My WODCON 2019 Presentation: Launching Robotic Dredging

Me, presenting my WODCON 2019 contribution.
Me, presenting my WODCON 2019 contribution.

Yesterday, I gave my presentation at the WODCON 2019 in Shanghai1. The WODCON is the triannual world dredging conference, were everybody in the dredging industry meets and exchanges knowledge and ideas. Just as I mentioned in my New Year’s post, I sometimes like to delve into some old archives, get inspiration and hatch new ideas2. So did I for this presentation.

Overview of the ‘Ketelmeer’ (Credit: Google Maps).
Overview of the ‘Ketelmeer’ (Credit: Google Maps).

A seminal dredging project concerning environmental dredging is the ‘Ketelmeer’ clean up dredging project, resulting in the creation of the contaminated sediment storage depot in the artificial island ‘IJsseloog’3. As careful removal of the contaminated sediment required novel dredging techniques, the government challenged the dredging industry to test four innovative concepts. The results were evaluated by the institute now called ‘Deltares’ and published in a report4. The original conclusion of the report was, that the auger dredge was the best in reducing the turbidity. Later, the bigger auger dredge ‘HAM291’ was constructed and used to actually clean up the lake. With the knowledge and the experience of the auger we also developed a range of auger attachments for our DOP pumps5.

Traditional auger attachment for a DOP pump excavator combination.
Traditional auger attachment for a DOP pump excavator combination.

Reading the Ketelmeer report again, it occurred to me, that one parameter had not been properly accounted for: the size of the dredge. The auger dredge was by far the smallest dredge in the game. With a weighed scoring method, the dredges were also compared in size and installed power. The reasoning is that a bigger dredge has more interaction with its environment. Naturally, the environment gets more disturbed and turbidity levels should be higher for a bigger dredge. And the data was there to support this hypothesis. Smaller is better! Still, this does not undermine the initial results of the concept, as that was evaluated for turbidity per cubic meter. The bigger dredges also delivered more production. But when comparing dredges of the same concept might the smaller ones will perform better on turbidity. And this is in accordance with our experience. Every project where we’ve supplied these auger dredge units, the contractor and the client where surprised and happy about the achieved turbidity levels. Now, we know why: smaller is less turbidity.

The next step in performance might be reached by further decreasing the size of the dredge. However, the DOP is already as small as it is for a viable application on an excavator. The conclusion is to have an auger operating directly on the bottom: an unmanned submarine dredging machine!

Possible general arrangement of a robotic dredge submarine.
Possible general arrangement of a robotic dredge submarine.

This machine should navigate by itself and self-supporting. The wear parts of the auger should be exchanged by itself and solar panels can provide extra energy for extended missions. It has only a small hopper and discharge should also be done quick and automatically. An unmanned barge or even a dump truck trailer at the shore of the waterway can be replaced at longer intervals. Obstacles and other tricky spots can be alerted to a human supervisor for later intervention. One machine alone does not have an impressive production. The real power is in applying them in numbers. As we are standing on the brink of a revolution in robotics and artificial intelligence, this scenario may be not as farfetched as from your first impression. Imagine a whole school of these mechanical fishes cleaning up your waterway, while you sleep…

Working method of robotic dredge submarine (Credit: Judith Korver).
Working method of robotic dredge submarine (Credit: Judith Korver).

References

  1. WODCON, Damen
  2. New Year’s post 2019, Discover Dredging
  3. Ketelmeer project, Wikipedia
  4. Rapportage baggerproeven Ketelmeer. RIZA Rapport, 97.023, ISBN 9036950708
  5. DOP Pumps, Damen

See also

Novel Density Measurement By A Little Dredging Engineer

 

Design of an experiment to test a novel density measurement.
Design of an experiment to test a novel density measurement.

Innovative ideas need an open mind, not hindered by trodden paths of thought that limit one’s ability to be creative. And whose mind is open and flexible enough to trust such a task? Children! In order to keep me sharp, I sometimes discuss tricky problems at the dinner table with my daughter. Lately I presented her with the problem of density measurement in a slurry pipeline. There is a well-known technology: nuclear radiation dissipation detection. But there are already two words in that concept that we are asked to replace with a safer and more environmentally friendly alternative1. Her solution was to measure how much particles would stick on a glue lathered rod held in the mixture flow. During one of my daughter’s holidays from school, we took the opportunity to test her novel concept.

Preparations for the density measurement experiment.
Preparations for the density measurement experiment.

If this is going to be a novel measuring technique, we should prove, that there is a relation between the amount of particles sticking to the glue and the density of the mixture. This had to be tested in a controlled experimental environment: a bucket. We filled it with a known amount of water and by adding sand to it, we gradually increased the density. There were several types of glue to be tested, which we applied to short pieces of PVC tube. Glues we considered were: hobby glue, mounting kit, wall paper glue, contact adhesive and duct tape. There was one reference tube with no glue. We kept the samples in the mixture for 30 seconds, while vigorously stirring the slurry with a mixing drill.

Density measurement in progress.
Density measurement in progress.

Before and after the sample was dipped in the slurry, we measured the mass of the sample on a scale. The difference in mass should represent the amount of particles that stuck to the glue. These differences were recorded and plotted in a graph. Surprise! There is a distinctive trend in the data points that would support our hypothesis on the relation between mixture density and particle capture.

Results of the density measurement.
Results of the density measurement.

Some remarks on these results. There is a wide variation in trend lines. Some of the samples became lighter than they started. Probably, because they dissolved in the mixture. The measurements for the 600 and 1200 grams of diluted sand is notable. This is probably related to an observer variability. The strongest relation between mixture density and grain capture is with contact adhesive and duct tape. Interestingly, the best predictor for the average capture over the glue types was the reference tube. Probably, the water clinging to the tube already contained enough particles to result in a measurable signal.

It was fun to do this exercise and the results are interesting. But, I doubt we can make a viable product out of this concept. In the dredging industry, we prefer a non-invasive technology, the sample rod would be knocked out in seconds. Several manufacturers are already developing alternatives2,3. In our ¡VAMOS! project, we tested one concept, that seems very promising and you will definitely hear more about it in due time. It is time to say good bye to the nuclear density sensor and adopt technologies that are future proof. Our children will be grateful if we leave them a better world.

Non-radioactive density measurement (Credit: ¡VAMOS!).
Non-radioactive density measurement (Credit: ¡VAMOS!).

References

  1. Regeling bekendmaking rechtvaardiging gebruik van ioniserende straling, Staatcourant NL
  2. Magnetic resonance multiphase flowmeters, Krohne
  3. ITS exhibiting at Europort and CEDA Dredging Days 2017, ITS

See also

Production management, Damen

Discussion at LinkedIn post

Sensing My Audience, Feedback On How To Drive DOP Pumps

DOP pump and power pack, ready for commissioning.

This website provides me a platform to share my knowledge about dredging as far as my experience reaches. Sometimes, I do training sessions with colleagues or clients. They give me the opportunity, as an academically educated dredger, to experience how our figments of imagination work out in real life. People from different countries and lifestyle share their experience with our products. I really enjoy those sessions, they give me feedback on my normal work. Some of the topics on this website are directly related to the topics I present on those sessions. One of the recurring issues might not be that obvious at first hand. So, this post results from this feedback, aptly concerning the feedback loop in a load sensing circuit in a hydraulic drive system, specifically for a DOP.

The most common arrangement for the traditional DOP drive is to have a separate (preferably co-purchased) hydraulic power pack1. The power pack we supply is specifically designed to drive the dredge pump in the DOP. This requires the extra investment of a hydraulic power pack. Whereas most hydraulic cranes and excavators, that can handle the DOP pump by weight, will also be able to divert some hydraulic power to an auxiliary connection or so called Power Take-Off or Aux PTO. Especially when the excavator is working with a standard suction head, not much motion and forces are required for other functions. So, naturally clients tend to propose to use their DOP driven on the hydraulic PTO. This could be possible, but only under very strict conditions. And here comes the advice after commissioning hundreds of those units: ‘Don’t use the load sensing feature of the hydraulic PTO!’ Why? Let’s study the hydraulic diagrams.

Hydraulic diagram DOP pump and power pack.

The standard arrangement of a DOP is quite straightforward. As long as you do connect them in the right fashion. By controlling the angle of the swash plate, you control the delivered flow from the hydraulic pump and the rotational speed changes accordingly. Any load changes at the dredge pump result in pressure changes in the circuit, without changing the pump speed.

Typically, excavators do have a different power characteristic. When freely raking the stick, they encounter little resistance at preferably a high speed to reduce cycle time. When they hit the ground, the resistance or load increases and they typically lower the flow to maximize power delivery. They do this by connecting the pump line to a control line that influences the angle of the swash plate. Effectively, this results in a feedback control loop.

Infographic feedback control loop in a load sensing circuit.

And here comes the trick. Feedback control loops are designed to be stable. But the varying load at the dredge pump, in combination with the masses in the hydraulic lines, the springs in the rubber hoses and the damping of the flow losses, make the sensitivity of the control loop irrelevant. The signal from the dredge pump load is lost and becomes out of sync with the required action at the LS port of the pump. Any random static will make the swash plate rattle and usually in an instant, the drive stops. Other than maybe a blocked dredge line, no harm is done and service calls to our colleagues2 will resolve the issue: ‘modern excavators will allow you to turn of the Load Sensing feature’. Or, purchase said dedicated hydraulic power pack.

Damen hydraulic power pack mounted on a crane.

I would like to acknowledge Wim Roeterdink for his advice and review of this post. He is an expert on hydraulics engineering and he is always available for assistance in your particular hydraulics problem.

References

  1. Hydraulic power packs, Damen
  2. Service, Damen

See also

Parker hydraulic pumps