Design of Automatic Dough Feeder Control System on Modified

Cassava Flour-based Noodle Extrusion using Fuzzy Logic Controller

Eko Kuncoro Pramono and Umi Hanifah

Center for Appropriate Technology, Indonesian Institute of Sciences, Jl. K.S. Tubun No 5, Subang, Indonesia

Keywords: Extruder, Fuzzy Logic Controller, Noodle Extrusion, Modified Cassava Flour.

Abstract: Production of modified cassava flour-based noodle has been done by many researchers using the extrusion

process. An extruder, an extrusion-based machine had been developed by Research Center for Appropriate

Technology to produce modified cassava flour-based noodle. The unstable feeding rate of the dough caused

the motor had fluctuation load during driving the main screw of the extruder. The main impacts were the

throughput of the noodle had become unstable too, the quality of the noodle also have an effect regarding the

density of the noodle, and the worst case was the higher feeding rate of the dough means heavier load for the

motor and can lead to jam of the extrusion process and damage of the main motor. In this paper, an automatic

dough feeder has been designed to control the feeding rate of the dough to the extruder using Fuzzy Logic

Controller (FLC). The inputs of the FLC were error and delta error of the main motor current and the output

of the controller was the rotational speed of the dough feeder motor. The design and simulation were done

using Matlab toolbox.

1 INTRODUCTION

Modified cassava flour (mocaf) is a derivative

product of cassava flour which uses the principle of

cassava cell modification by fermentation for 12 - 72

hours. This flour has the advantage of having a higher

protein content and better physicochemical properties

compared to cassava flour. Mocaf product

development has been carried out in terms of

improving the quality of mocaf (Kardhinata et al.,

2019; Kurniati & Aida, 2012; Nusa, Suarti & Alfiah,

2015) and increasing production (Hamidi &

Banowati 2019). One of the uses of mocaf is to

substitute wheat flour, which is not produced in

Indonesia, in making noodles. Some research on the

use of mocaf as a noodle-making material has been

done by many researchers (Afifah & Ratnawati,

2017; Indrianti et al., 2014; Yulianti, Sholichah &

Indrianti, 2019).

In line with the research on these mocaf flour-

based noodle products, the development of extruder

equipment as equipment to produce mocaf has been

widely developed. One of them, in 2013 the Research

Center for Appropriate Technology began developing

a single screw-type extruder machine (Siregar et al.

2013). The main aspect of a single screw extruder was

the screw pump or screw press, where the food dough

is compressed to form a semi-solid mass in a

cylindrical chamber (barrel) using a single screw that

is driven by an electrical motor and forced out

through a limited opening (die) at the tip of the barrel

(Riaz, 2000). When the food is heated, it is called

extrusion cooking or hot extrusion. The heat used in

the cooking process can come from the steam

injection (directly), from the heating jacket

(indirectly), and can also be done by heating the

dough before being put into the extruder, as well as

the heat energy arisen from the friction of the dough

during the extrusion process (Riaz, 2013).

On the process of making noodles, the dough, a

mixture of mocaf and other ingredients went through

the cooking process before being fed into the extruder

machine. The heating process was a critical step for

pre-gelatinization to maintain the noodle strands

because mocaf is a gluten-free flour which not able to

form a cohesive dough structure5. As a result, the

dough material from the extruder was a half-cooked

dough that had an irregular shape and size. The

feeding process of the dough into the extruder

machine was done manually by the operator so that

the feeding rate the dough becomes unstable. This can

give result in variations in physical characteristics of

the noodles produced.

312

Pramono, E. and Hanifah, U.

Design of Automatic Dough Feeder Control System on Modiﬁed Cassava Flour-based Noodle Extrusion using Fuzzy Logic Controller.

DOI: 10.5220/0009981900002964

In Proceedings of the 16th ASEAN Food Conference (16th AFC 2019) - Outlook and Opportunities of Food Technology and Culinary for Tourism Industry, pages 312-316

ISBN: 978-989-758-467-1

The use of fuzzy logic as a controller had been

done by many researchers. It was said that fuzzy logic

has been used in many industrial control applications.

It had several advantages including simple and low

cost for installation, capability to handle non-linear

systems (Abeykoon et al., 2011), free plant models

(Albertos & Sala, 2004; Driankov, Hellendoorn &

Reinfrank, 1993; Fileti et al., 2007). Some studies

related to the use of a fuzzy logic controller (FLC)

include FLC in the washing machine to control the

voltage for the inverter claimed can reduce the time,

electricity consumption, and washing water

(Wulandari & Abdullah, 2018). An FLC also was

used to control the expansion volume of dough pieces

similar to its standard conditions during the proofing

process claimed can provide significantly better

result, not require a mathematical model and had

better to reject any than PID controller (Yousefi-

Darani, Paquet-Durand & Hitzmann, 2019). PID-

fuzzy algorithms were applied to a polymerization

process control and compared to conventional PID

controllers, proved to be more suitable and reliable

(Fileti et al., 2007).

Regarding the many advantages of the fuzzy logic

controller mentioned above, in this study, an FLC was

designed to control the speed of feeding noodle-based

mocaf doughs that have non-uniform shapes so that it

is expected to provide motor workload to the extruder

constantly to produce homogeneous noodle products.

2 MATERIALS AND METHODS

The extruder that was used in this research had a

specification which showed in figure 1.

Figure 1: The extruder used in this research.

The Extruder was developed by the Research

Center for Appropriate Technology on 2018, the scale

down from the corn base noodle extruder Prototype 1

(Putra, Novrinaldi & Kurniawan, 2013; Siregar et al.,

2013). The main motor was a three phases induction

motor which had a nominal power of 5 HPs. The

motor was connected to a 40:1 reducing gearbox

system which operated at 1500 rotations per minute.

The motor was electrically driven by a Variable

Speed Drive (VSD) inverter. The extrusion process

was set at 75 °C which measured by a thermocouple

sensor located on the barrel. Both a heater and a

blower were placed to control the temperature to its

set point.

A preliminary experiment was done to investigate

the optimum current of the motor which referred to

the optimum load of the motor. There were 3

measurements using 3 same dough set of 2 kgs. The

productions of noddle were conducted according to

the proper process and the feeding of the dough to the

extruder was done manually by the operator. During

the extrusion process, the current measurements of

the extruder motor were done using a clamp ampere

and were recorded every 30 seconds. This experiment

gave an optimum value of motor current which set as

a setpoint to the proposed system.

The automatic dough feeder system was designed

to eliminate the inconsistency feeding rate of dough

which done manually by the operator. The system

was consisted of dough crusher and a motor drive

screw to feed the dough to the extruder automatically.

The proposed dough feeder system is shown in figure

Figure 2: Illustrated design of dough feeder system.

The rate of dough feeding was done by controlling

the angular speed of the feeder motor. The angular

speed of the motor was controlled based on the

feedback of the main motor load which was

represented by the current measurement. The control

system of the proposed feeding system was Fuzzy

Logic Controller (FLC). Figure 3 shows the block

diagram of FLC. The design and simulation of the

FLC system was done using Matlab R2016a software

(Trial Version).

Figure 3: Block diagram of FLC.

Design of Automatic Dough Feeder Control System on Modiﬁed Cassava Flour-based Noodle Extrusion using Fuzzy Logic Controller

313

Inputs of FLC were error and delta_error which can be

represented as follow,

    





(1)

d      1

(2)

where setpoint (SP) is the average current of the main

extruder motor. The steps of designing FLC as

follow:

2.1 Fuzzification

In the fuzzification process, each input value

consisting of crisp numbers is mapped in a fuzzy set

by determining the degree of membership in each

FLC input as described in the curve graph as follows

Figure 4: Membership function of error (top) and

delta_error (bottom).

2.2 Determining Fuzzy Rules and

Inference System

Table 1: Fuzzy rules pair set.

Remark:

NB = Negative Big

NS = Negative Small

Z = Zero

PS = Positive Small

PB = Positive Big

VS = Very Slow

S = Slow

N = Normal

F = Fast

VF = Very Fast

Each pair of the fuzzy set uses a logical AND

operator, so a set of fuzzy rules is obtained as shown

in Table 1, where the implication process uses Min

rules and the aggregation process uses the max

method, commonly known as the Min-Max method

(Mamdani Method). There were totally 25 rules made

of error and delta_error membership function pairs.

2.3 Defuzzification

The defuzzification process used in the Mamdani

method is the center of gravity method, where the

output value of crisp is calculated based on the center

of gravity of the aggregation of the fuzzy sets of each

output produced, as follows:



∗





.

























(3)

The membership function of output is described in

Figure 5 as follows,

Figure 5: Membership function of output.

3 RESULT AND DISCUSSION

Based on the result of the preliminary experiment of

the main motor current measurement during the

mocaf-based noddle production process, the

measurement data can be showed in Figure 6. It can

be showed that the average motor current is about 9

Ampere, that becomes the setpoint. In measurements

1 and 2, there is a time span in which the motor got

shut down due to overload. Recorded motor currents

before shutting down were at 13 ampere and 15

amperes. Therefore, the proposed system should have

a range of the motor current not exceed 13 Ampere.

16th AFC 2019 - ASEAN Food Conference

314

Figure 6: Measurement of motor current during noodle

production.

Applying set point and all fuzzy parameter

described above into Matlab fuzzy toolbox would

give simulation the calculation of the output based on

given inputs as shown in Figure 7.

Figure 7: Simulation of the proposed fuzzy at error -1.3 A

and delta_error -2.4 A.

Each crisp value of inputs will be mapped into its

membership function. Each pair of input membership

function will be mapped into the output membership

function based on the fuzzy rules. And all output

membership function will be aggregated and

calculated into crisp value of output based on formula

(3). On the example above, error -1.3 ampere and

delta_error -2.4 amperes will give output rotation

speed of the feeder’s screw at 7.63 Rpm. Figure 8

shows a surface response relationship between inputs

and output. During the positive value of error and

delta_error the controller gives a higher value of the

output, which means the load of the main motor was

low and need to be fed dough faster. On the other

hand, whenever error and delta error had negative

values the output gave less value, which means the

load of the main motor was high and need to be fed

by dough slower. The FLC maintained the rate of

dough feeding to its proper value based on the set

point.

Figure 8: Surface response relationship between inputs and

output.

4 CONCLUSIONS

A FLC was applied to control the feeding rate of

dough on an extruder during noodle production was

proposed and simulated under Fuzzy toolbox on

Matlab. The controller determines the rotational

speed of the screw-feeder motor based on the main

motor current on extruder which describes the load of

the motor. It is shown that the FLC can maintain the

feeding rate of dough on the extruder by manipulating

the screw speed of the feeder system. The controller

performances can be further improved by improving

the model’s accuracies, adding more fuzzy rules, etc.

The implementation of the proposed controller on the

dough feeder of the extruder together with different

fuzzy parameter setting and method will be addressed

under future work.

ACKNOWLEDGEMENTS

We would like to gratefully acknowledge Ministry of

Research, Technology and Higher Education through

Insentif Riset Sistem Inovasi Nasional (INSINAS)

Riset Pratama for funding this research.

REFERENCES

Abeykoon, C., Li, K., McAfee, M., Martin, P.J. & Irwin,

G.W. 2011, ‘Extruder Melt Temperature Control With

Fuzzy Logic’, IFAC Proceedings Volumes, vol. 44, no.

1, pp. 8577–82, viewed 26 August 2019,

<https://www.sciencedirect.com/science/article/pii/S1

474667016449888>.

Design of Automatic Dough Feeder Control System on Modiﬁed Cassava Flour-based Noodle Extrusion using Fuzzy Logic Controller

315

Afifah, N. & Ratnawati, L. 2017, ‘Quality assessment of

dry noodles made from blend of mocaf flour, rice flour

and corn flour’, IOP Conference Series: Earth and

Environmental Science, vol. 101, no. 1.

Albertos, P. & Sala, A. 2004, ‘Perspectives of fuzzy

control: lights and shadows’, 2004 2nd International

IEEE Conference on ‘Intelligent Systems’. Proceedings

(IEEE Cat. No.04EX791), IEEE, pp. 25–32, viewed 24

August 2019, <http://ieeexplore.ieee.org/document/

1344631/>.

Driankov, D., Hellendoorn, H. & Reinfrank, M. 1993, An

Introduction to Fuzzy Control, 1st edn, Springer Berlin

Heidelberg, Berlin, Heidelberg, viewed 24 August

2019, <http://link.springer.com/10.1007/978-3-662-

11131-4>.

Fileti, A.M.F., Antunes, A.J.B., Silva, F.V., Silveira, V. &

Pereira, J.A.F.R. 2007, ‘Experimental investigations on

fuzzy logic for process control’, Control Engineering

Practice, vol. 15, no. 9, pp. 1149–60, viewed 15 August

2019, <https://linkinghub.elsevier.com/retrieve/pii/

S0967066107000317>.

Hamidi, M.A. & Banowati, E. 2019, ‘Utilization of mocaf

flour (modified cassava flour) for revitalization of the

use tapioca flour in communities for empowering hulu-

hilir human resources in wonogiri regency’, IOP

Conference Series: Earth and Environmental Science,

vol. 243, no. 1, p. 012081, viewed 24 August 2019,

<http://stacks.iop.org/1755-1315/243/i=1/a=012081?

key=crossref.b837103ec4ff32d5d559b51fe67803a7>.

Indrianti, N., Kumalasari, R., Ekafitri, R. & Darmajana,

D.A. 2014, ‘Pengaruh Penggunaan Pati Ganyong,

Tapioka, dan Mocaf Sebagai Bahan Substitusi

Terhadap Sifat Fisik Mie Jagung Instan’, agriTECH,

vol. 33, no. 4, pp. 391–8, viewed 24 August 2019,

<https://jurnal.ugm.ac.id/agritech/article/view/9534>.

Kardhinata, E.H., Purba, E., Suryanto, D. & Rusmarilin, H.

2019, ‘Modified cassava flour (Mocaf) content of

cassava ( Manihot esculenta Crantz) in North

Sumatera’, IOP Conference Series: Earth and

Environmental Science, vol. 260, no. 1, p. 012088,

viewed 24 August 2019, <https://iopscience.iop.org/

article/10.1088/1755-1315/260/1/012088>.

Kurniati, L.I. & Aida, N. 2012, ‘Pembuatan Mocaf

(Modified Cassava Flour) Dengan Proses Fermentasi

Menggunakan Lactobacillus Plantarum,

Saccharomyces Cerevisea, Dan Rhizopus Oryzae’,

Paper and Presentation of Chemical Engineering, RSK

547.29 Aid p, 2013, vol. 1, no. 1, pp. 1–6, viewed 24

August 2019, <http://digilib.its.ac.id/ITS-paper-

230211300026795/25961>.

Nusa, M.I., Suarti, B. & Alfiah 2015, ‘Pembuatan Tepung

Mocaf Melalui Penambahan Starter Dan Lama

Fermentasi (Modified Cassava Flour)’, Agrium: Jurnal

Ilmu Pertanian, vol. 17, no. 3, pp. 210–7, viewed 24

August 2019, <http://jurnal.umsu.ac.id/index.php/

agrium/article/view/322>.

Putra, S.A., Novrinaldi & Kurniawan, Y.R. 2013,

‘Pengaruh Variasi Kecepatan Putar Ulir Mesin

Ekstruder Ulir Tunggal Terhadap Performa Mesin dan

Kualitas Mi Jagung Terhadap Performa Mesin dan

Kualitas Mi Jagung’, Proceeding Seminar Ilmu

Pengetahuan Teknik 2013 “Teknologi Untuk

Mendukung Pembangunan Nasional", pp. 157–61.

Riaz, M.N. 2000, Extruders in food applications, M.N. Riaz

(ed.), CRC Press Taylor & Francis Group, Boca Raton,

Florida.

Riaz, M.N. 2013, ‘Food Extruders’, Handbook of

Farm, Dairy and Food Machinery Engineering, vol. 2,

pp. 427–40, viewed 24 August 2019,

<https://www.sciencedirect.com/science/article/pii/B9

780123858818000161>.

Siregar, H.P., Putra, S.A., Taufan, A. & Kurniawan, Y.R.

2013, ‘Studi Eksperimental Prototip I Mesin Ekstruder

Mie Jagung’, Mekanika, vol. 12, no. September, pp. 39–

43.

Wulandari, N. & Abdullah, A.G. 2018, ‘Design and

Simulation of Washing Machine using Fuzzy Logic

Controller (FLC)’, IOP Conference Series: Materials

Science and Engineering, vol. 384, no. 1, p. 012044,

viewed 24 August 2019, <http://stacks.iop.org/1757-

899X/384/i=1/a=012044?key=crossref.160a9117ca4d

d9724848e66738a0db04>.

Yousefi-Darani, A., Paquet-Durand, O. & Hitzmann, B.

2019, ‘Application of fuzzy logic control for the dough

proofing process’, Food and Bioproducts Processing,

vol. 115, pp. 36–46, viewed 15 August 2019,

<https://linkinghub.elsevier.com/retrieve/pii/S0960308

518305236>.

Yulianti, L.E., Sholichah, E. & Indrianti, N. 2019,

‘Addition of Tempeh Flour as a Protein Source in

Mixed Flour (Mocaf, Rice, and Corn) for Pasta

Product’, IOP Conference Series: Earth and

Environmental Science, vol. 251, no. 1, p. 012037,

viewed 24 August 2019, <http://stacks.iop.org/1755-

1315/251/i=1/a=012037?key=crossref.8c79f60c191b1

36b378f3cf2a1ff7b85>.

16th AFC 2019 - ASEAN Food Conference

316