Search

A set of dynamic remote monitoring method of production performance based on Machine Learning is proposed for the production process of electric submersible pump (ESP) well with multi-dimensional parameters. Aiming at dealing with the characteristics of multi-dimensional parameters in the complex production system, to implement dynamic monitoring of production performance. Helping the engineers at data centers to find the anomaly remotely and make response in a timely manner. It puts forward a procedure for large amount, high dimension and low information density production data in complex production system, using the dimensionality reduction algorithm to reduce the dimensionality into one comprehensive parameter changing over time, time series analysis algorithm for the production anomaly detection and prediction based on Machine Learning. The Principal Component Analysis (PCA) is used to reduce the dimensionality and extract the crucial information. The Autoregressive Integrated Moving Average (ARIMA) model is used to conduct timing anomaly detection, and fbProphet model is used to analyze the dimensionality reduced data to provide prediction of the production. With the dimensionality reduction, time series comprehensive parameter analysis and anomaly detection method based on Machine Learning, more than 40 ESP wells with 15 dimensions production daily parameters up to 1,000 days were analyzed, which realized the comprehensive description of ESP wells with multiparameter. Although the PCA retained only 47.73% of the information in the first principal component, which may be related with the low information density of industrial big data, the subsequent analysis proved the effectiveness. The time series analysis realized many times anomaly detection during the life period of each ESP well, and visualized the production data and the anomalous events. More than 100 anomalous events were detected in advance and which were robust corresponding to the subsequence real production events, among which 95% agreement rate is achieved. The procedure proposed reported the anomaly events with high confidence up to 90%, and low misstatement rate and omission rate, realized the production perception and abnormal detection in a timely manner. Based on this algorithm, the best time for the well intervention is determined, so that the loss of production is avoided and the revenue is maximized. The novelty of the procedure of Machine Learning using the multiple production data is in the ability to provide a solution of dealing with the low information density and high noise in the complex multi- dimensional production data of production wells, realize the comprehensive description, analysis and prediction of the production. It is helpful for engineers find the abnormalities in time, and will support the decision making of production, optimization and well intervention for the production.

Nitrogen (N2) injection-assisted flowback with coiled tubing (CT), CT flowback for short, is more and more widely used due to its convenience in implementation and flexibility in control. The unique feature of gas injection with CT is that the injection depth can be changed continuously (Zhou et al., 2011), which makes flowback become a more complex dynamic process. However, scarce research has been done on such a dynamic process, especially in the optimization of operation parameters such as CT running into hole (RIH) speed and N2 injection rate. In this paper, numerical models are established using the transient multiple flow simulator to reproduce the dynamic process of CT flowback. And based on this, the characteristics of CT gas injection are explored by comparing fixed-point gas injection with variation-depth gas injection. The influence of formation flowback capacity, CT RIH speed and N2 injection rate on CT flowback process is analyzed. And finally the optimization suggestions for CT operating parameters are put forward. The simulation results show that, compared with fixed-point gas injection, variation-depth gas injection (i.e. CT gas injection) can achieve a more controlled flowback, significantly reduce the gas injection pressure, and realize the gentle unloading of bottom hole pressure. Formation flowback capacity has an important effect on the CT flowback process, different formation flowback capacities result in different gas well unloading degrees. With the increase of formation flowback capacity, the unloading degree of gas well decreases. CT RIH speed can to a large extent control wellbore unloading rate, i.e. the flowback intensity. The increase in RIH speed is at the expense of the increase in gas injection pressure and will lead to a more aggressive bottom hole pressure unloading. So in determining the RIH speed, both surface gas injection facility limitations and downhole pressure control requirements need to be taken into account. For a given CT RIH speed, there is a minimum N2 injection rate, which is required to achieve the continuous drainage, i.e., the lower limit for N2 injection rate. On the basis of this lower limit, further increasing the N2 injection rate will accelerate the wellbore unloading, and reduce the peak of gas injection pressure, but there is also an upper limit for N2 injection rate beyond which the benefits are no longer apparent.

Vortex drainage gas recovery is a new technology to unload liquid from gas wells. However, its operating mechanism and efficiency are still not clearly understood, the dynamic analysis model in the wellbore two-phase vortex flow is still lacking. In this study, the forces of the liquid film in the wellbore gas-liquid vortex flow are analyzed based on the two-phase fluid dynamic theory. A dynamic analysis film model for prediction of the pressure drop in the wellbore two-phase vortex flow and the optimal helical angle of vortex tools is established. Besides, the physical simulation experimental facilities were set up and the wellbore two-phase vortex flow was obtained. Various parameters with vortex tools, including the wellbore pressure drop, the operating envelope and the critical flow rate, were examined. The influences of structure parameters of vortex tools on the drainage effect were analyzed. These structure parameters include the flow section of the flow channel, the leak-tightness around the flow channel and the helical angle. The vortex tool which has the best drainage effect was chosen and tested, and the extent of reduction of the critical flow rate with the optimal vortex tool was determined. Then, a liquid loading gas well in the field was chosen. Vortex tools were optimized and used in the wellbore of the gas well. The dynamic analysis model is verified by the experiment and field test. The results show that the dynamic analysis model of the liquid film in the wellbore two-phase vortex flow is reasonable. The effectiveness of vortex tools can be improved by enhancing the leak-tightness on both sides of the flow channel and reducing the flow section of the flow channel. After installing the optimal vortex tool, the wellbore pressure drop decreased by about 9.6%, the liquid flow rate increased by about 12.4% and the critical flow rate decreased by about 20%. The helical angle of the vortex tool has an optimal value, which can be calculated by using the dynamic analysis model. Research results may provide theoretical guidance for the optimization and design of vortex tools.

We will report recent progress on shorter wavelength recombination X‐ray laser research at SIOFM, including the observation of lasing at 46.8 A of the lithium‐like Ti19+ 4f‐3d transition, which was conducted at LF12 laser facility of SIOFM with Ti slab targets driven by an about 110 ps‐duration FWHM quasi‐Gaussian pulse (λ=1.05 μm). Comparison of long‐pulse (∼900 ps) and short pulse (∼110) driving Li‐like Si and Ca ions and H‐like F ion recombination X‐ray lasers will also be presented.