Dr B Meenakshi
@sairam.edu.in
Professor/ Department of Electrical and Electronics Engineering
Sri Sairam Engineering College
RESEARCH INTERESTS
Wireless Sensor Network, Renewable Energy
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
B. Meenakshi and D. Karunkuzhali
Elsevier BV
U. Kavitha, Manjula Pattnaik, S. Jayaprakash, Yuvaraj S, Geetha Ponnaian, and B. Meenakshi
IEEE
The healthcare industry is one of the industries where the usage of smart refrigerators powered by the Internet of Things (IoT) is increasingly rapidly. To improve the performance of IoT smart refrigerators, this research proposes an efficient method that uses deep learning (DL) analytics, particularly Long Short-Term Memory (LSTM) networks. The refrigerator can assess and forecast medicine use trends from past data using LSTM networks, which allows for optimum storage conditions and timely refills. The system design incorporates IoT sensors to track humidity, temperature, and door status in real-time to view the fridge's surroundings fully. LSTM networks analyze the data gathered from these sensors to understand how people use their medications over time. Because of this, the fridge can calculate how much medicine will be needed in the future, notify users if there is a risk of running out, and keep medications at their most effective storage temperatures. The proposed system uses data-driven analytics to make medication management suggestions to each patient's unique profile and treatment regimen. The technology changes medicine storage settings on the fly to accommodate changing patient demands using LSTM networks' sequential data modeling capabilities. Experiment results show that the suggested method improves patient care by decreasing pharmaceutical waste and increasing medication adherence. One potential solution for smart medicine storage in healthcare settings is the integration of DL analytics with IoT smart freezers. This might lead to the development of data-driven pharmaceutical management systems.
Cheepurupalli Raghuram, Subramani Roychoudri, C. Viswanathan, C. Chandru Vignesh, Geetha Ponnaian, and B. Meenakshi
IEEE
Major source of increased healthcare settings morbidity, and death, hospital-acquired infections (HAIs) need urgent attention in healthcare facilities. Conventional approaches to HAIs detection and prevention mostly on retrospective analysis and manual surveillance, which can cause intervention delays and less-than-ideal results. In this research, it provides a new way to deal with these issues by automatically detecting and predicting HAIs using the power of the Internet of Things (IoT) and linear regression (LR) methods. Using IoT sensors throughout health care facilities to continually track vital signs, patient interactions, and healthcare procedures. Advanced statistical approaches, such as LR models, are used to the real-time data streams generated by these sensors to detect patterns and connections that may indicate the existence of HAIs and the dynamics of their transmission. The proposed method allows for the early detection of possible infection clusters and high-risk regions in the hospital by combining several data sources, such as patient demographics, medical history, and clinical processes. Predictive analytics to plan for the possibility of HAIs outbreaks and determine which preventative measures will be most effective. Healthcare institutions may improve their infection control policies, allocate resources more efficiently, and decrease the occurrence of automated systems. It shows that this strategy works, which bodes well for its future use in hospital settings where patient safety is paramount, and HAIs control is under constant scrutiny.
K. Lalitha, M. Xavier Suresh, R. Selvakumar, Prathik A, J. M. Sunitha, and B. Meenakshi
IEEE
This research focuses on sustainable crop protection through the integration of the Internet of Things (IoT) and drone-based spraying. The main goal is to improve agricultural practices' accuracy, efficiency, and environmental friendliness. Drones with spraying and IoT capabilities allow for the precise and controlled management of crop protection chemicals like insecticides. This approach minimizes the use of chemicals, reduces their environmental impact, and makes the most efficient use of resources. When used with IoT sensor data, support Vector Machine (SVM) analysis improves decision-making by providing predictive insights. SVM analysis allows for the accurate localization of treatment zones, which improves pesticide application while reducing damage to non-target species. This work proposes a potential way forward for long-term crop security using SVM analysis and IoT-enabled drone technology. It signals a new era of environmentally conscious crop protection techniques prioritizing effective pest control without sacrificing environmental preservation.
A. Sahaya Anselin Nisha, N. Venkatesvara Rao, G. Venkatesh, Sree Southry S, S. Murugan, and B. Meenakshi
IEEE
This research aims to find out how to manage congestion more effectively in cities by combining reinforcement learning (RL) with Internet of Things (IoT)-driven road user charging systems. Unlike typical static pricing models, this approach aims to improve traffic flow and reduce congestion by dynamically adjusting prices depending on real-time traffic circumstances. The system gathers and analyzes data on road use trends using IoT infrastructure, enabling adaptive pricing schemes. RL algorithms are then used to adjust billing policies in real time based on user preferences, congestion levels, and traffic volume. The evaluation of proposed approach is based on simulation studies performed in a genuine city setting. Compared to more traditional methods, the results show that it significantly improves commuters' journey times and reduces congestion. The results of this study highlight the promise of RL approaches in combination with IoT road user charging systems to alleviate traffic in cities and improve transportation efficiency generally.
M. Shafiya Banu, Arshadh Ariff Mohamed Abuthahir, R. Mohandas, D. Antony Joseph Rajan, B. Meenakshi, and N. Mohankumar
IEEE
Precision agriculture struggles with scalability, data integration, and decision-making, prompting innovative solutions. Recent advancements like CNNs and cloud computing provide interesting answers but confront challenges. Data quality, processing needs, interoperability, and real-time decision-making are issues. These difficulties demand innovative data synchronization, model dependability, and system scalability methods. This project uses sophisticated analytics and cloud technology to improve nutrient management systems. Improving scalability, accuracy, and efficiency while presenting unique data fusion and user-friendly interfaces are goals. Clear problems and goals guide the study's scope and purpose. Wireless soil nutrient sensors can capture real-time data on important soil factors such as pH levels, the amount of nitrogen, phosphate, and potassium present, and more. This information is then sent to a platform hosted in the cloud, where it is processed using sophisticated data analytics and machine learning algorithms, in addition to previous crop performance data and agronomic models. The program will create tailored suggestions for nutrient management, including topics such as the kinds of fertilizer, application rates, and the best time. These proposals give farmers more agency, allowing them to make educated decisions via an easily navigable interface. Because of the system's ability to react to changing circumstances and its continual monitoring, nutrient management techniques are kept in sync with those changes throughout the crop's life cycle.
D. Chandrakala, Chitra Sabapathy Ranganathan, E. Swarnalatha, R. C. Karpagalakshmi, B. Meenakshi, and S. Srinivasan
IEEE
The evaluation of the protections of soil carbon sequestration monitoring technique using cloud computing and machine learning. The implementation of these systems is essential for development of agricultural industry that is friendly to the environment for reducing the effect on the environment. As a consequence of recent developments in technology, including cloud computing, these surveillance systems are now far more effective than they were before. However, the storing of data on the cloud creates worries over the security of the data. Using Random Forest Regression (RFR) it potential these difficulties. Through the use of this method, the potential to improve the safety of data collection is utilized. With the examination of patterns within the data that has been collected to identify possible security. The capabilities of ensemble learning are responsible for the improved consistency and precision with which it operates. According to the findings of the studies, this method makes the process of collecting and analyzing data on soil carbon more secure. For maintaining confidentiality this is beneficial for both environmentally responsible farming and sustainable agriculture.
D. Karunkuzhali, B. Meenakshi, and Keerthi Lingam
Springer Science and Business Media LLC
E. Sudhakar, J. Thamilselvi, N Logeswari, B. Meenakshi, and K. Subhashini
IEEE
This paper proposes monitoring leakage in underground pipes using ZigBee in Wireless Sensor Networks. Leakage in underground pipes is a significant concern for water utilities as undetected leaks can result in water loss, leading to substantial financial losses and environmental concerns. The proposed system comprises ZigBee-enabled sensor nodes deployed strategically along the pipeline network. These nodes collect real-time data on various parameters such as pressure, flow rate, and temperature. By analyzing these parameters using advanced algorithms, the system can detect anomalies indicative of leaks or abnormalities in the pipeline network. To validate the effectiveness of the proposed system, simulations and experiments are conducted under various scenarios, including different leak sizes and locations. Results demonstrate the system’s high accuracy in detecting and locating leaks while minimizing false alarms.
A. Kathija Nasreen, M. Shenbagapriya, Senthil Kumar Seeni, Parthiban Veda, B. Meenakshi, and S. Murugan
IEEE
Public restrooms needs to be frequently cleaned to maintain public health and hygiene. Traditional bathroom cleaning techniques may not solve current cleanliness issues. This research proposes an advanced robotics, Internet of Things (IoT), and Recurrent Neural Networks (RNNs) solution to these problems. The proposed system uses autonomous robots with IoT sensors and cameras. Robots identify cleaning and maintenance needs in restrooms. A central system receives data from IoT sensors on cleanliness indicators, including toilet paper, soap, and foot movement. Recurrent Neural Networks (RNN) processes this data to predict and prioritize cleaning requirements. The RNN monitors toilet conditions in real time and reacts to changing use and cleaning needs. This dynamic technique optimizes cleaning resource allocation and maintains facility cleanliness. The device also warns cleaning personnel when certain areas need quick attention. This novel technique makes restroom care more cost-effective, responsive, and environmentally friendly by combining robots, IoT, and RNNs. This study advances smart facility management, which uses technology to improve public space cleanliness, user experience, and resource usage.
P. Santhuja, C. Selvi, C. Jehan, S. Aghalya, B Meenakshi, and K Mathivanan
IEEE
New automobiles must use Controller Area Network (CAN) architecture to meet automation needs and simplify point-to-point wire harnesses. This study offers a complete system that uses a CAN bus-based network to promote effective data transmission inside the automotive environment. Essential data, including speed, distance from other cars, in-car alcohol concentration, and unintentional lane changes, are all monitored by the system’s integrated sensors. The technology may successfully avoid accidents by constantly monitoring these factors and immediately alerting the driver if any move beyond the safe range. The system’s use goes much beyond simple warnings to drivers. The device uses a bump sensor to detect crashes and immediately sends an SMS message using GSM technology to seek fast help in distant places where accidents may occur. Changing lanes when another car is in the blind area is a prime example of a situation where this technology might be useful. The technology identifies the other vehicle’s presence using built-in sensors and alerts the driver before the lane change is initiated, reducing the likelihood of a collision. The suggested system’s reliance on CAN bus-based communication increases its adaptability and scalability, making it better able to accommodate the inevitable technological advances of the future. The combination of sensors and immediate warning mechanisms shows the potential of cutting-edge in-vehicle technology to improve road safety and accident avoidance.
Ramakrishnan Raman, Hanumaji Kantari, Atul A Gokhale, K. Elangovan, B. Meenakshi, and S. Srinivasan
IEEE
Storage, distribution, pricing, marketing, import/export, and other policy considerations all depend on accurate and timely crop output estimates. The directorates of economics and statistics issues official predictions advance estimates in the form of estimates and projections for key grains and commercial crops. However, these guesses in advance are not the actual projections. Using the K-Nearest Neighbors (KNN) method to estimate agricultural yields is a good choice since it is easy to understand, can handle non-linearity, and works well with small to medium datasets. Given these features, KNN is a good choice for agricultural stakeholders, including academics, decision-makers, and farmers who may not have a lot of experience with machine learning. With the use of the KNN method, agricultural yield estimate models may be trained to provide precise crop production predictions given pertinent input data. There is a lot of subjective judgment based on a variety of qualitative criteria involved in arriving at these estimations. As a result, it’s important to provide objective, statistically reliable projections of future crop areas and yields. Technological developments in computers and data storage have made an enormous amount of Information available. The difficulty has been in gleaning insight from this mountain of data, but recent advances in areas like data mining have opened a door from this data to more precise estimates of agricultural yield. Results show that the accuracy is 90%, and the error value is 10% using the KNN algorithm for the given agriculture datasets.
Hanumaji Kantari, M. Vadivel, Pavithra Jagadeesan, T. J. Nagalakshmi, B. Meenakshi, and S. Velmurugan
IEEE
The need for better control approaches in Permanent Magnet Synchronous Motors (PMSMs) has increased due to the increased usage of electric motors in many different areas. The unique characteristics of PMSM motors make it difficult to use conventional control techniques to dynamically adjust speed and torque. This research proposes a novel approach to real-time speed and torque optimization of PMSM motor drives by integrating Sliding Mode Control (SMC) in Reinforcement Learning (RL). This could alleviate these issues. This study introduces a dual-mode control approach for permanent magnet synchronous motors (PMSM) that leverages the robustness of SMC and the flexibility of RL. In order to improve the PMSM motor drive’s overall efficiency, a synergistic effect is achieved by creating an RL-based Maximum Torque and Speed Tracking (MTST) algorithm. Incorporating SMC ensures precise and stable control of speed and torque. Results from both theoretical and practical investigations using a PMSM motor operating system model in Matlab/Simulink validate the proposed method. The results reveal improved control over speed and torque, proving the efficacy and adaptability of the proposed method for PMSM actuators to function in different environments.
V G Sivakumar, N. Arunfred, N Anusha, C. Balakrishnan, B. Meenakshi, and S. Sujatha
IEEE
Energy-efficient solutions are in high demand in today's fast-developing technological environment, necessitating the use of sophisticated predictive modeling methods. Conventional forecasting systems have problems given the particular aspects related to home demand for energy. As a consequence, new strategies are constantly being created for more accurately foreseeing future occurrences. Weather, user habits, and advancements in technology interfere with precise forecasts even more. This study provides a novel application of the Gradient Boosting approach to solve such obstacles and enhance energy use prediction for residential applications. Through utilising the electrical energy of Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM), the suggested method takes into account prior use patterns, appliance-specific data, and external factors like as climate. The model's response to changes in the actual world provides exact estimates, providing homeowners with the information they require to make energy-saving decisions. This method also investigates the comprehending of Gradient Boosting models, showing key characteristics that affect energy usage patterns. Stakeholders focus heavily on interpretability to make informed decisions with regard to ways to save energy and resource allocation.
V. Vidya Lakshmi, S Giriprasad, S P Vimal, Hanumaji Kantari, B Meenakshi, and S. Srinivasan
IEEE
An efficient wind power forecasting system is essential to maximize the benefits of renewable energy’s incorporation into the grid. Thus, a new method for wind power forecasting by combining the Sparrow Search Algorithm (SSA) with Support Vector Machines (SVM) is presented. In order to fine-tune the settings of the SVMs, the SSA is employed, which uses the hunting behavior of sparrows. It improves the model’s capacity to capture fine fluctuations in wind power production. The proposed SSA-SVM model is tested using historical wind power data, and the findings show a considerable increase in predicting accuracy compared to standard SVM models. Time series analysis also shows how well the SSA-SVM model can adjust to wind power patterns. The effectiveness of the proposed system for wind power forecasting and its potential role in enhancing the prediction of wind power in the renewable energy environment are analyzed.
V. Dhayalan, Ramakrishnan Raman, N. Kalaivani, Ankit Shrirvastava, Ramireddy Sasidhar Reddy, and B Meenakshi
IEEE
Residential energy management is an important component of study in the effort for economical and green energy use. The majority of the time, traditional approaches create inadequate use of energy storage devices (ESS) and can't adapt to the ever-changing patterns of household energy usage. This research recognized these challenges and provided a new way to tackle them by integrating reinforcement learning (RL) techniques with IoT technology to make household ESS more efficient. The Internet of Things (IoT) allows the real-time capture of data, and RL shown promise in resolving optimization issues related to component mobility. The combination of these two elements could end up in a modern energy management system designed to meet the demands of residential environments. By employing a reinforcement learning approach that incorporates past data and adapts to new circumstances, efficient personalized energy optimization may be accomplished. By integrating IoT devices, real-time data can be collected, enabling the system to respond quickly to changes in demand and supply of energy. Key insights into the possibilities for intelligent and flexible systems for energy management in homes are provided by the study's actual demonstration of their application.
B. Meenakshi, D. Karunkuzhali, and Syed Muqthadar Ali
Wiley
SummaryIn this article, blockchain‐enabled hybrid Red Fox optimization and arithmetic optimization approach‐based cluster head selection along Hazelnut tree search algorithm (HTSA)‐based optimal trust path selection is proposed to secure data transmission at wireless sensor network. The proposed BC‐Hyb‐RF‐AOA‐HTSA‐WSN method consists of two phases: (i) to find optimum cluster head (CH) and (ii) to find optimal trust path. Firstly, hybrid Red Fox optimization approach and arithmetic optimization algorithm are employed to select cluster head accurately. After CH selection, HTSA is used to find trust route from several routes, which is finalized optimally with the joint trust that depends on trust parameters. Finally, blockchain is provided with optimized, carefully chosen trust routes for communication. The proposed BC‐Hyb‐RF‐AOA‐HTSA‐WSN method is activated in NS2 tool. The proposed technique achieves lesser delays of 98.38%, 92.34%, and 97.45%, better delivery ratios of 89.34%, 83.12%, and 88.96%, and lower packet drops of 91.25%, 79.90%, and 92.88% compared with the existing techniques, such as BC‐FA‐ROA‐WSN, BC‐RDA‐WSN, and BC‐HRDSS‐WSN.
B. Meenakshi and D. Karunkuzhali
Wiley
Meenakshi B., Logeswari N, Sivaprasad R, Brindha J, Janani N, and Gokula Subashree N
IEEE
India is a nation that is quickly developing globally. Waste management is crucial in a rising economy India It's critical to manage trash in a rising economy. 64 million tons of debris are created annually in India, placing it fifth in the world. Today, trash is sometimes tossed carelessly in many communities, and litter-filled highways are common. It calls for a management system that would address this issue and has a thorough understanding of how dustbins are inundated as waste flows out, contaminating the area in public spaces. Waste management is crucial in a rising economy India These unforeseen events lead to numerous issues. When there is too much waste, it can lead to dangerous diseases and increase the amount of infections because so many bugs can breed there. To combat this, smart bins were developed to track waste levels and perform waste separation. Existing components include automatic lid opening, level deduction, segregation, and sensing motion.
B Meenakshi, B Sumathy, Aouthithiye Barathwaj SR Y, Haariharan N C, Krishnakanth L, and M. Soumya
IEEE
In today's fast-paced world, where technological advances continue to transform various industries, agriculture is no exception. With the growing demand for sustainable food production and efficient resource management, there is an urgent need for innovative solutions that can transform traditional agricultural practices. Enter the world of artificial intelligence (AI) and machine learning (ML), which offer unprecedented possibilities in plant monitoring systems. Gone are the days of manual observations and guesswork to understand plant health and growth patterns. With the help of AI and ML, farmers and scientists can now immerse themselves in a new era of precision agriculture, where the well-being of each plant is monitored, analyzed and optimized in real-time.
B. Meenakshi, A. Vanathi, B. Gopi, S. Sangeetha, L. Ramalingam, and S. Murugan
IEEE
Wireless Sensor Networks (WSNs) are essential for IoT-enabled crisis management and executive response. This paper explores WSNs' function in catastrophe situations and proposes new ways to improve emergency response operations. The article begins with a discussion of disaster management and emergency response scenarios and the necessity for real-time data collecting, monitoring, and analysis. WSNs are suitable for such applications due to their dense and dispersed sensing, tolerance to hostile conditions, and low power consumption. The research provides unique WSN deployment, data aggregation, and resource optimization methodologies for disaster management. These solutions will improve network coverage, dependability, and sensor longevity. The research also examines the synchronization of AI and human brainpower calculations with WSNs to enable smart navigation and robotized crisis response. The research study addresses WSN security and privacy in disaster management. It identifies flaws and suggests ways to classify, verify, and share data. A reenacted fiasco is used to evaluate the proposed alternatives. Situational awareness, response coordination, and emergency operations efficiency increase with the provided techniques. This study sheds light on WSNs' application in IoT-enabled emergency response and disaster management. The proposed solutions advance WSN technology and enable more effective and efficient emergency response systems.
B. Meenakshi, Murugananth Gopal Raj, C. Pradip, and N. Saju
AIP Publishing
D. Karunkuzhali, Balasubramanian Meenakshi, and Keerthi Lingam
Springer Science and Business Media LLC
D. Karunkuzhali, B. Meenakshi, and Keerthi Lingam
Springer Science and Business Media LLC
L. Kurinjimalar, S. Pooja, R. Sivaprasad, B. Meenakshi, J. Shalini Priya, and T. Porselvi
IEEE
The design and development of a fifteen-level inverter for renewable applications is moving forward in our endeavor to reduce the environmental effect of solar power plant electricity generation. The produced DC must be converted to AC in order to connect an AC load to the grid without degrading grid performance. A multilayer inverter (MLI) is an appropriate option as it generates output voltage with stepped waveforms that are more similar to sinusoidal waveforms and have less harmonics. However, the DC voltage sources of switching devices expands and thus increasing the circuit design complexity and control. This paper employs just eight switches and one DC source in MLI to deliver a level output voltage for renewable energy applications.