sensors

Article
Democratization of PV Micro-Generation System Monitoring
Based on Narrowband-IoT
José Miguel Paredes-Parra 1 , Raquel Jiménez-Segura 2 , David Campos-Peñalver 1 , Antonio Mateo-Aroca 2 ,
Alfonso P. Ramallo-González 3 and Angel Molina-García 2, *

1 Technological Center of Energy and Environment (CETENMA), 30353 Cartagena, Spain;

jmparedes@cetenma.es (J.M.P.-P.); david.campos@cetenma.es (D.C.-P.)
2 Departament of Automatics, Electrical Engineering and Electronic Technology, Universidad Politécnica de
Cartagena, 30202 Cartagena, Spain; raquel.jimenez@edu.upct.es (R.J.-S.); antonio.mateo@upct.es (A.M.-A.)
3 Department of Information and Communications Engineering, Computer Science Faculty,
University of Murcia, 30100 Murcia, Spain; alfonsop.ramallo@um.es
* Correspondence: angel.molina@upct.es; Tel.: +34-968-32-5462

Abstract: Power system configuration and performance are changing very quickly. Under the new
paradigm of prosumers and energy communities, grids are increasingly influenced by microgen-
eration systems connected in both low and medium voltage. In addition, these facilities provide
little or no information to distribution and/or transmission system operators, increasing power
system management problems. Actually, information is a great asset to manage this new situation.
The arrival of affordable and open Internet of Things (IoT) technologies is a remarkable opportunity
to overcome these inconveniences allowing for the exchange of information about these plants. In this
paper, we propose a monitoring solution applicable to photovoltaic self-consumption or any other



microgeneration installation, covering the installations of the so-called ’prosumers’ and aiming to


provide a tool for local self-consumption monitoring. A detailed description of the proposed system
Citation: Paredes-Parra, J.M.;
at the hardware level is provided, and extended information on the communication characteristics
Jiménez-Segura, R.; Campos-
Peñalver, D.; Mateo-Aroca, A.;
and data packets is also included. Results of different field test campaigns carried out in real PV
Ramallo-González, A.P.; Molina- self-consumption installations connected to the grid are described and analyzed. It can be affirmed
García, A. Democratization of PV that the proposed solution provides outstanding results in reliability and accuracy, being a popular
Micro-Generation System Monitoring solution for those who cannot afford professional monitoring platforms.
Based on Narrowband-IoT. Sensors
2022, 22, 4966. https://doi.org/ Keywords: photovoltaic microgeneration; NB-IoT technology; monitoring network; Internet of
10.3390/s22134966 Things
Academic Editor: Grzegorz Fusiek

Received: 19 May 2022

Accepted: 25 June 2022
1. Introduction
Published: 30 June 2022
In recent years, most countries have established ambitious renewable energy targets
Publisher’s Note: MDPI stays neutral
for their electricity generation to foster sustainable and low-emission initiatives. In line
with regard to jurisdictional claims in
with the European and COP21 decarbonization environmental objectives [1], Spain has
published maps and institutional affil-
promoted new legislation providing support and an optimized framework for the Renew-
iations.
able Energy Sources (RES) integration, mainly wind power plants and solar photovoltaic
(PV) installations [2]. These new regulations, together with a significant reduction in the
cost of PV technology as a result of their own technological maturity, have addressed the
Copyright: © 2022 by the authors.
enormous growth in the integration of such solutions. Subsequently, a relevant number
Licensee MDPI, Basel, Switzerland. of PV installations can be found at different voltage levels; mainly under the alternative
This article is an open access article cooperatively-owned organizations giving a substantially different model of energy dis-
distributed under the terms and tribution and provision [3]. In Europe, and more specifically in Spain, PV installations
conditions of the Creative Commons have been widely integrated by prosumers as feasible strategies to decrease power demand,
Attribution (CC BY) license (https:// fulfill emission agreements and accelerate self-consumption growth rates [4]. These policies
creativecommons.org/licenses/by/ can have relevant consequences on the design and operation of distribution networks to
4.0/). which most prosumers are connected [5]. Under this scenario, guaranteeing a safe and

Sensors 2022, 22, 4966. https://doi.org/10.3390/s22134966 https://www.mdpi.com/journal/sensors

Sensors 2022, 22, 4966 2 of 20

reliable RES integration is a remarkable challenge to be solved by power systems in the

coming years [6]. Actually, high RES integration is included in the Strategic Energy Tech-
nology Plan (SET Plan) of the European Union—see Activity 4: Operation and diagnosis of
Photovoltaic Plants, highlights the digitization of the energy sector as a fundamental and
crucial element for the EU [7]. In parallel to the current developments of technology, digiti-
zation offers new opportunities for system operators and facility managers and maintainers
by optimizing the operation of their assets, reducing their operating costs and allowing
greater renewable integration. In this context, real-time medium- and low-voltage grid
monitoring and control under the new technologies will lead to significant improvement
of grid efficiency and hosting capacity [8]. Moreover, local and timely management of
distributed resources and self-consumption installations will be the key to ensure efficiency
and stability at the distribution and the operation level [9]. Among the different solutions,
PV self-consumption systems constitute a remarkable challenge for distributed generation
contributing to future smart grids [10].
With regard to solar PV monitoring technologies, various review contributions can
be found in the specific literature from different points of view [11–13]. Nevertheless, both
environmental and technical factors can significantly affect the PV power generation—partial
shading due to moving clouds, module temperature, humidity, the mounting angle, etc.
Further explorations are then required to design an effective solar PV monitoring tech-
nology [14]. In addition, the use of monitoring systems in large-scale PV systems can
be justified despite their high costs [15]. However, monitoring systems in medium and
small PV installations may address additional and high costs unacceptable to most users.
For this reason, data acquisition and monitoring solutions using low-cost platforms have
been widely proposed in recent years [16]. The advance of technologies such as IoT and
Industry 4.0 have allowed the implementation of new monitoring solutions as an alter-
native to traditional centralized systems. These solutions mainly aim to detect anomalies
of PV solar plants and for maintenance purposes. Different approaches can be found in
recent literature. Madeti et al. [11] provided a comprehensive review of PV monitoring
systems, comparing most of PV monitoring evaluation techniques in terms of their relative
performances. This contribution also examined sensors, controllers in data acquisition
systems, data transmission methods and data storage and analysis. In [17], Sunarso et al.
proposed a low-cost PV monitoring system based on an open-source Arduino platform
and demonstrated its use for the assessment of PV monitoring and supervision potential.
The system developed by Sunarso et al. can monitor solar irradiance, electric outputs
and temperature of multiple solar panels. A similar approach can be found in [18], where
the authors described a prototype for a portable data logger that integrates standalone
instruments with open-source hardware technologies for monitoring PV systems. The
proposed datalogger accomplished the accuracy requirements of the IEC standards for
PV systems. In 2019, López-Vargas et al. [19] improved the design with a new version of
this low-cost datalogger to overcome the shortcomings related to power consumption, the
limited voltage range and the user interface for stand-alone PV systems located in remote
areas deprived of telecommunications networks. Data were stored on an SD card to work
autonomously. This work continued with a new design that included allowing 3G technol-
ogy to monitor the PV systems remotely via web or via mobile application [20]. Regarding
communications, Ansari et al. [14] reviewed various monitoring technologies for PV power
plants focused on data processing modules and data transmission protocols. They showed
key issues and limitations of this technology. In terms of transmission protocol, this work
selected LoRa as a relevant data transmission solution, but it was not suitable for large
payloads. Moreover, a continuous data packet sending process should be avoided due to
some rules and constraints regarding the frequency band, which is in line with other recent
contributions [21].
From the limitations of the techniques found in the specific literature, and considering
the requirements of the application at hand, the authors previously proposed a solution
using a Raspberry Pi that served as a comprehensive solar PV monitoring following the
Sensors 2022, 22, 4966 3 of 20

IEC-61724 requirements. The solution was used at the module level [22]. It was tested
and assessed at real PV solar installations with remarkable results. The tests took some
months in CETENMA SOLAR installations facilities. A 250 Wp module connected to the
grid with a Soltec Solarfighter microinverter was monitored, as well as a 5 kWp ver-roof
installation connected to the grid and located in the Universidad PolitÃl’cnica de Cartagena
(Spain) during some weeks. All variables were collected with standard equipment at both
facilities. The results showed that the estimated error rate was lower than 2%. From this
preliminary design, a low-cost LoRa-based solar PV monitoring system to communicate
solar PV power plants located in remote locations was subsequently proposed in [23]. This
topology used an Arduino Board with an RFM95W transceiver fabricated by HOPE RF
configured as a LoRa TM modem available for the EU-868 MHz band [24]. The solution
stored data in a packet size of 38 bytes with a transmission power of 14 dBm and spreading
factor (SF) metric ranging from 10 to 12. An alternative system to be implemented in remote
PV power plants for monitoring and dispatching electrical and meteorological data was
then proposed and evaluated. This study demonstrated that the line of sight between
source and destination and propagation issues had a clear influence on the suitable data
reception process. As a main drawback, this solution only allowed a limited number of
transmissions, due to the over-air time of works with high spreading factor value (SF11 and
SF12) to ensure an accurate reception of data packets with a restricted duty cycle (1%). In
addition, a LoRa sensitive analysis was conducted by the authors in [25] for PV short-time
forecasting accuracy estimation.
By considering this preliminary background, as well as the lack of contributions
focused on monitoring self-consumption PV solar installations and providing information
to prosumers in terms of consumption/production profiles [26], the main contributions of
this paper are described below:
• A novel monitoring solution for the operation and maintenance of self-consumption
PV systems is proposed and assessed.
• The solution is based on Internet of Things (IoT) applications by using Narrowband
IoT (NB-IoT).
• The new design uses the Pycom IoT platform with NB-IoT as data transmission
technology.
The Pycom development board FiPy and sensor shield Pysense have recently been
proposed to monitor environmental variables such as humidity, temperature, altitude,
pressure, or light [27].
The rest of the paper is structured as follows: Section 2 introduces NB-IoT technology;
Section 3 discusses materials and methods; Section 4 describes the use case; Section 5
discusses both collected data and results; finally, conclusions are given in Section 6.

2. Communications: NB-IoT
Low-power, wide-area (LPWA) technologies are targeting most emerging markets
and applications. LPWA is a recent generic term involving different technologies focused
on enabling wide area communications, minimizing power consumption and cost [28].
LPWA has become a relevant growing space in IoT since in general LPWA solutions are
perfectly suitable for such IoT applications that only need to transmit low amounts of data
over a long range. Other researchers proposed a ZigBee network tested with WiFi for
smart grid applications. However, these tools have limited capability for analyzing real
scenarios and, according to Sultania et al. [29] ZigBee uses the same frequency band as WiFi,
which can lead to potential radio interference. Consequently, most of the developed LPWA
technologies have arisen in both unlicensed and licensed markets, such as SigFox, LTE-M,
NB-IoT and long range (LoRa). Among them, Shina et al. [30] affirmed that NB-IoT and
LoRa solutions are the two leading emergent technologies,despite the important technical
differences in terms of network architecture, physical features and MAC protocol between
them. Table 1 compares different network schemes and communication technologies.
Sensors 2022, 22, 4966 4 of 20

Table 1. Network schemes and communication technologies properties.

Network Data Transfer Typical

Technologies Latency
Schemes Rate Coverage Range
Ethernet, PLC,
HAN Zigbee, WLAN, 10–100 kbps up to 500 m 10 ms–1 s
Z-Wave
Ethernet, PLC, 100 kbps–1
BAN/IAN up to 1 km 10 ms–2 s
Wimax, WLAN Mbps
Ethernet, PLC,
DSL,
100 kbps–10
NAN Fiber–Optics, 0.1–10 km 10–50 ms
Mbps
WiMax, NB–IoT,
LoRa
PLC, Ethernet,
WAN Fiber–Optics, 10 Mbps–1 Gbps 10–100 km 10 µs–20 ms
LoRa, WiMax

Recently, different comparative studies are available in the specific literature evaluat-
ing LoRa and NB-IoT technologies [31]. Examples of new network architectures combining
both technologies can be found in [32,33]. In terms of vulnerabilities, LoRaWAN and
NB-IoT give sufficient security guarantees, but according to Coman et al. [34] both tech-
nologies need to be properly enforced. The NB-IoT technology was standardized in 2016
by the Third Generation Partnership Project (3GPP), considerably increasing the NB-IoT
applications for wireless data communication purposes [35]. In addition, the presence
of NB-IoT modules in the IoT device market share by 2030 will considerably increase
too [36]. NB-IoT is compatible with LTE-M, with transmission speed of 150 kbps and a
coverage range of 15 km. It is a licensed technology deployed in 79 countries, with an
investment of $598 billion [37]. In this case, NB-IoT is selected by the authors for PV power
plant monitoring and communication network evaluation. This proposal is in line with
other recent network applications, such as Smart Water Grid (SWG) [38], water quality
monitoring [39] or sustainable farming irrigation [40]. In all cases, NB-IoT was selected as
a suitable candidate due to high scalability in comparison to other technologies, such as
LoRaWAN and SigFox. Li et al. [41] demonstrated that NB-IoT satisfied both qualitative
and quantitative requirements in terms of security, reliability and scalability. The relevant
presence of IoT–infrastructure devices, the remarkable variety of applications and the use
of various data processing solutions have recently led to the fact that traditional data center
architectures do not allow the software configuration nor the physical scaling for tasks to be
solved. Moreover, they are no longer able to provide the current required indicators focused
on stability, controllability or productivity [42]. On the other hand, and regarding industrial
applications, Ballerini et al. [43] concluded that NB-IoT offered the highest quality of service
(QoS) while also ensuring data delivery, being a potential replacement to LoRaWAN when
communication reliability was a required factor. In addition, NB-IoT was also proposed as
an alternative solution for real-time demand response, able to monitor, control and connect
electrical appliances [44].
In this paper, we propose an NB-IoT solution for PV power plant monitoring and
communication purposes focused on self-consumption installations. Our approach is in
line with recent contributions focused on PV installations [45–47]. We extend the use of
IoT technology toward prosumers, providing an alternative solution for future distribution
system scenarios in which a high renewable penetration would lead to higher peaks of
generation as a consequence of potential reverse power flows at the medium voltage/low
voltage (MV/LV) distribution transformers [48].
Sensors 2022, 22, 4966 5 of 20

3. IoT Monitoring Solution

3.1. General Architecture
The proposed system uses a FiPy development board as the main hardware. FiPy is a
low-cost ESP-32 development board produced by Pycom manufacturer, which includes
WiFi, Bluetooth, LoRa, Sigfox and dual LTE-M (CAT M1 and NBIoT) technologies. This
platform allows us to test different communication technologies excluding any hardware
modification. Table 2 summarizes the most relevant technical features of these boards [49].

Table 2. FiPy specifications.

CPU
- Xtensa© dual–core 32–bit LX6 microprocessor(s), up to 600 DMIPS
- Hardware floating point acceleration
- Python multi–threading
- An extra ULP–coprocessor that can monitor GPIOs, the ADC channels and control most of
the internal peripherals during deep–sleep mode while only consuming 25 µA.
Networks
- WiFi (1 km range)
- BLE
- Cellular LTE-CAT M1/NB1 (Total world–wide support)
- LoRa
- Sigfox
Memory
- RAM: 520 kB + 4 MB
- External flash: 8 MB
- GPIO: Up to 22
- Hardware floating point acceleration
- Python multi–threading
Interfaces
- 2 ×UART, 2 × SPI, I2C, micro SD card
- Analog channels: 8 (12–bit ADC), 2 (8–bit DAC)
- Timers: 2 of 64 bit with PWM with up to 16 channels
- DMA on all peripherals
- GPIO: up to 22
Hash/Encryption
- SHA, MD5, DES, AES
RTC
- Running at 32 kHz
Range
- Node range: up to 50 km
Power
- Voltage Input: 3.3 V–5.5 V
- 3v3 output capable of sourcing up to 400 mA
Size
- 55 mm × 20 mm × 3.5 mm (excluding headers)

Pycom is programmed in MicroPython language [50]. For this purpose, it is necessary

to use Pymakr, which is an Integrated Development Environment (IDE) plugin for popular
code editors, such as Atom and Visual Studio Code. Pymakr thus makes the development
of IoT edge devices running microPython easier [51]. Although it is possible to program
FiPy via Pybytes, Visual Studio Code is an IDE widely used for this purpose. With regard
to the code developed by the authors, Figure 1 shows the structure and main blocks of such
code schematically. Firstly, the sensors are configured and set, the FiPy microcontroller
being able to collect data from such sensors. Secondly, a reading process of data from
the sensors is carried out, and subsequently, the corresponding message is sent via LTE
Sensors 2022, 22, 4966 6 of 20

including the gathered data. It is then necessary to verify if the device is connected to the
Pybytes platform. Indeed, eventual disconnections were detected during the testing, and a
reconnection process was included in the following terms: Once the connection is verified,
the message can be sent to the platform via LTE. This process is repeated with one-minute
sample time during the total of sunlight daily hours. FiPy is set on deep-sleep under low
sunlight conditions, in line with other contributions [52].

Figure 1. Flow diagram for the microcontroller.

Pybytes is a free cloud-based device management platform for all Pycom development
boards and modules. It provides a mobile app, allowing us to manage devices directly from
a smartphone. It is possible to change the priority of the used networks: WiFi, Bluetooth
and LPWAN networks. Pybytes also provides an easy to use dashboard to quickly create an
application that sends data to the platform and allows us to choose among different types
of data visualization. That dashboard updates automatically, and subsequently, the user
can visualize current status of the devices and received data in real time. It is also possible
to integrate Pybytes with a Cloud provider: AWS, Microsoft Azure, Webhooks and Google
Cloud. A recent infrastructure based on Pycom development board FiPy and sensor shield
Pysense to collect and send data to the remote cloud over Wi-Fi and Long Range (LoRa)
Sensors 2022, 22, 4966 7 of 20

protocols is also described in [27] for remote monitoring conditions. In this case, to provide
greater reliability and synchronicity to the data, a DS1302 RTC was incorporated. Thanks
to this module, the sending of packets to Pybytes was controlled every minute as were
the deep-sleep mode periods for suitable energy saving. This module provided us with
a temporary time stamp, helping us when drawing conclusions from the measured data.
In addition, and due to the constant disconnections of the Pycom LTE network, different
solutions were tested by the authors. The connection to the network was configured by
following the official documentation. The script was set to send packets every minute
over NB-IoT. In parallel, the firmware of the FiPy microcontroller used was developed to
perform a ping to the Pybytes platform and thus verify correct communication. If that
ping was not answered due to any connection drop, then it would wait 10 min for the
next ping. If this second ping was not answered either, then the watchdog timer was
triggered, and the system rebooted to re-establish communication. Consequently, under a
communication failure event, 10 min of information would be lost as the microcontroller
remains on hold. This operation cannot be modified in the program, but it is a matter of
firmware configuration.

3.2. Hardware
With regard to hardware components, the selected sensors as well as the calibration
process and the prototype encapsulation are now described. In fact, the selected sensors
provide an analog output necessary to calibrate them, in our case by using the Transmille
3000A Series calibrator. The sensors are in charge of monitoring PV installations under the
IEC-61724 requirements. The IEC-61724 standards were previously proposed to analyze
PV system performance under a variety of climates [53]. In our solution, PV electrical data
and weather parameters are gathered to estimate PV operating conditions and exchange
meteorological and electrical data. More specifically, the ACS758 current sensor [54] and
the YHDC HV25 voltage sensor were selected.

3.2.1. DC-Output Current: Acs758 Hall-Based Effect Sensor

A breakout board for the ACS758 Hall effect linear current sensor was selected for
measuring the DC current of the PV module string. The thickness of the copper conductor
of this sensor allows the device to survive under high current conditions. The ACS758
provides an analog voltage output signal that varies linearly with the current value. The
output voltage of the sensor varies significantly depending on the input current value.
Therefore, different currents were applied within the available measurement interval,
ranging from −50 A to 50 A DC. In this case, we used a calibration current range from 0 A
to 20 A, in steps of 1 A. Subsequently, the sensor output voltage was measured through
the FiPy microcontroller, see Figure 2. With this aim, a script was defined to convert the
measured value into an estimated current value. The sensor sensitivity provided by the
manufacturer was 40 mV/A, for a channel up to 5 V. In our case, the microcontroller channel
input was 3.3 V, and subsequently the sensitivity to be used was 26.4 mV/A. The offset
voltage was 1.65 V for 0 A, corresponding to 50% of the FiPy microcontroller analogical
offset range. The relative error was estimated by comparing the real input current to the
measured current values, with a maximum error of 4%. Different current sensors were
calibrated and compared to evaluate their performance and accuracy within the expected
current range. Figure 3 compares two current sensor calibration processes. An additional
source of potential errors in the current sensor output is the sensitivity of the FiPy analog
input channel, which is 20 mV/A.
Sensors 2022, 22, 4966 8 of 20

Figure 2. Current sensor calibration assembly: Transmille 3000A Series calibrator, ammeter, regulated
power supply and current sensor.

Figure 3. Current sensor calibration process: comparison of input current vs measured current.

3.2.2. DC-Output Voltage: Yhdc Hv25 Sensor

Regarding the DC voltage measurement, the low-cost YHDC HV25 sensor was selected.
It allows us to measure voltages up to 1000 V, scaling them to 5 V. Since the analog inputs
were used according to the Pycom FiPy microcontroller—working at 3.3 V, it was necessary
to include a voltage divider circuit between the sensor output and the analog–pin input.
This voltage divider consists of a 56 kΩ resistor and two 56 kΩ resistors in series. In this
way, an output of 3.3 V is obtained for an input of 1000 V, with a sensitivity of 3.3 mV/V.
With the aim of calibrating the sensor, a set of measurements were carried out from from
0 V to 1000 V, in steps of 50 V. For this purpose, a symmetrical voltage supply of ±12 V and
a variable voltage output signal were used together with the 3041A precision multi-product
calibrator, see Figure 4. A set of input voltage measurements were compared to the real DC
voltage to obtain a calibration curve for this voltage sensor. In a similar way to Section 3.2.1,
an additional set of measurements were also collected to estimate the relative error under
different boards. By considering that each solar panel had an open-circuit voltage (Voc ) of
50 V, the determined relative error was lower than 3%.
Sensors 2022, 22, 4966 9 of 20

Figure 4. Voltage sensor calibration assembly. Transmille 3000A Series calibrator and voltage sensor.

Due to the high error obtained in the low range of measurement—lower than 50 V—the
option of recalculating the voltage divider was then proposed by the authors. The target
was to equate the maximum input of 3.3 V to 800 V to test if the sensor precision could
be improved below 50 V. After a series of testing measurements, the option was finally
discarded since they did not significantly improve the results, and there was a risk of
obtaining a voltage higher than 800 V that could damage the equipment. It was observed
that the selected sensor did not provide a suitable voltage output for inputs lower than 50 V.
Therefore, a voltage divider was included in the laboratory tests to allow collecting voltage
data below such a threshold instead of using the YHDC HV25 sensor. This voltage divider
was formed by two resistances: 10 kΩ and 1 kΩ, respectively. The output signal was then
collected from the 1 kΩ voltage value. Given the high variability of the error for each YHDC
HV25 plate, the option of calibrating each board individually, with its own calibration line,
was selected. With this aim, Figure 5 compares the calibration process for three different
voltage sensor boards. In addition, Table 3 shows the estimated relative errors for these
three different voltage sensor boards. As shown in these results, the proposed board has
minor accuracy at low voltage levels but high accuracy over the expected operating voltage
ranges in medium and small PV installations. Therefore, it is accepted for our proposals.

Table 3. Relative Error YHDC HV25.

Range (V) Board 1 Board 2 Board 3

50–950 2.93% 2.66% 2.83%
150–1000 2.51% 2.57% 1.49%

3.2.3. Solar PV Module Temperature: Ds18b20 Digital

The DS18B20 is a digital thermometer providing from 9-bit to 12-bit Celsius temper-
ature measurements up to 125 ◦ C [55]. There is also a watertight package that protects
the sensor and allows it to be submerged in a liquid without damage available. Since it is
a digital sensor, the read signal does not degrade due to wiring distance. It can work in
Sensors 2022, 22, 4966 10 of 20

one-wire mode with an accuracy of ±0.5 ◦ C and 12-bit resolution. Various sensors can also
be used on the same pin, as they can be internally programmed with a unique 64-bit ID to
identify and differentiate them. The operating range is 3 to 5 V, being able to be used in
virtually any system by the use of microcontrollers. Further information of temperature
sensor analysis for PV solar module temperature measurement can be found in [56]. The
DS18X20 class from the onewire.py library [57] was also included to properly use this
sensor with the FiPy. The code can be found in the Appendix A.

Figure 5. Voltage sensor calibration process: comparison of input voltage vs measured voltage.

3.2.4. Environmental Variables: Sht3x Sensor

The SHT31 is a combined humidity and temperature sensor specially designed for
outdoor applications. Recent uses can be found in the building of monitoring applica-
tions [58,59]. Its performance characteristics allow the sensor for outdoor applications.
It can work on 3.3 V and 5 V systems with very low power demand, offering fast and
accurate measurement via I2C bus. The I2C class found in the micropython machine library
was included to use this sensor, in addition to the SHT3X.py library available in [60]. The
developed script can be found in the Appendix B. The irradiance is measured by a reference
solar panel (W/m2 ). This small PV module of 300 mW is directly connected with a shunt
resistor for sampling the voltage generated by the sunlight—68 × 37 mm, mono-crystalline
cells; 19% efficiency, 5 V, 60 mA. The analog signal is directly dependent on the irradiance.
Further information can be found in [61]. It is calibrated using a pyranometer located close
to the plate and in the same position, being then possible to relate the current generated by
the reference plate and the available solar irradiance, see Figure 6.

Figure 6. Solar panel calibration:lLaboratory environment.

Sensors 2022, 22, 4966 11 of 20

3.3. Economic Evaluation: Cost-Effectiveness

Finally, and in terms of cost-effectiveness, the proposed monitoring system is in line
with other contributions discussed in Sections 1 and 2. The proposed system is flexible to
be configured in different locations and PV installations. With the aim of offering a low-cost
system, the different hardware components are selected from open-source projects with a
high cost effectiveness threshold. Table 4 gives the monitoring node cost, which is lower
than other commercial solutions but higher that previous developments carried out by the
authors [22,23]. Nevertheless, the global cost could be reduced by using other open-source
platforms—such as Arduino—and depending on the number of nodes to be produced
and/or purchased.

Table 4. PV monitoring node cost.

Description Unit Price (Euro)

Mainboard: FiPy 59.4
Sensor: Irradiance 2.45
Sensor: PV DS1820 temperature sensor 1.5
Sensor: DHT temperature and humidity sensor 0.18
Sensor: Voltage sensors (DC) 18.4
Sensor: Current (DC) 17.5
Supply: PV module 9.95
Supply: Battery 6.9
Supply: DC/DC converter 7.9
Supply: DC Booster (24 V) 0.18
Outdoor enclosure and wiring 12
Total cost 136.36

4. Use Case Description

With the aim of evaluating the capability of the proposed solution in terms of measur-
ing, collecting and transmitting data, a preliminary prototype with na FiPy was developed
by using a protoboard. Some additional modifications and requirements were detected dur-
ing this evaluation process. As an example, the inclusion of a potential divider to measure
the voltage sensor output, as well as the necessity of increasing an additional input voltage
for this sensor were considered, see Figure 7. After the preliminary assembly was tested,
a first printed circuit board prototype was designed and milled to minimize the number
of wires required by the system by using a computer numerical control (CNC) milling
machine. In this case, the printed circuit board (PCB) design was carried out by using
Fritzing [62]. It is an electronic design software including libraries for different elements.
Flatcam was also used to create the GCode, finally used on the CNC machine that milled
the PCB. Different versions were proposed and tested to improve different aspects of the
design, mainly reducing sizes and elements required for the solution. Figure 8 shows the
connection diagram schematically.
Electronic components need to be protected from moisture, direct sun radiation, dust
and other external effects that could affect their normal operation. Moreover, to avoid
any interference in control signals and to isolate those components under high voltage
or intensity values, two different external boxes were designed using Onshape. It is a
free-to-use CAD platform in which assemblies can be created by using different items to
make a more complex design. Different recent applications can be found in the specific
literature [63].
With the aim of providing power to the node, a PV solar module (6.5 W rate power)
with a 2000 mAh Li-Ion Polymer battery was also included. The battery was selected to
ensure system operation at night or during periods when solar radiation is not enough to
provide the required energy under operating conditions. To ensure a relevant autonomy
of the system, the micro-controller goes into deep sleep at night and under low radiation
conditions. Figure 9 shows the prototype battery and power supply scheme.
Sensors 2022, 22, 4966 12 of 20

Figure 7. Prototype voltage and current sensors.

Figure 8. General scheme of the system electronics.

Figure 9. Power supply scheme of the system and prototype example.

5. Results
The proposed solution—both including hardware components and sensors—was
tested in a laboratory environment, aiming to assess its performance under different
Sensors 2022, 22, 4966 13 of 20

conditions. The solution was connected and assembled under outdoor conditions in a PV
self-consumption installation to evaluate the solution feasibility to be implemented in a
variety of real situations and conditions. The proposed system was initially tested at the
solar laboratory of CETENMA, located in the Industrial Park of Cartagena (Spain). Such a
facility includes measurement equipment to check the performance of PV power plants
and modules. For testing purposes, a single 250 Wp monocrystalline PV module connected
to an SF 250 W Soltec SolarFighter microinverter was used—V MPP = 28.5 V, I MPP = 8.8 A,
Voc = 34.6 V, Isc = 9.4 A. Figure 10 shows some examples of field test campaigns and
Figure 11 shows results corresponding to this calibration process.

Figure 10. Case study. Example of field test campaign. Charging solar panel and reference cell
assembled to the monitored PV module panel.

Figure 11. Calibration of the reference cell: comparison of data for two days.

From the power supply scheme shown in Figure 9, Figure 12 depicts voltage evolution
and current demanded (positive) or supplied (negative) by the battery. The Pycom module
consumption is around 200 mA over WiFi and 250 mA when the LTE communication
protocol is selected. During the information sending process, up to 350 mA can be achieved.
Figure 13 shows a detail of Pycom and battery power demand comparison. Regarding the
received data through NB-Iot communications versus inverter data, voltage and current
have enough precision, 2.45% and 3.9%, respectively, to evaluate the operation of the
plant with security. A total of six packages were sent with one-minute sample time:
8 bytes/package for five packages and 10 bytes/package for the additional package. In
Sensors 2022, 22, 4966 14 of 20

terms of electrical collected data, Figures 14 and 15 show both voltage and current data
respectively. Even though the selected reference cells achieved suitable results during the
calibration process, the received data show divergences during some periods of the day.
These divergences are due to both material and design of the enclosure of the reference
cell that, after some time, does not ensure that it remains in the plane of installation with
modules of the plant, reducing the received radiation with low values of solar angles, see
Figure 16. A new design for this enclosure is currently in progress. In terms of temperature
gradients, Figure 17 compares PV module temperature to ambient temperature These data
give additional information regarding PV operating temperature, which can be useful to
estimate other parameters, such as the performance ratio. In addition, it is also possible to
analyze the effect of irradiation and ambient temperature on PV system performance, as
was suggested by other authors [64,65].

Figure 12. Example of battery charge and discharge time periods.

Figure 13. Examples of Pycom demand consumption and battery power monitoring: (a,b) discharging
battery period; (c,d) charging and discharging battery period.
Sensors 2022, 22, 4966 15 of 20

Figure 14. Example of collected voltage data: PV module monitored.

Figure 15. Example of collected current data: PV module monitored.

Figure 16. Example of collected data received in-plane irradiance.

Sensors 2022, 22, 4966 16 of 20

Figure 17. Comparison of PV module temperature vs. ambient temperature.

Regarding communication quality of the proposed solution, additional metrics used in

previous works of the authors [22] are also discussed because no data of works with similar
systems have been found to compare the results. These metrics are Packet delivery ratio,
defined as the ratio between the packets successfully received and the total data packets
sent by the end nodes and time intervals between different data packets (inter-arrival time)
that is determined by the time interval value corresponding to each packet received. In
the previous work that used Lorawan communications [22], the payload had a length of
38 bytes, almost equal size of this work, where the size of the sent string was 37 bytes,
4 × 5 floating bytes (measurements of each sensor) and 17 bytes of temporary fingerprint.
With these similar conditions, our study reveals that the proposed system provides a
reliable connectivity with a packet delivery error around 4.6% and stable time interval
between packets of 60 s. These results validate the feasibility and reliability of our proposal
improving the results obtained in previous works and overcoming limitations related to
missing data packets and frequency of received data. Figure 18 shows these results.

Figure 18. Communication relative errors: Average daily values.

6. Conclusions
The proliferation of PV installations either for self-consumption or with the aim of
alleviating dependency on the grid is moving the energy system to more decentralized
power generation. This new scenario allows users (who now are prosumers) to generate
and/or consume their own energy. The lack of information about these PV systems emerge
Sensors 2022, 22, 4966 17 of 20

as a relevant challenge for the management of micro- and medium-size grids, as the
generation is more distributed. This paper describes a monitoring solution suitable to
be applied for self-consumption or any other micro-generation installation, covering the
installations of the so-called ’prosumers’ and aiming to providing them with a tool that
informs them about their local self-consumption. The proposed system allows us to monitor
both electrical and environmental variables. A prototype was calibrated and successfully
evaluated in a real PV self-consumption installation. Both current and voltage sensors were
calibrated, determining 3% and 4% relative error, respectively, under laboratory conditions.
The system includes a 6.5 W PV solar panel and a battery for energy requirements. Charge
and discharge battery cycles were also monitored and included in the paper. Regarding the
received data through NB-IoT communications versus inverter data, voltage and current
have enough precision, 2.45% and 3.9%, respectively, to evaluate the operation of the plant
with security. In addition, the proposed system provides a reliable connectivity with a
packet delivery error around 4.6% and stable time interval between packets of 60 s. The
size of the string was 37 bytes: 17 bytes of temporary fingerprint, 4 × 5 floating bytes.
This solution is also able to be implemented in large PV power plants, as well as other
alternative renewable installations.

Author Contributions: Conceptualization, J.M.P.-P. and A.M.-G.; methodology, A.P.R.-G. and A.M.-
A.; software, R.J.-S. and D.C.-P.; formal analysis, R.J.-S. and J.M.P.-P.; resources, A.P.R.-G. and A.M.-A.;
data curation, R.J.-S. and D.C.-P.; writing—original draft preparation, J.M.P.-P. and A.M.-G.; writing—
review and editing, A.P.R.-G. and A.M.-G.; supervision, J.M.P.-P. and A.M.-G. All authors have read
and agreed to the published version of the manuscript.
Funding: This work was partially supported by the Spanish agreement (2021) between the Institute
for Development of the Region of Murcia (INFO) and the Technological Center for Energy and
Environment (CETENMA). The paper includes results of the activity conducted under the ’Research
Program for Groups of Scientific Excellence at Region of Murcia (Spain)’ Project PVBRAIN (grant
2021.08.CT01.0035). A.P.R-G would like to thank European Commission Horizon Project PHOENIX
(Project number: 893079).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data that support the findings of this study are available from the
corresponding author, A.M.-G., upon reasonable request.
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript:

CNC Computer numerical control

DSO Distribution System Operator
IEC International Electrotechnical Commission
IoT Internet-of-Things
NB Narrow band
LPWAN Low-power wireless area network
LTE Long-Term Evolution
MAC Media Access Control
PCB Printed circuit board
PV Photovoltaic
Quality of service QoS
RES Renewable Energy Sources
SET Plan Strategic Energy Technology Plan
TSO Transmission System Operator
Sensors 2022, 22, 4966 18 of 20

Appendix A. Sensor DS18X20: Python Code

from onewire import DS18X20 , OneWire
# Inicialization
ow = OneWire ( Pin ( ’ P11 ’ ) )
temp = DS18X20 (ow)
# Reading
Tmod = temp . read_temp_async ( )
temp . s t a r t _ c o n v e r s i o n ( )

Appendix B. Sensor SHT3X: Python Code

from machine import I2C
import SHT3X
# FiPy − I2C − Pin a s s i g n m e n t
i 2 c = I2C ( 0 , p i n s =( ’ P9 ’ , ’ P10 ’ ) )
i 2 c . i n i t ( I2C . MASTER, baudrate =20000)
# SHT3X
s h t 3 x = SHT3X . SHT3X ( bus_obj= i 2 c , address =0 x44 )
# Temp and Hum r e a d i n g
measurements = s h t 3 x . get_measurement ( )

References
1. Di Silvestre, M.L.; Favuzza, S.; Sanseverino, E.R.; Zizzo, G. How Decarbonization, Digitalization and Decentralization are
changing key power infrastructures. Renew. Sustain. Energy Rev. 2018, 93, 483–498. [CrossRef]
2. Cabrera, P.; Carta, J.A.; Lund, H.; Thellufsen, J.Z. Large-scale optimal integration of wind and solar photovoltaic power in
water-energy systems on islands. Energy Convers. Manag. 2021, 235, 113982. [CrossRef]
3. Heras-Saizarbitoria, I.; SÃaez,
˛ L.; Allur, E.; Morandeira, J. The emergence of renewable energy cooperatives in Spain: A review.
Renew. Sustain. Energy Rev. 2018, 94, 1036–1043. [CrossRef]
4. Rosales-Asensio, E.; de Simón-Martín, M.; Borge-Diez, D.; Pérez-Hoyos, A.; Comenar Santos, A. An expert judgement approach to
determine measures to remove institutional barriers and economic non-market failures that restrict photovoltaic self-consumption
deployment in Spain. Sol. Energy 2019, 180, 307–323. [CrossRef]
5. Mateo, C.; Cossent, R.; Gómez, T.; Prettico, G.; Frías, P.; Fulli, G.; Meletiou, A.; Postigo, F. Impact of solar PV self-consumption
policies on distribution networks and regulatory implications. Sol. Energy 2018, 176, 62–72. [CrossRef]
6. Infield, D.; Freris, L. Renewable Energy in Power Systems; John Wiley & Sons: Hoboken, NJ, USA, 2020.
7. Eikeland, P.O.; Skjærseth, J.B. The Politics of Low-Carbon Innovation; Springer International Publishing: Cham, Switzerland, 2020.
[CrossRef]
8. Repo, S.; Ponci, F.; Dede, A.; Della Giustina, D.; Cruz-Zambrano, M.; Al-Jassim, Z.; Amaris, H. Real-time distributed monitoring
and control system of MV and LV distribution network with large-scale distributed energy resources. In Proceedings of the 2016
IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), Ljubljana, Slovenia, 9–12 October 2016; pp. 1–6.
[CrossRef]
9. Paiho, S.; Kiljander, J.; Sarala, R.; Siikavirta, H.; Kilkki, O.; Bajpai, A.; Duchon, M.; Pahl, M.O.; Wüstrich, L.; Lübben, C.; et al.
Towards cross-commodity energy-sharing communities—A review of the market, regulatory, and technical situation. Renew.
Sustain. Energy Rev. 2021, 151, 111568. [CrossRef]
10. Haegermark, M.; Kovacs, P.; Dalenbäck, J.O. Economic feasibility of solar photovoltaic rooftop systems in a complex setting: A
Swedish case study. Energy 2017, 127, 18–29. [CrossRef]
11. Madeti, S.R.; Singh, S. Monitoring system for photovoltaic plants: A review. Renew. Sustain. Energy Rev. 2017, 67, 1180–1207.
[CrossRef]
12. Rahman, M.; Selvaraj, J.; Rahim, N.; Hasanuzzaman, M. Global modern monitoring systems for PV based power generation: A
review. Renew. Sustain. Energy Rev. 2018, 82, 4142–4158. [CrossRef]
13. Triki-Lahiani, A.; Bennani-Ben Abdelghani, A.; Slama-Belkhodja, I. Fault detection and monitoring systems for photovoltaic
installations: A review. Renew. Sustain. Energy Rev. 2018, 82, 2680–2692. [CrossRef]
14. Ansari, S.; Ayob, A.; Lipu, M.S.H.; Saad, M.H.M.; Hussain, A. A Review of Monitoring Technologies for Solar PV Systems
Using Data Processing Modules and Transmission Protocols: Progress, Challenges and Prospects. Sustainability 2021, 13, 8120.
[CrossRef]
15. Mellit, A.; Kalogirou, S. Artificial intelligence and internet of things to improve efficacy of diagnosis and remote sensing of
solar photovoltaic systems: Challenges, recommendations and future directions. Renew. Sustain. Energy Rev. 2021, 143, 110889.
[CrossRef]
Sensors 2022, 22, 4966 19 of 20

16. Rus-Casas, C.; Jiménez-Castillo, G.; Aguilar-Peña, J.D.; Fernández-Carrasco, J.I.; Muñoz-Rodríguez, F.J. Development of a
Prototype for Monitoring Photovoltaic Self-Consumption Systems. Electronics 2020, 9, 67. [CrossRef]
17. Sunarso, A.; Andriani, D.; Iswanda, D.; Nurhaidah, N.; Pratomo, T.; Sihombing, H.; Muharram, Y.; Yuniarto, W.; Rusman, R.
Assessment of Pv Potential Using a Low-Cost Monitoring System. SSRN Electron. J. 2022. [CrossRef]
18. Fuentes, M.; Vivar, M.; Burgos, J.; Aguilera, J.; Vacas, J. Design of an accurate, low-cost autonomous data logger for PV system
monitoring using ArduinoTM that complies with IEC standards. Sol. Energy Mater. Sol. Cells 2014, 130, 529–543. [CrossRef]
19. López-Vargas, A.; Fuentes, M.; García, M.V.; Muñoz-Rodríguez, F.J. Low-Cost Datalogger Intended for Remote Monitoring of
Solar Photovoltaic Standalone Systems Based on Arduino. IEEE Sens. J. 2019, 19, 4308–4320. [CrossRef]
20. López-Vargas, A.; Fuentes, M.; Vivar, M. IoT Application for Real-Time Monitoring of Solar Home Systems Based on Arduino
With 3G Connectivity. IEEE Sens. J. 2019, 19, 679–691. [CrossRef]
21. Zorbas, D.; Caillouet, C.; Abdelfadeel Hassan, K.; Pesch, D. Optimal Data Collection Time in LoRa Networks—A Time-Slotted
Approach. Sensors 2021, 21, 1193. [CrossRef]
22. Paredes-Parra, J.; Mateo-Aroca, A.; Silvente-Niñirola, G.; Bueso, M.; Molina-García, A. PV module monitoring system based on
low-cost solutions: Wireless raspberry application and assessment. Energies 2018, 11, 3051. [CrossRef]
23. Paredes-Parra, J.; García-Sánchez, A.; Mateo-Aroca, A.; Molina-García, Á. An alternative internet-of-things solution based on
LOra for PV power plants: Data monitoring and management. Energies 2019, 12, 881. [CrossRef]
24. Fuentes, A.F.; Tamura, E. LoRa-Based IoT Data Monitoring and Collecting Platform. In Smart Cities; Nesmachnow, S., Hernán-
dez Callejo, L., Eds.; Springer International Publishing: Cham, Switzerland, 2020; pp. 80–92.
25. Bueso, M.C.; Paredes-Parra, J.M.; Mateo-Aroca, A.; Molina-García, A. Sensitive Parameter Analysis for Solar Irradiance Short-Term
Forecasting: Application to LoRa-Based Monitoring Technology. Sensors 2022, 22, 1499. [CrossRef] [PubMed]
26. Gautier, A.; Hoet, B.; Jacqmin, J.; Van Driessche, S. Self-consumption choice of residential PV owners under net-metering. Energy
Policy 2019, 128, 648–653. [CrossRef]
27. Voisin-Grall, A.; Malaolu, O.O.; Zhu, Y.; Ahmed, T.; Al-Ahmed, S.A.; Shakir, M.Z. Remote Condition Monitoring: A Prototype
Based on Pycom Development Board FiPy and Pysense. In Proceedings of the 2019 UK/China Emerging Technologies (UCET),
Glasgow, UK, 21–22 August 2019; pp. 1–6. [CrossRef]
28. Sornin, N.; Luis, M.; Eirich, T.; Kramp, T.; Hersent, O. LoRa Alliance, LPWA Technologies Unlock New IoT Market Potential.
LoRa Alliance. 2015. Available online: https://docplayer.net/23382599-Lpwa-technologies-unlock-new-iot-market-potential-a-
white-paper-prepared-for-the-lora-alliance.html (accessed on 18 May 2022).
29. Sultania, A.K.; Mahfoudhi, F.; Famaey, J. Real-Time Demand Response Using NB-IoT. IEEE Internet Things J. 2020, 7, 11863–11872.
[CrossRef]
30. Sinha, R.S.; Wei, Y.; Hwang, S.H. A survey on LPWA technology: LoRa and NB-IoT. ICT Express 2017, 3, 14–21. [CrossRef]
31. Lombardo, A.; Parrino, S.; Peruzzi, G.; Pozzebon, A. LoRaWAN Versus NB-IoT: Transmission Performance Analysis within
Critical Environments. IEEE Internet Things J. 2022, 9, 1068–1081. [CrossRef]
32. Leenders, G.; Callebaut, G.; Ottoy, G.; Van der Perre, L.; De Strycker, L. Multi-RAT for IoT: The Potential in Combining LoRaWAN
and NB-IoT. IEEE Commun. Mag. 2021, 59, 98–104. [CrossRef]
33. Peruzzi, G.; Pozzebon, A. Combining LoRaWAN and NB-IoT for Edge-to-Cloud Low Power Connectivity Leveraging on Fog
Computing. Appl. Sci. 2022, 12, 1497. [CrossRef]
34. Coman, F.L.; Malarski, K.M.; Petersen, M.N.; Ruepp, S. Security Issues in Internet of Things: Vulnerability Analysis of LoRaWAN,
Sigfox and NB-IoT. In Proceedings of the 2019 Global IoT Summit (GIoTS), Aarhus, Denmark, 17–21 June 2019; pp. 1–6. [CrossRef]
35. Dangana, M.; Ansari, S.; Abbasi, Q.H.; Hussain, S.; Imran, M.A. Suitability of NB-IoT for Indoor Industrial Environment: A
Survey and Insights. Sensors 2021, 21, 5284. [CrossRef]
36. Al-Sarawi, S.; Anbar, M.; Abdullah, R.; Al Hawari, A.B. Internet of Things Market Analysis Forecasts, 2020–2030. In Proceedings
of the 2020 Fourth World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4), London, UK, 27–28 July
2020; pp. 449–453. [CrossRef]
37. Andrade, R.O.; Yoo, S.G. A Comprehensive Study of the Use of LoRa in the Development of Smart Cities. Appl. Sci. 2019, 9, 4753.
[CrossRef]
38. Lalle, Y.; Fourati, L.C.; Fourati, M.; Barraca, J.P. A Comparative Study of LoRaWAN, SigFox, and NB-IoT for Smart Water Grid. In
Proceedings of the 2019 Global Information Infrastructure and Networking Symposium (GIIS), Paris, France, 18–20 December
2019; pp. 1–6. [CrossRef]
39. Huan, J.; Li, H.; Wu, F.; Cao, W. Design of water quality monitoring system for aquaculture ponds based on NB-IoT. Aquac. Eng.
2020, 90, 102088. [CrossRef]
40. Cardoso, J.; Gloria, A.; Sebastiao, P. A Methodology for Sustainable Farming Irrigation using WSN, NB-IoT and Machine Learning.
In Proceedings of the 2020 5th South-East Europe Design Automation, Computer Engineering, Computer Networks and Social
Media Conference (SEEDA-CECNSM), Corfu, Greece, 25–27 September 2020; pp. 1–6. [CrossRef]
41. Li, Y.; Cheng, X.; Cao, Y.; Wang, D.; Yang, L. Smart Choice for the Smart Grid: Narrowband Internet of Things (NB-IoT). IEEE
Internet Things J. 2018, 5, 1505–1515. [CrossRef]
42. Petrenko, A.S.; Petrenko, S.A.; Makoveichuk, K.A.; Chetyrbok, P.V. The IIoT/IoT device control model based on narrow-band IoT
(NB-IoT). In Proceedings of the 2018 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering
(EIConRus), Moscow and St. Petersburg, Russia, 29 January–1 February 2018; pp. 950–953. [CrossRef]
Sensors 2022, 22, 4966 20 of 20

43. Ballerini, M.; Polonelli, T.; Brunelli, D.; Magno, M.; Benini, L. NB-IoT Versus LoRaWAN: An Experimental Evaluation for
Industrial Applications. IEEE Trans. Ind. Inform. 2020, 16, 7802–7811. [CrossRef]
44. Tiwari, A.; Pindoriya, N.M. Automated Demand Response in Smart Distribution Grid: A Review on Metering Infrastructure,
Communication Technology and Optimization Models. Electr. Power Syst. Res. 2022, 206, 107835. [CrossRef]
45. Pereira, R.I.; Jucá, S.C.; Carvalho, P.C. IoT embedded systems network and sensors signal conditioning applied to decentralized
photovoltaic plants. Measurement 2019, 142, 195–212. [CrossRef]
46. Ma, D.; Xie, X.; Chang, Y. Low Voltage Distributed Photovoltaic Power Station Connected Estimation Model Based on Dispatch
Automation System. IOP Conf. Ser. Earth Environ. Sci. 2020, 440, 032028. [CrossRef]
47. Emamian, M.; Eskandari, A.; Aghaei, M.; Nedaei, A.; Sizkouhi, A.M.; Milimonfared, J. Cloud Computing and IoT Based
Intelligent Monitoring System for Photovoltaic Plants Using Machine Learning Techniques. Energies 2022, 15, 3014. [CrossRef]
48. Ciocia, A.; Amato, A.; Di Leo, P.; Fichera, S.; Malgaroli, G.; Spertino, F.; Tzanova, S. Self-Consumption and Self-Sufficiency in
Photovoltaic Systems: Effect of Grid Limitation and Storage Installation. Energies 2021, 14, 1591. [CrossRef]
49. Pycom Ltd. FiPy 1.0 Datasheet; Guildford: Surrey, UK, 2017; pp. 1–3. Available online: https://pycom.io/wp-content/
uploads/2018/08/fipySpecsheetAugust2017n2-1.pdf (accessed on 18 May 2022).
50. Bell, C. MicroPython for the Internet of Things; Springer: Berlin/Heidelberg, Germany, 2017.
51. Dias, J.P.; Restivo, A.; Ferreira, H.S. Designing and constructing internet-of-Things systems: An overview of the ecosystem.
Internet Things 2022, 19, 100529. [CrossRef]
52. Pietrosemoli, E.; Rainone, M.; Zennaro, M. On extending the wireless communications range of weather stations using LoRaWAN.
In Proceedings of the 5th EAI International Conference on Smart Objects and Technologies for Social Good, Valencia, Spain, 25–27
September 2019; pp. 78–83.
53. Klise, K.A.; Stein, J.S.; Cunningham, J. Application of IEC 61724 Standards to Analyze PV System Performance in Different
Climates. In Proceedings of the 2017 IEEE 44th Photovoltaic Specialist Conference (PVSC), Washington, DC, USA, 25–30 June
2017; pp. 3161–3166. [CrossRef]
54. Karsenty, A. A Comprehensive Review of Integrated Hall Effects in Macro–, Micro–, Nanoscales, and Quantum Devices. Sensors
2020, 15, 4163. [CrossRef]
55. Nurazizah, E.; Ramdhani, M.; Rizal, A. Rancang Bangun Termometer Digital Berbasis Sensor DS18B20 Untuk Penyandang
Tunanetra. eProc. Eng. 2017, 4, 3294–3301.
56. Jovanovic, U.; Jovanovic, I.; Mancic, D. Overview of Temperature Sensors for Temperature Measurement of PV Modules. In
Proceedings of the 2018 26th Telecommunications Forum (TELFOR), Belgrade, Serbia, 20–21 November 2018; pp. 1–8. [CrossRef]
57. pycom-Open Source Libraries. Available online: https://github.com/pycom/pycom-libraries/blob/master/examples/DS18X2
0/onewire.py (accessed on 27 April 2022).
58. Frei, M.; Deb, C.; Stadler, R.; Nagy, Z.; Schlueter, A. Wireless sensor network for estimating building performance. Autom. Constr.
2020, 111, 103043. [CrossRef]
59. Kuzmenkov, A.; Kuvshinov, D.; Buryachenko, S.Y.; Kaychenov, A.; Karachentseva, I.; Voronin, Z. Monitoring system for
temperature and relative humidity of the experimental building. J. Phys. Conf. Ser. 2021, 2131, 052070. [CrossRef]
60. SHT3X.py-Open Source Libraries. Available online: https://github.com/dvsu/Sensirion-SHT3X-MicroPython/blob/main/sht3
x.py (accessed on 27 April 2022).
61. Orsetti, C.; Muttillo, M.; Parente, F.; Pantoli, L.; Stornelli, V.; Ferri, G. Reliable and Inexpensive Solar Irradiance Measurement
System Design. Procedia Eng. 2016, 168, 1767–1770. [CrossRef]
62. Knörig, A.; Wettach, R.; Cohen, J. Fritzing—A tool for advancing electronic prototyping for designers. In Proceedings of the 3rd
International Conference on Tangible and Embedded Interaction 2009, Cambridge, UK, 16–18 February 2019.
63. Cunis, T.; Bronz, M. EDURA: An Evolvable Demonstrator for Upset Recovery Approaches with a 3D-printed Launcher. In
Proceedings of the 9th International Micro Air Vehicle Conference and Competition, Toulouse, France, 18–21 September 2017.
64. Dubey, S.; Sarvaiya, J.N.; Seshadri, B. Temperature Dependent Photovoltaic (PV) Efficiency and Its Effect on PV Production in the
World—A Review. Energy Procedia 2013, 33, 311–321. [CrossRef]
65. Eke, R.; Kavasoglu, A.S.; Kavasoglu, N. Design and implementation of a low-cost multi-channel temperature measurement
system for photovoltaic modules. Measurement 2012, 45, 1499–1509. [CrossRef]