摘要:
Changes occurring downhole during a fracturing process can create or reflect pressure signals. Capturing and evaluating such pressure waves during fracturing enables personnel to monitor, in real time or later, what happens downhole. When a fracture extends, a burst of acoustic noise is embodied in a pressure wave or signal, as is noise coming from other sources. By transforming time-based pressure signals to a frequency base, one can monitor this acoustic noise. In a particular implementation, a waterfall plot of frequency spectra at successive time slices of the original signal is used to determine frequency ridges, such as a ridge of decreasing frequencies indicates fracture extension and a ridge of increasing frequencies indicates either closure or proppant backing up in the fracture. Filtering, such as wavelet filtering, can be used. A fracturing process can be controlled in response to determining whether the fracture is extending.
摘要:
Changes occurring downhole during a fracturing process can create or reflect pressure signals. Capturing and evaluating such pressure waves during fracturing enables personnel to monitor, in real time or later, what happens downhole. When a fracture extends, a burst of acoustic noise is embodied in a pressure wave or signal, as is noise coming from other sources. By transforming time-based pressure signals to a frequency base, one can monitor this acoustic noise. In a particular implementation, a waterfall plot of frequency spectra at successive time slices of the original signal is used to determine frequency ridges, such as a ridge of decreasing frequencies indicates fracture extension and a ridge of increasing frequencies indicates either closure or proppant backing up in the fracture. Filtering, such as wavelet filtering, can be used. A fracturing process can be controlled in response to determining whether the fracture is extending.
摘要:
Development of an oil or gas reservoir is controlled using a neural network and genetic algorithm program to define a neural network topology and the optimal inputs for that topology. The topology is defined from identified and selected parameters (1) associated with the formation or formations in which actual wells are drilled in the reservoir, and (2) associated with the drilling, completion and stimulation of those wells, and (3) associated with the oil or gas production from the wells. Subsequent drilling, completion and stimulation of the reservoir is determined and applied based on hypothetical alternatives input to the topology and resulting outputs.
摘要:
Development of an oil or gas reservoir is controlled using a neural network and genetic algorithm program to define a neural network topology and the optimal inputs for that topology. The topology is defined from identified and selected parameters (1) associated with the formation or formations in which actual wells are drilled in the reservoir, and (2) associated with the drilling, completion and stimulation of those wells, and (3) associated with the oil or gas production from the wells. Subsequent drilling, completion and stimulation of the reservoir is determined and applied based on hypothetical alternatives input to the topology and resulting outputs.