Recently, the team of Zhang Kai has made new progress in the field of well placement optimization, and the related research has been published in the SPE Journal. The paper is titled " An Evolutionary Sequential Transfer Optimization Algorithm for Well Placement Optimization Based on Task Characteristics ".
Innovation: Traditional well placement optimization uses the optimization algorithm to call the reservoir numerical simulator to evaluate the pros and cons of the scheme, and iterates to get the best well locations. The biggest limitation is that each optimization starts from scratch, and the number of numerical simulation calls is too many, so the solving speed is difficult to meet the needs of the project. In this study, a sequential evolutionary transfer optimization algorithm based on task characteristics and autoencoder is proposed to solve the above problems. Because reservoir permeability, saturation, reservoir thickness and other characteristics have a recessive relationship with the optimal well locations, the relationship is the same or similar in each reservoir. Therefore, this algorithm can learn the solving experience from the completed well placement optimization task, extract and select the reservoir characteristics to obtain a characteristics matrix, and then use the single-layer denoising autoencoder to construct the mapping, map the optimal solution of the past reservoir to a dominant solution of the current task, so as to accelerate the optimization.
Abstract: Evolutionary transfer optimization (ETO) algorithms with the ability to learn from past tasks have made breakthroughs in more and more fields, when the experience embedded in the past optimization tasks is properly utilized, the search performance will be greatly improved compared to starting from scratch. Autoencoding Evolutionary Search (AEES) is an efficient evolutionary transfer optimization paradigm proposed in recent years, the solutions of each task are configured as input and output of a single-layer denoising autoencoder, and the across-problem mapping is established by minimizing the reconstruction error, which makes it possible to explicitly transfer the solutions across heterogeneous problems. However, despite the success of AEES, the population of the optimization task contains little information about the characteristics of the task and it is highly stochastic, especially in the early stages of searching, this restricts the effectiveness of the mapping constructed via AEES. On the other hand, most tasks do not save all candidate solutions in the search, which greatly limits the possibilities of traditional AEES applications, for example, a common engineering optimization problem in the oil industry, well placement optimization problems (WPO). To overcome such limitations, a sequential ETO algorithm for WPO problems based on task characteristics and an autoencoder is developed in this paper, it uses the implicit relationship between reservoir characteristics and optimal well locations to learn from past tasks, a mapping is calculated to transfer knowledge across tasks. The proposed algorithm aims to speed up the search for the optimal well locations and reduce the required time for WPO. The learned mapping is established by configuring the characteristics of past and current tasks as input and output of a single-layer denoising autoencoder, the derived mapping holds a closed-form transformation matrix across heterogeneous tasks, and the optimal solution of the past task can be easily transferred to a dominant solution of the current task by matrix calculation, thus it will not bring much computational burden in the evolutionary search while improving search performance. Furthermore, according to the specific task, the construction scheme of the matrix of characteristics can be flexibly extended to achieve effective search enhancement. Lastly, comprehensive empirical studies of well placement optimization and statistical analysis are carried out to verify the effectiveness.
The SPE Journal covers theories and emerging concepts spanning all aspects of engineering for oil and gas exploration and production, including reservoir characterization, multiphase flow, drilling dynamics, well architecture, gas well deliverability, numerical simulation, enhanced oil recovery, CO2 sequestration, and benchmarking and performance indicators. The latest impact factor of the journal is 3.602. This journal currently appears in the JCR Q1 or the Chinese Academy of Sciences ranking 3, ENGINEERING, PETROLEUM ranking 2.
Paper link:
Citation:
