初探神经网络在电动洗胃中的应用

Preliminary Study on Application of Neural Network in Electric Gastric Lavage

提出一种智能洗胃的设计方案。利用微处理器采集和计算模拟洗胃试验中的洗胃压力、压力变化率、液位、液位变化率、浑浊度和PH值等信号,通过神经网络算法对信号进行处理,判断洗胃工况。采用神经网络算法能区分管路堵塞、不完全堵塞、故障、洗胃正常和清洁等运行状态。神经网络算法可用于柔性密闭容器的清洗工作。

A method is proposed for intelligent gastric lavage design in this paper. The microprocessor collects and computes gastric lavage pressure and its rate of change, liquid level and its rate of change, turbidity, and PH signals in the analogue stomach lavage test, which are processed by neural network algorithm, and then infer gastric lavage work status. Pipe plug, incomplete pipe plug, fault, normal operating, clean etc can be distinguished by neural network. Neural network algorithm can be used for flexible airtight container cleaning.