Elevated-temperature performance of concrete with expanded perlite fine aggregate: experimental and ANN analysis
Frontiers in Built Environment, cilt.12, 2026 (ESCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 12
- Basım Tarihi: 2026
- Doi Numarası: 10.3389/fbuil.2026.1795730
- Dergi Adı: Frontiers in Built Environment
- Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus, INSPEC, Directory of Open Access Journals
- Anahtar Kelimeler: artificial neural network, concrete, dynamic elasticity modulus, expanded perlite aggregate, high temperature, thermal conductivity
- Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
- Atatürk Üniversitesi Adresli: Evet
Özet
This study investigated the thermal and mechanical properties of concrete incorporated with expanded perlite aggregate (EPA) exposed to elevated temperatures. The concrete specimens were prepared by replacing normal fine aggregate (0–2 mm) with EPA at varying proportions of 0, 25, 50, 75, and 100% and tests were conducted to analyse the thermal conductivity, compressive strength, flexural strength, dynamic elasticity modulus (DEM), dry unit weight and microstructural properties after exposure to temperatures of 23, 100, 200, 300, 400 500, 600°C and 700 °C for 2 h, by air cooling method. All specimens were initially cured in 23°C ± 1 °C lime-saturated water for 28 days. Results showed a consistent reduction in thermal conductivity, compressive strength, UPV, flexural strength, DEM, and dry unit weight as a function of replacement ratio. Specifically, compressive strength reductions of 11%, 19%, 25%, and 36% were observed for 25%, 50%, 75%, and 100% EPA replacement, respectively. All the concrete specimens exposed to a temperature of 500°C and 700 °C exhibited a significant reduction in thermal conductivity, compressive strength, UPV, flexure strength and DEM. Five property-specific ANN surrogates, for thermal conductivity, compressive strength, flexural strength, UPV and DEM, were trained on the 5 × 8 factorial dataset. The models achieved coefficients of determination in the range R 2 = 0.83–0.98 on the combined data, with test MAPE values between 3.4% and 7.9%. Sensitivity analysis showed that target temperature (sensitivity ratio 4.0–6.7) dominated over EPA replacement ratio (1.5–3.3) as the primary driver of residual performance. The novelty of the study lies in coupling a dense elevated-temperature program for fine-fraction (0–2 mm) EPA replacement with a five-output ANN system built from a single experimental dataset; extrapolation of the models beyond the tested scope (w/cm = 0.35, 0–2 mm EPA, 23°C–700 °C) is not claimed.