The following is an extract from D J Brake’s report, Quality Assurance Standards for Mine Ventilation Models and Ventilation Planning, which discusses the application of quality assurance in ventilation planning with particular respect to the ‘basis of design’, as well as the standards for validating a ventilation model.
A ventilation model can, in the right circumstances, be produced in only a day or two. However, the model is not an end in itself; in all cases it is the means to an end, which is to solve a ventilation problem or assess a new or modified ventilation design. In this sense, the model is only as good as the validity of the data on which it has been built and the process that has been used in its development.
In this author’s experience, there are three areas in which the ventilation design process fails because of failure to:
- Understand the scope, battery limits or deliverables of the exercise; recommendations in this regard have already been presented (Brake, 2008)
- Obtain or use the appropriate inputs and assumptions for the study or to understand the correct ventilation operating standards that need to be achieved by the design
- Develop a valid (ie accurate) ventilation model(s).
In addition, the use of a ventilation designer with insufficient skill or experience is a major contributing factor to the above three problems. However, this is not always the case. Often the mine design or operating staff do not understand the impact of certain design or operating practices on the ventilation system. If the wrong questions are asked by the ventilation engineer, or the right questions are not asked (two different situations), then it is possible even for competent persons to arrive at a design that is unsatisfactory, but which may not be recognised until the mine has spent millions of dollars adopting the system.
In this respect, there are two particular quality assurance (QA) issues that ventilation engineers needs to be familiar with. These are:
- How to validate a ventilation model
- How to prepare a basis of design (BOD) for a ventilation design.
The process of validating a ventilation model refers to the QA process, which ensures the model will give reliable predictions of the performance of the ventilation system, either ‘as built’ or at some future point in time. However, a valid ventilation model does not necessarily mean a good or optimised ventilation design. This comes about by careful and comprehensive definition of the inputs to the ventilation BOD as well as the knowledge and experience of the ventilation engineer.
These two facets of the design process will be discussed separately.
Ventilation model validation
It is very high-risk to use a ventilation model that has not been validated (effectively ‘certified as free of material errors’) as an input to decisions that may involve millions of dollars in capital or operating costs, or may either support (or compromise) critical future mine production, or may result in serious occupational health and safety (OH&S) consequences relating to dust, gases, fires, etc.
For this reason, only properly validated ventilation models should be used for major planning exercises and any model validation should have a documented paper trail back to original source documents that show all these measurements or justify all the key assumptions. In other words, the model must be an auditable document.
For a model to meet these criteria, it must correlate with more than just the measured airflows. In fact, a high correlation between measured airflows and modelled airflows can often cover up an invalid model.
The reason for this is that any model can be adjusted, massaged or fudged so that it reflects measured airflows. This is often done with the best of intentions and can be achieved by adjusting friction factors or shock losses or airway dimensions or lengths or regulator settings, etc. In addition, during audits this author has often found compensating errors such as an incorrect fan curve being used with incorrect shock losses but still producing a ‘correlated’ airflow. Therefore, to assume that a good airflow (volume) correlation means that the model is valid can and often does cover up fundamental underlying problems such as: incorrect fan curves (wrong fan type) or blade or variable inlet vanes (VIV) angles or impeller speed, incorrect air density, incorrect friction factors, shock losses or fixed resistances or fixed flows or airway lengths, shapes or dimensions, flow reversals or missing airways or incorrectly modelled regulators or leakage or recirculation paths.
The issue is that a massaged model or one with compensating errors will look correct and may in fact be fully satisfactory for examining minor ventilation changes to the network, ie whilst it is only being used for assessing minor or incremental changes then it may be fit for purpose. There may therefore even be confidence on-site in ‘the ventilation model’. However, if such a model is then used to examine wholesale or major changes to the network (eg reversals of airflow through main airways, new major airways, blocking off existing main airways, new major fans or fan relocations, significant changes in existing fan duties requiring higher or lower pressure/ flow, etc) then it can give very incorrect results that may not be detected until the changes are made, which may be after the expense of thousands or millions of dollars and have potential consequences on production schedules and the like. This author is therefore very reluctant to accept any ventilation model ‘as is’, without any validation process being conducted.
In fact, it is better to have a ventilation model that doesn’t reflect measured airflows quite as well, but has a better overall correlation with all of the above, than one that has been massaged to indicate a good airflow correlation but for which none of the other important correlations have been checked.
Therefore, for a ventilation model to be considered to be valid, it should meet the requirements in Table 1. Where any criteria cannot meet the standard, the risk must be assessed via simple sensitivity analysis to ensure the model will still be ‘fit for purpose’ with the non-compliance and if not, the measurements and/or the model are further examined to bring the criteria into compliance with the standard. Note that getting a good correlation between actual and model values will require using compressible airflow and, in some cases, taking natural ventilation pressure into account.
In Table 1, ‘major’ is undefined but refers to selecting a sufficiently representative sample of high airflow airways dispersed throughout the entire mine. What is sufficient will depend on the size of the mine and the extent of the ventilation circuit. However, as a general rule, the following airflows and differential pressures should be checked.
TABLE 1 – Key validation criteria for a ventilation model[table id=3 /]
- All regulators and circuit (district or booster) fans (as well as primary fans)
- The entry and exit of air into and out of ventilation districts or major splits.
- All mine primary and circuit (district or booster) fans
- All regulators and most other ventilation controls which, if they did not exist, would result in a significant short- circuit between intakes and returns.
In practice, any airflow split that is carrying more than (say) five per cent of the total airflow or more than 4 m/s should probably be checked. In some cases, it can be useful to categorise ventilation measurement stations in a system using the criteria in Table 2.
The above validation criteria is true for all ventilation modelling software.
TABLE 2 – A classification system for ventilation measurement stations[table id=4 /]
Ventilation basis of design
As noted earlier, a ventilation model may be valid in the sense that it accurately predicts how the network will perform, but the ventilation design/strategy itself may nevertheless still be seriously flawed. In this author’s experience, there are two reasons for this:
- The ventilation designer does not have the knowledge or experience to develop a sound design, or;
- The inputs used in the design are incorrect.
Peer review, especially when the peer reviewer is involved in the design from an early stage, is a helpful process to avoid the former of these two problems. Peer review is also a very helpful mentoring tool much like the traditional artisan’s’ approach to developing skill and competency in the apprentice.
However, for experienced ventilation designers, the latter of the above points is of most concern. One reason is that the mine planning engineers or senior management often only have a vague understanding themselves of how some important details of the mine will work, or they have conflicting understandings of these (between design and operations). In many cases, even senior mine planning engineers and operating managers have only a rudimentary understanding of ventilation, and in some cases, quite erroneous understandings. Therefore producing an auditable ventilation BOD, whilst it can often be a lengthy process, is a remarkably effective way to ensure all stakeholders are under the same understanding of how the mine will operate, and what ventilation standards will be achieved in that operation.
This author has had many experiences where the process of producing the ventilation BOD has drawn out critically important disagreements between key persons in the mine design and operations, which has meant that these can be resolved before the design is finalised. In many cases, without a detailed and explicit BOD, the problems would not have been recognised even after the ventilation design had been completed and approved, until operations actually commenced under the new design, when the problems with some of the details would have then become apparent.
A small example of the sorts of key inputs that may not be agreed include:
- How many workplaces need to be ventilated at any time to achieve operational flexibility and targets
- Whether persons will need to be working inbye (downwind) of a production loader what the operating temperature limits are for persons outside air-conditioned cabins.
The actual items to be included in a ventilation BOD will vary with the particular circumstances of the mine or the ventilation design.
There are four key elements to obtaining high quality ‘fit for purpose’ ventilation designs:
- A good understanding of the scope, battery limits, exclusions and deliverables from the work. These need to be critically reviewed before the study commences as sometimes the restriction of the scope of the design may so impact on the design that it renders any conclusions unsound or at least heavily ‘non-optimum’.
- A documented BOD, which ensures all the necessary inputs (factual and assumptions) are agreed, the standards for the resulting ventilation operations are agreed, and an auditable paper trail is established for every key ‘ventilation driver’ within the BOD.
- A validated ventilation model. Again, this must be an easily auditable document that can be clearly referenced back to the BOD or ventilation measurements audits.
- Competent, skilled ventilation engineers in the design development process. Achieving this is a separate matter to the content of this paper, but it is clear that no amount of process or standards, by itself, will result in an optimised ventilation design if the designer does not have the skills or experience to do a high quality job.
Templates have been provided for ventilation model validation and the BOD. These are not prescriptive as they will need to be adjusted for specific circumstances depending on the scope of the work; however, they provide examples of what is required.
Reprinted with the permission of The Australasian Institute of Mining and Metallurgy.
D J BRAKE
Dr D J (Rick) Brake is a chartered practising mining engineer with 30 years experience in underground and open cut operations in senior planning and operating roles in Australia and North America. He graduated with First Class Honours from the University of Queensland, completed a Master of Business Administration from Deakin University in Victoria in 1991, and a PhD in physiology at the School of Public Health at Curtin University in the area of human heat stress in 2002.
He has published extensively in the areas of mine ventilation, refrigeration and cooling, emergency egress and entrapment and human heat stress, and is a Fellow of the Australasian Institute of Mining and Metallurgy, a member of the Mine Ventilation Society of South Africa, and a Member of the Minerals Industry Consultants Association of Australia.
Rick was ventilation superintendent for the four Mount Isa underground mines in the mid-1980s. He was a member of the Editorial Committee for the Fourth International Mine Ventilation Congress in 1988, a member of the editorial and organising committees for the Eighth International Mine Ventilation Congress in 2005 and a member of the editorial and/or organising committees for the 12th and 13th US/North American Mine Ventilation Symposia in 2008 and 2010.
Brake, D J, 2008. A protocol and standard for mine ventilation studies, in Proceedings 12th US/North American Mine Ventilation Symposium 2008 (ed: K G Wallace), pp 3–11 (University of Nevada: Reno).