An improved
system and method for producing replacement sensor signals for failed sensors, and inferred sensor signals for non-instrumented physical parameters, in processes and equipment having one or more sensors in place for monitoring physical parameters. A data history from a fully-instrumented prototype provides a representative
data set of anticipated operating parameters for forming an empirical model in a
computer module for generating “virtual” signals for a process or
machine in real-time. Replacement or inferential sensor signals can be advantageously used in downstream control
processing or analysis. A memory for storing the representative
training set, or a transformation thereof, is coupled to a processor. The processor receives from an input data signals embodying real values from sensors actually on the process or
machine, and may receive these in real-time. The processor is disposed to take a set of readings of the actual sensors from the input, mid generate an estimate of one or more desired inferred sensors, using a linear combination of the representative
training set sensor data, as weighted by the result of a measure of similarity of the input sensor data to the representative
training set sensor data.