A new modeling method is presented in this study for optimizing the thermal and electrical energy scheduling for a multiple energy carrier microgrid (MECM) so as to minimize operations costs while meeting restrictions. A combination of microgrid (MG) network's thermal and electrical loads are supposed with the model using a day-ahead forecast (24 h). A digital twin of the MG incorporating different thermal and electrical devices is considered which can represent the real behavior of the MG, effectively. Electricity production from wind turbines has been estimated by using a day-ahead forecasting as well. Wind power production are estimated by a Monte Carlo simulation because of the uncertainty of day-ahead forecasting. Moreover, non-essential loads are shifted through the real-time demand response programs. The hybrid teacher learning with particle swarm optimization algorithm minimizes the operating costs of the MG. Using simulations, it is demonstrated that the suggested modeling method reduces operating costs and computing burden more than traditional centralized optimum scheduling methods found in other writing. Additionally, the outcomes will be examined and confirmed by scenarios test after implementing the suggested modeling structure.