矿井通风机实时监测与故障诊断的智能系统

针对矿井通风机系统参数难以测定和故障率较高的特点，提出将多层前向传递神经网络应用于通风机系统控制的方法，利用BP算法训练神经网络，从而实现矿井通风机数据的实时监测及系统的故障诊断并报警。该控制算法对被控对象的数学模型依赖程度较低，为非线性系统的控制提供了一种行之有效的研究方法。

Intelligent system of fault diagnosis and data real-time observation of mine ventilator

This paper is directed at the characteristics of parameters more difficult to determine and faults more likely to occur in the mine ventilator system. The back-propagation neural network is applied to ventilator system. Data real-time observation and fault diagnosis for ventilator system are realized by training the neural network using BP algorithm. The paper presents a novel learning linear control mechanism for a class of nonlinear system, which shows less dependence on the model of non-linear control system.