Personal Sound Zones and Shielded Localized Communication through Active Acoustic Control

The ability of active control systems to reduce propulsion-related noise tones in aircraft cabins is well documented. However, for a variety of reasons, analogous solutions for broadband random noise have yet to be demonstrated. In contrast to tonal noise problems, broadband noise sources (such as turbulence) tend to be distributed both spatially and spectrally, and cannot be isolated to simple structural-acoustic transmission paths. Consequently, the design of a broadband noise control system poses additional and more demanding constraints than are encountered in tonal controller design. In this paper, we describe initial laboratory tests of a multichannel active control approach being developed for broadband random noise control in aircraft cabins. The tests were conducted in a transmission loss (TL) facility with a flat aluminum panel (1.75m x 1.12m) mounted in the test and exposed to broadband acoustic excitation. The noise transmitted through the panel was measured with a spatially distributed array of microphones. A multichannel controller, coupled with an acoustic sensor and eight piezoceramic actuators (bonded to the panel surface), was used to reduce the transmitted noise in the 200-2000 Hz band throughout a large quiet zone. Overall noise reductions of up to 18 dB were demonstrated. Introduction In the last 10 years, active control of low-frequency tonal noise in aircraft cabins has progressed from the laboratory to commercial market acceptance. Active noise control techniques can successfully reduce aircraft cabin noise caused by engine or propeller tones in the low-frequency region (approx. 50-300 Hz). However, broadband random noise (e.g., due to boundary layer turbulence acting on the exterior of the aircraft) is a major contributor to cabin noise in many commercial aircraft, and is dominant in the mid-frequency range (approx. 300-1500 Hz). Conventional passive noise control treatments, though effective in the highfrequency range, impose unacceptable weight penalties when extended to treatment of low-frequency noise. It is therefore desirable to consider active control alternatives. Active acoustic noise control approaches employ cancelling sound sources (i.e., speakers) to control sound in enclosed spaces such as an aircraft cabin. In the mid-frequency range, acoustic control requires many cancelling sources to match the temporal and spatial sound patterns radiated from complex fuselage structures. Active structural-acoustic control (ASAC), on the other hand, employs control forces—applied directly to the vibrating structure—to minimize the sound transmitted through the structure. There is some indication that (for tonal noise control problems, at least) structural actuators are preferable to acoustic actuators, due to their closer match with the primary source patterns in the cabin. Also, structural transducers are reasonably light weight, compact, and amenable to integrated, smart structure designs. In this paper, we describe results of broadband ASAC applied to a large aluminum panel— representative of a fuselage section—using various test configurations, including details of the transmission-loss test facility and measurements, the controller architecture and optimization criteria, and real-time broadband ASAC results for both random and repetitive noise sources. The issue of time-delay, introduced by the actuator/panel dynamics, is treated in an appendix. Test facility and measurements Figure 1 shows a plan-view schematic of the McDonnell Douglas transmission-loss facility in Long Beach, CA, where the tests were performed. The facility has two anechoic rooms separated by a window in which a fuselage test panel is mounted. Acoustic noise is generated by a loudspeaker in the source room and transmitted through the panel into the receiver room, where the transmission loss of the panel is measured. For the broadband ASAC tests, the following electromechanical transducers complemented the facility: • An acoustic loudspeaker (the primary noise source) • A microphone sensor, mounted directly in front of the loudspeaker • An array of 14 microphone receivers (R1-R14), mounted in two coplanar arcs • An array of 8 electronically controlled piezoceramic (PZT) actuators (A1-A8) bonded to the panel. The microphone array was used to sample the noise pattern in a wide arc through the receiver room, roughly coplanar with the opposing loudspeaker source. Coarse sampling in the out-of-plane dimension was done by lowering the entire array. The placement of actuator patches on the panel is shown in Figure 2. Figure 3 and Figure 4 show the source and receiver rooms, respectively. Real-time implementation of the ASAC controller was performed by SRI's Advanced Signal Processor (ASP). For these tests, the ASP was configured with 1 control input, 8 control outputs, and up to 16 error inputs. Data were sampled at 13.3 kHz, and control filters were 2048 taps (-150 ms) long. Copyright ©1996 by the American Institute of Aeronautics and Astronautics, Inc. Noise Source (loudspeaker)

A demonstration of active acoustic control of radiation from a simply supported plate is presented. First, open‐loop characteristics of a 0.5×0.6‐m rectangular plate are shown to validate the assumption of nearly simply supported boundary conditions. Next, a state‐space design method for closed‐loop control of acoustic radiation is discussed. The design procedure is used to generate controllers that reject either narrowband or broadband point‐force disturbances of the plate's first four resonant modes. Simulation and experimental results are provided to illustrate the difference between active vibration control and active acoustic control. Radiation filters, derived from the radiation efficiencies of the modal basis, are implemented for these subsonic acoustic control experiments. During the acoustic cost experiments, the controller places a priority on actively damping only those modes that are contributing significantly to the acoustic radiation. The results of this active acoustic control experiment validate the built‐in selectivity of this control approach for suppression of structural acoustic radiation. The theoretical foundation for these experimental results is given in a companion paper, “Active acoustical control of broadband structural disturbances” [J. Acoust. Soc. Am. Suppl. 1 88, S 149 (1990)]. [Work supported by DARPA/ONR.]

The acoustic impedance at the diaphragm of an electroacoustic transducer can be varied using a range of basic electrical control strategies, amongst which are electrical shunt circuits. These passive shunt techniques are compared to active acoustic feedback techniques for controlling the acoustic impedance of an electroacoustic transducer. The formulation of feedback-based acoustic impedance control reveals formal analogies with shunt strategies, and highlights an original method for synthesizing electric networks ("shunts") with positive or negative components, bridging the gap between passive and active acoustic impedance control. This paper describes the theory unifying all these passive and active acoustic impedance control strategies, introducing the concept of electroacoustic absorbers. The equivalence between shunts and active control is first formalized through the introduction of a one-degree-of-freedom acoustic resonator accounting for both electric shunts and acoustic feedbacks. Conversely, electric networks mimicking the performances of active feedback techniques are introduced, identifying shunts with active impedance control. Simulated acoustic performances are presented, with an emphasis on formal analogies between the different control techniques. Examples of electric shunts are proposed for active sound absorption. Experimental assessments are then presented, and the paper concludes with a general discussion on the concept and potential improvements.

Generally, global active noise control is to minimize the sum of the energy of the residual sound field. But on some occasions, we are more concerned that the energy of the residual noise at every direction does not exceed a certain value. In this paper, an algorithm for global active noise control is introduced to achieve a global active acoustic radiation noise control by minimizing the maximum residual sound energy of the far field after active acoustic radiation control. The proposed algorithm adjusts the weights of the secondary sound sources based on the min-max optimization. Simulation results show that the proposed algorithm can reduce the maximum of the far-field residual sound pressure level 2.2 dB more than the maximum of the residual sound pressure level based on the traditional global acoustic radiation control algorithm.

Piezoelectric sensor and actuator has been highly concerned in active control with its perfect properties. In this paper, a refined hybrid enhanced assumed strain piezoelectric shell element and acoustical isoparametric element were introduced to investigate coupled problem of piezoelectric-structure acoustic. First, a formulation to calculate the response of coupled problem was presented. Second, the negative-velocity constant gain feedback strategy was used for control algorithm; Third, an enclosed-cavity with one flexible wall was adopted to investigate active structure acoustic control. Numerical results demonstrate that the application of piezoelectric materials on control of the acoustic radiation is rational and efficient in the low frequency range.

The concept of active acoustic control was recently introduced by Mansfield and Haywood (MAGMA 2000:10:147-151) to ameliorate the problem of acoustic noise from MRI, particularly that from high-speed EPI. A 30 dB reduction in noise was previously achieved with the use of acoustic control operating at spot frequencies within a narrow band. In this work, a new acoustic gradient pulse is introduced that comprises an oscillating gradient of finite duration, incorporating a combination of frequencies within this band designed for use as the switched read gradient in echo-planar imaging (EPI). Employing this pulse with active acoustic control results in a reduction of acoustic noise by 50 dB.

In this paper, the active control of sound transmission through double-panel partitions by inserting acoustic sources in the air gap between the double panels is studied through a computer simulation. The work is an extension of the previous analytical work in which a theoretical framework was developed to reveal the control mechanisms involved in active acoustic control. In this work, a computer simulation based on the theoretical framework is used to assess the control performance under different conditions. It is shown that the property of the radiating panel plays an important role in active acoustic control. The control performance can be improved by increasing the panel damping or decreasing the panel modal density. The reasons for this are discussed.

The active acoustic and structural noise control characteristics of a double wall cylinder with and without ring stiffeners were numerically evaluated. An exterior monopole was assumed to acoustically excite the outside of the double wall cylinder at an acoustic cavity resonance frequency. Structural modal vibration properties of the inner and outer shells were analyzed by post-processing the results from a finite element analysis. A boundary element approach was used to calculate the acoustic cavity response and the coupled structural-acoustic interaction. In the frequency region of interest, below 500 Hz, all structural resonant modes were found to be acoustically slow and the nonresonant modal response to be dominant. Active sound transmission control was achieved by control forces applied to the inner or outer shell, or acoustic control monopoles placed just outside the inner or outer shell. A least mean square technique was used to minimize the interior sound pressures at the nodes of a data recovery mesh. Results showed that single acoustic control monopoles placed just outside the inner or outer shells resulted in better sound transmission control than six distributed point forces applied to either one of the shells. Adding stiffeners to the double wall structure constrained the modal vibrations of the shells, making the double wall stiffer with associated higher modal frequencies. Active noise control obtained for the stiffened double wall configurations was less than for the unstiffened cylinder. In all cases, the acoustic control monopoles controlled the sound transmission into the interior better than the structural control forces.

Within the Brite-Euram Project “Anrava” several active control systems have been evaluated in order to reduce the road noise generated inside the cabin of a mid size station wagon during driving condition on different road surfaces. Different control strategies (feedforward with local and global control, pure feedback) and different approaches (active structural acoustic control and active noise control) have been tested on the car in laboratory conditions and the control performances have been evaluated. The final control strategies that have been selected based on the results of this study, are based on an adaptive feedforward control algorithm: six accelerameters provide the reference signals and 4 microphones placed inside the cabin give the error feedbacks. Two control configurations, each using a different kind of control source, have been retained : a structural acoustic control system which works with 6 inertial shakers positioned at the main vibration transmission paths of the car suspension, an anti-noise system with 4 loudspeakers inside the cabin. The paper describes the approaches utilised to design the different control systems and presents the results obtained during laboratory and road tests, comparing them with the performances predicted by numerical simulations.

The development of active acoustic control experiments demonstrates the need to calibrate our view of the physical world with regular testing. We started with a desire to use feedback-based active control on acoustic systems with a system model based on the wave equation and the opinion that all work would flow smoothly from that model of physical reality. We have now realized that the hardest issues associated with active acoustic control are not well described by our simplistic initial model. Physical testing is the vehicle through which we tested our initial model. With physical testing and experimentation, we have focused on those elements of acoustics that are most important to the success of our acoustic control methods. The control experiment caused a change in research direction from control design to actuator design.

Actuator performance is a critical part of active noise and acoustic control. The loudspeakers that are normally used as actuators in many active noise and acoustic control applications add significantly to the dynamics of the control loop and are detrimental to the controller’s performance. By compensating a loudspeaker with a technique similar to motional feedback, the loudspeaker performance is enhanced in applications such as control of acoustic enclosures. In this work, a method to easily and reliably compensate a loudspeaker in order to approximate constant volume velocity behavior over the piston-mode frequency range is presented and demonstrated. This reduces the influence of the environment upon the actuator, reduces low-frequency distortion of the speaker, and minimizes magnitude and phase shifts of the intended control signal. Numerical simulations and experimental results of the proposed methods are included. [Work supported in part by the Lord Corporation and the National Science Foundation under Grant No. CMS 95-01470.]

The PADK is a highly versatile DSP platform for implementing real-time audio applications. It features the Texas Instruments TMS320C6727, a powerful floating-point DSP processor and can process audio from eight input channels simultaneously. This makes it an ideal platform for developing multi-channel active acoustic noise control algorithms. This paper demonstrates results of system identification and feedback control using this kit. Using the Least Mean Square (LMS) algorithm, the filter coefficients are updated in real time to obtain an accurate impulse response. Potential applications of this particular platform are in the implementations of active acoustic noise control (ANC) systems in real time.

An active acoustic power control method for three-dimensional space was investigated both theoretically and experimentally. In this paper, the focus is on the typical case of a single primary and two secondary point sources, and methods of minimizing the power output were considered. Suitable arrangements of the sources and microphone were proposed on the basis of theoretical estimations. The authors call this the acoustic nodal point method. The power reducing performance was verified by applying the method to the noise radiating from a duct opening.

We present the first demonstration of active acoustic control using optical fiber sensors. The fiber sensors are oriented symmetrically about the vertical centerline of a thin, simply supported baffle plate and are configured for the most effective detection of the odd-odd plate modes. The output from the fiber sensor is used as an error signal, a least-mean-square algorithm is used for digital processing, and control is achieved using a single piezoelectric actuator. Preliminary results obtained from the use of elliptical-core, two-mode fiber sensors attached to the plate show effective acoustic attenuation of 25 dB in on-resonance cases and 10 dB in off-resonance cases.

The purpose of this research is to investigate the feasibility of utilizing the adaptive eigenvector optimization algorithm to find the optimal combination of left and right eigenvectors for active structural acoustic reduction. As depicted in the previous studies, the structural acoustic radiation depends on the structural vibration behavior, which is strongly related to the mode shape distributions (represented by the right eigenvectors of system) as well as the ability of disturbance rejection (represented by the left eigenvectors of system). In this research, a novel adaptive optimization algorithm is developed to determine the optimal combination of left and right eigenvectors of the structural system for active structural acoustic control. The sound suppression performance index (SSPI) is defined by combining the orthogonality index of left eigenvectors and the modal radiation index of right eigenvectors. Through the proposed adaptive eigenvector optimization algorithm, both the left and right eigenvectors of the structural system are adjusted such that the SSPI quantity decreases, hence the optimal combination of the corresponding left and right eigenvectors of the closed-loop system can be found for reducing the structural acoustic radiation. The optimal combination of left-right eigenvectors is then synthesized to determine the feedback gain matrix of the closed-loop system. The result of the active acoustic control shows that the proposed method can significantly suppress the sound pressure radiated from the vibrating structure.