SMART ENERGY METER DESIGN BASED ON DYNAMIC PRICING
Fayza Putri Safira1
1
Fakultas Pendidikan Teknologi dan Kejuruan, Universitas Pendidikan Indonesia (UPI), Indonesia (E-mail:
Fayza.putri@upi.
Abstract
In this research, design of a smart energy
metering device using dynamic pricing method
with a microcontroller is proposed. An Arduino
UNO board is used as the system’s main data
processing device. The purpose of this research is
to design a demand-side management system
which is aplied through the proposed smart meter
device. The system is consisted of supply side
meter device, load side meter device, an MQTT
server, and telemetry devices. This research is
done through literatural and devices datasheets
study which is then used as the base knowledge
for designing the system data flow and algorithm.
The system designed is then go through several
simulation and actual test repetition until the test
resulting a satisfactory output. The said test result
showed that the meter device is able to read and
charge energy usage, display the readings,
process the dynamic pricing method, and display
reading values through the telemetry devices with
relative error value averaging at 4 %. Some of the
major concerns since the implementation of smart
meters (prepaid meters) in some parts of Ghana is
how electricity consumers have benefited from
data obtained from these meters by providing
important statistics on electricity-saving advice;
this is one of the key demand-side management
methods for achieving load reduction in
residential homes. Appliance shifting techniques
have proved to be an effective demand response
strategy in load reduction. The purpose of this
paper is therefore to help consumers of
electricity understand when and how they can
shift some appliances from peak to off-
peak and vice versa.
Keywords : Smart meter, dynamic pricing,
microcontroller
I. INTRODUCTION
In the world of electricity, we recognize the
grid system as an energy distribution system. In its
development, the grid system used in its development,
the grid system used has taken a variety of new
technology applications with a variety of automation
systems in it. However, the overall energy distribution
system that is usually carried out still moves
unidirectionally (Bari, et al. 2014). The electricity
system that we know in general has two parties
involved in the distribution of electrical energy,
namely the providers and users of electrical energy.
Control of electrical energy in this grid system works
in one direction where the controller acts as a full
system controller.
On the electrical energy generation side, to
achieve good energy generation cost effectiveness,
energy generation with the lowest generation cost is
carried out as a plant that serves the load at low demand
times to the use of plants with the highest generation
cost at high demand times. The coordination of the use
of power plants commonly referred to as the economic
dispatch process aims to reduce the total operational
cost of electricity generation (Syafii & Putri, 2018).
On the consumer or user side of electrical
energy, the level of energy demand must be fully met
1
by the energy provider. The problem arises when the through simulation trials with software. The final result
demand for energy use is high or commonly referred to of the energy meter design was then tested with real
as peak load. To meet the needs at this peak load, the devices as the final product of the research. The energy
electrical energy provider must have a generator that is meter product was then compared with existing
able to provide energy at least as large as the maximum commercial meters to test its workability and compiled
load plus energy distribution and transmission losses as the final research report.
(Wahyuda & Budi 2015). However, with the state of
peak load that takes place at In a relatively short
timeframe, this large investment to meet peak load
requirements can be considered ineffective because
most of the time This energy facility is not fully
utilized. Therefore, it is necessary to have a system that
regulates the use of load on the consumer side to
increase the ratio of average load to peak load, in order
to increase the efficiency of investment in electric
energy generation.
The modernized grid system or better known
as the smart grid is an automated electrical energy
distribution network built based on the objectives of
increased efficiency, improved reliability, economy,
and security. The Smart Grid combines distributed
computing and communication capabilities to create an
electricity system with a near real-time balance of
supply and demand for electrical energy. In this
system, energy providers can incentivize users who
reduce energy use during peak load times, and can be
a way of load-side management to facilitate balancing
of energy generation. With the communication and
computational capabilities of the Smart Grid system,
the demand-side management process can be carried
out by designing consumer meter devices that are able
to know the condition of other loads connected to the
same network (Bari, et al. 2014).
I. RESEARCH METHOD
This research was conducted with a literature
study and device datasheet, then continued with the Figure 1 Flowchart of the research
creation of process algorithms and data flow and trial I. ANALYSIS AND DESIGN
application on the device until satisfactory results were
found based on testing. The purpose of this research is The designed system is divided into four
to design a smart energy meter with the ability to read function blocks that work together, as shown
power usage, system voltage, current, power factor and in Figure III 2. The server and client blocks
be able to calculate the costs charged to consumers in are blocks that contain meter devices placed
different load populations. The research began with a at the source and load. The server block
literature study on the general purpose of Smart Grid, placed on the source side functions to create a
the working principle of energy meters, price level value by reading and calculating
microcontrollers and the MQTT communication the connected sensors, then sending it to the
protocol. The study results obtained serve as the MQTT server block. The client block is the
author's basis for designing energy meters in main meter device that measures the use of the
accordance with the research objectives. The energy connected electrical load and calculates the
meter design is then implemented and modified price based on the energy used and the price

2
 level received from the MQTT server block.
The client block also functions to send the
reading data to the MQTT server block for
telemetry functions. The MQTT server block
serves as the main line of communication for
the system which is a server service of
PubNub. The telemetry device serves to
remotely display the results of sensor readings
from client blocks and can be configured to
work on various media such as smartphones
or personal computers.
Figure 2 The system workflow
The designed system is divided into four function
blocks that work together, as shown in Figure 2. The
server and client blocks are blocks containing meter
devices placed at the source and load. The server block
placed on the source side functions to create a price
level value by reading and calculating the connected
sensors, then sending it to the MQTT server block. The
client block is the main meter device that measures the
use of the connected electrical load and calculates the
price based on the energy used and the price level
received from the MQTT server block. The client block
also functions to send the reading data to the MQTT
server block for telemetry functions. The MQTT server
block serves as the main line of system communication
which is a server service from PubNub. The telemetry
device serves to remotely display sensor readings from
the client block and can be configured to work on
various media such as smartphones or personal
computers.
III.1. CURRENT SENSOR DESIGN
The current sensor device used in this study is a current
transformer device manufactured by Talema Group
with product code AC1050. The transformer has a coil
ratio of 1000: 1 so that for every 1 A of current through
the transformer device, a current of 1 mA will be
generated.
Figure 3. Current sensor circuit
The transformer output in the form of current
is then converted into voltage in accordance with the
input rating received by the ADC channel on Arduino,
which is 0-5 V by connecting the load resistor. With
consideration of the accuracy of the reading on the
Arduino ADC channel of 10-bit, the input current is
limited to an RMS value of 10 A with a maximum input
I value of:
I_imax=I_rms √2
I_imax=10√2
I_imax=14.142 A
To eliminate negative elements at the output,
the current sensor circuit is given a DC offset of half
the ADC input rating of 2.5 V. This DC offset of 2.5 V
is obtained by using the 5V channel on the Arduino
with an additional voltage divider circuit. To reduce the
use of power on the Arduino, a relatively large
resistance in the voltage divider circuit is selected,
namely 470 kΩ. The final circuit of the current sensor
interface device is shown in Figure 3.
III.2. VOLTAGE SENSOR DESIGN
The voltage sensor device on the designed energy
measuring instrument functions to provide voltage
adjustment from the voltage source that is read with the
voltage that can be read on the Arduino ADC channel.
The Arduino ADC channel is capable of reading
voltages of 0-5 V and converting them into digital data.
To get a maximum voltage of 5 V from the source
voltage, it is done by connecting a voltage reducing
transformer and connected to a voltage reducing
circuit.
3

Figure 4. Voltage sensor circuit
III.3. FERIFER DEVICE DESIGN
In the smart meter device designed, peripheral
devices are needed as a medium for visualizing digital
data received and processed on the MCU. The data that
needs to be visualized includes the reading values of
voltage, current, real power, apparent power, power
factor, and frequency. The display device used in this
research is a 16×2 character STN-LCD (Super Twisted
Nematic - Liquid Crystal Display) device that is not
able to display all the data that needs to be visualized
so it requires a way to arrange pages that display certain
data or groups of data.
To be able to set or select pages to display, an input
device in the form of a keypad is required. In addition
to the selection of the display page, the selected keypad
is also used to input data to make changes to the KWh
deposit stored on the smart meter device. With this
consideration, the keypad device chosen is a matrix
keypad with 4×3 buttons or 12 buttons consisting of
numbers (0-9), the symbol "*" and the symbol "#". The
following is the key map (keymap) used on the keypad
device.
III.4. DESIGN OF COMMUNICATION DEVICES
Gambar 1 Alur data proses dynamic pricing
In addition to pricing data transactions,
communication is also needed for telemetry or remote
measurement features. Telemetry is a control feature
that allows users of electrical energy services to read or
know the electrical conditions of installations
connected to electrical energy supply services. The
telemetry application designed is targeted at users of
electrical energy with data access including voltage,
current, real power, apparent power, power factor,
frequency, and electricity rates, all of which can be
monitored in real-time.
To be able to apply these two features, the author
chose an Internet communication method that
implements the MQTT communication protocol. The
MQTT communication protocol allows lighter and
faster data access than the commonly used HTTP
protocol. The wide coverage of Internet
communication allows telemetry over very long
distances. The communication hardware chosen is an
ESP8266 Wi-Fi device that can be programmed and set
up to process data and manage all data transactions
carried out via the internet so that it does not interfere
much with the number of sensor sample records by
MCUs that have limited working frequencies. The
server service (Broker) of PubNub is used as a
communication facilitator and debugging media for
data transactions.
III.5. TESTS
After the designed meter device was successfully
simulated in the ISIS Proteus 8.0 program, the research
continued with the realization of the meter device
based on the design that had been made. Then function
testing was carried out meter device,after All device
functions are declared to be running properly, the test
4
continues by testing the accuracy of the readings by current, power, power
comparing the readings from the smart meter device factor, and frequency.
with other measuring devices.
This test is conducted to test the ability of the 2 The server device is OK
smart meter device to read and calculate the value able to process energy
detected on the sensor and show the results on the usage on its network into
display object based on the page settings that have been price variables as the basis
previously set. The results of testing the sensor reading for the dynamic pricing
function of the smart meter device can be stated that process and send the data to
the smart meter device designed is able to read and
the client device.
show the readings of voltage, current, power, power
factor, and frequency.
3 The client device is OK
Testing the dynamic pricing function is done by able to receive variable
testing the function of creating and sending price levels price data from the server
on the server side smart meter device as well as the device, process it into a
function of receiving price level data, creating prices
based on the price level data received, and calculating
new energy usage price and
the value to be paid on the smart meter device on the calculate energy usage
client side. This test is done by checking the display on based on the dynamic
the server and client meter devices. Based on the test pricing process.
results table, it can be stated that the server device is
able to process energy usage on its network into price 4 The client device OK
variables as the basis for the dynamic pricing process allows telemetry or remote
and send the data to the client device, and the client monitoring of other devices
device is able to receive price variable data from the located at a distance.
server device, process it into a new energy usage price
and calculate energy usage based on the dynamic
pricing process.
From the previous design and testing, we can see
that the smart meter device designed is able to detect
the value of electrical element readings, namely
I.KESIMPULAN
voltage, current, real power, apparent power, power In this research, the dynamic pricing-based smart
factor and frequency, in a simple loading circuit, meter device designed using Arduino as MCU
perform pricing calculations based on dynamic pricing successfully reads energy usage and shows the results
principles that have been designed, and display the on the display device as the basic function of
results on the display device via LED display and conventional energy usage measurement tools. The
telemetry device. Based on the device testing check-list smart meter device designed also successfully applies
table in Table 1, the test results of the smart meter the dynamic pricing process which is the basis for
device design are as follows: designing the device, and allows remote monitoring via
Table 1. Check-list of device testing results the internet. The designed smart meter is successfully
simulated using ISIS Proteus 8 software and can
No Requirement perform all planned functions properly. Similarly, in
the real assembled device, all functions that have been
Completion
designed can work well. However, in the real circuit,
the reading results on the smart meter device show
1 The smart meter device OK error values that, at some reading values, are too high
is able to read and show the to be applied in real applications.
readings of voltage,

5
 REFERENSI
Alexander, C. K., & Sadiku, M. N. O. (n.d.).
Fundamental of Electric Circuit.
Amin, S. M. (2011). Smart Grid: Overview, Issues and Sutrisno. (1986). Elektronika Teori dan Penerapannya
Opportunities. European Journal of Control Jilid 1. Bandung: Penerbit ITB.
(2011)5-6:547-567.
Syafii., Putri, K. I. (2018). Analisa Operasi Pembangkit
Bari, A., Jiang, J., Saad, W., Jaekel, A. (2014). Termal untuk Melayani Beban Puncak Sistem
Challenges in Smart Grid: an Overview. Kelistrikan Sumbar. Jurnal Nasional Teknik
International Journal of Distributed Sensor Elektro, Vol. 7, No.1.
Network, Volume 2014/974682.
U.S. Departement of Energy. (2015). Metering Best
Barnett, R. H., O’Cull, L., & Coxm S. (2007). Practices: A Guide to Achieving Utility
Embedded C Programming and the Atmel Resource Efficiency, Release 3.0. Virginia:
AVR. New York: Thomson Delmar Learning. U.S. Department of Commerce.
Enokela, J. A. (2007). A Comparison of Performances Wahyuda., Santosa, B. (2015). Dynamic Pricing in
of Electronic and Electromechanical Energy Electricity: Research Potential in Indonesia.
Meter. Nigerian Journal of Technology, Vol. Industrial Engineering and Service Science.
26, No. 2, June 2007.
Iqbal, H. M. Z., Waseem, M., Mahmood, T. (2014).
Automatic Energy Meter Reading using Smart
Energy Meter. International Conference on
Engineering and Emerging Technologies
(ICEET-2014).
Jithin, J. K., Leneesh, M., Nijeesh, U. K., Tony, C. B.
(2015). Smart Energy Meter. International
Journal of Engineering Trend and Technology
(IJETT) – Volume 22 Number 4.
Lander, C. W. (1993). Power Electronics. London:
McGraw Hill.
Lappeenrannan Energia. (2011). Smart Metering
Handbook. Jyska: Personal Book Publishing.
Liun, E., Sunardi. (2014). Perbandingan Harga Energi
Dari Sumber Energi Baru Terbarukan dan Fosil.
Jurnal Pengembangan Energi Nuklir Volume
16, Nomor 2.
Raharjo, B. (2009). Pemrograman C++. Bandung:
Informatika.
Rashid, M. H. (2001). Power Electronics Handbook.
Academic Press Series in Engineering