Tyréns - Passenger Gangway Weight Reduction Project

Passenger Gangway Optimization

Customer

Tyréns

Challenge

The adjustable passenger g angway is used to boa rd passengers into ferries at Gothenburg harbor . This steel structure is subjected to wind, snow and live loads and was designe d and modeled to meet specifications for stiffness, natural frequency and maximum deflection. However during the manual design process only a limited number of selections for beam cross - sections were considered. The need to reduce mass, and hence cost, was the basis to incorporate optimization methodology for selection and dimensioning of beam cross - sections.

Solution

The gangway was mode led in three different configurations in STAAD - Pro engineering software. HEEDS MDO and its proprietary hybrid optimization algorithm SHERPA were used to select the best combination of beam - cross sections to meet the require ments . In total 25 beam s were included as design parameters , to be dimen sioned from a pool of 92 standardized cross - sections . All the beams were subjected to constraints on maximum stress . The structure was also subjected to maximum deflection and natural frequency constraints . In total 78 constraints were defined in HEE DS with a single objective: Minimize total mass. In just 27 hours, 2600 design iterations were completed. In each iteration 3 different configurations were analyzed simultaneously. Only 9.7% of the designs were deemed feasible, a testament to the complexity of the optimization problem. The optimal design resul ted in reduction of 5500 K g of steel from the structure with b aseline design weight of 83000 Kg.

Automatic Correlation Project

Autocorrelation of virtual suspension characteristics to K&C measurement

Customer

Premium Automotive OEM

Challenge

Kinematic and compliance (K&C) measurements are performed on physical car s with different load cases to estimate the suspension characteristics. The same tests are performed virtually to assess the vehicle dynamics attributes of the car. However t he virtual car suspension model always differ from the real car suspension , primarily due to manufacturing tolerances . T uning the virtual model involves adjusting hundreds of suspension harpoints and bushing properties. This task is tedious, time consuming and often extremely diff icult due to interdependencies in parameters and responses . The aim was to develop an indus try standard procedure for automatic tuning of suspension parameters in the virtual model to match the K&C test data .

Solution

In this project, the suspension was modeled in ADAMS Car , an d HEEDS MDO optimization software was employe d as a tool for auto - tuning of suspension parameters. For the front suspens ion model, 81 different design variables were selected. These included all hardpoint locations and bushing stiffnesses. The objective of correlation was to minimize the root mean square (RMS) value between K&C measurement curve s and ADAMS simulation curve s . From the K&C measurement data , 23 curves were defined as tar get curves in the autocorrelation process. In 2000 iterations, SHERPA, the proprietary hybrid optimization algorithm in HEEDS MDO, achieved a good correlation of the simulation model to the physical test data . Hence the simulation model is correlated and can now be used in the decision - making process and assessing vehicle dynamic c haracteristics , therefore eliminating further need for physical testing.

Virtual prototyping

Agricultural equipment virtual prototyping project

Customer: Agricultural Equipment Manufacturer

Challenge: Predict stresses in welded frame

Solution:

  • Virtual model of machine
  • Model of each pin with attachment and spring
  • Flexible frame 

Virtual prototyping

Seat rail and steering wheel vibration assessment virtual prototyping project

Customer: Premium Automotive OEM

Challenge: Predict vibrations in seat rail + steering wheel prior to availability of mule

Solution:

  • Engine model in A/Car with controllers and parametrized engine mounts
  • FE structural model
  • Optimization loop identifying key design parameters