hello and welcome to a presentation on sonodyne's latest acoustic inertial position reference systems used for dp first let's start by looking at a modern oil field on the right the semi-sub is positioned using acoustics and inertial for dp combined with gps it'll be supported by other vessels in close proximity using the same sensors and maybe simultaneous operations continuing close by new vehicle types are starting to appear such as the wave glider wave glycers are persistent unpowered vehicles maneuvering through the field collecting data from sub-sea assets using acoustics for comms and gps for positioning in addition auvs that we're already familiar with are starting to be used for intervention tasks and most vessels these days operate rovs both of which position sub c using acoustics on the monitoring side increasingly we're being asked to measure pressure temperature strains stresses inclinations from bops and wider monitoring tasks such as the defamation of the seabed as hydrocarbons are extracted there are also new sensor types appearing for subsea monitoring such as the automatic leak detection system this monitors for hydrocarbons in the water column and reports cases using acoustic data links so the analogy we can use is the wireless environment that we're used to at home and in the office is now moving offshore and subsea and whilst this is great for situational awareness from a dp point of view we need to make sure these additional tasks are fully integrated and don't interfere with our operations and that we can free up bandwidth from acoustic positioning to allow for monitoring and communication activities these challenges are combined with the fact that there's no such thing as a perfect position reference gps signals are weak equivalent to a 40 watt light bulb from a distance of 10 000 miles and they're also easy to interfere with a 1 watt jammer being able to destroy commercial gps for a radius of 100 kilometers more likely commercially is interference from unintentional jamming due to faulty electronics poor antenna terminations and interference from communication systems sub c we've already looked at the challenges of integrating comms and control tasks with our dynamic positioning and making sure they're compatible uh other issues sub c is that although we've put great effort into making acoustic positioning systems reliable with things like lusbl availability is still not 100 and we can have outages in our acoustic positioning due to interference due to noise in the water and thruster activity so what we need is a complementary technology that helps us mitigate the risks and challenges with gps and acoustics so this presentation looks at the latest positioning systems used for deep water drilling real world operation and how we address some of the challenges discussed and then the lessons learned so marksman ranger 2 dpins system looks like this in the center we've got the acoustic transceiver deployed through the gate valve used to measure vessel position from transponders deployed on the seabed co-located with the acoustic transceiver is the inertial navigation system in this case on top of the pole transceiver and ins are connected through the vessel up to the bridge where there's a navigation sensor hub dealing with all timing and communications and a pc running either the marksman or the ranger 2 software gps is shown connected but this is only required to set the latitude in the gyro and to see timing there are also outputs using all the major telegram types to the dp desks the part of interest here is what's happening with the inertial navigation and the acoustic positioning so let's start with usbl usbl or ultra short baseline is a basic positioning method that's measuring range horizontal and vertical angle from a transceiver mounted on a vessel to a seabed transponder it's been around for more than 20 years but over time solid has moved from analog to digital processing which adds integrity precision and robustness to both the wideband 2 ranging and the data telemetry in addition 6th generation hardware has additional quality metrics that are used for integration with ins limitation of all usbl systems is that sound travels through water at approximately 1500 meters per second so in deep water it can take three four or maybe five seconds to get an update inertial navigation has complementary characteristics so ins is a self-contained sensor it's measuring earth's rotation and gravity and using that to compute acceleration velocity and then change in position it's very good over the short term and it's very precise but over the long term it drifts so it's said to have complementary characteristics to the acoustic positioning system we can see on the chart here three different sensor types the blue line is our raw ins the red line is our raw usbl and the green line is an integrated ins usbl solution we can see that with time the blue line drifts so after approximately five minutes there's a 14 meter position error this happens with all inertial navigation systems and it's due to the way the errors accumulate in the integration process however if we were to zoom in on that blue line over the short term it's extremely precise and has a very high update rate in contrast the red line which is our usbl position is as accurate after five minutes as it is at the start however in the middle there we've got short term disruption in this case due to high gain dp maneuvers putting aeration noise and thruster wash into the water column and causing position jumps in some cases in excess of five meters what we're trying to achieve is the green line which is our best of both worlds so here we're taking the long-term accuracy of the acoustic system and combining it with the short-term precision and high update rate of our ins and that's what we call an acoustically aided inertial navigation solution so whilst the principles of acoustically aided inertial navigation are the same across sensor types and manufacture the implementations are different the two main types of implementation are either loosely or tightly coupled integration loosely coupled is shown in yellow on the diagram here with the acoustic transceiver measures ranges and bearings to the seabed transponders and passes those to the top side for position computation and it's the usbl position that is then sent to the ins in contrast the red lines are the tightly coupled integration here the transceiver measures ranges and bearings in the same way but passes the raw data to the ins for aiding so tightly coupled has some significant advantages ranges we know we can measure quite accurately we're starting and stopping a clock bearings and vertical angles on the other hand require measurement of phase across the face of the acoustic transceiver this has to be done in quite a noisy environment and is sometimes subject to interference so the bearings get the next amount of weight and the vertical angles get the least amount of weight because of ray bending effects meaning the angle measured at the transceiver face is often different to the true angle so being able to weight these raw measurements separately makes for a more accurate and robust solution the other advantage of tightly coupled is that we don't need a full observation set as each epoch so at one update we can take a range measurement the next update we could take a bearing measurement the next update we could take a horizon a vertical angle measurement and still use those over time to constrain the drift in the ins so let's take a look at how these systems are used in practice and it's worth noting that the solutions are scalable from quite small construction and intervention vessels all the way up to high integrity drill ships so the first application is the ocean intervention 2 operating out in the gulf of mexico in three thousand meters of water now sonadine have fitted uh single ranger two dpins systems to all three of the intervention class vessels uh and in this case operating in three chance in 3000 meters of water with one transponder at a four second update rate now these vessels would typically use just gps for dp deploying a full lusbl array can be quite costly and time consuming considering the nature of the work they do so installing the dpins system has allowed them to improve vessel utilization and dp reliability which in turn reduces downtime so the system installed on the oi2 looks something like this we've got an acoustic transceiver deployed through a gate valve with a load star ins mounted on top of the pole one seabed transponder in use for dp second one deployed as a wet spare now the transceiver and the ins are both powered and communicated with separately which means should the ins fail we can fall back to using the vessels gyros and vius data is passed up through the ship's cabling to the bridge where all i o and timing is dealt with in the nsh and then the position is complete computed on a pc running either ranger or marksman software so this diagram shows the performance of the system operating in 3000 meters of water we're achieving approximately a two meter accuracy which is equivalent to 0. 08 percent of water depth with the usbl system alone this data would be much too noisy for use as a dp reference and we can see here over time that our position accuracy is consistent and and stable at around the two meter mark so in addition to the accuracy achieved we can now bridge through acoustic outages using the ins and we can provide a fast update rate into the dp desk somewhere between one and five hertz to match the gps rather than it being four or five seconds this is what we would have with acoustics in three thousand meters of water so moving on from a lightweight uh construction intervention vessel uh to a semi-sub doing work over activity again in the gulf of mexico class of this vessel was dp3 so we need a dual system in this case it was a dual independent lusbl system and one half of that was upgraded to tightly coupled dpines another advantage of the ins system is it allows us to reduce the update rate to the acoustic transponders and in this case it was reduced from what would normally be three or four seconds down to 15 seconds which extends the battery life of those transponders and the acoustic system on this vessel was a fifth generation it's also compatible with ins but it's one generation old so this is what the system looked like on the left hand side is our dpins system we can tell this because we've got a load style mounted on top of the pole and fewer seabed references in this case three on the right hand side is our traditional lusbl system with no ins and five c bed transponders yellow hardware means it's the fifth generation system and the the benefits were the same we're now operating independent from gnss we've got a one to five hertz update rate going into the dp desk we can bridge through the acoustic outages and we've got fewer seabed transponders on our dpins system the other advantage of this setup is diversity of technology so we've got two fundamentally different positioning methods on each side of our dual independent system which means there's no software single points of failure this is the performance of the ins system installed on one of the poles so in a thousand meters of water we can see that with two transponders our position accuracy is about 0. 5 of a meter and that's over an extended period of time and we're looking at the the green line which is our ins 1d rms figure when we move to three transponders the accuracy is now down at 0.
2 of a meter which is approaching gnss levels of performance with three seabed transponders when we look at the dp desk we can see here that we've got three gps inputs gps one gps2 gps three our ins solution is hpr one and our l usbl solution is hpr2 the first gps was artificially de-weighted by the dp operators because they had lost confidence in it but we can see with the other references they're all achieving an equal weighting in the dp desk which makes for a very reliable and robust dp solution then when we simulate a gnss outage we can continue to operate on two independent and different pme types so the dpi and s solution and the l usbl solution in the absence of gnss and the vessel held station on dp whilst we did various heading changes and box maneuvers so now we move on uh in scale again to a drill ship this is vantage's tungsten explorer came out of the shipyard in september 2013 and is now operating in myanmar in about a thousand meters of water so this is a state state-of-the-art acoustic inertial position reference world-class because it's got two acoustic inertial systems one on each pole again they can extend transponder battery life by reducing update rate and it's fitted with two of the latest sixth generation wide band transceivers and is using six seabed transponders three per system so this is what their system looks like ins is installed on both poles and three seabed transfer transponders uh associated with with each system so again they're getting the fast update rates into the dp they're bridging through acoustic outages uh they're potentially saving on seabed transponders compared to a traditional l usbl which would have five on each side and both of these systems will gracefully fall back to lu spl should there be an issue with the inertial navigation performance we're achieving here is very similar to what we had on the semi-sub with three transponders we're getting gnss levels of performance independent of water depth and when you zoom in you can see that the track of the inertial solution follows very closely with the gnss so our marksman and ranger 2 dpi and s systems have been installed on many vessels now worldwide this next section is just a short look at some of the lessons we've learned so first of all for drill ships work over vessels or vessels involved in diving operations we should not be using a single seabed transponder for our dpins solution and this slide shows why on a vessel operating out in the gulf of mexico in 2800 meters of water we can see that over time our usbl position goes from quite stable to quite noisy for an extended period of time so that noisy usbl data found its way through into our ins solution and the green line in the bottom chart shows our ins error increasing and becoming unstable that resulted in a the dp ins solution being rejected from the desk so the sudden and then long-term degradation in usbl was because of a changing heading uh and the fact that the transponder was obscured from view of the transceiver due to the riser in the water column so to overcome this we need to be operating with three or more transponders when we're doing drilling diver operations or work over operations the next lesson learned is the fact that we need to use a tightly coupled integration of our acoustics and ins the two charts show the same data processed twice with the same vertical scale so the loosely coupled solution has an accuracy of around five or six meters this is using two transponders is a poorly configured system when we move to the tightly coupled solution the accuracy is improved to about two meters using exactly the same data so tightly coupled is more accurate and is more likely to be accepted in the the dp desk and in fact this loosely coupled data is not suitable for dp another thing we get asked about a lot is how many transponders should we use in our dpi s solution and this slide gives some indication of an example uh in the gulf of mexico operating in 2800 meters of water so we can see on the graph number of transponders against a position error and with one transponder our position error is about 2. 3 meters with 2 is 0. 6 with 3 it's 0.
2 with 4 it's 0. 18 so we can see a law of diminishing returns there and the optimum number is between 3 and 4. then uh when we look at ins installation we've learned that we need to co-locate our acoustic transceiver and inertial navigation system this is a test done on a vessel with stem tubes where we've got an ins co-located with your content the acoustic transceiver in a configuration we call gyro usbl and a second ins installed on the bridge now the dpos did a test by pointing the thrusters at the pole at about 70 power to see how much the the pole would move and we can see in the charts that in roll the pole starts to resonate so this resonating motion is sensed by the ins on the pole it's not sensed by the ins on the bridge in actual fact resonating about the true value is not too much of a problem the ins can smooth that out quite well but in pitch we can see that we've got a significant bias as the pole is pushed aft now that bias is measured by the ins on the pole it's not measured by the ins on the bridge so if we were dependent on an ins installed on a bridge or in a gyro room we'd see a 0.
3 degree error in pitch which in 3000 meters of water relates to a 15 meter position error so that's unlikely to be usable for dp purposes so the lesson learned is we must co-locate our inertial navigation system with our acoustic transceiver if the pole is rigid the load star can go on top of the pole if it's a stem tube over the side pole or there's doubt about the rigidity then we need to mount the ins with the acoustic transceiver on the bottom of the pole in a configuration that we call gyro usbl this is what that test looked like in the dp desk we can see we're tracking using our ins solution the thrusters were pointed aft towards the pole and we continue to navigate and hold station during that test and that's thanks to the instrument on the right which is our gyro usbl so we've gone through a lot of data a lot of detail in time series plots which can be quite complicated to understand but fundamentally there are two things to remember if we want accurate high integrity acoustic inertial navigation a we need to co-locate our ins with our acoustic transceiver considering using a gyro usbl and b we need a tightly coupled integration so we can take those ranges and bearings weight them differently thank you for watching please email any questions to sales at sonodyne.
