TIET Digital Repository Thapar Institute of Engineering & Technology
Not a member yet
6889 research outputs found
Sort by
An Improved Construction of Büchi Automata for Scientific Applications
No full textOmega automata or infinite words automata is a finite machine that works on words or strings of
infinite length. There is no acceptance condition as we have for NFA or DFA because of the words
or strings of infinite length. Omega automata can be classified into five categories based on
acceptance criteria: Büchi, Co-Büchi, Muller, Rabin, and Streett automata. All of these listed
omega automata have equal power but different acceptance conditions. Omega automata play an
essential part in verifying and synthesizing reactive systems. Omega automata is characterized by
two aspects acceptance condition and determinism. This thesis investigates the various classes of
omega automata and their applications by implementing multiple algorithms and models. The main
objective is to generate an algorithm for transforming omega regular expression to Büchi
automata.
This dissertation is dedicated to designing and expanding an algorithm that converts an omega
regular expression to its corresponding omega automata using a minimal approach. The proposed
algorithm improved the traditional canonical derivatives approach and introduced the canonical
factors reducing the no. of states in the resultant Büchi automaton. Ilie & Yu’s approach for
generating the Büchi automaton from -regular expressions was also revised.
Additionally, we employed the Büchi automaton model to illustrate the stages of cancer
progression. Furthermore, this thesis proposes a model that elucidates the process of tumor
formation by focusing on the activation/deactivation of the tumor suppressor gene as a result of a
mutation in the DNA nucleotide sequence, employing Büchi automata. The model
comprehensively explains the DNA mutation process and the origin of mutated cells, which are
subsequently followed by cellular proliferation.
We have also applied the concept of a Quantum Support Vector machine on the Breast Cancer
Wisconsin Data using Pennylane Default Simulator, Quantum Amazon Simulator State Vector,
and Quantum Amazon Simulator Density Matrix. A comparative study of amplitude encoding,
angle encoding, Z-Feature Map, ZZ-Feature Map, and poly feature map was conducted
थापर इंस्टीट्यूट ऑफ इंजीनियरिंग एंड टेक्नोलॉजी में फर्स्ट रिस्पॉन्डर कार्यक्रम और प्राथमिक चिकित्सा प्रशिक्षण आयोजित
No full text
Approaches to solve linear programming problems with imprecise parameters
No full textIn this thesis, it is pointed out that
(i) The existing results (Marimuthu and Mahapatra, 2021; Jeevaraj, 2022) are not correct. Also, it is pointed out that the existing RM (Marimuthu and Mahapatra, 2021) as well as the existing RM (Jeevaraj, 2022), fails to distinguish two distinct GTrFNs. Hence, the existing RMs (Marimuthu and Mahapatra, 2021; Jeevaraj, 2022) are not appropriate. Furthermore, the correct results, corresponding to the existing results (Marimuthu and Mahapatra, 2021; Jeevaraj, 2022) are stated and proved. Finally, it is shown that the existing RMs (Marimuthu and Mahapatra, 2021; Jeevaraj, 2022) will never fail to distinguish two distinct GTrFNs having same heights. However, both the existing RMs (Marimuthu and Mahapatra, 2021; Jeevaraj, 2022) may fail to distinguish two distinct GTrFNs having different heights. Hence, these existing ranking methods can be used only to solve such LPPs with GTrFCs in which height of all the GTrFNs is same.
(ii) The existing methods (Jeevaraj 2021; Bihari et al., 2025) to solve IVFFMCDMPs with known attribute weights and to solve IVFFLPP is not valid. Therefore, to propose an IVFFLPP based method to solve IVFFMCDMPs with unknown attribute weights, firstly, there is a need to resolve the drawbacks of (i) the existing methods to solve IVFFMCDMPs with known attribute weights (ii) the existing method to solve IVFFLPP. Also, modified methods to solve IVFFMCDMPs with known attribute weights are proposed to resolve drawbacks of existing methods to solve IVFFMCDMPs with known attribute weights. Furthermore, it is pointed out that due to some challenges, it is not possible to resolve the drawbacks of existing method to solve IVFFLPP. Hence, it is not possible to propose an IVFFLPP based method to solve IVFFMCDMPs with unknown attribute weights.
(iii) The existing approach (Saghi et al., 2023) fails to find the correct TrHFN (representing the OV of FT HFLPP, LPP in which each element of the ObF is represented by a TrHFN and each of the remaining parameters is represented by a ReN). Also, the reasons for the failure of the existing approach (Saghi et al., 2023) are discussed. Furthermore, to overcome this limitation, a new approach (named as Mehar approach) is proposed to solve FT HFLPP. Finally, the correct TrHFN, representing the OV of the considered FT HFLPP, is obtained by the proposed Mehar approach.
(iv) Much computational efforts are required to solve ST HFLPPs (LPPs in which each decision variable as well as each element of resource vector is represented by a TrHFN and each of the remaining parameters is represented by a ReN) by the existing approach (Saghi et al., 2024). Also, to reduce the computational efforts, an alternative approach is proposed to solve ST HFLPPs. Furthermore, some other advantages of the PrAlApp
over Saghi et al.’s approach are discussed. Finally, a ST HFLPP, considered by Saghi et al. to illustrate their proposed approach, is solved by the PrAlApp.
(v) The existing approach (Ranjbar et al., 2020) fails to find correct OS of TT HFLPPs (LPPs in which each parameter except decision variable is represented by a TrHFN). Hence, it is inappropriate to use the existing approach (Ranjbar et al., 2020). Also, the reason for this inappropriateness is pointed out. Furthermore, to resolve the inappropriateness of the existing approach (Ranjbar et al., 2020), a modified approach is proposed to solve TT HFLPPs. Finally, the modified approach is illustrated with the help of a numerical example.
(vi) Tamilarasi and Paulraj (2022) have used incorrect definition of a single-valued TNeN to propose their RFn. Therefore, Tamilarasi and Paulraj (2022)’s RFn is not valid and hence, it is inappropriate to use Tamilarasi and Paulraj (2022)’s method for solving LPPs with NCs and CrDVrs. Also, it is shown that if in the method, used by Tamilarasi and Paulraj (2022) to obtain their proposed RFn, the correct definition (Seikh and Dutta,
2022) of a single-valued TNeN is considered then Tamilarasi and Paulraj (2022)’s method fails to find a RFn. Hence, it is not possible to resolve the inappropriateness of Tamilarasi and Paulraj (2022)’s RFn.
(vii) Hemalatha and Venkateswarlu (2023)’s ranking approach fails to distinguish two distinct PnFNs. Therefore, it is inappropriate to use Hemalatha and Venkateswarlu (2023)’s ranking approach
Design and Development of Charging Schemes for Light Electric Vehicles
No full textIn this research work, the design and implementation of various converter topologies integrated with dual energy sources for charging of light electric vehicles (LEVs) are presented. The topologies are broadly classified as unidirectional and bidirectional DC to DC converters. These converters are additionally classified into non-isolated, isolated, and bridgeless types.
This work presents a novel architecture for an on-board charging (OBC) system that integrates dual energy sources, viz., single-phase AC grid and solar PV. The system employs a Modified Single-Ended Primary-Inductor Converter (SEPIC) converter topology to facilitate Light Electric Vehicle (LEV) charging. A diode bridge rectifier is used to convert AC to DC from the AC mains. An improved CC-CV control technique is developed to ensure robust operation of the converter, maintaining unity power factor (UPF) operation. In the event of a grid outage, an integrated solar photovoltaic (PV) system efficiently charges the LEV battery using a Maximum Power Point Tracking (MPPT) converter, adapting to varying environmental conditions. The Modified SEPIC converter manages LEV charging, emphasizing enhanced efficiency, low conduction losses, reduced component count, and high gain. The designed system offers soft-starting features of the BLDC drive in propulsion mode without using any current and voltage sensors on the motor side. The performance of the system is tested by using the MATLAB simulation and validated by a hardware prototype, the results prove the improved performance of the advanced charging methodology by the proposed converter.
This work also proposes an efficient configuration for a solar-powered on-board charging system utilizing a coupled inductor and switched capacitor bidirectional high-gain DC to DC converter with Grid-to-Vehicle (G2V) and Vehicle-to-Grid (V2G) operations. The bidirectional power flow capability of an on-board charger (OBC) benefits utilities and enhances the functionality of light electric vehicles (LEVs). The design of an OBC consists of an active front-end converter (AFC) for bidirectional power flow and unity power factor (UPF) operations. A proposed coupled inductor bidirectional high-gain SEPIC converter and a switched-capacitor bidirectional high-gain ZETA converter are designed and developed for the DC-DC stage. The AFC restricts the THD of supply current within the limits specified in international standards. In the event of a grid outage, an integrated solar photovoltaic (PV) system efficiently charges the LEV battery using a Maximum Power Point Tracking (MPPT) converter, adapting to varying environmental conditions. In addition, the brushless DC (BLDC) motor is used as a traction motor in this work due to its unique features, such as high density, low cost, simple control, etc. The presented LEV with a charging system is simulated in the MATLAB/Simulink platform, and real-time validation is performed using the OPAL-RT platform. The results obtained through both the simulation and real-time prototype indicate the effectiveness of the developed charging schemes with the coupled inductor and switched capacitor converter. Moreover, it introduces the design and implementation of a high-efficiency bidirectional isolated integrated DC to DC converter intended for the optimal charging and discharging of Light Electric Vehicle (LEV) batteries, utilizing dual power sources. The proposed system supports both Grid-to-Vehicle (G2V) and Vehicle-to-Grid (V2G) operations, ensuring stable performance even during grid voltage disturbances, including sags, swells, and outages. To enhance the robustness of the controller, an advanced mixed second-order–third-order generalized integrator (IMSTOGI) control algorithm is introduced to facilitate reliable operation of the Active Front-End Converter (AFC) under grid disturbances. During normal grid conditions, the converter ensures unity power factor (UPF) and constant current performance. In the event of a grid outage, an integrated solar photovoltaic (PV) system efficiently charges the LEV battery using a Maximum Power Point Tracking (MPPT) converter, adapting to varying environmental conditions. The functionality and power management strategy of the system are validated through real-time experiments, showcasing its effectiveness, reliability, and potential for seamless integration with the smart grids and renewable energy sources. Both simulation and experimental results from an OPAL-RT prototype support the system’s economic and operational advantages, confirming the efficiency of the proposed advanced charging methodology with the isolated integrated converter.
Additionally, this work introduces the design and implementation of a modified bridgeless SEPIC AC to DC converter topology with single-stage operations to facilitate LEV charging. The developed system utilizes two energy sources such as solar PV and single-phase grid. In the event of a grid outage, an integrated solar photovoltaic (PV) system efficiently charges the LEV battery using a Maximum Power Point Tracking (MPPT) converter, adapting to varying environmental conditions. The developed bridgeless converter manages LEV charging, with an emphasis on enhanced efficiency, low conduction losses, reduced component count, and high gain. The designed system offers soft-starting features of the BLDC drive in propulsion mode without using any current and voltage sensors on the motor side. The performance of the system is tested by using the MATLAB simulation and validated by hardware prototype, the results prove the improved performance of the advanced charging methodology by the
proposed converter.
This research presents an in-depth exploration of advanced DC-to-DC converter architectures integrated with dual power sources, namely solar photovoltaic (PV) systems and single-phase AC grid supply. The proposed solutions, which include modified SEPIC, bridgeless SEPIC, and high-gain bidirectional converters utilizing coupled inductors and switched capacitors, support both unidirectional and bidirectional power transfer—enabling efficient Grid-to-Vehicle (G2V) and Vehicle-to-Grid (V2G) functionality. Advanced control strategies such as Maximum Power Point Tracking (MPPT), Improved Mixed Second-Third Order Generalized Integrator (IMSTOGI), and Constant Current-Constant Voltage (CC-CV) ensure stable and efficient performance under varying grid and environmental conditions. The integration of smart grid capabilities alongside BLDC motor propulsion demonstrates the system’s flexibility. Simulation studies conducted in MATLAB/Simulink, along with real-time validation using the OPAL-RT platform, confirm the reliability, efficiency, and practicality of the proposed converter designs for Light Electric Vehicle (LEV) charging applications
Enhancing Performance and Energy Optimization in Serverless Computing
No full textServerless computing has been recognized as a transformative paradigm within cloud computing, offering Function-as-a-Service (FaaS) capabilities that allow developers to deploy applications without managing underlying infrastructure. Despite its advantages in scalability and cost-effectiveness, serverless computing still faces significant challenges related to workload unpredictability, inefficient resource utilization, energy consumption, and a lack of intelligent performance modeling. These issues are especially critical in serverless environments that demand dynamic autoscaling and precise workload management.
This thesis presents a comprehensive study of performance modeling and energy optimization in serverless systems, focusing on autoscaling mechanisms based on learning-driven approaches. Initially, a detailed literature review has been conducted to investigate the performance metrics in serverless computing—such as response time, cost, energy consumption, cold start frequency, resource utilization, and fault tolerance—and to assess the limitations of existing autoscaling strategies. The findings emphasize the need for intelligent, adaptive autoscaling models to efficiently manage fluctuating workloads to enhance Quality of Service (QoS) adherence.
The conventional approaches often fail to adapt effectively to sudden workload variations and lack the ability to learn from past performance data, which motivated the design of a more adaptive, learning-based autoscaling mechanism. Several models have been proposed and systematically evaluated throughout this research to address these concerns. Firstly, an auto-scalable model based on Q-learning has been introduced, enabling dynamic adjustment of compute resources in response to varying workload intensities. This model proves helpful in maximizing resource utilization by automatically scaling resources up or down as needed. The model continuously monitors incoming request rates and the current state of function instances, selecting scaling actions based on learned policies derived from historical performance data. The effectiveness of this model has been demonstrated on AWS Lambda, showing improvements in key metrics, including average response time reduced by 35.62\%, the mean number of idle instances minimized by 3.37\%, the probability of cold starts decreased by 38.5\%, and energy consumption lowered by 46.15\%.
While the Q-learning–based autoscalable model improved performance and energy consumption, its single-agent nature limited scalability and hindered coordinated decision-making across distributed instances. To overcome this, a Multi-Agent Deep Q-Learning (MADQL) model has been proposed to overcome the limitations of single-agent methods by enabling cooperative learning among agents. This model effectively mitigates issues of overutilization and underutilization by allowing agents to make real-time scaling decisions. Through extensive experimentation on a real-world e-commerce dataset using AWS Lambda, significant improvements in metrics have been revealed, with average response time reduced by 0.96\%, cost lowered by 1.46\%, energy consumption minimized by 2.43\%, throughput increased by 0.44\%, and CPU utilization improved by 15.79\% compared with the existing model.
Although MADQL provided cooperative learning and better workload distribution, it lacked predictive capabilities to anticipate workload surges, leading to reactive rather than proactive scaling. Building upon this, a hybrid learning model, LMP-Opt, has been introduced that integrates Long Short-Term Memory (LSTM) for workload prediction, Multi-Agent Deep Q-Learning (MADQL) for resource autoscaling, and Proximal Policy Optimization (PPO) for optimizing energy consumption through fine-tuning policy decisions. The LSTM component captures temporal workload patterns to facilitate predictive autoscaling. At the same time, MADQL dynamically allocates jobs by scaling resources up or down in response to workload fluctuations, and PPO has been introduced to refine these discrete actions into continuous ones, optimizing energy consumption and enhancing convergence.
The proposed model has been further validated on AWS Lambda and ServerlessSimPro using dynamic e-commerce workloads, demonstrating improvements of up to 6.09\% in response time, 6.14\% in energy consumption, and 7.82\% in cost, while improving CPU utilization by 4.93\% and reducing the required number of nodes by 5.59\%
Isolation of catechin-producing endophytic fungi from Camellia sinensis and influence of catechin nanoparticles on gut microbiota
No full textTea or Camellia sinensis (L.) Kuntze is the second most renowned and consumed beverage, after water. Tea has garnered significant interest as a functional food because of its numerous health advantages, especially in combating non-communicable chronic metabolic conditions. The health-promoting properties of tea are mainly ascribed to its bioactive catechins, viz., catechin, epicatechin (EC), epigallocatechin (EGC), epicatechin gallate (ECG), epigallocatechin-o-gallate (EGCG), and gallocatechin gallate (GCG). Clinical and preclinical investigations have demonstrated that catechin-rich green tea alleviates diet-induced metabolic distress through gut-level mechanisms, involving the attenuation of dysbiosis and strengthening of the intestinal barrier function to minimize endotoxin-induced inflammation along the gut-liver and gut-adipose axis. The drawbacks of plant-based catechin extraction, including lengthy cultivation periods, seasonal availability, climate-dependent yields, and the overharvesting of plant resources that jeopardize biodiversity, necessitate the search for sustainable and effective alternative sources of bioactive catechins due to their increasing demand for metabolic health-promoting purposes. One such alternative source of catechins could be the endophytic fungi that live in tea plant tissues without harming the host plants. Such symbionts represent an underexplored resource of novel bioactive chemicals and can generate a variety of secondary metabolites that are found in the host plants.
In this work, we examined the variety of endophytes from C. sinensis that can produce catechins. Various fungi were isolated from C. sinensis leaves, procured from the northern Himalayan region, India. However, only four isolates (CSPL6, CSPL6’, CSPL4, and CSPL5b) were found to produce catechin (381.48, 81.75, 12.28, and 166.40 μg/mg of extract, respectively) and EGCG (484.41, 67.29, 277.34, and 281.99 μg/mg of extract, respectively), as validated by high-performance liquid chromatography (HPLC). The isolated fungal strains were distinguished based on colony characteristics and molecular approaches as Pseudopestalotiopsis camelliae-sinensis, Aspergillus aculeatus, Phyllosticta capitalensis, and Didymella sinensis. These provide the first evidence of fungal endophytes that were able to synthesize catechins and EGCG from C. sinensis plant leaves. The gas chromatography-mass spectrometry (GC-MS)-based untargeted metabolomics indicated several pharmacologically important phytochemicals, mostly belonging to classes of citrates, tyrosols, pyridoxines, cinnamic acids, fatty acids, aminopyrimidine, and benzenetriol. The isolates that produced catechins had greatly enriched metabolic pathways related to the formation of butanoate, linoleic, and other fatty acids. The isolates were able to scavenge different intracellular free radicals to varying degrees. This study provides valuable insights regarding catechin-producing endophytes from the tea plant and their free-radical scavenging bioactivities, that could potentially serve to alleviate chronic diseases. Although all four isolates demonstrated effective scavenging activity and antioxidant potential against key intracellular free radicals, CSPL5b showed comparatively higher bioactivities than CSPL6, CSPL6’, and CSPL4. All four fungal extracts enhanced the growth of various probiotic Lactobacillus strains: L. sporogenes, L. rhamnosus, L. plantarum, and L. reuteri at low concentrations (1-8 μg/mL), indicating prebiotic effects that are typically linked to catechins.
The catechins are of immense scientific and industrial attention due to their prebiotic and antioxidant applications and have remarkable effects on gut health. But their effectiveness mainly depends on their absorption, bioavailability, stability, and their interaction with gut microbiota. So, to enhance the functions and properties of the catechins, the EGCG-chitosan nanoparticles (EGCG-CNPs) with sodium tripolyphosphate (TPP) were prepared by the ionic cross-linking method. The nanoparticles synthesized under optimum conditions demonstrated a 53% encapsulation effectiveness, an average particle size of 188.14±21.86 nm, polydispersity index (PDI=0.398), and zeta potential (38.15±2.56 mV). The synthesis of the composite nanoparticles and the formation of new hydrogen bonds between EGCG and chitosan were further demonstrated by the findings of the Scanning Electron Microscope (SEM), X-Ray Diffraction (XRD), and Fourier Transform Infrared (FTIR) analyses. EGCG-CNPs were more effective in preventing EGCG from degrading quickly, as validated by HPLC. Compared to its free form, EGCG-CNPs facilitated time-controlled sustained release from 0 to 12 h. The conformation and structure of chitosan may be altered due to the presence of EGCG. In vitro anaerobic fermentation of synthesized particles affected gut microbial composition, abundance, diversity, richness, and metabolic processes. EGCG-CNPs significantly boost microbial diversity and beneficial short-chain fatty acids (SCFAs)-producing commensals (e.g., Lactobacillus and Bifidobacterium), emphasizing the potential health benefits. The GC-MS-based untargeted metabolomics enabled the detection of diverse gut microbial metabolites such as indoles, amino acids, carbohydrates, phenolics, and sugars, which the EGCG-CNPs impacted.
Collectively, this study reports catechin-producing endophytes from tea leaves that not only possess potent bioactivities but can also be utilized as an alternative and sustainable source of bioactive phytochemicals, especially catechins. EGCG-loaded chitosan particles were produced using an ionic gelation technique. These nanoparticles can act as a unique delivery system for catechins, have great potential to improve the stability of EGCG, protect from gut microbiota-dependent metabolism, and boost the beneficial gut microbiota
Segregation, Flow Properties and Usability of Construction Materials Using Power Plant Solid Waste
No full textA vacuum-based segregation tester was designed to understand how bulk solid materials separate under controlled lab conditions. The tester uses a vacuum-driven airflow system to show how particles can be sorted into fine and coarse groups. This helps us grasp segregation behaviors and material handling traits. The design provides a basis for studying how particles interact and finding ways to improve segregation efficiency through better airflow and structural changes. This concept points to the potential for using such a tester in future experiments for classifying materials and construction purposes.
This study further investigates the flow properties and segregation tendency of pond ash (PA), pond fly ash (PFA), and flue gas desulfurization gypsum (FGDG), and further investigates the influence of these properties on the mechanical and durability properties of mortar made using these industrial by
products. Low-cohesion materials such as PA and FGDG improve workability and transport efficiency. Segregation analysis reveals that PA replacement in mortar leads to particle redistribution, affecting bulk density and particle size distribution, particularly at higher replacement levels. However, there is no significant segregation that affects the dry-mix for construction applications. The major focus of this study is to replace fine river sand with pond ash in the production of plaster. In this study, the physical, mechanical, and microstructure characteristics of mortar are investigated for different replacement levels of sand with pond ash. The findings demonstrate that raising the replacement level lowers the dry density of pond ash mixes. This is because pond ash is more porous and absorbs more water. The study confirms that up to 20% PA replacement enhances compressive and flexural strengths while reducing density, making it suitable for lightweight concrete (Novak 2017) and prefabricated elements. However, excessive replacement (≥30%) results in strength reduction due to increased porosity and ettringite formation. The findings indicate that raising the replacement level lowers the dry density of pond ash mixes, which is caused by the porosity and increased water absorption of pond ash. Notably, the compressive and flexural strengths are greater than those of control specimens at a 20% replacement rate. Conversely, the achieved strength exceeds the strength required by the Indian code, but the drying shrinkage is 31% less than the control. SEM-EDXS demonstrates that samples with a 20% pond
replacement have a higher C-S-H gel density, which results in higher compressive and flexural strengths. According to XRD results, the presence of sodium sulphate is the reason for the 40% pond ash replacement's decreased strength. Based on this study, it concluded that fine pond ash can partially replace river sand for internal plastering, as well as for making mortar and blocks.
On the other hand, adding FGDG (more than 5%) to the mixture produces most of the ettringite, sulfate and aluminate hydrates, and it shows a higher degree of expansion in the mixture compared to the control, which also reduces the strength. Therefore, the potential of combined pond fly ash (PFA) and flue gas desulfurization gypsum (FGDG) is investigated as a partial replacement for cement by mixing them with fine river sand for mortar development. Due to their finer particles and higher surface area,mixes with PFA and FGDG require more water for the same consistency, yet still show reduced flowability once hydrated. While dry mixes remain uniformly distributed with negligible segregation effect, water addition increases cohesion and lowers workability, which directly affects strength. Experimental results show that mortar containing pond fly ash, FGDG, and cement concentrations of 5 wt.%, 5 wt.%, and 90 wt.% yields the maximum compressive strength (8.9 MPa) and flexural strength (3.4 MPa) after 28 days. This mortar has 12.6% higher compressive strength and 48% lower shrinkage in comparison with control specimens. A reduction in shrinkage is attributed to the denser structure of the mortar, as the calcium silicate hydrate (C-S-H) gel is found to exist in the SEM image and is also identified through XRD studies. Higher concentrations of SO3 were observed to decrease the strength of mortar, highlighting the significance of maintaining the recommended level. FGDG causes an increase in the amount of CaSO4.2H2O, which dilutes the cementitious matrix, reducing its compressive strength. This is because FGDG has sulfate ions that react with some parts of the green mortar, like the aluminite phase of cement, to make ettringite (CaSO4.2H2O). The formation of expansive ettringite results in volume expansion, cracking, and deterioration of green mortar. The application of this mortar can be in the internal plaster, bricks, and masonry. These findings underscore the strategic incorporation of industrial by-products in sustainable construction, optimizing resource efficiency while maintaining mechanical performance and durability. Future research should explore higher replacement levels, long term field performance, and life cycle analysis, promoting the broader application of PA, PFA, and FGDG in eco-friendly infrastructure