Search

During industrial offshore deep-water drilling process, gas kick event occurs frequently due to extremely narrow Mud Weight (MW) window (minimum 0.01sg) and negligible safety margins for the well control purposes. Further, traditional gas kick detection methods in such environments have significant time-lag and can often lead to severe well control issues, and occasionally to well blowouts or borehole abandonment. In this study, firstly, the raw field data is processed through data collection, data cleaning, feature scaling, outlier detection, data labeling and dataset splitting. Additionally, a novel data labeling criterion for gas kick risks is proposed where five kick risks (Indicated by different colors in this study) are defined based on three key indicators: differential flow out (DFO), kick gain volume (Vol), and kick duration time (Time). Kick risk status represents one of the following cases: Case 0 - No indicators are activated (Green), Case 1 - Multi-drilling parameters deviation or DFO is activated (Orange), Case 2 - DFO and Vol are simultaneously activated (Light Red), Case 3 - DFO and Time are simultaneously activated (Light Red), Case 4 - DFO, Vol and Time alarms are simultaneously activated (Dark Red). Then, a novel data mining method using Long Short-Term Memory (LSTM) Recurrent Neural Network (RNN) is presented for early detection of gas kick events by analyzing time series data from field drilling process. The network parameters such as number of hidden layers and number of neurons are initialized to build the LSTM network. The learned LSTM model is evaluated using the testing set, and the best LSTM model (six (6)-layers eighty (80)-nodes (6 L*80 N)) is optimally selected and deployed. The accuracy of deployed LSTM model is 87 % in the testing dataset, which is reliable enough to identify the kick fault during the deep-water drilling field operation. Lastly, the LSTM model detected the gas kick events earlier than the "Tank Volume" detection method in several representative case studies to conclude that the application of LSTM model can potentially improve well control safety in the deep-water wells with narrow MW windows. [ABSTRACT FROM AUTHOR]

Cover crops have recently been generating a lot of interest in agriculture for their agronomic benefits, including erosion control, nutrient restoration, and increased cash crop yield. However, the value of these benefits is not well established and likely highly variable. The factors that influence a farmer’s decision to adopt or not adopt cover crops are also a point of interest to conservationists and policy makers. As part of a long-term study estimating the benefits and costs of cover crops, this study analyzed five years of historic information collected from Indiana farmers, who either grew or did not grow cover crops. No significant impact was found from cover crops on cash crop yields, but several interesting correlations in cover crop use were found. Cover crop use was higher on steeper fields. There was also a high correlation between the use of cover crops and no-till. [ABSTRACT FROM AUTHOR]

The ability of hydropower to adapt the electricity generation to the demand is necessary to integrate wind and solar energy into the electrical grid. Nowadays, hydropower turbines are required to work under harsher operating conditions and an advanced condition monitoring to detect damage is crucial. In this paper the methodology to improve the condition monitoring of Pelton turbines is presented. First, the field data obtained from the vibration monitoring of 28 different Pelton turbines over 25 years has been studied. The main types of damage found were due to fatigue, cavitation and silt erosion. By analyzing the vibration signatures before and after maintenance tasks, the symptoms of damage detected from the measuring locations were determined for each case. Second, a theoretical model using numerical methods (FEM) was created in order to simulate the dynamic behavior of the turbine. The model was validated with the results obtained from on-site tests that were carried out in an existing turbine. The deformations and the stresses of the runner under different operating conditions could then be computed. The calibrated model was used to analyze in detail the effect of misalignment between nozzle and runner. In historic cases, this abnormal operating condition lead to severe damage in the turbine, due to the effect of fatigue in some locations of the buckets. The model reproduced well the symptoms detected in the field measurements. The stresses could be calculated, which eventually can be used to estimate the remaining useful life of the turbine. [ABSTRACT FROM AUTHOR]

• Striking vehicle type for pedestrian crashes has changed markedly from 20 years ago. • Pedestrian chest injuries increased significantly caused by pickup trucks and SUVs. • PedAEB is less effective in reducing chest injuries than head and leg injuries. • This study highlights the importance of chest injury prevention for pedestrians. The objectives of this study were to 1) collect and analyze recent pedestrian crash cases for better understanding of the pedestrian injury distribution and mechanism, 2) use computational simulations to reconstruct pedestrian cases and estimate potential benefit of pedestrian automatic emergency braking (PedAEB) in reducing pedestrian injury risks, and 3) estimate how future pedestrian crash distribution might influence priorities for pedestrian protection. Analyses of national crash-injury dataset showed that the overall number of pedestrians in crashes as well as the serious and fatal pedestrian injuries in the U.S. have been increasing in recent years. Striking vehicle type has changed (i.e., decreased proportion of passenger cars and increase of SUVs and pickup trucks) from 20 years ago mirroring changes in the fleet distribution of vehicle sales. A total of 432 pedestrian injury cases were generated by linking the Michigan trauma data and police-reported crash data from 2013 to 2018. Among the linked cases, pickup trucks and SUVs were involved in crashes with more injuries across body regions. Notably, AIS 3+ chest injuries occur at almost the same rate as lower extremity injuries. A method, combining MADYMO simulations (n = 3,500), response surface model, and data mining, was developed to reconstruct 25 linked pedestrian crash cases to estimate the effectiveness of PedAEB. Based on national field data and MADYMO simulations, PedAEB was estimated to be effective in reducing the risk of head and lower extremity injuries but is relatively less effective in reducing the risk of chest injuries. The increased proportions of SUVs and pickup trucks in the vehicle fleet and the higher penetration of PedAEB may highlight the importance of future research into chest injury risk for pedestrian protection. [ABSTRACT FROM AUTHOR]

The simplified two-zone model to describe fluid flow in steady-state horizontal holes with multistage hydraulic fracturing is proposed, and the analytical solution in order to calculate sizes of hydrofractures is presented. The new procedure to determine size of multistage hydrofractures in horizontal holes is tested using field data on China deposits. The calculation results prove the efficiency of the procedure in the determination of steady-state conditions. The procedure can be used in optimization of hydraulic fracturing. [ABSTRACT FROM AUTHOR]

• Successful risk assessment of shallow gas hazards during industrial deep-water drilling process based on neural networks. • First data-driven modeling study analyzing huge valuable full-scale unbalanced seismic data from field by SMOTE-ENN method. • Fully connected DNN model classify the shallow gas hazard smartly. The geological conditions of deep water in the South China Sea are complex. Shallow gas is often encountered during deep-water drilling, which is likely to cause serious accidents such as blowouts and fires. This paper studied the identification methods of shallow gas during deep-water drilling based on neural networks as follows. First, the identification criteria of shallow gas were obtained from the seismic characteristics of shallow gas. Three-Dimensional (3D) seismic data from the field was used to identify the shallow gas. A dataset of thirteen seismic attributes was collected, including seismic amplitude, frequency, and velocity. The features were refined to optimize the seismic attributes. Secondly, this study employed Back Propagation (BP) neural network, BP neural network optimized based on Particle Swarm Optimization (PSO-BP), probabilistic neural network (PNN), and fully connected Deep Neural Network (DNN) as the algorithms for shallow gas identification according to the characteristics of the dataset. Results showed that the fully connected DNN performed better than the other three neural network algorithms. The predicted shallow gas based on neural networks is consistent with its actual distribution estimated by seismic data, proving the feasibility and effectiveness of the shallow gas identification. Our study pointed out that the fully connected DNN had great advantages in fitting the highly nonlinear mapping relation of the multi-level multi-source dataset, which provided a reference for similar classification problems. [ABSTRACT FROM AUTHOR]

A single-porosity model is developed to deal with a fracture zone slug test in a large dip angle by assuming the fracture zone causes a downward regional flow. For the oscillatory response, a larger dip angle causes larger amplitude while introduces little impact on period. The effective water length, an important parameter necessary for analyzing the oscillatory response, is proven to be independent of the dip angle and can be evaluated using the available horizontal formation methods. The dip angle effect is more pronounced for a larger storage coefficient. An empirical relationship is developed to evaluate the limiting dip angle, below which the dip angle effect is negligible. Field data analysis of a slug test in a 47° dip angle fracture zone indicates that neglecting the dip angle can result in a 27% transmissivity over estimation and a 53% storage coefficient under estimation. [ABSTRACT FROM AUTHOR]

Abstract: The BMW Group has launched two micro-hybrid functions in high volume models in order to contribute to reduction of fuel consumption in modern passenger cars. Both the brake energy regeneration (BER) and the auto-start-stop function (ASSF) are based on the conventional 14V vehicle electrical system and current series components with only little modifications. An intelligent control algorithm of the alternator enables recuperative charging in braking and coasting phases, known as BER. By switching off the internal combustion engine at a vehicle standstill the idling fuel consumption is effectively reduced by ASSF. By reason of economy and package a lead–acid battery is used as electrochemical energy storage device. The BMW Group assembles valve-regulated lead–acid (VRLA) batteries in absorbent glass mat (AGM) technology in the micro-hybrid electrical power system since special challenges arise for the batteries. By field data analysis a lower average state-of-charge (SOC) due to partial state-of-charge (PSOC) operation and a higher cycling rate due to BER and ASSF are confirmed in this article. Similar to a design of experiment (DOE) like method we present a long-term lab investigation. Two types of 90Ah VRLA AGM batteries are operated with a test bench profile that simulates the micro-hybrid vehicle electrical system under varying conditions. The main attention of this lab testing is focused on capacity loss and charge acceptance over cycle life. These effects are put into context with periodically refresh charging the batteries in order to prevent accelerated battery aging due to hard sulfation. We demonstrate the positive effect of refresh chargings concerning preservation of battery charge acceptance. Furthermore, we observe moderate capacity loss over 90 full cycles both at 25°C and at 3°C battery temperature. [Copyright &y& Elsevier]

In the last two decades it has become clear that component lifetime distributions are seldom exponential. Early failures due to flaws have been of major concern. Various models to cater for this observation have been suggested. The models have been developed from a component rather from a system point of view. Furthermore, no substitute for the heavily criticized MIL-HDBK-217-type prediction handbooks has been offered. This paper describes a recently developed methodology to arrive at realistic component reliability characteristics, which take into account the non-exponential component lifetime distributions. These characteristics are developed from a system's point of view, which makes it easier for the system manufacturer to assess the reliability of their systems in the field. The methodology is based on six years of research in field performance of electronic systems. The research has been carried out in close co-operation with industrial companies. [ABSTRACT FROM AUTHOR]