قاطع دائرة ذكي بتحكم عبر GSM باستخدام إنترنت الأشياء وتحسين الطاقة بالذكاء الاصطناعي

Traditional circuit breakers provide protection against overcurrent and short-circuit conditions but have no inherent provision for remote supervision, demand awareness or intelligent load management. This paper introduces the design, implementation and experimental testing of a smart circuit breaker that combines independent electrical protection, hybrid Global System for Mobile communication (GSM)–Internet of Things (IoT) connectivity, and an energy optimization layer based on artificial intelligence (AI). An edge controller measures voltage, current, active power and energy in real time and publishes telemetry through a Message Queuing Telemetry Transport (MQTT) link or, when Internet connectivity is lost, through the GSM short-message service (SMS). A hybrid Long Short-Term Memory–Extreme Gradient Boosting (LSTM–XGBoost) model forecasts short-term demand, and a weighted objective function schedules deferrable loads to minimize electricity cost and peak demand while satisfying circuit-capacity and user-priority constraints. Software and communication functions never substitute for hardware protection; a certified miniature circuit breaker (MCB) remains the primary fault-interruption device. the prototype measured the electrical quantities with a maximum loaded-condition deviation of 0.76% from the calibrated reference values; the forecaster achieved an RMSE of 0.083 kW and a coefficient of determination (R²) of 0.9899; and the hybrid channel maintained a 97.1% command-success rate during controlled Internet-outage trials, whereas the IoT-only channel recorded 0%. Compared with conventional operation, the AI-driven strategy reduced peak demand by 39.6% and electricity cost by 8.0%, whereas total energy consumption decreased by only 3.0%, indicating that its main advantage is peak-demand reduction rather than energy reduction. These findings indicate the feasibility of a deployable device that combines protection, resilient communication and predictive energy management.