{"title":"利用PSO-LSSVM方法预测凝析气井井口流道亚临界两相流","authors":"Azim Kalantariasl , Arash Yazdanpanah , Ehsan Ghanat-pisheh , Negar Shahsavar","doi":"10.1016/j.upstre.2021.100057","DOIUrl":null,"url":null,"abstract":"<div><p><span>Several empirical correlations for prediction of wellhead<span><span> gas flow rate<span> have been presented in the literature. In this study, subcritical wellhead choke<span><span> flow data of gas condensate wells that cover a wide range of flow rates (5.4–113 MMSCF/D) and choke sizes (32–192 64th in) were used to develop an intelligent prediction method. Subcritical two-phase flow wellhead choke data from 193 tests of gas condensate wells in 10 fields have been used. Measured pressure drop across the choke, </span>gas to liquid ratio (GLR) and choke size were the input parameters. PSO-LSSVM method was applied to field-measured test data and optimized model parameters were obtained for prediction of gas flow rate as objective function. In addition, the results were compared with recently published empirical correlations developed for </span></span></span>subcritical flow. Accuracy of the proposed model were evaluated with error parameters; AARD (average absolute relative deviation), RSME (relative square mean error), and R-squared. Results show the superiority of the proposed model with high accuracy. Observed data and model prediction matched very well with R</span></span><sup>2</sup><span> of 0.9996 and RMSE of 1.46. In addition, five test data that have not been used in the process of model development (training and testing) were used to assess the generality of the proposed mode. Very good agreement between model prediction and observed gas flow rate data was obtained and can be used for estimation of gas flow rate of subcritical chokes with high confidence.</span></p></div>","PeriodicalId":101264,"journal":{"name":"Upstream Oil and Gas Technology","volume":"7 ","pages":"Article 100057"},"PeriodicalIF":2.6000,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Prediction of sub-critical two-phase flow through wellhead chokes of gas condensate wells using PSO-LSSVM method\",\"authors\":\"Azim Kalantariasl , Arash Yazdanpanah , Ehsan Ghanat-pisheh , Negar Shahsavar\",\"doi\":\"10.1016/j.upstre.2021.100057\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p><span>Several empirical correlations for prediction of wellhead<span><span> gas flow rate<span> have been presented in the literature. In this study, subcritical wellhead choke<span><span> flow data of gas condensate wells that cover a wide range of flow rates (5.4–113 MMSCF/D) and choke sizes (32–192 64th in) were used to develop an intelligent prediction method. Subcritical two-phase flow wellhead choke data from 193 tests of gas condensate wells in 10 fields have been used. Measured pressure drop across the choke, </span>gas to liquid ratio (GLR) and choke size were the input parameters. PSO-LSSVM method was applied to field-measured test data and optimized model parameters were obtained for prediction of gas flow rate as objective function. In addition, the results were compared with recently published empirical correlations developed for </span></span></span>subcritical flow. Accuracy of the proposed model were evaluated with error parameters; AARD (average absolute relative deviation), RSME (relative square mean error), and R-squared. Results show the superiority of the proposed model with high accuracy. Observed data and model prediction matched very well with R</span></span><sup>2</sup><span> of 0.9996 and RMSE of 1.46. In addition, five test data that have not been used in the process of model development (training and testing) were used to assess the generality of the proposed mode. Very good agreement between model prediction and observed gas flow rate data was obtained and can be used for estimation of gas flow rate of subcritical chokes with high confidence.</span></p></div>\",\"PeriodicalId\":101264,\"journal\":{\"name\":\"Upstream Oil and Gas Technology\",\"volume\":\"7 \",\"pages\":\"Article 100057\"},\"PeriodicalIF\":2.6000,\"publicationDate\":\"2021-09-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Upstream Oil and Gas Technology\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S266626042100027X\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENERGY & FUELS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Upstream Oil and Gas Technology","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S266626042100027X","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
Prediction of sub-critical two-phase flow through wellhead chokes of gas condensate wells using PSO-LSSVM method
Several empirical correlations for prediction of wellhead gas flow rate have been presented in the literature. In this study, subcritical wellhead choke flow data of gas condensate wells that cover a wide range of flow rates (5.4–113 MMSCF/D) and choke sizes (32–192 64th in) were used to develop an intelligent prediction method. Subcritical two-phase flow wellhead choke data from 193 tests of gas condensate wells in 10 fields have been used. Measured pressure drop across the choke, gas to liquid ratio (GLR) and choke size were the input parameters. PSO-LSSVM method was applied to field-measured test data and optimized model parameters were obtained for prediction of gas flow rate as objective function. In addition, the results were compared with recently published empirical correlations developed for subcritical flow. Accuracy of the proposed model were evaluated with error parameters; AARD (average absolute relative deviation), RSME (relative square mean error), and R-squared. Results show the superiority of the proposed model with high accuracy. Observed data and model prediction matched very well with R2 of 0.9996 and RMSE of 1.46. In addition, five test data that have not been used in the process of model development (training and testing) were used to assess the generality of the proposed mode. Very good agreement between model prediction and observed gas flow rate data was obtained and can be used for estimation of gas flow rate of subcritical chokes with high confidence.