Vehicle active noise reduction method based on heuristic dynamic programming

Abstract

The invention discloses a car active noise reduction method based on heuristic dynamic programming, and belongs to the field of active noise reduction methods of noise in car rooms. According to the car active noise reduction method, massive data in a large database of noise information in a virtual environment of a cloud server are initiatively fitted, the rule of noise frequency spectrum characteristics in the database is looked for through a model network and an execution network trained by heuristic dynamic programming, the noise data with the similar frequency spectrum characteristics arefitted together, an off-line expert library having the less data size and reduced and generated from a huge database is established, so that optimization of the noise data is achieved, and the optimal control quantity is provided for a system. According to the innovative mode, the method can overcome the drawback that an existing car active noise reduction method needs to conduct one-by-one real-time operation on the massive data, and through practical testing, the method can fully meet requirements of noise reduction of an actual traffic car under a real road condition, so that the car active noise reduction method has great market prospect and potential economic value.

Description

technical field [0001] The invention belongs to the field of active noise reduction methods for indoor noise in automobiles, and in particular relates to an active noise reduction method for automobiles based on heuristic dynamic programming. Background technique [0002] Because of its ability to effectively reduce low-frequency noise, active noise reduction technology has attracted people's attention and has been gradually applied in reducing low-frequency noise. The principle of active noise reduction is to generate an artificial noise reduction signal with the same amplitude and frequency as the noise source signal and the opposite direction through the electro-acoustic system that can generate artificial noise reduction after the identification and analysis of the noise source signal. The artificial noise reduction and silencing signal is superimposed on the noise source signal to realize the cancellation of the source noise. [0003] The propagation of sound in the ai...

AI Technical Summary

Problems solved by technology

[0011] In order to solve the problem that the existing adaptive filter based on the principle of digital circuit and the multi-bandpass filter based on the principle of analog circuit cannot carry out the massive noise information required for the noise reduction calculation with the cloud server through wireless transmission in the state of offline driving Real-time interaction prevents the existing on-board offline noise reduction system from being supported by the storage and computing power of the cloud server; the existing active noise reduction methods are only for the theoretical model of a single area, and cannot be effectively applied to the real environment with multiple seats. Driving road conditions, it cannot timely and accurately perform qualitative classification judgments on noise type switching under various driving conditions in a timely and effective manner, and thus cannot quickly provide the optimal noise reduction parameter processing scheme under corresponding working conditions. At the same time, the old The method fails to classify and screen the vehicle cabin noise under various actual road conditions based on the spectral characteristics of the noise, and cannot form a greatly simplified offline noise reduction expert database. The effect is unsatisfactory, the coping strategy cannot be optimal; and the existing active noise reduction method cannot meet the technical problem that high-end vehicles can adjust and set the noise reduction area arbitrarily. The present invention provides a heuristic dynamic programming based Active Noise Cancellation Method for Cars

Images