Correction Method
Mathematical modeling of sensor responses using two independent input variables allows the simultaneously occurring effects of temperature and pressure to be corrected in a single step. Through bivariate calibration, a multi-dimensional array of measurements is constructed to map the transducer response across both physical ranges.
Measurement Process
Data acquisition during this procedure involves placing the sensor in a series of controlled thermal states while systematically varying the hydrostatic load. The derived coefficients populate a polynomial equation that represents the true output as a function of the raw frequency or voltage inputs. This technique isolates the primary target variable from environmental influence.
It avoids the compounding errors that arise when correcting for temperature and pressure in separate, sequential operations.
Error Source
Residual variations after this correction are typically evaluated using root mean square calculations across the calibration grid. System drift or long term aging of the sensor elements can degrade the validity of the computed surface, which necessitates periodic recalculation to maintain precision.
Hardware Requirement
Deploying these multi-variable algorithms requires sufficient memory and processor capability within the remote instrument housing to execute the real-time matrix math. In low power oceanographic applications, this computational demand must be balanced against battery life and sampling frequency.
