Study On Implantable Left Ventricular Assist Device Control Technology

As a means of effective treatment of heart disease, artificial heart research has been paid great attention by the medical profession at home and aboard. Left Ventricular Assist Device (LVAD) as a branch of artifical heart research has been regarded as the mainstream of the future artifical heart due to its advantages. With LVAD as the research subject, this paper focuses on the drive technology and control strategy of blood pump. The results obtained include the following four parts:In the first part, the motor drive coil have been designed in terms of the characteristics of the blood pump. By comparison with traditional blood pump of LVAD, a new blood pump has been analysed with respect to specified performance and structural characteristics; The relevant parameters of the drive coil including shape, number of turns, size, installation location, and part of the electromagnetic parameters have been determined which not only meet the demand of drive capability but aiso overcome the temperature rise problem.In the second part, the hardware configuration and software structure of the blood pump motor control system have been built up. The hardware configuration includes the controller election, power management, alarm indication and emergency response, especially involving design of driver circuit for blood pump motor; On the other hand, after a detailed requirement analysis, the LCD display in the graphical user interface as well as upper machine software have been developed respectively.In the third part, a control algorithm has been provided base on operating point of both blood pump output pressure and flow rate in view of conventional control system for blood pump only dealing with blood pump motor speed. According to the physiological characteristics of different subjects, blood pressure and blood supply have been set which can be regarded as a control target to achieve automatic control for blood pump. By using knowledge of hemodynamics and physiology, the computation problem of different operating point for different subjects has been solved; Then blood pump output pressure and flow rate which is unavailable online are estimated in terms of neural network algorithms; Operating point control algorithm, recovering control algorithm and releasing suction control algorithm are represented for three different working conditions; Finally, double close-loop control system has been design to control the blood pump speed. In the fourth part, a series of blood pump experiments and control system simulations have been carried out, including the drive coil back EMF waveform analysis, the phase current and terminal voltage analysis, the drive coil temperature rise analysis and three control algorithms simulation. It has been shown that the blood pump performance and the control algorithms have meet the requirements.