|
() |
|
| 05.PR040 DNV Thruster CM Contract Set Up |
1.1 Classify Equipment
Classify all equipment packages as per OREDA
Review reliability statistics with client including any historical data available
|
|
| |
1.2 Identify failure mechanisms
Review OREDA data, client data and consult with OEM to identify the failure mechanisms typical for the machines being monitored |
|
| |
2.1 Identify existing monitoring
Identify any existing monitoring in place and obtain historical data for reference |
|
| |
2.2 Identify appropriate CM techniques & frequency
Assess CM techniques against p-f interval
Use historic data and / or OEM consultation
Notes: Metron vibration / oil only and minimum specified interval for analysis by DNV of 3 months.
|
|
| |
2.3 Identify monitoring positions & analysis types for vibration
Review design of machines to ensure best transfer of vibration signal from component(s) to be monitored. |
|
| |
2.4 Identify monitoring positions & analysis types for oil
Obtain oil sample points from OEM where available.
Review existing oil sample points and ensure they are appropriate / correct. In particular check that there is sufficient flow of oil when sampling (with machine running) and that instructions are included to ensure any dead legs etc are managed.
|
|
| |
3.1 Create strategy document
Create and update DNV thruster monitoring strategy document with monitoring points, monitoring setup and frequency.
Update equipment monitoring register.
Update standard report template.
REFER TO documents.
|
|
| |
5.1 Establish baseline levels
REFER TO Sub-task Baseline Equipment Levels |
|
| |
6.1 Configure data collectors
Configure data collection for routine monitoring.
Monitoring parameters
Configure limit setpoints for automatic error checking
|
|
| |
6.2 Confirm oil sampling requirements with lab
Confirm requirements with Oil lab for routine sampling |
|
| |
7.1 Update strategy documents and issue to client
Update Strategy document and issue to client |
|
| |
8.1 Obtain feedback from client personnel - routine
Obtain monthly feedback and incorporate in monthly KPI report |
|
| |
8.2 Obtain feedback from client personnel - annual
Review performance at annual contract review |
|
| |
8.3 Review industry developments in monitoring applications
Review developments via conferences, desktop studies, discussion with other operators etc
Include in IMS annual review
|
|
| |
8.4 Review OEM service bulletins
Obtain latest service bulletins and update strategy documents as appropriate
Include in IMS annual review
|
|
| |
8.5 Identify NCs / IOs
Report through existing Metron processes |
|
| 05.PR041 DNV Equipment Baseline and Alarm Level Setup |
1.1 Prepare to acquire baseline data
This can be for:
1. New / overhauled machines, or;
2. Where existing machines have been taken over and no baseline exists, or;
3. Where there are concerns about the veracity of the existing baseline data.
There can often be a ‘wear in’ period for new and overhauled machines. It is recommended that if baseline data is acquired during commissioning of a new or newly overhauled machine due to convenience that this is verified by further measurements after a further period of operation.
|
|
| |
1.2 Acquisition setup
The baseline data collected should reflect the expected routine monitoring requirements or better. For example:
1. The monitoring points to be used should be identical (extra points may be taken for verification)
2. The raw data sample rates and frequency ranges should be as a minimum equivalent to that expected to be used (this can be greater if the ability to post process is available).
The more comprehensive the initial definition of baseline, the greater the likelihood of properly detecting, analysing and tracking the deterioration of the machine
Operating Parameters
Amongst other things, baseline vibration data will be dependent on the type, location and number of transducers available for routine monitoring (e.g. prox probes, accelerometers etc) and the ability to control the machines of interest. Baseline monitoring may include:
* broadband magnitude (displacement, velocity and/or acceleration) - typically 10Hz to 10kHz for vibration;
* time signal and waveform;
* rotational frequency;
* amplitude at once-per-revolution;
* vibration vectors (amplitude and phase);
* frequency spectrum analysis of the vibration signals at steady state;
* run-up/coast-down frequency response data (e.g. Bode plots, waterfall plots, polar plots);
* shaft orbit analysis;
* shaft centreline position.
|
|
| |
1.3 Acquire data
Data shall be acquired in all normal operating conditions.
The operating parameters of the machines should, as a minimum, include the normal operating conditions where routine monitoring will be performed. Extra operating conditions may be acquired for verification and understanding of machine operation.
|
|
| |
2.1 Obtain supplied oil sample
Obtain a sample of clean oil for each machine type from storage as supplied e.g. Totes or drums. |
|
| |
2.2 Obtain post treated oil sample
Obtain a sample of oil as supplied to each machine type following any polishing treatment |
|
| |
2.3 Obtain sample from operating machines
Obtain a sample from each machine in operation |
|
| |
2.4 Send samples to lab for analysis
Clearly mark machine tag and sample date and send to lab for analysis. |
|
| |
3.1 Review baseline data
Data collected during baseline readings should be reviewed for any anomalies. This should be compared to any previously available data or OEM / Industry standards (e.g ISO 10816, ISO 4406 etc). In particular any OEM limits for vibration and oil cleanliness should be reviewed. |
|
| |
3.2 Record current baseline settings
Baseline levels should be identified for each monitoring point (this may include more than simple overall levels e.g. this may include specific narrowband fault frequency levels for the particular machine)
The monitoring configuration register should be created / updated.
REFER to monitoring configuration register template
A database copy of baseline readings should be stored and controlled in the client job folder.
|
|
| |
3.3 Identify and record reference settings
Once baseline readings have been acquired, these should be reviewed against historical data, typical values for the specific type of machine and OEM values to create initial alarm thresholds and expected trend development for remaining useful life calculations - specific bearing life calculations and OREDA reliability data can also be used for this.
These should be updated in the configuration register and reflected in the monitoring report.
Note: These are not the same as client alarm levels and are used to aid on-site analysis and data verification.
|
|
| 05.PR042 DNV Thruster Data collection & Analysis |
1.2 Acquire vibration data
Collect data from routine monitoring points or ad hoc points as required |
|
| |
1.3 Validate vibration data
Verify data quality by performing the following:
If data collector does not have signal verification (bias voltage), confirm the bias voltage is within range.
If data collector does not have value limit checking set, confirm values are within expected range.
Check signal for signs of ski-slopes
Check frequency spectrum against previous readings to confirm frequency content is consistent with expected readings.
Check time waveform for quality of data
If the overall level has increased or reduced by greater than 25%, repeat readings to confirm.
If limited historical data is available, review against 10816 standard limit identified in strategy document
|
|
| |
1.4 Sample oil
Sample oil from defined samples points
Wipe sample point with a clean cloth prior to commence sampling.
Ensure uncontaminated sample equipment and bottles are used
Ensure sample bottle is clearly marked with sample point, date and time of sample (use pre-printed bottle labels wherever possible).
Ensure sufficient oil is drawn from sample point prior to actual sample - ensure that any dead leg volumes are removed and fluid is not contaminated by sample point itself.
|
|
| |
1.5 Validate oil sample
Ensure sample bottle is at least 75% full.
If fluid shows unexpected signs of high volumes of visible water or contaminants, draw a further sample (maintain both samples for analysis)
Ensure sample bottle lid is secure
Ensure sample bottle is clearly marked with sample point, date and time of sample (use pre-printed bottle labels wherever possible).
Pack for return to lab in appropriate cases - ensure cases are marked with the correct returns address (Metron Oil & Gas Ltd)
|
|
| |
1.6 Perform initial on-site analysis
Perform initial on-site review and identify any immediate areas of concern to offshore client personnel - ensure this is included in daily report to client.
|
|
| |
2.1 Consider data quality throughout analysis
Data quality shall be reviewed.
Process conditions shall be reviewed. Note varying and unavoidable operating conditions that may have been prevalent at the time of collection e.g. sea state, vessel loading etc
Syntactic data quality
Syntactic quality is the degree to which data conforms to its specified syntax. For example vibration data is collected from a point in mm/s but is returned in g, data is out of range, or to a different level of accuracy.
Semantic data quality
Semantic quality is the degree to which the data corresponds to what it represents i.e. it must represent something meaningful and real about the thruster application. Issues may include missing data, duplicated data etc.
Pragmatic data quality
Pragmatic quality is the relevancy of the data being collected. It must be timely and useful to the end user. Data must be collected and reported in an appropriate time for an operating response to be implemented and effective. A schedule for routine monitoring is in place, with ad hoc samples made in response to developing faults. This is instigated by the CM Engineer and validated by the Responsible Engineer.
|
|
| |
2.2 Perform detailed analysis of data acquired offshore
Vibration shall be assessed using three factors: vibration magnitude; any significant change in vibration, and; the rate of change of vibration. Reference shall be made to Metron baseline and historical values (which reference historical data, ISO 10816 etc as appropriate). This shall include specific fault frequencies and time waveform data, not simply overall levels. Any unexpected frequencies or other issues shall be reviewed and discussed with the OEM if required.
Oil shall be reviewed for typical causes of contamination and general degradation due to age and use. Particular attention should be paid to the use of ‘clean’ oil without proper pre-treatment / polishing to meet OEM specification.
Any notes from the engineer acquiring the data shall also be considered (anecdotal or otherwise) with suitable weighting.
The analysis of all data collectively will support an indication of confidence in diagnosis. This shall be based on the robustness of analysis and number of indicators.
Remaining useful life shall be estimated based on historical trend development. Review of bearing component life and consultation with the OEM may also benefit.
|
|
| |
2.3 Review and approve report
Review analysed data and verify findings / recommendations.
Any urgent recommendations should be communicated directly - preferably by phone but if necessary by acknowledged email.
|
|
| |
2.5 Review baseline and alarm levels
If there are any concerns as to the integrity of baseline and alarm level data this should be reviewed. This may be as a result of a delayed detection, a failed detection, a false positive or new / overhauled machine. |
|