Fault detection system and method using augmented data and fuzzy logic

An apparatus comprising:

• a) a processor;
• b) a memory coupled to the processor;
• c) a fault detection program residing in memory and being executed by the processor, the fault detection program including:
• i) a sensor data processing program, the sensor data processing program configured to receive sensor data from a turbine engine and determine a difference between the sensor data and expected values of the sensor data to generate residuals from the sensor data and determine a rate of change of the residuals, the residuals from the sensor data and the rate of change of the residuals providing an augmented data set;
• and
• ii) a fuzzy logic inference program, the fUzzy logic inference program configured to receive the augmented data set, and wherein the fuzzy logic inference program includes a plurality of membership functions and wherein each of the plurality of membership functions is associated with at least one data type in the residuals from the sensor data and the rate of change of the residuals, and wherein the fuzzy logic program is configured to fuzzify the residuals from the sensor data and the rate of change of the residuals using the plurality of membership functions and analyze the residuals from the sensor data and the rate of change of the residuals to determine a likelihood that a fault has occurred.

A fault detection system for detecting faults in a turbine engine, the fault detection system comprising:

• a sensor data processor, the sensor data processor configured to:
• receive sensor data from the turbine engine, the sensor data including exhaust gas temperature data, engine speed data, and fuel flow data;
• generate exhaust gas temperature residuals by determining differences between the exhaust gas temperature data to expected values of exhaust gas temperature;
• generate engine speed residuals by determining differences between the engine speed data to expected values of engine speed;
• generate fuel flow residuals by determining differences between the fuel flow data and expected values of fuel flow;
• determine a rate of change of the exhaust gas temperature residuals;
• determine a rate of change of the engine speed residuals;
• determine a rate of change of the fuel flow residuals;
• and
• a fuzzy logic inference system, the fuzzy logic inference system configured to receive the exhaust gas temperature residuals, the engine speed residuals, the fuel flow residuals, the rate of change of the exhaust gas temperature residuals, the rate of change of the engine speed residuals, and the rate of change of the fuel flow residuals, and wherein the fuzzy logic inference system includes a plurality of membership functions, and wherein the fuzzy logic system is configured to fuzzify the exhaust gas temperature residuals, the engine speed residuals, the fuel flow residuals, the rate of change of the exhaust gas temperature residuals, the rate of change of the engine speed residuals, and the rate of change of the fuel flow residuals using the plurality of membership functions to determine a likelihood that a fault has occurred in the turbine engine.