Abstract: An active heating system is often necessary for the normal growth of crops in a greenhouse, particularly with the severe winter weather in northern China. Among them, an air-source heat pump system can be favored for the low energy consumption, cost saving, and less pollution. Much effort was mostly on the working performance of heat pumps. Specifically, an advanced control systems of greenhouse environment with a high level of automation cannot better fit the current situation of family-oriented greenhouse cultivation, particularly at a high cost. It is highly urgent to combine the temperature control systems for the regulation of the greenhouse with the air source heat pump in a reasonable and simple way. In this study, an intelligent monitoring system was established to combine the management needs of modern greenhouse using the typical IoT structure. The system also consisted of on-site monitoring subsystem, IoT cloud box, and remote monitoring subsystem. The integrated heat transfer coefficient was then calculated under the rolled/unrolled greenhouse, according to the basic principles of heat transfer. The measured data was utilized to determine the performance parameters of the air source heat pump heating system and fogging spray. The heat balance equation was selected to establish the relationship equation between the control time of regulating equipment and environmental parameters of greenhouse. As such, the temperature control strategy was designed to collect the environmental data in real time by sensors. The obtained data was set as the input in the process, while the output was required the working time of equipment to control the equipment start and stop, according to the time to regulate the greenhouse temperature. At the same time, the crop light compensation point was used to control the action of the rolled quilt, in order to fully meet the light environment that required for the crop growth. An outdoor light irradiance was also used to correct the indoor light irradiance, because the sensor value cannot accurately reflect the real value, due to the rolled quilt before the unrolling action. Once the indoor irradiance correction value was greater than the crop light compensation point, the volume was opened, otherwise, the volume was closed. A series of experiments were also carried out with the greenhouse cucumber cultivation. There were the stable changes and small fluctuations of temperature around the target value in the improved model under the heat balance equation. The single-day power consumption of the control system saved 9.06 kW·h under similar external weather conditions, compared with the set value-based control system, accounting for 10.95% of the power consumption on the test day. Consequently, the control system under the heat balance equation can be easily implemented to regulate via the different temperature target values, according to the crop demand. The finding can provide a strong reference for the application and promotion of temperature control strategy for the greenhouse in winter.