CNN-LSTM based deep learning application on Jetson Nano: Estimating electrical energy consumption for future smart homes
Yazarlar (3)
Dr. Öğr. Üyesi Abdulkadir GÖZÜOĞLU
Tokat Gaziosmanpaşa Üniversitesi, Türkiye
Ondokuz Mayıs Üniversitesi, Türkiye
Ondokuz Mayıs Üniversitesi, Türkiye
| Makale Türü | Diğer (Teknik, not, yorum, vaka takdimi, editöre mektup, özet, kitap krıtiği, araştırma notu, bilirkişi raporu ve benzeri) |
| Makale Alt Türü | SCI, SSCI, AHCI, SCI-Exp dergilerinde yayınlanan teknik not, editöre mektup, tartışma, vaka takdimi ve özet türünden makale |
| Dergi Adı | INTERNET OF THINGS |
| Dergi ISSN | 2543-1536 Wos Dergi Scopus Dergi |
| Dergi Tarandığı Indeksler | SCI |
| Dergi Grubu | Q1 |
| Makale Dili | İngilizce |
| Basım Tarihi | 03-2024 |
| Cilt No | 26 |
| Sayı | 1 |
| Sayfalar | 17 / 0 |
| DOI Numarası | 10.1016/j.iot.2024.101148 |
| Makale Linki | https://doi.org/10.1016/j.iot.2024.101148 |
| Özet |
| Smart home applications have witnessed significant advancements, expanding beyond lighting control or remote monitoring to more sophisticated functionalities. Our study delves into pioneering an advanced energy management system tailored for forthcoming smart homes and grids. This system harnesses deep learning methodologies to predict consumer energy consumption. Leveraging a Wireless Fidelity (Wi-Fi) connection, we established an Internet of Things (IoT) network supported by Message Queuing Telemetry Transport (MQTT) for efficient data transfer. Our approach integrated the Jetson Nano Developer Kit for deep learning tasks, utilized Raspberry Pi as a home management server (HMS), and employed Espressif Systems' microcontrollers (ESP-01, NodeMCU, ESP32) to impart intelligence to household devices. Actual house measurements were collected and rigorously analyzed, demonstrating promising outcomes in deep learning, control, and monitoring applications. This management system's potential extends to empowering future smart homes and is a crucial component for demand-side energy management in forthcoming intelligent grids. |
| Anahtar Kelimeler |
| Cnn-lstm method | Deep learning | Demand-side management | Energy consumption estimation | Esp-01 | Esp32 | Future smart grids | Future smart homes | Home automation | home management server | Nodemcu | Wi-Fi |
BM Sürdürülebilir Kalkınma Amaçları
| Atıf Sayıları | |
| WoS | 28 |
| SCOPUS | 23 |
| Google Scholar | 56 |