Frontiers in Signal Processing

Download PDF (439.3 KB) PP. 22 - 29 Pub. Date: January 5, 2020

DOI: 10.22606/fsp.2020.41004

• Wanqiang Wu
  College of Electrical & Information Engineering, Southwest Minzu University, Chengdu 610041, China
• Liangfu Peng^*
  College of Electrical & Information Engineering, Southwest Minzu University, Chengdu 610041, China
• Gui Gan
  College of Electrical & Information Engineering, Southwest Minzu University, Chengdu 610041, China

In order to improve the problem that the traditional PID control oven controlled crystal oscillator cannot be adjusted in real time in the process of clock taming, a BP neural network tuning PID control algorithm is proposed. BP neural network tuning PID control algorithm can learn the rule of PID control online, and can adjust the parameters of PID control in real time. The results of simulation in MATLAB show that there is no obvious overshoot and oscillation in step response controlled by BP neural network tuning PID, and the system is stable faster than the traditional PID control. Therefore, BP neural network tuning PID control has better control effect than traditional PID control in oven controlled crystal oscillator control, and has a strong self-adaptability in parameter adjustment.

Clock taming, oven controlled crystal oscillator (OCXO), incremental PID control, BP neural network (BPNN), adaptive adjustment

[1] Zhu Xuetian, Xia Xu, Qi Fei. 5G key technologies and business forecast[J]. Application of Electronic Technique, 2018, 44(9): 1-4, 8.

[2] Liu Song, Lu Jingfei. Evolution of synchronous network architecture for 5G bearer [J]. Telecom Network Technology, 2017 (9): 22-26.

[3] Xue Yicong, Gong Hang, LIU Zengjun, et al. Analysis of Disciplined Crystal Oscillator Method Based on GNSS. GNSS World of China, 2017, 42(4): 38-42.

[4] Ma Yanqing. Research on adaptive taming and holding technology of oven controlled crystal oscillator OCXO [D]. Xi'an University of Electronic Science and Technology, 2011.

[5] Zhang Yongzhen, Su Hansong, Liu Gaohua, et al. PID controller parameter adjustment based on BP neural network [J]. Journal of Nankai University (Natural Science Edition), 2018, 51 (3): 28-32.

[6] Yu Geng, Fang Hongtao. Improved UKF integrated navigation algorithm based on BP neural network[J]. Application of Electronic Technique, 2019, 45(4): 29-33.

[7] Sun Jiangtao. GPS-based 1PPS secondary frequency standard taming technology [D]. Xidian University, 2010.

[8] Cai Xinpu. Research on time-frequency processing technology of cesium atomic frequency standard [D]. Xidian University, 2014.

[9] Tian Lei. Development of oven controlled crystal oscillator synchronous control device based on adaptive PID[D]. Jilin University, 2009.

[10] Cominos P, Munro N. PID controllers: recent tuning methods and design to specification[J]. IEE Proceedings-Control Theory and Applications, 2002, 149(1): 46-53.

[11] Zhuang M, Atherton D P. Automatic tuning of optimum PID controllers[J]. 1993, 140(3): 216-224.

[12] Shi Chunchao. Improvement of BP neural network algorithm and its application in PID control[D]. Tianjin University, 2006.

[13] Zhu Haifeng, Li Wei, Zhang Lin. PID Control Based on BP Neural Network Tuning[J]. Journal of Dynamics and Control, 2005, 3(4): 93-96.

[14] Jiangming K, Jinhao L. Self-Tuning PID Controller Based on Improved BP Neural Network[C]. Second International Conference on Intelligent Computation Technology & Automation. IEEE Computer Society, 2009: 95-98.

[15] Liu Jinwei. Advanced PID Control and Its MATLAB Simulation [M]. Publishing House of Electronics Industry, 2003.