Minimal sensor count approach to fuzzy logic rotary blood pump flow control.
Minimal sensor count approach to fuzzy logic rotary blood pump flow control.
复制标题
模糊逻辑旋转血泵流量控制的最小传感器计数方法。
DOI：
10.1097/01.mat.0000250786.56697.f2
复制
发表时间：
2007
期刊：
影响因子：
--
通讯作者：
Reeves,Andrew
中科院分区：
文献类型：
--
作者：
Casas,Fernando;Ahmed,Nisar;Reeves,Andrew
研制了一种无流量传感器的旋转血泵模糊逻辑流量控制器，并进行了体外实验。在LabView中实现的控制器被设置为在存在外部压力扰动的情况下维持流量设定点。使用稳态非线性近似，将流量估计为测量的泵的Δ P和速度的函数。模糊控制器采用泵的流量估计和Δ P作为反馈变量。去模糊控制输出操纵泵速度。隶属函数包括流量误差、Δ P和泵速。在模拟回路（水/甘油3.5cPs，37 ℃）中使用估计流量的实验运行与使用跨音速流量计的那些进行比较，用于九种流量和Δ P条件（4至6L/min，150至350 mm Hg）。伺服夹管阀产生的压力扰动范围为±23至±47 mm Hg。结果表明，即使在压力波动较大的情况下，模糊控制器也能将流量设定值调节到基线的±10%以内。超声波流量测量和估计流量计算场景之间的控制器性能没有差异。这些测试表明，模糊控制器是能够拒绝干扰和调节流量到可接受的限制，同时使用流量估计。
A rotary blood pump fuzzy logic flow controller without flow sensors was developed and tested in vitro. The controller, implemented in LabView, was set to maintain a flow set point in the presence of external pressure disturbances. Flow was estimated as a function of measured pump’s delta P and speed, using a steady-state, nonlinear approximation. The fuzzy controller used the pump’s flow estimate and delta P as feedback variables. The defuzzified control output manipulated the pump speed. Membership functions included flow error, delta P, and pump speed. Experimental runs in a mock loop (water/glycerin 3.5 cPs, 37 C), using the estimated flow, were compared with those using a Transonic flow meter for nine conditions of flow and delta P (4 to 6 L/min, 150 to 350 mm Hg). Pressure disturbances generated by a servo pinch valve ranged from±23 to±47 mm Hg. Results indicated that the fuzzy controller ably regulated the flow set point to within±10% of the baseline even under large swings in pressure. There was no difference in controller performance between the ultrasonic flow measurement and the estimated flow calculation scenarios. These tests demonstrated that the fuzzy controller is capable of rejecting disturbances and regulating flow to acceptable limits while using a flow estimate.