基于时延神经网络模型的舰船辐射噪声目标识别*

水声目标被动识别是水声信号处理领域的研究热点之一。海洋环境中存在的不规则噪声干扰,使得基于传统方法的水声目标被动识别技术在实际的应用场景中效果不佳。本文采用一种基于时延网络(Time Delay Neural Network,TDNN)模型的舰船辐射噪声目标识别方法,该方法利用目标的短时平稳特性和长时关联特性对目标的声纹特征进行建模,使用梅尔谱图提取目标信号的初级特征,再通过融合注意力机制和时延神经网络的深度学习模型实现高级特性提取,最后再利用余弦相似度实现不同目标的类别划分。该方法在ShipsEar数据集和自行采集的数据进行测试验证,目标识别准确率分别达到79.2%和73.9%,可证明本文方法的有效性。     

Implementation of active noise control in a multi-modal spray dryer exhaust stack

Tonal noise emitted from large-diameter spray dryer exhaust stacks used in the dairy industry can give rise to complaints from nearby communities. In many cases, the tone at the fan blade passing frequency is characterized by a frequency above the first mode cut on frequency of the exhaust stack and both its amplitude and the frequency are time varying. The variation in amplitude is a result of turbulence and temperature variations in the duct which cause angular variations in the nodal plane of modes with diametrical nodes. This in turn results in large fluctuations in sound pressure with time at any specified location in the duct, thus presenting a significant challenge for an ANC system with fixed control source and error sensor locations. In many food processing industries, the use of sound absorptive materials in silencers is not acceptable and, particularly when the fan speed is variable, it is difficult to achieve an acceptable passive solution at a reasonable cost. Here, the design and implementation of an active noise control system for tonal noise propagating above the cut-on frequency of the first higher order mode in large size cylindrical industrial exhaust stack is discussed, where the frequency and amplitude vary significantly and relatively rapidly with time. Physical system design principles and control algorithm optimization for a practical active noise control system are presented. Finally, real time control results which were achieved by a prototype installation on a large-diameter, in-service exhaust stack are given. Significant noise reductions were achieved in the community.     

宽频带海洋环境噪声矢量场空间相关特性建模

基阵的信噪比增益与噪声场空间特性密切联系,海洋环境噪声空间特性建模始终是水声学研究的热门问题.声纳功能不同,其工作频段和带宽通常也不相同,因此,任意频带噪声场的空间相关系数对声纳系统设计具有重要参考价值.依据海洋环境噪声场的产生过程,在高频近似条件下,本文提出一种噪声场时域建模方法,给出了水平分层介质中表面噪声时域声压和质点振速的积分表示,为噪声矢量场宽带模型的建立奠定了基础.根据风成噪声谱结构,数值计算了不同频带、不同谱斜率的噪声场空间相关系数,揭示了带宽、谱结构对风成噪声空间特性的影响规律.随着阵元间距和带宽增大,噪声矢量场各分量的空间相关系数的振荡周期数逐渐减少,振荡幅度逐渐减小,这是由于噪声场相关系数频域平均的结果.当谱斜率小于零时,宽频带噪声场的空间相关半径大于窄带噪声场的相关半径,这是由于低频段噪声起主要贡献的结果,实测海洋环境噪声声压场竖直方向空间相关特性变化规律与理论结果一致.本文模型对换能器成阵技术研究以及环境参数反演具有潜在应用前景.     

Measurements of a periodically driving qubit using a quantum point contact

Using the method developed by Gurvitz [1996 Phys. Rev. B 53 15932], we obtained the Bloch-type rate equations describing the entire system of a periodically driving qubit monitored by a quantum point contact detector. The results demonstrate that the isolated qubit can be kept in its initial state with a large driving frequency due to more difficult electron tunneling in qubit, and this initial state can always be measured at a small measurement-induced decoherence rate during a short time.     

Study on sound transmission characteristics of a cylindrical shell using analytical and experimental models

A circular cylindrical cavity enclosed by a thin elastic shell is found in many practical devices such as expansion volume mufflers, hermetic compressors and aircraft cabins. Analytical and experimental studies are conducted in this work to understand the characteristics of sound transmission through the cylindrical wall of such a system. Using an infinitely long circular cylindrical shell subjected to a plane incident wave, an exact solution is obtained by solving the classical shell vibration equations and the acoustic wave equations simultaneously. Transmission losses obtained from the solution are compared to the transmission losses that are measured for a cylindrical shell of finite length and the same cross-sectional dimensions. The comparison suggests that the theoretical model can be used as an effective design tool despite considerable simplifications involved.     

水下近水面锥柱组合壳声固耦合效应的半解析求解方法

针对水下近水面锥柱组合壳声固耦合多借助于数值方法求解的现状,本文提出一种半解析方法从机理上分析此类问题。首先基于能量泛函和Sanders壳体理论、虚拟弹簧法以及力与力矩平衡条件建立锥柱组合壳的结构模型;然后采用Legendre谱元法和二维傅里叶变换得到含自由液面的水下声场模型;最后由非线性迭代法和高斯积分求解耦合系统声振控制方程。通过与参考文献和数值方法结果的对比,验证了本文方法的收敛性、正确性和可靠性。研究结果表明,结构参数、浸没深度和激励频率与远场辐射声压密切相关。本文工作可推广到水下含内部结构的复杂旋转组合壳在不同结构边界及声边界下的声固耦合问题。     

Predictions and experimental studies of the tail pipe noise of an automotive muffler using a one dimensional CFD model

The tail pipe noise from a commercial automotive muffler was studied experimentally and numerically under the condition of wide open throttle acceleration in the present research. The engine was accelerated from 1000 to 6000 rpm in 30 s at the warm up condition. The transient acoustic characteristics of its exhaust muffler were predicted using one dimensional computational fluid dynamics. To validate the results of the simulation, the transient acoustic characteristics of the exhaust muffler were measured in an anechoic chamber according to the Japanese Standard (JIS D 1616). It was found that the results of simulation are in good agreement with experimental results at the 2nd order of the engine rotational frequency. At the high order of engine speed, differences between the computational and experimental results exist in the high revolution range (from 5000 to 6000 rpm at the 4th order, and from 4200 to 6000 rpm at the 6th order). According to these results, the differences were caused by the flow noise which was not considered in the simulation. Based on the theory of one dimensional CFD model, a simplified model which can provide an acceptable accuracy and save more than 90% of execution time compared with the standard model was proposed for the optimization design to meet the demand of time to market.     

Prediction of noise inside an acoustic cavity of elongated shape using statistical energy analyses with spatial decay consideration

In this paper, Statistical Energy Analysis (SEA) is used to predict the interior noise of an acoustic cavity of elongated shape. The disadvantage of the conventional SEA method, which quantifies the response in terms of the energy averaged over each subsystem, is overcome by introducing a one-dimensional spatial decay relation, through which information about the acoustic energy variation in the elongated direction is taken into account. The modified SEA is experimentally validated using a 1:5 scaled space station prototype, having the longitudinal dimension much larger than the cross-sectional dimension. It is also compared with a model reported in the literature. It is shown that, in the region where the acoustic pressure level decays at a constant rate, the two models agree well with each other and are capable of estimating the acoustic pressure variation along the space station cabin. However, near the end walls where the decay rate of the acoustic pressure level is not constant, the proposed model provides better accuracy.     

Identification of noise sources on a residential split-system air-conditioner using sound intensity measurements

This paper presents the results of experimental studies of the noise of a residential split-system air-conditioner unit. The compressor and condenser and associated fans were removed from the unit and did not form part of the studies. Care was taken with the unit to separate the inlet and exhaust noise from the noise radiated from the cabinet. The measurements were made with a two-microphone sound intensity probe and these resulted in sound power level data. The sound power levels produced by radiation from the inlet, exhaust and cabinet were obtained for five different volume flow rates. The effect on the sound power generated by removing the coil was investigated. Measurements and subjective studies show that the low frequency sound is predominantly radiated from the exhaust and inlet. At high frequency, the cabinet noise dominates.     

Effect of a Gaussian white noise on the charge density wave dynamics in a one dimensional compound

The effect of a white Gaussian noise (WGN) on the charge density wave (CDW) dynamics properties of a one-dimensional conductor is studied numerically in the weak pinning limit. In agreement with a recent experimental work on the charge density wave conductor K_0.3MoO₃, the addition of a Gaussian noise affects the CDW dynamics of the system. The results are discussed in the context of CDW dislocations.     

Geometric phase of a qubit driven by a phase noise laser under non-Markovian dynamics

Robustness of the geometric phase (GP) with respect to the environmental effects is a basic condition for an effective quantum computation. Here, we study quantitatively the GP of a two-level atom system driven by a phase noise laser under non-Markovian dynamics in terms of different parameters involved in the whole system. We find that with the change of the damping coupling, the GP is very sensitive to its properties exhibiting long collapse and revival phenomena, which play a significant role in enhancing the stabilization and control of the system dynamics. Moreover, we show that the GP can be considered as a tool for testing and characterizing the nature of the qubit–environment coupling. Due to the significance of how a system is quantum correlated with its environment in the construction of a scalable quantum computer, the entanglement dynamics between the qubit with its environment under external classical noise is evaluated and investigated during the time evolution.     

Synthesis and rheological investigation of a magnetic fluid using olivary silica-coated iron particles as a precursor

A new type of magnetic fluid was prepared by dispersing monodispersed iron–silica (Fe–SiO₂) composite particles in polyethylene glycol (PEG) 400. The composite particles Fe–SiO₂ were synthesized by hydrogen reduction from α-Fe₂O₃–SiO₂ spheres. Their microstructures were observed by a high-resolution transmission electron microscope (HRTEM) and the magnetism was characterized with a superconducting quantum interference device (SQUID) magnetometer. Both steady-state and dynamic rheological properties of the magnetic fluid under different magnetic fields were studied by using a rheometer. Experimental results show that this magnetic fluid has a relatively high magnetoviscous effect at low shear rates. The yield stress of this material shows an increasing trend with a magnetic flux density. Also, viscoealstic properties of such materials are different from conventional ones.     

Computation of internal aerodynamic noise from a quick-opening throttle valve using frequency-domain acoustic analogy

Recently, plastic products in air-intake parts of automotive engines have become very popular due to advantages that include reduced weight, constricted cost, and lower intake air temperature. However, flow-induced noise in air-intake parts becomes a more serious problem for plastic intake-manifolds than for conventional aluminum-made manifolds. This is due to the fact that plastic manifolds transmit more noise owing to their lower material density. Internal aerodynamic noise from a quick-opening throttle valve is computed by using a frequency-domain acoustic analogy, which is based on the integral formula derived by using the General Green Function, Lighthill’s acoustic analogy and Curle’s extension of Lighthill’s. The integral formula is arranged in such a manner that allows a frequency-domain acoustic signal to be predicted at any location in a duct through the use of unsteady flow data in space and time, which can be obtained by applying Computational Fluid Dynamics techniques. The prediction of the acoustic pressure level from the quick-opening throttle valve shows good agreement with actual measurement. Through the detailed analysis of the flow-noise generation mechanism it was found that the anti-vortex lines, formed behind the throttle valve during the quick-opening behavior, feed the large-scale coherent turbulence and, as a result, play a crucial role in generating the dipolar sound from unsteady loading of the quick-opening throttle valve. From this, it can be inferred that the low-noise design of the throttle-duct system can be made by using this concept to break the large-scale vortex structure.     

An inversion modelling method to obtain the acoustic power of the noise sources in a large factory

A common problem for large factories that wish to decrease their environmental acoustic impact on neighbouring locations is to find out the acoustic power of every noise source. As these factories cannot stop their activity in order to measure each source individually, a procedure is needed to obtain the acoustic powers with the factory under normal operating conditions. Their contribution to the overall sound pressure level at each neighbouring location can then be found and it is possible to calculate the improvements obtained after any modification of the sources. In this paper an inversion modelling method is used to do so. Acoustic powers are obtained by means of field sound pressure level measurements and with the use of a sound propagation software. A careful analysis of the solution has been carried out by simulating errors on the measured data in order to detect possible correlations between the acoustic power of different sources and avoid misleading interpretations of the results. The whole methodology has been applied to a liquid-gas production factory.     

格子玻尔兹曼方法对不同张角聚焦声束的建模

高强度聚焦超声(HIFU)是一种新型的无创治疗肿瘤新技术,其中换能器声场数值计算能够为HIFU治疗提供重要的依据。传统非线性KZK和SBE模型广泛应用于换能器声场数值计算,但依然存在某些不足。我们采用一种介观尺度的新型流体力学方法,即格子Boltzmann方法(LBM),基于2维9离散速度(D2Q9)格子构建了轴对称多弛豫参数LBM模型,并通过调节弛豫参数分析其对模型的影响;利用该模型对两个具有不同张角的球面聚焦换能器的声场进行数值模拟,并与KZK和SBE模型的计算结果进行比较。结果表明LBM模型能够很好地描述超声波的激发和传播机制,从流体力学的角度描述聚焦声场的分布,具有清晰的物理意义,且计算过程不受换能器张角的限制,在换能器声场的理论分析和模拟计算及其在HIFU治疗中的应用有着积极的意义。     

Investigation of rail noise and bridge noise using a combined 3D dynamic model and 2.5D acoustic model

Bridge noise and rail noise are two major sources of an elevated rail transit bridge in the low and medium frequency range (20–1000 Hz). However, in most of the existing literature, the noise radiated from the bridge and rail was investigated separately or using a simplified source model. In this study, an accurate method is proposed to simulate both the rail noise and bridge noise simultaneously. First, the dynamic responses of the rail and multi-span bridge are obtained using a three-dimensional (3D) vehicle-track-bridge interaction analysis model. Then, the two-dimensional (2D) infinite element model is used to calculate 3D modal acoustic transfer vectors of the rail and bridge based on the wavenumber transformation, in order to overcome the singularity and non-uniqueness of the conventional boundary element method and reduce the computation cost. Third, a field test is conducted, and the accuracy of the proposed simulation procedure is verified. Finally, the contribution of the rail and bridge noise to the total noise level is investigated in the whole space near the bridge. Generally the bridge noise occupies a higher contribution in the space beneath the girder due to the shielding effect of the bridge shape on the rail noise, while the rail noise is dominant in the upper space above the bridge. It is found the presence of the vehicle bodies has considerable effect on the rail noise but little influence on the bridge noise. The slope of the roughness level spectrum has significant influence on the dominant field of bridge noise and rail noise. For the excitation of the assumed ISO roughness level used in this study, the difference between the rail noise and bridge noise is only about 3 dB at field points 15–30 m away from the track center, which indicates both the bridge and rail noise should be included in the noise prediction for an elevated rail transit bridge.     

利用单驱动变量实现一类混沌系统的线性和非线性广义同步

研究了一类混沌系统的线性和非线性的广义同步.对此类混沌系统,通过设计一个合适的响应系统,可以仅通过传递一个驱动变量实现系统的线性和非线性广义同步.给出了该响应系统设计的一般方法.由于只需传递一个信号,比已有方法具有更高的实用价值,理论推导和数值仿真进一步表明了该方法的有效性. 关键词： 混沌系统 广义同步 单驱动变量     

利用单驱动变量实现一类分数阶混沌系统的修正投影同步

通过设计一个合适的响应系统,提出了一种修正投影同步方案,该方法通过传递一个驱动变量实现了一类分数阶混沌系统的修正投影同步,由于只需在驱动系统和响应系统之间传递一个驱动变量,具有较高的实用价值,最后以分数阶统一混沌系统为例进行了数值仿真,理论和数值仿真表明了该方法的有效性.     

Quantum noise analysis of an amplifying TCRR using a continuous-modes formalism

A quantum-mechanical model of the amplifying two-coupler ring resonator based on the continuous-modes theory of quantum optics is used to show certain noise resonances of the output fields.     

Improvement of noise reduction performance for a high-speed elevator using modified active noise control

A high rise building demands a high-speed elevator. Since a high-speed elevator has various transfer paths of noise transmitted from motor and rope to cabin interior, it is very difficult to solve the noise problem. Most research for noise reduction has been performed regarding passive noise control by using mainly absorption material and insulation material. In this study, while it is modeling as multiple-input and single-output with respect to transfer paths of high-speed elevator on conditions of stationary and driving states, the characterized frequency in the cabin is discovered through a contribution technique. It is able to replace by 1-dimensional model to control noise at a major contributed frequency. Also, a new active noise control technique has been proposed to control the cabin noise effectively at unpleasant area that is required to make quite zone for passenger. The Correlation Filtered X-LMS (Co-FXLMS) algorithm has been applied to control the dominant frequency noise that it has a high contribution. Simultaneously, this study has a proposed Moving Band Pass Filter (MBPF) to improve the performance of active noise control in the cabin which is able to apply a dynamic system with time variant states. Finally, we obtained the 8 dB noise reduction in the cabin at ear level and it has been proved that the modified active noise control using Co-FXLMS algorithm and MBPF is available to improve the performance of noise reduction.