Arduino-based PID control of humidity in closed space by pulse width
modulation of AC voltage
SEIICHI NAKAMORI
Professor Emeritus, Faculty of Education
Kagoshima University
1-20-6, Korimoto, Kagoshima, 890-0065
JAPAN
Abstract: My previous work explored a method for constant temperature proportional-integral-derivative (PID)
control in a closed space using an Arduino. Similarly, this paper proposes a method of PID control to keep the
humidity constant in a closed space using Arduino. PID control by a pulse width modulation (PWM) signal with
Arduino is used to keep the humidity constant due to the heat generated by an incandescent light bulb to which
an AC voltage of 100 (V) is applied. Here, the humidity is measured using the temperature/humidity sensor
DHT11. Specifically, the target value of humidity is set to 30 (%), and the experimental results of P, PD, PI, and
PID controls are compared from the points of quick response and stationary performance. As an evaluation of
stationary performance, the mean-square value of 1,000 data of the deviation (target value – measured humidity)
is calculated, and the PID control is found to be the preferred control.
Key-Words: PID control, constant humidity control, Arduino, temperature/humidity sensor, zero cross-type SSR
Received: May 15, 2021. Revised: February 17, 2022. Accepted: March 12, 2022. Published: April 21, 2022.
1 Introduction
Temperature and humidity are controlled in various
places such as egg incubators, baby incubators, rice
husk dryers, etc. In Latif et al. [1], the temperature
and humidity measured by the temperature/humidity
sensor DHT11 in the baby incubator are controlled
by the microcontroller ATmega8535 in the range of
33 (° C) to 35 (° C), humidity 40 (%) to 60 (%). A
fan is used to control the humidity, and a heater is
used to control the temperature. Matamoros et al. [2]
use proportional-integral-derivative (PID) control of
temperature and humidity in an open-atmosphere
system for bioprinting. The PID controller stabilizes
the temperature and humidity in the open-atmosphere
system at 311 (s) and 65 (s), respectively. In Nugroho
et al. [3], PID control of chamber temperature is used
to obtain test conditions for isolators that separate
conductors in power system transmission. The
temperature is controlled by an Arduino R3
microcontroller, and the PID controller drives a
thermoelectric cooled actuator (TEC12706) based on
the temperature measured by the DHT22 sensor. The
experimental results show that the chamber
temperature is maintained at the target value with
proportional coefficient , integral
coefficient and derivative coefficient
. The objective of Okpagu et al. [4] is to model,
design and develop an egg incubator system. This
system uses temperature sensor LM35 to measure the
condition of the incubator and automatically change
the condition to be suitable for the eggs. In this study,
light bulbs are used to regulate the temperature of the
eggs and water, and control fans are used to ensure
humidity and ventilation. Mathematical models of
the incubator, actuator and PID controller are
developed. The controller design based on the model
is validated by Matlab Simulink simulation, and the
Zeigler-Nichol tuning method is employed to adjust
the controller for temperature control. The
parameters of the PID controller are tuned to obtain
the desired transient response to the unit step input.
Desirable overshoot, rise time, peak time, and settling
time are obtained to improve the robustness and
stability of the system. In Rahman et al. [5], a PID
controller is suggested to control the temperature and
humidity of an egg incubator. The controller
parameters can be adjusted from a PC and an
implemented panel (keyboard). The system consists
of an Arduino Mega board, humidity sensor, heat
sensor, temperature sensor, heat source, humidity
source, servo motor, LCD module, and keyboard
with ATmega1280 microcontroller. In Veldscholte et
al. [6], a PID controller-based humidistat is proposed
using a microcontroller Arduino Uno, a small
WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS
DOI: 10.37394/23201.2022.21.6
Seiichi Nakamori
E-ISSN: 2224-266X
49
Volume 21, 2022
proportional solenoid valve, a gas flush bottle and a
humidity sensor. The settling time is about 30 (s) and
the relative humidity range of 10-95% can be
achieved. In Nakamori [3], a PID constant
temperature control is proposed to maintain a
constant temperature in a closed space of a styrofoam
box using Arduino. An incandescent light bulb with
an applied voltage of 100 (V) AC is pulse-width
modulated by Arduino, and the temperature in the
closed space is controlled by PID to be constant. An
IC temperature sensor LM35DZ is used to measure
the temperature.
Similar to Nakamori [3], in this study, we propose
a method of PID control to maintain a constant
humidity in the closed space of the styrofoam box.
PID control is applied so that the humidity due to heat
emitted from an incandescent lamp bulb to which an
AC voltage of 100 (V) is applied becomes a constant
value with a pulse width modulation (PWM) signal
from the Arduino. Here, the humidity is measured
using the temperature/humidity sensor DHT11.
Specifically, the target value of humidity is set to 30
(%), and the experimental results of P, PD, PI, and
PID controls are compared. From the experimental
results, the P control (), PD control (
, ), PD control ( , ), PID
control (, , ), and PI control
(, ) are preferred in order of quick
response. As an evaluation of the steady-state
characteristics, the average of the squares of 1,000
data of the deviation (target value - measured
humidity) is calculated. As a result, the performance
of steady-state characteristics is preferable in the
order of PID control (, , ),
PI control (, ), P control (
), PID control (, , ), PD
control (, ), and PD control (,
). Here, , and represent the PID
parameters as shown in (1).
2 Block Diagram of Feedback Control
for Constant Humidity Adjustment
Fig.1 Block diagram of the constant humidity feedback control system.
The purpose of the control in this paper is to perform
PID control so that the humidity in the closed space
is as close as possible to the target value of humidity.
The proportional, integral, and derivative coefficients
of the PID controller are adjusted appropriately based
on experiments. The overall control function is given
by
.
In constant control of temperature in the closed space,
the temperature increases when an incandescent light
bulb is heated. Usually, deviation is given by “target
value - measured temperature(t)”. In the case of
humidity control, when the incandescent light bulb is
heated, the humidity in the closed space decreases.
For this reason, the amount of – deviation(t) is input
to the control element.
Section 3 presents PID control experiments for
constant humidity.
3 PID Control Experiments for
Constant Humidity Adjustments
Final
controlling
element
(Actuator)
Controlled
object
(Process)
Detecting
element
(Transducer)
Humidity target value
Reference input
+
_
Controlled variable
Control
element
(Arduino)
Reference input
Measured humidity
Humidity target value
(Setpoint)
+
_
WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS
DOI: 10.37394/23201.2022.21.6
Seiichi Nakamori
E-ISSN: 2224-266X
50
Volume 21, 2022
Fig.2 illustrates the circuit diagram of the constant
humidity control using the temperature/humidity
sensor DHT11 with an incandescent light bulb as the
heat source. The outline of the operations is as
follows.
(1) PWM pulses are output from port 9 of the
Arduino to the Base pin of transistor NPN
2SC1815. During the PWM control signal is 5
[V], electric current flows through the
Phototriac and a trigger pulse is generated from
the Phototriac.
Fig.2 Circuit diagram for constant humidity control.
(2) The DHT11 temperature/humidity sensor
converts heat from a 40 (W) incandescent light bulb
into a DC voltage. The sensitivity of DHT11 at
25(°C) is ±5(%) (RH) for humidity and ±2(°C) for
temperature.
(3) The Phototriac and Triac function as zero-
crossing solid-state relay (SSR). Following zero-
cross switching of the SSR, an AC voltage with a
zero-crossing waveform is supplied to incandescent
light bulb.
(4) The humidity value measured by the
temperature/humidity sensor is compared with the
reference input level, and the Arduino operates the P,
PD, PI, and PID controllers so that the humidity value
approaches the reference input level. If the PWM
value is negative, it is set to 0; if the value exceeds
255, it is set to 255.
Fig.3 shows the inside of the closed space of a
styrofoam box for constant humidity control. There
are DHT11 temperature/humidity sensor and a 40(W)
incandescent light bulb. The depth, width, height, and
thickness inside the styrofoam box are 145 (mm), 350
(mm), 280 (mm), and 25 (mm). Styrofoam belongs to
the insulation category. A material whose thermal
conductivity does not exceed 0.05 (W/m-K) is called
Here, (see the block diagram of
Fig.1), : the desired humidity at discrete time
, , M(k): the measured humidity at
Fig. 3 Internal layout of a closed space in a constant
humidity control experiment. There are DHT11
temperature/humidity sensor and a 40(W)
incandescent light bulb.
a heat insulator. The constant value PID control
program for humidity with the temperature/humidity
sensor DHT11 is shown later. In this program, the
control input is of the form
(1)
discrete time . Usually, the deviation is given
by , but in this paper is set to
, as shown in the block diagram.
The justification for this can be explained as follows.
In the constant temperature control by Nakamori [3],
increasing the PWM value of the Arduino causes the
temperature of the incandescent light bulb to rise and
the temperature in the closed space to increase. In the
constant humidity control, increasing the PWM value
of the Arduino increases the temperature of the
incandescent light bulb, while decreasing the
humidity in the closed space.
In this experiment, the PID parameters , , and
are adjusted based on the deviation. The constant
value PID control program for humidity is shown in
Program listing for PID constant control for humidity.
In the case of PID control of temperature in Nakamori
[3], the more the incandescent light bulb is heat, the
higher the temperature of the closed space. In the case
of PID control of humidity, the more the
incandescent light bulb is heated, the lower the
humidity. Thus, in the humidity PID control program,
the line of italicized bold face is different from the
temperature PID control program. In the PID
WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS
DOI: 10.37394/23201.2022.21.6
Seiichi Nakamori
E-ISSN: 2224-266X
51
Volume 21, 2022
program, dt represents Its value is 1.01 (s). the
value of dt is calculated in the program by
"dt=(micros()-pretime)/1000000;".
Program listing for PID constant control for humidity.
#include <dht.h> //inclusion of header file “dht.h”
dht DHT;
#define DHT11_PIN 2
const float moku=30; //desired humidity 30 [%]
// kp=1; kp=5; kp=100
const float kp=100;
// ki=0.8; ki=5
const float ki=0.8;
// kd=1; kd=5
const float kd=5.0;
float dt;
float d;
float pretime;
float P,I,D;
float PreP = 0;
int co;
int pre;
float PID_error = 0;
long PID_p = 0;
int i;
void setup() {
Serial.begin(9600);
pinMode(9,OUTPUT);
co=10;
}
void loop() {
int chk = DHT.read11(DHT11_PIN);
analogWrite(9,co);
d=float(DHT.humidity);
dt=(micros()-pretime)/1000000;
//Serial.print("dt=");
//Serial.println(dt);
pretime=micros();
PID_error=d-moku;
I+=PID_error*dt;D=(PID_error-PreP)/dt;
PreP=PID_error;
// PID controller:
co = kp*PID_error+kd*D+ki*I;
// P controller:
//co = kp*PID_error;
// PD controller:
//co = kp*PID_error+kd*D;
// PI controller:
//co = kp*PID_error+ki*I;
// I controller:
//co = ki*I;
delay(500);
if (co <0) {
co=0;
}
else if (co >255) {
co=255;
}
else {
co=co;
}
Serial.print("PWM value=");
Serial.println(co);
Serial.print("humidity=");
Serial.println(d);
analogWrite(9,co);
delay(500);
}
Fig.4 illustrates the humidity (%) vs. discrete time ()
with the P controller for . After the
humidity repeats the values of 30 (%) and 31 (%).
Fig.5 illustrates the humidity (%) vs. with the P
controller for . After the humidity
repeats the values of 30 (%) and 31 (%). In view of
quick response, P control for is superior to
that for . Fig.6 illustrates the PWM control
signal vs. by the P controller for and
. In Fig.6, the PWM control signal takes the value
of 100 or 0 after when . For
WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS
DOI: 10.37394/23201.2022.21.6
Seiichi Nakamori
E-ISSN: 2224-266X
52
Volume 21, 2022
, the PWM control signal has a value of 9 at 0 (s)
and converges to 1 after 500 (s). Fig.7 illustrates the
humidity (%) with the PD control vs. when
and . At , the humidity is 31 (%),
followed by repeating values of 31 (%) and 30 (%).
Fig.8 illustrates the PWM control signal vs. by the
PD controller for and . At ,
the PWM value is 5 and then repeats between 5 and
0. Fig.9 illustrates the humidity (%) vs. by the PD
controller for and . At , the
humidity is 31 (%), and then the values of 31 (%) and
30 (%) are repeated. Fig.10 illustrates the PWM
control signal vs. by the PD controller for
and . At , the PWM value is 5, and
then 5, 0, and 9 repeat as the main values. Fig.11
illustrates the humidity (%) vs. by the PI controller
for , and , . For
and , the humidity values vary
widely up and down from 30 (%) and does not
converge near the target value 30 (%) even after a
period of time. For and , at
, the humidity is 30 (%) from 31 (%) one second
earlier, and then it takes values of 30 (%), 31 (%), and
29 (%) randomly. Fig.12 illustrates the PWM control
signal vs. by the PI controller for ,
and , . For and
, the PWM signal randomly takes the values 0 and
255. The maximum value of PWM for ,
is 113 at and the minimum value
is 0. The PWM value changes randomly with time.
Fig.13 illustrates the humidity (%) vs. with the PID
controller for , , and
, , . For , ,
, the humidity is 31 (%) at and changes
randomly to 31 (%), 30 (%), and 29 (%) thereafter.
For , , , the humidity swing
is almost as large as for , in the PI
control. Fig.14 illustrates the PWM control signal vs.
by the PID controller for , ,
and , , . The PWM
control signal waveform for , ,
is almost the same as that for PI control for
, . The PWM control signal
waveform for , , is similar
to that of the PI control for , . From
the above experimental results, the P control (
), PD control ( , ), PD control
(, ), PID control (, ,
), and PI control (, ) are
preferred in order of quick response.
Fig.4 Humidity (%) vs. discrete time () by P controller
for .
Fig.5 Humidity (%) vs. discrete time () by P controller
for .
Fig.6 PWM control signal vs. discrete time () by P
controller for and .
WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS
DOI: 10.37394/23201.2022.21.6
Seiichi Nakamori
E-ISSN: 2224-266X
53
Volume 21, 2022
Fig.7 Humidity (%) vs. discrete time () by PD controller
for and .
Fig.8 PWM control signal vs. discrete time () by PD
controller for and .
Fig.9 Humidity (%) vs. discrete time () by PD controller
for and .
Fig.10 PWM control signal vs. discrete time () by PD
controller for and .
Fig.11 Humidity (%) vs. discrete time () by PI controller
for , and , .
Fig.12 PWM control signal vs. discrete time () by PI
controller for , and , .
WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS
DOI: 10.37394/23201.2022.21.6
Seiichi Nakamori
E-ISSN: 2224-266X
54
Volume 21, 2022
Fig.13 Humidity (%) vs. discrete time () by PID
controller for , , and
, , .
Fig.14 PWM control signal vs. discrete time () by PID
controller for , , and ,
, .
Table 1 shows the mean-square values (MSVs) of
deviations for P, PD, PI and PID controllers. The
MSV value is calculated by
. Here,
. To investigate
the steady-state performance of the PID control, the
MSV is calculated using the humidity data from
discrete time to . From the
MSVs, the performance of steady-state
characteristics is preferable in the order of PID
control (, , ), PI control
(, ), P control (), PID
control ( , , ), PD control
(, ), and PD control (, ).
Table 1 Mean-square values of deviations for P, PD, PI
and PID controllers. Mean-square value is calculated by
,
.
Types of
PID control
PID parameters
Mean-square
values of
deviations
P control
0.2070
0.7660
PD control
,
0.3320
,
0.3180
PI control
,
0.1990
,
3.2090
PID control
,
,
0.1620
,
,
3.1530
Fig.15 Pulse width modulated waveform (upper figure),
for the PWM value 128, of the AC voltage waveform
(lower figure) [3].
Fig.15 shows the PWM waveform of the AC voltage
by the zero-crossing SSR when the PWM value is
128. Fig.15 shows the root-mean-square (RMS) and
peak-to-peak 1/10 values of the AC voltage measured
with the channel 1 and 2 probes, respectively.
4 Conclusions
This paper proposed an Arduino-based constant-
humidity PID control method with the incandescent
light bulb as a heat source in the closed space of the
styrofoam box, using the humidity measured by a
temperature/humidity sensor DHT11. In PID
constant value control of humidity, the PWM signal
output from the Arduino is adjusted to the AC voltage
applied to the incandescent light bulb through the
WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS
DOI: 10.37394/23201.2022.21.6
Seiichi Nakamori
E-ISSN: 2224-266X
55
Volume 21, 2022
NPN transistor and zero-crossing type SSR so that
the humidity approaches the target value. From the
above experimental results, P control (),
PD control (, ), PD control (,
), PID control (, , ),
and PI control (, ) are preferred in
order of quick response. From the MSV, the
performance of steady-state characteristics is
preferable in the order of PID control (,
, ), PI control (, ),
P control (), PID control (,
, ), PD control (, ), and PD
control (, ).
References:
[1] A. Latif, H. A. Widodo, R. A. Atmoko, T. N.
Phong, and E. T.Helmy, Temperature and
Humidity Controlling System for Baby
Incubator, Journal of Robotics and Control
(JRC), Vol.2, No.3, 2021, pp. 190-193. doi:
10.18196/jrc.2376.
[2] M. Matamoros et al., Temperature and
Humidity PID Controller for a Bioprinter
Atmospheric Enclosure system, Micromachines,
Vol.11, No.11, 2020, pp. 1-15. doi:
10.3390/mi11110999.
[3] S. Nakamori, Arduino-based PID Control of
Temperature in Closed Space by Pulse Width
Modulation of AC Voltage, International
Journal of Computer and Systems Engineering,
Vol.1, No.2, 2021, pp. 1–18.
[4] A. Nugroho, T. Andromeda, and M. A. Riyadi,
PID Controller Implementation for Temperature
Control in Leakage Current Test Chamber 20kv
Insulator, International Journal of Advanced
Engineering Research and Science, Vol.8, No.2,
2021, pp.106–112. doi: 10.22161/ijaers.82.14.
[5] P. E. Okpagu and A. W. Nwosu, Development
and Temperature Control of Smart Egg
Incubator System for Various Types of Egg,
European Journal of Engineering and
Technology, Vol. 4, No. 2, pp. 13–21, 2016.
[6] Z.-A. S. A. Rahman and F. S. A. Hussain, Smart
Incubator based on PID Controller,
International Research Journal of Engineering
and Technology, Vol.4, No.3, 2017, pp. 2501–
2509.
[7] L. B. Veldscholte, R. J. Horst, and S. de Beer,
Design, Construction, and Testing of an
Accurate Low-Cost Humidistat for Laboratory-
Scale Applications, Eur. Phys. J. E, Vol.44,
No.4, 2021, Article no.48, pp.1-8. doi:
10.1140/epje/s10189-021-00062-5.
WSEAS TRANSACTIONS on CIRCUITS and SYSTEMS
DOI: 10.37394/23201.2022.21.6
Seiichi Nakamori
E-ISSN: 2224-266X
56
Volume 21, 2022
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
The author contributed in the present research, at all
stages from the formulation of the problem to the
final findings and solution.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
No funding was received for conducting this study.
Conflict of Interest
The author has no conflict of interest to declare that
is relevant to the content of this article.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
