Report
Teaser, summary, work performed and final results
Periodic Reporting for period 1 - Foresight (Foresight: Autonomous machine monitoring and prognostics system for the Oil and Gas and Maritime sectors)
Teaser
Today most vessels and offshore platforms still implement Time-Based Maintenance (TBM) strategy, which consists in periodic replacement of machinery parts regardless of their conditions. The main alternative solution to TBM is Condition Based Maintenance (CBM), i.e. fixing...
Summary
Today most vessels and offshore platforms still implement Time-Based Maintenance (TBM) strategy, which consists in periodic replacement of machinery parts regardless of their conditions. The main alternative solution to TBM is Condition Based Maintenance (CBM), i.e. fixing machinery just in time prior to functional failure. This approach, based on machinery diagnostics performed on data collected by sensors estimating the machinery health status, beats TBM in cost-efficiency; yet TMB is still the defacto maintance strategy in the maritime sector. The Maritime market demands a maintenance monitoring system able to provide a cost-efficient real-time machine monitoring technology able to manage different machinery while minimising connectivity bandwidth needs and expert involvement performing data analysis.
During our Phase 1 Feasibility Study, we analysed the technical, practical and economic feasibility of Foresight for the Maritime industry. Foresight is the ideal solution, it is agnostic of the type of machinery and type of data input, works without constant data connection to the shore and performs machinery prognostics without human interaction. Foresight can be applied to almost any equipment onboard a vessel. Foresight combines a novel sensor, a bridge to connect 3rd party sensors and a network gateway. Our sensors and bridges have embedded a set of complex algorithms digitalising and processing the signals from the machinery. Foresight calculates for each individual component a set of Condition Indicators (CI) using mathematical algorithms based on physical principles that capture the fault signatures of known components from the raw data. Each CI is linked to a different failure mode, which allows to directly look for the problem rather than analysing all data generated by the sensors. Foresight has been designed to work autonomously without supervision of the maintenance crews of data analysis experts to process the data.
In the design of Foresight, we have also considered the interests of classification agencies and insurance companies, who have great interests on accessing to the information regarding maintenance operations. Foresight proposes a win-win-win business model where shipowners, classification agencies and insurance companies directly benefit from data sharing incentives.
Work performed
During our Phase 1 Feasibility Study, we analysed the technical, practical and economic feasibility of Foresight for the Maritime industry:
Product development feasibility study
During the Q1 2019 Machine Prognostics (MP) manufactured a batch of 50 sensors at its manufacturing warehouse in Oslo (Norway) with the support of its current manufacturing partner. This sensors batch was used to investigate and demonstrate Foresight technical feasibility by introducing faults in small and large bearings, planetary gearboxes and shafts. The only issues found in the first batch were due to ground noises of the power supply. The laboratory testing demonstrated the good performance of our CIs, which detected all the failures intentionally introduced into the machinery.
To confirm the practical feasibility of the Foresight system with actual machinery, MP together with a major crane OEMs component supplier installed Foresight on a 1 MW High-Pressure Unit used on cranes. During the tests, we shown that: 1) Foresight could be used without data analysis expertise; 2) the information collected of the machinery health data was complete; 3) the installation process was simple and effective, the OEM team was able to complete it without integration problems.
In parallel to these laboratory tests, we performed some demonstrations to different potential end-users, as seen in Figures 1 and 4, in machinery on board vessels during their stay in harbours.
Regulatory feasibility study
Thanks to our collaboration with DNV GL, we started the analysis of the certification under the DNVGL-CP-0484 rule, which had been re-edited in February 2019 to allow the class approval of digitally based service supplier scheme, valid for Oil & Gas and shipping. This is a major milestone for Foresight as it opens the door for class processes without an expert performing all the data processing.
Commercialisation readiness and communication feasibility study
The tests performed confirmed the strategy to start the sales on cranes and progressively grow into other machinery installed on board vessels and platforms. The meetings with the potential customers confirmed this approach, as they have not found a solution to properly ensure the constant maintenance monitoring of cranes, which are critical for many operations in the Oil and Gas (O&G) segment. In order to quantify our competitive advantages, we performed an in-depth analysis of the sector, the main competitors, their solutions and competitive advantages. As the first solution that offers an autonomous machine monitoring solution, Foresight demonstrated manyUnique Selling Points over the competitor offerings.
Financial feasibility study and Detailed business plan
The results obtained by MP during the tests with the major OEM have attracted the interest of potential end-users and other OEMs. The companies that accepted to participate in the pilots have not only confirmed their interest in piloting the solution during the future Accelerator project but also the intention to purchase the system once it is available. In order to ensure the financial feasibility, we met with different investors. In those meetings, investors have seen a great potential in Foresight, but the financial risk for them is still too high. A €70-million-AUM Venture Capital firm has confirmed their interest in participating in the investment round associated to the Accelerator project.
Final results
Aiming to keep improving the performance of Foresight, we continued our R&D work and we obtained the following improvements since the Phase 1 application, in parallel to the tests performed at the laboratory and with OEMs:
• Improvements over the data connection to handle the data transfer over high latency networks.
• Recoding the HI algorithms improving local processing of the data.
• Complete new design of the Printed Circuit Board assemblies and integrated circuit board for sensor.
• New design of the printed circuit board of the bridge for 3rd party sensor integration.
• Modified the sensor faraday cage allowing 3D printing, enhancing sensors resonance frequencies and halving manufacturing costs.
During the Feasibility Study, we were able to obtain feedback from the potential customers to estimate how Foresight will help shipowners to save up to 25% of their current Operative Expenditures, in average, if installed in all the critical machinery (crane, main and auxiliary engine, electrical generator or main air compressor): Foresight:
1) dramatically reducing the installation costs;
2) cut the replacement of healthy expensive parts;
3) reduce , potentially eliminates all unscheduled downtime;
4) eliminates the need of data analysis experts onboard;
5) lowers the cost of the periodic reclassification works; and,
6) obtains savings in Hull & Machinery premiums.
Website & more info
More info: https://www.machineprognostics.no.