Production performance forecasting method based on multivariate time series and vector autoregressive machine learning model for waterflooding reservoirs
ZHANG Rui,JIA Hu
Table 4 Forecast results of all producers in well group No. 6.
Well
name
Actual average
daily output/m3
Average daily output from
machine learning/m3
Average daily output from history matching/m3 Relative error of
machine learning
Relative error of
history matching
X49-6 2.75 2.58 3.29 0.066 0.196
X50-5-1 0.59 0.57 1.37 0.035 1.404
X50-7-2 1.55 1.57 1.43 0.013 0.077
X51-5-1 2.79 2.48 3.38 0.125 0.211
X51-6 3.49 3.25 3.98 0.074 0.140