Water Management Models Save Time and Money
Using the software to create multiple design options, engineers actively manage Roy Hill’s dewatering program.

Modeling the hydraulic network allowed the Roy Hill water management team to cost effectively balance dewatering
needs with operational consumption.
Water management represents a significant area for cost control for mining operations. A properly engineered network will cost less to build, operate and maintain, lowering capital expenses (capex) and operating expenses (opex). Moreover, ensuring all water management activities are completed in timely fashion, they should not affect the mine plan or mining operations, avoiding additional costs to the business.

Take the new $A10 billion Roy Hill iron project in the Pilbara region of Western Australia as an example. It will produce 55 million metric tons per year (mt/y) of iron ore from multiple surface pits and production benches. Before mining operations can begin, the site water table must be lowered to provide dry digging conditions. The mine will also require water to process ore. The mine’s water system will consist of an evolving, integrated network of water supply and dewatering operations extending across most of the 300 km2 mining lease.

Within the water supply scope there are various consumer facilities with differing feed requirements in terms of both water quality and quantity. Up to 22 megaliters per day (ML/d) of high-quality water will be required for final ore processing and potable water (after treatment) purposes. This is supplied by 24 dedicated supply bores from the Stage 1 borefield. A further 34 ML/d of lower quality water will be required for ore processing makeup and dust suppression supplies. This is currently delivered by eight separate construction water supply bores and the greater, progressively expanding dewatering network.

Over the 20-year life of the mine, there will be two distinct dewatering systems, raw and saline. Currently on site, there are approximately 48 active raw dewatering bores. The raw dewatering system is estimated to peak at 105 active bores in 2023. As mining operations deepen and progress southwest, the dewatering system will turn progressively more saline; requiring the separate network. Saline dewatering operations are not expected to begin until 2018, with a peak of 75 bores expected in 2021.

During peak operations, around 213 bores (138 raw and 75 saline) will pump approximately 140 ML/d through a network of 220 km of high density polyethylene (HDPE) pipework ranging in diameter up to DN800. These peak quantities will be maintained as dewatering borefields are progressively installed and removed as mining operations migrate.

Engineering the Network
Roy Hill’s Water Management Team, as part of Operations Engineering Services, is responsible for all facets of site raw water supply and dewatering, such as planning, design, construction and operations. This includes but is not limited to:
• Ensuring adequate supplies to the various locations;
• Providing water of acceptable quality to the desired locations;
• Preserving limited good quality groundwater supplies;
• Supplying water to new mining locations;
• Maintaining the ground water level a minimum 1 m below the base of active mining operations;
• Ongoing expansion of dewatering operations sufficiently ahead of mining;
• Spatial fit of the water infrastructure within the operational footprint to minimize conflict; and
• Disposing of excess dewatering by acceptable methods during periods of water surplus.

They are achieving this using Water- GEMS software, a mapped, fully integrated hydraulic model. The model allows a detailed representation of the water system providing a full understanding of the dynamic nature of the network; both through system expansion and changing yield and quality characteristics of the bores over time (Immature -> Intermediate -> Mature).

Water management at Roy Hill has a total sustaining capital budget on the order of A$440 million over the course of the mine’s life. The Water Management team is currently looking at ways to optimize the design of future systems to reduce the potential size/rating of infrastructure and its overall cost. For example, a pipe with a lower pressure rating can be as much as 33% cheaper than the equivalent size at a high-pressure rating. Through smart design, this may be the most effective way of reducing costs while confidently maintaining the desired safety factor.

As far as operational expenses, the project team has looked to minimize the cost from the fueling and servicing of gensets powering the pumps, which is estimated at A$4 million per year. The two methods identified are an overall reduction in system run hours and optimization of bore running based on fuel use. The two methods could provide an overall 20% reduction in service cost and fuel usage.

Water management must operate ahead of mining operations. Given the dynamic nature of the mining operations, the project team needs to be in a position to make system alterations at very short notice in response to mine plan changes while retaining overall efficiency and cost effectiveness.

By reducing the size of the pump and genset, the mine could save $20,000 per installation.
Developing a Hydraulic Model
The WaterGEMS hydraulic model contains all key items, both as constructed and to be constructed in the field; including all bores, pump packages, storage tanks and Turkey Nests (water storage ponds for dust suppression), pipework and isolation, and other control valves. Operating scenarios have been created in line with system expansion over time (generally quarterly), allowing quick transition between the current and future state. For all chronological steps, a steady state operating scenario and Extended Period Simulation (EPS) are modeled to fully understand system operation. The steady state model allows the team to run specific operating scenarios while the EPS simulates dayto- day operations. Together, they give an accurate representation of the likely range of flows and pressure the system will be subject to.

By fully modeling the future scenarios within the integrated system, the team can optimize pipeline size and pressure ratings alongside pump duties. The model allows the engineers to better understand the flow and pressure combinations over the life of the mine. Often within the mine dewatering industry, systems are designed in an isolation building in a level of conservatism to account for unknown scenarios. By looking at the current and future scenarios and running the EPS in the hydraulic model, the team gains a better understanding of the system operation, answering some of the unknowns and removing an element of conservatism from the design process, reducing the overall cost of the project.

By planning and designing the entire dewatering network in WaterGEMS, Roy Hill does not need to engage external resources when the mine plan changes. The same applies to on-site changes required by mining operations. Having the information immediately available means quick decisions can be made based on a better understanding of the potential implications; therefore not delaying mining operations or putting at risk the dewatering network.

The EPS is developed by creating the on-site system controls within the hydraulic model and simulating the running of the system for a period of time (e.g., a week). By analyzing the results, inefficiencies in the running of the system can be identified, for example, bores turning on and off to fill tanks/turkey nests where-by the existing supplies can be sufficient. By removing inefficiencies in control logic, a reduction in the overall number of bores operating and hence the overall run hours can be achieved.

Applying Energy Costs to the Model
The Darwin scheduler function within WaterGEMS optimizes the running of the system to achieve the lowest energy consumption while maintaining water supply requirements. It uses algorithms to run a high number of operating combinations, to determine the best combination based on preset criteria.

This tool is generally used by Water Utilities to optimize running of pumps based on an energy tariff charged by an Energy Provider (i.e., a rate from a power company that changes throughout a day).

In the Roy Hill case, the Darwin Scheduler is set up to optimize the number of bore pumps running to maintain the water level in the main supply tanks while using the lowest amount of energy possible. The energy calculation is based on the fuel burn rate of the diesel generators at each bore; the lowest energy consumption would lead to the lowest fuel burn rate on-site. The fuel burn rate for each bore is loaded into the model as the energy tariffs; with individual bore sites subject to different energy tariffs based on the Pump/Diesel Generator combination at the bore. As an indication, within the Stage 1 Borefield, there are five different pump sizes and four different generator sizes.

The Darwin Scheduler is set up to output the best three scenarios, which can then be checked against the EPS in the model to prove the minimum tank levels are maintained. It can be run weekly/ monthly/quarterly in line with changing water supply demands on-site or dewatering network expansion.

Sharing the Data
By outputting design and operational information from WaterGEMS, it was easy to provide modeled data for discussion. This enabled the site water team to develop a greater understanding of the hydraulic system; from engineers on-site constructing to operators responsible for the running of the system.

WaterGEMS imports aerial imagery, LIDAR, pit profiles, survey data and other reference data from other internal Roy Hill departments to enable a clear work interface. Similarly, proposed design layouts generated by WaterGEMS were exported in compatible formats for other departments to interpret and approve. While not the main driver, a byproduct of this is the environmental savings associated with an overall reduction in the use of raw materials for construction and fuel burn during operations.

With the majority of the pipelines constructed from HDPE, reducing the pipe diameter can save up to 37% by weight. To date, it is estimated that Roy Hill has used 635 metric tons (mt) of HDPE. Lowering the operating pressure also reduces the pump and associated generator sizing/ load. This leads to a reduction in fuel burn with a 25% drop in generator load reducing fuel burn up to 30%, and dropping a generator size reduces fuel burn by up to 20%.

Accurate system simulations ensures that accidental or unapproved water discharge to the environment can be prevented or controlled with more confidence. The operational optimization from the EPS and Darwin Scheduler targets a 20% reduction in fuel use. A 20% reduction equates to a savings of 3.6 ML (A$2.4 million) of diesel over the mine life (based on current projections); equivalent to 10 million mt of CO2 emissions.

By carefully modeling and designing the entire network, the operating pressure of the system has been reduced to less than 500 kPa. This means that 80% of all pipelines installed by October 2016 on-site are PN6.3 or PN8; 46 km of the 56-km network. It is estimated that this alone has contributed to a savings of A$1.6 million within the first two years of operations. Projecting this forward, it is likely to see savings in the order of $16 million across the mine’s life.

Reduction in equipment size costs is difficult to estimate. Given there were 48 bores installed by October, and reducing pump motor and genset size by one size can save as much as A$20,000 per site, it is estimated to have saved in the order of A$1 million. With anywhere up to 500 bores being installed, this could lead to a total savings of A$10 million.

Having a fully modeled and well-understood system has enabled the Water Management team to avoid costly redesigns and alterations, which in itself can cost upward of A$1 million per project stage (materials and labor). By owning and controlling the hydraulic model, Roy Hill can complete all designs internally. It can take up to three weeks to arrange a consultant to complete works, which given the usual short turnaround timeframes required, is not feasible. Given the team is currently completing five to 10 projects per year and a consultant can cost $10,000-$15,000 per project, there is a saving of as much as A$2 milion over the mine life plus multiple weeks of delays.

This article was adapted from a case study submitted by Bentley Systems. They recognized Roy Hill with a Be Inspired Award in 2016. www.bentley.com

As featured in Womp 2017 Vol 01 - www.womp-int.com