Sensor error identification and elimination method based on Laplacian eigenmapping algorithm

Abstract

The invention discloses a sensor error identification and elimination method based on a Laplacian eigenmapping algorithm, and relates to the field of engine testing. An annular symmetrical temperature measurement structure, Laplacian eigenmapping and a wavelet basis noise reduction algorithm to complete continuous and accurate measurement of the temperature of an engine are adopted. The method comprises the following steps: 1, in order to cope with airflow disturbance generated by deflection of tail flame of an engine, arranging an engine combustion chamber annular symmetrical structure measuring device; 2, in order to timely and effectively weaken and remove measurement interference, further obtaining a target temperature value through dimension reduction processing of a Laplacian feature mapping algorithm; and 3, performing sensor temperature signal noise reduction processing through wavelet transform filtering, and selecting a wavelet basis Coif5 to perform temperature signal noise reduction processing so as to eliminate a measurement error caused by a complex engine tail flame temperature change rule. According to the invention, the problems of poor temperature measurement accuracy caused by complex temperature change rules and difficulty in timely and effective removal of measurement interference can be effectively solved.

Description

technical field [0001] The invention relates to a sensor error identification and elimination method, in particular to a Laplacian feature mapping algorithm. Background technique [0002] Engine research and development generally needs to be verified by ground tests, and finally flight tests are carried out. There are many types of test benches, among which wind tunnel test is the second only to flight test. It depends heavily on energy and is not suitable for a large number of experiments. The test bench can provide all the conditions required for the actual operation of the engine. It can adjust the intake, exhaust, fuel flow, pressure, temperature and other parameters of the engine to the same data as the flight process, and collect the data of the engine under different working conditions. Changes in internal temperature, pressure, thrust, vibration, speed, etc., engine accessories, such as generator output voltage, current changes, by analyzing these data to check whet...

AI Technical Summary

Problems solved by technology

[0008]Due to the complex change law of the engine exhaust plume temperature, the combustion chamber exhaust will deflect during the start-up process. Traditional temperature sensors usually use mean value filtering or Fourier transform The accuracy of temperature measurement results is poor in filtering and noise reduction processing, and at the same time the generated airflow disturbance factors will cause measurement interference. Effectively weaken and remove

In order to solve the problem of poor temperature measurement accuracy due to complex temperature changes, a ring-shaped temperature measurement structure is designed to reduce and remove measurement interference in a timely and effective manner. Wavelet transform filtering is used to denoise the sensor temperature signal. Based on the above problems, a A Laplacian eigenmapping algorithm based on the elimination of sensor temperature measurement errors, thereby accurately improving the accuracy of the engine test bench measurement and control system

Images