In 2000 the Australian Coal Association Research Program announced the allocation of industry funding for a Landmark project targeting major advances in longwall mining automation. The aim of the project is to develop a longwall automation system that will reliably carry out the routine functions of cutting and loading coal, maintaining face geometry and manipulating roof supports without human intervention.

The Landmark Longwall Automation project extends over an initial three year period beginning in July 2001 with CSIRO and CMTE selected as the two research providers.

This project differs from many previous longwall automation attempts in that it is industry funded, it is focussed on practical (and ultimately commercialisable) outcomes rather than commercial imperatives, it has gained the support and cooperation of the major longwall equipment manufacturers but is not exclusive to an original equipment manufacturer (OEM).

The importance of communication design

The Landmark automation system integrates new enabling technologies, such as inertial navigation and industrial data networks, with existing longwall equipment and associated controllers from multiple manufacturers. In the early scoping and planning stages of the project several issues relating to the choice of data communication protocol and network architecture were identified as crucial to the practical success of the project.

The communication protocol needed to have widespread industry support and proven reliability. These protocols needed to be open-system to ensure it remained available, expandable and maintainable. A common communication design needed to apply across all components of the automation system to ensure mix and match capability with commonly used longwall equipment. Communication hardware equipment needed to affordable, reliable, maintainable, scalable, configurable and readily available. One of the implications of this was that future communication bandwidth requirements would be orders of magnitude greater than presently available on longwall equipment and control systems

Why broadband?

The immediate goal of the Landmark project is to achieve longwall automation to the level of on-face observation which requires operators to remotely monitor the extraction process and to intervene and handle exceptions as necessary. This level of remote monitoring requires the collection of considerably more sensor and monitoring information than would be the case in a completely hands-off automated system.

The Landmark control strategy calls for the use of on-face video cameras and extensive use of equipment condition sensors. The video images and equipment condition information is displayed at a remote monitoring station to allow the operator to continuously assess the performance of the mining process and to identify present or impending exception conditions.

The maximum achievable rate at which data can be transferred is limited by the bandwidth of the communication channel. The bandwidth associated with typical industrial-specific control networks is sufficient for the transfer of simple sensor and control information and remote diagnostics but generally inadequate for remote video and data intensive monitoring. Furthermore these networks do not easily integrate with external data networks and so do not readily offer the convenience of remote administration and data collection.

In response to the need for greater bandwidth and network integration there is increasing interest in the use of Ethernet for industrial networks.

By far the most commonly used data communications protocol is Ethernet which has become the de facto standard for business and personal computer networks. Because of this, Ethernet hardware is now relatively inexpensive, widely available and is designed to meet the high bandwidth demands of large commercial networks. This combination of price, availability and bandwidth has ensured the success of Ethernet and in turn its continuing development and improvement.

For these same reasons, Ethernet has found increasing application in factory automation where there has been a significant shift away from proprietary closed-system networks which inherently limit interoperability, maintainability and system integration across manufacturers.

Ethernet was selected for the Landmark project to satisfy the immediate data transmission and integration requirements and to meet the anticipated bandwidth demands for the foreseeable future.

Why EIP

While Ethernet defines the physical and data link layers of the network, it doesn’t define the protocol or language to be used by the networked devices or equipment. There are a number of commonly used network protocols but none of these suit the device level industrial control requirements such as interoperability and device self-identification.

To meet the growing need, industry developed a standard known as Control and Information Protocol (CIP). CIP allows for control of low-level devices on the factory floor and exchange of high-level control system information and is the same applications layer protocol used by ControlNet and DeviceNet.

Although CIP is actually independent of the underlying network there is increasing motivation to combine the proven reliability of CIP with the speed, affordability and off-the-shelf convenience of Ethernet for industrial control applications. The encapsulation of CIP over Ethernet is referred to as Ethernet/IP (EIP) and is being actively promoted and supported by the Open DeviceNet Vendor Association (ODVA).

Current status

Field trialing of the first stage developments of the Landmark project is ongoing at South Bulga Colliery. These trials have included the implementation of a wireless Ethernet link to the moving shearer which has allowed remote logging of data from shearer-mounted control equipment. It has also provided a direct Ethernet link between the shearer and the CSIRO research facilities in Brisbane via the mine underground network.

Two of the major longwall equipment manufacturers are developing Landmark compliant Ethernet/IP interfaces to their equipment and implementing the equipment specific control algorithms. The first major deliverable of the Landmark Longwall Automation project is the demonstration of automated face alignment which is scheduled for early 2003.