pub struct PcaObservation {
pub run_index: usize,
pub t2: Option<f64>,
pub q_stat: Option<f64>,
pub n_components: usize,
pub pc1_loading: Option<Vec<f64>>,
pub residual_vector: Option<Vec<f64>>,
pub sensor_labels: Vec<String>,
}Expand description
A single multivariate process observation expressed in terms of the PCA residual statistics already computed by the upstream FDC system.
All fields are received from the FDC system; the DSFB observer never writes back to it.
Fields§
§run_index: usizeZero-indexed run / lot identifier.
t2: Option<f64>Hotelling’s T² statistic for this run (Mahalanobis distance² in the PC subspace).
q_stat: Option<f64>Q-Statistic (Squared Prediction Error / SPE) for this run.
n_components: usizeNumber of principal components retained by the upstream FDC model.
pc1_loading: Option<Vec<f64>>Loadings of the first principal component, one entry per sensor.
Must have length == number of sensors selected for PCA.
None when the FDC model does not expose loadings.
residual_vector: Option<Vec<f64>>Raw normalised residual vector (one entry per sensor).
None when the FDC system does not expose individual residuals.
sensor_labels: Vec<String>Sensor labels corresponding to entries in pc1_loading and
residual_vector.
Trait Implementations§
Source§impl Clone for PcaObservation
impl Clone for PcaObservation
Source§fn clone(&self) -> PcaObservation
fn clone(&self) -> PcaObservation
1.0.0 · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
source. Read more