#solar #algorithm #sunrise #azimuth #zenith

no-std spa

The Solar Position Algorithm module (SPA) for Rust calculates the sunrise-sunset and azimuth and zenith-angle for specific geo-position and time (UTC); for example for solar-panel-alignment or automotive

8 releases (4 breaking)

0.5.1 Feb 11, 2024
0.5.0 Feb 11, 2024
0.4.0 Jun 11, 2023
0.3.1 Apr 18, 2023
0.1.1 Nov 19, 2017

#37 in Date and time

Download history 682/week @ 2023-12-22 841/week @ 2023-12-29 1057/week @ 2024-01-05 798/week @ 2024-01-12 2659/week @ 2024-01-19 1794/week @ 2024-01-26 1793/week @ 2024-02-02 954/week @ 2024-02-09 811/week @ 2024-02-16 729/week @ 2024-02-23 671/week @ 2024-03-01 655/week @ 2024-03-08 665/week @ 2024-03-15 854/week @ 2024-03-22 727/week @ 2024-03-29 1014/week @ 2024-04-05

3,344 downloads per month
Used in 2 crates

Apache-2.0

125KB
361 lines

Solar Position Algorithm (SPA)

The Solar Position Algorithm module (SPA) for Rust calculates the sunrise-sunset and azimuth and zenith-angle for specific geo-position and time (UTC); for example for solar-panel-alignment or automotive.

Calculating the Sunrise and Sunset

The following function is calculating the sunrise-sunset at geo-pos lat/lon at time t (UTC).

The algorithm has been ported to Rust from http://lexikon.astronomie.info/zeitgleichung/neu.html Its accuracy is in the range of a few minutes.

Arguments

  • utc - UTC time-point (DateTime)
  • lat - latitude in WGS84 system, ranging from -90.0 to 90.0.
  • lon - longitude in WGS84 system, ranging from -180.0 to 180.0

The function returns a result of type SunriseAndSet.

pub enum SunriseAndSet {
    PolarNight,
    PolarDay,
    Daylight(DateTime<Utc>, DateTime<Utc>),
}

The polar night occurs in the northernmost and southernmost regions of the Earth when the night lasts for more than 24 hours. This occurs only inside the polar circles. The opposite phenomenon, the polar day, or midnight sun, occurs when the Sun stays above the horizon for more than 24 hours. "Night" is understood as the center of the Sun being below a free horizon, represented by the variants SunriseAndSet::PolarNight or SunriseAndSet::PolarDay.

Since the atmosphere bends the rays of the Sun, the polar day is longer than the polar night, and the area that is affected by polar night is somewhat smaller than the area of midnight sun. The polar circle is located at a latitude between these two areas, at the latitude of approximately 66.5 degrees. The function is approximating the atmospheric bend using a height of 0,833333 degrees.

The variant SunriseAndSet::Daylight(DateTime<Utc>, DateTime<Utc>) represents time of sunrise and sunset.

In case latitude or longitude are not in valid ranges, the function will return Result::Err(BadParam).

pub fn sunrise_and_set<F: FloatOps>(utc: DateTime<Utc>, lat: f64, lon: f64) -> Result<SunriseAndSet, SpaError> {..}

Calculating the Solar Position

The following functions is calculating the solar position (azimuth and zenith-angle) at time t and geo-pos lat/lon

The algorithm has been ported to Rust from http://www.psa.es/sdg/sunpos.htm The algorithm is accurate to within 0.5 minutes of arc for the year 1999 and following.

Arguments

  • utc - UTC time-point (DateTime)
  • lat - latitude in WGS84 system, ranging from -90.0 to 90.0.
  • lon - longitude in WGS84 system, ranging from -180.0 to 180.0

The function returns a result of type SolarPos.

pub struct SolarPos {
    // horizontal angle measured clockwise from a north base line or meridian
    pub azimuth: f64,
    // the angle between the zenith and the centre of the sun's disc
    pub zenith_angle: f64,
}

In case latitude or longitude are not in valid ranges, the function will return Result::Err(BadParam).

pub fn solar_position<F: FloatOps>(utc: DateTime<Utc>, lat: f64, lon: f64) -> Result<SolarPos, SpaError> {..}

Platform specific Float Operations

The SPA library supports both std and no_std target builds via the FloatOps trait:

pub trait FloatOps {
    fn sin(x: f64) -> f64;
    fn cos(x: f64) -> f64;
    fn tan(x: f64) -> f64;
    fn asin(x: f64) -> f64;
    fn acos(x: f64) -> f64;
    fn atan(x: f64) -> f64;
    fn atan2(y: f64, x: f64) -> f64;
    fn trunc(x: f64) -> f64;
}
  • On std target builds (default), the built-in implementation StdFloatOps is provided for your convenience, gated behind the std feature.
[dependencies]
spa = "^0.5"
  • On no_std target builds, you need to provide your own implementation for floating point operations, for example using libm. The default-features = false option must be specified in a dependency declaration.
[dependencies]
spa = { version = "^0.5", default-features = false }

Dependencies

~1–1.3MB
~21K SLoC