Feed-Forward Neural Network Based Petroleum Wells Equipment Failure Prediction

In the oil industry, productivity of wells depends on performance sub-surface equipment system. These systems often have problems stemming from sand, corrosion, internal pressure variation, or other factors. order to ensure high and avoid high-cost losses, it is essential identify source possible failures in early stage. However, this requires additional maintenance fees human power. Moreover, losses caused by these may lead interruptions whole production process. minimize costs, paper, we introduce a model for predicting failure based processing historical data collected multiple sensors. The state system predicted Feed-Forward Neural Network (FFNN) with an SGD Backpropagation algorithm applied training Our model’s primary goal potential malfunctions at stage process’ continued performance. We also evaluated effectiveness our against solutions currently available industry. results study show that FFNN can attain accuracy score 97% given dataset, which exceeds models provided.