Noise Control Problems Research Articles

Multichannel active noise control and acoustic equalization using fast affine projection algorithms

In the field of adaptive signal processing, it is well known that affine projection algorithms or their low-computational implementations, fast affine projection algorithms, can produce a good trade-off between convergence speed and computational complexity. Although these algorithms typically do not provide the same convergence speed as recursive-least-squares algorithms, they can provide a much improved convergence speed compared to stochastic gradient descent algorithms, without the high increase of the computational load or the instability often found in recursive-least-squares algorithms. In this presentation, multichannel fast affine projection algorithms are introduced for active noise control or acoustic equalization. Multichannel fast affine projection algorithms have been previously published for acoustic echo cancellation, but the problem of active noise control or acoustic equalization is a very different one, leading to different structures. The computational complexity of the new proposed algorithm is evaluated, and it is shown through simulations that not only can the new algorithm provide the expected trade-off between convergence performance and computational complexity, it can also provide the best convergence performance (even over recursive-least-squares algorithms) when non-ideal noisy acoustic plant models are used in the adaptive systems.

The performance surface in nonlinear mean square estimation: application to active noise control problems with correlated signals

This paper investigates the properties of the performance surface for the problem of nonlinear mean-square estimation of a random sequence. The problem studied has direct application to the study of active noise control (ANC) systems when the transducers are driven into a nonlinear behavior. A deterministic expression is derived for the mean-square error (MSE) surface as a function of the system's degree of nonlinearity for Gaussian correlated input signals. It is shown how the presence of the nonlinearity deforms the MSE surface. It is demonstrated that the surface is unimodal, and the expression for the optimum weight vector is determined. The new results are then used to quantify the behavior of ANC systems employing the LMS adaptive algorithm. Important algorithm properties are derived from this study. Examples are presented which verify the analytical models derived.

Reduction of electronic delay in active noise control systems— A multirate signal processing approach

Electronic delay has been a critical problem in active noise control (ANC) systems. This is true whether a feedforward structure or a feedback structure is adopted. In particular, excessive delays would create a causality problem in a feedforward ANC system of a finite-length duct. This paper suggests a multirate signal-processing approach for minimizing the electronic delay in the control loop. In this approach, digital controllers are required in decimation and interpolation of discrete-time signals. The computation efficiency is further enhanced by a polyphase method, where the phases of low-pass finite impulse response (FIR) filters must be carefully designed to avoid unnecessary delays. Frequency domain optimization procedures based on H1, H2, and Hinfinity norms, respectively, are utilized in the FIR filter design. The proposed method was implemented by using a floating-point digital signal processor. Experimental results showed that the multirate approach remains effective for suppressing a broadband (200-600 Hz) noise in a duct with a minimum upstream measurement microphone placement of 20 cm.

THE USE OF MULTIPLE ACTUATORS IN THE ROBUST CONTROL OF AN ACOUSTIC DUCT

The paper considers the problem of robust active noise control in an acoustic duct. In particular, the paper considers how the number of actuator speakers affects the performance of the system. The robust controller design methodology being considered is based on some recent results on minimax LQG control. The paper also uses a structured approach to uncertainty modelling as part of the controller design scheme.

Active Control of Noise Transmission using Self-Organizing Neural Network.

The neural network has been applied to many engineering problems such as pattern recognition, optimization, system identification and control because of its nonlinear mapping capability. Although the network has a great ability to acquire solution for a given problems through learning, however, several problems are known such as the convergence to a local minimum solution, and the tuning of network parameter manually to obtain desired accuracy. In the present paper, we deal with the self-organizing neural network where the spatially scattered neurons are connected automatically using Cellular Automata. The network is applied to the active noise control problem as an adaptive system, where the model consists of a simply supported plate embedded in an infinite rigid baffle above a rectangular cavity. The fully coupled acoustic-plate interaction equations including characteristics of piezoelectric transducers are solved using time-varying Green's function techniques, and the performance of the controller is discussed through numerical investigation as well as experiment.

Structural Acoustics and Vibration

First Page

Applicability of Thompson boundary conditions to obliquely incident acoustic waves

A numerical study employing the Euler equations in a two-dimensional duct flows is presented. Thompson boundary conditions imposed at the inflow/outflow boundaries indicated bounded growth of absolute error with increasing angle of incidence of oblique acoustic disturbances, supporting the suitability of these boundary conditions for acoustic propagation and noise control problems.

Energy source identification using reactive structural intensity

The analysis of wave propagation phenomena in structures, especially energy flux, can be a powerful approach to noise and vibration control problems. The active mechanical intensity corresponds to the period-averaged value and is related to the propagative wave field. The reactive intensity is related to the reverberance of energy in the structure, which originates the standing waves or mode shapes. Properly separated into its wave components, reactive intensity maps agree well with operational modes, indicating the location of the nodal lines. On the other hand, active intensity is the quantity usually addressed when solving vibration and noise problems, since it provides information about the main energy flow paths, thus enabling identification of the energy sources and sinks. For highly reverberant structures, however, measuring active intensity becomes awkward, and intensity plots fail to indicate the region where the energy is injected or dissipated in the structure. A new technique to localize energy sources based on the divergence of the reactive intensity and the distribution of the potential and kinetic energy densities within the structure is presented. Numerical results are given for a beam with a point excitation using finite-element and spectral element models based on the Bernoulli–Euler beam theory.

Active Shielding and Control of Noise

We present a mathematical framework for the active control of time-harmonic acoustic disturbances. Unlike many existing methodologies, our approach provides for the exact volumetric cancellation of unwanted noise in a given predetermined region of space while leaving unaltered those components of the total acoustic field that are deemed friendly. Our key finding is that for eliminating the unwanted component of the acoustic field in a given area, one needs to know relatively little; in particular, neither the locations nor structure nor strength of the exterior noise sources need to be known. Likewise, there is no need to know the volumetric properties of the supporting medium across which the acoustic signals propagate, except, perhaps, in the narrow area of space near the boundary (perimeter) of the domain to be shielded. The controls are built based solely on the measurements performed on the perimeter of the region to be shielded; moreover, the controls themselves (i.e., additional sources) are also concentrated only near this perimeter. Perhaps as important, the measured quantities can refer to the total acoustic field rather than only to its unwanted component, and the methodology can automatically distinguish between the two. In the paper, we construct a general solution to the aforementioned noise control problem. The apparatus used for deriving the general solution is closely connected to the concepts of generalized potentials and boundary projections of Calderon's type. For a given total wave field, the application of Calderon's projections allows us to definitively split its incoming and outgoing components with respect to a particular domain of interest, which may have arbitrary shape. Then the controls are designed so that they suppress the incoming component for the domain to be shielded or alternatively, the outgoing component for the domain, which is complementary to the one to be shielded. To demonstrate that the new noise control technique is appropriate, we thoroughly work out a two-dimensional model example that allows full analytical consideration. To conclude, we very briefly discuss the numerical (finite-difference) framework for active noise control that has, in fact, already been worked out, as well as some forthcoming extensions of the current work: optimization of the solution according to different criteria that would fit different practical requirements, applicability of the new technique to quasi-stationary problems, and active shielding in the case of the broad-band spectra of disturbances. In the future, the aforementioned finite-difference framework for active noise control is going to be used for analyzing complex configurations that originate from practical designs.

F-1032 自己組織化ニューラルネットワークによる能動騒音制御(G10-5 音響・騒音制御)(G10 機械力学・計測制御部門一般講演)

The neural network has been applied to many engineering problems such as pattern recognition, optimization, system identification and control because of its nonlinear mapping capability. Although the network has a great ability to acquire solution for a given problems through learning, however, several problems are known such as the convergence to a local minimum solution. In the present paper, we deal with the self-organizing neural network where the spatially scattered neurons are connected automatically using Cellular Automata method. The network is applied to the active noise control problem as an adaptive system, where the model consists of a simply supported plate embedded in an infinite rigid baffle above a rectangular cavity. The performance of the controller is discussed through numerical investigation.

Active noise control studies using the Rayleigh–Ritz method

The Rayleigh–Ritz method is used to model a smart composite plate and the piezoelectric patches attached to it. Classical plate theory is used to model the composite plate and electroelastic theory is used to model the segmented piezoelectric patches. Eigenfunctions of a clamped–clamped beam are used as the Ritz functions for the panel and the rigid cavity modes are used to model the acoustic cavity. The dynamic equations of motion for the coupled smart panel-cavity system are derived using Hamilton’s principle. The forcing term due to the cavity acoustic pressure is determined by using virtual work considerations. Five collocated pairs of actuator/sensor pairs are attached to the plate at a predetermined placement scheme. A (multi-input–multi-output) MIMO controller is designed for the coupled vibroacoustic system. The output feedback controller is then employed to emulate the optimal controller by solving the Riccati equations. The performance of the control system in reducing the vibration and the noise transmitted into the enclosure is studied. It is shown that the Rayleigh–Ritz procedure can be used more efficiently than the finite-element method for cabin noise control problems.

ACTIVE CANCELLATION OF NOISE IN A CAR CABIN USING THE ZERO SPILLOVER CONTROLLER

In active noise control problems, spillover will occur because of the Bode integral constraint if the performance sensors and measurement sensors are collocated, or if the control speakers and the noise sources are collocate. This paper is focused on the implementation of a multiple-channel zero-spillover ANC controller using a spatially feedforward structure for a car cabin. The plant model is realized by using a frequency-domain procedure. The linear-quadratic-Gaussian algorithm is used in controller design. Experiments are carried out to verify the proposed technique, where random noise and engine noise are chosen as the primary noise. The results indicate that the proposed controller yields a broadband attenuation for the random noise. The proposed method is also effective in attenuating the engine noise if an assumed model of the noise is incorporated into controller design.

Le Chatelier’s Principle in noise control

Le Chatelier’s Principle is perhaps better known to those who studied chemistry at the university. Therefore, I find that few in the noise control community are familiar with this principle. This paper features some examples of noise control problems in which ‘‘the Principle’’ plays a role. For noise control purposes and for the record, Le Chatelier’s Principle states that ‘‘when a dynamic system is modified with the intention of achieving a high degree of change in the noise issued by this dynamic system, the achievement is mitigated by the implementation of the modification.’’

BROADBAND DISTURBANCE ATTENUATION OVER AN ENTIRE BEAM

This paper considers a problem of disturbance attenuation for a pinned–pinned flexible beam. The beam is modelled as a distributed parameter system and a procedure is developed to solve the corresponding disturbance attenuation problem for such a distributed parameter system. A controller is designed in a way that minimizes the effect of disturbances over the entire beam. The controller design approach of this paper is applicable to a large number of active noise and vibration control problems.

The robust SPR problem: Design algorithms and new applications

Practical methods to solve the robust strictly positive real (SPR) problem are presented in this paper. Passivity conditions, in the form of an SPR requirement, are sufficient for the convergence of the hyperstable family of adaptive IIR algorithms. The robust SPR problem tries to get the satisfaction of the mentioned passivity condition for an uncertain set describing the unknown system. We show how this robust problem can be practically solved, considerably extending previous works on the topic. We also find another important area for the robust SPR problem in the active noise control problem. Adaptive algorithms for active noise control are more challenging than those used in classical scenarios; the error cannot be accessed, but a filtered version of it. This imposes a certain knowledge of the frequency response which filters the error, in such a way that its phase must be estimated with an error in the range (− π/2, π/2) . The robust SPR problem can take care of different transfer functions corresponding to varying operating conditions, thus yielding robust adaptive algorithms.

Quiet! Okay, But How?

This article focuses on pneumatic tools that are among the most damaging sources of noise pollution. The National Institute for Occupational Safety and Health estimates that more than 30 million workers in the United States are exposed to hazardous noise, costing the economy about $1 billion every year. There is no magic solution to all noise control problems, and this remains true even when we are dealing just with pneumatic tools. Each case must be considered separately, and the solution to be used must satisfy the often conflicting demands of at least five criteria simultaneously. The sound power level produced by a pneumatic tool is a product of many interrelated parameters: operating pressure, pressure drop, expansion volume, exhaust air velocity, speed of device, exhaust air, airflow, length of exhaust path, and tool power. Good standards are available for evaluating the noise of many components, and suppliers provide data on their products that can be used for comparison.

Evaluation of an Actuator Placement Method for Active Noise Control Applications

In our previous work, we developed a new actuator placement algorithm that is capable of selecting the best actuator placement for active noise control problems over a broad band of frequencies. The actuator selection algorithm is based on a novel extension of the Householder QR subset selection algorithm. The QR algorithm uses the l2 matrix norm as a performance measure. In this paper, numerical results generated by that algorithm are compared with numerical results generated using five different performance measures. These measures, which are based on different matrix norms and functions of the actuator frequency responses, yield actuator placements that result in active noise control systems with improved performance and robustness.

Plant uncertainty analysis in a duct active noise control problem by using the H∞ theory

Plant uncertainty is one of the major contributing factors that could affect the performance as well as stability of active noise control (ANC) systems. Plant uncertainty may be caused by either the errors in modeling, computation, and measurement, or the perturbations in physical conditions. These factors lead to deviations of the plant from the nominal model, which will in turn affect the robustness of the control system. In this paper, the effects due to changes in physical conditions on the ANC system are investigated. The analysis is carried out in terms of performance and robustness by using a general framework of the H∞ robust control theory. The size of plant uncertainty is estimated according to the infinity norm of the perturbations to physical conditions, which provides useful information for subsequent controller design that accommodates both performance and stability in an optimal and robust manner. The guidelines for designing the ANC systems with reference to plant uncertainties are also addressed.

ACTIVE NOISE CONTROL OF ACOUSTIC SOURCES USING SPHERICAL HARMONICS EXPANSION AND A GENETIC ALGORITHM: SIMULATION AND EXPERIMENT

A general method is presented to optimize transducer location in an active noise control problem. The method includes two parts. First, the actuator configuration is determined by using a model of the primary field which is a spherical harmonics expansion. In the second part, a genetic algorithm is used to determine the error sensor configuration. This method is then applied to two real acoustic sources: a dipole and an electrical transformer. In numerical simulations, the primary field of both sources measured in an anechoic room was used to determine the active control configurations. Then, the actuator and error sensor arrangement was tested in an active control experiment involving both primary, sources.

Noise control under successive fulfillment of jobs specified by various degrees of labor tension

The Standards of occupational noise (A-weighted) in Russia, the Ukraine, and other countries in the former USSR are differentiated by degrees of labor tension (40–80 dB). Frequently there is a real necessity to solve the problem of noise control, which is summed up for a full-time day during which labor is exposed to different noise levels. The problem is addressed by the definition of equivalent sound shift level—the noise working loudness. This definition correctly sets forth one noise standard during a shift of one action, i.e., the fulfillment for a particular period by a worker of identical labor tension. The decision of this task is complicated by the presence during the shift of work distinguishing not only the noise levels but different labor tension. Under the circumstances, it is advisable to sum up noise control according to the quotient of noise loudness (QNL). It is defined by the overage in time exceeding the standards of sound level for every separate period of labor with different degrees of labor tension. The quotient of noise loudness (QNL) is calculated by the proposed formula. It allows one to make noise control under successive fulfillment of jobs specified by various degrees of labor tension.