Application And Research On Non-Straight Well Trajectory Optimization Design And Decision-making Analysis

With the development of drilling technology, more and more high difficulty well had been explored in oil field. Mostly these are non-straight wells of which including direction wells, horizontal wells, sidetracking wells and long-reach wells of which mean the inclination angle unequal to zero. The design of well trajectory is very import of which effect the output and economical benefit. The design of well trajectory includes optimization design and real-time drilling design. The exploitation of well is very complicated engineering including not only well trajectory controlling technology but also decision experience which calls the formerly experience, for there are subjective factors among them which is difficult to conclude by mathematic module like well trajectory optimization but need new knowledge expression. The decision-making during the design makes the more signification of aided decision-making in drilling well field. For the better aided decision support of better non-straight well controlling, based on the system of horizontal well and sidetracking well computation and controlling software, making the research of non-straight well track in oil engineering as background, this paper inducts and improves on particle swarm optimization algorithm to make parameters optimization of designed function of well trajectory, and inducts and mends decision tree technology to make research on experiences decision, makes research on the position for cluster of well from optimization problem.Firstly, The trajectory design is very important to the optimization design. This paper mends the standard particle swarm optimization algorithm and makes parameters optimization of designed function of well trajectory. For the sake of predigesting optimization function module and improving optimization efficient, and also overcoming the demerit of former module that considering every kinds restricting factors so that it takes on roundly but complexity, the paper puts up new optimization target function of well trajectory based on the optimization theory. By importing standard particle swarm optimization algorithm, adopting punishing function method, the paper makes parameters optimization of target function. Aim at the disadvantage of standard particle swarm optimization algorithms that calls earliness, through adopting punishing function method, the paper makes improving to the algorithms that it will be forced to leave the position when it is nearly approaching to the local particle point by punish function, so to conquer the limitation of the standard particle swarm optimization algorithms. The new optimization target function has been put up that is conciseness of which makes the shortest length of space curve as the optimization target, adopts mathematic disposing of turning one of well trajectory curvature, and utilizes the sum of subsection equals to the target length to make sure the hitting target precision so to conclude expression of subsection parameters from the position parameters of target. The method is simple and creditability, and the example in practice of the method validates that it matches the requirement of engineering not only on the computing precision but also on the speed.Secondly, for realizing furthest optimization of well trajectory through predicting uncertain factors during the optimization course, by combining the Maximal Entropy Theory this paper makes improving toward classical decision tree method to put up new method of which validated by the engineering examples and gets the better effect. After building up the new method, the paper further improves the well trajectory optimization by machine learning of the decision habits of the operators to provide better decision support for the manager and operator.Thirdly, for the sake of perfect of well trajectory design, the paper makes research on the condition of ground which is also the important factor to the well trajectory design of which the signify base of designing of horizontal and directional well. This paper analysis the problem...