基于BP神经网络预测模型的轧花机转速控制系统的设计与试验

Design and experiment of cotton ginning machine rotational speed control system based on BP neural network prediction model

控制棉纤维中的短纤维率是提高棉花品质的关键因素,利用籽棉回潮率进行轧花机转速的控制能减少成品棉中的原棉短纤维指数,提高成品棉的品质。针对这一问题,利用BP神经网络控制思想,建立以籽棉回潮率与轧花机转速为输入层的BP神经网络预测模型,通过对原棉纤维数据的训练处理,预测出原棉短纤维指数值相对应的电机转速控制频率与回潮率的设定值,并以此数据为基础对轧花机控制系统进行设计,通过预测值提前获知最优的转速控制频率,优化轧花机的转速。通过对试验样机进行测试试验,结果表明,轧花机试验样机系统稳定,能通过回潮率、短纤维指数预测出轧花时所需要的实际转速。原棉短纤维指数实测值的变化曲线与预测值接近,通过回潮率的预测调节,轧花机的转速随着回潮率与原棉短纤维指数进行了相应的调整,验证了BP神经网络预测模型具有较好的控制效果。

The control of short fiber index in cotton fiber is an key factor to improve the cotton quality.The seed cotton moisture regain used to control the rotational speed of the cotton ginning machine can reduce the short fiber index of the raw cotton,and improve the quality of the finished cotton.Aiming at this problem,the BP neural network control thought is used to set up the BP neural network prediction model taking the seed cotton moisture regain and cotton ginning machine′s rotational speedas its input layer.The raw cotton fiber data is trained to predict the motor rotational speed control frequency corresponding to theraw cotton short fiber index and set value of the moisture regain.On these basis,the cotton ginning control system was designed to obtain the optimal rotational speed control frequency in advance with the prediction value to optimize the rotational speed ofthe cotton ginning machine.The test results of the experimental prototype show that the experimental prototype system of the cot?ton ginning machine is stable,can predict the practical rotational speed needed by cotton ginning machine with the moisture regain and short fiber index.The variation curve of the measured value of the raw cotton short fiber index is close to that of the prediction value.The rotational speed of the cotton ginning machine is adjusted with the prediction adjustment of moisture regain andraw cotton short fiber index correspondingly.It is verified that the BP neural network prediction model has better control effect.