You’ll find other mining/blasting related scientific papers here.
A pattern expansion process demands several steps in order to determine the best blast parameters (to be changed) and avoid production and safety issues. With O-Pitblast this process is very easy and generated automatically. If you’re interested, in this post we’ll explain you how we do it and to get a trial and see by yourself just contact us directly on our website or through firstname.lastname@example.org
So, let’s start from the beginning…
Naturally, mining and explosives engineers need to understand what they are blasting. The first step of a blast process is to try to understand the rock that is meant to be blasted. Identify the exact parameter, in terms of geology and rock structures, that affect the blast and determine the easiness of a rock to break when submitted to an explosive stress, was always a complex process. The practice field experience still plays the major role when the discussion is about the future blast results (Persson, Holmberg, & Lee, 1993). In the next chapters the authors will present a new methodology to, statistically identify this rock factor or rock influence in the process of fragmentation prediction.
The second step on blast planning is the definition of its volume and general dimensions. This should be limited by operation characteristics like blasted volume needed, drilling and explosive supplier capacity Load&Haul availability and production. The general planning department generates a blast polygon with certain characteristics. Holes are distributed inside the polygon in order to provide the best energy or powder factor (kg of explosives per m3/t of rock) distribution. The sequence of diagrams shows the overall process.
Blast Design Process
As mentioned before, based on primordial geology analysis and blast pattern characteristic it’s possible to infer (with a determinate degree of confidence) the size distribution of blast fragments. This first approach allows engineers to assess if their blast will achieve operation needs. Since it depends on a rock factor or rock constant, and the knowing that the crust can be very heterogenic, the prediction model needs to be constantly calibrated in order to provide reliable results. It will be explained afterwards.
The way authors found to calibrate the fragmentation curve was by comparing the predicted fragmentation with the actual one. The last one can be obtained by photo analysis. There are several tools in the market that provide the needed technology to estimate the block size in a muckpile. In this research was used the iPad and iPhone WipWare’s application, which turned to be a very useful and accurate tool.
Fragmentation Analysis with Wipware – Wipfrag
Based on a linear optimization method (download full artice here), the process to calibrate the rock factor/rock influence constant, analyses the predicted and measured X20, X50, X80 and X90 to obtain a perfect match between the two fragmentation curves.
Rock Factor Calibration Process
Results demands and application
On the changes application stage, there is the need to define the fragmentation restrictions. The model will find the best design parameter (optimum global points), such as burden, spacing, stemming, subdrilling, taking into account the restriction defined, to reduce the blast cost (objective function) and this last one based on the fragmentation restrictions calculated by the Kuz-Ram model. The design parameters restrictions, are based on empirical ranges that can be inspired by the investigation results of the researchers mentioned on the background chapter.
Optimized Design Parameters
CASE STUDY AND RESULTS ANALYSIS
The next points detail each step of the optimization process (based on the methodology described before) and presents some of the achieved results.
Initial stage (IS)
The original situation’s benchmarking is a very important point to record. Every field change must be gradual and studied individually to identify potential issues or deep improvements on the process. The initial stage of the blast designs and results were recorded.
IS Design and Results
In terms of design, the analysed operation parameters and fragmentation results are presented on the following tables.
Rock factor calibration
Rock factor (rock blastability influence) parameters are present in the 3rd table (of the previous image). These values were used to predict further designs and pattern expansion plans.
It is possible to observe that the obtained fragmentation from photo analysis is slightly smaller than the prediction. Since Kuz-Ram models retrieve higher values of fragmentation when rock factor is higher (meaning the higher the rock factor the hardest is to break that rock) is understandable that the best fit factor must be smaller.
With the calibrated rock factor, applied on the described on the non-lineal optimization model process, the design parameters, that best fulfills the empirical restrictions and match the fragmentation demands (X90 ≤ 400,00mm), were determined (following table)
This first approach must be treated as any other non-linear problem, considering that this solution can be an optimum local and not the global one. Knowing this, the practical methodology is presented below.
The authors defined a plan to achieve the obtained results in order to avoid too much changes in the terrain and manage the results at every stage. Small changes were applied on each stage and fragmentation results were evaluated. The pattern was expanded until the limits of the desired fragmentation were acceptable. On the next table is possible to analyse the evolution on each stage.
Analysing previous table the authors incremented 10 cm on burden and spacing on each stage. Up to Stage 4 no fragmentation issue, however when the Stage 5 was applied some oversizes were observed (X90 = 481,53mm). The authors took the decision to select the Stage 4 as the “optimum global”.
This blast pattern was used to blast 5 020 000 m3 and, on next the first graphic, is presented the Drill and Blast improvements in terms of holes reduction (were estimated a reduction of 2779 holes applying this methodology). On the second one, the savings for drilling, explosives and accessories represents an overall saving of 826 019,59€.
The cost benefits and the quality of blast results prove by themselves the utility of this kind of numerical approaches on blast pattern definition. With this research is proved that it’s possible to build mathematical models that simulate results for a blast geometric variables. This methodology proved to be very useful in setting strategies for cost reduction and blast optimization. It’s always important to combine mathematic models with field experience to avoid excessive changes and end up with productivity and safety issues.
This kind of approaches can be used not only for pattern expansion but also for patter adjustments (sometime closing the pattern) to fulfill mine to mill demands in terms of blast results.