Crude oil, or simply known as oil in the liquid state, is a result created after organic matters have transformed into this black substance over millions of years. The use of oil accelerated less than two centuries ago when it became a viable and efficient alternative to coal as the key source of energy.
Today, companies are still drilling into the ground to source more oil and natural gas (in its gaseous form), and the production process has always evolved to operate in more efficient fashion even though the era of alternative renewable energy is within the horizon.
Due to its viscous nature, this black gold, as it was known in the rise of the industrialization age, requires a lot of process to convert into a form we, the consumers, use. Juggling between gas, liquid and solid state, petroleum can bring an avenue of challenges to engineers of the oil and gas sector.
In the production process, oil viscosity can be difficult for machinery to handle. And in a critical case, this complication transforms into unwanted unplanned shutdown, resulting in machine breakdown, maintenance and restoration, and more importantly, halt in revenue which translates to yielding loss.
In other words, malfunction to machine parts can arise at any given time which is why it is crucial that engineers are constantly vigilant to detect issues in the production stage — or even better, stay ahead of a potential problem.
In the modern era of today, we have the technology capable of predicting possible issues in a form of mechanical maintenance which heavily relies on software-based artificial intelligence, known as statistics analytics maintenance.
This tool empowers the engineer users to run computational algorithms on data collected, from sensors which must be previously installed in their respective machines, and with the support of the pre-programmed commands, any data that deviates from the preset input. In addition, the algorithm is equipped with self-learning capacity, allowing it to evolve and adapt to new changes in order to aptly prepare for possible unexpected failure.
By utilizing this tool correctly, it would help the oil production and refinery plant avoid small and large-scale breakdown, and transitively boost production rate which means more yield. Despite a substantial setup cost, this is for companies to weigh their own abilities to balance if this cost could be compensated by the error-avoiding output. If so, the predictive analytics-based maintenance is an excellent option to consider as means to prevent unplanned shutdown.
How does the predictive characteristics of analytics maintenance compare to other types of maintenance.
The future approach of one of Thailand’s biggest energy conglomerates, PTT and how adopting machine learning can promote sustainability.