Available Technology

Chatelet: More Accurate Modeling for Oil, Gas or Geothermal Well Production CR-3194

Researchers at Berkeley Lab have developed a probabilistic algorithm for the modeling of exponential curves generated by the movement of multiphase material (liquids and gases) through multiple permeability regimes during oil, gas or geothermal well production. The algorithm, called Chatelet, can fit and clean data in a method analogous to a Kalman filter. Chatelet captures not only the properties of the overall curve, but also localizes short-time events, including transients. This greatly reduces biases and uncertainty and allows transients to be identified and studied in isolation. The current approach is to fit an ad-hoc modified exponential function to these curves. Models are fit to data using a non-linear, least-square algorithm. Once a curve has been fit to the data, it is projected forward to estimate future, rate-time behavior. However, these estimates can vary widely in response to the parameterization of the model and the method used to fit the data.
Truer to collected data - Accurately determines the start and end times of transient curves - Reveals the consequences of transients - Improves the ability to extrapolate curves beyond the initial data
Internal Laboratory Ref #: 
Patent Status: 
Available for licensing as a stand-alone application or as code modules to be incorporated with other analysis tools.
Share to Facebook Share to Twitter Share to Google Plus Share to Linkedin