摘要
在金川公司选矿厂2006年大型浮选设备工业实验中,为保证矿浆液面自动控制系统液位的稳定性,提出了一种符合现场复杂情况,具有线性流量特性调节阀阀芯曲线的智能拟合方法。在基于过程补余量算法中,结合BP神经网络模型,用不同形状的调节阀阀芯曲线改变调节阀的原有流量特性,从对控制的稳定性、控制响应的时效性等因素考虑后,确定了这种阀芯的最佳拟合曲线,经现场实际使用验证,效果良好。
The study was conducted on large flotation equipment in ore plants of Jinchuan Group Ltd during 2006, the intelligent fitting method of valve core curve is proposed based on the linear flow characteristic of regulating valve. Optimal curve fitting is determined based on replenishment quantity of flotation process and model of BP neural network with using the different shape of valve core curve. Industrial experiments have shown that the method has a wide using prospect with high reliability, high stability of the flotation machine liquid level and high effectiveness of level control system.
出处
《中山大学学报（自然科学版）》
CAS
CSCD
北大核心
2009年第1期22-25,共4页
Acta Scientiarum Naturalium Universitatis Sunyatseni
基金
国家高技术产业发展计划资助项目(发改办高技[2005]1899号,甘发改高技[2005]291号)
关键词
BP神经网络模型
液位
阀芯曲线
流量特性
智能拟合
model of BP neural network
liquid level
valve core curve
flow characteristic
intelligent fitting
