發(fā)布時間：2018-02-14 13:01

  本文關鍵詞： TCP信號 參數(shù)估計 粒子濾波 相似度 穿深　出處：《長江大學》2017年碩士論文　論文類型：學位論文

【摘要】：隨著經(jīng)濟的持續(xù)迅速發(fā)展,國家對石油能源的需求量已經(jīng)明顯不斷的在增加。因此,石油作為重要的能源之一,其占的地位也越來越重要。而油井射孔是石油開采中的一項非常重要的技術,它是為了實現(xiàn)井筒和預測產(chǎn)油地層之間的連通,以便提取石油。而射孔質(zhì)量的好壞直接關系著油井產(chǎn)能的高低,因此,評價分析射孔質(zhì)量的方法就顯得非常重要。本文在已有研究成果的基礎上,借鑒地震波形反演的思想,利用粒子濾波算法對油管傳輸射孔(TCP)信號進行分析處理。然后,提出了新的計算射孔穿深的方法。主要工作如下:1.了解了TCP監(jiān)測識別系統(tǒng),分析了TCP信號的特征和采集過程中的噪聲。最終得到射孔信號的幅值不超過5V,頻率范圍為40Hz~300Hz左右。2.了解并分析了信號參數(shù)估計的常用方法,提出和驗證了應用粒子濾波算法估計信號的相位。具體地研究了算法中部分參數(shù)的確定,有粒子的數(shù)目、粒子的移動步長、計算粒子與目標間相似度的方法。對比分析大量的MATLAB仿真結(jié)果得到,粒子數(shù)目確定為64、移動步長為0~1的隨機數(shù)、相似度方法為余弦相似度。當信噪比大于20d B時,得到的相位估值的相對誤差低于0.7%。3.利用粒子濾波算法估計射孔振動波表達式中的參數(shù),引用前人對振動波表達式中衰減系數(shù)的研究成果,將已獲取的射孔信號作為目標,將理論上分析的振動信號作為粒子。截取兩段不同時間段的射孔信號,并分別采用有衰減的粒子和無衰減的粒子,利用粒子濾波算法分別估計射孔信號的相位和頻率。最終得到信號的相位約為-0.6713,頻率約為59.9999Hz。4.了解并分析射孔的參數(shù)設計和射孔質(zhì)量的評價標準,以及與射孔穿深有關的因素,如混凝土靶的回彈值和強度、波速等。結(jié)合力學和地震學等方面的相關知識,估算出射孔彈爆破后波的速度3v≈1.6660×10 m/s,從而估算出射孔穿深d≈578.6010mm,然后進行穿透深度折算,結(jié)合其他相關的參數(shù)進行射孔質(zhì)量分析。
[Abstract]:With the continuous and rapid development of economy, the national demand for petroleum energy has been obviously increasing. Therefore, oil is one of the important energy sources. And well perforation is a very important technology in oil production, which is designed to connect the wellbore with the formation that is expected to produce oil. In order to extract petroleum, the quality of perforation is directly related to the productivity of oil wells. Therefore, it is very important to evaluate and analyze perforation quality. The particle filter algorithm is used to analyze and process the perforation signal transmitted by tubing. Then, a new method for calculating perforation depth is proposed. The main work is as follows: 1. Understand the TCP monitoring and identification system. The characteristics of TCP signal and the noise in the acquisition process are analyzed. Finally, the amplitude of perforation signal is less than 5V and the frequency range is about 40Hz / 300Hz. This paper proposes and verifies the application of particle filter algorithm to estimate the phase of signal. The determination of some parameters in the algorithm, the number of particles, the moving step size of particles, are studied in detail. By comparing and analyzing a large number of MATLAB simulation results, the results show that the number of particles is 64, the moving step is 0 ~ 1 random number, and the similarity method is cosine similarity. When the SNR is more than 20dB, The relative error of the phase estimation obtained is lower than 0.7. 3. The parameters of the expression of the perforated vibration wave are estimated by particle filter algorithm. The obtained perforation signal is taken as the target by citing the previous research results on the attenuation coefficient of the vibration wave expression. The vibration signals analyzed in theory are regarded as particles. The perforation signals of two different periods of time are intercepted, and the attenuated particles and the non-attenuated particles are used, respectively. The phase and frequency of perforation signal are estimated by particle filter algorithm respectively. Finally, the phase of the signal is about -0.6713 and the frequency is about 59.99Hz.4.The parameter design of perforation and the evaluation standard of perforation quality are understood and analyzed, as well as the factors related to perforation depth. For example, the rebound value and strength of concrete target, wave velocity and so on. Combined with the relevant knowledge of mechanics and seismology, the velocity of the wave after blasting is estimated to be 3v 鈮，

本文編號：1510761