1、SEQUENCE OPTIMIZATION IN LONGWALL COAL MININGBHP Billitons Illawarra Coal operates several longwall coal extraction systems in the Bulli and Wongawilli coal seams in the Southern Coalfields of the Sydney Basin, Australia. IC has applied a proprietary mixed integer linear programming open pit strateg
2、ic mine planning tool called Blasor to rapidly evaluate, for a number of longwall mining domains, the jointly optimal underground development strategy and mining sequences necessary. For each development scenario, the Blasor optimizer maximizes the discounted operating cash flow as the objective fun
3、ction, subject to mining, processing, and transportation capacity constraints and product blend constraints. When applied to the underground longwall domain sequence optimization problem, Blasor evaluates a set of carefully considered scenarios, each describing a feasible underground development and
4、 transport strategy. In this application, Blasor plays the role of a fair valuation tool for each major scenario, by determining the most valuable extraction schedule for each development scenario. This establishes a basis for comparing the economic merits of competing scenarios that are structurall
5、y different.Longwall mining, sequence optimization, economic evaluation, mixed integer programmingINTRODUCTIONBHP Billitons Illawarra Coal (IC) mining operations are focused on the mining of two seams, the Bulli and Wongawill seams of the Southern Coalfields of the Sydney Basin, and centered around
6、the coal export facilities at Port Kembla in New South Wales, Australia. Illawarra Coal typically produces a high quality metallurgical coal and lesser quantities of steaming coal at annual rates of about 8Mt. Coal is produced from three longwall mining systems at Appin, West Cliff, and the recently
7、 commissioned Dendrobium Colliery. Two Coal Preparation Plants (CPP) are supplied with coal from these longwalls, one located at West Cliff, the other located near BlueScope Steels Port Kembla blast furnaces. IC has a contract to supply 2.5 3.0 Mt per annum to BlueScope, with the balance being deliv
8、ered to the Port Kembla Coal Terminal (PKCT) for transport to Australian and overseas customers. Resource modeling and mine planning activities are conducted using the Mincom Minescape software and the Runge XPAC software. The XPAC scheduler requires the user to determine a “strategic” path through
9、the longwall domains, where these comprise a number of longwall blocks. Longwall blocks are typically 300 metres wide and 1 to 3 km in length. XPAC then defines the detailed extraction sequence which follows this strategic direction by using a myopic search of the different legal short term extracti
10、on options, and choosing the one which best meets the objective and relaxed constraints. XPAC usually has a difficult time in meeting hard blend constraints, but it can apply penalties as a method of forcing compliance.Run of Mine (ROM) and washed coal product transportation logistics are complex. R
11、OM coal from Dendrobium is transported via rail to Port Kembla with CPP rejects transported by road for disposal at the West Cliff Colliery. Product coal for export is then transported by road to the PKCT. ROM coal from the Appin Colliery is transported a short distance by road to the West Cliff CPP
12、, with wash plant products subsequently transported by road to the PKCT or BlueScope Steel. Fig. 1. Longwall domains for Appin/West Cliff CollieriesDefinition of longwall domains within the relevant coal seam is also a complex matter. As longwall mining results in varying degrees of subsidence, and
13、the Wollongong escarpment is a heavily populated area, definition of Coal Extraction Zones (CEZ) consistent with efficient longwall operations and minimal public disturbance requires a considerable planning effort. Typically, further data must also be gathered about the coal resource ahead of mining
14、, which when coupled with the various planning approvals, can require significant amounts of time. Underground transport options for coal produced by a longwall are also complex, particularly for the Appin/West Cliff Colliery (Fig. 1) where several domains could be transported to surface by one or o
15、ther of the hoisting systems, each of which has its own cost structure and capacity constraints. The Dendrobium Colliery is less complex, with a single coal hoisting system (Fig. 2).Fig. 2. Longwall domains for the Dendrobium CollieryTABLE 1. Coal Quality Variation by CollieryCoal qualities vary by
16、domain within each colliery. The ranges across all blocks in all domains are presented in Table 1. On a block-by-block comparison, the West Cliff domain offers the highest coking coal yields (the more valuable product), whilst Dendrobium tends to offer the lowest. Vitrinite, ash and specific energy values also vary quite widely between collieries. On a weighted average basis, the Appin domains
