| CPC G05B 23/0283 (2013.01) [G05B 23/0232 (2013.01); G05B 23/0243 (2013.01); G05B 23/0267 (2013.01)] | 18 Claims |

|
1. A method for correction of sensors in an industrial process, the method comprising:
collecting first data via one or more online first sensors with respect to the industrial process, the first data corresponding to measured variables for at least one respective process component;
aggregating the first data in data storage as respective input data points with corresponding contextual tags comprising time series identifiers;
for a given sample collected from a production stage of the industrial process and having a respective time series identifier, obtaining at least one respective input data point of the first data having a corresponding time series identifier;
collecting second data via one or more second sensors, the second data comprising measured variables corresponding to the at least one respective process component from the given sample;
determining a difference between at least one measured variable from the first data and the corresponding at least one measured variable from the second data;
selectively generating a feedback signal to at least one of the one or more first sensors as an error value based on the determined difference; and
automatically calibrating the at least one of the one or more first sensors to correct further first data collected from the at least one of the one or more first sensors based on the generated feedback signal.
|