Incorporating fuel, lube oil and coolant chemistry data into Big Data matrix for Analytics algorithms allows savings in fuel consumption, increased productivity and engine reliability improvement
A large coal mine in South America needed to improve operating costs and decided to look in detail into mining equipment reliability and efficiency. Haul trucks premature engine degradation and failures were having an impact on productivity.
The trucks had typical Machine Health Monitoring Systems (MHMS), which collected robust data on engine and electrical system parameters, however this data had not helped to mitigate the problem. Even more, operating practices were closely monitored, and no real value opportunity had been identified in this aspect.
I-Optia proposed an innovative approach to the problem, incorporating real time monitoring data of physical and chemical parameters of lube oil, fuel and cooling fluid, aiming to early detection of deviations in combustion process, to correct them before accelerated degradation could lead to visible loss of engine performance.
Firstly, is less than optimal fuel atomization, which is usually detected at a stage when incomplete combustion residues have already caused abnormal mechanical deterioration of turbochargers, piston rings, camshafts, valve guides, etc.
A rapid change in chemistry of lube oil, detected in real time, is a good call for immediate attention, which can lead to a small corrective repair such as fuel injector replacement or injection pump recalibration.
Secondly, fuel contamination with humidity, corrosion particles, or other micro particles is a very frequent source of premature degradation of injection systems and combustion reaction.
A rapid change in chemical or physical parameters of fuel, detected in real time, is a good call for immediate attention, which can dictate simple correctives such as further fuel treatment before introducing it into the engine.
Similarly, a rapid change in chemical or physical parameters of coolant, is an indication of fluid degradation, that may lead to heat transfer deviations and cylinder hot spots, which affect combustion reaction.
Real time detection of these changes can lead to simple correctives such as replacing coolant or service to the cooling system.
Ai-Relia-BDt & Predictions of Combustion Deviations
Incorporating chemical and physical real time data of the mentioned fluids, with other information from MHMS, for AI Analysis enable real time predictions of the mentioned degradation modes.
This timely proactive approach helps to achieve significant improvement in fuel consumption as well as fleet reliability and availability.
Savings from implementing this innovation, for a fleet of 250 haul trucks, were calculated in the range of 4 to 8 million USD per year, just from fuel consumption reductions.