Modeling of the Thermoelectric Modules Driven by Solar Photovoltaic

ed
by using MATLAB/SIMULINK

Hassan M. M. Mustafa1*, Amal E. M. Elnaggar1, Soliman Sharaf 2, Zeinab S. Abedel Rehim1,

M. A. El-Bayoumi1 and Helmy M. El Zoghby2

iew
1 Mechanical Engineering Department, Engineering and Renewable Energy Research Institute, National
Research Centre, Giza, Egypt
2 Electrical Power and Machines Engineering Department, Faculty of Engineering, Helwan University.
* Email: hassanabuhashish@gmail.com

Abstract

v
The increasing need for sustainable and energy-efficient technologies has spurred

re
extensive research into renewable energy applications. Among the promising options,
photovoltaic (PV) cells stand out for their ability to convert solar energy into electrical power,
which can be utilized in many systems. One exciting application of PV power is in
thermoelectric cooling, specifically through Peltier modules. Peltier coolers offer a solid-state,
er
compact, and environmentally friendly alternative to conventional cooling methods. Utilizing
the Seebeck effect, these modules can efficiently transfer heat from one side of the device to
pe
another, making them suitable for off-grid or energy-limited environments. This work aims to
analyze, model, and simulate a Peltier element cooling system that uses photovoltaic (PV)
technology under different operating conditions. The Peltier element is represented by its data
sheet the adaptive neuro-fuzzy inference system (ANFIS) is the basis for this which can predict
ot

the temperature differences at different voltages of the PV system under rapidly changing
environmental conditions. Two types of controllers {proportional-integral derivative (PID) and
tn

fuzzy-proportional-integral derivative (FPID)} are adopted in this study. This study

demonstrates the control system's efficiency in improving refrigeration response by decreasing
overshoot, rise time, settling time, and steady-state error, supported by experimental validation
rin

under various conditions. The proposed practical system is implemented using the MATLAB
program.
Keywords: Thermoelectric Modules, Controller, MATLAB Simulink, FPID, PID, Solar
Photovoltaic.
ep

1 Introduction

Conventional refrigeration and cooling systems predominantly depend on electricity and

Pr

refrigerants, which significantly contribute to environmental challenges. The reliance on

refrigerants such as chlorofluorocarbons (CFCs) and hydro fluorocarbons (HFCs) leads to

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ozone layer depletion and the intensification of global warming due to their high global
warming potential (GWP). Additionally, the heavy electricity consumption of these systems

ed
exacerbates greenhouse gas emissions, further impacting the environment, [1].

Refrigeration alone accounts for approximately 15% of household energy usage, making it one

iew
of the largest contributors to domestic energy consumption. This demand increases
substantially with industrialization, as cooling requirements in commercial, industrial, and
residential sectors grow. The cumulative effect is a considerable rise in energy consumption
and a parallel increase in the environmental footprint,[2].

v
In [3] developed a refrigeration system utilizing a thermo-electric Peltier module enabling the

re
user to effectively maintain and regulate the cooling system's temperature.

In [4]an automated refrigerator powered by solar energy, highlighting the global adoption of
er
solar power for electricity generation. Instead of a conventional condenser-based refrigerator,
a Peltier module is employed due to its advantages, such as the absence of CFC emissions,
portability, and ease of transport. Additionally, smart technologies are integrated into the
pe
refrigerator's design to enhance its functionality. The compact size of the Peltier module allows
for convenient portability, making it suitable for use in various locations,

In [5] utilizing a cooler for dual purposes: air cooling and refrigeration. The objective is to
ot

propose a modified design where a refrigeration box, constructed from mild steel, is integrated
within the cooler tank. To minimize heat loss, the cooler tank is insulated externally with
tn

rubber pads. All surfaces of the refrigeration box are in contact with the tank water, except for
the front face, which serves as the door of the refrigeration box. This design aims to enhance
functionality and improve energy efficiency.
rin

In [6] a novel dynamic PV model leveraging artificial intelligence (AI) is introduced, making use
of an ANFIS (adaptive neuro-fuzzy inference system). ANFIS combines the strengths of neural
networks and fuzzy systems, benefiting from the capabilities of both approaches. The design
ep

process is thoroughly detailed, including extensive testing of various membership functions,

training iterations, and membership function quantities through MATLAB simulations to meet the
AI requirements of the ANFIS model. In [7] by contrasting the correctness of the produced model
with classical dynamic models that have been previously proposed in the literature, the model's
Pr

performance is assessed. Innovative DVR and DSTATCOM based on MLI with ANFIS control
for improved power quality.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 In [8] Predictive energy control with gap -ANFIS for grid-connected industrial PV-battery
systems, in [9] Distributed energy systems powered by solar and connected to the grid using hybrid

ed
ANFIS based on Aquila arithmetic optimization to reduce harmonics. In [10], the integration of
renewable energy sources was optimized through the use of a seven-level converter overseen by
ANFIS-CS-GWO.

iew
|In [11] carried out experimental and numerical studies on a tiny, mobile solar-powered absorption
chiller. The chiller's coefficient of performance (COP) was determined to be 0.31 through
simulation using Simulink, a modeling tool available in MATLAB. However, both studies

v
encountered issues with sudden temperature peaks within the collectors, impacting the COP. In
[12], a new cooling system that combines the Peltier effect with solar-concentrating photovoltaic

re
(PV) systems was introduced. This analysis was carried out using a simulation model developed
in MATLAB, allowing for a comprehensive evaluation of the cooling system's performance and
its impact on the overall efficiency of the solar PV setup.
er
In [13] the optimization of thermoelectric modules (TEM) for a novel active
photovoltaic (PV) cooling method they developed a mathematical model for TEM considering
pe
temperature-dependent material properties. In [14], A Peltier module with a heat sink and a
solar dish were used to concentrate solar heat energy and produce electricity. The study
involved experimental testing as well as simulations using MATLAB (V7.6.0). This research
aims to study, model, and simulate a cooling system (Peltier) powered by a photovoltaic (PV)
ot

system. This work proposes an innovative artificial intelligence method for (Peltier) based on
the Adaptive Neuro-Fuzzy Inference System (ANFIS), which can estimate the temperature
tn

differences at different using the datasheet for the Peltier module. Utilizing the MATLAB
program, the practical suggested system was implemented.
2 Materials and Method
rin

2.1 Refrigerator box

A representation of the portable refrigerator can be found in Figure 1. It is now feasible
to incorporate PV technology to offset refrigerator energy usage thanks to advancements in
photovoltaic systems. Even with 1000 W/m² of solar radiation on a bright day, a solar panel
ep

can produce a comparatively small amount of electricity. For example, in clear sky
circumstances at noon, a 4 m² panel with 25% efficiency can provide up to 1 kW. However,
even during the hottest parts of the day, a single panel cannot supply the 300–2000 W of
Pr

electricity needed for typical tropical refrigeration systems. A refrigerator box with a PV
module and control system attached is shown in Figure 3. The photovoltaic system is connected

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 to the refrigerator box. The three primary parts of the Solar Thermoelectric Refrigerator
(STER) are the control unit, PV module, and refrigerator box that are connected to the Peltier

ed
unit to enable effective heat dissipation. In October 2023, the STER's performance was
assessed in hot weather in Giza, Egypt, where average temperatures outside ranged from 35°C
to 55°C. From sunrise to sunset, the temperature inside the refrigerator box was continuously

iew
recorded during the study.

The configuration contains of a refrigerator box (RB) of 8 × 8 × 8 cm, with four thermoelectric
units (TU) mounted on its walls; these are positioned on the refrigerator box's four sides [15].

v
To stop cooling loss, the RB comprises two boxes: an inner stainless-steel box [16] and an

re
outside wooden box, separated by thermal insulating foam. Every thermoelectric unit has a
fan, heat sinks, and a Peltier unit. The Peltier unit's hot side must be effectively cooled for the
cold side to cool, and a fan usually helps with this. A digital millimetre is used to measure
er
voltage and current, and a digital thermometer is used to track temperature.
pe
ot
tn
rin

Fig. 1: Photographic picture of thermoelectric unit powered by PV

ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 2.2 Experimental Components' Material and Properties
A Peltier unit, specifically the TEC1-12706 type [17], which employs N-P junction

ed
semiconductors connected to a copper conductor. Table 1 contains comprehensive
performance details for the TEC1-12706 Peltier device.

Table 1 performance requirements for the TEC1-12706 Peltier unit

iew
Model Size, (mm) Power, Qmax, (W) Current, Imax, (A) Voltage, Vmax, (V)
TEC1-12706 40 × 40 50 6.4A 14.4

v
re
Fig. 2: Peltier unit
2.3 PV Systems in Refrigeration Applications
er
The performance of a photovoltaic (PV) cell can be described using several equations
depending on the context. However, the core equation for the output current and voltage of a
pe
PV cell is based on the Shockley diode equation. The general equation for the current produced
by a PV cell, PV Cell Output Current Equation

I𝑝𝑣 = Iph ― Id ― Ish (1)

IPh = 𝐺 ∗ [Iphref +α ∗ (T ― T𝑟𝑒𝑓)] (2)
ot

[
Id = I0 ∗ exp [
qv
𝑛𝑘𝑡
] ― 1] (3)
𝑉
tn

Ish = Rsh (4)

IPV = Iph ― I0exp [q(v𝑛𝑘𝑡

+ IR )
] ―1 ― 𝑉 +R 𝐼R
s
sh
s
(5)
rin

Depending on the cell temperature (𝑇), incident solar irradiance (𝐺), and PV current
temperature factor (𝛼𝑖), the PV array transforms incident solar irradiance into photoelectric
current (𝐼𝑝ℎ), photoelectric current (𝐼𝑆𝐶_𝑅). and solar irradiation (𝐺𝑅), at STC. The PV array's
ep

output voltage (𝑉𝑃𝑉). and current (I𝑝𝑣) are influenced by the load situation. The output current
(I) is equal to the photoelectric current (I𝑝ℎ) less the diode current (𝐼𝑑)and the shunt current (
Pr

𝐼𝑠ℎ). according to equation (1).

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 The diode current (𝐼𝑑) depends on the diode saturation current (𝐼𝑜), electron charge (q),
Boltzman constant (𝑘), and cell temperature (T), according to equation (3, 4). From previous

ed
equation the simulation in MATLAB done Figure 3 shows the MATLAB Simulink model
used to simulate PV cell.

v iew
re
er
pe
Fig. 3 The MATLAB Simulink model of photovoltaic cell

In MATLAB Simulink, a Photovoltaic (PV) model typically consists of several

interconnected blocks that represent different aspects of a PV module. The explanation of the
ot

key components is often found in a Simulink PV model. The common way for depicting the
electrical properties of a PV module is the current versus voltage (I-V) curve and power against
voltage (P-V) curve. The I-V and P-V curves for a typical PV module are displayed in Figure
tn

4, where Voc stands for Vmpp stands for maximum power point voltage, Impp for maximum
power point current, Pmpp for maximum power point, open circuit voltage, and Isc for short
circuit current.
rin
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 Pmp p
P-V c urve

ed
Isc I-V c urve

Current (A)

Power (W)
Imp p

iew
Voltage (V) Vmp p Voc

v
Fig. 4 PV module I-V curve and P-V curve

re
This is the main block that models the PV module or array of modules. It simulates the
electrical characteristics of a PV cell based on an equivalent circuit model, including a
current source, diode, series resistance (Rs), and shunt resistance (Rsh). The inputs to
PV array are Solar irradiance (G) this represents the intensity of sunlight (in W/m²),
er
typically ranging from 0 to 1000 W/m². Figure 4 demonstrations the MATLAB
Simulink model used to simulate PV Characteristic. These display blocks plot the (I-
pe
V) and (P-V) curves of the PV module based on the output current and voltage
ot
tn

Fig. 5 The MATLAB Simulink model used to simulate PV characteristics

rin

2.4 Modeling of Peltier using Adaptive Neuro Fuzzy Inference System (ANFIS)
A MATLAB model of a Peltier module provides a powerful simulation environment to analyze
ep

and optimize the performance of thermoelectric cooling systems. This model typically
represents the complex thermal and electrical characteristics of the Peltier device, which relies
on the Peltier effect to generate a temperature difference across its surfaces. The model can be
Pr

developed using MATLAB Simulink

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 Fuzzy Rule Interpretation the model provides a set of fuzzy rules that can offer insights into
the underlying behavior of the Peltier module. Applications Temperature Control, in cooling

ed
systems where precise temperature regulation is required. Energy Efficiency Optimization, to
improve the performance of thermoelectric coolers by predicting and adjusting inputs for
optimal efficiency. Performance Prediction For simulation and design purposes in systems

iew
using Peltier modules.

Table (1) displays the training data from Peltier data sheet

v
Table 1 Training data from Peltier data sheet

re
Volt (v) 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

dT (ºC) 0 0.5 1.8 2.6 3.99 5.17 6.15 6.9 7.79 7.83 7.92 7.73 7.53 7.02 6.3 5.37
er
Figure 6 shows the training date in MATLAB ANFIS edit platform, figure 7 displays the
pe
training error after training the network, and figure 8 shows actual Peltier output and ANFIS
Peltier emulation output.
ot
tn
rin
ep

Fig. 6 Training date in MATLAB ANFIS

Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
re
Fig. 7 Training error after training the network
er
pe
ot
tn

Fig. 8 Training error after training the network

rin

It is clear from the previous figures and from table (1) that the highest voltage for the Peltier

through the data sheet is 10 V therefore, when designing the boost converter the voltage will
ep

be set to 10 V to obtain the highest performance from the Peltier

Open-loop designs, a DC-DC converter is used to match the PV modules output voltage to the
Pr

operating requirements of the Peltier module. This converter does not regulate the power flow
based on cooling needs but merely aligns voltage and current to maximize energy transfer

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 efficiency from the PV module to the Peltier. Figure 9 demonstrations the MATLAB Simulink
model is used to simulate Peltier module powered by PV characteristics.

ed
v iew
re
Fig. 9 Open loop control of Peltier module powered by PV
er
Open-loop control is often chosen for its simplicity and low cost. This setup requires no sensors,
feedback loops, or complex controls, reducing the system's complexity. The direct connection between
pe
the PV and Peltier modules makes installation straightforward and eliminates the need for an external
power supply, making it ideal for remote or portable applications where minimal maintenance is
essential

2.5 Closed Loop Control of the Proposed Cooling System

ot

Closed-loop control for a photovoltaic (PV)-powered Peltier cooling system introduces a

feedback mechanism to regulate the Peltier modules cooling output, improving stability, efficiency, and
tn

adaptability under varying solar conditions. Unlike open-loop control, closed-loop control continuously
monitors system parameters, such as temperature, current, and voltage, and adjusts the power supply to
the Peltier module based on the cooling requirements.
rin

2.5.1 Cooling System Performance using PID Controller

A Proportional-Integral-Derivative (PID) controller is commonly used to maintain a
stable voltage in closed-loop control. The PID controller calculates an error value, which is the
ep

difference between the target and actual voltage measured in the PV output. Based on this error,
the controller adjusts the power supplied to the Peltier to bring the voltage closer to the set
point. This system diagram has a setup that integrates a Photovoltaic (PV) panel, a Peltier
Pr

module, and a Proportional-Integral-Derivative (PID) controller. This system is designed to

utilize solar energy to power a Peltier cooling device, commonly used in portable refrigeration
or temperature control applications. A detailed explanation of each component is shown in

10

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 Figures 10, and 11. The Block diagram shows the components of a closed loop system (PV,
buck-boost, Peltier, PID controller, feedback).

ed
v iew
re
er
pe
Fig. 10 Peltier module powered by PV with PID controller
ot
tn
rin
ep

Fig. 11 PID block diagram

Pr

The PID parameters are fine-tuned based on system requirements using trial and error,
ensuring rapid, stable, and accurate response to temperature changes. This tuning is essential

11

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 in applications where even small temperature deviations must be controlled, such as medical
or laboratory cooling.

ed
2.5.2 Buck – Boost Converter
Figure 12 depicts the buck-boost converter's construction. The converter is made up of

iew
a DC input voltage source (V), load resistance R, diode D; filter capacitor C, inductor L, and
controlled switch S. The diode is kept off when the switch is on, but the inductor current rises.
When the switch is off, the diode gives the inductor current a path. Take note of the diode's
polarity, which causes its current to be pulled from the output. Figure 13 shows the waveforms

v
of the buck-boost converter. In a steady state, the inductor's zero volt-second product condition
offers.

re
VSDT = ― VO(1 ― D)T er (7)
pe
ot

Fig. 12 Circuit diagram of buck-boost converter

tn

Therefore, the dc voltage transfer function of the buck-boost converter is

rin

VO D
GV = =― (8)
VS 1―D
ep

The output voltage Vo is negative with respect to ground. The magnitude of the converter, as
its name implies, can be greater or less than the input voltage (equal at D = 0.5).
Pr

12

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
re
Fig. 13 Waveforms of buck- boost converter

The buck – boost converter topology has advantage in its capability of performing under
er
the case when the output voltage more or less than the input source voltage. But Output voltage
polarity opposite to that of input and voltage stress on switching device in the chopper is high.
pe
For the reasons mentioned above, the boost topology is selected in this research because of
their greater efficiency and simplicity based on our system conditions. In the other cases the
topology should be selected based on the application conditions and limitations to get the
greater efficiency.
ot

3 Result
tn

3.1 Simulation Results at (MPPT) for Optimized PV Utilization using PID Controller
Closed-loop control systems often incorporate an MPPT controller to ensure the PV
module operates at its maximum power point, extracting the most power possible under current
rin

irradiance conditions. This is important for maintaining adequate power supply to the Peltier
module, especially under low-light or varying irradiance levels. Adjust the PV modules'
operating point to maximize energy capture, which is then directed to the Peltier system. By
optimizing PV output. There are three tests to show the validity of the PID controller the first
ep

test represents the low radiation and low voltage, the second test represents the medium
radiation and medium voltage, and the third test represents the high radiation and high voltage.
Pr

13

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 First test when solar radiation is low the voltage is low. Figure 14 (a, b, c) shows the variation
in cooling temperature (Tc), temperature difference (DT), and useful power (Qc) respectively.

ed
Figure 15 (a, b) displays the variation in duty ratio and buck-boost output voltage respectively.

v iew
re
er
Fig. 14 Peltier inputs by using PID controller variation in (a) Cooling temperature (b)

Temperature difference (c) Useful power

pe
ot
tn
rin
ep

Fig. 15 The input voltage by using PID controller (a) variation of duty ratio (b) variation of

buck boost output voltage

Pr

14

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 Second test when solar radiation is medium the voltage is medium. Figure 16 (a, b, c) shows
the variation in cooling temperature (Tc), temperature difference (dT), and useful power (Qc)

ed
respectively. And Figure 17 (a, b) displays the variation on duty ratio and buck boost output
voltage respectively.

v iew
re
er
pe
Fig. 16 Peltier inputs by using PID controller variation in (a) cooling temperature (b)

temperature difference (c) useful power

ot
tn
rin
ep
Pr

Fig. 17 The input voltage by using PID controller (a) variation of duty ratio (b) variation of

buck boost output voltage

15

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 Third test when solar radiation is high the voltage is high. Figure 18 (a, b, c) shows the
difference in cooling temperature (Tc), temperature difference (dT), and useful power (Qc)

ed
respectively. Figure 19 (a, b) displays the variation in duty ratio and buck-boost output voltage
respectively.

v iew
re
er
pe
.

Fig. 18 Peltier inputs by using PID controller variation in (a) cooling temperature (b)

temperature difference (c) useful power

ot
tn
rin
ep
Pr

16

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
re
er
Fig. 19 The input voltage by using PID controller (a) variation of duty ratio (b) variation of

buck boost output voltage

pe
3.2 Fuzzy Controller Block Diagram for a Peltier Module
A fuzzy controller is an intelligent control system designed to handle the nonlinear and
complex behavior of systems like a Peltier module. In this case, the fuzzy controller is used to
ot

maintain a specific temperature or temperature difference by adjusting the electrical input to

the Peltier module based on temperature feedback. Figure 20 shows Fuzzy Controller Block
tn

Diagram for a Peltier Module.

rin
ep
Pr

Fig. 20 Fuzzy Controller Simulink for a Peltier Module

17

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 Feedback Loop the output temperature of the Peltier module is continuously
monitored and fed back into the fuzzy controller. This forms a closed-loop control system,

ed
where the fuzzy controller continually adjusts the input to minimize the error. Figure 21 shows
the proposed system with the ANFS Program

v iew
re
er
pe

Fig. 21 Proposed systems With FPID controller

ot

3.3 Simulation Results at (MPPT) for Optimized PV Utilization using Hybrid Fuzzy-
PID Controller
tn

The Fuzzy logic adjusts the PID gains (Kp, Kd, Ki) based on real-time error analysis.
In scenarios with sudden changes in irradiance, the Fuzzy component increases (Kp) to enhance
rin

response speed, while it reduces (Kd) to minimize overshoot. This dynamic adjustment leads
to faster convergence to the optimal temperature and power output. Performance Enhancement:
The Hybrid Fuzzy-PID controller effectively manages cooling to enhance the PV system's
ep

efficiency. It dynamically adjusts to changes in environmental conditions, providing stable and

optimal temperature control. Figure 22 (a, b, c) shows variation in cooling temperature,
temperature difference, and useful power respectively,
Pr

Future Scope Real-time testing and integration with adaptive thermal management systems
could further improve performance. Additional research on hybrid cooling systems (e.g., phase

18

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 change materials, water-cooled systems) combined with advanced controllers could optimize
PV performance even more. There are three test shows the validity of fuzzy logic controller

ed
the first test represent the low radiation and low voltage, the second test represent the medium
radiation and medium voltage, the third test represent the high radiation and high voltage.

iew
First test when solar radiation is high the voltage is high. Figure 23 (a, b, c) shows the
variation in cooling temperature (Tc), temperature difference (dT), and useful power (Qc)
respectively. And Figure 24 (a, b) displays the variation on duty ratio and buck boost output
voltage respectively.

v
re
er
pe
ot
tn

Fig. 22 Peltier inputs by using FPID controller variation in (a) cooling temperature (b)

temperature difference (c) useful power

rin
ep
Pr

19

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
re
er
Fig. 23 The input voltage by using FPID controller (a) variation of duty ratio (b) variation of
buck boost output voltage
Second test when solar radiation is medium the voltage is medium. Figure 24 (a, b, c) shows
pe
the variation in cooling temperature (Tc), temperature difference (dT), and useful power (Qc)
respectively. And Figure 25 (a, b) displays the variation on duty ratio and buck boost output
voltage respectively.
ot
tn
rin
ep

Fig. 24 Peltier inputs by using FPID controller variation in (a) Cooling temperature
Pr

(b) Temperature difference (c) Useful power

20

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
re
Fig. 25 The input voltage by using FPID controller (a) variation of duty ratio (b) variation of
er
buck boost output voltage
Third test when solar radiation is high the voltage is high. Figure 26 (a, b, c) shows the
pe
variation in cooling temperature (Tc), temperature difference (dT), and useful power (Qc)
respectively. And Figure 27 (a, b) displays the variation on duty ratio and buck boost output
voltage respectively.
ot
tn
rin
ep

Fig. 26 Peltier inputs by using FPID controller variation in (a) cooling temperature (b)
temperature difference (c) useful power
Pr

21

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
re
Fig. 27 The input voltage by using FPID controller (a) variation of duty ratio (b) variation of
buck boost output voltage
er
3.4 PID controller and PID fuzzy logic controller comparison for the suggested system
Designing a fuzzy PID control with MATLAB, we next examine the control effect and
pe
contrast it with the PID controller's effect. Comparatively speaking, fuzzy control outperforms
PID control. In particular, it can focus more on a number of factors, including overshoot, steady
state error, and response time. When the control response from these two systems was
compared, it was shown that the fuzzy logic controller greatly decreased steady state error and
ot

overshoot. The PID controller is simpler but struggles with nonlinear behavior and requires
precise tuning. It works well for linear systems but may not handle the dynamics of a PV-
tn

Peltier cooling system effectively, especially when environmental conditions vary. The Fuzzy
Logic controller provides better performance in nonlinear systems and adapts well to varying
conditions. It doesn't require a mathematical model, making it easier to apply to complex
rin

systems like Peltier modules driven by PV panels. Figure 28 displays the input voltage, Figure
3.47 and figure 3.48 show the comparison between PID and FPID on first test
ep
Pr

22

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
Fig. 28 The input Voltage

re
er
pe
ot
tn
rin

Fig. 29 First test variation in (a) cooling temperature (b) temperature difference (c) useful

power
ep
Pr

23

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
re
er
Fig. 30 First test (a) variation of duty ratio (b) variation of buck boost output voltage

Figure 31 and figure 32 show the comparison between PID and FPID on second test
pe
ot
tn
rin
ep
Pr

24

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 Fig. 31 Second test variation in (a) cooling temperature (b) temperature difference (c) useful
power

ed
v iew
re
er
pe
Fig. 32 Second test (a) variation of duty ratio (b) variation of buck boost output voltage

Figure 33 and figure 34 show the comparison between PID and FPID at third test
ot
tn
rin
ep
Pr

25

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 Fig. 33 Third test Variation in (a) cooling temperature (b) temperature difference (c) useful
power

ed
v iew
re
er
Fig. 34 Third test (a) variation of duty ratio (b) variation of buck boost output voltage
pe

4 Conclusions
The fuzzy logic controller offers significant advantages over traditional controllers when
applied to systems like Peltier modules. Due to its ability to handle nonlinear behavior, the
ot

fuzzy controller provides faster response and higher accuracy in temperature regulation. Unlike
conventional controllers, which rely on precise mathematical models and may struggle with
tn

changing operating conditions, the fuzzy controller adapts effectively using a rule-based
approach. This flexibility allows it to maintain stable performance even under varying
environmental factors, making it a superior choice for application requiring precise and reliable
rin

control

References
[1] M. Rahman, … M. I.-J. of E. and, and undefined 2024, “Design and Implementation of a Solar-
ep

Powered DC Refrigerator Based on Peltier Effect,” researchgate.netMR Rahman, MJ Islam, IB

ChowdhuryJournal of Electrical and Power System Engineering, 2024•researchgate.net,
Accessed: Nov. 28, 2024. [Online].
Pr

[2] J. Mao, G. Chen, and Z. Ren, “Thermoelectric cooling materials,” Nature Materials 2020 20:4,
vol. 20, no. 4, pp. 454–461, Dec. 2020, doi: 10.1038/s41563-020-00852-w.

26

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 [3] M. Kalimuthu, … R. S.-… on A. in, and undefined 2021, “Peltier based temperature controlled
smart mini refrigerator,” ieeexplore.ieee.orgM Kalimuthu, R Subashini, UM Al Rasheeth, M

ed
Sabaeshwaran2021 International Conference on Advancements in Electrical,
2021•ieeexplore.ieee.org, Accessed: Nov. 28, 2024.

[4] I. Sarma, G. Reddy, … V. Y.-… on I. C., and undefined 2024, “Solar Energy-Powered Automated
Refrigerator,” ieeexplore.ieee.orgIVS Sarma, G Reddy, V Yogesh, V Sailaja2024 International

iew
Conference on Integrated Circuits, 2024•ieeexplore.ieee.org, Accessed: Nov. 28, 2024.
[Online]. Available: https://ieeexplore.ieee.org/abstract/document/10602977/

[5] B. Sahare, C. S.-I. J. of R. T. and, and undefined 2020, “Design and Development of a Cooler
used for Air Cooling and Refrigeration,” academia.eduB Sahare, C SahuInternational Journal

v
of Recent Technology and Engineering, 2020•academia.edu, Accessed: Nov. 28, 2024.
[Online]. Available: https://www.academia.edu/download/112987371/E6886018520.pdf

re
[6] A. Ramadan, S. Kamel, I. Hamdan, A. A.- Mathematics, and undefined 2022, “A novel
intelligent ANFIS for the dynamic model of photovoltaic systems,” mdpi.comA Ramadan, S
Kamel, I Hamdan, AM AgwaMathematics, 2022•mdpi.com, Accessed: Nov. 28, 2024.
[Online]. Available: https://www.mdpi.com/2227-7390/10/8/1286
er
[7] B. R.-E. P. S. Research and undefined 2024, “Novel MLI-based DVR and DSTATCOM with ANFIS
control for enhanced power quality improvement,” ElsevierB RekhaElectric Power Systems
Research, 2024•Elsevier, Accessed: Nov. 28, 2024.
pe
[8] M. Bakare, A. Abdulkarim, … A. S.-P.-A. in, and undefined 2024, “PREDICTIVE ENERGY
CONTROL FOR GRID-CONNECTED INDUSTRIAL PV-BATTERY SYSTEMS USING GEP-ANFIS,”
Elsevier, Accessed: Nov. 28, 2024. [Online]. Available:
https://www.sciencedirect.com/science/article/pii/S2772671124002274
ot

[9] J. Z.-E. P. S. Research and undefined 2024, “Hybrid aquila arithmetic optimization based
ANFIS for harmonic mitigation in grid connected solar fed distributed energy systems,”
ElsevierJ ZahariahElectric Power Systems Research, 2024•Elsevier, Accessed: Nov. 28, 2024.
tn

[10] N. Bhavani, A. Singh, D. K.-P.-A. in Electrical, and undefined 2024, “Optimized integration of
renewable energy sources using seven-level converter controlled by ANFIS-CS-GWO,”
Elsevier, Accessed: Nov. 28, 2024. [Online]. Available:
rin

https://www.sciencedirect.com/science/article/pii/S2772671124002699

[11] Y. Yin, X. Zhai, R. W.-A. thermal engineering, and undefined 2013, “Experimental investigation
and performance analysis of a mini-type solar absorption cooling system,” ElsevierYL Yin, XQ
ep

Zhai, RZ WangApplied thermal engineering, 2013•Elsevier.

[12] H. Najafi, K. W.-S. Energy, and undefined 2013, “Optimization of a cooling system based on
Peltier effect for photovoltaic cells,” ElsevierH Najafi, KA WoodburySolar Energy,
2013•Elsevier.
Pr

27

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 [13] N. Khadka, A. Bista, B. Adhikari, A. Shrestha, and D. Bista, “Smart solar photovoltaic panel
cleaning system,” IOP Conf Ser Earth Environ Sci, vol. 463, no. 1, Apr. 2020, doi:

ed
10.1088/1755-1315/463/1/012121.

[14] T. Prantor, … M. H.-2016 I. R. 10, and undefined 2016, “Generation of electricity using
concentrated solar power and thermo-electric module,” ieeexplore.ieee.orgTT Prantor, M
Hasan, CA Hossain2016 IEEE Region 10 Conference (TENCON), 2016•ieeexplore.ieee.org.

iew
[15] A. E. M. Elnaggar, S. Sharaf, Z. S. Abedel Rehim, M. A. El-Bayoumi, H. M. M. Mustafa, and H.
M. El Zoghby, “Enhancing COP and Cooling Temperature for Peltier with Decreasing Power,
and Chemical Composition Negative Effect by Optimizing Connection, Position Angle, and
Voltages,” Egypt J Chem, vol. 67, no. 13, pp. 453–459, Dec. 2024, doi:

v
10.21608/EJCHEM.2024.257857.9051.

[16] A. E. M. Elnaggar, S. Sharaf, Z. S. Abedel Rehim, H. M. M. Mustafa, and H. M. El Zoghby,

re
“Enhancing the Coefficient of Performance (COP) of Mini Refrigerators Based on
Thermoelectric Units (Peltier),” polityka Energetyczna, 2024.

[17] A. E. M. Elngar, Z. S. A. Rehim, W. M. Refai, M. El-Bayoumi, and A. L. Marzouk, “Modeling and

er
Power Control for Integrated System of Photovoltaic and Thermoelectric Element,” Journal of
Scientific and Engineering Research, vol. 5, no. 12, pp. 202–210, 2018.
pe
Abbreviations Nomenclature
ANFIS Adaptive Neuro Fuzzy Inference System
FPID Fuzzy-Proportional-Integral- Derivative Ti Initial temperature, °C
TE Thermoelectric Cp Specific heat of water, J kg−1 K−1
ot

TEC Thermoelectric cooler I Electric current, A

TU Thermoelectric unit M Mass of water, kg
COP Coefficient of performance Q˙ Heat transfer rate, W
tn

Cp Specific heat of water, J kg−1 K−1 Tc The temperature of Cold junction,

CFCs Chlorofluorocarbons Th The temperature of Hot junction, °C
PV Photo voltaic T Time, s
Q˙ Heat transfer rate, W W˙ Power consumption, W
rin

Tc The temperature of Cold junction, °C T Temperature, °C

Th The temperature of Hot junction, °C V Voltage, V
RB Refrigerator box
TU Thermoelectric unit
ep

S.S. Stainless steel

∆T The cooling temperature difference
PID Proportional-Integral- Derivative
DEC Direct evaporative cooling
Pr

28

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478
 ed
v iew
re
er
pe
ot
tn
rin
ep
Pr

29

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=5105478