基于Type-2 FNN数据融合的双进双出磨煤机料位检测

BBD ball mill material measure based on type-2 FNN data fusion

针对双进双出磨煤机料位准确检测的难题,提出一种基于二型模糊神经网络(Type-2 FNN)数据融合的双进双出磨煤机料位检测方法.首先将多传感器采集的变量参数数据按照二型模糊规则进行模糊化处理;然后构造神经网络进行数据融合,所得融合结果即为检测的料位值.该方法具有较好的自组织、自学习、并行分别处理能力,保证了检测结果具有较高的准确性.仿真实验表明,该方法可有效弥补单一测量方法的不足,实现料位更为准确的测量.

For the material accurate measure problems of double input and double output（BBD） ball mill, a method of BBD ball mill＇s material measure based on type-2 fuzzy neural network（Type-2 FNN） data fusion is proposed. Firstly, the variables parameters are dealed with fuzzily according to the type-2 fuzzy rules, which are acquisitied by multi-sensor. Then the neural network is structured for data fusion, and the result is the material data. The method not only has good capabilities of self-organization, self-learning and respectively processing, but also ensures that detecting result has higher accuracy. Simulation results show that, the application of this fusion system can effectively remedy the lacks of single measurement method to achieve more accurate measurement.