基于标准误差算法的油井工况识别技术研究

抽油机示功图包含了大量的采油设备状态、采油工艺和采油工况的多维信息特征。如何精确地、快速地、方便地识别这些重要信息对指导油井设备的安全可靠运行、采油工艺的改进、实时监测采油工况具有非常重大的现实意义。提出了基于标准误差算法的油井工况识别方法,并建立了标准误差匹配算法的数学模型。首次提出匹配算法和图形差异程度具有关联性的论点,并对标准误差算法和经典匹配算法的可靠性进行了详细的对比研究,研究结果表明标准误差匹配算法和图形差异程度具有极高的关联性,并且标准误差算法的可靠性要优于目前广泛应用的经典匹配算法。通过对大量示功图的识别表明,基于标准误差匹配算法具有非常高的识别可靠性。特别是对微小差别的示功图具有极高的识别精度。

Study on Oil Production Conditions Recognition Based on Standard Error Algorithm

The dynamometer card contains plenty information features of equipment status, technology and conditions in oil production. It is of great significance to identify such important information accurately, rapidly and conveniently for safe and reliable operating of the pumping unit, improving of oil production technology, real time monitoring of oil production performance. This paper presents a new method to recognize oil production equipment conditions based on standard error algorithm and also develops mathematical model of standard error matching algorithm. An argument about matching algorithm correlating to the degree of difference between graphics is first proposed. The reliability of standard error matching algorithm and classic matching algorithm are compared in detail. Intensive study shows that standard error matching algorithm highly correlates to the degree of difference between graphics, and the reliability of standard error algorithm is better than that of the classical matching algorithms. A large number of dynamometer cards recognitions have shown that standard error matching algorithm has a fairly high reliability. Furthermore, this algorithm has terrific recognition accuracy especially to those dynamometer cards with minute differences.