Denoising Speech Based on Deep Learning and Wavelet Decomposition

The work proposed a denoising speech method using deep learning. The predictor and target network signals were the amplitude spectra of the wavelet-decomposition vectors of the noisy audio signal and clean audio signal, respectively. The output of the network was the amplitude spectrum of the denoised signal. Besides, the regression network used the input of the predictor to minimize the mean square error between its output and input targets. The denoised wavelet-decomposition vector was transformed back to the time domain by the output amplitude spectrum and the phase of the wavelet-decomposition vector. Then, the denoised speech was obtained by the inverse wavelet transform. This method overcame the problem that the frequency and time resolution of the short-time Fourier transform could not be adjusted. The noise reduction effect in each frequency band was improved due to the gradual reduction of the noise energy in the wavelet-decomposition process. The experimental results showed that the method has a good denoising effect in the whole frequency band.

Acoustic hologram optimisation using automatic differentiation

Acoustic holograms are the keystone of modern acoustics. They encode three-dimensional acoustic fields in two dimensions, and their quality determines the performance of acoustic systems. Optimisation methods that control only the phase of an acoustic wave are considered inferior to methods that control both the amplitude and phase of the wave. In this paper, we present Diff-PAT, an acoustic hologram optimisation platform with automatic differentiation. We show that in the most fundamental case of optimizing the output amplitude to match the target amplitude; our method with only phase modulation achieves better performance than conventional algorithm with both amplitude and phase modulation. The performance of Diff-PAT was evaluated by randomly generating 1000 sets of up to 32 control points for single-sided arrays and single-axis arrays. This optimisation platform for acoustic hologram can be used in a wide range of applications of PATs without introducing any changes to existing systems that control the PATs. In addition, we applied Diff-PAT to a phase plate and achieved an increase of > 8 dB in the peak noise-to-signal ratio of the acoustic hologram.

Cascaded Nonlinear Mass Fluctuation Stochastic Resonance System and Its Application in Bearing Fault Diagnosis

Stochastic resonance (SR) can realize bearing fault signal diagnosis by transferring noisy energy. In order to enhance the output response of the system and realize effective signal extraction, the nonlinear mass fluctuation SR system caused by nonlinear asymmetric dichotomous noise is cascaded to obtain the cascaded nonlinear mass fluctuation SR system. First, the output amplitude gain of the first-stage of the system is derived, and the influence of different parameters on it is explored; then the effects of different parameters of the cascaded system on the output amplitude gain and the output SNR are studied separately, which proves that the cascaded system can effectively double enhance the output response of the system; finally, the adaptive genetic algorithm is used to solve the difficulty of parameter adjustment, and the cascaded nonlinear mass fluctuation SR system is applied to the bearing fault diagnosis. The system proposed in this paper takes into account the effects of nonlinear asymmetric dichotomous noise and cascaded systems and performs waveform smoothing and double enhancement of the output signal. It can better extract fault signals and has effective engineering value.

INVESTIGATION OF AN ELECTROMAGNETIC TWO-STROKE VIBRATING ACTIVATOR IN OSCILLATORY MODE

The article considers the results of the study of electromagnetic vibration exciter with sequentially included capacitor in the electrical circuit, consisting of mechanical and electrical subsystems. It is shown that by means of the Lagrangian-Maxwell equation the interconnection between mechanical and electric subsystems can be realized. The relations describing processes of establishment of amplitudes and phases of oscillations both in mechanical and in electric subsystems are deduced. The equations connecting the output (amplitude) of vibration of the vibrating exciter with its input (voltage) of the network are presented. As a result, the formulas allowing making corrections at the solution of the system describing operation of the electromagnetic vibrating exciter in two-stroke mode are presented.

Cooperative mechanism of resonant behaviors in a fluctuating-mass generalized Langevin system with generalized Mittag–Leffler memory kernel

In this study, we investigate the resonant behaviors in the fluctuating-mass generalized Langevin equation (GLE) with generalized Mittag–Leffler (M–L) memory kernel. By using the stochastic averaging method and Laplace transform, we obtain the exact expression of the first-order moment of system steady response, based on which we analyze the dynamical mechanism of the various non-monotonic phenomena. Based on tbe numerical results, we further discuss the dependence on various parameters systematically and study the interplay and cooperation between the generalized M–L memory kernel and trichotomous noise in terms of output amplitude amplification. The results reveal the coexistence of non-monotonic phenomena in the proposed system, such as bona fide stochastic resonance (SR), conventional SR and wide-sense SR. We even observe the stochastic multi-resonance (SMR) behaviors with five or six peaks in the evolution of output amplitude amplification varying with the driving frequency. It is worth emphasizing that quintuple-peak and sextuple-peak bona fide SR phenomena had never been observed in the previous literatures. Thus, these results will provide more extensive support for manipulating the resonant behaviors through system parameter control in the potential applications.

Study on a Large-Scale Three-Dimensional Ultrasonic Plastic Welding Vibration System Based on a Quasi-Periodic Phononic Crystal Structure

The uniformity of amplitude distribution and amplitude gain are two main factors affecting the performance of ultrasonic welding vibration system. In order to improve the uniformity of amplitude distribution and amplitude gain of welding surface to enhance the performance of the vibration system, a new design method of a large-scale three-dimensional ultrasonic plastic welding vibration system based on a quasi-periodic phononic crystal structure is proposed. In this method, the composite horn combined with a conical section and a cylindrical section can effectively improve the output amplitude gain of the welding surface. In addition, the method forms a quasi-periodic phononic crystal structure by slotting in a large-scale three-dimensional tool head, and utilizes the band gap property of the structure to effectively suppress lateral vibration of the tool head and improve the amplitude distribution uniformity of the tool head’s welding surface. However, when the size of the tool head is relatively large, the quasi-periodic phononic crystal structure cannot suppress the lateral vibration very well. Therefore, the paper processes fan-shaped slopes on the output surface of the tool head which can further improve the uniformity of the amplitude distribution and amplitude gain. Finally, the simulation analysis and experiments show that the design method can optimize the large-scale three-dimensional ultrasonic plastic welding system, improve the uniformity of the vibration distribution and increase the output amplitude gain of the welding surface.

Trifold frequency multiplier based on mixer for the feedback system of BEPC II

This paper will introduce the structure, development process, test and experiment of trifold frequency multiplier by using mixer and its application in accelerator fields. This invent includes a power splitter, a frequency doubler, a mixer, a bandpass filter, an amplifier and several RG233 cables. One 500MHz signal which always is provide by the timing system of the accelerator is input to a power splitter, a power splitter could divide one signal into two signals with equal magnitude, one is directly input to the LO port of mixer, the other is input to frequency doubler then 1.0 GHz signal will be got. This 1.0 GHz signal is input to the RF port of the mixer, then output at IF port of mixer, the frequency of the IF port signal is sum and difference of 500MHz and 1.0GHz, so 500MHz and 1.5GHz signal will be got. Then this signal is input to a bandpass filter whose centre frequency is 1.5GHz, so a clean 1.5GHz signal is got. The trifold frequency multiplier by using mixer is simple to operate, and it has good performance. In the experiments, the 500MHz signal was passed through the Trifold Frequency Multiplier to get the 1.5GHz signal, and its output amplitude stability is 0.49%, and its synchronization stability is 10.3 ps compared with the 500MHz input signal.