Development of a foss- based hardware- in- the- loop platform for control engineering education

Artigo Submetido em xx/xx/xx

1a. Revisão em xx/xx/xx; 2a. Revisão em xx/xx/xx; 3a. Revisão em xx/xx/xx; 4a. Revisão em xx/xx/xx;

Aceito sob recomendação do Ed. Consultor Prof. Dr. Yyyy Yyyyyyyy yy Yyyyyyyy



DEVELOPMENT OF A FOSS- BASED HARDWARE- IN- THE- LOOP PLATFORM FOR CONTROL ENGINEERING EDUCATION

Flavio Padilha Queiroz [e-mail]	Francisco José Gomes [e-mail]
Luka Parma de Freitas [e-mail]	Vinicius Athouguia Gama [e-mail]

Electrical Energy Department, Electrical Engineering Course

Engineering College – Federal University of Juiz de Fora – MG - Brazil

Abstract: This work deals with the development of a laboratory platform based on the Hardware in the Loop - HIL simulation techniques, designed and assembled for utilization as a didactic tool in the control engineering education area. It allows simulation and analysis of industrial control loops and controllers dynamics belonging to the shop floor operations, but with potentiality for operation in the supervisory level, incorporating safety and optimization techniques. The developed module, based on Java language and Eclipse compiler, that are Free Open Source Software - FOSS tools, encompasses a three-dimensional digital environment that simulates an industrial plant dynamics, but whose operation is based on hardware components: an external PID controller and an industrial inverter. Being a FOSS-based development this module has no restriction for its utilization. The interface between the hardware components and the three-dimensional digital environment is based upon a PIC 16F877A design. The developed module simulates, partially, the dynamics of a real pilot plant belonging to the Industrial Process Laboratory of the Engineering College of the UFJF, incorporating some of its complexities and nonlinearities and operates with the same time constants. The developed HIL platform, totally FOSS-based, can reinforce the educational laboratory practices for improving engineering education especially that one related with industrial process control, supervision and optimization.
Keywords: Hardware in the Loop – HIL, FOSS, industrial process control, control engineering education.

1Introduction

Industrial process control loops, if adequately controlled, can guaranty comparative advantages like the increase in the gain and stability margins, decrease in energy consumption, faster response and stabilization times, increase of production rates and minimization of environmental impacts, among others. Considering the extremely competitive and innovative environment faced by industrial processes in the actuality, it is easier to understand that even these small details can make all the competitive difference. In such a situation, adequate industrial process control strategies have become efficient ways for optimizing process productivity, production efficiency and the quality of their final products; it is a well-known fact that even marginal improvements can impact positively and deeply their economic and environmental aspects, enlarging the dimensions associated with the correct utilization of industrial process control strategies.

It is known, however, that in the industry environment reality a significant number of industrial control loops operates with inappropriate strategies, incorrect structures and inadequate tuning procedures. Even for the omnipresent PID controllers, it has been reported that, in the operators jargon, the derivative mode, D, is associated with the words “Disaster”, “Dangerous” or “Do not utilize...” Cooper, 2010). If one considers the amount of conceptual analysis and references that can be found in the literature concerning the PID strategy (Aström, 2006; Aström 1995; Visioli, 2006; Normey-Rico, 2007; Sung, 2009, just to quote a few) it becomes clear that, in a great number of situations, the theoretical knowledge repository and its practical utilization by the industrial operators, does not match. A possible explanation for such a situation may be found in a lack of understanding of the PID modules functionalities, misunderstanding of the complexities associated with the industrial process dynamics and conservative - or even wrong-, understanding of the PID tuning procedures. This situation can convey to fearful attitudes of the industrial processes operators, concerning the appropriate utilization and tuning of the PID controllers.
Another important point can also be taken into consideration: the engineering is a practical profession, hands-on, and since the very beginning of its education, didactic laboratories have been playing one of the main roles inside the education process, especially in the industrial process control area. One can even assert that, prior to the scientific component of engineering education, the majority of the engineering professional profiles are worked out in the didactic laboratories (Feisel, 2005). The explanation is obvious: the laboratory activities, inside the educational process, have explicit characteristics and objectives for establishing the engineering professional profile as, for instance, to understand and utilize conceptual models; to know how to collect, analyze and interpret data; perform projects on demand; to learn with mistakes and errors; to acquire a creative view; to know how to communicate effectively, in both oral and written conditions, the results and conclusions of his work and, finally, how to deal with a teamwork project. (Peterson, 2007). A more intensive utilization of the laboratory practices inside the educational process, however, faces problems that vary from a misconception of the educational process until operational situations, like the costs associated with the acquisition, operation, maintenance and updating of the laboratory equipment. This situation imposes the necessity of finding alternative, or complementary, ways and new postures for a better and more intensive utilization of laboratory practices in the engineering educational process.
As a consequence of this situation, some alternatives for reinforcing laboratory educational practices have been tried. Among them, the Hardware- in- the- Loop (HIL) techniques have been viewed as a possible solution for design, analysis and assessment of complex systems. With the help of these techniques, it is possible to simulate complex processes utilizing real time conditions that resemble very closely the dynamics and complexities of the real physical systems under analysis. The HIL techniques utilize three-dimensional digital environments which reproduce the real systems dynamics, but whose interaction with the operators occurs through a physical hardware interacting with the digital environment. The three-dimensional digital environment, based on high-order models of the process dynamics, sensors and actuators, attempts to simulate, as closely as possible, the real working conditions of the real processes (Pfister, 2010, Coelho, 2007, Egel, 2009 ).

Based on its low-cost and versatility, the HIL technique has been becoming a pattern in the design and development of complex systems, in the starting steps of new project designs, for operators training situations and process assessment and, also, in the educational area. As it enables, in the very beginning steps of a process development or training situation, to simulate and forecast dangerous and risky situations, for both operators and systems, it enlarges the safety conditions; its convenience for application and utilization can be considered an additional factor that reinforces it as a powerful tool for design, development, control and assessment of industrial process and marketable wares (Gu, 2007; King, 2004).

A possible strategy for enlarging the HIL utilization in the educational area can be made possible if one takes on alternative solutions to some of its component: the software utilized for developing the three-dimensional digital environment. The impact of the software costs on the final price of a HIL platform can be exemplified with the well known digital environment LabView®. This software, largely utilized in the academic institutions, as well as in the industry environment, for simulations, design and analysis, including Control Prototyping strategies, has costs around U$ 20.000,00, even for its educational version, with the associated toolboxes.

Another example: educational simulators, like the ITS PLC Professional Edition^®, from Nova Didacta, have costs around U$ 5.000,00 for each desktop license, just for the digital environment, without the costs associated with the hardware components (Nova Didacta, 2009).

Facing this situation, the present proposal was to design and assembly a low cost, although technically consistent, HIL platform for engineering education utilizing Free Open Source Software – FOSS. Although the most known characteristic associated with FOSS is its low cost, practically null for the users, others not so known characteristics, and quite so important, must be taken into consideration and can explain its growing power and utilization, all over the world, and its widespread acceptance in the academic and industrial environments. They include safety, reliability and stability, open patterns and independence from the providers, reduced dependence from importation and the strengthening of the technological local capacity (OpenSource, 2010).
Inside this context, this work discusses the development of a laboratory HIL platform, FOSS-based, for utilization as an engineering educational tool in the industrial process control area, based on a real pilot plant existing in the Industrial Process Control Laboratory belonging to the Engineering College of UFJF. This pilot plant allows the analysis and control of the loops associated with the main variables of the industrial process area: temperature, flow, pressure and level. The proposal was to reproduce its nonlinear tank level-control dynamics, encompassing an electric pneumatic valve, a PID controller and the complexities associated with a hydraulic pump controlled by an industrial inverter. The PID controller and the industrial inverter are integrated inside a hardware block, while the plant dynamics is simulated through a three-dimensional digital environment. All the development was based on Java, on the Eclipse compiler and on Blender, FOSS environments. The future steps include the aggregation of the other loops and dynamics belonging to the pilot plant.
The article details the environment characteristics, the development of the digital environment and the necessary interfaces with the hardware equipment, the results obtained and, finally, conclusions and suggestions for future works.

2METhODOLOGy

The HIL platform conception arose from an industrial pilot plant located at the Industrial Process Control Laboratory utilized by the Electrical Engineering Course of the Engineering College, at the UFJF (Gomes, 2008): a double tank system for analysis and control of the main variables utilized in the industrial process control loops – level, flow, pressure and temperature. These four variables are inside a single circuit, but which can be configured according distinct topologies (Figure 1); this pilot plant operates with industrial signals, more specifically 4 to 20 mA, and some of the loops can also utilize the digital protocol Profibus^® PA.

Figure 1 – Industrial Pilot Plant utilized as a basis for the HIL platform development.

It operates with an industrial supervisory system that manages all the variables, including a Logical Programmable Controller - LPC. It is also possible to utilize analogical PID controllers instead of the LPC algorithms. A differential characteristic of this plant is the fact that it utilizes industrial sensors and actuators, with all their complexities and nonlinearities, allowing their calibration and adjustment which characterizes, undoubtedly, an industrial environment for control education.

Figure 2 – Structure and modules of the proposed HILS.

As an attempt to reproduce this reality, the proposed HIL platform encompasses 04 independent systems working interconnected (Figure 2): a three-dimensional digital environment that reproduces the double tanks, sensors and actuators dynamics; an interface system between the digital environment and the hardware components, more specifically the PID controller, the inverter and a manual valve. Utilizing these hardware components the operator can interact with the digital environment and make all the necessary operations he needs; the designed interface operates through the serial port.

3Project design

3.1Mathematical Modeling

The mathematical modeling of the plant dynamics, concerning the nonlinear tank, was based on the well known Bernoulli equation that relates the pressure, velocity and height in the steady motion of an ideal fluid (Lamb, 1953):

(1)

(2)

where v₁ and v₂ account for the velocity on the high and low points of the liquid, S₁ and S₂, the superior and inferior areas of the input and output valves of the tank, ρ the fluid viscosity, g the gravity acceleration and h the fluid level relative to the output valve in the tank, that represents also the desired controlled variable of the system:

(3)

The S₂ value is determined through the PID action that sends the signals for opening and closing the electrical pneumatic valve, which is the final actuator. The electrical pneumatic industrial valve modeling was based on the equations (4) and (5), presented in the sequence (Wade, 2004):

(4)

(5)

These expressions reproduce the flow inside a valve with linear inherent characteristics. The C_v coefficient in equation (4) varies with the valve position, from a minimum value until a maximum value, C_vmax, that occurs when the valve is totally open; m represents the opening percentage of the valve. In equation 5, f accounts for the flow inside the valve and f_max the maximum flow when the valve is totally open; the β coefficient express the relationship between the maximum and the minimum pressure drop inside the valve, the so called installed characteristics of the valve (Wade, 2004).

Utilizing the geometrical relations associated with the nonlinear cylindrical tank, it is possible to express the relationship between the fluid volume inside the tank, V, and the fluid level h, the controlled variable, according to the equation (6), where r is the tank radius and L its length:

(6)

At each sample interval the cylindrical tank volume V is updated based on the equations described: the Bernoulli equations, the valve characteristics and the geometrical relationship that express its volume.

Figure 3 – Three-dimensional digital environment of the HIL.

In order to develop the three-dimensional digital environment (Figure 3) that simulates the double tanks dynamics, the manual valve that controls the input flow to the cylindrical tank, the hydraulic pump propelled by the industrial inverter and the electric pneumatic valve controlled by the PID, the Blender, an open source, cross platform suite of tools for 3D creation environment has been utilized (Blender, 2010). The three-dimensional environment displays also, graphically, the PID control action and the dynamics of the controlled variable. The data of the hydraulic pump flow and the inverter action are also displayed for the user. The time constants of the digital environment dynamics are very similar to that of the real plant; additionally, the environment reproduces the sound characteristics of the hydraulic pump, of the liquid flow and of the pneumatic valve in such a way that the user has the same feeling as he were inside a real industrial plant.

Figure 4 – Operation of the HIL module: the operator changes manually the fluid flow control valve.
The digital environment also displays the tank level variations and the movement of the valves. The manual valve, utilized by the operator for controlling the input flow in the cylindrical tank and also for introducing load disturbances into the system, operates manually through an external potentiometer (Figure 4); the electric pneumatic valve is driven by the PID action and the fluid flow in the circuit can be altered, also manually, by the operator, through the industrial inverter.

3.2Communication Interfaces

It has also been developed, as an integral part of the project, a communication interface between the three-dimensional digital environment and the hardware components, more specifically the manual valve, the PID controller and the industrial inverter. With the help of this interface it is possible, for the user, to operate the system on a closed loop topology, to introduce load disturbance on the fluid flow entering the tank and to adjust the hydraulic pump flow with the inverter.
In this way, the user can operate manually, on the hardware block, the upper valve and adjust the fluid flow, in the digital environment, introducing manually load disturbance (Figure 4). It is possible also to change the level set point on the PID controller, adjust its tuning parameters and other functionalities, like soft start, antiwindup reset and, even, utilize the auto tuning option. By means of the inverter, the user can also control the hydraulic pump and adjust the fluid flow according to their needs.

Figure 5 – DA converter circuitry.
The circuitry diagrams are displayed on figures (5) and (6), where one can see that the communication interface between the software and hardware blocks has been separated into modules, with distinct functionalities, including the digital/ analogical/ digital conversion; this conversion utilizes an embedded PIC strategy through the CI MAX232, responsible for the communication between the PID and the digital environment. This interface can deal with any type of PID controller with industrial input-output characteristics based on voltage or current signal, as one can see in figure (6).

Figure 6 – Interface with inverter.

Figure 7 – General view of the HILS module.

The complete HIL platform can be seen on figure (7): it is possible to see, on the right side, the three-dimensional digital environment and, on the left side, the developed hardware block with its distinct functionalities: the manual valve for adjusting the fluid flow and also for introducing load disturbances into the system, the analogical PID controller, the industrial inverter, the emergency button for switching off the system and the general on-off button.

4RESULTS OBTAINED

The developed HIL platform was planned for utilization in the engineering educational area, more specifically in the industrial process control area, as a didactic tool for improving the laboratorial educational conditions. So, it is expected that the students can face important characteristics of the industrial process control area, especially some of the equipment and actuators complexities, as well as loop dynamics associated with their control variables: level, temperature, pressure and flow (Campos, 2007). To illustrate the potentiality and the flexibility of the HIL module some situations that can be worked out by the students will be discussed; our focus, when selecting these examples, was just to point out the applicability of the HIL platform for creating, in the laboratory activities, some of the several industrial process control complexities.

For the first situation it was selected the classical PID tuning process, utilized by any engineer in the industrial area: the Ziegler-Nichols – ZN ultimate gain procedure (Sung, 2009). Although very common among the industrial engineers, this solution is not recommended for tuning PID controllers in level loops, considering their dynamic characteristics and the nonlinearities associated with the storage tanks (Campos, 2007). With the ZN ultimate gain the following tuning parameters were obtained: P = 0.1; I = 7 e D = 1. 8 and utilized in the test procedures, as described in the sequence: beginning with a level value of 5%, the set point was then changed for 50% of the tank level, for a fluid flow of 2.72 m³/h. As soon as

Figure 8 – Results with the Ziegler-Nichols tuning procedure.
the process had been established, a load disturbance was then applied, changing manually the fluid flow to 1 m³ /h; after new stabilization, the set point was then changed again to 10% of the tank level. The first results (figure 8), for a ZN tuning procedure, show its inadequacy for a level-control loop: although the behavior of the controlled variable seems to be acceptable, one can see that the control variable behaves in an extremely oscillatory mode, practically with an on-off strategy, not recommended for industrial actuators like electric pneumatic valves. When dealing with industrial process control, one must care not only about the behavior of the controlled variables, but especially with the actuators dynamics and with correct strategies for extending their useful lifetime.
A new tuning procedure, recommended for level-control loops, as suggested by Friedman (Campos, 2007), was then utilized. The method suggests to deactivate the D module, resulting in the new calculated tuned parameters P = 4.7 and I = 18; the results obtained are shown in figure (9). It is possible to see that the controlled variable dynamics is now very sluggish, a feature expected for the first tuning attempt according to this tuning procedure method (Campos, 2007).

Figure 9 – Friedman tuning procedure results (first attempt).
One can see that, while the tuning procedure according ZN method has demanded 250 sec for reaching the new set point, the Friedman procedure, utilizing its first tuning parameters, has demanded around 850 sec. In such a situation, the tuning procedure suggests the utilization of the half of the integral time value, that was then changed to I = 9. The final results, with these new parameters, are displayed in figure (10).

Figure 10 – Results for the Friedman tuning (final attempt).
The results now obtained show that the dynamics of the controlled variable, although still a little “lazy”, is very close to that obtained when the ZN method was utilized. The control variable dynamics, however, has changed drastically: its behavior is now totally accepted, with a dynamics that do not wear down the actuators. It is also possible to see the occurrence of low magnitude oscillations, a very common situation for the level-control loops (Campos, 2007).
Another situation which is becoming very familiar to the industrial operators can also be illustrated and worked out in the HIL platform: the auto tuning functionality exhibited, in an increasing way, by the industrial PID controllers. This situation is depicted in figure 11. The results show the dynamics and behavior of the PID auto tuning functionality: initially, one can see the two oscillations induced by the controller in order to get the tuning parameters and, in the sequence, with the gains automatically adjusted, the behavior of the control and controlled variables, showing a dynamics close to the one which was possible to get with Friedman´s tuning procedure.
The auto tuning procedure adjusted the parameters gains for the values P = 4, I = 20 and D = 4. It is possible also to see, as in the former situation, the low magnitude oscillations, typical of the level-control loops. It is also worth to mention the nonlinear characteristics of the cylindrical tank that can be observed in the behavior of the control variables, according to the point of tuning procedure, process operation and disturbance introductions.

Figure 11 – Auto tuning of the PID controller.

5CONCLUSIONS

The didactic HIL platform, planned and developed as an educational tool for industrial control engineering education, totally FOSS-based, models and controls the dynamics of an industrial process, more specifically a level-control loop of a double tanks system. The potentiality of the HIL simulation technique as a powerful environment for engineering control education and training has been displayed in this work, although, for space reasons, only some situations could be presented and illustrated. If the HIL technique can, on the one hand, reproduce the physical real systems, encompassing their complexities and nonlinearities, without the high costs associated to the acquisition of industrial pilot plants, on the other hand allows students and operators to face the difficulties, complexities and specificities related with the understanding, utilization and tuning of industrial controllers, especially for complex industrial process dynamics.

A differential feature of this work is the fact that it is totally FOSS-based, resulting in an educational environment that, although reproducing the complex dynamics of a real industrial pilot plant, has practically no costs for its utilization. These characteristics suggest the continuity of this work, with the modeling of the remaining temperature, pressure and flow loops. In this way, it will be possible to get a HIL platform, FOSS-based, where the students can deal with the main process control variables, enlarging their knowledge and skills for actuating in the industrial process control area.

The results obtained and here discussed, although limited for space reasons, show that conditions, procedures, solutions and complexities associated with the industrial process control loops can be simulated and worked out in a HIL environment. These features include points like choosing the adequate controllers tuning for specific loops, utilizing auto tuning industrial controllers, working with industrial inverters, just to quote a few. The example utilized for the correct tuning procedure for a level-control loop can really exemplify the applicability of the HIL platform to control education.

Practically without costs, the FOSS-based HIL modules can be utilized for optimization and reinforcement of the laboratory activities in the control engineering education, since the students have the possibility to utilize HIL modules before working with a real plant. In doing so, they arrive at the laboratories with a better understanding of the processes dynamics and how to deal with them, improving and facilitating their learning. In extreme situations, where there is no real pilot plant for the students to work with, the HIL simulation techniques can be utilized as an option that, although not the best solution, can help the students to improve their knowledge acquisition and professional skills development in the industrial process control area.

These low-cost characteristics of the FOSS-based HIL modules cannot be neglected, especially in countries like Brazil, where the laboratory modules generally come from foreign countries, with high costs for their acquisition, maintenance and replacement. And even when these laboratory modules are imported, their number is generally not enough for the student’s necessities, with great drawback for their professional profile education, which lacks important concepts, knowledge and skills if they are not worked out during their university courses.

Finally, it is worth to say that this HIL platform is being already utilized for some extension courses and training procedures in the industrial process control area, inside the activities of the electrical engineering course of the Engineering College of UFJF, within the Programa de Educação Tutorial activities context.

6Acknowledgments

The authors thank the Programa de Educação Tutorial - PET/ MEC for the support in doing this work.
bibliography
Aström, K. J and T. Hägglund (2006). Advanced PID Control, ISA Instrumentation, Systems and Automation Society, Research Triangle Park, NC, USA.
Aström, K. J and T. Hägglund (1995). PID Controllers: Theory, Design and Tuning, ISA Instrumentation, Systems and Automation Society, Research Triangle Park, NC, USA.
Campos, M. C. M.M e H. C. G. Teixeira (2007). Controles Típicos de Equipamentos e Processos Industriais, Editora Edgar Blücher, São Paulo.
Coelho, A. A. F. (2007) Modelagem, Controle e Simulação Hardware In The Loop de um Míssil com Vôo Rasante à Superfície do Mar, M. Sc. Dissertation, COPPE, UFRJ.
Cooper, D. (2010) Practical Process Control: Proven Methods and Best Practices for Automatic Process Control. Available at . Accessed in 12/06/2010.
Available at . Accessed in 06/12/2010.
Available at Accessed in 06/12/2010.
Egel, T. (2009) Real Time Simulation Using Non-causal Physical Models. Paper Number: 2009-01-1021.SAE World Congress & Exhibition, April, USA.
Feisel, L. D. and A. J. Rosa (2005). The Role of the Laboratory in Undergraduate Engineering Education, Journal of Engineering Education, January, p. 121-130.
Gomes, F. J and D. P.Pereira (2008). Laboratórios Integrados para Controle de Processos e Análise da Eficiência Energética de Sistemas Industriais. Procedings of the XXXVI Congresso Brasileiro de Educação em Engenharia, São Paulo, Brazil.
Gu, F, W. S. Harrison, D. M Tilbury and C. Yuan (2007). Hardware-In-The-Loop for Manufacturing Automation Control: Corrent Status and Identified Needs.. IEEE Conference on Automation Science and Engineering, Scottsdale, p. 1105-1110, AZ, USA.
King, P. J. and D. G. Copp (2004). Hardware In The Loop For Automotive Vehicle Control Systems Development. UKACC Control 2004 Mini Symposia (2004/11105), Bath, UK, pp. 75-78.
Lamb, H (1953). Hydrodynamics, 6th ed., Cambridge Univ. Press, UK.
Normey-Rico, J. E. and E. F. Camacho (2007), Control of Dead-time Processes, Springer Verlag, London Limited, London, UK.
Nova Didacta (2009), Sistemas Didáticos de Medição, Catálogo Técnico, São Paulo, 2009.
Peterson, G. D and L. D Feisel (2002) A Colloquy on Learning Objectives For Engineering Education Laboratories, Proc. American Society for Engineering Education, Annual Conference & Exposition.
Pfister, F; C. Reitze and A. Schmidt. (2010) Hardware in The Loop technology for development and test of vehicles control systems, IPC Automotive Engineering Software and Consulting GmbH, Karlsruhe, Germany. Available at . Accessed in 06/12/2010

Sung, W. S, J. Lee and I. Lee, (2009) Process Identification and PID Control, John Wiley & Sons (Asia) Pte Ltd.,

Visioli, Antonio (2006) Practical PID Control, Springer Verlag, London Limited, London, UK.
Wade, H. L.(2004) Basic and Advanced Regulatory Control: System Design and Application. ISA Publications, Research Triangle Park, NC, USA.

Resumo: O trabalho aborda o desenvolvimento de uma plataforma laboratorial baseada nas técnicas de simulação Hardware in the Loop – HIL, projetada e construída para utilização como ferramenta didática na área de educação em engenharia de controle. O módulo permite simulação e análise das dinâmicas das malhas de controle industriais, no chão de fábrica, mas com potencialidade para operar no nível supervisório, incorporando técnicas de segurança e otimização operacionais. O módulo desenvolvido, baseado na linguagem Java e no compilador Eclipse, ferramentas Free Open Source Software – FOSS, engloba um ambiente digital tridimensional que simula a dinâmica de uma planta industrial, mas cuja operação é totalmente baseada em componentes físicos: um controlador PID externo e um inversor industrial. Constituindo um desenvolvimento baseado em FOSS, não possui qualquer restrição para sua utilização. A interface entre os componentes físicos e o ambiente tridimensional digital é baseada no módulo PIC 16F877. O módulo desenvolvido simula, parcialmente, a dinâmica de uma planta piloto real pertencente ao Laboratório de Processos Industriais da Faculdade de Engenharia da UFJF, incorporando algumas de suas complexidades e não linearidades, e operando com as mesmas constantes de tempo. A plataforma HIL desenvolvida, totalmente baseada em FOSS, permite reforçar as práticas laboratoriais para melhoria da educação em engenharia, especialmente aquelas associadas ao controle, supervisão e otimização de processos industriais.

Palavras-Chave: Hardware in the Loop – HIL, FOSS, Controle de Processos Industriais, Educação em Engenharia de Controle.

Comentários dos Revisores e Ações efetuadas

- O revisor #8 recomendou este artigo como: REQUER NOVA VERSÃO!

Comentário do Revisor:

1. 
   O artigo é de boa qualidade porém, faltou colocar o resumo e as palavras-chave.
   

Resposta: Embora o artigo já apresentasse o Resumo e as palavras-chave em inglês, como o restante do texto, foi efetuada a inserção do Resumo e das Palavras-Chave em português, ao final do artigo. Esclareço que tentei, por duas vezes, através de email para a Editoria da Revista, esclarecer a necessidade da inserção do Resumo em português, sendo o artigo em inglês. Como não obtive resposta, foi então efetuada a inserção do Resumo e Palavras-Chave ao final do texto.

Comentário do Revisor:

2. 
   A cor de fundo das figuras 8, 9, 10 e 11 deveria ser branca para tornar os resultados mais visíveis.
   

Resposta: A solicitação foi atendida. Novas figuras foram geradas de forma a atender a demanda do revisor.

- O revisor #7 recomendou este artigo como: REQUER NOVA VERSÃO!

Comentário do Revisor: O artigo apresentado é bem redigido e organizado, retratando, através da aplicação de um sistema de controle PID de nível de líquidos em tanques acoplados, o desenvolvimento de uma plataforma de simulação com arquitetura aberta e baseada em software gratuito. Alguns aspectos do desenvolvimento do modelo matemático não são muito detalhados. As figuras 8, 9, 10 e 11 apresentam palavras em português e, portanto, incompatíveis com o idioma escolhido para o texto, no caso, o inglês. Isto deve ser facilmente corrigido. A contribuição deste trabalho está na motivação para o desenvolvimento de plataformas semelhantes, como alternativa para a realização de experimentos em controle de processos. Isto se dá através da experiência relatada.  

Resposta: No tocante à questão da modelagem, já foi explicado anteriormente que o enfoque do artigo não é na modelagem dos processos, mas sim no desenvolvimento de um ambiente FOSS para utilização como ferramenta educacional para a prática laboratorial, na educação em controle de processos. Dado o espaço limitado para a apresentação do trabalho, consideramos mais importante descrever a aplicabilidade do módulo desenvolvido e suas características operacionais, pois a modelagem utilizada encontra-se na bibliografia referenciada, sendo associada aos procedimentos operacionais normais na indústria. Os textos em português foram corrigidos, sendo necessário para isto gerar novas figuras, conforme solicitado pelo revisor.

SBA Controle & Automação Vol. 00 no. 00 / Jan., Fev., Mar, Abril de 0000