Developing an Active Vibration Cancellation System with LabVIEW Real-Time and PXI

Author(s):
Dr. Martin Pflüger - AVL

Industry:
Automotive

Products:
LabVIEW, PXI/CompactPCI,

The Challenge:
Developing an active vibration cancellation (AVC) system to reduce vibrations that were transferred from the car engine to the car body.

The Solution:
Providing a PXI-based system that uses LabVIEW Real-Time to perform an active vibration cancellation in real time for a flexible test stand.

An Innovative Approach to Noise Control
As the automotive industry uses increased numbers of industrial equipment such as engines, blowers, fans, transformers, and compressors, acoustic noise problems become more evident. The traditional approach to acoustic noise control uses passive techniques such as enclosures, barriers, and silencers to attenuate the undesired noise. These passive silencers prove useful for their high attenuation over a broad frequency range; however, they are relatively large, costly, and ineffective at low frequencies. Mechanical vibration is a related type of noise that commonly creates problems in all areas of transportation and manufacturing.

AVC involves an electromechanical system that cancels the primary, unwanted vibrations based on the principle of superposition. This occurs when an antivibration of equal amplitude and opposite phase generates and combines with the primary vibration, thus resulting in the cancellation of both vibrations. The AVC system efficiently attenuates low-frequency noise where passive methods are either ineffective or tend to be expensive or bulky. The AVC obtains information about the possible improvement of some reading points. We can immediately measure decimation of the sound level in the car interior.

AVC System Requirements
We based our AVC system on the LabVIEW Real-Time platform to perform a deterministic and reliable execution of the algorithm, which uses NI PC-based data acquisition (DAQ) for hardware in the loop (HIL). In addition to NI hardware, we use sensors, actors, and amplifiers to measure and control the vibration propagation from the car engine to the car body in different directions.

With the AVC system, we can cancel vibrations depending on the engine speed. We can cancel three selectable orders of frequencies at three analog channels simultaneously. This configuration requires high-execution performance in a deterministic way by a LabVIEW Real-Time application.

The AVC uses an opposite feedback to counteract the undesired vibration of the engine. The AVC algorithm works as an adaptive method and generates a sine wave depending on engine speed, signal input, and the result of the system identification to ensure an appropriate magnitude, frequency, and phase-correct sine wave for analog output to cancel the undesired vibration.

NI-Based System
We based the system on the LabVIEW Real-Time PXI-8170 embedded controller with a PXI-6711 analog output board, a PXI-6071 multifunction I/O board for the analog inputs implemented in a PXI-1000B chassis and the connector block CA-1000. In addition to the National Instruments components shakers, we used accelerometers, power amplifiers, measurement amplifiers and a laptop computer.

Because we developed the software with LabVIEW Real-Time and the Application Builder, the test engineer experienced quick and easy operation.

We chose LabVIEW Real-Time as the programming language so we could control and process data in realtime and in a secure way. We also selected LabVIEW Real-Time because the software operates on a regular Windows computer, and the compiled software code downloads to the embedded real-time controller.

A Flexible Software Solution
The software integrates three basic steps: setup, system identification and AVC routine. The system is self adjusting through digital adaptive filters.

During the software development phase, we placed high priority on designing a flexible and easy-to-use system for daily use on the chassis roller test stand. With the setup menu, we can select the required channels and channel numbers and enable and disable shakers and accelerators.
Furthermore, we can vary critical parameters like execution speed and algorithm-relevant process parameters.

System Identification
The employment of adaptive filters for AVC applications is complex because the summing junction of the disturbing and the canceling vibrations represents mechanic superposition. Therefore, we must compensate the secondary-transfer path functions, which include the D/A converters, reconstruction filters, power amplifiers, shakers, and the mechanical paths from the shakers to the accelerometers. The system identification establishes these paths.

The system identification virtual instrumentation (VI) displays the adaptation process of the adaptive filter to the mechanical system under investigation. The error signal gives an indication of the quality of the adaptation process. Ideally, the error signal converges towards zero. Additionally, the calculated filter coefficients of the 300 coefficient adaptive filter are shown. In the given case, we executed 20,000 adaptation steps to obtain the filter coefficients.

Depending on the number of the selected accelerometers and shakers, we must execute a specific number of runs for the system identification.

Active Vibration Cancellation Routine
The system can cancel up to three vibration directions – each having up to three frequencies simultaneously. The AVC VI shows the system at a special breadboard construction with a pure sine wave stimulation. The error signal is the indicator for the quality of the cancellation process. In addition to accelerometer and shaker signals, the rotation speed of the engine is important for the cancellation routine.

Reducing Airborne Noise in the Passenger Compartment
We used LabVIEW Real-Time and National Instruments PXI hardware to build a compact and flexible AVC system for the cancellation of vehicle vibrations. With this system, we can reduce three selectable engine orders for each of the three analog input channels. The algorithm runs in a parallel execution for each channel. Results at the vehicle acoustics chassis test stand show an achievable airborne noise reduction of 3 dB in the passenger compartment.
With LabVIEW Real-Time, we found it relatively simple to develop a parallel execution of the real-time software code.
The graphical user interface, which monitors the time-critical data and processes of the system, is a valuable tool during the development process.

For more information, contact
Dr. Martin Pflüger, AVL
Tel: 43-316-787-2304
Fax: 43-316-787-1450
E-Mail: martin.pflueger@avl.com

Browse All Case Studies »