Formation Grain Size Profile Prediction Model Considering the Longitudinal Continuity of Reservoir Using Artificial Intelligence Tools

Grain size characteristics (d50, UC) of formation sands are crucial parameters in a sand control design. UC and d50 are commonly derived from sieve or laser particle size analysis (LPSA) techniques on a limited number of core samples in the process of drilling, which cannot represent the variations of grain sizes in the formation by the limited number of core samples. Moreover, staged and hierarchic design of sand control usually needs the whole longitudinal distribution profile of grain size. The grain size characteristics of the reservoir are formed in the process of a long history and have a good correlation with the formation environment of the sediments. Sand control design can only use test well data, because of lacking actual producing position cores. The vertical and horizontal anisotropy and heterogeneity of reservoirs bring difficulties and greater risks to the design of sand control schemes. Therefore, it is very important to find a simple and effective reservoir granularity prediction method. The existing prediction models by artificial intelligence method use single point logging data as eigenvalues to predict d50 and UC without considering the longitudinal continuity of data. This paper presents an efficient solution to predict grain size profile based on conventional logging curves by using four machine learning method (ANN, Random forest, XGBoost, SVM). In order to make full use of the geological continuity of the reservoir, the longitudinal continuous points according to the spatial correlation are adopted as the machine learning feature parameters from the perspective of geological analysis and the data-driven grain size profile prediction model are established by using the logging curve trend and background information, which further improves the prediction accuracy of the model and provides basic data for sand control. The ANN model of five point mapping has the best prediction effect in predicting d50 with a highest correlation coefficient 0.819 and a lowest error MAE 9.59. The XGBoost model of five point mapping has the best prediction effect in predicting UC with a highest correlation coefficient 0.402 and a lowest error RMSE 1.15. This method has been successfully used in offshore oil field in sand control optimization.

Low Profile Beam Steerable Antenna using PI Axis Posts for Massive-MIMO 5G Applications

5G networks promised to provide increased coverage, capacity and end-user throughput. Advanced antenna radiating system plays a very important role for the establishment of 5G networks. In this work, a novel low size profile enforced terminating boundary patch antenna array is proposed. This antenna array achieves a Hybrid beam steering with a low size through PI axis posts up to 33%. The antenna has a three-layer structure where a single layer PTFE composite based Rogers 5880 is utilized with a dimension of 15× 16 mm2. It operates at frequency band of 35 GHz with a wide bandwidth of 530 MHz and with a mean return loss of -28 dB. This radiating device is designed with the microstrip feed for the massive-MIMO 5G requirements.

A computer vision-based, in-situ springback monitoring technique for bending of large profiles

Bending processes have various advantages, such as less processing time, lower number of tooling parts, and cost compared to other manufacturing processes. However, one of the disadvantages of a bending process is the inevitable springback problem, which entails geometrical inaccuracy. Many researchers have made attempts to effectively measure springback in-line to control product quality and compensate for variability. While measurement tools and machines are available to measure springback, they might not be able to accommodate large products due to the size limit of measurement devices. Nevertheless, sensor-based monitoring is becoming critical to control product quality and to move towards Industry 4.0. In this paper, an in-situ springback monitoring technique for bending of large-size profiles is proposed to overcome the measurement restrictions for such profiles. A computer vision technique with the circular Hough transform was used to evaluate springback. The marked points on a profile were used to track the deformation of the workpiece. However, a weakness with image processing is to recognize the points from the complex background. Instead of employing global search for the points in an image frame, the marked points were detected by locally setting regions based on forming parameters such as a bending angle and stretching level. Springback was calculated by the change of position of those points. The results of springback monitoring were validated with the physically measured data from experiments. Based on this measurement technique, the feasibility of a computer vision-based springback monitoring in large-size profile bending is discussed in detail.