Abstract: The invention provides a collaborative optimization method for gas injection huff-n-puff parameters in tight oil reservoirs. The method relates to the technical field of oilfield development parameter optimization, including: (1) establishing the numerical simulation model that accurately describes the actual oil reservoirs; (2) determining the optimization parameters of gas injection huff-n-puff, giving the optimization range of gas injection huff-n-puff parameters and other variable constraints, and establishing an optimization objective function; and (3) using the particle swarm optimization algorithm to solve the objective function constructed based on a collaborative optimization model for gas injection huff-n-puff parameters. Then the optimal gas injection rate, gas injection time, soaking time, and production time after the collaborative optimization for gas injection huff-n-puff parameters of the reservoir are obtained.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for predicting the intent of an agent in an environment. One of the methods includes obtaining context data characterizing an environment, the context data comprising data characterizing a plurality of agents in the environment; and generating, based on processing the context data using a neural network having a plurality of network parameters, a respective predicted stationary state classification over a plurality of categories for each of one or more target agents of the plurality of agents in the environment, the categories including one or more categories that indicate that the target agent is a stationary non-participant of traffic in the environment.