MS1®, a multi-language modeling and simulation environment for continuous systems, is designed to support the unique requirements found within the various technical and research departments of industry. MS1 accepts different non-causal model representations such as Bond Graphs, Block Diagrams and Differential-Algebraic Equations (DAEs). Bridges also exist to call FORTRAN or C subprograms from within a model (depending on the chosen solver capabilities). These features have been included to meet the needs of particular modeling problems, allowing MS1 to handle complex models specific to many industry sectors. Models with up to 80 000 state variables have been successfully coped with by MS1 customers.
MS1 interfaces with major solvers on the market (Esacap™, Matlab®, ACSL®...) and can be easily adapted to customer specific solvers (for instance, MS1 has been interfaced to the FORTRAN-based LEDA solver used by EDF). Models can alternatively be exported to Simulink® (as C-coded S-Functions), Maple® or Mathematica®. MS1 performs a symbolic manipulation of the model (using especially a powerful causality analysis engine) and generates for the user the corresponding simulation code.
Results generated by the solver can be displayed from within MS1 as plots of any variables versus time or another model variable. Specific to the Bond Graph representation, the symbolic animation and effort-flow diagrams yield a deep insight into the model energy structure.
MS1 provides tools (via a highly visual system) allowing the user to check the validity of his model throughout the development phase. MS1 detects processes and manages algebraic constraints as well as functional discontinuities thanks to its analysis and evaluation modules. This ensures a coherent and valid model at each modeling stage. MS1 also permits topological analyses of the model (looking for causal loops and paths between variables).
MS1 is flexible and modular and allows multi-level hierarchical decomposition of models. Users can "navigate" straightforwardly from sub-model to sub-model. Each model can have several coexistent versions, allowing "polymorphic modeling". Furthermore, an MS1 study (models and submodels organized in a single file) can invoke submodels stored in other MS1 studies, permitting an easy development of bespoken model libraries.
Most of the above-mentioned features (acausal representation, hierarchical decomposition, polymorphic modeling and model libraries) aim to provide enhanced model reusability, which was one of the MS1 project major objectives.
MS1 is different from most other programs due to the following characteristics :
As stated hereabove, these features aim to provide enhanced model reusability, which was one of the MS1 project major objectives, permitting an easy development of bespoken model libraries
MS1® is an interactive environment for modeling, analysis and simulation of linear and non linear dynamic systems.
The description of models is done in a intuitive and simple way with the mouse, by "drag & drop" of basic components available in MS1. The whole process, from modeling to simulation, is controlled from the same interface. The control is made either via menus, icons, short cuts or keyboard commands. MS1 can be used in "batch", which yields an easy integration in a larger simulation environment.
MS1 is available for platforms using standard operating systems and interfaces: Unix HP-UX 10.20, Sun Solaris 2.5 to 7, RedHat Linux 2.0.32 (or further), Windows 95/98/NT and 2000. The interface presentation and the commands are homogeneous whatever platform is used. In the same way, the compatibility of files is guaranteed to transfer 'ms1' files from an operating system to another, for instance if you use an heterogeneous data-processing environment.
Moreover, MS1 answers the preoccupations of backward compatibility in guaranteeing recovery of your models described under former versions.
MS1 and its documentation are available in English and in French.
MS1 is a environment describing and analyzing the dynamics of linear or non-linear systems, very powerful to describe complex components of systems and to simulate huge models.
MS1 is an open environment, very flexible which integrated your practices of designer.
You describe and build your models with the mouse, by "drag & drop" of the components predefined in MS1. These components are sorted by "family" : sources, sinks, algebraic operators, integrators, connections, inputs and outputs and can be adapted. It is possible for you to create and develop your own library of components and sub-systems, from the proposed elements.
The phenomenological laws and the parameters are indicated in dialog boxes accessible from the components.
Unlike more classical modeling tools, MS1 allows you to choose the representation mode (formalism) closest to your practices or your needs to describe your systems. MS1 provides ranges of Bloc-Diagram components and Bond Graphs components (for energy studies), but also allows you to use the description by equations.
MS1 answers the model durability preoccupations by making it fully possible to call subroutines in C and Fortran. As you preserve and reuse your existing models, you capitalize your investments.
Building model is made by hierarchical description: you can describe a whole system with subsystems, without limitation of the number de subsystems created, using indistinctly top-down and bottom-up approaches.
This freedom of choice for description of your models induces that subsystems can be described in various formalisms and that, you can thus describe a complete system with subsystems using different modes of description.
MS1 proposes to you a permanent visual help during the construction of models and deals with actions, releasing you from tiresome handling or from constraints relating to the numerical phase. MS1 thus enables you to center yourselves on your job of engineer.
MS1 informs you that the model is structurally coherent and complete, that means that you introduced all required information such as laws, parameters, links and nodes variables, initial conditions, characteristics specific to calculations,...
At the beginning of the modeling phase, you can also call upon the function of generation by default of laws and parameters for a first level of study of your system.
You can personalize your models, whole or part, by giving by yourself specific names to the variables. By default, MS1 automatically generates them, just like it automatically propagates them through the hierarchical structure of the model.
MS1 is a flexible and open tool, which fits to you environment by proposing interfaces towards standard solvers of the market or solvers specific to your company.
MS1 automatically undertakes the shaping of equations for the calculation phase: equations are analytically handled if necessary and are automatically ordered and sorted for the selected solver. Thus, the algebraic loops, derivations and integration are systematically identified and treated by MS1. You can thus write equations in a declarative way, which frees you from the numerical aspects.
Simulation of a same model with various solvers, without changing the code of simulation submitted to the solver, is easy with MS1, provided that the solver is able to cope with the characteristics of your model. If necessary, MS1 informs you of the solver limitations, after analysis of the structure and characteristics of your model.
This analysis is possible for systems described in Bond Graph. The Bond Graph method of representation allows study of the energy phenomena whatever the concerned physical discipline is. It is a true technological "plus" proposed in standard in MS1. Powerful and federating method, it is spread more and more within industries, promoted by research and teaching centers of high schools.
After selection of the solver and generation of the simulation code by MS1 for said solver, the launching of solver in made transparently and automatically from general interface. You can also launch simulation in batch.
Interfacing MS1 with various solvers (Matlab®-Simulink®, ACSL®, Esacap™ or solvers specific to the customers) has several advantages.
MS1 fits your environment if you already have solvers with which MS1 interfaces.
In the same way, you can transparently migrate form a solver to another in order to compare several simulations. "Touchy" models for instance require this type of comparison.
You can exploit for each model or each class of model the best adapted solver and consequently the most efficient.
In MS1, interfaces with formal equation manipulators (Maple®, Mathematica®) make possible the use of models for other aims than numerical simulation.
When the simulation phase is achieved, MS1 recovers the results to display them. The use of the display module presents standard functions such as buttons "on/pause/stop", selection of the starting point in time to launch display of the results.
Among standard display functionalities, MS1 proposes for instance manual or automatic selecting / deselecting of the variables directly on the model or in the list of the variables, state or phase diagrams (plot of a state variable versus its derivative), possibility to display a variable versus another one, multi-scales loops.
MS1 integrates innovative capacities of display particularly developed for models described in Bond Graphs . These animations are automatically generated by MS1 on the basis of the system characteristics introduced in the modeling phase, without having to give any further information.
Symbolic animation, which is a "living" and didactic animation, allows display of the energy fluxes in the studied system. On the other hand, it will give you a first approach of the results in the phase of adjustment of the model. Animation is particularly used by our users when they communicate with their colleagues specialists in other fields or their hierarchical superiors. The global and systemic vision of the studied system enables a quick locating of modeling errors and an intuitive understanding of operation.
The representation of effort/flow graphs illustrates the "trajectory" of each pair of energy variables and proves to be particularly useful in the dimensioning phase.
Following summary presents selection criteria of a solver interface, taking into account the characteristics of the systems that you model and simulate.
All these interfaces "MS1+solver"allow:
MS1 also exploits particular characteristics, proper to the solvers, allowing to exceed these minimal characteristics:
|Advantage of the combination||Solver characteristics|
|MS1 + Esacap||
- steady state calculation
- transients + stationary initialization
- transients + non stationary initialization
- DAE implicit (algebraic loops) allowed
- derivation allowed
|- interpreted model|
MS1 + Matlab
MS1 + Simulink
- autonomous matlab code
- export of submodels to Simulink through C-coded S-functions
- call of C subprograms (S-functions only)
- interpreted model
- no display of results for the S-functions
- needs a C compiler (for the S-functions)
|MS1 + ACSL||
- call of Fortran sub-programs
- DAE implicit (algebraic loops) allowed
- needs a Fortran compiler
- compiled model: very fast for big models
MS1 | Esacap
Products | Information | Contacts | Home