Neural Network Minimun Parameter Sliding Control of Blood Pump Based on Heart Rate

Authors

Huibo Liu, School of Information Engineering, Inner Mongolia University of Science and Technology, Baotou 014010, China;
Liu Mu, School of Information Engineering, Inner Mongolia University of Science and Technology, Baotou 014010, China;

Abstract

Abstract: An intelligent algorithm is designed to control the blood pump to assist natural cardiac blood pumping and to meet the needs of blood perfusion. The algorithm combines sliding mode control which does not depend on object model and RBF neural network which has optimal approximation properties to design controller. Minimum parameter learning method is used to design the adaptive law, adjust the weights of neural network. In order to ensure the real time, the speed signal is given by the function of blood flow, blood pump speed and heart rate. Through comparing with PID algorithm, the research shows that: under the RBF sliding mode control, the biggest blood pump speed regulating time is 0.23 s, steady-state error is 5 RPM, flow maximum relative error is 0.89%, both dynamic and static performance of this system is good, so the minimum parameters of RBF neural network sliding mode control algorithm fully meets the requirements of the blood pump control system.

Recommended Citation

Liu, Huibo and Mu, Liu (2019) "Neural Network Minimun Parameter Sliding Control of Blood Pump Based on Heart Rate," Journal of System Simulation: Vol. 30: Iss. 5, Article 22.
DOI: 10.16182/j.issn1004731x.joss.201805022
Available at: https://dc-china-simulation.researchcommons.org/journal/vol30/iss5/22

First Page

1796

Revised Date

2016-07-27

DOI Link

https://doi.org/10.16182/j.issn1004731x.joss.201805022

Last Page

1802

CLC

TP391.9

Recommended Citation

Liu Huibo, Liu Mu. Neural Network Minimun Parameter Sliding Control of Blood Pump Based on Heart Rate[J]. Journal of System Simulation, 2018, 30(5): 1796-1802.

DOI

10.16182/j.issn1004731x.joss.201805022