欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出了基于特征选取与多RBF神经网络的冷凝器污垢软测量方法。该方法采用基于BP神经网络的灵敏度计算，从受污垢影响的冷凝器各个性能指标中，提取最能反映冷凝器污垢状态的特征变量。在此基础上，针对冷凝器变工况、冷凝器空气漏人量等因素对污垢特征变量的影响，研究基于多RBF神经网络的智能建模方法，有效实现冷凝器污垢与其他参数变化对特征变量影响的分离。根据此方法，进行了现场试验，试验结果表明：该方法能较准确地在线监测冷凝器污垢，并在冷凝器出现堵管、空气漏人量较大、工况参数大范围变化时，取得比热阻法、传热系数法、模糊软测量法更可靠的测量结果。
英文摘要：
A novel approach based on feature selection and multiple RBF neural network for online soft-sensing of fouling in condenser is proposed in this paper. In the approach, sensitivity calculation based on BP neural network is introduced to determine the most suitable feature variable that reflects the fouling state, multiple RBF neural network is employed to separate the influences of fouling and other factors, such as off-design condition on feature variable. Based on this method, experiment on an actual condenser was carried out. The results show that the approach measures the fouling correctly and is more effective than thermal resistance method, heat transfer coefficient method and fuzzy soft-sensing method under the condition of blocked tubes, excessive amount of air in condenser and great change of condition parameters.
同期刊论文项目
同项目期刊论文
Environmental/economic power dispatch problem using multi-objective differential evolution algorithm
Intelligent Foreign Particle Inspection Machine for Injection Liquid Examination Based on Modified P
Existence and exponential stability of almost periodic solution for shunting inhibitory cellular neu
位置：