When Electric Drives Change the Rules of NVH Analysis

A leading international OEM conducts extensive NVH comparative measurements on production, competitor, and prototype vehicles. With increasing electrification and model diversity, the requirements for the correct interpretation of measurement data are rising sharply.

While key reference values such as crankshaft speed or ignition sequence can be recorded directly and in high resolution in combustion engines, electrification significantly increases the complexity of NVH reference generation and signal interpretation.

From an NVH perspective, the electric drive is often only accessible to a limited extent as a clearly referenceable excitation source:

• Relevant rotor speeds are often not directly measurable, but only indirectly available via vehicle buses (e.g., CAN FD) or substitutively via output or wheel speeds.
• In multi-motor architectures, electromagnetic motor orders, gear mesh excitations, inverter switching harmonics, and structural resonances may overlap spectrally.
• Gear ratios, differentials, and wheel-induced slip lead to additional decoupling between the internal rotor arrangement and the external measured variable.

This shifts the NVH analysis from a speed-triggered order evaluation to a more signal-driven, path-based, and model-supported interpretation.

The outcome is a labor-intensive, error-prone process with limited comparability across vehicle programs.

• For combustion engines: Engine speed usually directly measurable > simple order analysis
• For electric vehicles:
  • Speed often only indirectly available (e.g., indirect measurement of the outputs or via ECU/CAN where available)
  • Multiple motors > superimposed NVH signatures
  • Gear ratios, differential + wheel slip > deviations between wheel speed and drive speed

Important: Relevant excitation references often need to be derived from indirect vehicle parameters rather than being directly available as measurement channels. Without DBC/file descriptions (third-party vehicles), direct interpretation is not possible – indirect or model-based references are required.

Each vehicle has individual parameter combinations:

• Transmission ratios
• Number of motors
• Axle concepts
• Differential types

• Perform measurement
• Research transmission parameters
• Manually update parameters
• Manually reconfigure analyses
• Risk of inconsistent results

Important: Particularly critical for benchmark measurement programs across vehicle fleets.

Our customer implemented a cloud-based NVH parameter pipeline that combines vehicle knowledge, measurement data, and analysis automation.

• Automated reconstruction of relevant NVH references eliminates large parts of the manual analysis preparation effort
• Analysis configurations are derived automatically from centralized engineering knowledge
• Measurements can be processed immediately after data upload without manual reconfiguration

• Reduced misconfigurations
• Uniform evaluation standards across programs

• New vehicle variants are integrated through knowledge maintenance rather than analysis redevelopment
• The same analysis pipeline can be reused across large and continuously evolving vehicle portfolios
• Engineering expertise is captured once and applied consistently at scale

The solution establishes a scalable foundation for data-driven NVH engineering by combining cloud-based analysis pipelines, automated workflows, and centralized vehicle knowledge.

It enables consistent evaluation across increasingly complex vehicle architectures while creating the basis for software-defined vehicles, AI-supported engineering processes, and future connected, autonomous, shared, and electrified mobility concepts.