Komatsu on intelligent longwall mining in the 4th Industrial Revolution

At APCOM 2019 in Wroclaw, Komatsu’s Willem Fourie outlined the company’s vision and reality for the digital mine ecosystem. The latter part of the 20th century witnessed the speeding up of the move to the 4th Industrial Revolution with Komatsu Mining Corporation (KMC) and its customer mining companies using machine learning and deep learning to support longwall systems. Today already available longwall automation features include face alignment, personal proximity detection, 3D geometric control and visualisation as well as shearer pitch steering and face conveyor variable speed drives. Going forward, further advances using the Faceboss control system are adding to today’s technology in areas such as longwall visualisation, data solutions in real time, longwall planning and the remote management interface.

The internet, big data and improvements in processing power mean that more can be measured, monitored and analysed. The mining industry, however, has generally struggled to integrate these technologies into the mining environment. KMC has pioneered many of the systems and processes to capture machine data and transform it into information and useful decision-making knowledge.

Fourie discussed how digital enabling technologies allow the transition from a reactive unit based monitoring approach to a proactive system of systems methodology utilising cutting edge modern technologies. System performance monitoring also enables the linking together of various data sources, not just specific machine detail, but also the operator performance, training, maintenance, logistics and most importantly safety systems. KMC says it has been pioneering the age of smart, connected products and complementing that with advanced analytics to create a digital clone equivalent of its physical counterpart.

The smart connected products of a KMC longwall system include a Joy shearing machine, armoured face conveyor (AFC), roof supports, stageloader, crusher, and mobile belt tailpiece. The ongoing aim is to detect patterns and systematic relationships amongst the operator inputs, control system parameters, sensor data and alarm/event information. Statistical techniques, such as regression analysis, are utilised to optimise cutting profiles, detect roof cavities and improve roof support cycle times. Predictive diagnostics and rule use cases are implemented to analyse real-time data and alarm of changing longwall characteristics.

Equally for a room and pillar section there are many pieces of equipment that need to operate in harmony to deliver steady production, keep cost in control and ensuring a safe workplace. For this purpose, KMC has developed a product called JoyConnect to capture, channel and deliver information to the right place at the right time helping to address decision support challenges such as bottlenecks, downtime, and dependencies between equipment.

KMC has implemented an analytics platform capable of ingesting data from machines all around the world to analyse data and assist in decision making. “This knowledge is used to drive safety, productivity and reliability at many operations. It provides a deeper link between the boardroom and the coal face enabling improved decision making and management. The availability of information for automation has traditionally been used in a reactive manner, but with new technology, its real power is as a pro-active and predictive intervention in the mining operation.”