This study examines the potential of generative artificial intelligence (AI) in architectural design processes, focusing specifically on the energy-efficient renovation of building envelopes. The research explores the use of AI-based image generation tools to assist in the conceptual design phase of ventilated façade systems aimed at improving the thermal performance of existing buildings.

Based on photographic documentation of façades—obtained through conventional photography and photogrammetric imaging (SFM) to ensure geometric accuracy—the proposed methodology generates multiple design alternatives with AI assistance. These alternatives are constrained by a predefined catalog of construction products and materials provided by façade system manufacturers. The objective is to evaluate the current capability of generative AI tools to produce, within a reduced timeframe, multiple façade design options that are both visually coherent and technically feasible, in accordance with specific constructive and material parameters.

The results are assessed according to both architectural criteria—such as aesthetic consistency, contextual adaptation, and formal diversity—and technical parameters, including constructability, compliance with energy efficiency requirements, and correspondence with manufacturers’ specifications.

This research seeks to establish a framework for integrating generative AI into the early design stages of energy retrofit projects, providing architects with intelligent and responsive tools that enhance creative exploration without compromising technical rigor. Ultimately, this work contributes to the ongoing reflection on the role of AI in intelligent design processes, proposing its use as a collaborative and adaptive system that bridges aesthetic decision-making with energy-conscious architectural design.

The integration of renewable energy sources into modern power systems is essential for reducing carbon emissions and improving energy sustainability. Hybrid renewable energy systems, especially photovoltaics (PVs) and wind turbines (WTs), present an attractive solution due to their complementary generation profiles. However, their optimal placement and sizing in electrical distribution networks remain challenging due to nonlinear constraints and multiple objectives such as minimizing power losses, improving voltage profiles, and reducing operational costs. Advanced optimization techniques are therefore critical to achieving these goals.

This work presents an Adaptive Fuzzy Particle Swarm Optimization (AFPSO) algorithm applied for optimal integration of hybrid PV and wind energy systems in Direct Current electrical distribution networks. The proposed method, tested on a multi-objective function that adressed three objectives, minimization of active power losses, improvement of voltage deviation, and reduction of cost, is applied to the IEEE 69 bus system. AFPSO incorporates fuzzy logic to adaptively regulate PSO parameters, enhancing exploration and exploitation capabilities.

Comparative numerical comparison with Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), Grey Wolf Optimizer (GWO), and Whale Optimization Algorithm (WOA) algorithms is performed; the comparison reveals that AFPSO achieves the best results in terms of power loss reduction, voltage profile enhancement, and economic performance. The approach demonstrates strong potential for supporting efficient and reliable renewable integration in DC distribution systems.

In this study, a novel optimization framework was developed to minimize the Net Present Cost (NPC) of hybrid renewable energy systems under realistic Somali conditions. The study evaluated four configurations using hourly load, solar radiation, and temperature data: Diesel-only, Diesel/PV, PV/Battery, and Diesel/PV/Battery systems. The main objective was to identify the most techno-economically viable and environmentally sustainable option capable of supplying a commercial load with a 15.8 kW peak demand. The optimization was conducted in Python through a single-objective cost function constrained by system reliability, with a Loss of Power Supply Probability ≤ 0.05 and total capacity maintained between 21–25 kW. The analysis considered capital, replacement, and fuel costs with component lifetimes ranging from 10 to 20 years and a discount rate of 8%. The results indicated that the Diesel + PV + Battery configuration achieved the best overall performance, with an NPC of $286,746, LCOE of $0.296/kWh, reliability of 95.5%, and a renewable fraction of 74.6%, while reducing CO₂ emissions to 987 t. In comparison, the diesel-only system recorded the highest cost, with an NPC of USD 345,140 and emissions of about 1,338 t CO₂. The proposed optimization method demonstrates a practical and scalable approach for low-cost hybridization in resource-constrained regions, enhancing both economic efficiency and environmental sustainability of local microgrids.

As global urbanization accelerates, the demand for energy-efficient building environments continues to rise. Heating, ventilation, and air-conditioning (HVAC) systems account for nearly half of a building’s total energy consumption, highlighting the urgent need for intelligent optimization. This paper presents Gree Electric’s advancements in integrating artificial intelligence (AI) into HVAC systems to enhance energy efficiency and environmental comfort in built environments. The study introduces a multi-layer AI optimization framework combining real-time data analytics, adaptive control algorithms, and predictive maintenance to minimize energy waste across residential, commercial, and urban applications. Utilizing deep learning models and IoT-enabled sensor networks, Gree’s system dynamically adjusts temperature, humidity, and airflow based on occupant behavior, ambient conditions, and energy pricing signals. Experimental simulations and pilot installations demonstrate reductions of up to 25% in overall energy consumption while maintaining high user comfort and system reliability. This research also discusses Gree’s broader vision of AI-enhanced urban climate management, including the integration of smart HVAC with renewable energy systems and building energy management platforms. By aligning AI design strategies with sustainable architecture and smart city initiatives, Gree’s approach illustrates how data-driven climate control technologies can significantly contribute to low-carbon, intelligent urban environments. Future research will explore large-scale implementation models across diverse climatic zones, integration with district energy networks, and the development of autonomous control architectures for multi-building systems. The study contributes to advancing AI-based energy efficiency strategies and supports the global transition toward intelligent, resilient, and environmentally sustainable cities.

Energy monitoring and management represent fundamental components in modern electrical systems, particularly for applications demanding high precision, efficiency, and intelligent control capabilities. This research presents an advanced AI-enhanced smart DC energy management system that seamlessly integrates real-time monitoring with predictive control and dynamic load optimization. The system employs sophisticated machine learning algorithms to analyze energy consumption patterns, predict future usage trends, and automatically optimize load distribution across connected devices.

Through the implementation of neural network models and time-series analysis, the system achieves remarkable accuracy in forecasting energy demands while maintaining optimal power distribution efficiency. The architecture incorporates IoT connectivity through platforms like Blynk, enabling remote monitoring, real-time data visualization, and intelligent control capabilities. Designed specifically for renewable energy applications, including solar power systems and DC microgrids, the solution demonstrates exceptional performance in maximizing energy utilization while minimizing waste.

Experimental results validate the system's capability to reduce energy consumption by up to 30% through intelligent load scheduling and predictive optimization. The integration of artificial intelligence with conventional energy management approaches represents a significant advancement in sustainable energy technologies, offering both economic and environmental benefits. This innovative system provides a comprehensive solution for modern energy challenges, bridging the gap between traditional power management and cutting-edge AI technologies while maintaining reliability and user accessibility.

Courtyards have long been integral to vernacular architecture in Tamil Nadu, offering passive cooling, natural light, and social connectivity. However, modern residential buildings often overlook their climatic and energy benefits, leading to increase reliance on mechanical systems. This study explores how artificial intelligence (AI) can be integrated with Building Automation Systems (BAS) and responsive facade technologies to enhance the environmental performance of courtyard-based architecture across different climatic zones of Tamil Nadu.

By analyzing traditional courtyard geometries, proportions, and orientations, this research identifies key parameters influencing thermal comfort, ventilation, and daylight performance. By reducing energy demand and increasing energy efficiency, AI adds overall impact by making the system smart. These parameters are then simulated and optimized using AI algorithms to derive climate-specific courtyard configurations. The integration of BAS enables dynamic environmental control—adjusting facade openings, shading devices, and airflow in real time based on sensor data.

The proposed AI-BAS-Facade framework bridges vernacular design logic and digital intelligence, transforming static courtyards into adaptive, energy-efficient microclimates. The results aim to establish data-driven design guidelines for low-rise housing that reduce operational energy consumption while preserving regional architectural identity. This approach demonstrates how AI-enhanced traditional design strategies can lead to sustainable, context-responsive built environments in a rapidly urbanizing climate.

The increasing penetration of wind and solar resources in interconnected grids has made multi-area dynamic economic–emission dispatch (MADEED) a significantly more challenging optimization problem due to multi-temporal coupling, stochastic renewable fluctuations, inter-area power exchanges, and complex nonlinear constraints such as ramp limits, valve-point effects, and prohibited zones. Existing intelligent optimization techniques often show limited scalability, slow convergence, or poor constraint feasibility when applied to large multi-area systems. To overcome these limitations, this study proposes a hybrid Co-variance Guided Artificial Bee Colony (CG-ABC) and NSGA-II framework equipped with boundary-update implicit constraint handling and BWM–TOPSIS decision analytics for identifying the Best Compromise Solution (BCS) across conflicting economic and emission objectives.

The proposed CG-ABC enhances the original ABC by introducing a co-variance learning matrix, enabling bees to adapt search directions based on population distribution and correlated variable dynamics. This significantly improves global exploration and local exploitation in large-scale dispatch spaces. The hybridization with NSGA-II ensures robust non-dominated sorting, crowding-distance-based diversity management, and stable Pareto-front generation under multi-objective competition. An implicit constraint-handling mechanism based on boundary-repair and feasibility-preserving updates ensures strict adherence to tie-line flow limits, dynamic ramping limits, and inter-area balance constraints without requiring penalty parameter tuning.

The method is validated on multi-area test systems referenced in the recent literature, including two-area (6 units), three-area (10 units), and four-area (40 units) setups incorporating wind and solar generation modeled using Weibull and lognormal distributions. Comparative studies demonstrate that the proposed hybrid algorithm provides smoother, denser Pareto fronts, faster convergence, and superior feasibility preservation compared to recent high-performance algorithms reported in the literature. With BWM–TOPSIS aiding operator-level decision-making, the proposed framework offers a robust and sustainable tool for renewable-integrated multi-area power scheduling aligned with future low-carbon grid objectives.

Buildings consume nearly 40% of global energy and produce 36% of CO₂ emissions, making net-zero energy buildings (NZEBs) critical for climate mitigation. However, empirical validation of NZEB performance at commercial scale remains limited. This study presents a data-driven computational analysis of the Bullitt Center—a six-story, 50,071 sf office building in Seattle—to quantify its operational energy balance and ecosystem service value using AI-enhanced monitoring analytics.

Over 12 months, machine learning algorithms analyzed 8,760 hourly data points from IoT sensors monitoring energy consumption, solar generation, water usage, and occupant behavior. Statistical modeling and regression analysis identified performance drivers, while predictive algorithms optimized HVAC operations and demand response. Multi-criteria ecosystem services valuation employed computational modeling to quantify economic benefits over the building's 250-year design life.

The Bullitt Center achieved an Energy Use Intensity (EUI) of 9.4 kBTU/sf/year, performing 77% better than Seattle's baseline and producing 114,085 kWh annual energy surplus. AI-powered analytics revealed daylighting strategies reduced artificial lighting by 82%, while predictive HVAC controls maintained thermal comfort with minimal energy use. Smart rainwater harvesting and composting systems eliminated municipal water and wastewater connections. Ecosystem services valuation estimated USD 18.45 million in total benefits: energy efficiency (45%), carbon reduction (38%), and water conservation (12%). Benchmarking against 350 regional buildings confirmed superior performance.

This study demonstrates that integrated design processes, rigorous commissioning, and AI-driven monitoring enable genuine net-zero performance at commercial scale. The computational framework provides evidence-based guidelines for architects, engineers, and policymakers advancing climate-resilient intelligent building systems.

Introduction
The building and urban sectors account for a major share of global energy use, yet current energy-efficient design remains constrained by linear workflows and limited design-space exploration. This paper proposes an AI-driven parametric design framework that couples generative form-finding with performance-based evaluation to support the creation of ultra-low-energy buildings and districts. The focus is on retaining architectural authorship while leveraging artificial intelligence to navigate complex trade-offs between form, comfort, and energy demand.

Methods
A multi-stage workflow is developed using a parametric modelling environment linked to energy simulation and AI-based optimisation. Parametric building and district models encode key geometric and material variables (massing, orientation, envelope articulation, glazing ratios, and passive systems). Surrogate models trained on simulation datasets (e.g., regression and neural networks) predict energy and comfort metrics in real time. Multi-objective genetic algorithms and reinforcement learning agents generate and refine candidate configurations, while an architect-in-the-loop interface allows designers to steer and constrain the search.

Results
Applied to prototypical mid-rise building typologies and mixed-use district scenarios, the workflow rapidly produces large families of design variants and visualises Pareto fronts between energy use intensity, daylight availability, and floor area efficiency. Compared to conventional iterative design, the AI-enhanced approach uncovers solutions with substantially lower predicted annual energy demand while preserving spatial quality and façade diversity. At the district scale, coordinated optimisation of building forms and orientations demonstrates improved load smoothing and reduced peak demands.

Conclusions
The study shows that AI-driven parametric architectures can transform energy-oriented design from a late-stage validation exercise into an integrated generative driver of form at building and urban scales. The proposed workflow reinforces, rather than replaces, architectural judgement, and suggests a path towards explainable, performance-informed design cultures for ultra-low-energy built environments.

Introduction
Timber has re-emerged as a primary structural material in low-carbon architecture, yet its full potential remains underutilised when energy performance, structural behaviour, and fabrication constraints are treated in isolation. This paper presents a bio-inspired, AI-optimised design framework that couples material intelligence in engineered timber systems with energy-driven morphogenesis. Drawing on biological analogues of branching, lattices, and shells, this approach seeks to derive envelope and structural logics that simultaneously minimise operational, embodied energy while preserving architectural expression.

Methods
A parametric environment is developed for CLT and glulam assemblies, encoding grain direction, cross-section variation, connection logic, and panelisation rules. Bio-inspired typologies (e.g., venation-like shading lattices, ribbed shells, and branching supports) are translated into editable generative scripts. Energy and structural performance are evaluated via linked simulation engines for dynamic thermal behaviour, daylight, and FE analysis. Multi-objective evolutionary algorithms and machine learning surrogates (e.g., gradient-boosted trees) search and approximate the design space, optimising annual energy use, daylight autonomy, structural utilisation, and material volume. An architect-in-the-loop interface allows control over formal language, tectonic articulation, and programmatic requirements.

Results
When applied to pavilion-scale and mid-rise prototypes, the workflow yields timber morphologies that reduce annual heating and cooling loads by 20–30% relative to conventional rectilinear envelopes, while cutting material mass by 10–18% through structurally efficient, bio-inspired geometries. Shading lattices derived from venation logics improve daylight uniformity and reduce glare without sacrificing transparency. Sensitivity analyses show that joint topology and grain-aware orientation strongly influence both stiffness and thermal bridging, highlighting the necessity of integrated material and energy modelling.

Conclusions
This study demonstrates that bio-inspired, AI-optimised timber architectures can transform wood from a merely “sustainable” substitute into an active driver of energy-aware form-finding. Embedding material intelligence within AI-assisted parametric workflows enables architects to negotiate structural, environmental, and fabrication criteria as co-equal agents in design, advancing a holistic paradigm for high-performance timber architecture.