کنترل کیفیت صدا فعال حفره ناشی از صدای موتور

Active control solutions appear to be a feasible approach to cope with the steadily increasing requirements for noise reduction in the transportation industry. Active controllers tend to be designed with a target on the sound pressure level reduction. However, the perceived control efficiency for the occupants can be more accurately assessed if psychoacoustic metrics can be taken into account. Therefore, this paper aims to evaluate, numerically and experimentally, the effect of a feedback controller on the sound quality of a vehicle mockup excited with engine noise. The proposed simulation scheme is described and experimentally validated. The engine excitation is provided by a sound quality equivalent engine simulator, running on a real-time platform that delivers harmonic excitation in function of the driving condition. The controller performance is evaluated in terms of specific loudness and roughness. It is shown that the use of a quite simple control strategy, such as a velocity feedback, can result in satisfactory loudness reduction with slightly spread roughness, improving the overall perception of the engine sound.

The successful development of new products relies on the capability of assessing the performance of conceptual design alternatives in an early design phase. In recent years, major progress was made hereto, based on the extensive use of virtual prototyping, particularly in the automotive industry. The state-of-the-art in CAE modeling techniques which can be used for the analysis of time-harmonic acoustic problems is presented in [1]. An overview is given, with automotive interior noise applications, on recently investigated extensions and enhancements to enlarge the application range of different techniques. The efficiency of present CAE techniques allows the use of optimization, e.g., in improving the NVH characteristics of a full-scale engine [2] or a vehicle body [3]. The novelty on this framework is to account for the human perception when defining product performance criteria as in [4] and [5]. Additionally, active control has shown the potential to enhance system dynamic performance which allows lighter and improved products. Research done in the previous years on smart materials and control concepts has led to practical applications with promising results for the automotive industry [6]. However, to make the step to the design of active sound quality control (ASQC), the control schemes, along with appropriate simulation procedures, need to become an integral part of the product development process [7]. In other words, this requires: (i) the product performance metrics to be based on human perception attributes and (ii) the simulation models to support the specific aspects related to smart structures (active systems, actuators, sensors and control logic). In order to demonstrate the proposed simulation procedure and evaluate the effect of active control on the perceived sound quality (SQ), a vibro-acoustic cabin mock-up is selected (Fig. 1). It consists of a simplified car cavity with rigid acoustic boundary condition. The passenger compartment (PC) and the engine compartment (EC) are connected through a flexible firewall which allows noise generated in the EC to be transmitted to the PC. A sound source placed in the EC works as a primary disturbance source. The primary source is driven by a real-time engine simulator, capable of delivering a harmonic excitation based on the engine orders’ amplitude and phase [8]. The controller is based on a collocated structural sensor/actuator pair (SAP) connected to the firewall in a time-invariant velocity feedback loop.The simulation procedure and experimental validation of the active structural acoustic control (ASAC) system are presented in Section 2. The SQ metrics and algorithms used in this paper are reviewed in Section 3. The results, concerning roughness, specific loudness and Zwicker loudness are treated in Section 4. Finally, some general conclusions are addressed in Section 5.

This paper describes a modeling procedure for ASAC, which allows the use of standard vibro-acoustic FE models in the control design. The modeling procedure is experimentally validated, with a vehicle mock-up, for both passive and active configurations. The proposed experimental setup is acoustically excited using a SQ equivalent engine simulator, typically employed in auralization. The use of such scheme allows repeatable measurements with engine-like excitation signals, furnishing results that can be directly correlated to automotive applications. The selected control strategy, collocated velocity feedback, presents satisfactory results for global noise reduction in the PC. The results for loudness attenuation are presented in terms of specific and Zwicker loudness, the latter being linearly related to the human sensation of volume. The results indicate that the effect of the controller is not just noticed locally by the occupants, but is improved rather globally. Finally, the simulation scheme and analysis tools presented here, such as the 3D SQ plots, enables the quantitative assessment of important design parameters, helping the decision making process in an early design phase. The feasibility of these analyses rely on the compact SS representation of the FE model and the fast algorithms used to calculate SQ metrics. The way to improve roughness is through order balancing. The desired order profiles (amplitude and phase vs. rpm) can be defined with the aid of SQ equivalent models. Though, the control strategy must be capable of order balancing. Therefore, a next step in this study will investigate other control strategies, such as adaptive feedforward controllers and their inclusion in the real-time engine simulator framework, for an eventual fully numerical ASQC design platform.