Methods for balancing aircraft engines based on flight data

A method for balancing a gas turbine aircraft engine, comprising:

• (a) acquiring flight parameter data, balance weight data and vibration data from one or more flights by each of a multiplicity of reference aircraft, wherein the flight parameter data includes engine revolutions per minute, altitude, Mach number, temperature and thrust;
• (b) using the flight parameter data, balance weight data and vibration data for each flight by each reference aircraft of the multiplicity of reference aircraft to train a respective associated artificial neural network of a multiplicity of artificial neural networks to output vibration data representing predicted vibrations for an engine in response to input of respective flight parameter data for the associated flight, wherein each artificial neural network is trained to establish a mapping or correlation between a multiplicity of inputs in the form of engine revolutions per minute, altitude, Mach number, temperature, thrust, and magnitude and phase angle of a balance weight configuration and an output in the form of amplitude and phase of a predicted engine vibration;
• (c) acquiring flight parameter data, balance weight data and vibration data from a flight by a test aircraft having an engine to be balanced, wherein the flight parameter data includes engine revolutions per minute, altitude, Mach number, temperature and thrust;
• (d) using the artificial neural networks to predict the engine vibrations that would be produced by each reference aircraft based on the input of flight parameter data and balance weight data from the flight by the test aircraft;
• (e) identifying engine vibration predictions output by the artificial neural networks which match or nearly match the engine vibration measurements acquired from the test aircraft during its flight;
• (f) for each artificial neural network that produced matching or nearly matching vibration predictions, determining the magnitude and phase angle of a mass vector for each balance weight configuration that was in place on an engine for flights of reference aircraft whose associated artificial neural network produced matching or nearly matching engine vibration predictions;
• (g) generating a report that includes magnitudes and phase angles of mass vectors determined in step (f);
• (h) selecting a magnitude and a phase angle included in the report;
• and
• (i) reconfiguring the balance weights attached to the engine to be balanced on the test aircraft in accordance with the selected magnitude and phase angle.

A gas turbine aircraft engine comprising a fan spinner and one or more balance weights attached to the fan spinner, wherein the engine was balanced using a method comprising:

• (a) acquiring flight parameter data, balance weight data and vibration data from one or more flights by each of a multiplicity of reference aircraft, wherein the flight parameter data includes engine revolutions per minute, altitude, Mach number, temperature and thrust;
• (b) using the flight parameter data, balance weight data and vibration data for each flight by each reference aircraft of the multiplicity of reference aircraft to train a respective associated artificial neural network of a multiplicity of artificial neural networks to output vibration data representing predicted vibrations for an engine in response to input of respective flight parameter data for the associated flight, wherein each artificial neural network is trained to establish a mapping or correlation between a multiplicity of inputs in the form of engine revolutions per minute, altitude, Mach number, temperature, thrust, and magnitude and phase angle of a balance weight configuration and an output in the form of amplitude and phase of a predicted engine vibration;
• (c) acquiring flight parameter data, balance weight data and vibration data from a flight by a test aircraft having an engine to be balanced, wherein the flight parameter data includes engine revolutions per minute, altitude, Mach number, temperature and thrust;
• (d) using the artificial neural networks to predict the engine vibrations that would be produced by each reference aircraft based on the input of flight parameter data and balance weight data from the flight by the test aircraft;
• (e) identifying engine vibration predictions output by the artificial neural networks which match or nearly match the engine vibration measurements acquired from the test aircraft during its flight;
• (f) for each artificial neural network that produced matching or nearly matching engine vibration predictions, determining the magnitudes and phase angles of mass vectors for balance weight configurations in place on the engines for the flights of reference aircraft whose associated artificial neural network produced matching or nearly matching engine vibration predictions;
• (g) selecting a magnitude and a phase angle from the magnitudes and phase angles of mass vectors determined in step (f);
• and
• (h) reconfiguring the balance weights attached to the engine to be balanced on the test aircraft in accordance with the selected magnitude and phase angle.