K. K. Wagh Institute of Engineering Education and Research

Environmental sensor systems are essential for monitoring infrastructure and environmental quality but are prone to unreliability caused by sensor faults and environmental anomalies. Using Environmental Sensor Telemetry Data, this study introduces a novel methodology that combines unsupervised and supervised machine learning approaches to detect anomalies and predict sensor failures. The dataset consisted of sensor readings such as temperature, humidity, CO, LPG, and smoke, with no class labels available. This research is novel in seamlessly blending unsupervised anomaly detection using Isolation Forest to create labels for data points that were previously unlabeled. Finally, these generated labels were used to train the supervised learning models such as Random Forest, Neural Network (MLP Classifier), and AdaBoost to predict anomalies in new sensor data as soon as it gets recorded. The models confirmed the proposed framework's accuracy, whereas Random Forest 99.93 %, Neural Network 99.05 %, and AdaBoost 98.04 % validated the effectiveness of the suggested framework. Such an approach addresses a critical gap, transforming raw, unlabeled IoT sensor data into actionable insights for predictive maintenance. This methodology provides a scalable and robust real-time anomaly detection and sensor fault prediction methodology that greatly enhances the reliability of the environmental monitoring systems and advances the intelligent infrastructure management.•Combines Isolation Forest for anomaly labeling and supervised models for anomaly prediction. •Scalable and adaptable for diverse IoT applications for environmental monitoring. •Provides actionable insights through anomaly visualization, revealing patterns in sensor performance.

Industrial control systems (ICS) are crucial for automating and optimizing industrial operations but are increasingly vulnerable to cyberattacks due to their interconnected nature. High-dimensional ICS datasets pose challenges for effective anomaly detection and classification. This study aims to enhance ICS security by improving attack detection through an optimized feature selection framework that balances dimensionality reduction and classification accuracy. The study utilizes the HAI dataset, comprising 54,000 time series records with 225 features representing normal and anomalous ICS behaviors. A hybrid feature selection approach integrating wrapper and filter methods was employed. Initially, a Genetic Algorithm (GA) identified 118 relevant features. Further refinement was conducted using filter-based methods—Symmetrical Uncertainty (SU), Information Gain (IG), and Gain Ratio (GR)—leading to a final subset of 104 optimal features. These features were used to train classification models (Naive Bayes (NB), Random Forest (RF), and Support Vector Machine (SVM)) with a 70:30 train-test split and tenfold cross-validation. The proposed feature selection method significantly improved classification accuracy, achieving 98.86% (NB), 99.91% (RF), and 97.97% (SVM). Compared to the full dataset (225 features), which yielded 97.51%, 99.93%, and 96.17%, respectively, our optimized feature subset maintained or enhanced classification performance while reducing computational complexity. This research demonstrates the effectiveness of a hybrid feature selection approach in improving ICS anomaly detection. By reducing feature dimensionality without compromising accuracy, the proposed method enhances ICS security, offering a scalable and efficient solution for real-time attack detection.

The purpose of this research is to determine how well bamboo-based anti-hailstorm structures provide stability and resilience in the face of a variety of unfavorable weather conditions, such as wind, rain, and hail events. The study also aims to investigate the benefits that using bamboo as a primary construction material has for the environment, as well as the structural integrity and material durability. Software analysis and simulations were used to evaluate the performance of these structures under simulated hailstorms, heavy rainfall, and high wind speeds. The results show that bamboo, because of its strength and flexibility, offers a viable, sustainable alternative to conventional materials.

The Machine scheduling problems are solved to optimize the objectives of material flow time which depends upon processing time, setup time and material handling time. In flexible manufacturing system, automated guided vehicles (AGV) are used for material handling purpose from one machine to another machine. However, in most of the cases, machine scheduling problems and AGV scheduling problems are solved separately. Some authors have solved machine scheduling and AGV scheduling problem simultaneously, but they have considered fixed number of AGVs. However, as the number of AGV increases, it will minimize the material flow time, but this will increase the idle time and cost of the AGVs. This material flow time will be stabilized after reaching to the certain number of AGVs. Hence, this problem leads to identify the optimum number of AGV required for the material handling system. In this paper, an attempt is made to identify the best sequence of operations of the products on different machines. From this sequence of operations of all the products, the processing time, sequence-dependent setup time, material handling time and waiting time will be calculated. Hence, in this paper, the objective function is set to minimize the material flow time which depends upon processing time, setup time, material handling time and waiting time. The Artificial Intelligence (AI) techniques used in this paper are (i) Metaheuristics and their applications in intelligent automation: discrete artificial bee colony algorithm (DABC) and (ii) Industrial experiences in the application of the above techniques, e.g. case studies of a flexible manufacturing system with AGVs for material handling. The discrete artificial bee colony algorithm (DABC) algorithm is applied on the case studies. There is around 10% improvement in the results obtained by discrete artificial bee colony algorithm (DABC). The proposed approach thus can be effectively implemented to reduce the material flow time in industries such as automotive industries, electronics industries, and consumer goods manufacturing industries especially while operating in flexible environment.

This work focuses on ER (Electrorheological) fluid-based hybrid hole-entry journal bearings, which studies the effect of ER fluid with partial electric field on the tribological nature of two-lobe bearings with bionic textured surfaces. The design of bionic textures, by virtue of their capacity to influence fluid flow, load capacity, and friction reduction for the enhancement of bearing performance, has been analytically investigated. The integration of ER fluid subjected to partial electric field along with bionic textured surfaces displays an optimistic improvement in performance metrics of bearings, which clearly manifests through preliminary results. Notably, the optimization of bionic textures through genetic algorithms resulted in a 13.04% reduction in friction coefficient and a significant increase in load capacity. The combination of partial ER fluid lubrication and bionic textures also improved fluid film stiffness by up to 66.11%, further enhancing stability and overall bearing efficiency. The obtained results are expected to further enhance design and development activities of high-performance hybrid journal bearings for numerous industrial applications.

A major issue brought on by land degradation, increased agricultural productivity, and other human activities is soil erosion. Planning and conservation efforts within a basin or watershed benefit from an assessment of soil erosion. Soil erosion is primarily influenced by rainfall patterns and land cover, and modeling soil erosion is crucial for assessing the degree of land degradation. Under a variety of circumstances, modeling can offer a quantitative and reliable method for estimating soil erosion and sediment output. In the current work, soil loss in the Konkan region of India of the Vashisthi Basin has been estimated using the Revised Universal Soil Loss Equation (RUSLE), a soil loss model integrated with GIS rainfall data for previous 16 years (2006–2021) referred to analyses study.The RUSLE model, in conjunction with elements such as the Rainfall Erosivity Factor (R), for case study of Vashisthi River Basin (VRB) of Konkan, Maharashtra (India) given results with average value for Chiplun (30978.105 MJ.mm/ha/hr/y), Dapoli (30701.941 MJ.mm/ha/hr/y), Khed (28807.257 MJ.mm/ha/hr/y),Guhagar (24210.344 MJ.mm/ha/hr/y)

The adsorptive separation of Acid Red 33 (AR33) dye by groundnut shell-based adsorbent obtained using phosphoric acid as an activating agent was investigated. The adsorbent characteristics were determined by Brunauer–Emmett–Teller (BET), scanning electron micrograph (SEM), and Fourier transformed infrared spectroscopy (FTIR) analysis. Thermochemical activation yielded a significant enhancement in surface area from 1.971 to 1150.252 m²/g. The effects of pH (2–10), adsorbent loading (1–5 g/L), concentration (50–200 mg/L), temperature (288–318 K), and time (15–270 min) were investigated on the extent of adsorption. Isotherm and kinetic analysis of adsorption trials revealed best fitting of experimental data by Langmuir isotherm and pseudo-second-order models. Maximum Acid Red 33 uptake by modified adsorbent was obtained as 107.53 ± 0.91 mg/g at dose of 3.5 g/L and pH of 2 in 180 min. Thermodynamic characteristics demonstrated spontaneous and endothermic adsorption. The reusability study implied the drop in the removal efficiency from 98.26 ± 0.97% at first cycle to 79.57 ± 1.04% to the third cycle. These results demonstrated the reusability potential up to three cycles. The presented research demonstrated that physio-chemically activated groundnut shell powder is a promising sorbent for adsorptive separation of Acid Red 33. Graphical abstract

In this work we report the hydrothermal method for the synthesis of NiTiO3 perovskite nanoparticles (NPs). The influence of reaction time variation (6 h, 12 h, 18 h and 24 h) on the synthesis of NiTiO3 NPs was studied. The synthesized material with variation in reaction time was characterized by various characterization methods like UV-Visible Spectroscopy, X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), Field emission scanning electron microscopy (FESEM), Energy dispersive X-ray spectroscopy (EDS), Transmission electron microscopy (TEM), High resolution transmission electron microscopy (HRTEM) and Selected area electron diffraction (SAED). Band gaps of the material synthesized by varying reaction time were studied and it was found that band gap decreases on increase in reaction time. The various structural parameters of the NiTiO3 like crystallite size, dislocation density and microstrain were calculated from XRD data. From the FESEM analysis, average particle size was estimated and found to increase in the range of 32 nm to 37 nm as reaction time increased from 6 h to 24 h. The electrical properties of the material were studied. The gas sensing performance of NiTiO3 thick films was tested on static gas sensing unit for various gases such as H2S, CO, CO2, Cl2, NH3, Ethanol and LPG. It showed maximum response to CO2 gas at room temperature (RT) for sample synthesized at 24 h reaction time which is attributed to the disappearance of TiO2 phases. The repeatability and long term stability of the sensor were also studied.

Background ML-powered Internet of Medical Things (MLIoMT) is a burgeoning framework poised to transform healthcare, particularly in the timely identification of heart disease. Objectives This article proposes an innovative MLIoMT structure aimed at leveraging machine learning (ML) algorithms for heart disease detection. Materials and Methods Through the integration of wearable sensors, mobile applications, cloud computing, and advanced ML techniques, MLIoMT enables continuous monitoring of vital signs and cardiac health indicators in real time. By analyzing this data stream, abnormalities indicative of heart disease can be detected early, facilitating timely intervention and personalized healthcare recommendations. The MLIoMT framework employs diverse ML methods, such as deep learning and ensemble techniques to enhance the accuracy and reliability of heart disease prediction models. Results The proposed structure holds promise for revolutionizing preventive healthcare, enabling proactive management of cardiac health, and ultimately reducing the burden of heart disease. Results in terms of accuracy, precision, recall and F1 score show that the proposed system has better performance and efficiency. Conclusion Overall, MLIoMT represents a significant advancement in healthcare technology, with the potential to improve patient outcomes and enhance overall quality of life.

In this work we are presenting the hydrothermal method to synthesize NiTiO3 perovskite nanoparticles (NPs). The effect of variation in reaction temperature on the structural, optical, electrical and gas sensing properties of NiTiO3 nanoparticles was investigated. The nanoparticles synthesized at different reaction temperatures were characterized by various characterization methods like XRD, FTIR, UV–Visible Spectroscopy, FESEM, TEM, HRTEM and SAED. The results of UV–Visible analysis revealed that band gap of NiTiO3 decreased from 2.90 to 2.56 eV on increase in reaction temperature from 140 to 200 °C. The XRD analysis showed that crystallite size decreased in the range of 21 to 12 nm on increase in reaction temperature. The various parameters of the material like dislocation density, microstrain and crystallinity were also calculated from XRD data. The average particle size was estimated by FESEM analysis and found to be decreased on increase in reaction temperature. FTIR analysis confirmed the formation of NiTiO3. Study of electrical properties proved the semiconducting behaviour of NiTiO3. The detail analysis of NiTiO3 sensor characteristics in terms of sensitivity, selectivity, response and recovery time was carried out. The study of gas sensing performance of NiTiO3 revealed that NiTiO3 synthesized at 140 °C showed maximum sensitivity to CO2 gas at room temperature, whereas NiTiO3 synthesized at 200 °C showed maximum sensitivity to H2S gas at 250 °C.

This chapter addresses the limitations of single-rate sampling in designing discrete-time state feedback-based controllers for two-time-scale systems. It proposes a multirate state feedback (MRSF) control approach, i.e. sampling slow and fast states at different rates. The method involves constructing a block-triangular form of the continuous two-time-scale system, discretizing with a smaller sampling period for the fast subsystem, and block-diagonalizing for the slow subsystem. A derived MRSF control stabilizes the full-order system, employing a two-stage observer to estimate slow and fast states. Simulation results demonstrate computational savings without compromising closed-loop performance compared to the single-rate sampling method.

This chapter addresses the challenges of designing discrete-time state feedback control for a multi-time-scale system, where lower sampling rates fail to capture information from fast and very fast states, while higher rates significantly increase computation. The proposed solution is a multirate state feedback (MRSF) control for a linear time-invariant three-time-scale system, involving sampling of slow, fast, and very fast states at different periods. The method employs a block-triangular form of the continuous three-time-scale system, successive discretization, and simultaneous feedback control design at various sampling rates to formulate a combined MRSF control, proven to stabilize the full-order system. A sequential three-stage observer is introduced for state estimation, and simulations demonstrate the superiority of the approach over single-rate sampling.