Optimization of Bulk Handling Control System using Sensor Force Feedback with Altair EDEM, Inspire, and MotionSolve Co-simulation


CHALLENGE

Often bulk handling systems used in Heavy Equipment and the Healthcare and Life Sciences Industries are developed with advanced control systems to automate the movement of the bulk material.  Force based sensors like force-sensing resistors, strain gauges, or load cells are typically used in bulk handling systems to detect the presence of bulk material in the system due to their relatively low cost and high reliability.  The development and tuning of the software used to control the bulk handling system is typically completed after the bulk handling system is built which makes modifications to the physical design impractical due to the cost associated with late-stage changes.  This means that the control system must operate within the bounds of an already existing system that was not optimized for control and automation resulting in sub-optimal performance.

 

WHY OPTIMIZE PERFORMANCE VIRTUALLY

Simulations of the bulk handling system can be completed in which the signal from force-based sensors are predicted allowing for early design modifications to be implemented prior to the fabrication of the bulk-handling system enabling optimal performance of the system.  Simulations of bulk handling systems are often completed with a Discrete Element Method (DEM) software like Altair’s EDEM, in which the interaction of the bulk material and the system can be predicted.  Typically, the motion of the components in the bulk handling system can be defined in the DEM software, but when the kinematics are a function of the force exerted by the bulk material on a sensor a co-simulation with a multi-body dynamics (MBD / MBS) tool like ADAMS, RecurDyn, or Altair’s MotionSolve are required.  A 3D computer-aided design (CAD) model of the bulk handling system is required as an input to the DEM simulation, Altair Inspire can be used to generate such a model. Users that are unfamiliar with the creation of MBD models benefit from the intuitive workflow in Altair Inspire (Inspire Motion) to setup and run MBD models, which can be easily exported to MotionSolve for co-simulation with EDEM.

 

FILES

The simulation files can be downloaded here: (all files used 2023.0 version of software):

 

SOFTWARE REQUIRMENTS

It is recommended to use the 2023.0 version of all the software to ensure compatibility with the shared files.

 

SIMULATION WORKFLOW

For step-by-step directions to setup the models please see the following video: How to Implement Force Based Motion Control in EDEM and MotionSolve

 

Step 1 – CAD Model Altair Inspire

Altair Inspire was used to create the 3D model of the hopper and inclined auger.  Open the Altair Inspire Model Generic Auger Hopper Model.stmod, shown in Figure 1.  The stationary components (all components except for the auger and sensors) were also exported in a step file Generic Auger Hopper Model_Stationary.step and imported into the EDEM file Generic Example Motion Control Based no Geometry.dem.    In the next step the motion of the auger will be set up.

A transparent hopper and inclined auger housing with yellow force sensors and a gray inclined auger.

Figure 1 Hopper and inclined auger with two yellow force sensors shown in Inspire.

 

Step 2 – Define MBD model in Altair Inspire Motion

Inspire Motion was used to introduce a motor to rotate the inclined auger, as shown in Figure 2.  The completed setup of the auger motion is available in the Inspire Model Generic Auger Hopper Model Motion Definition.stmod.  After verifying the correct motion of the auger the model was exported to an mdl format which can be opened and modified in Altair MotionView, the pre/post processor for the MBD solver MotionSolve.  In the next step the motion of the motor will be updated so that it is dependent on the force exerted by the bulk material on the force sensors.

Figure 2 Motion of the inclined auger within the auger housing and hopper in Inspire.

 

Step 3 – Update the Auger Motion in MotionView

The exported model from Inspire Motion Generic Auger Hopper Model Motion Definition_MMKS.mdl was updated in MotionView, shown in Figure 3.  A solver variable Sensor_Force was created in which the product of the magnitude of the forces on both sensors was calculated. An Output Variable was added so that the Sensor_Force could be visualized after the co-simulation was complete, and the motor motion constraint was updated to be a function of the Sensor_Force variable.  Finally, a DEM sub system was created enabling the coupling of EDEM and MotionSolve.  When creating the DEM subsystem the auger and force sensors were added to the EDEM model.

MotionView model with highlighted DEM subsystem, Solver Variable & Output Force_Sensor, and the updated motor motion constraint

Figure 3 MotionView model with highlighted DEM subsystem, Solver Variable & Output Force_Sensor, and the updated motor motion constraint.

 

Step 4 – Complete the coupled co-simulation between EDEM and MotionSolve

In the final step, the EDEM simulation deck Generic Example Motion Control Based with Geometry.dem was coupled with the MBD model Generic_ Auger_ Hopper_ Model_ MMKS_ Updated.mdl and the results are shown in Figure 4.  The plot of motor speed and sensor force are included in Figure 4 to highlight the dependance of the auger motor speed on the sensor force.  In this example the auger was initially stationary because there was no force exerted by the bulk material on the force sensors.  As the simulation progressed bulk material was added to the hopper and as a result forces from the bulk material were exerted on the sensors.  This caused the auger speed to increase to a maximum of 200 rpm.  In the middle of the simulation (at approximately 7 seconds) the introduction of the bulk material into the hopper was interrupted, eventually allowing the height of the material in the hopper to drop to the point where the auger stopped rotating.  After approximately 14 seconds the introduction of the bulk material was resumed and the bulk material accumulated in the hopper. Eventually, the force exerted by the bulk material on the sensors reach a threshold that resulted in the acceleration of the auger motor to a maximum speed of 200 rpm.

Figure 4 Co-simulation of the auger and hopper bulk handling system in which the speed of the auger motor was dependent on the forces exerted by the bulk material on the yellow force sensors.

The objective of this example is to demonstrate how the motion of a component in a virtual bulk handling system can be controlled based on the bulk material exerting a force on a sensor. To take this example a step further and optimize the bulk handling system consider using Altair HyperStudy in combination with the EDEM-HyperStudy Connector.  If the simple implementation of the speed controller in MotionView based on the forces exerted by the bulk material on the force sensor is to limiting consider including a 1D simulation tool like MATLAB, Amesim, or Altair Twin Activate to implement and explore more advanced control logic.  Alternatively, if the power electronics system in under study including the motor, inverter and control loop consider adding Altair PSIM to the coupled simulation.  An example can be found here in which EDEM, MotionSolve, PSIM and Twin Activate (formerly Activate) were used together to model the drivetrain of a bulk solids mixer.

 

There are other examples of how EDEM and MotionSolve can be coupled:

 

Here are some frequently asked questions and answers specific to MotionView and Motion Solve

 

There is an ever increasing library of how-to videos available on YouTube, please consider subscribing to the channel so that you are aware of the new content as soon as it is available: 

For further information on EDEM Applications we have plenty more on the Altair Community: