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 Geoenergy Science and Engineering (Journal of Petroleum Science and Engineering). The paper is titled "A transfer learning framework for well placement optimization based on denoising autoencoder".
Innovation: In traditional well placement optimization, the optimization algorithm calls the reservoir numerical simulator to evaluate the producers and injectors of the scheme and iterates to get the best scheme. The biggest limitation is that each optimization starts from zero, so the number of calls to numerical simulation is too many, and the solving speed is difficult to meet the needs of engineering. In this study, we propose a transfer optimization framework for well placement optimization based on de-noising autoencoder, which can be combined with any population-based evolutionary optimization algorithm to extract experience from past optimization tasks and thus accelerate the solution of optimization tasks. In the history of reservoir well placement optimization for decades, a large number of successful well placement optimization cases have been accumulated. Using the transfer optimization framework to extract, transfer and reuse the knowledge in the accumulated cases is expected to break the current "data island" dilemma and improve the solving efficiency. To solve the three core questions in transfer learning: "When to transfer? ", "Transfer what? ", "How to transfer? ". In this paper, a similarity measurement method for reservoir is proposed respectively, which can be used to find the past tasks with similar reservoir properties and avoid the negative transfer risk caused by the inter-task transfer which is too different. A method of evolutionary transfer among individuals of well placement optimization was proposed. The evolutionary direction of the optimization task was extracted by learning from the optimal solution of the past task by the optimal individual in the current iteration number, and then the evolutionary direction was applied to the fixed number individual in the current iteration number, so as to improve the fitness of the individual. This paper proposes a de-noising self-coding transfer framework suitable for reservoir well placement optimization, which can transform the extracted evolutionary direction into a mapping, obtain a closed mapping matrix through simple matrix operation, and complete knowledge transfer without too much computational burden.
Abstract: Well placement optimization is directly related to the recovery factor of reservoir development, and at present, the mainstream solution is an evolutionary algorithm. However, time-consuming numerical simulators need to be called to evaluate each alternative well placement scheme. Since the rules of well placement problems are universal, similar reservoirs will have similar well locations. Thus, knowledge transfer across similar well placement optimization tasks can expedite searching effectively. To this end, this paper proposes a novel transfer learning framework for well placement optimization to extract the potential well placement rules based on the feature extraction capability of a single-layer denoising autoencoder. The reconstruction mapping between the previous and present tasks is established to make the randomly generated well locations inherit the knowledge from the optimal well locations of the previous task, which helps the search direction of the evolutionary algorithm quickly bias to the optimal solution, thus, the solving of present task can be accelerated. The simplified denoising autoencoder holds a closed-form solution after derivation of the loss function, and the corresponding reuse of knowledge will not bring much additional computational burden on the evolutionary search. In addition, a similarity measure method between well placement optimization tasks is proposed to avoid a negative transfer. At last, comprehensive experiments on benchmark functions and well placement optimization instances are presented to evaluate the effectiveness of the proposed framework.
The Geoenergy Science and Engineering (Journal of Petroleum Science and Engineering) covers the fields of petroleum and natural gas exploration, production and flow in its broadest possible sense. Topics include: reservoir engineering; reservoir simulation; rock mechanics; petrophysics; pore-level phenomena; well logging, testing and evaluation; mathematical modeling; enhanced oil and gas recovery; petroleum geology; compaction/diagenesis; petroleum economics; drilling and drilling fluids; thermodynamics and phase behavior; fluid mechanics; multi-phase flow in porous media; production engineering; formation evaluation; exploration methods; CO2 Sequestration in geological formations/sub-surface, etc. The latest impact factor of the journal is 4.346, and the average impact factor IF in the past 3 years is 3.646. This journal currently appears in the JCR Q1 or the Chinese Academy of Sciences ranking 2.
Paper link:https://doi.org/10.1016/j.geoen.2023.211446
Citation: Ji Qi, Yanqing Liu, Yafeng Ju, Kai Zhang, Lu Liu, Yuanyuan Liu, Xiaoming Xue, Liming Zhang, Huaqing Zhang, Haochen Wang, Jun Yao, Weidong Zhang. A transfer learning framework for well placement optimization based on denoising autoencoder.Geoenergy Science and Engineering, 2023: 211446.
