Petroleum analytics learning machine system with machine learning analytics applications for upstream and midstream oil and gas industry

A method for optimizing exploration, production and gathering from at least one well of oil and natural gas fields using a petroleum analytics learning machine system to maximize production while minimizing costs, comprising the steps of:

• collecting structured digital data and unstructured textual data from geological, geophysical, reservoir modeling, drilling, completion and production of crude oil, natural gas, ethane, butane, propane and condensate;
• receiving incoming data over a communications network and storing the incoming data into a system integration database by a processor based server or cloud based distribution of servers to provide collected data, the incoming data comprises digital exogenous data, real-time and historical endogenous data, historical data from surrounding production wells, hydraulic fracture completion data, and progress, status and maintenance data from new vertical and horizontal wells, including kickoffs, sidetracks, step-outs, pipeline gathering systems, compressor stations, and oil and gas sensor data including from public and private data sources;
• recording a time and depth for each data point of the collected data;
• cleaning the collected data to eliminate extraneous and noisy data;
• normalizing and storing the clean data;
• processing the normalized data to determine clusters of correlation in multi-dimensional space to identify a machine learned ranking of importance weights for each attribute;
• ranking said importance weights, and identifying patterns to enhance production of said at least one well of oil and natural gas fields;
• performing predictive and prescriptive optimization on the normalized data utilizing unique combinations of machine learning and statistical algorithm ensembles, including at least two of the following:
• linear and non-linear support vector machines and regressions, decision trees, hidden Markov models, random forests, neural networks, deep learning networks, bagging, boosting, feature selection, clustering, approximate and dynamic programming;
• classifying unstructured textual data to correlate with optimal production by utilizing progressive clustering with learned seeds, information extraction and retrieval, image recognition, textual mining, keyword and keyphrase extraction, semantic and sentiment analysis, entity and pattern recognition and knowledge discovery processing to capture the dynamics of said at least one or more wells of oil and natural gas fields and provide categorization results from labeled data sets to identify patterns;
• displaying data and analyses, transmitting recommendations, and receiving actual field actions and reactions on a graphical user interface on a network-enabled processing device over the communications network, the recommendations being based on the collected data of one or more wells, or one or more predicted conditions, communications with the one or more of the field systems is autonomous and personalized to steer disparate data simultaneously to operators working on vertical and horizontal wells, hydraulic fractures, or other field operations that are needed to improve future production from of a well in response to one or more trends, said one or more predicted conditions, or recommendations displayed on the graphical user interface connected to the petroleum analytics learning machine system;
• and
• wherein the petroleum analytics learning machine system utilizes:
• an exploration and production synthesizer of available data from wells in an area or play, in order to score and rank the combined importance weights of attributes to predict maximum production at minimum costs when convolved with specific attributes of each well;
• a real-time synthesizer to optimize drilling to match a designed pathway of a drilled well including hitting one or more target landing zones, while minimizing sinuosity and missed targeting of horizontal, diagonal and vertical components of the drilled wells;
• a real-time processor to:
• convolve importance weight values of attributes received by the petroleum analytics learning machine system from historical data and attribute data from each new well as it progresses in real time to predict future production of said each new well before oil and gas are delivered to the surface;
• and
• to utilize time-series attributes during each hydraulic fracturing stage to automatically classify production effectiveness of said each hydraulic fracturing stage and provide recommendations to maximize future production of said each new well;
• and
• wherein the recommendations are directed autonomously to optimize the production of oil, natural gas, and natural gas liquids while minimizing water production over time.