Artificial Intelligence for Sustainable Energy

This work compares aerodynamic loads, power outputs and wake fields of upwind and downwind wind turbines. To that end, wind tunnel experiments are performed with model wind turbines with cylindrical and ellipsoidal nacelles, to investigate whether the nacelle shape changes the performance or flow characteristics of the turbine configurations. The thrust and the power coefficients of both upwind and downwind turbines show similar trends with no significant difference between peak power coefficients of the two configurations. For the laminar and uniform inflow conditions, wind speed is slightly higher around the rotor centre in the wake of the downwind turbine. However, the effect of nacelle shape to the wake field is not significant. Turbulence intensity profiles for upwind and downwind configurations do not show significant differences and peak values of turbulence intensities can be observed around the rotor tip region. Differences between turbulence profiles for experiments with cylindrical and ellipsoidal nacelles are negligible. The study shows that downwind turbines with both types of nacelles showed comparable performance and wake characteristics to their upwind counterparts. Therefore, they can be scaled up with flexible and cheaper blades, without having to compromise in terms of turbine performance.

Transformer failure is a significant concern in electrical power systems as it can result in costly damages and endanger human lives. Early detection of defects can minimize damage before it becomes dangerous to work with. Predicting transformer health is essential for ensuring continuous quality of service to consumers through a predictive maintenance approach. Maintenance records of transformers often include temperature, oil quality, and vibration. The challenge lies in developing models that enable the prediction of the transformer health index (HI) from these maintenance records. Several research studies have formally reported various implementations of machine learning algorithms to predict transformer health index (HI). In this research report, authors introduce a machine learning algorithm and classification on transformer health detection approach using a support vector machine (SVM) and an artificial neural network (ANN). The direction of this study is to evaluate the complexity of this prediction domain using both machine learning models. Both SVM/ANN are commonly deployed machine learning models in most application domains. The authors investigate this problem from both regression and classification perspectives, implementing various kernel functions associated with the SVM such as radial basis function (RBF) and artificial neural network (ANN). Despite the good separation between classes and good regression on the training set using more nonlinear models, it is observed that overfitting occurs when evaluating on independent test sets for both regression and classification (especially in cases involving SVM). The authors found that more nonlinear kernels (SVM) yielded better performance thereby indicating that future research may benefit from more linear models. The study postulates that machine learning models should be chosen based on the general suitability, data linearity, and complexity to achieve accurate predictions of transformer health index (HI). The results of this study may provide insights for future research in developing models that can accurately predict transformer health.

As the world moves towards a greener future, adoption of renewable energy sources has become very popular and solar power has become one of the most desirable sources of energy. The performance and efficiency of solar panel modules are very much affected by environmental parameters such as temperature, irradiance and dust. Many studies found that the accumulation of dust on the surface of solar panels has seriously reduced the output power of a solar panel. Dust that has accumulated on the panels may reflect or refract the photons thus preventing them from reaching the surface of the panel. Thus, it has reduced the efficiency of solar panels by 3–25%. This study aims to design and fabricate a solar panel cleaning system. The system will be placed atop the solar panels. It consists of an on-board cleaning brush, water tank and control electronics. After the fabrication of the design, testing is done using an acrylic panel embedded with LDR sensors. The intensity of the light passing through the panel is measured before and after multiple cleaning passes. Based on the results, the cleanliness of the surface had significantly improved the efficiency of the panel from 60 to 98% after 3 cleaning passes using water jet.

This research study presents a computationally improved system using a pattern recognition (PR) algorithm to classify fingers movement based on data acquired by a surface EMG (sEMG) sensor when muscle is contracting. Ten subjects were involved in this investigation where the forearm’s muscle activities were acquired using two-channel sEMG placed at the flexor digitorum superficialis and extensor digitorum muscles. The focus of this study is to integrate the sensor with servo motors to control the movement of artificial limbs or prosthetics based on which muscle is at work. The work involved signal processing on raw sEMG signals, followed by multiple time domain feature extraction (TD). sEMG signal is then segmented using an overlapping window of size 250 ms and increments of 50 ms. The feature extraction was used to build up the convolutional neural network (CNN) which is used to train the classes of fingers movement. The dataset was split into 80% for training and 20% for testing the classifier. The CNN model was able to incorporate most of the data’s variability while maintaining an average classification accuracy of 98%.

Due to numerous data that is required to transfer in the information age, there is an increasing demand of computation and memory usage. Compressed sensing (CS) appears to be a promising technique in order to save both the computation and data storage. Iterative hard thresholding (IHT) is one of the signal recovery methods in the CS. Despite its fast computation, the IHT often delivers poor performance in the signal reconstruction accuracy. To solve this issue, we present an improved IHT in this work by using the fractional norm. Numerical simulation is demonstrated to illustrate better accuracy of the signal reconstruction than former approaches in various scenarios.

The drive for mobility has caused an increase in greenhouse gas emissions, leading to a shift toward the adoption of electric vehicles (EVs). These vehicles are powered by efficient electric motors, which offer reduced upkeep and enhanced operation. However, the unstable nature of electricity demand and the proliferation of EV charging equipment providers have posed significant challenges in EV charging. Furthermore, charging interruptions are commonplace, with electrical vehicle supply equipment (EVSE) serving as a mysterious black box for EV owners. By analyzing EVSE and EV data during charging, sensitive signals of charging termination can be discovered. The present research paper aims to investigate endurance data to uncover the root cause of EV charging interruption and to address uncertainties in EV charging.

Drones’ technology has become forefront in the industry development sector. It has been used in many applications such as forest fire monitoring, border security surveillance and delivery application. However, drone flights are limited to only a few minutes also high current consumption by drone’s motors resulting in requiring batteries with high capacity. This represents a challenge for the drones to efficiently serve their purpose, researchers show more interest in developing a drone wireless charging method for longer flight duration. Many factors are to be considered such as power transfer efficiency, number of transmitter coils, frequency range and transfer distance; all these factors have been reviewed in this research. This paper aims to investigate and review the methods of wireless power transfer (WPT) and its capability to charge a drone’s battery at all sizes. WPT is divided into three main types which are capacitive wireless power transfer (CWPT), radiative and non-radiative methods. Non-radiative technique is divided into three methods which are inductive wireless power transfer (IWPT), hybrid capacitive wirelesses and magnetic resonant wireless power transfer. The types of wireless charging technologies based on the distance between transmitter and receiver have been summarized. The advantages and disadvantages of each method have been introduced. Hybrid inductive and capacitive method using multiple transmitter antennas is a potential technology for the charging of drone’s battery. Design of the WPT station has been proposed, which consists of four transmitters’ antennas aims to increase the output power efficiency. Performance analysis of wireless power transfer module was conducted, and efficiency was calculated based on the distance between power transmitter (Tx) and power receiver (Rx). A maximum efficiency of 98% was achieved at 0 mm between Tx and Rx, when using 20Ω as a load resistance. A minimum efficiency of 86% was achieved at 0 mm between Tx and Rx, when 35 and 50 Ω were used as a load resistance respectively. The future development and future workforce strategies were proposed.

One such method can be used to assess the fuel rate for an internal combustion engine (ICE) vehicle within a specific area is to develop the driving cycle. For a full-battery electric vehicle (BEV), driving cycle is an important instrument for the evaluation of vehicle characteristics like energy consumption and range estimation. This research aims to experience the actual on-road driving of the selected routes and apply proper methods to construct a significant driving cycle that closely represents local driving patterns to evaluate the fuel rate of certain vehicles within the area. The study area was IIUM Gombak, and speed-time data was collected among selected local routes using a motorcycle. Two machine learning methods were used to construct the driving cycle which are k-means clustering and Markov chain. Results showed that the former method was better suited for the study area as it considered road geography as a significant aspect of traffic flow, where a fuel consumption of 4.50 L per 100 km was required. The study highlighted the importance of determining major aspects of traffic flow such as, but not limited to, road geography, work zones, and traffic volume, when choosing the method outcomes for a representative driving cycle.

Wireless communication networks for the Internet of Things (IoT) system face significant challenges in terms of reliability and security. Traditional networks rely on cryptographic protocols to address security, which primarily focus on upper layers in the OSI model without considering the physical layer. However, recent research has increasingly focused on physical layer security, as it offers a different approach to achieving secrecy in communication. The goal of physical layer security is to ensure that data transmission between legitimate nodes in a network remains confidential, even in the presence of malicious actors attempting to intercept the communication. One way to achieve secrecy is by utilizing the unique characteristics of the physical layer, such as thermal noise, interference, and fading channels. There are various approaches to physical layer security, and this paper focuses on information-theoretic methods for enhancing wireless system security against eavesdropping attacks. Error-correcting codes (ECC) also play a role in providing security, and various code classes can be employed for this purpose, such as polar codes and low-density parity-check (LDPC) codes. In this paper, we investigate the use of One-Step Majority Logic Decodable (OSMLD) codes in physical layer security for IoT applications. OSMLD codes are characterized by their low complexity in the majority decoding process. The main contributions of this work are the application of different types of OSMLD codes in the wiretap channel and the demonstration of the benefits of OSMLD codes in physical layer security for IoT applications.

Green wireless communications aim to reduce the environmental impact of wireless communication systems while maintaining or improving their performance. One technique used in wireless communication systems is spectrum sensing which is an enabling technique that provides information on spectrum availability for cognitive radio. Cyclostationary spectrum sensing is a particular sensing approach that takes use of the built-in periodicities characteristic to most man-made signals. However, when channel fading conditions are severe, the interference can affect primary users and the wireless communication systems consume a significant amount of energy and generate greenhouse gas emissions, leading to various environmental and health impacts to maintain quality. To combat this issue, intelligent reflecting surface aided cyclostationary spectrum sensing is proposed. Cases where the line of sight between the primary user and its destination is known, were investigated. Receiver Output Characteristic curves were produced with and without the use of intelligent reflecting surfaces in cyclostationary spectrum sensing to determine if it retains the information from the primary user in severe channel fading conditions. Simulation results verify that the use of intelligent reflecting surfaces can improve the performance of cyclostationary detection for spectrum sensing.

Long Short-Term Memory (LSTM) networks are widely recognized for their ability to capture and retain long-term dependencies within time series data, making them a valuable tool for dealing with complex relationships between elements over extended periods of time. This research proposes a KerasTuner-based LSTM network to predict future electricity consumption and maximum demand. Dataset used in this study is historical electricity consumption data of a plastic manufacturing plant in Malaysia, collected at 30-min intervals from 1st January 2017 to 31st December 2019. Both random selection and KerasTuner-based hyperparameter tuning were used to determine the best hyperparameters. The results demonstrated that the KerasTuner-based LSTM approach is effective in predicting future electricity consumption and captures the complex dependencies within the electricity consumption data. The evaluation metrics, training time, and limits of the future maximum demand indicated the effectiveness of the proposed model. This is proven when the proposed model outperformed other models and could improve the prediction accuracy while saving time. This research shows that the proposed model could serve as a valuable tool for predicting maximum electricity demand and could be applied in other industries to provide crucial insights for energy planning and management.

The emergence of large quantum computers running Shor’s algorithm threatens the security of several cryptographic schemes in current use, including electronic voting. As a consequence, many post-quantum candidates that are quantum-resistant are actively investigated. Post-quantum schemes-based hard lattice problems are particularly promising. In this paper, we present and implement a new quantum-resistant electronic voting scheme and prove its efficiency and security by studying its algebraic complexity. Our scheme is based on a combination of two techniques: hard lattice problems and homomorphic encryption with Fan and Vercauteren system.

Electromagnetic interference (EMI) produced by power electronic components like DC-DC converters and associated switching components is a major problem in electric vehicles. Most techniques for mitigation of EMI are to use a single filter. This study focuses on combining filter circuits to mitigate EMI. Two filter combinations are considered in this paper. Module 1 is a combination of snubber circuit and line impedance stabilization network (LISN), and Module 2 is a PI filter with snubber circuit and LISN. Simulation studies are done for a buck converter with the filter modules and the filter performances are compared.

Bioimpedance spectroscopy (BIS) is a frequency-based technique which has been extensively used in medical and agricultural applications. BIS involves estimation of the power spectral density (PSD), to determine the frequency response of biological material modelled as an impedance. This PSD is then used to determine other biological parameters of interest. A deep-learning-based BIS technique is proposed for body area networks (BAN) using the IEEE 802.15.6 Human Body Communications (HBC) specification. The proposed technique uses the start of frame delimiter (SFD) from the HBC physical layer (PHY) frame to perform BIS. A deep-learning-based strain gauge was developed based on this SFD-BIS framework as a proof of concept. Results show an average RMSE of around 3.87 N and an average absolute error of 1.20 N for all noise variations explored. The relatively low error showed the feasibility of this approach validating this proof of concept. This has great potential when applied to cyber-physical systems.

In this work, we present a new bit-flipping iterative Low-Density Parity-Check (LDPC) decoder based on reliability. The proposed decoder, called Reliability-based Single Bit-Flipping Decoding Algorithm, introduces the idea of separating the hard decision from the soft decision while processing flipping metrics, where the initial belief about received signals during decoding operations will efficiently contribute to the refinement of the final decision. We demonstrate through simulations over the binary-input Additive White Gaussian Noise (AWGN) channel that our decoder offers efficient trade-offs between BER performance and decoding complexity. It exhibits better decoding performance than known BF algorithms like Gradient Descent Bit Flipping (GDBF) and Single Bit Flipping (SBF) and presents a polynomial complexity, which proves its efficiency.

Small and medium-sized based manufacturing industry, despite being the most important industry in an economy is still grappling with unsteady processes, futile effort on controlling disturbances and erroneous deviation of its end-product resulting in waste of raw materials. Thus, necessitates to push Industry 4.0 (I4.0) higher at their agenda to increase manufacturing efficiency. Considering this, the manufacturer in this study is a mid-tier supplier of high-quality wood door for residential spaces and who frequently deals with modular and customizable door sets. This study makes the following contributions: (1) develop a microcontroller-driven automated system to accurately measure dimensions of door sets; (2) establish a communication to store-retrieve raw data using IoT technologies and (3) develop graphical user interface as diagnostic tool that generates statistical reports as data analytics. A low-cost ESP8266 (ESP) microcontroller Wi-Fi module interfaces with a rotary encoder used to monitor the displacement of door set for error deviation. Data is sent using IoT-based ThingSpeak application. Results satisfactorily record accuracy on error deviation which set between 0 and 0.6 mm based on the percentages of doors. Statistical reports, such as error deviation, percentage of doors within the allowed error, and production rate were remotely accessed to gauge productivity status. Emerging technologies of automation and Internet of Things that underpin concepts introduced by I4.0 are viewed as an antidote to manufacturing issues as it facilitates the creation of smart monitoring and controlling system for improved productivity yield.

Empty fruit bunch (EFB) is among the major agricultural waste from the oil palm industry, but its cellulose extract has the potential to clean up hazardous and large-scale oil spills. This research aims to extract cellulose from the oil palm empty fruit bunch fiber and study the water absorption rate of the cellulose. First, extraction processes were conducted using 15% sodium hydroxide (NaOH) and 10% hydrogen peroxide (H₂O₂). Then, the morphology of empty fruit bunch fiber, crystalline index, and phase identification was characterized using scanning electron microscopy (SEM), and X-Ray diffraction (XRD), respectively. The extracted cellulose was found to have a smaller diameter (8–50 µm). Alkaline treatment assisted in the removal of hemicellulose, while bleaching aided in the removal of lignin and discoloration. As a result, the extracted cellulose showed a high crystalline index of 30.67%. Moreover, the water absorption rate of the extracted cellulose is calculated, and the maximum water absorption rate of the extracted cellulose was recorded at 269.50%. Hence, this research provides an efficient method for extracting cellulose from oil palm empty fruit bunch fiber and the highest yield of water absorption from the extracted cellulose, primarily used as aerogel in oil spill cleaning.

Efficient detection and segmentation of wall cracks play a crucial role in building maintenance and construction. However, the implementation of vision transformers (ViT) for crack classification often presents challenges due to its high computational complexity, making it unsuitable for deployment on low-efficiency devices. To address this issue, we propose a novel approach that leverages knowledge distillation (KD) to create a computationally efficient ensemble model comprising a convolutional neural network (CNN) and ViT. In our framework, the teacher model is the ViT, which possesses exceptional classification capabilities, while the student model is a CNN designed to reduce complexity and enhance inference efficiency. By employing KD, we transfer knowledge from the ViT to the CNN, enabling the student model to approximate the performance of the more complex teacher model. This approach reduces the training time and computational requirements without significantly sacrificing classification accuracy. Following the classification stage, we employ a UNet segmentation model on the crack-detected images to accurately identify and delineate the damaged areas within a cracked surface. By analyzing the segmented images, we can calculate essential metrics such as the percentage of crack area and the length of the cracks. These metrics provide valuable insights into the severity of the cracks, facilitating the development of effective strategies for repair and prevention. Experimental results on a diverse dataset demonstrate that our ensemble model achieves competitive crack detection and segmentation performance while maintaining efficiency. The proposed approach not only reduces the complexity associated with ViT deployment but also provides an accurate and comprehensive analysis of wall crack severity. These findings have significant implications for building maintenance and construction industries, enabling proactive measures to mitigate structural damage and ensure safer and more sustainable infrastructure.

Titanium alloy (Ti-6Al-4V) is a biomaterial which has an incomparable weight-strength ratio, corrosion resistance, and thermal resistance than any other commonly used metals. It is one of the reasons for using it in implants, aerospace, defense applications. The induced cutting temperature during machining is higher as it is one of the hard materials. It is chemically active, and it tends to react with tool material. But it must be confronted as it has wide industrial applications. This research was intended to investigate the cutting temperature during turning of Ti alloy and analyze the significance of cutting parameters such as cutting speed, feed rate, and depth of cut. The Taguchi L₂₇ orthogonal array was utilized in conducting experiments. The range of parameters and their effect on cutting temperature were statistically analyzed. At this end, a predictive model using ANFIS was developed and validated. Performance metric RMSE was found to be very minimal (9.31994e-05).

Structural Health Monitoring (SHM) is a critical task in the management of buildings, bridges, and civil infrastructures. With the evolution of Artificial Intelligence (AI), SHM through AI is gaining popularity from past few years. Therefore, this work aims in providing SHM solution for wall defects through computer vision techniques. The work presents the development of mobile application to fully automate SHM procedure for inspection of building, finding defects, and its analysis. For crack detection purpose, MobileNetV2 as an algorithm is incorporated to determine two anomaly condition of whether the wall has crack or not. Mobilenet showed good result with an accuracy of 95%. The defect detected image is then segmented with the U-Net algorithm that resulted a Dice Similarity of 0.88, to determine the area of the damaged portion. The work also comprises real-time SHM of the walls using You Only Look Once version 5 (YOLOv5) for various defects such as crack, spalling, and seepage. To integrate these algorithms into mobile application, the models are converted into TensorFlow Lite (TFLite) format with the aim to reduce the size of the model and computational power. The proposed software has potential to significantly improve efficiency and effectiveness of SHM, providing a user-friendly and accessible solution for defect detection and analysis on mobile devices.

The utilization of federated learning in unmanned aerial vehicle (UAV) networks facilitates collaborative training of machine learning models by multiple UAVs while ensuring privacy preservation. However, the existing solutions for securing local model updates, such as heavy computation homomorphic encryption, secure multiparty computation, and differential privacy, are not feasible for UAV networks with limited computational resources and data capacity. To address this issue, a new lightweight protocol has been proposed, which protects the integrity of non-privacy sensitive local model updates in UAV networks using the MAVLink protocol over WiFi and LoRa communication technologies. The lightweight protocol has been designed using the SHA256 hash function and integrated with a public blockchain for integrity verification purposes. A proof of concept has been presented to demonstrate the proposed protocol’s capability of protecting the integrity of local model updates in UAV networks. Furthermore, the security of the proposed protocol has been analyzed and shown to be secure against adversary-in-the-middle and replay attacks. The computation cost of the proposed protocol has also been evaluated and found to be supported by UAV networks.

This research aims to investigate the machinability of a SiC-reinforced Al6061-T6 composite by wire-cut electric discharge machining (Wire-EDM) and to develop a predictive model using artificial neural network (ANN). The machine parameters such as current (I), pulse-on time (T_on), pulse-off time (T_off), and wire feed (EF) are examined and optimized for a high material removal rate (MRR). The experiments are designed using Taguchi L₁₆ orthogonal array, and experiments are conducted at room temperature. The MRR obtained at different experiments is analyzed using statistical tools. An ANN model was then developed, and its performance was evaluated by comparing prediction results with experimental results. Visual graphs were used to show the combined impact of Wire-EDM factors on machinability performance. The suggested model reduces the time needed to set the process parameter values, improving production rate and process effectiveness.

Potato is the most widely grown and consumed food throughout the world. There are a number of potato crop diseases that affect production, and these diseases differ in symptoms, circumstances, and controls. Early detection and recognition of disease information can aid in disease prevention and production. This paper presents deep learning models using proposed CNN and pre-trained models for potato disease detection and classification. The proposed model is more efficient and accurate at detection and classification. To perform classification, we used two datasets: PLD and PlantVillage. The Xception CNN model serves as the foundation for our proposed model. It achieved 1.00 accuracy, 0.99 precision, 1.00 recall, and 0.99 F1-score on the PLD dataset. On the PlantVillage dataset, it achieved 1.00 accuracy, 1.00 precision, 1.00 recall, and 1.00 F1-score. We also compared the results of Inception-ResNet-V2, MobileNet-V2, VGG-19, Inception-V3, and Xception models with the performance of our proposed model. For three classes of potato leaves, the proposed CNN model produced more accurate results than other pre-trained models.

The primary objective of this research project was to investigate the potential application of wind energy as a solution for reducing charging time and extending the driving range of electric vehicles (EVs). To harness the benefits of the aerodynamic characteristics around the vehicle, a vertical-axis wind turbine was designed and positioned beneath the vehicle, with an inlet located at the front grille. The turbine blades were designed using the NACA4415 airfoil profile and fabricated using 3D printing technology. Experimental measurements were conducted in a subsonic wind tunnel to evaluate the velocity-to-voltage relationship. The experimental results revealed a significant increase in speeds, reaching up to 50% improvement when a nozzle was incorporated into the system. However, at a wind speed of 11.14 m/s, the speed increment dropped to 10% due to wind speed instability. At lower wind speeds of 12 m/s, the power output was measured at 11.28 W, which is considered relatively low. The utilization of a nozzle at the turbine inlet successfully enhanced the wind velocity, resulting in a power output increase of up to 19 kW. The highest recorded wind velocity was 33.33 m/s. Consequently, the integration of a nozzle at the inlet of the wind turbine demonstrated its potential to provide a greater supply of wind energy, thereby generating higher power output within the EV system.

This study investigates the performance characteristics of a triboelectric generator utilizing a polydimethylsiloxane (PDMS)-aluminum (Al) bilayer system operating in contact-separation mode. The primary objective is to evaluate the influence of surface modification on the output voltage of the generator. The experimental procedure involves the preparation of an Al layer and PDMS samples with a 4 × 4 cm² area and a thickness of 2 mm, which were subjected to surface modifications using abrasive paper with grit sizes of 240 CW, 400 CW, 600 CW, and 800 CW. The findings of this study demonstrate that treating the PDMS sample with a 600 CW grade as the negative tribo-material, in combination with treating the Al layer with an 800 CW grade as the p-type material, resulted in a significant 50% increase in output voltage compared with the untreated configuration. This enhanced performance can be attributed to the formation of micro-structured surfaces on both the PDMS and Al layers through surface treatments which facilitates electron transfer within the triboelectric generator. The outcomes of this investigation provide valuable insights into the impact of surface treatment on the performance of PDMS-Al triboelectric generators. The findings contribute to the fundamental understanding of energy harvesting devices and offer opportunities for optimizing the design and fabrication of more efficient triboelectric generators.

An imbalance between generation and load often occurs in decentralized power generation systems. This imbalance causes voltage and frequency deviations, which affect the stable operation of the system. This paper proposes control strategies for the system sources to regulate the voltage and frequency by balancing the active and reactive power. Various scenarios of power generation and load-demand conditions are considered to investigate the performance of the proposed control strategies. A simulation was conducted using PSCAD/EMTDC software to assess the feasibility of the proposed control strategies. Results indicate that the suggested control strategies are effective in maintaining system voltage and frequency under various operating conditions. The embedded decoupled current control approach in the proposed control strategies realizes fast compensation for transient disturbances.

Small subcritical damage in composite materials can serve as warning signs for potential catastrophic failures, making their detection crucial. However, identifying these signs poses significant challenges due to the complex nature of composite structures. Meanwhile, recycled carbon fiber, which possesses inherent electrical conductivity, can be exploited as both a structural and sensing material in composite structures. By integrating recycled carbon fiber into composite materials, it becomes possible to monitor structural health and detect subtle damage, such as delamination, fiber breakage, or matrix cracking in real-time. This paper explores the potential of recycled carbon fiber as a dual-purpose material as both structural and damage sensing material, highlighting the opportunities it presents for enhancing safety and performance in various industries.