Samarth Rural Educational Institute

SAMARTH COLLEGE OF ENGINEERING, BELHE

A
Seminar & Technical Communication
Presentation On

A Framework for IoT Based Appliance

Recognition in Smart Homes

Under Guidance Of

Prof. Borhade R. B.

By

Lamkhade Sanika Machhindra

Department of Computer Engineering

2024-2025
•CONTENT
Abstract
• Introduction
• Objectives
• Literature Survey
• Problem statement
• Existing system
• System Architecture
• Analysis
• Advantages
• Disadvantages
• Real time applications
• Results
• Conclusion
• Future scope
 Abstract

This project presents an innovative IoT-based framework for

appliance recognition and load monitoring in smart homes,
addressing the growing need for efficient energy management
solutions.
By leveraging machine learning algorithms, the system aims to
accurately classify appliances based on their power consumption
patterns, thereby providing insights into energy usage and
enhancing user convenience.
 Introduction

The increasing integration of IoT in smart homes has led to a

demand for efficient load monitoring and appliance recognition
systems. This project explores the development of a flexible
framework that utilizes machine learning for real-time appliance
recognition.

• Tracks energy consumption

• Improve user experience
• Support energy saving initiatives
 Objectives

1. Develop a Robust Framework

2. Implement Machine Learning Algorithms

3. Enhance User Interaction

4. Real-Time Monitoring
 Literature Survey
Sr. Paper Title Author Summary Advantages Disadvantages Year
No. Name
1. Internet of Things applications
as energy internet in
Ersan Kabalci,
Yasin Kabalci
In this paper, we have comprehensively
analyzed Internet of Things (IoT)
Monitor key component
Identify solution to
Complex
implementation
2019

smart grids and smart applications enabled for smart grids and problems
environments. smart environments investigate the
development of the EI based IoT
applications.

2. New Challenges in the Design

of Microgrid Systems:
Yanpeng Wu,
Peilin Xie
Microgrids (MGs), referred to as next-
generation power systems, are receiving
Increased energy
efficiency
High cost 2020

Communication Networks, considerable attention from both industry

Cyberattacks, and Resilience and academia. Integrated with (DERs)
and a variety of loads, MGs function as a
localized power grid that can be operated
independently or connected to utility grids.

3. Convergence and
interoperability for the energy
Ying Wu,
Yanpeng Wu
The Energy Internet is proposed to
enhance the collaborative utilization of
Reduce energy
wastage
Data model
incompatibility
2020

internet: From distributed renewable energy resources;

ubiquitous connection to enable a flexible, customer-engaged
distributed automation energy transaction network; and achieve
real-time balancing of supply and demand
 Problem Statement

Current load monitoring systems often lack reliability and

flexibility in real-time applications, making it challenging to
accurately identify appliances and their usage patterns.
Additionally, existing solutions may not effectively support
the classification of activities of daily living, limiting their
practical applicability in smart home environments.
 Existing System

Existing systems primarily focus on energy monitoring

without integrating advanced machine learning
techniques for appliance recognition. Many solutions are
hardware-intensive and do not provide a comprehensive
framework for data processing and analysis. Furthermore,
they often fail to address the need for real-time monitoring
and user interaction.
System Architecture
 System Architecture
The proposed architecture consists of five layers:

1. Physical Things Layer: Includes appliances and metering

devices for data acquisition. Appliances
2. Perception Layer: Responsible for collecting and processing data
from the physical layer. Sensors
3. Communication Network Layer: Facilitates data exchange
between devices and applications. Wifi, Zigbee
4. Middleware Layer: Manages data processing and storage. Cloud
computing
5. Application Layer: Hosts the appliance recognition module and
user interface.mobile
 Methodology

1. Framework Components-
Training Framework: feature extraction
Inference Framework: Real-time appliance recognition.
2. Data Collection-
Description of data acquisition methods
3. Feature Extraction-
Implementation of a novel feature extraction algorithm.
4. Classification Models
Utilization of three machine learning models:-
Feed-forward neural network
Long short-term memory (LSTM)
Support vector machine (SVM)
5. Handling Imbalanced Classes
Techniques applied to manage class imbalance during
classification.
6. Feature Importance Analysis
Method to assess the impact of selected features on model
prediction
 Analysis

The performance of the proposed system is evaluated based on

several metrics, including accuracy, precision, recall, and F1-score. A
sensitivity analysis is conducted to determine the impact of different
parameters on classification performance. The results indicate that
the system can effectively classify appliances and monitor energy
consumption, with accuracy rates exceeding 90% in controlled
environments.
• Accurate
• Precise
• Real-Time monitoring
 Advantages

1. Real-Time Monitoring
2. Flexibility
3. User-Friendly Interface
4. Appliance Recognition
 Disadvantages

1. Initial Setup Complexity

2. Data Quality Dependency
3. Cost of Smart Appliances
4. Complex Implementation
 • Real-Time Applications

1. Smart Home Energy Management

2. Automated Energy-Saving Solutions
3. Health Monitoring Systems
 Results

The proposed system achieved a classification accuracy of

approximately 95% in laboratory settings, demonstrating its
effectiveness in recognizing individual appliances. The flexibility of
the framework allows for the integration of additional datasets and
machine learning models, enhancing its applicability in real-world
scenarios.
 Conclusion

The proposed framework effectively integrates IoT technology for

appliance recognition in smart homes, enhancing energy
management. Demonstrated the ability to support multiple appliance
types with customizable features.
 Future Work

1. Real-World Implementation: Conduct trials in diverse home settings to

validate the framework under varying conditions.
2. User Interface Development: Develop a user-friendly interface to
facilitate the customization and operation of the appliance recognition
system.
3. Data Privacy and Security: Address concerns regarding data privacy
and propose solutions for secure data handling in smart homes.
 Reference
1] Y. Kabalci, E. Kabalci, S. Padmanaban, J. B. Holm-Nielsen, and F. Blaabjerg, ‘‘Internet of Things
applications as energy internet in smart grids and smart environments,’’ Electronics, vol. 8, no. 9, p.
972, Aug. 2019.
[2] S. Tan, Y. Wu, P. Xie, J. M. Guerrero, J. C. Vasquez, and A. Abusorrah, ‘‘New challenges in the
design of microgrid systems: Communication networks, cyberattacks, and resilience,’’ IEEE Electrific.
Mag., vol. 8, no. 4, pp. 98–106, Dec. 2020.
[3] Y. Wu, Y. Wu, J. M. Guerrero, J. C. Vasquez, E. J. Palacios-Garcia, andJ. Li, ‘‘Convergence and
interoperability for the energy internet: From ubiquitous connection to distributed automation,’’ IEEE
Ind. Electron. Mag., vol. 14, no. 4, pp. 91–105, Dec. 2020
[4] J. Leitao, P. Gil, B. Ribeiro, and A. Cardoso, ‘‘A survey on home energy management,’’ IEEE Access,
vol. 8, pp. 5699–5722, 2020.
[5] B. Hartono, S. P. Mursid, and S. Prajogo, ‘‘Review: Home energy management system in a smart
grid scheme to improve reliability of power systems,’’ IOP Conf. Ser., Earth Environ. Sci., vol. 105, Jan.
2018, Art. no. 012081.
[6] C. Liu, K. T. Chau, D. Wu, and S. Gao, ‘‘Opportunities and challenges of vehicle-to-home, vehicle-
to-vehicle, and vehicle-to-grid technologies,’’ Proc. IEEE, vol. 101, no. 11, pp. 2409–2427, Nov. 2013.
[7] M. A. Devlin and B. P. Hayes, ‘‘Non-intrusive load monitoring and classification of activities of daily
living using residential smart meter data,’’IEEE Trans. Consum. Electron., vol. 65, no. 3, pp. 339–348, Aug.
2019.
[8] X. Yuan, P. Han, Y. Duan, R. E. Alden, V. Rallabandi, and D. M. Ionel, ‘‘Residential electrical load
monitoring and modeling–state of the art andbfuture trends for smart homes and grids,’’ Electr. Power
Compon. Syst.,vol. 48, pp. 1–19, Nov. 2020.
[9] D. Burbano, ‘‘Intrusive and non-intrusive load monitoring (a survey) inference and learning approach,’’
Latin Amer. J. Comput., vol. 2, no. 1, p. 9, 2015.
[10] A. U. Rehman, T. T. Lie, B. Vallès, and S. Tito, ‘‘Low complexity event detection algorithm for non
intrusive load monitoring systems,’’ in Proc. IEEE Innov. Smart Grid Technol.-Asia (ISGT Asia), May 2018, p.
751.
[11] A. U. Rehman, T. T. Lie, B. Valles, and S. R. Tito, ‘‘Event-detection algorithms for low sampling
nonintrusive load monitoring systems based on low complexity statistical features,’’ IEEE Trans. Instrum.
Meas., vol. 69, no. 3, pp. 751–759, Mar. 2020.
[12] R.-C. Mihailescu, D. Hurtig, and C. Olsson, ‘‘End-to-end anytime solution for appliance recognition
based on high-resolution current sensing with few-shot learning,’’ Internet Things, vol. 11, Sep. 2020, Art.
no. 100263.
[13] F. Paradiso, F. Paganelli, A. Luchetta, D. Giuli, and P. Castrogiovanni,‘‘ANN-based appliance recognition
from low-frequency energy monitoring data,’’ in Proc. IEEE 14th Int. Symp. World Wireless, Mobile
Multimedia Netw. (WoWMoM), Madrid, Spain, Jun. 2013, pp. 1–6.