Author(s)

Zu Wang¹, Liang Xia^1*, John Kaiser Calautit², Xinru Wang^1,3, Danwei Jiang¹, Song Pan⁴, Jinshun Wu⁵

¹Research Centre for Fluids and Thermal Engineering, University of Nottingham, Ningbo, China.
²Department of Architecture and Built Environment, The University of Nottingham, Nottingham, UK.
³College of Emergency Technology and Management, North China Institute of Science & Technology, Langfang, China.
⁴Beijing Key Laboratory of Green Built Environment and Energy Efficient Technology, Beijing University of Technology, Beijing, China.
⁵Hebei Province Renewable Energy Heating and Cooling Cogeneration International Union Research Center, North China Institute of Science & Technology, Langfang, China.

Proportional, integral and derivative (PID) control strategy has been widely applied in heating systems in decades. To improve the accuracy and the robustness of PID control, self-tuning radial-basis-function neural network PID (RBF-PID) is developed and used. Even though being popular, during the control process both of PID and RBF-PID control strategy are inadequate in achieving simultaneous high energy-efficiency and good control accuracy. To address this problem, in this paper we develop and report an enhanced self-tuning radial-basis-function neural network PID (e-RBF-PID) controller. To identify the superiority of e-RBF-PID, following works are conducted and reported in this paper. Firstly, four controllers, i.e., on-off, PID, RBF-PID and e-RBF-PID are designed. Secondly, in order to test the performance of the e-RBF-PID controller, an experimental water heating system is constructed for being controlled. Finally, the energy consumption for the four controllers under the three control scenarios is investigated through experiments. The experimental results indicate that in the three scenarios, the developed e-RBF-PID controller outperforms on-off controller as having higher accuracy. Compared to the PID controller, the e-RBF-PID controller has higher speed in control, and the experimental results show that settling time savings is between 12.6% - 49.0%. Most importantly, less control energy consumption is obtained if using the e-RBF-PID controller. It is found that up to 28.5% energy consumption can be saved. Therefore, it is concluded that the proposed e-RBF-PID is capable of enhancing energy efficiency during control process.

Energy-Efficient Control, RBF Neural Network, Enhanced Self-Tuning PID, Experimental Validation

Wang, Z. , Xia, L. , Calautit, J. , Wang, X. , Jiang, D. , Pan, S. and Wu, J. (2021) Development of an Enhanced Self-Tuning RBF-PID Controller for Achieving Higher Energy-Efficient Process Control. Journal of Building Construction and Planning Research, 9, 272-291. doi: 10.4236/jbcpr.2021.94017.