Connected Automobiles

Constantly changing market conditions, increased competition, higher focus on customized demand, and globalization in the automotive sector requires constant innovation.

REQUEST A DEMO

Connected Designs

DesignEdge uses machine learning and AI with real time data for virtual simulation, modeling, and customized design solutions, helping reduce cost and design to market cycles.

Predicting Component Failures

Integrating real trip data with driving behaviour for predicting remaining useful life (RUL) and optimized designs for local conditions.

Connected Real 3D Simulator

Virtual simulation and modeling for what-if analysis and training.

DesignEdge Case Study – Prediction of Brake Pad Wear by using Machine Learning

BUSINESS PROBLEM

  • Brake Pad Wear is a topic that concerns Car Users, OEMS and Tier1 suppliers, because it is related to occupant safety.
  • There was no tool available, which could accurately predict the Brake Wear and give you a prior warning when the brake pads need to be replaced.
  • The reason Brake Wear is so difficult to predict is that it depends on many factors like Car Model, No. of Passengers, Brake Disc Size, Road Conditions, Braking Severity, Pad Grade etc.

SOLUTION DEPLOYED

  • OBD devices were installed in 700 cars to record the Driving Behaviour, Traffic Conditions, Braking Severity and Ambient Conditions
  • Non-contact type temperature sensors were also installed to continuously record the brake disc temperature
  • Brake Dynamometer tests were also conducted on different grades of brake pad. Different braking regimes were used to collect data on disc temperature, brake pad wear and disc wear
  • The data collected is used to build an ANN model to predict the brake pad wear, the disc wear and the coefficient of friction with 86% accuracy for any car, any disc size, any pad grade, driven on any route with a particular driving behaviour

BUSINESS IMPACT

  • Currently, a new disc design is either tested on a dynamometer or on a test track driven for thousands of kilometres to determine the wear of the brake pads.
  • By using this application the design time can be significantly shortened
  • OEM and Tier1 suppliers can quickly find out whether an existing design of a brake pad and disc will work for a new model of a car or a new design of a brake pad and disc will be required

Air Flow over the Brake Disc

Temperatures of the Brake Disc

DesignEdge Case Study – Prediction of Fatigue Damage of a Truck Chassis by using IOT & Machine Learning

BUSINESS PROBLEM

  • The major focus of truck manufacturers today is to design a truck chassis with more payload capacity and possible less weight.
  • This makes a chassis more susceptible to fatigue damages and failures.
  • Such failures are usually very sudden, and therefore manufacturers want a system which can predict the fatigue damages and the remaining useful life of the chassis

SOLUTION DEPLOYED

  • FE Analysis of the chassis was done with different loading conditions on each of the wheels.
  • An ANN model was trained with the load values and the results of stress and deformation values obtained from the FE Analysis.
  • Four force transducers and four accelerometers were installed at the four loading points of the chassis. These sensors pick up the road load at the four different points of the chassis in real time.
  • These load values are uploaded on the iGloble server and the ANN model calculates the cumulative damages in the chassis and predicts how many miles the truck can travel before cracks develop in the chassis.

BUSINESS IMPACT

  • Cracks and failures in the chassis lead to huge downtime and impacts the safety of the driver and the vehicle.
  • By using this solution, fleet owners and manufacturers can predict which location of the chassis is likely to develop a crack and can take corrective action by replacement of parts or strengthening different locations of the chassis.
  • This increases the uptime of the fleet and improves safety of the driver and the vehicle

Road Loads measured by the sensors

Stress values obtained from FE Analysis

DesignEdge Case Study Thermal and Vibration Analysis of Head-lamps

  • Coupled Flow and Thermal Analysis of Head-lamps to determine if the temperatures are below HDT
  • Dynamic Response Analysis was done to determine the resulting stress and deflections due to Vibrations and Shock
  • The results of the Thermal Analysis were found to be within 1o C of the test results.

DesignEdge Case Study Durability Analysis of Wheel

    • To perform Cornering (CFT) and Radial load (RFT) test on the wheel and find it’s life.
    • The process of CFT and RFT was simulated using digital simulation and validated against the existing result.
    • Nonlinear static analysis was carried out. Contact between disk and drum was modeled and full preload was applied on the bolts connecting the two components.
    • Stress history obtained from the analysis was used to find the life of the component using fatigue analysis.
    • The results were found very close to the test results.

See how Bytedge can help you from your data

Know More