Active noise control algorithm for transformer

Abstract

The invention discloses an improved active noise control algorithm for a transformer. The active noise control algorithm has the remarkable advantages that: (1) reference signals are obtained via adopting DDS (direct digital synthesis) by only needing to collect fundamental frequency components in primary noises, thus the reference signals are acquired simply; (2) an simple off-line disturbance modeling method is adopted; and (3) the updating interval of the control weight of the algorithm is a multi-harmonic noise period, thus being low in calculated amount. During modeling, the added disturbance magnitude is over 5 times than variable quantity of primary harmonic noises or over 10% of primary harmonic noise amplitude, thus ensuring to obtain the preferable modeling result of transfer functions of a secondary path. The stability and the noise reduction performance of the algorithm are both considered for the dereferencing of a leakage coefficient (le) in an update formula of the control weight, thus the dereferencing is 0.9999; and the convergence rate and the stability of the algorithm are both considered for the dereferencing of a convergence coefficient (mu), thus the dereferencing is 0.05 when the distance between an error microphone and a control source is 20cm.

Description

1. Technical Field [0001] The invention relates to a noise control algorithm of an active noise control system. An improved active control algorithm is proposed for the harmonic characteristics of transformer noise. 2. Background technology [0002] Transformer noise mainly comes from the iron core. Under the action of alternating magnetic flux, the iron core emits humming due to the magnetostriction. Its characteristic is the even-numbered multiplication relationship of the power frequency, mainly 100Hz, 200Hz, 300Hz. This kind of low frequency AC noise uses passive control technology, the size of the noise reduction equipment is usually relatively large, and the noise reduction effect is not obvious. Active noise control technology can just make up for this deficiency. [0003] In order to achieve better noise reduction effects, adaptive algorithms are often used, such as the Filtered-X LMS (FXLMS) algorithm (V Martin, A Roure. Active noise control of acoustic sources using sph...

AI Technical Summary

Problems solved by technology

[0004] US Patent US4878188 proposes an active control method using adaptive notch filter technology to control single-frequency noise, but this method will lead to unstable control when the secondary channel has a large gain outside the control frequency band

A frequency estimation method for controlling harmonic noise based on adaptive notch filter technology was proposed (S Kim, Y Park. Active control of multi-tonal noise with reference generator based on on-line frequency estimation. Journal of Sound and Vibration1999; 227(3):647-666.), this method has the advantages of fast convergence, good stability, and no need for reference signals, but due to the need for online frequency evaluation, the amount of calculation is larger than the method proposed by US4878188

Images