O&M, Integration of tools and systems Deliverable number 4

Preparing to load PDF file. please wait...

0 of 0
100%
O&M, Integration of tools and systems Deliverable number 4

Transcript Of O&M, Integration of tools and systems Deliverable number 4

Logistic Efficiencies And Naval architecture for Wind Installations with Novel Developments
Project acronym: LEANWIND Grant agreement no 614020 Collaborative project Start date: 01st December 2013 Duration: 4 years
O&M, Integration of tools and systems Work Package 4 –Deliverable number 4.7
Lead Beneficiary: AAU Due date: 30th September 2017 Delivery date: 11th October 2017 Dissemination level: Restricted (RE)
This project has received funding from the European Union’s Seventh Programme for research, technological development and demonstration under grant agreement No. 614020.

LEANWIND deliverable D4.7 - project no. 614020

Disclaimer
The content of the publication herein is the sole responsibility of the authors and does not necessarily represent the views of the European Commission or its services.
While the information contained in the documents is believed to be accurate, the authors(s) or any other participant in the LEANWIND consortium make no warranty of any kind with regard to this material including, but not limited to the implied warranties of merchantability and fitness for a particular purpose.
Neither the LEANWIND Consortium nor any of its members, their officers, employees or agents shall be responsible or liable in negligence or otherwise howsoever in respect of any inaccuracy or omission herein.
Without derogating from the generality of the foregoing neither the LEANWIND Consortium nor any of its members, their officers, employees or agents shall be liable for any direct or indirect or consequential loss or damage caused by or arising from any information advice or inaccuracy or omission herein.

Document Information

Version V1.0-4.0

Date 01/09/2017

Description

Name/ Prepared by

Organis

ation

AAU

J. Nielsen

(AAU)

V5.0

05/09/2017 AAU

J. Nielsen

(AAU)

V6.0

06/10/2017 AAU

J. Nielsen

(AAU)

Reviewed by
J. Sørensen (AAU) I. Sperstad (SINTEF) T. Welte (SINTEF) C. Desmond (UCC) J. Giebhardt (IWES Fraunhofer)

Approved by
Jan Arthur Norbeck (MRTK)

0

LEANWIND deliverable D4.7 - project no. 614020

Authors information: Name Jannie Sønderkær Nielsen John Dalsgaard Sørensen Simon Ambühl Iver Bakken Sperstad Thomas Welte Magne L. Kolstad Lars Magne Nonås Pantelis Anaxagorou Jeroen De Neve Jan Goormachtigh Djamila Ouelhadj Chandra Irawan Blas J. Galván Øyvind Netland
Acknowledgements/Contributions: Name Rasmus Mølgaard Hviid

Organisation AAU AAU AAU SINTEF_ER SINTEF_ER SINTEF_ER SINTEF_Ocean NTUA GeoSea Maintenance GeoSea Maintenance UOPHEC UOPHEC ULPGC NAAS
Organisation KRT

Definitions
Acronym AI ANN API BoP BP CBM CCNN CM DBN DX.X ESN ESP FCTV FEA FFANN FFT FMEA FMECA HMM HSMM I2C+SPI

Description Artificial Intelligence Artificial neural networks Application programming interface Balance of plant Back Propagation Condition-based maintenance Correlation Coefficient Neural Networks Condition monitoring Dynamic Bayesian networks Deliverable number X.X in the LEANWIND project Echo state Networks Engine System Prognosis Fast Crew Transfer Vessel Finite element analysis Feed Forward Artificial Neural Networks Fast-Fourier Transform Failure Modes Effect Analysis Failure Mode Effect and Criticality Analysis Hidden Markov model Hidden Semi-Markov model Inter-Integrated Circuit + Serial Peripheral Interface

1

IDPS LCOE LED LPM MAS MILP MRM MSM MTTF NAN NN NPC O&M OSV PDF PM POD RAMS RMS ROV RPM RPN RUL SCADA SMC SPC SVP SWATH TCI TCM TeCoLog TeCoMan W2W WP

LEANWIND deliverable D4.7 - project no. 614020
Integrated Diagnosis and Prognosis System Levelized cost of energy Light-emitting diode Logistics planning module Moving Averaged Spectral Mixed integer linear problem Maintenance Routing Model Maintenance Scheduling Model Mean time to failure Not a number Neural networks Nominal Power Classification Operation and maintenance Offshore Service Vessel Probability density function Preventive maintenance Probability of detection Reliability, Availability, Maintainability and Safety Root Mean Square Remotely Operated underwater Vehicle Revolutions per minute Risk priority number Remaining useful life Supervisory control and data acquisition Sequential Monte Carlo Statistical Process Control Support vector machine Small Waterplane Area Twin Hull Technical condition indices Technical condition management module Technical condition based logistic planner Technical condition manager Walk to work Work package

2

LEANWIND deliverable D4.7 - project no. 614020
Executive Summary
This deliverable is the final deliverable within work package (WP) 4 ‘Operation and Maintenance strategies’ with the title “O&M; Integration of tools and systems”. Several tools, models, methodologies, concepts (henceforth simply referred to as "solutions") have been designed, developed, applied, or identified with the aim of optimizing O&M for offshore wind turbines. Each solution considers a limited part of the complex optimization problem, and the aim of this deliverable is to describe how the solutions can be used in combination, and how they can be integrated to increase the level of detail and capture effects not included in each solution originally. In summary the deliverable demonstrates that one by integrating different tools can increase the accuracy of the results solve problems that one could not previously solve in an adequate manner using each tool in isolation.
The solutions considered in this deliverable are: 1. Reliability-based design and degradation modelling: application of existing methods
for identification of critical components (e.g. FMECA and RAMS), web-based tool for reliability based design, and description of degradation modelling. 2. Degradation modelling through Fault Diagnosis and RUL prognosis: description of approaches for fault diagnosis and RUL prognosis. 3. O&M access: input data describing O&M access solutions including transfer limits. 4. O&M Strategy model: simulation based tool for strategy optimization regarding e.g. O&M vessel fleet composition and jack-up vessel charter strategy. 5. Risk-based O&M model: tool for optimization of inspection and repair strategy considering probabilistic models for deterioration and inspections. 6. Remote Presence system: prototype of robot on rails to be used inside the nacelle for remote inspections using high definition photos, thermography, and audio recording. 7. IDPS Web Service: web service for condition monitoring and diagnostics. 8. A Dynamic Scheduling Framework: tools for dynamic scheduling and routing of preventive and corrective maintenance tasks. 9. TeCoLog (technical condition based logistic planner): a concept using existing tools for an operational/tactical logistics decision support system for planning and scheduling of O&M activities using technical condition indexing.
Solutions 4, 5, 8. and 9. are decision support tools for O&M planning, the remaining ones can be considered as providing input to the decision support tools. The O&M Strategy model (4.) and the Risk-based O&M model (5.) are strategic decision support tools, to be used for long-term planning, and the Dynamic Scheduling Framework (8.) and TeCoLog (9.) are tools for tactical and operational decisions on shorter time scales.
The two strategic decision support tools have different strengths and weaknesses. For example, the O&M Strategy model is most applicable to decisions relating to maintenance logistics (e.g. vessel fleet composition and jack-up vessel charter strategy), as these aspects are modelled accurately. In contrast, the Risk-based O&M model is most applicable to optimization of the condition monitoring, inspection, and repair strategy, as the effect of this strategy on failure rates is modelled explicitly. The strategy tool considers condition-based maintenance only through high level performance data, whereas the Risk-based O&M model considers vessel strategies only through the costs.
3

LEANWIND deliverable D4.7 - project no. 614020
Due to these differences, the two tools are to some extent complementary. Thus, the tools can be used together to provide more accurate results.
Both the Dynamic Scheduling Framework and TeCoLog can make an optimal maintenance schedule with the objective of reducing the maintenance costs. TeCoLog includes a module for suggesting maintenance tasks based on condition monitoring information, whereas the Dynamic Scheduling Framework need the list of maintenance tasks, which could be provided by the Risk-based O&M model. The dynamic scheduling model considers optimal routing of each vessel, whereas TeCoLog consider which maintenance task to be made when by which vessel considering location and need for equipment.
In addition to describing the principles of how integration and combined use of WP4 tools can be done and add value, this deliverable includes four case studies demonstrating different levels of integration between tools.
Case study 1 concerns approaches for integration of deterioration models and riskbased decisions in the O&M strategy model. These approaches make it possible to include the effect of inspection and repair strategy on the need for repairs and failures in the O&M strategy model. Three approaches are considered: full code integration (accurate but time consuming), "loose" integration through setting up data interfaces (simple and flexible but without correct distribution of repairs and failures in time), and a Bayesian network based approach using data interfacing (computationally efficient and with correct distribution of events in time).
Case study 2 presents a cost-benefit analysis of condition monitoring systems. The analysis was performed with the O&M Strategy model and data from an industrial partner was used. The case study shows how the value of condition monitoring can be estimated based on high level performance data of the condition monitoring system.
Case study 3 presents the architecture and methodology of three models for remaining useful life (RUL) estimation for main bearings. One model proposal is physics based using a multi-sensor condition monitoring system, and two model proposals are data driven using vibration monitoring and temperature monitoring respectively. The vibration based model has been implemented and applies the spectral kurtosis for diagnosis. The model is demonstrated using vibration data, but the data available was not sufficient to validate the approach.
Case study 4 presents a purpose-built simulation-based tool for estimating the costs of repairs/exchanges requiring jack-up vessels. This tool is to be used in combination with the Risk-based O&M model. Also, the effect of jack-up contract and strategy (fix-onfailure or campaign) can be estimated. Alone, the Risk-based O&M model assumes that mobilization costs are paid for each repair, but when used in combination with this tool, mobilization costs can be shared between more repairs. For a case study with blade exchanges, the expected costs per blade exchange could be reduced by up to 44 % using a more detailed cost model considering bundling of repairs.
4
ToolsProjectAauToolDecision Support Tools