1.

INTRODUCTION TO MODELING AND SIMULATION

INTRODUCTION
Most systems from various branches of science (physics, chemistry, engineering,
economics, sociology, etc.) present a high degree of complexity, being described by a number
sea of variables and interactions. In numerous situations, attempts or direct measurements on
The phenomena in systems are difficult or even impossible. The causes are the most diverse.
it is too dangerous, too expensive, too slow, too fast, too complicated, it cannot be done
the study conditions, the influence of the environment is too strong, the necessary means do not exist, there are
ethical restrictions, must be repeated many times or the studied object exists only in a single
exemplar.
1.2. SIMULATION
In these situations, a relatively new technique for 'virtual' realization comes to the aid of specialists.
experiences: simulation. Etymologically, the word simulation comes from the Latin simulatio,
meaning the ability to reproduce, represent, or imitate something. Simulation is a field
of efficient multidisciplinary investigation. It is based on mathematics, systems theory,
Computer science, etc. Simulation provides a profoundly scientific and precise tool, offering
at the same time the possibility of studying the real states of the systems without their physical composition. The technique
simulation is frequently used in the field of machines and electric drives. The need to
the simulation of various electromechanical systems has arisen due to the need to reduce costs of
design and manufacturing of prototypes, as well as shortening the time frame from the phase
from conception to the realization of the finished product. The designer can operate and manipulate the system with
the assistance of simulation techniques in all phases of the design activity. The use of simulation in
the design of electric machines and electric drive systems offers a number of advantages,
as it would be: • it is suitable for educational purposes, as it allows learning how it works
of components, circuits by tracking the time variation of currents and voltages,
it can provide an accessible and documented perspective on the behavior and performance of the system,
shortening of the prototype development time due to the possibilities of studying the issues
specifies through system simulation, • allows the theoretical study of destructive tests, of
response to malfunctions and operation under abnormal conditions, • studying parasitic effects, such as
and capacities and leakage inductances, the saturation phenomenon, etc.
Modeling of electromechanical systems

Course 1/2
The waveforms resulting from the simulation can be easily monitored and analyzed by replacing
classic measurements subject to errors and disturbances, • new circuit concepts can be easily tested and
variations of circuit parameters (for example, component tolerances), • can optimize
the performance objectives proposed through simulations on a large number of variables, • in simulation not
it is necessary to indicate the nominal values of the components, • the interaction between can be studied
two converters powered from the same network, something almost impossible to achieve without simulation,
Certain parts of the circuit can be simplified in the model to allow focus on other aspects.
circuit portions.
1.3. THE MATHEMATICAL MODEL

The mathematical model is the description of certain phenomena, social or natural, or natural objects in
the physical elements are replaced with logical elements, of formal mathematics in view of
study of the respective phenomenon. The model in general must reflect the properties
main features of the phenomenon or object in a simplified form. A general concrete model
it is very complicated and even impossible to elaborate. For this reason, models are used
simplify, which reproduce only certain aspects of reality. In this way, instead of systems,
the phenomena and real objects are analyzed in a model more or less similar to the one
Real and from the behavior of the model, conclusions are drawn about the functioning of the entire real system.
The mathematical model must rigorously describe in the form of mathematical representations of
the relationships existing within the respective system, practically formalizing the laws of behavior of the system.
For a system, a series of models can be established. Specialists believe that there is no
there is no unique or correct model for a specific system, but rather there are good and bad models.
the model constructed must satisfy two requirements that are always in conflict: to be as complete as possible
and faithful. The correct choice is only possible through the application of quantification methods, observations
statistics, tests, measurement techniques, etc. When choosing the model, considerations must be taken into account
the limits of calculations to be performed on the given calculation platform (calculation speed and memory
available). The main properties of a mathematical model are as follows: •
acceptability • applicability • usability • completeness The classification of models can be carried out
according to several criteria, namely: • by the purpose of using the model: for description, for presentation,
of analysis and forecasting, • according to the nature of the modeled phenomenon: social, chemical, biological, of
production, etc., • according to the nature of the model: material, electric, mechanical, cognitive, computer-based,

Modeling of electromechanical systems

Course 1/3
• model type: of construction, of operation, formal depending on the values they take
state sizes (variables): model with discrete variable states, with continuous variable states, with
mixed variable states, • depending on the time dependency of the variables: dynamic model, dependent
time-dependent (at least one of the variables depends on time), static model, time-independent (no
the variable does not change over time), • depending on the output variables: deterministic model (the input and
the initial state unambiguously determines the obtained result), stochastic model (the input and the initial state
determines only the distribution of the final result values). • According to the model's dependence on the future:
unpredictable (which does not foresee the future) and predictable. According to systems theory, a system (and
its implicit mathematical model) can contain the following main elements: • parameters of
input • state sizes (variables) • output parameters The mathematical model can never
It is not the purpose of modeling itself, but it should be a useful tool in the analysis of the modeled phenomenon.
Model creation is based on the application of abstraction, that is, the adaptation of some simplifications of
of the studied system so that the basic character of the process does not change. In this case, it does not
a series of particular characteristics of the simulated system are taken into account. The models are
achieved through the following steps: • observation • hypothesis creation • actual modeling •
experimenting with the composed model • finalizing the model It is important to note that
the similarity between the constructed model and the modeled object (system) is only partial. Very
However, the existence of this similarity is important, because only in this way can it be ensured
determining the real states of the model. In principle, there are two distinct ways to solve the
equations of a mathematical model: through analog devices or through analytical-numerical calculations.
Based on these paths, there are two different simulation methods. • analog simulation (the simulation
on the system) with the help of reproducing the relationships from the studied system using some devices
name simulators. • simulation on model by performing calculations for analytical resolution of
the equations of a mathematical model of the system. The most commonly and efficiently used nowadays
this numerical simulation carried out with the help of computers, given the computing speed and
the large memory capacity of these electronic computing devices. In this course
we will exclusively focus on this category of simulations. The simulation procedure of a
the system includes several stages:

Modeling of electromechanical systems

Course 1/4
1. Identifying the problem. The correct and complete formulation of the problem to be solved must be made.
in the terms of the given discipline. The system to be studied is analyzed: the system is decomposed into parts
component for it to be analyzed in order to understand its nature and characteristics
essential. At this stage, the accepted assumptions about the phenomenon must be specified and estimated
the sizes and dominant variables of the system, as well as the input parameters. It is a stage
very important, as this is where the purpose of the activity and the clarifications made here are determined
determines the results obtained from the simulation. 2. The mathematical formulation of
the issues (the actual modeling). In this stage, the dependencies are established
among the different sizes determined in the previous point. In the vast majority of cases, it is obtained
ordinary differential equations or partial derivatives, upon which conditions are imposed
supplementary (initial, at limit, etc.) 3. Solving the mathematical problem. It is the most
an important stage of the simulation process. Only by choosing the correct method of solution and
by applying it correctly, the required and useful results can be obtained. At this stage, it is chosen
the programming language used and the actual simulation program is developed. Also in this
the stage tests and verifies the designed simulation program. The final phase of this stage is a
constitutes the practical execution of the simulation (running the simulation program). 4. Validation
simulation program. At this stage, the results obtained are compared with the help of
the simulation program with those obtained experimentally, considering both the errors of
measurement, as well as the accuracy of the mathematical solving methods used. The simplest and safest
The validation scale is testing the program on a particular case, for which the solution is known.
Testing is effective if all branches are covered and all subroutines of the program are requested.
5. Analysis and interpretation of results. This stage consists of collecting the simulated data and
their processing. The results of the simulation can be obtained in the form of tables, histograms, representations.
graphs, etc. The obtained data can be processed statistically (for example, tests can be performed)
significance, etc.). The development of simulation programs for electromechanical systems is actually
implementation phase of the mathematical model on the computer. The main requirements imposed
The simulation programs aim to: • have a user-friendly interface. • have possibilities
multi-level modeling (for example, the electric car and its task are characterized using
differential equations formulated in terms of state variables, and their control systems are
describe through transfer functions or logical functions). A good simulation program must
to allow each block to be described in the most advantageous way. • to be able to cover a wide
domain of simulation times, considering that in general the modeled systems are both
variables with rapid variations, as well as variables with slow changes, due to large constants of
time. • to allow the user to easily set the initial conditions. • the system parameters to
it is as easy to modify as possible, even during simulations, thus giving it a feature
general program
Modeling of electromechanical systems

Course 1/5
As previously stated, the simulations are based on the differential and algebraic equations that describe
the operation of the system. Simulation programs differ primarily in the way they
solve these equations that describe the studied system. Thus modern simulation programs are
I divide into two main categories at the user interface level: equation solving programs
and circuit-oriented programs. Equation solving programs are very useful in many
simulation problems, both in power electronics and in the field of electric machines.
The simplified system can be described by state variables. The state equations can be solved either
using a high-level language (Fortran, C or Pascal), either with the help of advanced environments
calculation (MATLAB, LabView, SIMNON, ACSL or MATRIXx). In case it was chosen for
the use of a general high-level programming language must be composed of a logical scheme
detailed, explained, which should contain all mathematical equations in an expanded form [2]. For
these programming languages have already developed a series of libraries aimed at solving
differential equations, as well as operations with matrices and graphs. Calculating devices are
designed to solve a general class of problems. In order to use them,
the user must describe the system using differential equations with state variables, and
the controllers should be specified with transfer functions (for example in the form of a transform
Laplace). Usually, these simulation programs allow modularity, meaning each part of
the model can constitute a module by itself, and the interconnection of modules can be done easily with
connection lines.
1.4. SIMULATION PROGRAM
There are special programs designed for simulation. Nowadays, the most widespread of these,
both in academic and industrial environments, is SIMULINK. This graphical interface has
The MATLAB environment is particularly aimed at simulating dynamic systems. Programs
oriented towards the circuit develops even the equations of the system from the information received from
user. The user in this case must provide the program only with the interconnections between
the models of circuit elements. The initial conditions are also established very quickly. There is
the possibility of being able to easily make changes to the circuit topology. Good programs
oriented towards the circuit are multi-level, which means that in addition to the orientation description
the circuit of the controllers allows the incorporation of system models defined by the user based on
differential equations. There are many circuit-oriented programs. Some of these are
dedicated to the simulation of electronic circuits (the most widespread being SPICE), or electrical networks
In addition to these, programs aimed at the circuit have emerged on the market, especially
simulations in the field of power electronics and electric drives (SIMPLORER or
CASPOC). These programs differ from each other by the quality of the user interface,
the numerical integration methods provided for solving differential equations, the way of handling
of nonlinearities, the choice of integration step, the ease with which controllers can be incorporated and
external models for advanced multi-level simulations or the approach for handling
electronic switches.
Modeling of electromechanical systems

Course 1/6
1.5. SIMULATION OF ELECTROMECHANICAL SYSTEMS

Electromechanical systems exhibit a series of particularities that must be taken into consideration.
the simulation time. In electromechanical systems, consisting of electrical drive systems and
the actual driven system, the energy that is taken from the source and transferred to the process
technological can undergo three types of conversion: • Electric energy parameters, in
the converters with semiconductor elements placed between the industrial network (source) and the machine
electric. Electronic converters do not change the nature of energy, but only its parameters (form,
frequency, average value, effective value, etc.). • Electromechanical, performed by the electric machine
of operation. Electric vehicles in electromechanical systems predominantly operate in
engine regime, absorbing electrical energy and supplying mechanical energy to the working machine. • A
parameters of mechanical energy, in the transmission located between the electric motor and the working machine
activated. Where the technological process requires parameters of mechanical energy different from those
the electric car can economically ensure the conversion of mechanical parameters
this is carried out by the transmission placed between the motor and the working machine. This transmission may be missing
in the case of direct coupling, but it can also be very complex when the nature of the motion changes
(rotation conversion - translation, etc.). Despite the great diversity of technological processes that
consumes mechanical energy provided by the drive motor, characterizing the working machines at
the level of the engine's crankshaft (or the moving part in translational motion in linear machines) is
satisfactory face through a limited set of parameters: torque, angular speed, angular position,
equivalent moment of inertia, road diagrams, etc. The study of electromechanical systems has a
an experience of almost a century, which has allowed the maturation of analysis methods and has influenced
the working procedures and in the adjacent fields. From this historical perspective, modeling
electromechanical systems present a special interest for all specialists working in
this domain [3].
1.6. BIBLIOGRAPHY
FOWLER, A.C.: Mathematical Models in the Applied Sciences, Cambridge Texts in Applied
Mathematics Series, Cambridge University Press, Cambridge, 1997. [2] DODESCU, GH. –
SCHIANU, Ş. – OGĂDESCU, I. – NĂSTASE, P.: Simulation of Systems, Military Press, Bucharest,
1886. [3] SORAN, I.F. – KISCH, D.O. – SÎRBU, G.M.: Modeling of conversion systems
energy, Ed. ICPE, Bucharest, 1998.