Wärtsilä – virtual fuel flow

by 24, Jan, 2021Innovation, Maritime 4.00 comments

Wärtsilä has developed a modelling approach that helps to measure fuel flow without a flow meter. This is valuable for shipowners, operators and technical managers who are looking for transparency without costly investments.

Shipping companies are increasingly having to make difficult decisions to increase efficiency. Most companies are still doing this based on noon reports.

It is difficult to achieve much of an understanding regarding the relationship between fuel consumption and speed using only data from noon reports. Until now, the only alternative option has been to install a flow meter, which gathers data about how much fuel is flowing to the engine. This comes at a high cost, not just in its installation and maintenance, but in managing the data streams it generates and establishing how it can be utilised in the decision-making process.

Wärtsilä, the Finnish manufacturer and developer of smart maritime technology, understood long ago that, in order to help shipowners to optimise fleet performance, it had to develop an alternative approach.: by fusing manual reports and automatically gathered data from the onboard navigational equipment a vessel’s fuel flow can be modelled.

The techniques behind the model are a combination of naval architecture and the latest statistical methods. This allows the model to be fine-tuned using operational data, making it self-adapting and vessel specific.

The model begins with a generic model based on vessel particulars. The model quickly learns from operational data, producing reliable results after one laden and one ballast ocean leg. As the performance of the vessel changes over time, so does the model. This allows the user to quantify effects like hull fouling.

The model can be used in multiple way to improve operational performance. This breadth is highlighted in the software modules developed by Wartsila, covering aspects ranging from voyage and fuel optimisation to improving safety or monitoring emissions.

Another benefit of the system is that it enables companies to easily compare performance across their entire fleet, even if the vessels have different equipment installed on them.

The model can be used to identify differences between very similar ships. This enables investigations into optimal paint selection or maintenance schedule.

This modelling technique can only be used for ships where there is a direct relationship between speed through water and fuel consumption. It won’t work if you have a complex propulsion system, where there is not a constant relationship between power and speed.

Gathering data

Noon reports are a key input for the model. Gaps and errors in reporting are minimised by the model’s ability to predicts the reported consumption before it is even submitted by the crew. Any errors in noon reporting can easily be discarded

“There should not be any additional work for the crew in gathering data to feed the system. If there was, that might make the system harder to implement. The work of seafarers is already complex enough,” says Carlos Losada, solutions manager at Wärtsilä Voyage.

An aggregated fuel consumption measurement is obtained via Wärtsilä SmartLog, an application designed to collect all the necessary data. SmartLog is readily available as part of Wärtsilä Fleet Operations Solution.

Wärtsilä collects data based on navigational equipment and noon reports to ensure that information is collected from already existing processes. This avoids adding additional work for the crew, which would be a roadblock when it comes to scalability, Carlos Losada, solution manager at Wärtsilä Voyage explained.

Comparing modelling techniques

Mr Losada shows three different examples of speed-fuel data. One solely based on noon report data, a second one based on a high-end auto-logging system, and a third one based on the Wartsila Virtual Flow Meter approach.

The noon report data alone just shows a few points with similar speeds and does not easily fit into a curve. Additionally, the accuracy of the reports adds uncertainty to the analysis.

The data from the auto-logging system shows an instance where the fuel flow readings became static. This shows potential challenges in maintaining the quality of the data and in processing it.

The Wärtsilä model output is similar to the output using the flow meter data but does not require any flow meter. Simplifying the setup and allowing the users to focus in making the right decision based on the results of the analysis.

So, the Wärtsilä model gives a similar result to the flow meter data-based model, but for much lower cost. The flow meters themselves have costs of installation, calibration and maintenance, and the data streams they generate need to be processed, filtered, analysed and allocated to the right part of the voyage. “You cannot use this data without going through some heavy processing of it,” he says. You also have no back-up if the flow meter fails.

“It is not necessarily true that the more we invest in data accuracy, the more we reduce the costs of running our business,” he says.

Using the technology

Typical goals for shipping companies are to understand how much fouling is affecting performance, better plan voyages to reduce fuel consumption, ensure fuel consumption is in accordance with the charter party (contract with customers) and understand the total cost of running ships.

“Each of those areas requires a different level of granularity [of data]”, Mr Losada says.

There are the many ways in which the model can be used to analyse and improve operational efficiency of a fleet. Some examples of these include optimising Charterparty descriptions, evaluating the efficiency of past voyages or improving the accuracy of commercial voyage planning. The solution is scalable, working immediately with standard equipment and processes.

By digitalising the process and making it a bit smarter, Wärtsilä takes the burden off the crew and the office. It helps to avoid ‘death by spreadsheet’. If it’s something you need to monitor periodically, there’s a dashboard for that. If it’s something that you could do with interfering there and then, there’s a notification on your tablet or phone.

Auto-updating

The model does not need to go through any specific calibration process. It is “self-calibrating”, in the sense that it is being continually improved over time.

The system becomes very hard to fool. For someone to be able to tweak the system, for example for it to believe the vessel had received more fuel than had been loaded, they would need to know what fuel consumption the computer system is expecting, then provide something slightly different to that. “You’d have to be a naval architect with a lot of free time,” Mr Losada says.

The approach has the added benefit of reducing the complexity and cost that can come with recording, storing and processing vast quantities of data. Wärtsilä’s aim with FOS is to give much of the value of a total fleet performance management system – one that relies on collecting and transmitting data continuously from every sensor of the ship – for a tiny fraction of the cost. There may be shipowners that prefer the comprehensive data solution, but for many a light, cost-effective installation that yields good results without requiring an army of analysts will be the most attractive option. There is a point at which more investment in vessel performance only leads to relatively small marginal savings, making it not commercially viable. Wärtsilä Voyage aims at the sweet spot between the level of such an investment and the associated commercial benefits.

There are always obstacles to progress, and digitalisation at sea is no exception. But Wärtsilä Voyage firmly believes that by maximising value, minimising cost and keeping complexity at bay, system designers can help shipowners to make sure these obstacles are speedbumps, not roadblocks.

This article is based on a Digital Ship webinar with Wärtsilä. Watch the webinar here https://youtu.be/3kIGiAjOgdQ 

A more technical webinar explaining how the system works can be viewed here https://www.wartsila.com/insights/webinar/virtual-fuel-flow-meter

Source

January 2021
M T W T F S S
 123
45678910
11121314151617
18192021222324
25262728293031