Marisol Delgado

Pseudo-bond graph model and simulation of a continuous stirred tank reactor

Abstract A dynamic model is developed that allows us to represent the behavior of a continuous stirred tank reactor (CSTR) for the saponification of ethyl acetate with sodium hydroxide. The model is based on the bond graph technique, which lets us represent the molar and energy balances of the system. Various simulations were done using the ENPORT 5.2 program, enabling us to validate the model with data taken directly from the real system. A satisfactory behavior was obtained for the thermal variables involved, but for the concentration, the model tends to deviate even for values at steady state. This is due to the difficulty in estimating the experimental values of the reaction rate constants.

A survey of bond graphs : Theory, applications and programs

Abstract This paper presents a survey of the bond graph method from its conception and creation by Prof. Henry Paynter to the present. A thematic classification of the bibliographical material is given depending on whether work deals with the definition of theoretical details of bond graphs, applications of bond graphs to particular fields, or if it deals with a computer program developed using the bond graph theory. A list of journals which publish bond graph related articles, as well as textbooks is also presented. The survey, though exhaustive, does not pretend to be all inclusive. Some works may have been unintentionally omitted due to obvious limitations on accessing and listing all the relevant literature concerning the topic.

Proportional-integral observer for systems modelled by bond graphs

Abstract In the present work a bond graph approach to build full-order proportional–integral (PI) observers is shown. The proposed approach is based on the bond graph method defined for designing classical linear observers (Luenberger observers). The method is extended to build a second type of integral observers more robust to the presence of sensor noise. As application, the method is used over the pseudo bond graph model of a continuous stirred tank reactor (CSTR). Simulation results show the performance of the PI observers with respect to the presence of modelling errors and the robustness of integral observers when sensor noise is present.

A bond graph approach to the modeling and simulation of switch regulated DC-to-DC power supplies

Abstract A bond graph approach is proposed for the modeling and simulation of switched DC-to-DC power converters of the boost type, regulated by means of a pulse-width-modulation (PWM) strategy. An average bond graph model of the PWM regulated DC-to-DC power converter is derived using ideal lossless transformer elements. The proposed bond graph model constitutes an “input” to a computer aided package, called DESIS, which automatically generates the system model as a set of differential equations with explicit control parameters. The computer package is also capable of simulating both the average and the actual discontinuous PWM regulated responses of the DC-to-DC power converter for a given feedback duty ratio policy.

Modeling and simulation of switch regulated dc-to-dc power converters of the boost type

This paper presents the bond graph modeling and simulation of a switched dc-to-dc power converter of the boost type, regulated by means of a pulse-width-modulation (PWM) strategy. Two models are presented, the average model, which represents the switch as an ideal linear transformer element and the actual (discontinuous) model, which represents it as a non-linear transformer element whose function is a non-periodic pulse train taking values in the discrete set {0,1}. Both transforms are modulated by the pulse width. The bond graphs were simulated and both the average and the actual discontinuous PWM regulated responses of the dc-to-dc power converter for a given feedback duty ratio policy were obtained.

Luenberger observers for linear time-invariant systems modelled by bond graphs

This paper shows how to build Luenberger observers for linear time-invariant systems modelled by bond graph. The methods are based on Luenbergers algebraic methods to design both full-order and reduced-order observers. The procedure for reduced-order observers uses the bicausality concept to simplify some classical matrix calculations (the calculation of matrix inverses is not needed), which is an important improvement mainly for large-scale systems. The calculation of the observer gains is based on the pole-placement techniques for linear systems modelled by bond graphs. As an application, both observers are designed for a vehicle suspension modelled by bond graphs.

Pseudo-bond graph model and simulation of an industrial flash separator

Abstract The present paper describes the modeling and simulation of a two-phase flash separator. This equipment is part of the natural gas liquefaction process of the San Joaquin extraction plant located in Anzoategui State, Venezuela. The principal function of the separator consists in taking the extracted propane to saturation conditions so as to take advantage of its refrigerant properties. The pseudo-bond graph methodology was used in modeling the system. Such methodology was very suitable for this chemical process since it allows good management of the non-linearities present in the system while also accepting the use of elements that do not yield to the rules of traditional bond graph methods. Several simulations were accomplished with 20-SIM program and the model was validated using real operating conditions of the plant and by using simulation results from a previously developed model.

Use of MATLAB and 20-sim to simulate a flash separator

Abstract This work presents the simulation of a two-phase flash separator, one of the processes of an industrial natural gas liquefaction plant. The model simulated was developed using the bond graph methodology, and MATLAB and 20-sim software were used to obtain the dynamic behavior of the flash. MATLAB was used to determine the input conditions of the separator as they were not provided by the plant information, and 20-sim was used to simulate the bond graph model. The simulation results were validated using real operating conditions of the plant. The paper shows how this two widely used computer programs can help in the understanding of the behavior of real process.

Bond Graphs in the Design of Adaptive Passivity-Based Controllers for DC/DC Power Converters

In this paper a new approach is proposed for the design of adaptive passive dynamical compensators regulating the output load voltage in pulse-width-modulation (PWM) DC/DC power converters. The approach is based on the bond graph description of the system. For the design of the passivity-based controller, a damping injection on the error dynamic bond graph is performed where dissipation is added in order to achieve asymptotic stability. Bicausality is used to determine the steady state value of the desire output variable. This controller is then modified to generate an adaptive feedback scheme. The non-minimum phase nature of the average output voltage variable is demonstrated by a graphical analysis performed directly on the average bond graph of the converter. The performance of the proposed controller of each converter is tested through computer simulations. The results obtained in the paper can be extended without any difficulty to other systems represented via bond graphs.

High Gain Observers for Non-Linear Systems Modelled by Bond Graphs

Abstract In the present work, a bond graph approach to build full-order high gain observers is shown. The proposed approach is based on the classical techniques to design high gain observers. Simulation results are used to show the performance of the high gain observer in chemical engineering applications as a distillation column modelled by pseudo bond graphs.