| Autor: |
Tan Chaodong, Qin Peng, Li Loulou, Gao Zhaomin, Hua Wu, Song Wenrong |
| Rok vydání: |
2019 |
| Předmět: |
|
| Zdroj: |
2019 IEEE Eurasia Conference on IOT, Communication and Engineering (ECICE). |
| DOI: |
10.1109/ecice47484.2019.8942765 |
| Popis: |
Aiming at poor effect caused by unreasonable production system based on the analysis of production characteristics in plunger gas lift, a data-driven model for optimal decision-making is established, and control method of production and optimal time of switching well and casing pressure value was optimized. The decision-making model in this paper is based on the massive data of plunger gas lift production. The data model and pattern matrix of 16 influencing factors are constructed. Considering the time and attribute weight, the Euclidean metric between the decision-making wells and the historical wells is calculated by K-nearest neighbor algorithm. According to the similarity weight, three most similar historical wells are selected. The reasonable operation parameters of the wells to be determined can be calculated by the easy average method. The experimental verification by recursive method shows that the shut-in time after decision-making adjustment is shortened, the production of water and gas is increased, and the drainage efficiency is obviously improved. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
