摘要
针对油田卧式水套加热炉热效率较低的情况,提出利用热管回收烟气余热,并根据其特性准确地控制水温。根据加热炉的工况需求进行热管的选型和安装。针对加热炉水温控制具有大滞后的特性,提出了单神经元SVM-Smith预测控制。利用支持向量机预测了被加热介质的流量,提高Smith预测补偿的准确性。采用单神经元PID控制和Smith预测补偿相结合的方式,整定控制系统的参数,取得了较好的控制效果。对构造的控制算法进行了计算机仿真,证明了算法的有效性,然后通过试验证明了热效率得到明显提高,热管余热回收的显著效果。
Aiming at the problem of lower thermal efficiency for the horizontal water jacket heating furnace in oilfield, the method of recovering the heat of exhaust gas with heat pipe, and controling water temperature accurately according to its characteristics is proposed. The selection and installation of heat pipes are carried out according to the working conditions of the furnace. Aiming at the characteristic of large lag of water temperature control of heating furnace, a single neuron SVM-Smith predictive control is proposed. The flow of heated medium is predicted by support vector machine. The accuracy of Smith prediction compensation is improved. The combination of single neuron PID control and Smith predictive compensation is used to adjust the parameters of the control system. Good control results is approached. The computer simulation of the control algorithm for the construction is conducted. The effectiveness of the algorithm is verified. The thermal efficiency is obviously improved by experiment. The heat recovery of the heat pipe is remarkable.
引文
[1] 刘国伟,谭凯中.高温低氧燃烧技术应用与火筒式油田加热炉的热工特性实验研究[J].科学技术与工程,2018(18):48-54.
[2] 梁海锋,殷诚宏,雷钧,等.长庆油田加热炉提效实践与认识[J].石油工业技术监督,2016(12):52-54.
[3] EI-Genk M S,Tournier J M.Conceptual Design of HP-STMC Space Reactor Power System for 110 kWe[J].AIP Conference Proceedings,2004(699):658-672.
[4] Shafii M B.Thermal Modeling of Unlooped and Looped Pulsating Heat Pipes[J].ASME J Heat Transfer,2002,123:1159-1172.
[5] 曹卫卫,任杰锶.基于某燃油加热系统的热管性能与应用分析[J].机械工程与自动化,2018(04):205-209.
[6] 吴存真,刘光泽.热管在热能工程中应用[M].北京:水利水电出版社,1993.
[7] 王显.有杆泵抽油井工况远程监测与故障诊断系统研究[D].武汉:武汉理工大学,2006.
[8] 刘炜.基于支持向量机的抽油机示功图工况判别[D].西安:西安理工大学,2009.
[9] 王凯,刘宏昭,穆安乐.基于最小二乘支持向量机的有杆抽油泵工况多分类研究[J].机械科学与技术,2010,29(12):1687-1691.
[10] 于德亮.基于支持向量机沉没度预测的潜油泵冲次优化研究[J].中国工程机械学报,2011,31(27):138-144.
[11] 王宏宇,糜仲春,梁晓艳,等.一种基于支持向量机回归的推荐算法[J].中国科学院研究生学院学报,2007(06):742-748.
[12] 赵一兵.基于支持向量机回归的应急物资需求预测[J].计算机仿真,2013(08):402-412.