Intro To Chemical Adsorption

Intro
Programming paradigms
Architecture
Use Cases
DAE Tools: An equation-oriented process

modelling and optimization software
Introduction
Dragan Nikolić
DAE Tools Project, http://www.daetools.com
1 December 2013
Dragan Nikolić DAE Tools - Introduction

Intro
Architecture
Use Cases
Outline
1 Intro
2 Programming paradigms
3 Architecture
4 Use Cases

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Outline
1 Intro
General Info
Motivation
Main features
General
DSL vs. DAE Tools
3 Architecture
Overview
4 Use Cases
Use Case 1 - High level modelling language
Use Case 2 - Low level DAE solver
Use Case 3 - Embedded simulator (back end)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
What is DAE Tools?
Process modelling, simulation, optimization and parameter

estimation software (www.daetools.com)
Areas of application:
Initially: chemical process industry (mass, heat and
momentum transfers, chemical reactions, separation processes,
phase-equilibrium, thermodynamics)
Nowadays: multi-domain
Hybrid approach between general-purpose programming
languages (c++, Fortran, Java) and domain-specific
modelling languages (Modelica, gPROMS...)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
What can be done with DAE Tools?
Simulation
Steady-State
Transient
Optimization
NLP problems: IPOPT, NLOPT, OpenOpt, scipy.optimize
MINLP problems: BONMIN
Parameter estimation: Levenberg–Marquardt algorithm
(scipy.optimize)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Types of systems that can be modelled
Initial value problems of implicit form: systems of linear,

non-linear, and (partial-)differential algebraic equations
Index-1 DAE systems
With lumped or distributed parameters: Finite Difference or
Finite Elements Methods
Steady-state or dynamic
Continuous with some elements of event-driven systems
(discontinuous equations, state transition networks and
discrete events)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Why yet another software?
Advantages:
1 Hybrid approach betwen DSL and GPPL
2 Programmatical generation of models
3 Runtime modification of objcts/models (operating procedures)
4 Introperability with 3rd party software packages/libraries
5 Code generation/Model exchange capabilities

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Not a modelling language
A set of software packages

API for:
Model development
Results processing (plotting, various file formats)
Simulation, optimization and parameter estimation
Code generation for other DSLs and programming languages
Report generation (XML+MathML) and model exchange
Large set of supported solvers (DAE, LA, NLP, MINLP)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Not a modelling language (cont’d)
Allows easy interaction with other software libraries (two-way

interoperability with other software, embedding in other
software etc.)
Free/Open source software (GNU GPL)
Cross-platform (GNU/Linux, MacOS, Windows)
Supports multiple architectures (32/64 bit x86, arm, any
other with the GNU toolchain)
Developed in c++ with Python bindings (Boost.Python)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Object-oriented modelling
Everything is an object (models, parameters, variables,

equations, state transition networks, simulations, solvers, ...)
Models are classes derived from the base daeModel class
(inheriting the common functionality)
Hierarchical model decomposition allows creation of complex,
re-usable model definitions
All Object Oriented concepts supported (such as multiple
inheritance, templates, polymorphism, ...) that are supported
by the target language (c++, Python), except:
Derived classes always inherit all declared objects (parameters,
variables, equations, ...)
All parameters, variables, equations etc. remain public

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Equation-oriented (acausal) modelling
Equations given in an implicit form (as a residual)

F (ẋ, x, y, p) = 0
Input-Output causality is not fixed:
Increased model re-use
Support for different simulation scenarios (based on a single
model) by specifying different degrees of freedom
For instance, equation given in the following form:
x1 + x2 + x3 = 0
can be used to determine either x1 , x2 or x3 depending on
what combination of variables is known:
x1 = −x2 − x3 or x2 = −x1 − x3 or x3 = −x1 − x2

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Separation of models definition from operations on them
The structure of the model (parameters, variables, equations

etc.) given in the model classes (daeM odel,
daeF initeElementM odel)
The runtime information in the simulation class
(daeSimulation)
Single model definition, but:
One or more different simulation scenarios
One or more optimization scenarios

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Hybrid continuous/discrete systems
Modelling of continuous systems with some elements of

event-driven systems
Discontinuous equations
State transition networks
Discrete events

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Code generation
Model export from DAE Tools to other

DSL/modelling/programming languages
Modelica
c99

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Model Exchange
Support for Functional Mock-up Interface for Model Exchange

and Co-Simulation (FMI): https://www.fmi-standard.org
FMI – a tool independent standard to support both model
exchange and co-simulation of dynamic models using a
combination of xml-files and compiled C-code
Still in experimental phase

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Model reports
Automatic model documentation

XML + MathML format
XSL transformation used to to generate HTML code and
visualize reports
Two types:
Model description report (contains model definition)
Runtime report with all values and equations expanded
(contains definition of the simulation)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Model reports (cont’d)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Model reports (cont’d)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Multi-domain
From chemical processing industry to biological neural

networks
DAE Tools is not a DSL but defines the basic modelling
concepts such as models, parameters, variables, various types
of equations (ordinary, differential, partial differential,
discontinuous), state transition networks etc. that can be
used as building blocks for a specific domain
Example: a reference implementation simulator for NineML
(xml-based modelling language for describing networks of
spiking neurons)
The key concepts from NineML are based on DAE Tools
concepts: Neurone, Synapse, Population of neurones, Layers,
Projections etc
Intro
General Info
Motivation
Architecture
Main features
Use Cases
DAE Plotter
2D plots (Matplotlib)
Animated 2D plots
3D plots (Mayavi)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Solvers
Supported DAE solvers@

Sundials IDAS
(https://computation.llnl.gov/casc/sundials/main.html)
Supported FE libraries:
deal.II (http://dealii.org)
Supported optimization solvers:

IPOPT (https://projects.coin-or.org/Ipopt)
Bonmin (https://projects.coin-or.org/Bonmin)
NLOPT (http://ab-initio.mit.edu/wiki/index.php/NLopt)

Intro
General Info
Motivation
Architecture
Main features
Use Cases
Solvers
Supported LA solvers:
Sundials dense LU, Lapack
Trilinos Amesos (http://trilinos.sandia.gov/packages/amesos)
Trilinos AztecOO (http://trilinos.sandia.gov/packages/aztecoo)
SuperLU SuperLU-MT
(http://crd.lbl.gov/ xiaoye/SuperLU/index.html)
Umfpack (http://www.cise.ufl.edu/research/sparse/umfpack)
MUMPS (http://graal.ens-lyon.fr/MUMPS)
CUSP (http://code.google.com/p/cusp-library)
Intel Pardiso (http://software.intel.com/en-us/articles/intel-mkl)

Intro
Programming paradigms General
Architecture DSL vs. DAE Tools
Use Cases
Outline
1 Intro
General Info
Motivation
Main features
General
DSL vs. DAE Tools
3 Architecture
Overview
4 Use Cases

Intro
Use Cases
Approaches to process modelling
Two approaches to process modelling:

Domain Specific Language (DSL)
General-purpose programming language (such as c, c++, Java
or Python)

Intro
Use Cases
Domain Specific Languages
Special-purpose programming or specification languages

dedicated to a particular problem domain
Designed to directly support the key concepts from that
domain
Specifically created to solve problems in a particular domain
(Usually) not intended to solve problems outside that domain
(although that may be technically possible in some cases)
Commonly lack low-level functions for filesystem access,
interprocess control, and other functions that characterize
full-featured programming languages, scripting or otherwise
Examples: Modelica, gPROMS, SpeedUp, Ascend, GAMS ...

Intro
Use Cases
General-purpose programming languages
Created to solve problems in a wide variety of application

domains
Do not support key concepts from any domain
Have low-level functions for filesystem access, interprocess
control etc.
Examples: c, c++, Fortran, Python, Java etc.
Typical scenario: solving a DAE system
Choose a solver (Sundials IDA, DASSL, RADAU5, DAEPACK
etc)
Implement user functions to manually calculate residuals and
derivatives for a Jacobian matrix, apply boundary conditions
etc
Create an executable program

Intro
Use Cases
DAE Tools approach
A sort of the hybrid approach:

Applies general-purpose programming languages such as c++
and Python
Offers a class-hierarchy/API that resembles a syntax of a DSL
as much as possible.
Provides low-level concepts such as parameters, variables,
equations, ports, models, state transition networks, discrete
events etc.
Concepts from new application domains can be added on top
of its low level concepts (for instance the simulator for
biological neural networks - NineML, as it will be shown later
in Use Case section)
Enables an access to the low-level functions and a large
number of standard libraries
Intro
Use Cases
gPROMS vs. Modelica
Model developed in gPROMS The same model in OpenModelica

http://www.psenterprise.com https://www.openmodelica.org
Intro
Use Cases
DAE Tools
The same
Dragan modelDAE
Nikolić in Tools
daetools
- Introduction
Intro
Use Cases
Key modelling concepts & grammar
DSL/Modelling languages
Domain-specific languages allow solutions to be expressed in
the idiom and at the level of abstraction of the problem
domain (direct support for all modelling concepts by the
language syntax)
DAE Tools
Modelling concepts cannot be expressed directly in the
programming language and have to be emulated in the API or
in some other way

Intro
Use Cases
Verbosity
Clean, concise, ellegant and natural way of building model
descriptions: the code can be self documenting
DAE Tools
The support for modelling concepts is much more verbose and
less elegant; however, DAE Tools can generate
XML+MathML based model reports that can be either
rendered in XHTML format using XSLT transformations
(representing the code documentation) or used as an
XML-based model exchange language

Intro
Use Cases
Maintainability & portability
Domain-specific languages could enhance quality, productivity,
reliability, maintainability and portability
DAE Tools

Intro
Use Cases
Simulators & programming languages
DSLs could be and often are simulator independent making a
model exchange easier
DAE Tools
Programming language dependent; however, a large number
of scientific software libraries exposes its functionality to
Python via Python wrappers

Intro
Use Cases
Need for a compiler/parser/interpreter
Cost of designing, implementing, and maintaining a
domain-specific language as well as the tools required to
develop with it (IDE): a compiler/lexical parser/interpreter
must be developed with all burden that comes with it (such as
error handling, grammar ambiguities, hidden bugs etc)
DAE Tools
A compiler/lexical parser/interpreter is an integral part of the
programming language (c++, Python) with a robust error
handling, universal grammar and massively tested

Intro
Use Cases
Need for a new language syntax
Cost of learning a new language vs. its limited applicability:
users are required to master a new language (yet another
language grammar)
DAE Tools
No learning of a new language required (everything can get
done in a favourite programming language)

Intro
Use Cases
Interoperability with the 3rd party software
Increased difficulty of integrating the DSL with other
components: calling external functions/libraries and
interaction with other software is limited by the existence of
wrappers around a simulator engine (for instance some
scripting languages like Python or javascript)
DAE Tools
Calling external functions/libraries is a natural and
straightforward Interaction with other software is natural and
straightforward

Intro
Use Cases
Runtime generation & modification
Models usually cannot be created in the runtime/on the fly
(or at least not easily) and cannot be modified in the runtime
DAE Tools
Models can be created in the runtime/on the fly and easily
modified in the runtime

Intro
Use Cases
Simulation setup
Setting up a simulation (ie. the values of parameters values,
initial conditions, initially active states) is embedded in the
language and it is typically difficult to do it on the fly or to
obtain the values from some other software (for example to
chain several software calls where outputs of previous calls
represent inputs to the subsequent ones)
DAE Tools
Setting up a simulation is done programmaticaly and the
initial values can be obtained from some other software in a
natural way (chaining several software calls is easy since a
large number of libraries make Python wrappers available)
Intro
Use Cases
Operating procedures
Simulation operating procedures are not flexible; manipulation
of model parameters, variables, equations, simulation results
etc is limited to only those operations provided by the
language
DAE Tools
Operating procedures are completely flexible (within the limits
of a programming language itself) and a manipulation of
model parameters, variables, equations, simulation results etc
can be done in any way which a user cosiders suitable for
his/her problem

Intro
Use Cases
Outputs
Only the type of results provided by the language/simulator is
available; custom processing is usually not possible or if a
simulator does provide a way to build extensions it is limited
to the functionality made available to them
DAE Tools
The results processing can be done in any way which a user
considers suitable(again within the limits of a programming
language itself)

Intro
Overview
Architecture
Use Cases
Outline
1 Intro
General Info
Motivation
Main features
General
DSL vs. DAE Tools
3 Architecture
Overview
4 Use Cases

Intro
Overview
Architecture
Use Cases
Available modules
pyDAE:
pyCore (key modelling concepts)
pyActivity (simulation, optimization)
pyDataReporting (results handling)
pyIDAS (DAE solver)
pyUnits (unit and quantity concepts)
FE Solvers:
pyDealII

Intro
Overview
Architecture
Use Cases
Available modules (cont’d)
LA Solvers:
pySuperLU
pySuperLU MT
pyTrilinos (Amesos, AztecOO)
pyIntelPardiso
NLP/MINLP Solvers:
pyIPOPT
pyBONMIN
pyNLOPT

Intro
Overview
Architecture
Use Cases
Available components
Model
Simulation
Optimization
DAE solver
LA solver
NLP solver
Log
Data reporter
Data receiver

Intro
Overview
Architecture
Use Cases
Available components (cont’d)

Intro
Architecture
Use Cases
Outline
1 Intro
General Info
Motivation
Main features
General
DSL vs. DAE Tools
3 Architecture
Overview
4 Use Cases

Intro
Architecture
Use Cases

Intro
Architecture
Use Cases

Intro
Architecture
Use Cases

Intro To Chemical Adsorption

Uploaded by

Copyright:

Available Formats

Intro To Chemical Adsorption

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Intro To Chemical Adsorption

Uploaded by

Copyright:

Available Formats

Intro

DAE Tools: An equation-oriented process

DAE Tools Project, http://www.daetools.com

Dragan Nikolić DAE Tools - Introduction

Dragan Nikolić DAE Tools - Introduction

Dragan Nikolić DAE Tools - Introduction

What is DAE Tools?

Process modelling, simulation, optimization and parameter

Dragan Nikolić DAE Tools - Introduction

What can be done with DAE Tools?

Dragan Nikolić DAE Tools - Introduction

Types of systems that can be modelled

Initial value problems of implicit form: systems of linear,

Dragan Nikolić DAE Tools - Introduction

Why yet another software?

Dragan Nikolić DAE Tools - Introduction

Not a modelling language

A set of software packages

Dragan Nikolić DAE Tools - Introduction

Not a modelling language (cont’d)

Allows easy interaction with other software libraries (two-way

Dragan Nikolić DAE Tools - Introduction

Everything is an object (models, parameters, variables,

Dragan Nikolić DAE Tools - Introduction

Equation-oriented (acausal) modelling

Equations given in an implicit form (as a residual)

Dragan Nikolić DAE Tools - Introduction

Separation of models definition from operations on them

The structure of the model (parameters, variables, equations

Dragan Nikolić DAE Tools - Introduction

Hybrid continuous/discrete systems

Modelling of continuous systems with some elements of

Dragan Nikolić DAE Tools - Introduction

Model export from DAE Tools to other

Dragan Nikolić DAE Tools - Introduction

Support for Functional Mock-up Interface for Model Exchange

Dragan Nikolić DAE Tools - Introduction

Automatic model documentation

Dragan Nikolić DAE Tools - Introduction

Model reports (cont’d)

Dragan Nikolić DAE Tools - Introduction

Model reports (cont’d)

Dragan Nikolić DAE Tools - Introduction

From chemical processing industry to biological neural

Dragan Nikolić DAE Tools - Introduction

Supported DAE solvers@

Supported optimization solvers:

Dragan Nikolić DAE Tools - Introduction

Dragan Nikolić DAE Tools - Introduction

Dragan Nikolić DAE Tools - Introduction

Approaches to process modelling

Two approaches to process modelling:

Dragan Nikolić DAE Tools - Introduction

Domain Specific Languages

Special-purpose programming or specification languages

Dragan Nikolić DAE Tools - Introduction

General-purpose programming languages

Created to solve problems in a wide variety of application