System and method for validating validity of sensor using control limit

A method for controlling an operation of a sensor installed in a gas turbine based on assessing a validity of the sensor, the method comprising:

• providing the sensor and a processor connected to the sensor by a communication network;
• generating, by the processor, a Bayesian inference model outputting a posterior distribution of a parameter of sensor data using a Bayesian sampling technique based on a prior distribution of the parameter of the sensor data and measured historical data of the sensor;
• determining, by the processor, control lines of the sensor data comprising an upper control line and a lower control line based on a credible interval of the posterior distribution, wherein each of the upper control line and the lower control line is a value being compared to real-time sensor data;
• receiving, by the processor, a performance reduction index of the gas turbine acquired by measuring a performance reduction degree of the gas turbine;
• updating, by the processor, the upper and the lower control lines by reflecting a trend of the performance reduction index;
• receiving, by the processor, the real-time sensor data from the sensor;
• monitoring, by the processor, whether the real-time sensor data deviates from the control lines;
• determining, by the processor, that the sensor is generating correct data until the real-time sensor data deviates from the control lines to ensure an extended continuity of the gas turbine in which the sensor is operated;
• triggering an alarm via a speaker and a display in response to determining that the real-time sensor data deviates from the control lines;
• and
• repairing or replacing the sensor when the real-time sensor data deviates from the control lines,
• wherein the parameter includes a mean and a standard deviation,
• wherein the credible interval of the posterior distribution comprises a first credible interval and a second credible interval,
• wherein the first credible interval is established by:
• determining a median value of the posterior distribution of the mean;
• and
• establishing the first credible interval as a first predetermined percentage of a highest posterior density of the posterior distribution of the mean with respect to the median value of the posterior distribution of the mean,
• wherein second credible interval is established by:
• determining a median value of the posterior distribution of the standard deviation, and
• establishing the second credible interval as a second predetermined percentage of the highest posterior density of the posterior distribution of the standard deviation with respect to the median value of the posterior distribution of the standard deviation,
• wherein the determining of the control lines comprises:
• selecting a first lower boundary value denoted as a μL and a first upper boundary value denoted as a μu for the first credible interval;
• selecting a second lower boundary value denoted as a σL and a second upper boundary value denoted as a σu for the second credible interval;
• determining a lower control limit range by combining the μL with each of the σL the σu, and an upper control limit range by combining the μu with each of the σL and the σu;
• and
• determining the lower control line within the lower control limit range and the upper control line within the upper control limit range.