Data-driven generation of phonetic broad classes, based on phoneme confusion matrix similarity

Abstract. Urban fire continues to be a persistent disaster, especially with the proliferation of highly dense urban settlements. As a response, several measures were established to help mitigate the losses caused by fire including simulating the fire spread. The cellular automaton system has been widely used to simulate the complex process of fire development along with Physics-based models. A data-driven approach has been rarely employed. This paper presents the result of incorporating machine learning techniques to the existing cellular automaton based urban fire spread models. Specifically, instead of manually calculating the ignition probability of each cell in the automaton, the Extreme Learning Machine (ELM) was used to learn the ignition probability from the historical data. After building the model, its performance was evaluated using the data collected from the four fires in Basak, Lapu-Lapu City. By using a confusion matrix to compare the actual and the predicted values, the Burned Actual – Burned Predicted relationship was derived. Results suggest that the proposed method can effectively describe the development of fire, and the model accuracy is quite good (i.e., the Burned Actual - Burned Predicted relationship ranges from 78% to 83%). Lastly, the study was able to demonstrate the possibility of using a data-driven approach in creating a simple cellular automaton fire spread simulation model for urban areas. Further studies utilizing more fire incident data on with varying properties is recommended.

Download Full-text

Performance Improvement of Wells Augmented Stuck Pipe Indicator via Model Evaluations

10.2523/iptc-21455-ms ◽

2021 ◽

Author(s):

Meor M. Meor Hashim ◽

M. Hazwan Yusoff ◽

M. Faris Arriffin ◽

Azlan Mohamad ◽

Tengku Ezharuddin Tengku Bidin ◽

...

Keyword(s):

Real Time ◽

Confusion Matrix ◽

Operation Time ◽

Oil And Gas Industry ◽

Data Driven ◽

Trend Detection ◽

Real Time Monitoring ◽

Operational Costs ◽

Drilling Parameters ◽

Drilling Data

Abstract The advancement of technology in this era has long profited the oil and gas industry by means of shrinking non-productive time (NPT) events and reducing drilling operational costs via real-time monitoring and intervention. Nevertheless, stuck pipe incidents have been a big concern and pain point for any drilling operations. Real-time monitoring with the aid of dynamic roadmaps of drilling parameters is useful in recognizing potential downhole issues but the initial stuck pipe symptoms are often minuscule in a short time frame hence it is a challenge to identify it in time. Wells Augmented Stuck Pipe Indicator (WASP) is a data-driven method leveraging historical drilling data and auxiliary engineering information to provide an impartial trend detection of impending stuck pipe incidents. WASP is a solution set to tackle the challenge. The solution is anchored on Machine Learning (ML) models which assess real-time drilling data and compute the risk of potential stuck pipe based on drilling activities, probable stuck pipe mechanisms, and operation time. The output of the analysis is built on a warning and alarm system that can be utilized by the engineers to refine and optimize the well construction activities; tackling the stuck pipe issue before it manifests. This solution is evaluated by comparing historical and real-time drilling parameters with the prediction data to generate an error analysis. On top of that, a confusion matrix is tabulated based on the analysis of warnings and alarms raised by the solution to rule out Type 1 and Type 2 errors. The WASP solution has demonstrated tolerably accurate predictions of drilling parameters with minimal warnings and alarms error. With the solution, the stuck pipe issue can be identified hours earlier before the actual stuck pipe was reported in the historical well. It is a powerful tool with the capability to pinpoint possible stuck pipe mechanisms for engineer's immediate analysis and intervention. Value creation from the WASP solution has been massive with a reduction in manhours of analysis, potential NPT events, and unexpected operational costs. Data-driven techniques are effective in preventing stuck pipe incidents and will be scalable to tackle other downhole issues such as loss of circulation, well control, and borehole instability.

Download Full-text