基于BP神经网络的抽油机节能调节算法研究

为了降低抽油机能耗,需要对抽油机在以往的工作中柄平衡块进行调节。当平衡块位于最佳平衡位置时可以达到节能的目的。以往,平衡块的调节都是前线工人凭经验进行调节,即无法大规模普及,也耗费大量时间和人员成本。为了解决这一问题,提出了一种基于BP神经网络的适用于抽油机平衡块快速准确调节的算法,并通过实践检验,验证了该算法的准确性和有效性。根据该算法调整后,耗电量单井减少2~13 kWh,平均减少5.8 kWh,平均省电比为6.47%。

Research on Energy Saving Adjustment Algorithm of Pumping Unit Based on BP Neural Network

In order to reduce the energy consumption of the pumping unit,it is necessary to adjust the handle balance block in the previous work of the pumping unit.The purpose of energy saving can be achieved when the balance block is located in the best balance position.In the past,the adjustment of balancers was based on the experience of front-line workers,which could not be popularized on a large scale and cost a lot of time and personnel.In order to solve this problem,an algorithm based on BP neural network is put forward,which is suitable for the rapid and accurate adjustment of the balance block of pumping unit.After adjusting the algorithm,the power consumption per well is reduced by 2~13 kWh,with an average reduction of 5.8 kWh,and the average power saving ratio is 6.47%.