基于BP神经网络和模糊PID的风机转速自适应控制

Adaptive control of fan speed based on BP neural network and fuzzy PID

为提高风机转速控制的自适应能力和稳态性能,提出一种基于BP神经网络和模糊PID的控制方法,将风机转速控制问题转化为风机风量调节问题。首先,构建风机数学模型,获取风机转速与风量的数学关系;然后,采用BP神经网络对需风量进行预测,将预测的需风量与当前风量的偏差及其变化率输入模糊PID控制器中,并进行模糊化,同时根据模糊规则控制表进行推理后,采用重心法解逆模糊;最后,通过调节风机风量来实现风机转速的控制。仿真结果表明,所提方法实现了风机转速的快速、平稳调节,相较于PID控制等方法,可快速调节煤矿机械中风机的转速,超调量下降了约20%,表现出良好的自适应能力和动态性能,满足风机转速控制需求。

In order to improve the adaptive ability and steady-state performance of the fan speed control,a control method based on BP neural network and fuzzy PID was proposed,which transformed the problem of fan speed control into the problem of fan air volume regulation.Firstly,the mathematical model of the fan was constructed to obtain the mathematical relationship between the fan speed and the air volume.Then,the BP neural network was used to predict the required air volume,and the deviation between the predicted required air volume and the current air volume and its change rate were input into the fuzzy PID controller for fuzzification.At the same time,the fuzzy rule control table was used for inference,and the centroid method was used to solve the inverse fuzzy problem.Finally,the fan speed was controlled by adjusting the air volume of the fan.The simulation results showed that the proposed method realized the fast and stable adjustment of the fan speed.Compared with PID control and other methods,it could quickly adjust the fan speed in coal mine machinery,with a reduction of about 20%,which showed good adaptive ability and dynamic performance,and met the needs of fan speed control.