Brighter bulk handling

07 Sep 2015
Learning: bulk handling is getting brainier

Learning: bulk handling is getting brainier

Bulk handling technology is learning from past mistakes finds Stevie Knight

Bulk operations aren’t anywhere near as fast or competitive as their box trade cousins and the focus has often been simply on minimising delays and downtime. However, after a while of asking what's wrong with bulk terminal productivity, a few companies have started to ask, ‘what's right?’

Five-ships-a-day bulk ports like Saqr put huge pressures on their machinery as the loads are often hard and abrasive, but there are other issues. The facility handles several different kinds of material, and while some are light, others have densities up to 2 t/m³.

Each load of material comes with unique characteristics: even if it’s from the same batch: “The stuff from the top of the hold has a different consistency from the stuff at the bottom, especially if it’s been at sea for three or four weeks,” says Günter Berthold of Liebherr.

It seems variety, far from being the ‘spice of life’, can be the cause of a drop in productivity at bulk ports.

It’s not easy to manage this kind of diversity, and many crane drivers, in their eagerness to get a good load in the grab, tend to underestimate the weight of the mouthful the crane has taken. When the overload alarm goes off the driver is forced to release before taking another, smaller bite “and he then normally gets over cautious, and underestimates”, says Mr Berthold. An inexperienced driver can keep this iterative process going for as many as 10 moves.

 

Taking a hit

It doesn’t just waste time, the cranes also take a big hit from getting overloaded, though many operators don’t realise how big an impact this has. “The FEM sensors on the crane will tell you what is going on, but you need to know that dropping just one classification point means you have halved the crane’s lifetime,” says Mr Berthold. Further, even the common limestone cargo created lurking issues for Saqr. While fine limestone poses no threat, some contains chips measuring up to 200mm that implode suddenly under the pressure of the closing grabs, shaking the whole machine.

Saqr did what it could: in an effort to stop overloads and to mitigate against the shaking it tried limiting the cranes to smaller, 22m³ volume grabs – but of course this hit productivity.

Enter Liebherr’s SmartGrip, installed on Saqr’s latest LHM 550 mobile harbour crane. The grab had its basic parameters, including closing speed and rope slack, adapted to the rocky limestone material and because it exerts a smooth increase in force rather than biting down in one go, the shaking dramatically decreased.

However, Mr Berthold explains the real trick is that the grab ‘learns’ in a similar way to experienced operators who quickly get a feel for the material they are working on. “Though it’s not actually as fast as the most skilled operators, taking around four loads to get to optimum instead of two cycles, it’s actually a lot more consistent and the grab can keep it up tirelessly,” says Mr Berthold.

All this has allowed Saqr to use a bigger 32m³ grab on the limestone, resulting in a 30% average load increase allowing even inexperienced drivers to work faster and in more comfort.

Given the benefits that accrue from taking humans out of the equation, some high turnover bulk facilities are now looking to full automation. While driverless stacker-reclaimer units are a little easier to design, automating grab ship-unloaders has meant resolving a mountain of variables. This includes not only different materials but also different ships with hatches that vary in size and location, “which are not always perfectly rectangular”, says Bernd Mann of iSAM.

 

Optimum moves

However even these are beginning to be optimised. Iron and coal transhipment specialist Hansaport in Hamburg has been running iSAM’s automation solution on its four, 35 tonne capacity units since 2013 and now the huge ship unloaders at the EMO facility in Rotterdam are following suit. At the time of writing an 85 tonne capacity grab was being wired up. And once again emulating human behaviour is the key to getting it right.

Mr Mann explains that iSAM’s solution has ‘sight’ and ‘touch’ supplied by a 3D scanner and an inertial navigation system – with the brain being a computer wired into the PLC. It works by recognising the main elements, a simplification akin to human processes. So, any holes in the deck are hatches; anything inside is cargo to be moved while structures above the deck are to be avoided. Added to this there’s a no-go area for the grab from the edge of the hatch to the waterside of the ship “as you can’t sanitise a working ship of people who need to keep on walking past”.

However, it does have an important advantage over a human operator, the control system is not only able to calculate the current position, but also the kinetic energy of the grab at any point on its travels. This makes sure that the grab does not collide with the hatch or the unloader structure, not even in case of ‘hard’ emergency stops.

However, interactions inside the hold can be unpredictable.

“It’s relatively easy to get a grab into the material, but it moves much more freely than a container spreader: a lift creates a small landslide from the edges of the hole and this means you can’t say in advance if the grab will come out smoothly, get a ‘kick’ to one side or even start to rotate. You just don’t know till it happens, and it’s a killer for cycle time,” says Mr Mann.

So, iSAM looked at the skilled operators to find out how they stabilise the grab before bringing it back to the hopper and they came up with more than the usual trolley motion: a sideways kick entails stilling with a travel move, and a rotation is stopped by resting the grab for a second on a level surface.

Different programmes are set up to take into account the different cargo characteristics, but the result is impressive and even gives the humans a bit of competition. "We won't be as fast as the top 10% in Hamburg or Rotterdam, but we can consistently deliver a repeatable performance on a par with an experienced operator," he says, adding that there’s no teabreaks or shift changeovers. Further, all this is kept up while still respecting the mechanical limits of the machinery.

So while bulk automation is, says Mr Mann, technically tricky and each facility needs more tailoring than a container terminal, it does promise to save a lot of wear and tear on expensive hardware.

 

Digging for data

Will mining for information become as commonplace as digging for coal? Recent news suggests lucrative returns may not be that far away.

The rise of ‘communicating’ machines means there’s now a huge amount of data coming back to manufacturers and designers who are beginning to put it to good use.

For example, a lot of the information that helped tailor the SmartGrip, says Günter Berthold, actually came from data mining. There are over 600 Liebherr cranes in bulk operations, most with a yearly throughput of over 10,000 cycles, “and while it’s just a slab of information to us, we’ve had some very clever people picking out the patterns”.

iSAM's Mr Mann has already seen a lot of improvements come directly from data mining: there’s been a reduction of annoying false positives in collision prevention and likewise improvements have been made in stacker-reclaimer performance by minimising step-wise ‘bench’ moves.

However, the most intense data mining has taken place on grab unloading. The key issue, says Mr Mann, is that analysing the moves separately leads to misleading results: “You can’t look at each grab cycle individually; you have to see the process for an entire hatch.” He explains: “A digging pattern that looks very promising at the beginning can lead to half-empty grabs later in the process or even make it practically impossible to reach underneath the coaming without the help of bulldozers.”

He goes on to say: “On top of this there was a multitude of different materials and ship sizes which required two years and roughly 1.5m cycles to get sufficient data for analysis.” That meant looking at moves for over 30m tonnes of material.

But it was worth it, resulting in a better than 20% increase in performance.

Links to related companies and recent articles ...

Liebherr Container Cranes

view more