The rapid proliferation of electric vehicles (EVs) necessitates advanced intelligent systems to manage their increasingly complex subsystems ranging from powertrain optimization to battery management and charging infrastructure. Machine learning (ML) has emerged as a transformative tool capable of addressing these challenges through data driven prediction, adaptive control, and intelligent decision making. However, existing reviews are still disjointed, broadly limited to point subsystems, and deficient in standardized performance assessment, unified benchmarking, and analytical understanding of algorithmic appropriateness. These limit the comparability, interpretability, and real world implementation of machine learning models in electric vehicle infrastructure. This review uniquely remedies these shortcomings in a performance oriented, cross domain synthesis of machine learning deployments for electric vehicle design, battery systems, and charging infrastructure. Systematically examining 240 peer reviewed publications based on unified statistical metrics like Root Mean Square Error (RMSE) and Coefficient of Determination (R2), it provides a quantitative benchmarking framework linking algorithmic groups to subsystem performance. The analysis reveals that hybrid deep learning models such as Long Short Term Memory with Double Deep Q-Network (LSTM-DDQN), Convolutional Neural Networks-Bidirectional Gated Recurrent Units (CNN-BiGRU), and transformer based frameworks consistently achieve superior accuracy (R2 > 0.95, RMSE < 0.05) compared to traditional algorithms. The paper also clarifies the rationale behind why ensemble hybrid deep learning models systematically outperform conventional methods, thus laying the groundwork for subsystem variant optimization and deployment tactics. Along with consolidation, the survey brings to focus crucial open research issues that include dataset standardization, interpretable models, interface to Controller Area Network (CAN)/On-Board Diagnostics (OBD) communication protocols, and federated learning towards ensuring privacy in electric vehicle networks. Overall, this work advances the field by turning disjoint literature into an analytically informed guide to intelligent, interpretable, and scalable electric vehicles. This review concludes by identifying promising future directions such as explainable artificial intelligence (XAI), digital twins, embedded ML, and federated autonomy that will underpin the next generation of intelligent, scalable, and sustainable EV ecosystems.
Abstract This in vitro study evaluated the dentinal tubule occlusion (TO), depth of penetration (DoP), and dentin permeability (DP) of oyster shell-derived nanohydroxyapatite (os-nHAp) with and without 15% proanthocyanidin (PA) pretreatment. os-nHAp was synthesized via the precipitation method and it was characterized. The morphology and particle size of os-nHAp were analyzed using a high-resolution transmission electron microscope (HRTEM). Cytotoxicity of os-nHAp, PA/os-nHAp, and casein phosphopeptide-amorphous calcium phosphate (CPP-ACP) was assessed by (3-(4,5-dimethythiazol-2-yl) 2,5-diphenyl tetrazolium bromide (MTT) assay using human osteosarcoma (MG-63) cell line. One hundred and ninety-seven dentin discs of 3 mm thickness were prepared from the crown portion of extracted human teeth. The dentinal surfaces of the discs were etched for 2 min with 6% citric acid to simulate dentin hypersensitivity. Five discs were randomly selected and the patency of dentinal tubules was confirmed using a scanning electron microscope (SEM). The remaining 192 discs were divided into four groups (n = 48) depending on the type of remineralization as follows: group 1: os-nHAp, group 2: PA/os-nHAp, group 3: CPP-ACP, and group 4: no treatment. The remineralization protocol was followed for 21 days. Out of the 48 dentin discs in each group, 32 discs were used to evaluate dentinal tubule occlusion (TO) and depth of penetration (DoP) using SEM. The remaining 16 discs were subjected to an assessment of dentin permeability (DP) using a hydraulic conductance model. TO, DoP and DP were evaluated after remineralization and acid challenge. Characterization studies confirmed the presence of pure phase apatite. HRTEM confirmed the nanometric particle size of os-nHAp. MTT assay results showed that all the tested materials exhibited >80% cell viability when tested up to a concentration of 100 µg/mL. The results demonstrated a significantly higher mean percentage of TO, DoP, and lesser mean DP after remineralization in groups 1, 2, and 3 (p < 0.05). After the acid challenge, group 3 showed a significant reduction in TO and DoP, and increased DP (p < 0.05). However, no such changes were observed in groups 1 and 2. Within the limitations of this study, it can be concluded that os-nHAp and PA/os-nHAp could serve as potential and durable therapeutic agents in the treatment of dentin hypersensitivity. Graphical Abstract
The exhaust emissions from automotive diesel engines are successfully controlled over the years by adopting different combustion strategies and after treatment devices, whereas the combustion induced vibration and noise are the major pollutant in off-road vehicle engines and yet to be optimized. In the present study, a twin cylinder, Simpson's S-217 tractor diesel engine was used to evaluate the performance, combustion, vibration, and noise characteristics, using biofuel blends. For this study, the blends of pine oil - soapnut oil biodiesel (P75SNB25), diesel - soapnut oil biodiesel (SNB20) and diesel were used as fuel. The pine oil used in this research was purchased and used in its neat form. The soapnut oil was extracted from the soapnut seeds by cold pressing method and trans-esterified in two stages by using methanol and catalysts. The experimental results revealed that the performance and combustion characteristics of the blend P75SNB25 was superior to diesel and the blend SNB20 was slightly inferior to diesel. For the blend P75SNB25, the amplitude of acceleration with respect to time was reduced by 19.48% and 11.58% at no load and full load conditions respectively, whereas for the blend SNB20, the amplitude of acceleration showed a reduction of 14.27% and 9.46% at no load and full load conditions respectively in comparison with diesel operation. But both the blends P75SNB25 and SNB20 showed a maximum reduction of noise by 2.34% at different engine loads compared to diesel operation.
Induction motor is used for different applications in industries such as grinding, milling, mining and automation. Induction motor performance degrades over time due to continuous operation, over load, transient unbalanced supply voltage and current. Hence, induction motor performance and efficiency are evaluated through measurement of motor torque. The torque measurement using various sensors such as acoustic, mems-vibration, strain gauge, in line sensors, never provides the inaccurate results due to contact type sensors and improper mounting of sensor on the motor. Moreover, existing methods never provides inaccurate torque measurement for different types of motor duty cycle such as continuous duty, intermittent periodic duties, short-time duty and continuous operation with intermittent load. In this paper, novel non-contact method measures torque with Dual-function ultra-wide band (UWB) (DFR) Radar sensor. DFR acquires polarized signal reflected from air gap of motor magnetic flux. Moreover, air gap magnetic flux radiation through motor ventilator is directly proportional to motor torque. In this paper, motor torque is measured through air gap magnetic flux radiation and DFR sensor. The electromagnetic waves from DFR is reflected by the air gap magnetic flux emitted from motor and the reflected waves are polarized. The reflected DFR sensor signal from air gap magnetic flux has high polarity charge and resonance. From received polarized signal, Instantaneous Frequency (IF) is obtained through Multi-Synchro Squeezing Transform (MSST) algorithm. The motor torque is measured with IF and Gaussian regression algorithm. Torque spikes in Induction motor is analyzed during frequent change in heavy, low load, and induced transient Torque prediction through MSST IF frequency and Gaussian Regression is compared with Torque predicted through different transform such as Discrete Wavelet Transform (DWT), Stationary Wavelet Transform (SWT), Dyadic Wavelet Transform (DyWT). The MSST based torque measurement provides 96% of accuracy, when compared with traditional strain gauge torque measuring instrument. Initial review.
The primary objective of speech enhancement algorithm is to enhance speech perception characteristics that have been damaged by background noise. Ensuring speech quality and intelligibility is essential in speech processing analysis for effective communication. To do this, a new temporal modulation processing speech enhancement approach is developed that reconstructs the temporal envelopes (TEVs) in the time frequency (T-F) domain using a one-dimensional feature learning convolutional neural network and long short-term memory (ID FLCNN-LSTM). This deep learning architecture is used to restore distorted envelopes. This approach is tested by utilising 2,700 words from nine different speech samples combined with babble noise and SSN (speech spectrum shaped random noise) at various SNRs. The proposed algorithm's performance is assessed using the Short time objective intelligibility (STOI) and Perceptual evaluation of speech quality (PESQ) metrics. The suggested model improved STOI and PESQ scores as 0. S2 and 2. S0 respectively. The experimental outcomes predicted for quality and intelligibility scores outperforms TEV algorithm, 1D-CNN, and other existing models in speech enhancement under noisy conditions.
