Unsupervised correction algorithm of engine inlet total temperature
In the process of measuring the total temperature of the engine intake,the measured value often deviates from the actual value due to factors such as air flow velocity,radiation,convection,and conduction.At present,some domestic institutions have obtained the total temperature recovery coefficient through blowing equipment to correct the error caused by air flow velocity.But this method only corrects the air flow velocity error,and does not consider the effects of errors like self-heating,anti-icing,thermal convection,thermal radiation,thermal conduction on the total inlet temperature,the correction results obtained are limited.A model-based total temperature correction method was proposed,which could search the total temperature label value through an unsupervised method and correct it based on expert experience.Then a model was established through training of historical data to improve the correction performance.Under specific inlet conditions and sensing parts,this method can take various error factors into account more comprehensively,thereby provide more accurate inlet total temperature correction results.
aero-enginecorrection of inlet total temperaturetotal temperature recovery coefficientunsupervised amendmentexpert experience
NETL
NSTL
万方数据
