VIETNAM NATIONAL UNIVERSITY, HANOI INTERNATIONAL SCHOOL GRADUATION PROJECT RESEARCH AND DESIGN OF A SYSTEM FOR EARLY FIRE DETECTION USING MACHINE LEARNING ON INDOOR LABORATORY DATA TRẦN THẢO CHI Hanoi - 2024. VIETNAM NATIONAL UNIVERSITY, HANOI INTERNATIONAL SCHOOL GRADUATION PROJECT RESEARCH AND DESIGN OF A SYSTEM FOR EARLY FIRE DETECTION USING MACHINE LEARNING ON INDOOR LABORATORY DATA SUPERVISOR: Assoc. HA MANH HUNG STUDENT: TRAN THAO CHI STUDENT ID: 21070065 COHORT: QH-2021 SUBJECT CODE: INS4011 MAJOR: Business Data Analytics Hanoi - 2024. Abstract This thesis presents the research, design, and development of an early fire detection system leveraging machine learning algorithms on indoor laboratory data.
With a focus on enhancing fire safety, this work addresses the critical need for timely fire detection to mitigate property damage and loss of life. The proposed system integrates multiple sensors, including temperature, humidity, and gas sensors, to gather comprehensive environmental data. Controlled fire experiments were conducted to collect a robust dataset for the training and evaluation of machine learning models. Two primary algorithms, the Support Vector Machine (SVM) and Artificial Neural Network (ANN) were assessed for their effectiveness in predicting fire events.
The ANN model demonstrated an accuracy of 87%, making it a viable option for practical fire forecasting applications. The SVM, particularly when combined with Principal Component Analysis (PCA), showed potential for commercial applications due to its enhanced performance with PCA. The findings highlight the importance of sensor selection and fusion in building an effective fire detection system. The contributions of this thesis include a practical technique for early fire fore- casting using ANN, recommendations for sensor configurations, and insights into the performance of SVM and ANN models.
This research lays the groundwork for future advancements in fire detection technology, aiming to improve safety and response times in home environments. Keywords: early fire detection, machine learning, sensor data, Principal Component Analysis, Support Vector Machines, and Artificial Neural Networks. 1 Declaration This graduation thesis is my original work conducted during my time at Vietnam National University Hanoi. Credit has been given to all the sources and inspirations that have helped me along the way.
All the support and assistance I received during this research are gratefully mentioned. I hereby declare that the graduation thesis ”Research and design of a sys- tem for early Fire Detection using Machine Learning on Indoor Laboratory Data” are the results of my own research and have not been published in any other work. Throughout the implementation process of this project, I have adhered strictly to research ethics. All findings are derived from my own research and surveys, and all references are clearly cited in accordance with regulations.
I take full responsibility for the accuracy of the figures, data, and other content in my graduation project. Student Tran Thao Chi 2 Acknowledgements I would like to express my profound gratitude to my advisor, Dr. Ha Manh Hung, for his invaluable guidance, support, and encouragement throughout this research jour- ney. His insights and expertise have been crucial in shaping this thesis.
I am also pro- foundly grateful to my associate professors at International School, Vietnam National University Hanoi, for providing me with the expertise and resources necessary to pur- sue this research. Special thanks to the Faculty of Applied Sciences for their support and the Business Data Analytics program for equipping me with the skills needed for this project. A heartfelt thanks to my research supporter for his collaboration, dedication, and for making this journey both productive and enjoyable. Your camaraderie have been a source of motivation.
I would like to extend my heartfelt gratitude to my family and friends for their unwavering support and understanding throughout this research. I wish to acknowledge all the researchers and professionals whose work has inspired and informed this thesis. Your contributions to the field have been invaluable. 3 Contents Abstract 1 Declaration 2 Acknowledgements 3 1 Introduction 8 1.3 Approach and Methodology .4 Scope and Limitation .1 Fire detection technologies .1 Overview of Machine Learning .2 Support Vector Machine .3 Principal Component Analysis .4 Artificial Neural Network .3 Related works on Machine Learning approach.
23 3 Experiment and Result 26 3.1 The Indoor Laboratory Fire Dataset (ILFD) .2 The NIST Report of Test FR 4016 Manufactured Home Tests (NIST) .3 Experimental Household Data .7 PCA-SVM Implementation .9 Results on collected Experimental Household Data. 35 4 Conclusion and Future Work 36 4. 37 5 List of Figures 2.1 Inside view of smoke detectors, showing the chamber and electronic com- ponents[29] .2 (a) Distance from a point to a line in 2D space, (b) Distance from a point to a plane in 3D space. The margins are shown as dashed lines.4 The primary concept of PCA is to identify a new orthonormal system where the most significant components are preserved in the first K com- ponents [6] .5 Model of an artificial neuron labeled k[19] .6 The data are initially recorded from the onset of the fire until the fire alarm is triggered.
As the fire starts, the risk level progressively escalates from low to medium and then to high.1 Hardware Deployment setup for Experiments: (a) setup physically sit- uated in the middle of the laboratory room; (b) experimental setup showing sensors connected to the microcontroller motherboard and the LoRa Node wireless system connecting it to the cloud.2 Indoor Laboratory Fire Dataset conducted experiments[18] .3 NIST experiments and activation time for non-modified smoke alarms, heat alarms, and sprinkler .4 Data acquisition system using IoT analytics platform service to aggre- gate live data streams on cloud .6 Time, frequency & wavelet transform .7 Example of original and filtered data .8 Cumulative % variance explained by principal components. 33 6 List of Tables 2.1 Fixed temperature and Rate-of-Rise(RoR) features .2 Comparison of UV, IR, and UV/IR flame detectors .1 Performance results of SVMs .2 Performance result of ANNs .3 SVMs (fine-tuned) on collected Experimental Household Data .1 Motivation Fires pose a significant threat to safety and property worldwide, and Hanoi, like many urban areas in Vietnam, is no exception[1][10][28][26][27]. Every year, fires cause extensive damage, resulting in the loss of homes, valuable assets, and, tragically, lives. Despite advancements in fire-fighting technologies and equipment, timely detection and response remain critical challenges.
The densely populated areas and rapid urbaniza- tion of Hanoi exacerbate these risks, making early fire detection not just a matter of safety, but a pressing necessity. In recent years, Hanoi has seen a number of high-profile fire incidents that have high- lighted the limitations of current fire detection systems. These incidents often escalate rapidly due to delayed detection, leading to devastating consequences. Traditional fire detection methods, which rely primarily on smoke and heat sensors, often fail to pro- vide the necessary early warning needed to prevent such disasters.
This is particularly concerning in residential areas and commercial buildings where the concentration of people and flammable materials is high. The motivation for this thesis stems from the urgent need to enhance fire detection capabilities in Hanoi and across Vietnam. By leveraging modern technologies such as machine learning and advanced sensor systems, there is potential to significantly improve early fire detection and response times. This research aims to create and implement a fire detection system that can provide accurate and timely alerts, thereby mitigating the risks and impact of fires.
Furthermore, the choice to concentrate on indoor laboratory data for this study is 8 driven by the desire to create a controlled environment that allows for precise calibra- tion and testing of fire detection algorithms. This approach ensures that the system is robust and reliable before deployment in real-world scenarios. The ultimate goal is to contribute to a safer environment in Hanoi and other urban areas in Vietnam, reduc- ing the incidence and severity of fire-related incidents through innovative technological solutions. The integration of machine learning into fire detection systems represents a signifi- cant advancement over traditional methods.
Machine learning algorithms are capable of examining patterns and identifying anomalies in environmental data, providing early warning signals that might be missed by conventional sensors. This thesis seeks to ex- plore these capabilities, developing a system that not only detects fires more quickly but also reduces false alarms, thereby increasing the overall efficiency and reliability of fire response strategies. In conclusion, the motivation for this research is rooted in the need to address the persistent and growing threat of fires in Hanoi and Vietnam. By harnessing the power of machine learning and advanced sensor technology, this thesis aims to pioneer a new approach to fire detection, offering a solution that is both innovative and impactful.2 Objectives The main goal of this thesis is to design, develop, and evaluate an advanced early fire detection system using machine learning techniques and a comprehensive set of sensors, with the goal of providing a more effective and reliable solution for detecting fires at their inception in indoor environments, thereby enhancing safety and mitigating damage.3 Approach and Methodology The approach undertaken in this research is systematic and multi-faceted.
Initially, data collection involves the meticulous gathering of sensor readings from controlled fire scenarios in a laboratory environment. This includes sensors measuring CO2 levels, hu- midity, temperature, MQ139, TVOC, and eCO2, ensuring a diverse set of data points for model training. The subsequent stage involves comprehensive data preprocessing, which encompasses data cleaning, normalization, and addressing any missing values 9 to maintain the integrity and suitability of the dataset for machine learning applica- tions. The research then progresses to model selection, where various algorithms such as Support Vector Machine (SVM) and Artificial Neural Network (ANN) are identified for evaluation.
The dataset is systematically partitioned into training, validation, and test sets to support robust model development and performance evaluation. Rigor- ous performance metrics, including accuracy, F1-score, and ROC-AUC are utilized to compare the models, ultimately determining the most effective method for early fire detection. The final stage involves the implementation of the best-performing model in a real-time detection system, followed by an in-depth analysis of the results to discuss the findings and pinpoint areas for future research and enhancement.4 Scope and Limitation This thesis focuses on designing, developing, and evaluating an early fire detection system using machine learning techniques tailored for indoor environments. The system integrates various sensors, including those measuring temperature, humidity, gases, and flames, and employs machine learning algorithms, specifically Artificial Neural Networks and Support Vector Machines.
Data will be gathered via controlled fire experiments to train and validate these models. The system’s performance will be evaluated using metrics such as precision, sensitivity, specificity, accuracy, and ROC- AUC, with real-world testing to ensure practical effectiveness. However, this research has certain limitations. The controlled laboratory environ- ment for data collection and initial testing may not fully represent real-world condi- tions, potentially impacting the system’s generalizability.
The reliability and accuracy of the sensors used are critical, and any limitations in sensor performance could affect the overall effectiveness of the system. Despite testing various fire scenarios, not all potential fire situations can be covered, which may limit the system’s applicability. Technical constraints related to the ESP8266 microcontroller and LPWAN for data transmission might also pose challenges. Additionally, the system may experience false positives or negatives, necessitating further refinement.
Scaling the system to larger or more complex settings, such as industrial sites or extensive commercial buildings, may require additional adjustments and optimizations.5 Challenges The research encounters several challenges during conduct: Collecting high-quality data is difficult due to controlled environment limitations, sensor precision issues, and data imbalance. Laboratory settings may not capture real-world fire variability, and ensuring precise sensor calibration is crucial to avoid inaccurate readings. Additionally, the rarity of fire incidents compared to non-fire events, leading to an imbalanced dataset that can bias the model. Model selection and training also pose challenges.
There is a balance to be struck between complexity and performance, with more sophisticated models such as deep neural networks requiring more resources and being prone to overfitting. Hyperpa- rameter tuning is time-consuming, and selecting relevant features from sensor data is critical for model accuracy.