High energy consumption in residential buildings poses significant challenges, prompting governments to regulate this sector through comprehensive energy assessments and classification strategies. This study introduces a multi-layer perceptron artificial neural network (ANN) model to grade and predict energy consumption levels in residential buildings in Tabriz, Iran, based on their geometric and functional characteristics. This study uses the K-Nearest Neighbors (KNN) algorithm to classify energy consumption grades based on energy ratio (R-value). Six sample buildings were modeled using Rhinoceros 3D version 7 and Grasshopper version 1.0.0007 software to extract key energy-influencing factors. A parametric geometric model was developed for rapid data generation and validated against reference buildings to ensure reliability. Building classifications spanned areas of 40 to 300 square meters and heights of up to six stories, with energy evaluations conducted using EnergyPlus. The collected data informed the ANN model, enabling accurate predictions for existing and future constructions. The results demonstrate that the model achieves a remarkable prediction error of just 0.001, facilitating efficient energy assessments without requiring extensive modeling expertise. This research emphasizes the role of geometric features and natural lighting in energy consumption prediction, highlighting the model’s practicality for early design evaluations and architectural validations.

Modeling complex, non-stationary dynamics remains challenging for deterministic neural networks. We present the Chaos-Integrated Synaptic-Memory Network (CISMN), which embeds controlled chaos across four modules—Chaotic Memory Cells, Chaotic Plasticity Layers, Chaotic Synapse Layers, and a Chaotic Attention Mechanism—supplemented by a logistic-map learning-rate schedule. Rigorous stability analyses (Lyapunov exponents, boundedness proofs) and gradient-preservation guarantees underpin our design. In experiments, CISMN-1 on a synthetic acoustical regression dataset (541 samples, 22 features) achieved R ² = 0.791 and RMSE = 0.059, outpacing physics-informed and attention-augmented baselines. CISMN-4 on the PMLB sonar benchmark (208 samples, 60 bands) attained R ² = 0.424 and RMSE = 0.380, surpassing LSTM, memristive, and reservoir models. Across seven standard regression tasks with 5-fold cross-validation, CISMN led on diabetes (R ² = 0.483 ± 0.073) and excelled in high-dimensional, low-sample regimes. Ablations reveal a scalability–efficiency trade-off: lightweight variants train in <10 s with >95% peak accuracy, while deeper configurations yield marginal gains. CISMN sustains gradient norms (~2300) versus LSTM collapse (<3), and fixed-seed protocols ensure <1.2% MAE variation. Interpretability remains challenging (feature-attribution entropy ≈ 2.58 bits), motivating future hybrid explanation methods. CISMN recasts chaos as a computational asset for robust, generalizable modeling across scientific, financial, and engineering domains.

This paper investigates the structural performance of flat double-layer grids with various constitutive units, addressing a notable gap in the literature on diverse geometries. Six common types of flat double-layer grids are selected to provide a comprehensive comparison to understand their structural performance. Parametric models are built using Rhino and Grasshopper plugins. Single- and multi-objective optimization processes are conducted on the considered models to evaluate structural mass and maximum deflection. The number of constitutive units, the structural depth, and the cross-section diameter of the members are selected as design variables. The analysis reveals that the semi-octahedron upon square-grid configuration excels in minimizing structural mass and deflection. Furthermore, models lacking a full pyramid form exhibit higher deflections. Sensitivity analyses disclose the critical influence of the design variables, particularly highlighting the sensitivity of structural mass to the number of constitutive units and cross-section diameter. These findings offer valuable insights and practical design considerations for optimizing double-layer grid space structures.

This study proposes a machine learning (ML) model to predict the displacement response of high-rise structures under various vertical and lateral loading conditions. The study combined finite element analysis (FEA), parametric modeling, and a multi-objective genetic algorithm to create a robust and diverse dataset of loading scenarios for developing a predictive ML model. The ML model was trained using a recurrent neural network (RNN) with Long Short-Term Memory (LSTM) layers. The developed model demonstrated high accuracy in predicting time series of vertical, lateral (X), and lateral (Y) displacements. The training and testing results showed Mean Squared Errors (MSE) of 0.1796 and 0.0033, respectively, with R2 values of 0.8416 and 0.9939. The model’s predictions differed by only 0.93% from the actual vertical displacement values and by 4.55% and 7.35% for lateral displacements in the Y and X directions, respectively. The results demonstrate the model’s high accuracy and generalization ability, making it a valuable tool for structural health monitoring (SHM) in high-rise buildings. This research highlights the potential of ML to provide real-time displacement predictions under various load conditions, offering practical applications for ensuring the structural integrity and safety of high-rise buildings, particularly in high-risk seismic areas.