The purpose of fluids management is two fold--coordinating haul frequency and spill prevention for both oil and gas. The output of the program is requesting gauges (measurements) by humans or hauls (liquid takeaway by trucks).
The current algorithm has some large limitations. For liquid projection it uses the last previous calculated liquid rate per day based on the last two gauges. For spill prevention it applies this rate to the last most full tank as it assumes worst case is all liquid goes into that tank.
The new algorithm will include:
• Historical production rates by gauges (not just the last gauge to gauge)
• Liquid gas ratio
• Historical haul frequency
• Any data that shows artificial lift was installed or optimized
One requirement will be to map out wells to their associated tanks and weight how liquids are being distributed across multiple tanks. Next would be to find a specific algorithm to accurately predict when hauls are necessary based on production rates. We will also look into whether increased gauge frequency is necessary based on current production rates. Being able to link dates when the artificial lift has changed will allow reasonable expectations for when production rates will increase, which may result in the need for frequent gauging.
Another project would be identifying the ratio of liquid rates between shared tanks-that is, when a well dumps into a series of tanks, what ratio of the liquid goes into tank 1 and what goes into tank 2-if we knew this we could get away from assuming that 100% goes to each tank and taking the worst case scenario.