Explosion caused by flashing liquid in a process vessel

Development of an AI-based image/ultrasonic convergence camera system for accurate gas leak detection in petrochemical plants

Outdoor pipeline leaks are difficult to accurately measure using existing concentration measurement systems installed in petrochemical plants owing to external air currents. Besides, leak detection is only possible for a specific gas. The purpose of this study was to develop an image/ultrasonic convergence camera system that incorporates artificial intelligence (AI) to improve pipe leak detection and establish a real-time monitoring system. Our system includes an advanced ultrasonic camera coupled with a deep learning-based object-detection algorithm trained on pipe image data from petrochemical plants. The collected data improved the accuracy of detected gas leak localization through deep learning. Our detection model achieves an mAP50 (Mean average precision calculated at an intersection over union (IoU) threshold of 0.50)score of 0.45 on our data and is able to detect the majority of leak points within a system. The petrochemical plant environment was simulated by visiting petrochemical plants and reviewing drawings, and an outdoor experimental demonstration site was established. Scenarios such as flange connection failure were set under medium-/low-pressure conditions, and the developed product was experimented under gas leak conditions that simulated leakage accidents. These experiments enabled the removal of potentially confounding surrounding noise sources, which led to the false detection of actual gas leaks using the AI piping detection technique.

Investigation of an accident in a resins manufacturing site: the role of accelerator on polymerisation of methyl methacrylate

This paper analyzes the effect of an accelerator on the polymerisation of methyl methacrylate (MMA). This study is based on the results of an investigation of an accident in a manufacturing site for resins located in the United Kingdom. As sequence of event to cause the accident the following was assumed: during an unattended batch process a runaway undesired polymerisation of methyl methacrylate occurred, generating rapid vaporisation of monomer, which in contact with an ignition source, led to an explosion followed by a fire. Since no initiator for the polymerisation reaction had been jet added to the blend, it was supposed that the accelerator contributed to the onset of the undesired polymerisation. The accelerator involved in the accident t has therefore been tested by differential scanning calorimetry and adiabatic calorimetry. The experimental data allowed the authors to prove the hypothesis made and to define safety ranges for the polymerisation reaction.

The relationship between automation complexity and operator error

One of the objectives of process automation is to improve the safety of plant operations. Manual operation, it is often argued, provides too many opportunities for operator error. By this argument, process automation should decrease the risk of accidents caused by operator error. However, some accident theorists have argued that while automation may eliminate some types of operator error, it may create new varieties of error. In this paper we present six case studies of explosions involving operator error in an automated process facility. Taken together, these accidents resulted in six fatalities, 30 injuries and hundreds of millions of dollars in property damage. The case studies are divided into two categories: low and high automation complexity (three case studies each). The nature of the operator error was dependent on the level of automation complexity. For each case study, we also consider the contribution of the existing engineering controls such as safety instrumented systems (SIS) or safety critical devices (SCD) and explore why they were insufficient to prevent, or mitigate, the severity of the explosion.