Recently, Dr Liang Xia and his PhD students have filed a new control system called ‘an enhanced self-tuning RBF-PID controller for bi-object optimization’, which not only can it make up for the shortcomings of traditional PID control system with high energy consumption, but it can also adjust itself to the best energy supply scheme in an unstable environment through machine learning.

In our daily life, there are lots of control systems, such as the air conditioner to control the room temperature and the valve to adjust the water flow rate. Take a simple case for example.

In the public space like shopping mall, to maintain a comfortable environment in the summer time, the temperature of air-conditioning will be set at around 26°C. When the temperature sensor of air conditioner detects that the room temperature is above 26℃, it will spontaneously accelerate the working frequency of the refrigeration system to reduce the room temperature. Conversely, when the room temperature is lower than 26℃, it will spontaneously slow down the operating frequency of the system to maintain the relative stability of the temperature in the building. This is the traditional PID control system.

However, unlike the traditional PID control system, this new system can find the best way for the energy consumption. According to Dr Liang Xia’s words, it is very important to reasonably reduce the non-steady-state energy consumption in the control process. In real life, due to the existence of many random and inevitable disturbances, the controlled quantity is often difficult to maintain at the target value continuously and steadily. Taking a common office and residential building as an example, due to changes in the number of people indoors, human behavior, and outdoor weather conditions, it is difficult for the indoor air conditioning system to stabilize the air temperature within an ideal range.

This poses a challenge to the control strategy of the air conditioning system: on the one hand, the control strategy of the air conditioning system needs to gradually control the air temperature to the comfortable temperature of the human body as much as possible; on the other hand, the unsteady energy consumption caused by the control system is also need as little as possible. Therefore, using a dual-objective control system that can simultaneously meet the control requirements and reduce the non-steady-state energy consumption of the control process can reduce the operating energy consumption of the equipment and improve the efficiency of energy use.