摘要
本文应用神经网络建立流量、温度、压力、空压机群数量、湿度等参数与空压机系统中空压机群运行状态之间的控制模型。并通过实验得到多组样本数据来训练神经网络模型,然后采用不同的算法来优化模型,最终确定L-M算法收敛和运算速度最快。最后利用建立好的神经网络模型优化空压站控制系统,并使用LabVIEW软件编写控制程序。使空压机产生的气体量与用气端的用气量相匹配,从而有效地解决了供气端与用气端供需不平衡问题,降低了空压站系统的总功耗。
This paper uses neural network to establish a control model between flow, temperature, pressure, number of air compressors, humidity and other parameters and the operating state of air compressors in air compressor systems. The neural network model is trained by experimentally obtaining multiple sets of sample data, and then different algorithms are used to optimize the model. Finally, the L-M is the fastest and the fastest. Finally, using the established neural network model to optimize the air compressor station control system and using LabVIEW software to write the control program, the gas volume generated by the air compressor group is matched with the gas consumption at the gas end, thus effectively solving the gas supply end and the use. The imbalance between supply and demand of the gas terminal reduces the total power consumption of the air compressor system.
作者
王二明
赵军
尚昆
吴丹
WANG Er-ming;ZHAO Jun;SHANG Kun;WU Dan(School of Energy and Power Engineering,University of Shanghai for Science and Technology,Shanghai 200093)
出处
《新型工业化》
2019年第9期42-46,共5页
The Journal of New Industrialization
基金
上海理工大学项目：“基于人工智能控制技术的工业空压站高效运行策略研究”（项目编号：2018KJFZ161）
