Abstract: To reduce the energy consumption of air conditioning system in cigarette production process, an intelligent control strategy based on temperature and humidity deviation regulation mechanism and machine learning algorithm was proposed for upgrading the K6 air conditioning system in the primary processing department of Baoding Cigarette Factory. On the basis of the modification of surface cooling valve and the addition of a high-performance energy-saving server, along with the improvement of PID control technology and the integration of the regional target control optimization module, the intelligent regulation module of wet equilibrium state and the critical operation state control module, the characteristic data, set values of temperature and humidity of the air conditioning system was adjusted in real time to achieve the optimization of the control strategy and establish an intelligent energy-saving monitoring system for air conditioning. Comparative experiments were conducted on the air conditioning system before and after the optimization. The results showed that both the temperature and humidity in the primary processing department before and after the optimization met the technical requirements. The optimized system featured remarkable energy-saving effect, showing decreases of 76.2%, 39.5% and 11.8% respectively in the consumption of steam, cold water and electricity. This study provides technical references for the energy-saving improvements in similar industrial facilities.