Frequency Response Functions Dataset of a 7-Story RC Shear-Wall Core Building (Bab Bab Ezzouar, Algiers) for Damage Identification

This dataset contains simulated ambient-vibration frequency response data from a finite element model of a seven-story reinforced concrete (RC) commercial building located in Bab Bab Ezzouar, Algiers, Algeria. The structure has a ground floor plus seven upper stories, totaling a height of 34.74 m with a constant inter-story height of 3.96 m. The lateral load-resisting system consists of a central RC core shear wall enclosing the stairwells and elevator shafts. The FEM model, bulit in SAP2000, represents the building's structural members and identifies 5 predefined damage zones located at the base of the structure, corresponding to critical regions where the highest stress levels occur and where stiffness degradation would most significantly affect the global dynamic response. Structural damage is simulated as a localized reduction in the material elastic modulus (E) applied simultaneously across all 5 damage zones, with an independent randomized severity level per zone ranging from 0 (no reduction) to 0.9 (90% loss of stiffness). The dataset comprises 5,000 randomly generated damage scenarios plus 1 undamaged (healthy) baseline scenario, for a total of 5,001 cases. For each scenario, the structural response was obtained through a steady-state (frequency-domain) analysis simulating ambient vibration excitation, sampled at 81 discrete frequency points over a 0–20 Hz bandwidth. Responses were recorded at 12 sensor locations across the building, 3 sensors per instrumented story, placed on alternating stories (i.e., every other story is instrumented, while the remaining stories carry no sensors), each capturing motion along three orthogonal directions (U1, U2, U3), giving 81 frequencies × 3 directions = 243 response values per sensor per scenario. File contents: dataset.npy — array of shape (5001, 12, 243), float32. Axis 0: scenario index (index 0 = healthy baseline; indices 1–5000 correspond to the damage scenarios in full_data.csv, row-aligned). Axis 1: sensor ID (12 sensors, 3 per instrumented story). Axis 2: flattened FRF response (81 frequency points × 3 directions, U1/U2/U3). full_data.csv — shape (5000, 5). Each row is a damage scenario; each column is one of the 5 predefined damage zones at the building base, with values representing the severity of the elastic modulus reduction applied to that zone (0 = undamaged, 0.9 = 90% reduction in E) for that scenario. Note on scenario alignment: row i of full_data.csv corresponds to index i+1 of dataset.npy (offset by the healthy baseline at index 0).