_cycles( series );
    /* cycles:
        { 3, -0.5, 0.5, 0, 1 },
        { 4, -1.0, 0.5, 1, 2 },
        { 4, 1.0, 1.0, 4, 5 },
        { 8, 1.0, 0.5, 2, 3 },
        { 9, 0.5, 0.5, 3, 6 },
        { 8, 0.0, 0.5, 6, 7 },
        { 6, 1.0, 0.5, 7, 8 }
    */
}

Running example

Build and run the example using docker:

docker build -t rainflow .
docker run --rm rainflow

or locally:

cmake -S . -B build
cmake --build build
build/example_rainflow example/samples.txt

Running tests

Build and run unit tests using docker:

docker build -t rainflow .
docker run --rm rainflow build/test_rainflow

or locally:

cmake -S . -B build
cmake --build build
build/test_rainflow

To run tests locally the GoogleTest library is required. Under linux it can be installed via package manager eg.:

apt-get install -y libgtest-dev