Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017) SCIENCE & TECHNOLOGY Journal homepage: http://www.pertanika.upm.edu.my/ Ziegler-Nichols First Tuning Method for Air Blower PT326 Mahanijah Md Kamal* and Muhammad Hanihazaim Abd Halim Faculty of Electrical Engineering, Universiti Teknologi MARA (UiTM), 40450 Shah Alam, Selangor, Malaysia ABSTRACT In this work, two types of controller were designed for the nonlinear air blower system PT326 used at the Instrumentation Laboratory Faculty Electrical Engineering, UiTM, Shah Alam. This work began with collection of data from the experimental work. Once the S-shape of the system response was obtained, the procedure of getting the process dead time, τ D and time constant τ C was applied to the S-shape form. By determining these two values, the optimum values of PI and PID controllers can be calculated. From the acquired data, the simulation model was developed in MATLAB/Simulink R2013a software using the transfer function obtained from the open-loop control system. The modelling system is based on the transfer function of open-loop air blower system PT326 before the design state of finding a suitable controller can be suggested. The controller design of PI and PID was obtained using the first method Ziegler-Nichols tuning rules. The result from the simulation shows that the Ziegler-Nichols first tuning rules can be applied in designing the PI and PID controller based on S-shape response obtained in open-loop test. Keywords: Air blower PT326, PI, PID, Ziegler-Nichols first method INTRODUCTION Proportional-Integral-Derivative (PID) controllers are the most adopted controllers in industry due to good cost and benefit ARTICLE INFO Article history: Received: 25 October 2016 Accepted: 17 March 2017 E-mail addresses: mahanijah@ieee.org (Mahanijah Md Kamal), hanihazaimwork92@gmail.com (Muhammad Hanihazaim Abd Halim) *Corresponding Author ratio they are able to provide (Antonio, 2004). On the other hand, PID (Ziegler & Nichols, 1942; Nims, 1950; Chien, Hrones, & Reswick, 1972) control algorithm is still continued to be widely used for most industrial control systems mainly because it is simple to maintain and tune. The PID controllers are still widely used in the process industries even though control theory has been developed significantly since they were first used decades ago (Zhuang & Atherton, 1993). In the process industries, temperature control plays an important role in order to produce ISSN: 0128-7680 2017 Universiti Putra Malaysia Press.
Mahanijah Md Kamal and Muhammad Hanihazaim Abd Halim good quality end products. The temperature control of different systems is required and the rates of reaction are controlled by heating and cooling the reactants. Lu and Tsai (2001) stated that in the plastic injection moulding process, the temperature in each temperature zone must be appropriately set and precisely controlled. According to Zhuang and Atherton (1993), and Astrom and Hagglund (1994), in 1942 Ziegler-Nichols presented a tuning formula based on time response and open-loop response rate of the system. The most frequently used experimental methods the Ziegler-Nichols open-loop and closed-loop design methods. In this work, the first method of Ziegler-Nichols was used to monitor the performance of the air blower system PT326. Open-loop process identification is the most common method used to obtain the information of the process dead time and also the process response rate. Normally, the system response is in the form of S-shape by drawing a tangent line on the response curve, where the optimum values of PID controller can be attained. Nowadays, there are a lot of publications on the tangent technique and the PID controller tuning. The analysis of open-loop response using the concept of S-shape or tangent method was reformulated by Ishak and Hussain (1998) who came out with a new algorithm that offers easier and faster calculation on the process response. Further experimental study was also conducted by Ishak and Hussain (1998) who designed a PID controller to control a flow of water to verify their proposed technique. Later, the concept of open-loop response was implemented on a PID controller tuning using Ziegler-Nichols by Kamaruddin et al. (2009) for a glycerine bleaching process, whereas Hambali et al. (2012, 2013, 2014) used the Ziegler-Nichols technique for flow, temperature and also pressure based on the tangent method for various PID tuning rules. PID controller using Ziegler-Nichols and Modified Ziegler-Nichols was designed to analyse the performance of speed control of DC motor (Meshram & Kanojiya, 2012). Mathematical model of the PT326, based on harmony search algorithm and Lyapunov using adaptive fuzzy controller, was studied by Sharma, Chatterjee, and Rakshit (2014) to identify and approximate the system behaviour to a first order with a delay. PT326 is a self-contained process and control equipment with the basic characteristics of large plant with transfer lag, system response, enable distance/velocity lag and proportional control. It models common industrial situations in which temperature control is required. The process contained in PT326 involves air that is drawn from the atmosphere by a centrifugal blower, and is heated as it passes over a heater grid before being released into the atmosphere through a duct (Alsofyani, Rahmati, & Anbaran, 2014). Figure 1 shows the methodology process of Ziegler-Nichols first tuning method proposed in this work. 260 Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017)
Ziegler-Nichols First Tuning Method for Air Blower PT326 Figure 1. The methodology process Hence, the objective of this work is to design a suitable closed-loop controller for air blower system PT326 where a model system of PT326 has been developed. The design of PI and PID controller based on Ziegler-Nichols first method was simulated in MATLAB/Simulink R2013a environment. Observation was made to monitor the response of both in order to determine the most suitable controller for air blower process PT326. METHOD In this equipment, air drawn from atmosphere by a centrifugal blower is driven past a heater grid and through a length of tubing to atmosphere again. The air steam velocity may be adjusted by means of an inlet throttle attached to the blower (Rahmat, Hoe, Usman, & Wahab, 2005). Process trainer PT326 is used as an example of industrial process because of the simplicity on temperature process operation. This is a teaching aid for control system laboratories that illustrates pure time delay added to a system with first order dynamics (Dadone & Landingham, 2001). This mimics the actual industrial process that utilises the temperature parameter as the control element, where the temperature of the air blower system is maintained at a certain level. Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017) 261
dynamics (Dadone & Landingham, 2001). This mimics the actual industrial process that utilises the temperature parameter as the control element, where the temperature of the air blower system is maintained at a certain level. Mahanijah Md Kamal and Muhammad Hanihazaim Abd Halim Figure 2. The experimental setup of PT326 Figure 2. The experimental setup of PT326 The experimental setup of Process trainer PT326, shown in Figure 2, consists of a self-contained process and control equipment whose main function is similar to a hair dryer (Ribeiro & The experimental setup of Process trainer PT326, shown in Figure 2, consists of a Cardoso, 1998), which is connected with an oscilloscope and signal generator. The process consists of heating the air flowing in the tube to the desired temperature and thetopurpose self-contained process and control equipment whose main function level is similar a hair of the control equipment is to measure the air temperature, compare it with a value set by the user and generate control signal determines the amount of electrical power dryer (Ribeiro & acardoso, 1998),which which is connected with an oscilloscope andsupplied signal to a correcting element. In this case, a heater mounted adjacent to the blower, as described in generator. The process consists of However, heating the flowing in PT326 the tube to the Feedback Instrument Limited (1996). the air Process trainer system onlydesired allows the gain to be adjusted in order to achieve the best characteristic outputs in terms of peak time, temperature and state the purpose control equipment is temperature to measureofthe settling time level and steady error. Thisofis the not sufficient to control the the air air blower system especially when it is operated under open-loop condition. Data measurements temperature, compare it with a value set by the user and generate a control signal which obtained from this experimental setup were used to design a suitable controller for temperature control of PT326. determines the amount of electrical power supplied to a correcting element. In this case, a Based on the open-loop test, a new closed-loop controller is therefore proposed using heater mounted tuning adjacent blower, as described in Feedback Instrument Limited Ziegler-Nichols rulestointhe order to improve the output characteristics. The signal generator is set with an amplitude of 2V peak to peak with the temperature set at 35 C. Initially, the open (1996). the Process PT326 system only allows theout gain to be adjusted in loop testhowever, on frequency responsetrainer with sinusoidal input signal is carried at starting frequency of 0.01 Hz to model the process in order to obtain the transfer function of the process. The order to achieve the best characteristic outputs in terms of peak time, settling time and tuning of both rules was carried out to obtain the proportional constant, Kp, integral constant KI, andstate derivative constant, derived fromofτdthe andairτcblower obtained from D. These constants steady error. This is notksufficient to controlwere the temperature system the output response of an open loop test with step input using Ziegler-Nichols first method where S-shape response was obtained. The constant values were used to simulate the PI and PID controller and the output characteristics were observed. 262 Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017)
RESULTS AND DISCUSSION Ziegler-Nichols First Tuning Method for Air Blower PT326 Several methods are available for designing PID controller. The most frequently used experimental methods are the Ziegler-Nichols open-loop and closed loop design methods. In this work, however, the work focuses on Ziegler-Nichols open-loop method where the system response to a step input is in S-shape form. The experiment was conducted at the Instrumentation Laboratory Faculty of Electrical Engineering, Universiti Teknologi MARA. Figure 3 shows the response curve of process trainer PT326 of air blower system in S-shape where the design of PI and PID controller can be developed. Figure 3. The S-shape response using Ziegler-Nichols first method The design procedures are as follows: a) Obtain an open-loop step response from the experiment. b) Draw the tangent line on the response curve. c) Using the cross-points of the tangent with x-axis and with the steady-state output line to determine time delay, τ D and time constant, τ C. d) PID controller tuning design can be used using the formulae stated in Table 1. Table 1 Ziegler-Nichols first method open-loop tuning parameters Controllers Kc τ i τ D P 0 PI 0 PID Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017) 263
Mahanijah Md Kamal and Muhammad Hanihazaim Abd Halim The Ziegler-Nichols first method tuning formula is based on the value of gain Kc, τ i and τ D to determine the output characteristics. By doing this, the Process trainer PT326 output characteristics can be improved by applying different types of controller. To illustrate the accuracy of the Ziegler-Nichols tuning formula based on Table 1, the output responses can be improved and analysed. The transfer function of the process is given as in Equation [1]. [1] where Kp is steady state gain of the system τ is time constant of the system t d is dead time of the system Table 2 shows the value of P controller, PI controller and PID controller design using the Ziegler- Nichols first method after obtaining the value of τ D and τ C. The air blower system PT326 only consists of P controller. Therefore, to analyse the parameters in Table 2, MATLAB/Simulink R2013a is used to model the air blower system PT326. Once the S-shape response has been obtained, the nest step is to draw the tangent line on the response curve. Here, the value of time delay, τ D and time constant, τ C can be determined. The tangent line at the inflection point of S-shape curve determining the intersections of the tangent line with the time axis and the voltage axis. From the graph shown in Figure 3, the values of τ D = 0.27s and τ C = 0.55s were obtained. Once the values of τ D = 0.27s and τ C = 0.55s are known, the next step is to design the suitable controller for the air blower system PT326 using the parameters indicated in Table 1. Table 2 shows the value of P controller, PI controller and PID controller design using the Ziegler-Nichols first method. The air blower system PT326 only consists of P controller. Therefore, to analyse the parameters in Table 2, MATLAB/Simulink R2013a is used to model the air blower system PT326. Table 2 Ziegler-Nichols first method open-loop tuning parameters Controllers Kc τ i τ D P 2.04 0 PI 1.833 0.9 0 PID 2.44 0.54 0.135 Based on the experimental results, the transfer function of air blower system PT326 was obtained. The open-loop transfer function is given in Equation [2]. [2] 264 Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017)
Ziegler-Nichols First Tuning Method for Air Blower PT326 The calculations obtained in Table 2 are used as the parameters setting in the simulation model connected with the transfer function of Process trainer PT326 as in Equation [2]. Figure 4 shows the MATLAB/Simulink model constructed where the step response is injected as the input signal. Figure Figure 4. The 4. model The model of PT326 of PT326 in the in Simulink the Simulink model model Figure 5 shows the simulation result of three controllers: P controller, PI controller and PID controller. From the observation, with P controller, the output cannot achieve the steadystate response, whereas PI controller has the ability to improve the steady state error, and the combination of PID will improve the transient response in terms of settling time, t s, peak time, t p and steady state error. Based on the output response, the rise time, tr of PID controller is faster than PI controller. However, PID controller has a higher overshoot compared with PI controller. Both controllers achieved steady state error, which indicates that the final response of Process trainer PT326 will be at a constant value at a certain length of time. 1.6 The output response of P, PI and PID using Ziegler-Nichols first method 1.4 1.2 1 Amplitude 0.8 0.6 0.4 0.2 Step input P OLT PI OLT 0 0 10 20 30 40 50 60 70 80 90 time,sec Figure 5. The controller response based on Ziegler-Nichols first method PID OLT Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017) 265
Mahanijah Md Kamal and Muhammad Hanihazaim Abd Halim By applying the transfer function obtained from Equation [2], the response of PI and PID controller in comparison with P controller shows that both the controllers can achieve steadystate condition. The difference between these two controllers is on the time taken to become stable. Figure 6 shows the results of output characteristics based the values obtained in Table 2, where the excitation input is a step response. Even though the output response oscillates more, the response will still achieve the steady-state condition. However, the time taken is longer from the original response, as shown in Figure 5. 1.8 1.6 1.4 1.2 Amplitude 1 0.8 0.6 0.4 0.2 Step input PI OLT PID OLT 0 0 10 20 30 40 50 60 70 80 90 time, sec Figure 6. The effect of integrator in PID controller In order to analyse the effect of integrator towards the process response of air blower PT326, the value of integral in the PID controller has been changed from τ I = 0.54 to τ I = 0.48. As shown in Figure 7, the response of PID controller is found to be almost similar with PI controller output. If the value of integrator in PID controller is higher than the value from the open-loop test, the overshoot ban be reduced and it will achieve a steady state error faster compared with the PID controller response in Figure 5 and Figure 6. 1.5 1 Amplitude 0.5 Step input PI OLT 0 0 10 20 30 40 50 60 70 80 90 time, sec Figure 7. The effect of integrator in PID controller PID 0.84 266 Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017)
CONCLUSION Ziegler-Nichols First Tuning Method for Air Blower PT326 This work focuses on designing a suitable controller for air blower system PT326 in order to achieve the steady-state response. The optimum values of PID controller can be easily accessed and gained from the open-loop response. By doing the open-loop test, the design of PID controller becomes much faster. From the simulation aspect, analysis in terms of rise time, overshoot, and delay time and steady-state condition is quicker and time consuming compared with the gain adjustment. The performance of air blower system Process trainer PT326 can be further improved instead of using P controller only. The responses of both the controllers of air blower system PT326 are better in terms of response time and settling time when the Ziegler-Nichols first method is applied. It was observed that the response of PID controller depends on the type of process and requirement settings by user. However, PID controller is the most common controller used in a slow process such as temperature control. Therefore, PID controller is more suitable for air blower system Process trainer PT326. ACKNOWLEDGEMENT The authors would like to acknowledge the Faculty of Electrical Engineering, Universiti Teknologi MARA for providing the financial support. REFERENCES Alsofyani, I. M., Rahmati, M. F., & Anbaran, S. A. (2014). A PID Controller Design for an Air Blower System. IRICT 2014 Proceedings, pp. 48-59. Antonio, V. (2004). A new design for a PID plus feedforward controller. Journal of Process Control, 14, 457-463. Astrom, K., & Hagglund, T. (1994). PID controller: Theory, Design and Tuning, Library of Congress Cataloging-in-Publication Data. Chien, K. L., Hrones, J. A., & Reswick, J. B. (1972). On the automatic control of generalized passive systems. Trans ASME, 74, 175-185. Dadone, P., & Landingham, H. V. (2001). Remote Control of Industrial Processes, Mountain Workshop on Soft Computing in Industrial Applications Virginia Tech., Blacksburg. Virginia, pp. 93-97. Feedback Instrument Limited. (1996). Process Trainer PT326, User Manual, Crowborough, 1-9. Hambali, N., Ang, A. A. R., Ishak, A. A. & Janin, Z. (2013). Various PID controller tuning for Air Temperature Oven System, International Conference on Smart Instrumentation, Measurement and Applications. KL. Hambali, N., Masngut, A., Ishak, A. A., & Janin, Z. (2014). Process Controllability for Flow Control System using Ziegler-Nichols (ZN), Cohen-Coon (CC) and Chien-Hrones-Reswick (CHR) Tuning Methods. International Conference on Smart Instrumentation, Measurement and Applications. KL. Hambali, N., Zaki, M. N. K., & Ishak, A. A. (2012). Reformulated Tangent Method of Various PID Controller Tuning for Air Pressure Control, International Conference on Control System, Computing and Engineering, pp. 17-21. Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017) 267
Mahanijah Md Kamal and Muhammad Hanihazaim Abd Halim Ishak, A. A., & Hussain, M. A. (1998). Open-loop Process Identification: Reformulation of Response Rate Calculation. Regional Symposium on Chemical Engineering (pp. 1-5). Ishak, A. A., & Hussain, M. A. (1998). Reformulation of the Tangent Method for PID Controller Tuning. TENCON 2000, pp. 484-488. Kamaruddin, N., Janin, Z., & Taib, M. N. (2009). PID controller tuning for glycerine bleaching process using well-known tuning formulas - a simulation study. In Industrial Electronics, pp. 1682-1686. Lu, C. H., & Tsai, C. C. (2001). Adaptive decoupling predictive temperature control for an extrusion barrel in a plastic injection molding process. IEEE Transactions on Industrial Electronics, 48(5), 968-975. Meshram, P. M., & Kanojiya, R. G. (2012). Tuning of PID Controller using Ziegler-Nichols Method for Speed Control of DC Motor, IEEE International Conference on Advances in Engineering, Science and Management, pp. 117-122. Nims, P. T. (1950). Some Design Criteria for Automatics Controls. Trans AIEE 70, 759-768. Rahmat, M. F., Hoe, Y. K., Usman, S., & Wahab, N. A. (2005). Modelling of PT326 Hot Air Blower Trainer Kit using PRBS signal and Cross-correlation Technique. Jurnal Teknologi, 42, 9-22. Ribeiro, B., & Cardoso, A. (1998). A Model-based Neural Network controller for a Process Trainer Laboratory Equipment. Artificial Neural Nets and GeneticAlgorithms, Springer, 601-605. Sharma, K. D., Chatterjee, A., & Rakshit, A. (2014). Harmony search algorithm and Lyapunov theory based hybrid adaptive fuzzy controller for temperature control of air heater system with transportdelay. Applied Soft Computing, 25, 40-50. Zhuang, M., & Atherton, D. P. (1993). Automatic tuning of optimum PID controllers, pp. 216-224. Ziegler, J. G., & Nichols, N. B. (1942). Optimum settings for automatic controllers. Trans ASME, 64(8), 759-768. 268 Pertanika J. Sci. & Technol. 25 (S): 259-268 (2017)