← Back to Use Cases

Ride Optimisation Considering Vehicle Mass Property Variation

Authors:
Name Affiliation Phone Number Email Address
Jacqui Morison Jaguar Land Rover arichar4@jaguarlandrover.com
Claire Freeman Jaguar Land Rover arichar4@jaguarlandrover.com
Andy Richardson Jaguar Land Rover arichar4@jaguarlandrover.com
Industrial Sectors:

Automotive Engineering

1. DESCRIPTION OF USE CASE

Introduction

At Jaguar Land Rover the ultimate objective of the Vehicle Dynamics Team is for the prototype vehicles to be ‘signed off’ by the Vehicle Evaluation Team. The Vehicle Dynamics attribute includes ride, steering & handling assessment. Due to the cost of prototypes the sign-off process has its limitations in that it is focussed on a discrete number of vehicle variants. It does not consider the performance variability that exists across the full range of variants and/or specification levels. Each individual vehicle will have different mass and inertia properties affecting the ride performance. Therefore it is currently difficult to optimise the ride performance of the vehicle across the entire variant/specification range of a vehicle programme. A potential extension of the use case could be to incorporate the influence of component tolerances and to   additionally optimise steering attribute performance. 

Design/Assessment ObjectivesTo optimise the ride performance of the vehicle to achieve the ride targets for the programme, within programme assumptions.Note: The objective ride target definition for a programme also has uncertainties associated with it since it is cascaded down from subjective evaluation. Design/Assessment Parameters

  • Spring Rate
  • Damper Rate
  • Bush Rates
  • Mass Properties

Assessment Process

The assessment process is to measure a vehicle virtually over a range of standard ride surfaces. From this data a series of metrics are extracted. These are used for quantifying the ride attribute performance during development and have been developed to reflect subjective vehicle assessment through objective measurement. The metrics can be divided into ‘primary ride’ and ‘secondary ride’.

  • Primary Ride:  The ability of a vehicle’s body to follow a road surface.  For example a vehicle exhibiting good primary ride will have less suspension motion as the body stays connected to the wheels that are following the road profile.  Primary ride is assessed vertically and in pitch and roll directions.
  • Secondary Ride:  The level of vibration experienced by the occupants in the frequency range 3-30Hz.  This is typically due to masses moving relative to the body, for example wheels, engine, exhaust, and structural vibrations within the body itself all being excited by the road input. Secondary ride can further be divided into ‘chop’ and ‘shake’.

UQ&M Aspirations/Objectives

Since it is not understood how the variation in mass properties affect the distribution of ride from vehicle to vehicle and because only a discrete number of prototypes are available for tuning, current ride tuning does not optimise all possible variants/specifications.

  • Determine a process to optimise vehicle ride performance for all potential customer vehicle variants & option specifications. Improved prediction of ride variation to minimise the number of component variables required to achieve desired ride performance across the programme.
  • Is there a better method to optimise ride performance across a range of uncertain mass properties?
  • Potential extension of the use case to incorporate the influence of component tolerances and to additionally optimise steering attribute performance.

2. KEY UQ&M CONSIDERATIONS
2.1 Process Inputs

Mass Properties

  • Mass (kg)
  • Mass Distribution (% Fr)
  • CG Height (mm)
  • Roll Inertia (kgm^2)
  • Pitch Inertia (kgm^2)
  • Yaw Inertia (kgm^2)
Possible additional uncertain inputs - Component Tolerance
  • Springs
  • Dampers
  • Bushes/Mounts
The mass properties are uncertain because predicting exact weight figures for any single variant to populate a simulation model is very difficult. Also, models are generated for assumed discrete derivatives and where these derivatives/variants are located in the distribution of ride performance is not understood. The customer uptake/frequency would also be of consideration when determining optimum ride behaviour. Figure 2.1.1 shows the variation of mass distribution on an example vehicle programme. IMAGE ONE Figure 2.1.2 shows the variation of front/rear mass distribution on an example vehicle programme. IMAGE TWO Figure 2.1.3 depicts the sales of vehicles with differing total mass and front/rear mass distribution on an example vehicle programme. IMAGE THREE Figure 2.1.4 & 2.1.5 indicate the linear relationship that exists between roll and pitch inertia and vehicle mass: IMAGE FOUR IMAGE FOUR_b The PDFs included are based on vehicle programme weight data. The quality of the data is uncertain and is known to require improvement.  

2.2 Propagation

Simpack Carwizard is used to simulate the models and generate the ride metrics. The series of ride simulations takes approximately 3 hours. Therefore, a DoE analysis of 36 runs would take approximately 108 hours. This would imply that because of the associated reduced accuracy of surrogate models they would not be considered. The preference would be to use analysis tools as current and for the UQ&M treatment to be a non-intrusive wrapper around the process. Inverse mapping is an important requirement.

2.3 Interpretation and Communication of Results

  • Optimised vehicle ride performance for all potential customer vehicle variants & option specifications.
  • Improved prediction of ride variation to minimise the number of component variables required to achieve desired ride performance across the programme.
  • Potential extension of the use case for optimised vehicle steering & ride attribute performance for all potential customer vehicle variants & option specifications. (Including the influence of component tolerance uncertainties.)

3. CURRENT STATE OF MATURITY

A six sigma green belt has been completed to understand that the various mass properties on a single vehicle programme influence the ride performance to a significant degree. The project involved the completion of a Design of Experiment analysis to investigate the affect of varying mass properties on the ride metrics of a specific vehicle programme. From these results a tuning/adjustment tool has been suggested to enable optimal ride tuning, see Figure 3.1.1 & 3.1.2. The black crosses on the ride metrics represent the performance of a tuning vehicle (prototype) The specific vehicle (variant) has been tuned to meet the target. However, overlaying the distribution of performance due to mass property variation from the study can guide the engineer to perform further tuning IMAGE FIVE IMAGE SIX This current process however, ignores popular customer sales choices and although may be useful to optimise ride performance across the mass variation range, has the possibility to still only be optimising for a smaller number of customers.

References:

http://support.prosig.com/2012/08/15/what-is-primary-and-secondary-ride/