Advances in Research

Thermal Management, Packaging, and Electro-thermal Co-design of Wide- and Ultra-Wide-Bandgap Power Devices

Neha Singh, Deep Sehgal — Wed, 14 Oct 2026 00:00:00 +0000

Background: Wide- and ultra-wide-bandgap (WBG/UWBG) power devices offer higher blocking-field capabilities and faster switching, but increasing power density presents significant thermal management and packaging challenges. Heat dissipation depends on semiconductor properties, thermal interfaces, package geometry and cooling conditions. However, translating material- and interface-level thermal improvements into reliable device and converter performance remains challenging because of differences in operating conditions and electro-thermal constraints.

Aims: To assess thermal transport, packaging and electro-thermal co-design constraints in WBG/UWBG power devices.

Study Design: Focused narrative review.

Methodology: Thirty journal articles were compared by structure, measurement or modelling methods, operating conditions and evidence limitations. The search was updated on 7 August 2026.

Findings: Interfaces, anisotropy, package geometry and duty cycle can outweigh bulk conductivity advantages. Thermal improvements can coexist with electrical degradation. Interface measurements, simulations and pulsed demonstrations do not establish continuous converter performance or service life.

Conclusion: Coupled thermal, electrical and packaging assessment is essential. The priority gaps are matched-condition comparisons, manufacturing variability and durability under representative converter operation.

Predictive Sentiment Analysis and Proactive Service Intelligence: Implications for Long-Term Customer Loyalty and Retention-A Scoping Review

Thu, 15 Oct 2026 00:00:00 +0000

Background: Predictive sentiment analysis and proactive service intelligence are increasingly used to identify deteriorating customer relationships and support targeted service responses, yet evidence is distributed across marketing, customer relationship management, service research and business analytics.

Objective: This scoping review maps the evidence linking sentiment analysis, predictive customer analytics and proactive service with customer loyalty and retention and develops an integrated conceptual synthesis.

Methods: The review was guided by established scoping-review methodology and PRISMA-ScR. Nine multidisciplinary databases were searched, supplemented by backward citation searching and citation tracking. Evidence was charted by study design, industry, technology, data type, analytical method, customer-intelligence construct, service mechanism and relationship outcome.

Results: A total of 515 records were identified, 60 full-text reports were assessed and 32 sources were included. The mapped evidence indicates that textual sentiment signals can add predictive value when combined with behavioural and transactional data; AI-enabled customer analytics can support churn-risk identification and personalised service; and proactive service may translate customer intelligence into timely interventions. However, prediction accuracy should not be equated with intervention effectiveness, and direct longitudinal evidence for durable loyalty and retention remains limited. Trust, privacy, explainability, human oversight and responsible data use emerged as important boundary conditions.

Conclusions: The literature supports a conceptual pathway from customer data to sentiment intelligence, predictive risk assessment, proactive intervention and relationship outcomes, but the pathway should be interpreted as an evidence-informed synthesis rather than a validated causal model. Future research should prioritise longitudinal and experimental evaluations of intervention effects and heterogeneous treatment response.

A Boundary-Adapted Crank–Nicolson-Induced Mamadu–Njoseh Basis Functions for the Solution of Parabolic Partial Differential Equations

Thu, 15 Oct 2026 00:00:00 +0000

A boundary-adapted Crank–Nicolson-induced Mamadu–Njoseh basis-functions (CN-MNBF) framework is developed for the numerical solution of one-dimensional parabolic partial differential equations. The method combines a weighted Mamadu–Njoseh polynomial approximation in space with Crank–Nicolson time discretization. To enforce homogeneous Dirichlet conditions exactly, the numerical implementation uses boundary-adapted trial functions obtained by multiplying each Mamadu–Njoseh polynomial (MNP) by the boundary factor (1 - x²) on the transformed interval [-1,1]. The resulting mass and spatial-operator matrices are assembled directly in this constrained space. For the diffusion problem, a weighted energy argument establishes dissipativity and the associated Crank–Nicolson update is contractive in the mass-matrix norm. The fully discrete error splits into spatial approximation and temporal discretization components, with the temporal component second-order under standard smoothness assumptions. Numerical illustrations show decreasing spatial error and, for a sufficiently rich spatial approximation, approximately second-order temporal convergence. We interpret the induced basis as an operator-adapted representation of the Crank–Nicolson evolution rather than a new finite-dimensional spatial span.

Automated Genetic Algorithm-Based Impedance Bandwidth Optimization of a Triangular U-Slotted Microstrip Patch Antenna Using Python and HFSS

Mitesh Upreti, Sanjay Mathur, Neha Singh — Thu, 15 Oct 2026 00:00:00 +0000

Microstrip patch antenna optimisation using full-wave electromagnetic simulations can become computationally expensive when multiple geometrical parameters are involved. This paper presents a Python–PyAEDT-based automated optimisation framework for enhancing the impedance bandwidth of a compact triangular U-slotted microstrip patch antenna. The antenna was initially designed for operation around 5 GHz on a 40 mm × 40 mm substrate and was subsequently modified to a compact 24 mm × 30 mm FR-4 substrate. The final configuration incorporates additional triangular and circular patch elements, a U-shaped slot, and a modified ground structure. Following the initial antenna modelling and parameterisation in Ansys High Frequency Structure Simulator (HFSS), the framework automatically updates six geometrical parameters, executes electromagnetic simulations, extracts the reflection coefficient (S₁₁), and calculates the continuous −10 dB impedance bandwidth. Genetic Algorithm (GA)-based optimisation and script-based random search were each evaluated using 100 HFSS simulations. The GA achieved an impedance bandwidth of 2357.65 MHz over approximately 3.202–5.559 GHz, with a minimum S₁₁ of −24.128 dB. In comparison, random search achieved a bandwidth of 2260.37 MHz over approximately 3.459–5.720 GHz, with a minimum S₁₁ of −25.852 dB. The GA-based solution provided an impedance bandwidth 4.30% greater than that of the random-search solution in the recorded runs, whereas random search produced a more negative minimum S₁₁. The recorded total wall-clock times were 32.42 minutes for GA and 50.34 minutes for random search. These results demonstrate the feasibility of automating antenna optimisation and comparing alternative search strategies under an equal simulation budget. However, because each method was evaluated in a single run and employed a different candidate-selection criterion, repeated trials are required to establish the robustness of the observed performance difference.