SAMTECH Group SAMTECH Group Machine Tools Applications Machine - - PowerPoint PPT Presentation

▶
samtech group samtech group machine tools applications
SMART_READER_LITE
LIVE PREVIEW

SAMTECH Group SAMTECH Group Machine Tools Applications Machine - - PowerPoint PPT Presentation

Mar 13, 2023 408 likes •1.04k views

SAMTECH Group SAMTECH Group Machine Tools Applications Machine Tools Applications August 10, 2004 August 10, 2004 29/06/04, Page 1 SL/04/SAM/MKG_ppt/28an_a SAMTECH Mechatronic Mechatronic Solution Solution SAMTECH using Finite Element

slide-1
SLIDE 1

29/06/04, Page 1

SAMTECH Group SAMTECH Group Machine Tools Applications Machine Tools Applications

August 10, 2004 August 10, 2004

SL/04/SAM/MKG_ppt/28an_a

slide-2
SLIDE 2

29/06/04, Page 2

SAMTECH SAMTECH Mechatronic Mechatronic Solution Solution using Finite Element Method using Finite Element Method for the Machine Tools Industry for the Machine Tools Industry

slide-3
SLIDE 3

29/06/04, Page 3

Scope of the presentation

What is the problem of High Speed Machine Tools? Why is it difficult to analyze numerically Machine Tools? Methodology suited to compute dynamic behavior of vibrating mechanical systems SAMCEF for Machine Tools Presentation of examples Conclusions and future investigations&developments Questions/Answers

slide-4
SLIDE 4

29/06/04, Page 4

In order to increase the productivity, it is more and more necessary to increase the velocity of Machine Tools moving parts If maximum velocities are higher and higher, accelerations are increasing significantly Accelerations prescribed by controllers are generating huge inertia forces due to the mass of moving parts If the mass of moving parts is reduced due to the removal of material, there can be a loss of stiffness and thus of precision Inertia effects are thus major excitations for vibrations of High Speed Machine Tools in working conditions Vibrations cannot be cut in pieces

What is the problem of High Speed Machine-Tools ?

slide-5
SLIDE 5

29/06/04, Page 5

Need of a coherent global model describing the whole dynamic flexible behavior of the Machine Tool (mechanisms + structures + controllers) Sensitive components are the ones transmitting the vibrations (guideways, linear motors, ball screws…) These components have to be modeled using advanced numerical computation techniques able to manage the interactions between moving flexible bodies NL Finite Element coupled with kinematical constraints and large relative motions contact techniques are appropriate to achieve this goal SAMTECH has been working on this innovative approach since 15 years and brought recently this new powerful and integrated technology to the market, starting with the sector of Machine-Tools

What is the problem of High Speed Machine-Tools ?

In conclusion:

slide-6
SLIDE 6

29/06/04, Page 6

Examples of possible sources of “Non-Linearities” involved in FEM models of Machine Tools:

  • Large relative displacements & rotations
  • Linear motors
  • Ball screw
  • Guideways
  • Gearing systems
  • Belts
  • Bushings
  • Friction effects
  • Interaction with controllers
  • Visco-elastic material behavior…

Need of an implicit unconditionally stable time integration scheme:

  • Control of time integration error during discontinuities
  • Reliable description of shocks due to sudden change of motion

sense

Why is it difficult to analyse numerically Machine Tools ?

slide-7
SLIDE 7

29/06/04, Page 7

Augmented Lagrangian form of the equations of motion

⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = − − = = + + ) , ( ) , , ( ) (

inertia internal externa

t q k g g g t q q g p k q

l T

φ φ λ r r r r & r r r & & r B M

[ ]

3 / 1 , ) ( ) ( ) 1 ( ) , , ( ) , , ( ) 1 ( ) 1 (

1 1 1 1 1 1 1

∈ ⎭ ⎬ ⎫ ⎩ ⎨ ⎧ = + − + − = + − +

+ + + + + + +

α αφ φ α α α α α

n n n n n n n n n T n n T n n

q k q k t q q t q q q & & & & g g λ B λ B M HHT-Form of discretised equilibrium equation

Equations are solved iteratively each time integration step: “HHT-Form” Mixed system of equations in FEM-DOF and Lagrange Multipliers:

M: Mass matrix B: Gradient of the constraint matrix φ: Kinematical constraints

Classical non-linear Finite Element equations ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = − − = = ) K( ) C( with ) , , (

internal inertia internal externa

q q g g g g t q q g q

l

& r r r r & r & & r M

λ: Lagrange multiplier p: Penalty factor k: Scaling factor

Coupling mechanism and structural analysis Mathematical background

Methodology suited to compute dynamic behavior of vibrating mechanical systems

slide-8
SLIDE 8

29/06/04, Page 8

Integrated Multi-Disciplinary Solution

General integrated environment giving access to different levels of model fidelity and different analysis types from the same CAD based model

SAMCEF for Machine Tools

slide-9
SLIDE 9

29/06/04, Page 9

Suits to all kinds of Machine Tools

Linear Motor Gantry Rotational axis Parallel Translation axis C-Frame Box-In-Box High speed dynamics Flexible Screw

SAMCEF for Machine Tools scope

slide-10
SLIDE 10

29/06/04, Page 10

Global solution for the modeling of Machine Tools Very general computation scheme

SAMCEF for Machine Tools

slide-11
SLIDE 11

29/06/04, Page 11

Finite Element Analysis using SAMCEF

  • SAMCEF Mecano: Mechanisms&Non-Linear Dynamic

Structure/Contact FE Analysis

  • SAMCEF Asef: Linear static FE analysis
  • SAMCEF Dynam: Modal FE analysis + Super-element

creation/restitution

  • Interface between mechanical analysis (mechanisms&structures)

and controller design:

  • Import of MATLAB Simulink controllers inside SAMCEF Mecano
  • Export of state space (A,B,C,D matrices) from SAMCEF Mecano to

MATLAB Simulink

  • SAMCEF Field: User Friendly CAD Based Modeling and Post-

Processing Environment

Optimization of parameterized mechatronic systems using BOSS quattro

SAMCEF for Machine Tools

Components of SAMCEF for Machine Tools

slide-12
SLIDE 12

29/06/04, Page 12

NL Finite Elements (linear motor…) and S.E. Flexible kinematical joints (Guideways, flexible ball screw…) Digital Control (Sensors and Actuators)

  • SIMULINK

SIMULINK

  • …

…

Rigid kinematical joints

Integrated Non-Linear Mechanical Analysis

SAMCEF Mecano

slide-13
SLIDE 13

29/06/04, Page 13

Description of all the parts and the connection devices of the mechanical system Different Model Fidelity levels (rigid body, detailed meshed structure or Super-Element) Different types of connection devices (rigid or flexible kinematical joints, contact/friction conditions) “Sensors” and “actuators” to be connected to the digital control boxes Output of linearized mechanical models for control design

Integrated Non-Linear Mechanical Analysis

SAMCEF Mecano

slide-14
SLIDE 14

29/06/04, Page 14

Linear Static Analysis (SAMCEF Asef) Modal Analysis (SAMCEF Dynam) Super-Element creation and restitution (SAMCEF Dynam) Integrated environment with MBS and NL FEA

Linear Analysis Modal Analysis (SAMCEF Dynam)

SAMCEF Asef & SAMCEF Dynam

slide-15
SLIDE 15

29/06/04, Page 15

SAMCEF Mecano Control Box

Next Sample Time Command Vector Updated State Variables Current Time Sensor Vector State Variable

Sensor Variables:

Positions, Displacements, Velocities, Accelerations and/or Reaction Forces

Command Vector:

Forces, Displacements, Velocities, Accelerations and/or functions

Inputs and outputs of Controllers Communication between SAMCEF and embedded controllers

SAMCEF-Digital Control Interface

slide-16
SLIDE 16

29/06/04, Page 16

Controllers can be imported from a Controller Design Software (for example MATLAB Simulink using Real Time Workshop (RTW) to generate C subroutine) They can also be fully written by the user (open to in-house controllers) Linked with SAMCEF Mecano

SAMCEF-Digital Control Interface

slide-17
SLIDE 17

29/06/04, Page 17

Positions, displacements, velocities, accelerations as well as reaction forces are provided independently by SAMCEF Mecano to the controller Activation and deactivation times Several controllers can be used at the same time The same controller can be used several times Different schematic boxes (sensors, controllers and actuators) can be connected between them Prescribed values and gains can be defined through the SAMCEF data set

SAMCEF-Digital Control Interface

slide-18
SLIDE 18

29/06/04, Page 18

The computation time step is bound to the time discretization of the Controller tComp <= t Controller Automatic time step choice between instants of “Rendez-vous”

Time step control

SAMCEF-Digital Control Interface

slide-19
SLIDE 19

29/06/04, Page 19

SAMCEF Field

Integrated Modeling Environment

slide-20
SLIDE 20

29/06/04, Page 20

Modeling defined directly on the CAD geometry Easy modifications of the modeling fidelity for bodies

  • Rigid Body
  • Meshed body with any type of material, including composites
  • Super-Elements generated by SAMCEF or imported from another

FE package

Easy modifications of the modeling fidelity for connections

  • Ideal kinematical joints
  • Rigid-flexible contact conditions
  • Flexible-flexible contact conditions

Easy switch from an analysis type to another one (SAMCEF Mecano, SAMCEF Dynam, SAMCEF Asef…) Import of controllers from MATLAB Simulink

SAMCEF Field

Integrated Modeling Environment

slide-21
SLIDE 21

29/06/04, Page 21

Five modules to manage the global to the detailed analyses SAMCEF Field

slide-22
SLIDE 22

29/06/04, Page 22

CAD creation

Geometry creation Geometry import

slide-23
SLIDE 23

29/06/04, Page 23

Structural elements of a Machine Tool

Parameterized components

slide-24
SLIDE 24

29/06/04, Page 24

Super-Element model reduction Super-Element creation Placement and use

Component creation and use

slide-25
SLIDE 25

29/06/04, Page 25

Linear and non-linear data

Data definition

slide-26
SLIDE 26

29/06/04, Page 26

Association of Data, Mechanism, Motion Unit using geometry Guideways: Flexible slider, Bushing…

Data creation

slide-27
SLIDE 27

29/06/04, Page 27

Machine Tool Mechanical Elements

Linear motors Ball screws Guideways Rack and pinion Ball screw Guideways & rack and pinion Linear motor Definition on geometry Multiple models for each physical component Only additional parameters between different models

  • f the same transmission

Transmission models Common features

slide-28
SLIDE 28

29/06/04, Page 28

Flexible model of a guideway assembly (node moving on a prismatic flexible slider (beam mesh)

Meshing: Guideways

slide-29
SLIDE 29

29/06/04, Page 29

Driven hinge joint SCRF stiffening elements MEAN stiffening MEAN nut node

COUPLING JOINT

ROTOR STATOR BEAM STICK BEARING elements stiffening elements MEAN STICK STICK

Meshing: Ball screw

Flexible model of a ball screw (node moving on a beam mesh and rotating due to the screw pitch)

slide-30
SLIDE 30

29/06/04, Page 30

Selection of rigid or flexible behavior

Meshing : Rigid – Flexible

slide-31
SLIDE 31

29/06/04, Page 31

Rigid and Flexible Model of C-Frame

slide-32
SLIDE 32

29/06/04, Page 32

Model linearization Integration of a SAMCEF model into a functional simulation tool used to design controllers Controller design

Efficient choice of the right controller Fine tuning of the initial values for the gains

Adapted controller

Controller design

slide-33
SLIDE 33

29/06/04, Page 33

Linearized system is described by M(mass), C(damping) and K(stiffness) matrices Automatic transformation into A-B-C-D Space State matrices

Effect of the linearization point

slide-34
SLIDE 34

29/06/04, Page 34

Internal or external controller

Integration of various Digital Controllers into SAMCEF Non-Linear Simulations in order to get a mechatronic FE model combining mechanisms, structures and controllers in one single model

Controller interface with model

slide-35
SLIDE 35

29/06/04, Page 35

Static stresses evaluation Vibration Modes calculation Super-Element reduction Time response (static, kinematical or dynamic)

Various analyses using same model description

Analyses

slide-36
SLIDE 36

29/06/04, Page 36

Non-linear structure analysis

Non Linear Results

slide-37
SLIDE 37

29/06/04, Page 37

0.1 0.2 0.3 0.4 0.5 0.01 0.02 0.03 0.04 0.05 0.06 time(s) position(m) step response: siemens controller step input position horizontal slider

Siemens Controller Position Controller Speed Controller Current Controller Controllers can be Imported from a Controller Design Software (for example MATLAB Simulink) Fully written by the user

Industrial Controller

slide-38
SLIDE 38

29/06/04, Page 38

System with varying structural flexibility

Bode Diagram Frequency (Hz) Phase (deg) Magnitude (dB)

  • 50
  • 40
  • 30
  • 20
  • 10

10 10

1

10

2
  • 720
  • 540
  • 360
  • 180

180 L = 34 cm L = 37.75 cm L = 41.5 cm L = 45.25 cm L = 49 cm

Bode Diagram Frequency (Hz) Phase (deg) Magnitude (dB)

20 40 60 80 10

1

10

2
  • 270
  • 225
  • 180
  • 135
  • 90
  • 45

L = 34 cm L = 37.75 cm L = 41.5 cm L = 45.25 cm L = 49 cm

Identification of the system: FRFs from the motor force input to

  • the motor position
  • the endpoint acceleration

Clear dependence of the eigenfrequencies on the arm length

Flexible beam with varying length

slide-39
SLIDE 39

29/06/04, Page 39

Motion Control and Vibration Control

Movement of the linear motor and the end-tip of the beam

Dynamic effect of the flexibility

slide-40
SLIDE 40

29/06/04, Page 40

0.5 1 1.5 2 2.5 3 3.5 4
  • 6000
  • 4000
  • 2000
2000 4000 6000 Tijd (seconden) Volgfout ( ) 탆 met regelaar zonder regelaar

Pole placement controller

0.5 1 1.5 2 2.5 3 3.5 4
  • 6000
  • 4000
  • 2000
2000 4000 6000 Tijd (seconden) Volgfout ( ) 탆 met regelaar zonder regelaar

LQR controller

0.5 1 1.5 2 2.5 3 3.5 4
  • 8000
  • 6000
  • 4000
  • 2000
2000 4000 6000 8000 Tijd (seconden) Volgfout ( ) 탆

MIMO controller

0.5 1 1.5 2 2.5 3 3.5 4
  • 6000
  • 4000
  • 2000
2000 4000 6000 Tijd (seconden) Volgfout ( ) 탆 SISO-motorregelaar MISO-motorregelaar

MISO controller

0.5 1 1.5 2 2.5 3 3.5 4
  • 5000
  • 4000
  • 3000
  • 2000
  • 1000
1000 2000 3000 4000 5000 Tijd (seconden) Volgfout ( ) 탆

HAC/LAC controller

0.5 1 1.5 2 2.5 3 3.5 4
  • 6000
  • 4000
  • 2000
2000 4000 6000 Tijd (s) Totale fout (μ m)

met regelaar zonder regelaar

PD controller

Vibration Controller Types

slide-41
SLIDE 41

29/06/04, Page 41

Synchronised time delay

0.2 0.4 0.6 0.8

  • 0.05

0.05 0.1 0.15 0.2 time(s) position(m) sinus response: synchrone horizontal position input horizontal position vertical position input vertical position 0.2 0.4 0.6 0.8

  • 0.05

0.05 0.1 0.15 0.2 time(s) position(m) sinus response: non synchrone horizontal position input horizontal position vertical position input vertical position

  • 0.05

0.05 0.06 0.08 0.1 0.12 0.14 0.16 X(m ) Y(m) Circle test : syn ch ron e traject in pu t traject ou tpu t

  • 0.05

0.05 0.06 0.08 0.1 0.12 0.14 0.16 X(m) Y(m) Circle test : non synchrone traject input traject output

Tuning individual controllers Concurrent Tuning of all controllers

Performance measures: Circle test

slide-42
SLIDE 42

29/06/04, Page 42

Rigid-body simulation, flexible structure analysis with or without Super-Elements

slide-43
SLIDE 43

29/06/04, Page 43

BOSS quattro

Task Management and Optimization

slide-44
SLIDE 44

29/06/04, Page 44

RamZ Fixed Part SliderX

9517

SliderY

9443 8343 9484

Model to be optimized

slide-45
SLIDE 45

29/06/04, Page 45

Definition of an optimization problem

slide-46
SLIDE 46

29/06/04, Page 46

Optimization iteration

slide-47
SLIDE 47

29/06/04, Page 47

Optimization loop

slide-48
SLIDE 48

29/06/04, Page 48

Automatic loop on controller parameters

Results of optimization problem

slide-49
SLIDE 49

29/06/04, Page 49

«Mechatronic Compiler for Machine Tool Design» (EU Project, 01/02/2001 to 31/01/2004)

DigitalRegulator.mdl parameter definition in PDigitalRegulator.m 4 Response2 3 Response1 2 Motor mass [kg] 1 Motor1 Force [N] Vin MVel a_curr force mass P.Ind=0.0018 P.Res=0.2 F.Const=57 V.Const=19 weightless motor position loop sampling z-1 den(z) numerical derivation
  • f velocity
z-1 den(z)
  • num. deriv.
  • f acceler.
v_cmd v_act forceFF f_cmd Kp=300000 tn=10 discrete speed controller without reference model Iref Iact Vcmd kp=5.65288 tn=2 limit=244.949 discrete current controller pos_act vel_est
  • dig. velocity derivator
[ContErr] contouring error [ActVolt] WVolt [SpeedCom] WSpeedCom [RefCurr] WRefCurr [ActPos] WPos [EstSpeed] WEstSpeed [ActCurr] WActCurr 1 VelLoopFlag SumV1 SumV SumP U U(E) Selector1 U U(E) Selector Exc: 4 RefSpeed Exc: 5 RefPos 1 PosLoop Flag Motor Locking
  • dig. lowpass ord.2
State= on CuttFreq=2000 RelDamp=0.7 SampleT=125 LowPass for simmetrization1
  • dig. lowpass ord.2
State= on CuttFreq=500 RelDamp=0.8 SampleT=1000 LowPass for simmetrization
  • K-
KProp Position Kv {Exc} Goto Tag Visibility {Exc} Goto [ContErr] From7 [ActPos] From6 [EstSpeed] From5 SpeedCom] From4 [RefCurr] From3 [ActCurr] From2 [ActVolt] From1
  • K-
Force2Current 1 FeedF Gain Exc: 3 CurrBefore Exc: 2 CurrAfter
  • rd.2 dig. bandstop
State 'Off' CentrFreq=0 DBand=0 NBand=0 BS_Freq=100 SampleT=31.25 BandStop filter F4
  • rd.2 dig. bandstop
State 'Off' CentrFreq=0 DBand=0 NBand=0 BS_Freq=100 SampleT=31.25 BandStop filter F3
  • rd.2 dig. bandstop
State 'Off' CentrFreq=300 DBand=40 NBand=20 BS_Freq=100 SampleT=31.25 BandStop filter F2
  • K-
AxisInertia Exc: 1 ActForce 4 Excitation input 3 Meas Position [m] 2 Motor Velocity [m/s] 1 Pos Com [m]

Reference: MECOMAT EU Partnership

slide-50
SLIDE 50

29/06/04, Page 50

Mechanical Modeling of Machine Tools

CAD Geometry defined in SAMCEF Field or recovered from CAD through STEP, IGES, CATIA exchange

X slide Y slide Portal Ram Electro spindle

Guides ways

SAMCEF Mecano assembly conditions (rigid bodies, rigid kinematical joints and flexible sliders, gluing) to connect moving flexible bodies

Super-Elements to manage large meshes

Agile transfer lines, COMAU

slide-51
SLIDE 51

29/06/04, Page 51

0.5 1 1.5 2 2.5 3 3.5 4

  • 6000
  • 4000
  • 2000

2000 4000 6000 Tijd (s) Totale fout ( μ m)

met regelaar zonder regelaar

Performance Hac-Lac controller for ramp position trajectory

0.5 1 1.5 2 2.5 3 3.5 4

  • 5000
  • 4000
  • 3000
  • 2000
  • 1000

1000 2000 3000 4000 5000 Tijd (seconden) Volgfout ( ) 탆

Vibration controller (LAC controller) Linear Motor controller (HAC controller)

Time Tracking error Tracking error

Without and with controller

Controller Design of Machine Tools

Time

slide-52
SLIDE 52

29/06/04, Page 52

Same mechanical modeling principles

as previous examples

Linear motors implemented in the

generic non linear force element of SAMCEF Mecano

Digital controllers acting on axes

modeled in MATLAB Simulink and integrated in SAMCEF Mecano

Comparison

between numerical results and experiments: less than 8%

  • f error on the speed and the

precision of execution

Mechatronic Application

slide-53
SLIDE 53

29/06/04, Page 53

Parallel machine using 3 linear motors and 1 rotational axis implemented in the non-linear pre- post processor SAMCEF Field Rigid and flexible behavior

Advanced Machine Tool Application

slide-54
SLIDE 54

29/06/04, Page 54

Rigid and Flexible behavior on parallel machine

Advanced Machine Tool Application

slide-55
SLIDE 55

29/06/04, Page 55

Parallel machine using 3 linear motors

control structure mechanism

Advanced Machine Tool Application

Three axes machine Based on a “delta” parallel structure Strokes : X 500 Y 500 Z 300 Linear speed: 100 m/min on all axes Acceleration: 35 m/s² on all axes Jerk: 5000 m/s3 on all axes Repeatability: 5 µm on all axes Spindle: 15 kW 40 000 rpm

slide-56
SLIDE 56

29/06/04, Page 56

Mechanism Mechanism + Structure Mechanism + Control Mechanism + Structure + Control

Mechanism + Structure

Interdependence: Structure - Mechanism - Control

Advanced Machine Tool Application

slide-57
SLIDE 57

29/06/04, Page 57

Robotic examples

slide-58
SLIDE 58

29/06/04, Page 58

General SAMTECH Product Portfolio

slide-59
SLIDE 59

29/06/04, Page 59

Product/Service Approach of SAMTECH Group

slide-60
SLIDE 60

29/06/04, Page 60

Future Developments & Perspectives

Introduction of harmonic response (SAMCEF Repdyn) in SAMCEF Field Introduction of thermal transient analysis (SAMCEF Thermal) in SAMCEF Field Extension of the Mechatronic approach by Finite Element Analysis to other industrial sectors (robotics, textile machines, automotive, aeronautics, space, defense…)

slide-61
SLIDE 61

29/06/04, Page 61

Other examples of mechanical systems modeled using SAMCEF

slide-62
SLIDE 62

29/06/04, Page 62

Conclusions

Machine Tools manufacturers currently develop high speed machines were inertial effects are becoming critical A Rigid Multi-Body model is clearly not sufficient Super-Elements are not always sufficient (inefficient management of contact conditions between moving flexible bodies) SAMCEF for Machine Tools proposes an implicit NL Dynamic solution using NL Finite Element/Contact techniques SAMCEF for Machine Tools is also a global and integrated CAD based solution covering all the Machine Tools components (linear motor, guideways, ball screw, gears, rack-and-pinions…), the model fidelity levels and analysis types (linear or not) requested by this industrial sector