Transfer learning for well logging formation evaluation using similarity weights

Machine learning has been widely applied in well logging formation evaluation studies.However,several challenges negatively impacted the generalization capabilities of machine learning models in practical imple-mentations,such as the mismatch of data domain between training and testing datasets,imbalances among sample categories,and inadequate representation of data model.These issues have led to substantial insufficient identification for reservoir and significant deviations in subsequent evaluations.To improve the transferability of machine learning models within limited sample sets,this study proposes a weight transfer learning framework based on the similarity of the labels.The similarity weighting method includes both hard weights and soft weights.By evaluating the similarity between test and training sets of logging data,the similarity results are used to estimate the weights of training samples,thereby optimizing the model learning process.We develop a double experts’network and a bidirectional gated neural network based on hierarchical attention and multi-head attention(BiGRU-MHSA)for well logs reconstruction and lithofacies classification tasks.Oil field data results for the shale strata in the Gulong area of the Songliao Basin of China indicate that the double experts’network model performs well in curve reconstruction tasks.However,it may not be effective in lithofacies classification tasks,while BiGRU-MHSA performs well in that area.In the study of constructing large-scale well logging processing and formation interpretation models,it is maybe more beneficial by employing different expert models for combined evaluations.In addition,although the improvement is limited,hard or soft weighting methods is better than unweighted(i.e.,average-weighted)in significantly different adjacent wells.The code and data are open and available for subsequent studies on other lithofacies layers.

Experimental Analysis and Optimization of Timing Belt Guide Plate Material on Engine Operational Friction Loss

We focus on the friction loss measurement and reduction of the timing belt guide plates of a fourcylinder gasoline engine. To minimize the friction loss caused by the dynamic friction of the guide plates during timing belt motion and improve the efficiency of the internal combustion engine(ICE), we adopt four different plastic materials in fabricating the guide plates. With controlled engine boundary conditions and operational modes, an AVL electric dynamometer is used to measure the output of the engine. The results indicate that selecting polytetrafluoroethylene(PTFE) as the additive in fabricating the guide plates can effectively reduce the friction loss, so that the output torque and output power of the engine can be improved, thus reducing fuel consumptions. This work also has positive impact on the efficiency optimization of similar ICEs.

Application of resist-profile-aware source optimization in 28 nm full chip optical proximity correction

As technology node shrinks, aggressive design rules for contact and other back end of line（BEOL）layers continue to drive the need for more effective full chip patterning optimization. Resist top loss is one of the major challenges for 28 nm and below technology nodes, which can lead to post-etch hotspots that are difficult to predict and eventually degrade the process window significantly. To tackle this problem, we used an advanced programmable illuminator（FlexRay） and Tachyon SMO（Source Mask Optimization） platform to make resistaware source optimization possible, and it is proved to greatly improve the imaging contrast, enhance focus and exposure latitude, and minimize resist top loss thus improving the yield.