Exploration of Monte Carlo Method for Optimization of Energy Consumption in Industrial Enterprises in Energy Efficiency Diagnosis

Energy consumption, as one of the most concerned parts of industrial manufacturing costs, has an important impact on the overall operation and development of industrial enterprises. The article takes an air-conditioning manufacturing enterprise as an example, constructs its energy consumption prediction model (ARIMA model), optimizes its parameter estimation method through Bayesian estimation and Markov Monte Carlo method, and finally constructs the ARIMAMCMC model for energy consumption prediction of industrial enterprises. The accuracy of its prediction effect is verified by comparing the load prediction results of this model with other prediction models. The effect of energy consumption optimization under fixed cooling demand and the effect of energy consumption optimization on continuous periods before and after model optimization are analyzed respectively. The model has the highest accuracy in predicting the cooling load of central air-conditioning, and the difference between its predicted and actual values is the smallest. The energy consumption is lower when there is high cooling demand above 70%. At lower levels of 60% and 50% of cooling demand, energy wastage is higher. About 6.09% of the optimized model’s energy consumption is reduced. During the duration period (9:00-20:00), the total energy consumption before and after optimization is 97.218kW and 90.706kW, respectively, and the system energy saving is 6.70%. The period with the worst energy-saving effect is 12:00-13:00 on the 21st, saving energy consumption by 2.27%. The best energy saving is 18:00-19:00 on the 21st, saving 21.05% of energy consumption. During the two days before optimization, the average water temperature was 26.31 and 23.66°C, and the average water temperature after optimization was 24.88 and 23.06°C, respectively.