PREDICITIVE MAINTENANCE (CBPMM)

 Condition Based Predictive Maintenance Monitoring (CBPMM)

Condition Based Predictive Maintenance Monitoring (CBPMM) is a strategy that uses equipment data to detect changes in the machine's behaviour, identify potential failures, and then schedule maintenance before a breakdown occurs. CBPMM aims to minimise unplanned downtime, increase equipment efficiency, and improve operational reliability. This white paper will discuss the various methods used in CBPMM, including Vibration Monitoring and Analysis, Thermal Imaging, and Lube Oil Analysis, as well as the equipment, procedures, and record-keeping necessary to implement this maintenance strategy.

Introduction

Condition Based Predictive Maintenance Monitoring (CBPMM) is a strategy that monitors equipment and its surrounding environment to detect changes in the machine's behaviour, identify potential failures, and then schedule maintenance before a breakdown occurs. This strategy uses various methods, such as Vibration Monitoring and Analysis, Thermal Imaging, and Lube Oil Analysis, to detect problems before they cause significant damage.

1.1 What is Condition-Based Predictive Maintenance Monitoring and Purpose

CBPMM is a maintenance strategy that aims to reduce unplanned downtime, increase equipment efficiency, and improve operational reliability. The goal is to schedule maintenance before equipment failure occurs based on the analysis of data collected from the equipment and its environment. By detecting potential problems early, maintenance can be scheduled at the most convenient time, thus reducing the impact of downtime on the operation.

1.2 What is Vibration Monitoring and Analysis?

Vibration Monitoring and Analysis is a technique used to detect changes in the vibration signature of machinery, which can indicate potential problems. Vibration analysis is typically done by monitoring the machine's vibration frequency, amplitude, and direction. These measurements can help identify problems such as misalignment, unbalance, bearing wear, and other issues. The analysis is done using specialised equipment, such as the CMT VIB. MONITOR A4900, which is used to collect and analyse vibration data.

1.3 What is Thermal Imaging?

Thermal Imaging is a technique used to detect changes in temperature, which can indicate potential problems. Thermal Imaging is done by using an infrared camera, such as the FLIR E6, to detect temperature changes. These changes can indicate loose connections, electrical faults, and other problems. The analysis is done by comparing the temperature readings from different points on the equipment to identify potential problems.

1.4 What is Lube Oil Analysis?

Lube Oil Analysis is a technique used to detect changes in the oil composition, which can indicate potential problems. The analysis is typically done by collecting oil samples and analysing them in a laboratory like Shore Laboratory. The analysis can identify contamination, wear particles, and other problems. Onboard Lube Oil Analysis is also possible with Testing Equipment that can provide quick results. Cylinder Scrape Analysis is another method for analysing the piston ring and cylinder condition.

 

2. Equipment

2.1 Equipment for Vibration Monitoring (e.g.CMT VIB. MONITOR A4900)

The CMT VIB. MONITOR A4900 is a specialised device used for Vibration Monitoring and Analysis. It can collect and analyse vibration data from machinery and provide diagnostic information to help identify potential problems.

2.2 Equipment for Thermal Imaging (e.g.FLIR E6)

The FLIR E6 is a specialised device used for Thermal Imaging. It can detect temperature changes and provide diagnostic information to help identify potential problems.

2.3 Equipment for Lube oil Analysis

Different methods of Lube oil Analysis require different equipment, as described below:

2.3.1 Analysis by Shore Laboratory - Landing Samples

Shore Laboratory is a specialised laboratory for analysing Lube Oil Samples collected from the equipment. The laboratory provides detailed reports on the oil's composition

2.3.2 Onboard Lube Oil Analysis -Testing Equipment

Onboard lube oil analysis is a technique that allows maintenance teams to monitor the condition of lubricants in real time without having to send samples to a shore laboratory for analysis. This technique involves using specialised testing equipment that can be installed on the machinery, allowing for continuous monitoring.

The testing equipment typically includes a sensor that measures the acidity or alkalinity of the lubricant, as well as its viscosity and water content. The data collected by the sensor is analysed to determine the condition of the lubricant and whether any corrective action is needed.

Onboard lube oil analysis is a valuable tool for maintenance teams because it provides them with immediate feedback on the condition of the lubricant, allowing them to take corrective action quickly if necessary. This can help prevent severe equipment damage and reduce the likelihood of unexpected downtime.

2.3.3 Cylinder Scrape down Analysis - Testing Equipment

Cylinder scrape-down analysis is a technique that is used to monitor the condition of the cylinders in large diesel engines. This technique involves removing a small amount of material from the surface of the cylinder and analysing it under a microscope to determine the condition of the cylinder walls.

The testing equipment used for cylinder scrape-down analysis typically includes a specialised microscope capable of magnifying the surface of the cylinder to a high degree. The technician will take a sample of the material using a scraper tool and place it under the microscope for analysis.

The data collected from cylinder scrape-down analysis is used to determine the condition of the cylinder walls and whether any corrective action is needed. For example, suppose the analysis reveals significant wear or damage to the cylinder walls. In that case, maintenance teams may need corrective action, such as replacing the cylinder or adjusting the engine settings to reduce stress on the cylinders.

 

Cylinder scrape-down analysis is a valuable technique for monitoring the condition of diesel engines, particularly those used in marine applications. By detecting potential problems early, maintenance teams can take corrective action before significant damage occurs, helping to ensure the reliability and longevity of the equipment.

3.1 Procedure for Vibration Monitoring and Recordkeeping

Vibration monitoring is an essential part of predictive maintenance. The following steps outline the procedure for vibration monitoring and recordkeeping:

3.1.1 Vibration Monitoring Data Collection

The first step is to collect vibration data from the equipment. The vibration monitoring equipment, such as the CMT VIB. MONITOR A4900 can be used to measure the vibration levels of various machinery parts. The data collected should include the following information:

·         Date and time of measurement

·         Equipment identification number

·         Measurement point location

·         Vibration amplitude in all three directions (x, y, z)

·         Machine speed (if applicable)

·         Any other relevant information, such as temperature, pressure, and load.

Collecting data from the same measurement point regularly is important to establish a baseline and detect any deviations. In addition, the data should be recorded in the PMS (Planned Maintenance System) for analysis and future reference.

3.1.2 Vibration Monitoring Data Analysis

Once the vibration data is collected, it is analysed to detect any abnormalities or deviations from the baseline. The data can be analysed using specialised software, which can identify the root cause of the problem and provide a recommendation for corrective action.

·         The analysis should include the following steps:

·         Comparison with baseline data

·         Identification of any trends or changes in vibration amplitude

·         Identification of any unusual vibration patterns, such as high-frequency noise or low-frequency vibration

·         Identification of any critical frequency ranges

·         Determination of the root cause of the problem.

3.1.3 Corrective Action –Maintenance

After the analysis, corrective action is taken to address the identified problem. The corrective action may involve maintenance or repair of the equipment, replacement of damaged parts, or adjusting operating parameters. The following are some examples of corrective actions for vibration monitoring:

3.1.3.1 Corrective Action – Motor

·         Rebalancing or realigning the motor

·         Repairing or replacing the bearings

·         Adjusting the belt tension

·         Replacing the motor if necessary.

3.1.3.2 Corrective Action – Pump

·         Replacing or repairing the impeller

·         Realigning the pump

·         Replacing the pump bearings

·         Adjusting the pump speed or discharge pressure.

3.1.4 Record-Keeping in PMS

Finally, all the data collected and analysed should be recorded in the PMS. The record should include the following information:

·         Date and time of measurement

·         Equipment identification number

·         Measurement point location

·         Vibration amplitude in all three directions (x, y, z)

·         Machine speed (if applicable)

·         Corrective action taken

·         Date and time of corrective action

·         Any other relevant information.

Recording and analysing vibration data regularly can help detect problems early, prevent equipment failure, and increase equipment reliability and uptime.

3.2. Procedure for Thermal Imaging and Recordkeeping

Thermal Imaging is a non-destructive testing technique for detecting temperature variations in machinery and electrical systems. It is an effective tool for detecting potential problems such as loose connections, faulty insulation, and component overheating.

3.2.1 Thermal Imaging Data Collection

The data collection process for thermal Imaging involves using a thermal imaging camera, such as the FLIR E6. The camera is used to capture images of the equipment being monitored. To ensure accurate and reliable data collection, the following steps should be taken:

·         Turn off any equipment that is not needed for the test.

·         Ensure that the equipment being monitored is in its normal operating condition.

·         Take a thermal image of the equipment from a safe distance, ensuring the entire equipment is in the field of view.

·         Take multiple images of the equipment from different angles to capture potential issues from different perspectives.

3.2.2 Thermal Imaging Data Analysis

Thermal imaging data analysis involves the interpretation of the thermal images taken during the data collection process. Therefore, the following steps should be taken during the data analysis process:

·         Review the thermal images for any hotspots, which are equipment areas with higher temperatures than the surrounding areas.

·         Compare the hotspots' temperature readings to the equipment's baseline temperatures to determine if the temperatures are within normal operating ranges.

·         Evaluate any temperature differences between similar equipment to determine if any discrepancies exist.

Determine if any corrective actions are necessary based on the analysis.

3.2.3 Corrective Action – Maintenance

If corrective actions are necessary based on the thermal imaging data analysis, the following steps should be taken:

3.2.3.1 Corrective Action – Machinery and Electrical Panels

·         Identify the issue and the equipment affected.

·         Determine the cause of the issue.

·         Develop a plan to resolve the issue.

·         Schedule maintenance activities as necessary.

Perform corrective actions to resolve the issue.

3.2.3.2 Corrective Action – Motor

If an issue is identified with a motor during thermal imaging data analysis, the following corrective actions should be taken:

·         Identify the issue and the motor affected.

·         Determine the cause of the issue.

·         Develop a plan to resolve the issue.

·         Schedule maintenance activities as necessary.

Perform corrective actions to resolve the issue, including repairing or replacing the motor.

3.2.4 Record-Keeping in PMS

All thermal imaging data, analysis, and corrective actions should be recorded in the PMS. This includes the thermal images taken, the analysis performed, and any corrective actions are taken. The data should be stored in a centralised location that is easily accessible for future reference. This information can be used for trend analysis and to identify recurring issues requiring additional attention.

3.3. Procedure for Lube Oil Sample Collection, Analysis and Recordkeeping

3.3.1 Lube Oil Sampling and Data Collection

A proper sampling technique is essential for accurate and representative oil analysis. Therefore, the following steps should be taken during lube oil sampling:

·         Use clean sampling tools to prevent contamination of the sample. The sampling tools should be dedicated to specific oil types to prevent cross-contamination.

·         Sample from the exact location each time. This will help ensure that the sample represents the oil system.

·         Use the proper sampling method for the specific equipment. For example, some equipment requires a running sample, while others require a static sample.

·         Record the sample location, equipment type, and date on the bottle or container.

·         Sample at the recommended interval, typically based on equipment operating hours or time in service.

·         Submit the sample to the laboratory as soon as possible to ensure accurate analysis.

3.3.2 Lube Oil Testing – Reports Data Analysis

The laboratory will report the oil's physical and chemical properties, such as viscosity, flash point, water content, and particle count. Additionally, the report will provide information on the oil's wear metals, contaminants, and additives.

The following are the typical tests conducted on lube oil samples:

·         Viscosity test: This test measures the thickness of the oil.

·         The flash point test determines the temperature at which the oil will ignite.

·         Water content test: This test determines the amount of water in the oil.

·         Particle count test: This test measures the number and size of particles in the oil.

·         Wear metal analysis: This test determines the amount of wear metals in the oil and can indicate the condition of the equipment.

·         Contaminant analysis: This test determines the presence of contaminants such as fuel, coolant, or dirt.

·         Additive analysis: This test determines the amount of additives in the oil.

3.3.3 Corrective Action –Maintenance

The laboratory report should be reviewed by maintenance personnel to determine if corrective action is required. The following are examples of corrective actions that may be required:

·         Change the oil: If the oil has exceeded its recommended service life or contains high levels of wear metals, contaminants, or additives, the oil should be changed.

·         Investigate the equipment: If the wear metal levels are high, the equipment should be inspected for signs of wear or damage.

·         Clean the equipment: If the oil contains high contaminants, the equipment should be cleaned to prevent further contamination.

·         Repair or replace equipment: If the equipment is damaged, it should be repaired or replaced.

3.3.4 Record-Keeping in PMS

All lube oil analysis reports should be retained in the vessel's Planned Maintenance System (PMS) for future reference. In addition, the following information should be recorded:

·         Sample location and date of sampling.

·         Equipment type and identification number.

·         Laboratory report results.

·         Corrective actions are taken.

·         Follow-up actions are required.

·         Next scheduled sampling date.

4. Job Postponement Request Procedure with Examples

While predictive maintenance is critical for ensuring optimal equipment performance and preventing unscheduled downtime, there may be instances when it is impossible to conduct a scheduled maintenance check on time. For example, a vessel may be in the middle of a vital operation or experiencing harsh weather conditions, making it challenging to perform routine checks.

In such cases, it is essential to have a job postponement request procedure to ensure that maintenance checks are rescheduled and completed as soon as possible. Below are the steps involved in the job postponement request procedure for vibration monitoring, thermal Imaging, and lube oil analysis:

·         Identify the Need for Postponement: The first step is to identify why the scheduled maintenance check needs to be postponed. This could be due to operational reasons, equipment unavailability, weather conditions, or any other reason hindering the maintenance check completion.

·         Notify the Relevant Parties: Once the need for postponement has been identified, the relevant parties must be informed. This includes the maintenance team, the operations team, and any other stakeholders affected by the postponement.

·         Request for Postponement: The maintenance team must submit a formal request for postponement to the vessel management team. The request should include the reason for the postponement, the proposed new date for the maintenance check, and any other relevant details.

·         Approval or Rejection of Request: The vessel management team will review the postponement request and either approve or reject it. If the request is approved, a new date for the maintenance check will be scheduled. If the request is rejected, the maintenance team must justify and propose an alternative date for the maintenance check.

·         Recordkeeping: Once the postponement request has been approved, it must be recorded in the vessel's planned maintenance system (PMS) to ensure that the new date is tracked and the maintenance check is completed as soon as possible.

4.1 Example of Job Postponement Request Procedure for Vibration Monitoring:

·         Identify the Need for Postponement: Due to a sudden change in operational requirements, performing the scheduled vibration monitoring check on the designated date is impossible.

·         Notify the Relevant Parties: The maintenance and operations teams are notified of the need for postponement.

·         Request for Postponement: The maintenance team submits a formal request for postponement to the vessel management team. The request includes the reason for the postponement, the proposed new date for the maintenance check, and any other relevant details.

·         Approval or Rejection of Request: The vessel management team reviews and approves the request, rescheduling the vibration monitoring check for the proposed new date.

·         Recordkeeping: The postponement request is recorded in the vessel's PMS to ensure the new date is tracked, and the vibration monitoring check is completed as soon as possible.

 

 

4.2 Example of Job Postponement Request Procedure for Thermal Imaging:

·         Identify the Need for Postponement: Due to adverse weather conditions, performing the scheduled thermal imaging check on the designated date is impossible.

·         Notify the Relevant Parties: The maintenance and operations teams are notified of the need for postponement.

·         Request for Postponement: The maintenance team submits a formal request for postponement to the vessel management team. The request includes the reason for the postponement, the proposed new date for the maintenance check, and any other relevant details.

·         Approval or Rejection of Request: The vessel management team reviews and approves the request, rescheduling the thermal imaging check for the proposed new date.

·         Recordkeeping: The postponement request is recorded in the vessel's PMS to ensure that the new date is tracked and the thermal imaging check is completed as soon as

5. Conclusion

In conclusion, condition-based predictive maintenance monitoring is a powerful tool that can help industries save time and money by reducing the need for costly repairs and downtime. The various techniques, including vibration monitoring and analysis, thermal Imaging, and lube oil analysis, can help detect potential problems before they become critical, allowing maintenance teams to take corrective action proactively.

Procedures for each technique discussed in this white paper have been outlined, including data collection, analysis, and corrective action. By following these procedures, maintenance teams can improve the reliability and efficiency of their equipment and reduce the likelihood of unexpected failures.

Job postponement requests can be made when necessary. However, they should be handled according to a specific procedure to ensure that they do not compromise the effectiveness of the predictive maintenance program.

By implementing a condition-based predictive maintenance program, industries can significantly reduce their maintenance costs, extend the life of their equipment, and improve overall operational efficiency. Therefore, it is crucial to recognise these techniques' importance and ensure they are utilised to their full potential.

 For more information and guidance contact us.

Disclaimer:

Out of Box Maritime Thinker © by Narenta Gestio Consilium Group 2022 and Aleksandar Pudar assumes no responsibility or liability for any errors or omissions in the content of this paper. The information contained in this paper is provided on an “as is” basis with no guarantees of completeness, accuracy, usefulness, or timeliness or of the results obtained from the use of this information. The ideas and strategies should never be used without first assessing your own company situation or system, or without consulting a consultancy professional. The content of this paper is intended to be used  and must be used for informational purposes only

AI-PREDICTIVE MAINTENANCE

 AI-Based Predictive Maintenance - Start

The start of AI-based predictive maintenance is data collection and pre-processing. This involves gathering data from various sensors placed on the machines or embedded in the machine itself. The data can include temperature, pressure, vibration, and load. This data is then pre-processed to remove any outliers or noise and to ensure the data is in a format suitable for analysis.

One common approach to anomaly detection in rotating machinery is to use statistical models such as Gaussian Mixture Models or Hidden Markov Models. These models are trained on historical data collected from the equipment to establish normal operating conditions. Then, real-time data can be compared to the model to identify deviations from normal behavior, which can be flagged as an anomaly.

Another anomaly detection approach is using machine learning algorithms such as clustering or deep learning. For example, a clustering algorithm can be used to group similar patterns in the data and then identify any significantly different patterns. Deep learning algorithms such as autoencoders or recurrent neural networks can also detect anomalies by learning patterns in the data and comparing new data to the learned patterns to identify any deviations.

Once the data is pre-processed, the next step is to perform feature engineering, which involves selecting and transforming the relevant variables from the data to help improve the accuracy of the predictive models.

Once anomalies have been detected, it is vital to interpret the data and determine the root cause of the problem. This information can then be used to make informed decisions about the maintenance that needs to be performed, such as replacing a worn part or performing a routine check.

After feature engineering, the next step is to choose a suitable machine learning algorithm to develop the predictive maintenance model. This can include algorithms such as random forests, decision trees, support vector machines, or neural networks. The choice of algorithm will depend on the complexity of the data and the problem being solved.

Rotating Machinery - Anomaly Detection

Through regular use, the deterioration of a rotating machine, such as a pump or fan, can produce anomalies, which should be viewed as a warning of suboptimal conditions rather than a complete shutdown. Sensors play a critical role in detecting these anomalies, as the sensors and sensor outputs used in anomaly detection significantly impact their performance. Therefore, choosing sensors with the appropriate capabilities and features is essential. However, there is still a risk of error if the data from these sensors is not accurately transferred to the model. Incorrect data transfer can result in false-positive alarms, causing a waste of time and effort or even more severe consequences such as missed alarms and material and moral damage. Supervised learning is impossible when no labelled data from machines with historically defective and healthy signals exist. In this situation, a machine learning model can be trained on a training set consisting only of "normal" samples. Then, an anomaly alarm can be triggered based on measuring the distance between the original signal and the predicted signal.

However, we need to define "normal" and "abnormal" at this stage, which can be challenging. One way to overcome this challenge is to use unsupervised learning methods such as clustering or autoencoders. Clustering methods group similar data points together, and if there is an anomaly in the data, it will be placed in its cluster, making it easy to identify. Autoencoders are neural networks that can reconstruct the input data, and any deviation between the input data and the reconstructed data can indicate an anomaly.

Another technique that can be used in anomaly detection is time-series analysis, where the behavior of the rotating machine over time is analysed to detect any deviations from normal behavior. For example, the vibration signal of a machine can be analysed over time to detect any changes in the frequency spectrum or any sudden spikes in vibration. These changes can indicate a potential issue with the machine, and maintenance can be performed to prevent a complete failure.

Feature Extraction

Rotating equipment operates by generating vibrations and maintaining these vibrations at acceptable levels for reliable production. To detect anomalies in these machines, sensors play a crucial role in the predictive maintenance process. Vibrations measured by accelerometers are the most basic units for detecting anomalies. Two main features can be extracted from the acceleration signals collected: time domain features and frequency domain features.

Metrics such as RMS, Crest, Kurtosis, and Peak can be obtained from the time domain. In contrast, total harmonic distortion, harmonic indications, spectral centroid, and sideband energy are extracted from the frequency domain. It is crucial to transfer these frequency domain features to the machine learning model to create a reliable model. The frequency content is a signature of the normal or abnormal state of the machine, and the speed information should also be included in the model for independent harmonic analysis. With each frequency amplitude from a 3-axis vibration sensor entered as a feature, the machine learning model can compare thousands of features in a multi-dimensional mathematical space, allowing even the most subtle anomalies to be detected based on the model's resolution.

Machine Mode Analysis

Numerous types of machines are utilized in various industries, each designed to perform different functions. These machines operate in cycles powered by various factors like fluctuating production speed, raw materials, and processes. For instance, the speed of a rolling mill may vary based on the desired material quality, thickness, and production speed. As a result, the vibration data from the machine also changes with the speed.

In traditional predictive maintenance applications, anomaly alarms may trigger with every speed change, which is inaccurate. In such cases, it is crucial to include process parameters like speed, power, etc., in the model and determine the different machine modes first. For example, during a test on a roller, vibration data was collected at four different speeds, and the machine learning model revealed six distinct modes. Four of these modes operate at different speeds, one is the mode in which the machine is idle, and the last mode is the mode in which anomalies occur.

The graph shows five groups if the measurements are not coloured, but the model has six modes because the machine's operating cycle in one cluster, divided into two groups, is the critical speed. This leads to mechanical looseness and coupling over time, resulting in a new abnormal mode (purple measurements in Figure 2). Although the standard deviation of the data in the cluster is evident compared to other clusters, it is challenging to distinguish two different modes in that cluster.

 

Two measurements operating in the same mode can be differentiated by their spectral information, as shown by the blue and purple plots. However, it may not be easy to distinguish these two measurements by looking at their average Vrms values. The reason is that Vrms may not be sufficient for high-frequency vibrations and may not detect anomalies in some fault types. Hence, transferring all spectral information to the machine learning model ensures optimal results in anomaly detection.

Conclusion

AI-based predictive maintenance is crucial in detecting and preventing failures in rotating machinery. The first step in implementing AI-based predictive maintenance is data collection and pre-processing, followed by feature engineering and selecting the appropriate machine learning algorithm to develop the predictive maintenance model. Vibration signals measured by accelerometers are the most commonly used sensors in predictive maintenance, and features such as RMS, Crest, Kurtosis, Peak, total harmonic distortion, harmonic indications, spectral centroid, and sideband energy can be extracted from the signals for analysis. Furthermore, time-series analysis and unsupervised learning methods like clustering and autoencoders can detect rotating machinery anomalies. Additionally, it is essential to include machine mode analysis and process parameters in the predictive maintenance model to increase its accuracy.

References

1.       www.rolls-royce.com. (n.d.). Rolls-Royce opens first Ship Intelligence Experience Space. [online] Available at: https://www.rolls-royce.com/media/press-releases/2017/27-11-2017-rr-opens-first-ship-intelligence-experience-space.aspx [Accessed 13 Feb. 2023].

2.       Medium. (n.d.). Medium. [online] Available at: https://medium.com/@connect.hashblock/the-ultimate-guide-to-decision-tree-algorithms-2ff42d7cf6c [Accessed 13 Feb. 2023].