9.4) Spectral Envelope

Vibration analysts can use several tools for fault analysis within the Web Platform. One of the main ones is the Spectral Envelope, useful for identifying high frequency faults through signal demodulation.

When accessing a spectrum in the Web Platform, the screen includes a button in the upper right corner of the screen to request an envelope (demodulation) of the signal.
 


Figure: Spectral Envelope Selection Button


By clicking the button ““, a window will be displayed asking for the frequency range in which you want to apply the envelope. The Web Platform has predefined envelope values, parameterized by the most common frequency ranges in vibration analysis. However, it is also possible to define a custom envelope in the same window. Simply select the “Customizable” option and define the desired frequency range.

 


Figure: Customizable envelope configuration


After selecting the frequency range, the envelopes (spectral and waveform) will be available at the top of the page. Both have the same functionalities as the spectral analysis.


Figure: Spectral envelope with filter from 3500 to 4300 Hz, pointing BPFO bearing failure (dark mode of the Web Platform activated, for better contrast)


 

 

9.5) Waveform autocorrelation

Autocorrelation is an analysis in the time domain that correlates waveforms with themselves to calculate a correlation factor at each time instant. It can be useful in helping the vibration analyst identify periodic frequencies, improving visualization and also, analogous to the previous section, allowing it to be displayed in its circular form.
To activate the waveform autocorrelation plot, simply, in the desired spectral screen, activate the option “” located in the upper part of the screen.
 


Figure: Activation of autocorrelation


The graph will be displayed just above the spectrum. The same tools available for the standart waveform are also available.
 

Figure: Waveform Autocorrelation


 

9.6) Spectral Waterfall

The waterfall is a three-dimensional graph where up to 10 spectra of a monitoring point are displayed, helping the analyst to identify fault trends from the growth of vibration amplitudes. 
It can be accessed via the Cascade button present at the top of the “Spot Viewer” and “Spectral” pages.
 

Figure: Spectral Cascade selection button


In the window that will be displayed, you must select which spectra will be used to form the cascade. The available filters are: date of spectra, settings (it is necessary that all spectra have the same frequency settings, monitored axes and duration) and acceleration threshold. The acceleration threshold filters spectra whose RMS acceleration is below a user-defined value, in order to only use graphs generated while the machine is running.
With the filters set, simply select the spectra to be used by clicking on the “” ssymbol to the right of each spectrum listed at the bottom of the window and then clicking ““.
 


Figure: Selection of Spectra to build the waterfall

The “Spectra Cascade” screen, which will open next, shows at the top some general information about the monitoring point and the spectra, such as machine, RPM, duration and number of lines.
Below, the waterfall chart and the available tools are displayed. The time period appears on one of the axes, in an increasing manner for a comparison of possible failure evolution.
 

To facilitate the visualization of the graph, it can be moved to the most suitable position for analysis. The following are the available forms of movement:
  • CTRL + Click + mouse movement: lateral movement;
  • Click + mouse movement: rotation;
  • Scroll mouse button: zoom in or out (graphic).
In addition, the “” button allows you to change the spectra used to generate the graph.
Picture: Metric options for cascading

Waterfall is also available for envelope and cepstro view, as shown above. Enabling either of these options will update the chart to reflect these types of metrics.
In the right corner of the chart some analysis tools are available  that can be applied to the waterfall, such as:

Figure: Spectral cascade visualization tools


1. Zoom tools, which allow you to analyze a specific frequency range of the spectra.

2. Cursor tools, which allow you to add single, harmonic or sideband cursors at specific frequencies.

3. Defining the metric on which the graphs will be displayed below (acceleration, velocity or displacement) and their respective units.

4. Definition of the analyzed axis (X, Y or Z). In the case of triaxial spectra, the user must define which axis will be analyzed each time.

5. Filter tool. High-pass, low-pass and band-pass filters are available for each spectrum of the waterfall.

6.Notes. Tool that allows you to add comments on a specific graph and frequency.

7. Color visualization tool. Definition of the colors of the graphics in a way that facilitates visualization, available in 4 modes.

     7.1 Global maximum amplitude: color gradient that takes all spectra into account. Peaks of higher amplitude will be displayed in warm colors.

     7.2 Local maximum amplitude: color gradient for each spectrum. Higher amplitude peaks will be displayed in warm colors.

     7.3 Pattern per axis: solid color, same for all spectra;

     7.4 Individual pattern: solid color, individual for each spectrum.

Picture: patterns of the Spectral Cascade
 
8. Show/Hide spectra. Allows the user to view only a portion of the spectra defined when generating the cascade;

9. Option to hide/display grid at the bottom of the graph and view the waterfall in logarithmic scale.


 

9.7) Side-by-Side Spectral Comparions

In addition to spectral comparison using a reference spectrum, it is also possible to view two vibration spectra side by side, making it easy to compare vibration spectra collected some time ago at the same spot, and even between different spots.

This option is available to the user on the spectral screen, using the “” button on the upper right of the screen.
 

Figure: Spectral screen with side-by-side comparison option


Clicking on this button will display a window showing the spectra available for side-by-side comparison.


To select a spectral window, click on the “” button, displayed right next to the window and confirm the selection by clicking on ““. To find the desired spectrum more easily, you can search by machine name, Spot, or sensor number.
 


Figure: Filter to choose spectral of comparison
 
This tool can be useful in fault diagnosis. The analyst can, for example, compare side by side spectra or vibration waveforms of the same spot, on different collection dates, making clear the evolution of vibration amplitude or even the appearance of new failure frequencies at the spot.
 

Figure: Waveform spectral comparison 
 
In addition, after viewing the spectra, you can access all the tools to do a more detailed analysis on each of the viewed spectra.


 
 

 

12) Gateways

The DynaGateway is the automated data collector for the Dyna-Predict and its purpose is to interact with the DynaLoggers within its bluetooth range, collecting measurements and requesting spectra according to user-defined time intervals. The collected data is sent directly to the Web Platform via Wi-Fi, Ethernet and Mobile Networks (according to the model chosen).

In the Web Platform, the user has the possibility to manage the Gateways, associated to the Desktop, through the Gateways screen. To access this screen, use the predictive section, from the side menu, and select the “Gateways” option.


Picture: Information tab of the Gateways of the selected workarea.

Accessing this screen you will find a list of all the Gateways that have been associated as well as information about their operation.
For more information about the prerequisites for operating the device, as well as the system configuration and management screens, refer to the Gateway User Manual.

 

11) Integrated Dashboard

The integrated Dashboard, available in the side menu of the Platform (figure bellow) takes into account both the predictive module (vibration and temperature monitoring) and the sensitive module (DynaSens) that has checklists and inspection routes.

This part of the Web Platform is divided into two sections:
▪ Health indicators
▪ Reports
 


Figure: Integrated Dashboard


 

11.3) Machine Viewer

The Machine Viewer is a tool for viewing information about the company’s machines, joining information from the sensitive and predictive modules. There are two ways to access the functionality: through the side menu, in the “Machines” option, or by clicking on the desired machine through the asset tree. This screen encompasses all information relating to a particular machine in the plant, grouping Subassemblies, Components and Spots common to it.

Applicable information and filters

The top of the Machine Viewer shows the path within the asset tree where this machine is located, a table with machine information, and filters for Subsets and Components. If any of these filters are applied, the information shown in the rest of the screen will refer only to the filtered items. In addition, you must define the starting instant and the number of revolutions displayed in the graph.
 

Figure: Machine information and applicable filters.

Timeline

Just below the filter options, a timeline is displayed with the events of all the monitoring points registered on the machine, such as Spots creation/deletion, Spots parameterization, spectral analysis, A2 alerts triggered, checklists performed, and reports issued. By default the events of the last seven days will be shown, but this interval can be changed through the calendar in the upper left corner.
 

Picture: Machine’s timeline.

It is worth noting that the events presented in the machine timeline are for the time period selected above. In the same way presented in the Spot Viewer, one can comment on each event individually, generating a history of conversation between users

Monitoring Points

Below the timeline, in the Machine Viewer, a reduced version of the DMA is shown, with only the Spots referring to that machine (or subsets/components, if a filter is applied). The information follows the same pattern as that displayed in the traditional DMA.
 

Figure: Reduced version of the DMA Dashboard on the machine screen

Anomalies – Sensitive Module

The “Anomalies” field lists all the anomalies reported by users in the  checklists’s answers filled out during the sensitive module routes, for the customers who have purchased it. Listed in this section are: the anomalies, observations, user responsible, subset/component, and date/time. By clicking on one of these anomalies, the user is redirected to the checklist report.
 

Picture: Anomalies – Sensitive Module

As in the other features on this screen of the Platform, the information displayed follows the pattern according to the time period chosen on the machine timeline.

Manual Measurements

Also belonging to the sensitive module, below the “Anomalies” field, a graph is displayed with the results of the manual measurements performed on the routes of the sensitive module. To select which magnitudes will be displayed in the graph, simply click on the “” option and select the desired magnitude.
 

Picture: Manual Measurement Graphs – Sensitive Module

The Sensitive Module is a tool in the DynaPredict solution for managing inspection routes on field. For more information about the usability of the solution, please see the Sensitive Inspection Manual.

Predictive Analysis

At the bottom of the page is the “Predictive Analysis” section, referring to the performance of technical reports from multiple monitoring points on the machine. All reports performed in the period are presented.
 

Figure: Predictive Analyses performed on the machine, per monitoring point

To make new reports, the user simply selects the icon “” and selects which Spots will be analyzed.


 

 

11.2) Viewing Reports (Predictive Analysis Reports)

The “Reports” division, present in the Integrated Dashboard, presents all the reports (predictive and sensitive) issued by the system users.
 

Picture: Sensitive and predictive reports

Regarding the columns shown:

  • Status: shows how critical the report performed by the user is.
    • Red: intervention required;
    • Yellow: alert condition;
    • Green: no problems found.

▪ Resolution: presents information about the treatment given to the report.

  • Open: the report was issued, but no corrective action was registered for the report;
  • In progress: a corrective action for the reported problem is in progress;
  • Resolved: the reported problem was corrected through some action / action of the team. By clicking on the line of the generated report, you can check what action was taken, if the user who changed the status has filled in details of the action.

▪ Location in the asset tree: points to where the Spot or machine belongs within the company’s hierarchical structure.

▪ Diagnosis: informs the diagnosis filled in by the user for the failure found.

▪ Detected Failure: informs which failure category was filled in by the user who created the report.

▪ Type: This screen shows predictive and sensitive reports. The “type” option indicates the origin of each report.

▪ User: information about which user filled out each report. Hovering the mouse shows the user’s name and e-mail.

▪ Deadline: filled in by the user in the report itself, represents the deadline for the action recommended by the report creator.

▪ Creation date: date on which the report was created.

In addition, if any of the reports have undergone changes, the “” symbol (placed on the far right of the line of the modified report) will allow you to check which information has changed and by which users. By clicking on this symbol, a new tab will be displayed showing the changes.
 


Picture: Historical changes in performed report 


 

11.1) Health Indicators

The Health Indicators, within the Integrated Dashboard, display the health indices of different plants or work areas of the company. Each is represented by a pie chart indicating the health of the assets in the area, based on the number of Spots and inspection routes that triggered A1 or A2 alerts.
 


Picture: Health indices by sectors or plants


In this same screen, you can click on the side arrows in each pie chart so that, within the organization of the asset tree, a new Dashboard is generated with the sublevels and machines related to the selected one. In this way, it is possible to perform an integrated and specific visualization of all areas and sub-areas of the plant.
The purpose of this screen is to show which areas need more attention from the maintenance or reliability team, i.e., which of them are theoretically in more compromised health and are impacting the overall health levels.

 

10) Creation of reports (predictive analyses)

The analysis of the data collected by DynaLoggers can be done in several ways through the Web Platform. The Predictive Analysis tab has as main goal to enable the elaboration of technical reports about failures, alerts or recommendations regarding the health of the monitored asset. The Predictive Analysis section is found at the end of the “Spot Viewer” and “Machine Viewer” pages.


Performing Predictive Analysis

To make a technical report of a Spot, just access the section through the Spot Viewer (at the bottom of the page) and select the “” option.


Picture: Creating Predictive Analysis


A new screen will be shown, requesting Spot analysis information:
 

Picture: Filling in the predictive report

The information tab has the following fields:

  • Status: selection of the current asset condition. Three options are available to inform the severity of the report;
  • Diagnosis: brief description of the problem encountered by the analyst;
  • Detected failure: selection of the failure found at the point among a series of categories;
  • Recommended action: indication of which actions should be taken based on the analysis performed;
  • Note: addition of infos that have not yet been cited;
  • Date on which the failure was detected: optional info to register from which moment the failure could be identified; 
  • Deadline for Action: add deadline for action based on the “recommended action”.

The Spot Viewer tab, on the other hand, brings the graphs referring to the history of the analyzed spot data, with the objective of providing background for the analyst when making his report. It is possible to add temperature, acceleration, and speed graphics with customized time periods. It is also possible to add an explanatory comment below the graph.


Picture: Data history tab available when creating a predictive analysis

The “Spectral Analysis” tab allows the addition of spectra that prove the detected failure. The selected spectra will be part of the report.



Picture: Adding spectra where failures are identifiable


Finally, the “Images” tab allows the user to add pictures to the report. By selecting the “Add Image” option, files saved on the user’s computer can be added and will become part of the report.



Picture: Adding images to the predictive report


Predictive Analysis Visualization

The predictive analysis performed are available for viewing and access in several ways, as presented below. To access the full version of the report, click on the report icons in any of the options listed below.

1. In the Spot viewer screen (data history): at the bottom of the screen, below the RMS acceleration graph.
 

Picture: Predictive Analysis in the Spot Viewer
 

2. In the timeline of the Spot Viewer screen: to access the complete report, simply click on the icon in the timeline.
 

Picture: Reports shown on the timeline of the data history screen

3. Through the Integrated Dashboard’s Reports screen 

On this screen you have all the reports history for all the Spots for which this type of report has already been generated. More details can be found in the “Viewing Reports” section below.

4. Through the DMA Dashboard, in the “Report” column. In the case of visualization of the analyses through the DMA dashboard, the last report performed and the criticality assigned is shown.
 

Picture: Column for viewing reports from the DMA

Predictive Analysis for multiple monitoring points

In the Machine Viewer screen, it is possible to perform technical reports for more than one monitoring point simultaneously. At the bottom of the page, you will find the option ““.
 

Picture: Option to issue report by machine 
 
Clicking on this option will result in the opening of a window similar to the one discussed in the previous section. The difference, in this case, is the need to include which Spots the user wants to add to the report.
 

Picture: Predictive Analysis for Multiple Spots
 

When the list of Spots for the machine appears, simply select, in the rightmost column, which ones will be added to the analysis using the “” button. Going forward, the diagnostics, the appointed failure, status, and recommended action options will be the same as those already present in the Predictive Analysis for just one monitoring point, detailed in the previous section.
When performing a batch predictive analysis, as in this case, the selected Spots will inherit the report, which will be available through the options listed in this section.