A Step-by-Step Guide to Evaluating the Credibility of Overpressure Scenarios
If you are evaluating overpressure scenario credibility, and also applying overpressure protection by system design, you need a rigorous tool to help you in that process. In this newsletter, Neil Prophet shows step-by-step how to determine the credibility of an overpressure scenario and the magnitude of an overpressure event using Process Safety Office? software from ioMosaic Corporation . Compliance guidelines are included.
ASME Section XIII
ASME Section XIII is a set of standards by the ASME (The American Society of Mechanical Engineers) for overpressure protection of pressurized equipment, such as boiler pressure vessels and piping systems. ASME Section XIII addresses two ways of overpressure protection by system design:
In the case of where pressurized equipment for which the pressure is self-limiting, Process Safety Office? SuperChems? software can help us determine the maximum coincident pressure and temperature for the various applicable overpressure scenarios. In situations where we are protecting pressurized equipment, where the pressure is not self-limiting, we can again use Process Safety Office? SuperChems? software to determine the maximum coincident pressure and temperature for the applicable overpressure scenarios. We can also use Process Safety Office? PSMPro? software to determine overpressure scenario credibility.
In this newsletter, I will focus on pressurized equipment where the pressure is not self-limiting. According to ASME Section XIII, there are a number of conditions that need to be met to be able to apply this process:
The system may be protected by a combination of system design and pressure relief devices if the following conditions are met:
(a) Pressurized equipment is not exclusively in air, water, or steam service, except for:
(b) The decision is the responsibility of the user.
(c) The user shall conduct a detailed analysis to identify and examine all scenarios that could result in an overpressure condition, and magnitude of the overpressure. A multidisciplinary team shall conduct the analysis.
(d) The overpressure scenario shall be readily apparent so that operators or protective instrumentation will take corrective action to prevent operation above the MAWP.
(e) There shall be no credible overpressure scenario in which the pressure is not self-limiting. Credible events or scenario analysis as described in WRC Bulletin 498 shall be considered.
(f) The results of the analysis shall be documented and signed by the individual responsible for management of operation of the pressurized equipment. Documentation shall include:
Because we will have to model those scenarios over time dynamically, Process Safety Office? SuperChems? software is very effective. It is not just a case of modeling release requirement, steady state calculation, or steady state release capacity calculation.
It's also worth noting that ASME Section XIII and UG 140 (when it was in use up until 2021) don't specifically require Safety Instrumented Systems for overpressure protection by system design. But in the process of evaluating the overpressure scenario credibility, what you'll find is that a safety instrumented system is usually required to reduce the overpressure scenario frequency to an acceptable level.
WRC 498
In 2005, WRC 498 was published by The Welding Research Council to provide guidance on the application of Code Case 2211 for Overpressure Protection by System Design. It was fully incorporated into ASME VIII UG-140 in 2008. The bulletin had been referenced in previous editions of the Assembly Section VIII.
The WRC 498 bulletin follows Recognized and Generally Accepted Good Engineering Practices (RAGAGEPs). It promotes consistency in overpressure protection by system design and recommends following a five-step process:
Step 1 ― Identify and document all initiating events or causes and the resulting overpressure. All scenarios are considered credible until evaluated and specifically designated to be non-credible.
Step 2 ― Analyze the actions and consequences of the causes from Step 1 to identify scenarios that are not credible.
Step 3 ― For each action and consequence deemed credible, determine a PDP (process design pressure) and MAWP.
Step 4 ― Document the design features and the governing scenarios used to establish these design features.
Step 5 ― Identify all process equipment that is critical to preventing pressure above the MAWP and document appropriate operating, inspection and maintenance procedures.
Step 1 and Step 2 are the main activities that will be discussed here. Step two addresses the activity of evaluating overpressure scenario credibility. We will also look at how to determine maximum coincident pressure and temperature for overpressure scenarios. I will use a case study as the basis of these discussions.
Overpressure Scenario Credibility
After reviewing their relief system design basis, an operating company discovered that it was inadequate for specific scenarios of concern:
The reactors were also fitted with Safety Instrumented Systems, which can be taken into account per Annex E (Informative) of API - American Petroleum Institute Standard 521.
The changes may have occurred in the operating conditions over the lifetime of the facility, or changes in throughput, or the fact that certain scenarios were missed during the initial design, or were inadequately modeled. The company operates multiple (four) fluidized bed reactors:
In this diagram of one of the reactors, we see the flow coming in, we see the reactor itself, and we see the effluent being discharged. We could have a blocked outlet, provided both of these paths are closed, and we could have a loss of cooling on the reactor itself. There are also Safety Instrumented Systems. The valves are shown, but the logic is not.
Therefore, we can apply the WRC 498 process, which does list several techniques for evaluating overpressure scenarios and determining their credibility, such as HAZOP, What-If, or Failure Mode, Effects, and Criticality Analysis (FMECA). The ioMosaic Corporation team applied the Fault Tree Analysis (FTA) technique as the most efficient and effective method to assess the relief systems. This method meets pressure protection standards API Standard 521, ISA 84.01, and WRC 498. Within WRC 498, there is a suggested annual probability of less than 1 in 10,000 as the maximum annual probability for an overpressure scenario to be considered non-credible. Just bear in mind that it's worded as a suggested value.
We have three different cases of blocked outlets and two cases of loss of cooling. That implies that five different fault trees needed to be created.
The Fault Tree Analysis (FTA) technique can take into account factors such as initiating event modifiers such as reactor operating phases, the number of reactors operating at any given time, or switch permissives to control the position of valves. The technique can also take into account Independent Protection Layers. Those would be defined as either active or passive engineered controls that prevent the chain of events, leading to an undesirable outcome. In this case, the undesirable outcome would be the overpressure scenario. Those could be Safety Instrumented Systems or the basic process control systems.
PSMPro? Fault Tree Tutorial
This next section is a step-by-step tutorial on how to set up a fault tree with PSMPro? software. Prior to constructing the fault tree, you must have a good understanding of the failure scenario being analyzed.
Develop a Fault Tree
Upon opening the fault tree application, only the Top Event is displayed on a blank worksheet.
The Top Event can be defined by selecting the “Modify” option from the Operations menu, or by right-clicking.
The next step is to name the Top Event and the relevant fault tree symbol assigned to it.
The fault tree can then be further developed by selecting “Add Branch” from the Operations menu, or by right-clicking.
As the fault tree grows, PSMPro? will optimize the layout.
Failure Rate Data can be built into PSMPro?and accessed during fault tree construction.
The fault tree can be as large or small as desired and can span multiple pages using the Transfer function.
PSMPro? will check to ensure that the correct logic is applied when defining frequency or probability inputs.
OR Gate
AND Gate
Once the fault tree is defined, there are a number of results that can be calculated from the Math menu, or by right-clicking.
Compute
Get Cut Sets
Get Path Sets
A Cut Set is a route through a tree between an initiator and the Top Event. The shortest credible way through the tree is called a Minimal Cut Set. Cut Sets are shown graphically in red and in tabular format.
Path Sets are the opposite of the cut sets. If none of the events in a path set occur, the top event will not occur. Path Sets are shown graphically in red and in tabular format
The ‘Compute’ option will calculate the probability or frequency for the selected node – whether this is for part of the fault tree, or the Top Event.
Generate Reports
In addition to the visual fault tree, several types of reports can be generated in PSMPro? to support the fault tree itself:
Fault Tree Report provides a summary of frequency and probability data used in the analysis.
Detailed Fault Tree Report provides a summary of frequency and probability data used in the analysis, as well as equipment and data details.
The Fault Data Report provides a summary of frequency and probability data available within PSMPro?.
The Filtered Fault Data Report provides a filtered summary of frequency and probability data available within PSMPro?.
Fault Tree Results
Multiple Fault Trees were developed to represent each overpressure scenario; three scenarios for blocked outlets and two fault trees for loss of cooling. These took into account the initiating events, valve positions, and safeguards, and the logic was laid out visually to help explain the sequence of events leading to the overpressure scenario.
One of the key reasons for performing fault tree analysis is to be able to determine the frequency or probability of the top event. This means that each of the contributing steps has a frequency or probability assigned to it, so there's a lot of work involved prior to setting up the calculation in PSMPro?. It helps to have a sketch of your fault tree in mind. And you need to do some research to find various probabilities or frequencies to be able to input them into PSMPro?.
The calculated frequencies of the potential overpressure scenarios are summarized below. We can see that in each case, the scenario frequency is less than our target frequency of 1 times 10 to the minus 4. That helps us determine that these overpressure scenarios could be considered not credible.
SuperChems? Tutorial
The ability to model the maximum coincident pressure and temperature would be another aspect to consider when evaluating these overpressure scenarios. This would be required for credible scenarios rather than non-credible scenarios with the aim of ensuring that the pressure of accumulation doesn't reach certain specified limits. In the case of pressurized equipment for which the pressure is self-limiting, then we would need to demonstrate that the maximum coincident temperature and pressure do not exceed the MAWP of that temperature. Or for the case where we have pressurized equipment for which the pressure is not self-limiting, then the maximum possible limit is the test pressure for the credible scenarios. Rather than just doing steady-state API sizing equations, we need to do more dynamic, advanced calculations instead. This next section is a step-by-step tutorial on how to set up the dynamic calculations with Process Safety Office? SuperChems? software using the dynamic two-phase model.
Create the Scenarios
Upon opening SuperChems?, create the scenarios by selecting the “Scenario” option in the “Define” tab.
This takes us to the Scenario object window in SuperChems?. Here, create scenarios and give the scenarios meaningful names. The scenario key is of particular importance for the final reports. The format of the scenario will be the Vessel Tag, Scenario Key, and Descriptor.
Next, create the mixture by clicking on the “Chemical” option in the “Define” tab, and then selecting the “Mixture” option in the pop-out window.
The compounds can easily be added to the mixture from the built-in database of over 4,000 chemicals in SuperChems?. Just click the “Green Plus Icon” at the top right corner of the mixture window or right-click within the mixture pop-out window.
Before we go into the inputs, let’s review the color coding system. Red will represent an input necessary for the calculation, blue will represent useful documentation, and green will represent miscellaneous notes.
Next, create the vessel object by clicking on the “Equipment” option in the “Define” tab. In the vessel objects window, click on the “Green Plus Icon” to create a new vessel. The top section of the vessel object inputs is outlined in blue, which means it is useful information, but not necessary for the calculation.
By scrolling down you can find the vessel dimensions, height above grade, and head type, which are required inputs.
Inside the vessel object, we can also define the fire loading by clicking on “External Fire” on the left side and checking the box to “Enable Fire Loading”. We can specify fire duration, an API 521 fire flux, liquid level in terms of volume percent, and additional wetted area. As these are outlined in red, they are all required inputs for a fire.
Now apply the mixture and vessel to the scenario by navigating to the objects window, then right-click on the mixture or vessel, and select “Apply” in the pop-out window.
All sizing models are located in the Flow and Source Term context menu, which is accessed by right-clicking on “Models” within the scenario. Since the vessel is 25 vol% full, select “Vessels Containing Two Phases (Dynamic)” model.
Before running any calculations, a dynamic model needs to be initialized. That is, the pressure, temperature, volume fill percent, and mass of contents need to align.
Several tools within this model will help with this step, but for this example, we’ll only go over one tool. The “Specify Fill Fraction” tool is located in the Vessels Containing Two Phase (Dynamic) model’s toolbox. By specifying a volume fill percent, it will determine the corresponding mass and initial temperature.
Once the “Specify Fill Fraction” tool finishes running, confirm the starting temperature agrees with the normal operating temperature. The figure below shows in this case, it is close enough.
One final step before running this calculation is to set it as a required calculation by clicking on “Required”. This enables the reporting utility to recognize it (very important for printing reports).
Now the calculation is ready to run. In this first example, we’ll run the calculation with no relief device. As expected, the pressure and temperature continue to rise. This vessel may fail due to overpressure or overtemperature.
Now we will run another dynamic model with a relief device layout. But first, we need to create a relief device layout. Before creating a relief device layout, first create a PSV by going to the “Define” tab, clicking “Piping Layout” option, and selecting “Pressure Relief Valve”.
The important data entries in the PSV object are the inlet and outlet size, the discharge coefficient, the area, and the set pressure.
With the PSV object created, we can then create a piping layout by clicking on the “Piping Layout” option in the “Define” tab.?
In the “Piping Layout” object window, create a new piping layout. The PSV can be selected from the available units box and moved over to the selected units box. Additional piping may be moved over, but in this example, we only include the PSV. For layouts containing only one object, the discharge coefficients also need to be manually defined.
With the piping layout defined, we simply need to copy the scenario with no PSV since it is already initialized, and apply the piping layout to it. To do this, right-click on the scenario and select “Copy” then give it a meaningful name.
With the copied scenario selected, right-click the “Top Primary Piping Layout” and select “Change” to switch to the piping layout with the relief device.
Next, navigate to the “Vessels Containing Two Phases (Dynamic)” model and change the flow type of the piping layout to vapor. Click on “Update” to register the changes and then “Run”.
In this case with the PSV applied we can see that the pressure builds up and then cycles up and down as the pressure relief device opens and closes. Similarly, we could plot temperature vs. time to make sure that the temperature build-up is not excessive and doesn't exceed MAWP. The other graph on the right shows us the vessel level vs. time. Provided that the accumulation is kept below our target pressure, then this system would be sufficiently protected in the event of a credible overpressure scenario.
Note that the applied relief device by itself is adequate in limiting the pressure below MAAP, but around one hour, the vessel becomes vapor full, so it may still fail due to overtemperature.
Overpressure Scenario Credibility Results
So we can therefore evaluate any system for overpressure protection by system design to see if the scenario is credible or non-credible. We can also model the pressure build-up to make sure it is within acceptable accumulation limits.
Report Generation
The reporting requirements for overpressure protection by system design are pretty stringent. We have put a lot of effort into making the Process Safety Office? reporting utility user-friendly and simple to use. This section addresses how we generate a report to accompany the documentation step-by-step. Please note that certain check box APIs need to be selected if specific graphs are to be included.
We have created two scenarios: One dynamic model without a PSV and one with a PSV. To print the report, we go to the “Reports” tab and click on the “Prepare Report” option.
Select the vessel associated with the scenarios from the dropdown list in the “Select Vessel” tab and click on “Next”.
Select the piping layout associated with the scenarios in the “Piping & Vessels” tab and click on “Next”.
Progress to the “Final Step” tab and then select the “Generate Report” option. Click on “Finish”.
This will take you to the main “Report Writer” window. We could click on “Generate report” now, but in this case, we want to include the pressure history and vessel fill history graphs.
To display pressure history in the report, navigate to the “Graphic Settings” tab, select the “Results” option, and then select “Pressure History”.
Similarly, to display the vessel fill level history, navigate to the “Graphic Settings” tab, and select the “Mass Flow” and “Vessel Fill Level History” options.
Finally, navigate back to the “Report” tab, and click on the “Generate Reports” option.
The report will output to the same folder as the .COR file. Below are some screenshots from the printed report showing the summary table with the two scenarios and some other graphical outputs.
In Conclusion
Overpressure protection by system design requires a rigorous approach. Following compliance guidelines published by the ASME (The American Society of Mechanical Engineers) and helps ensure interoperability, safety, reliability, and quality. The tutorials showed step-by-step how to determine the credibility of an overpressure scenario and the magnitude of an overpressure event.
To learn more about Process Safety Office? software, visit our website > https://bit.ly/3R785kf
Process and Storage Vessels Exposed to Fires: Failure Assessment Webinar
In this 60-minute AIChE webinar presented by Georges Melhem , Ph.D., FAIChE, identify best practices for assessing the failure potential of pressure and storage vessels exposed to external fire. You’ll also explore mitigation considerations, including pressure relief and de-pressuring systems, water sprays, and fire proof insulation.
To register for the webinar, visit the AIChE website > https://bit.ly/3TBCkU9
How We Can Help You
Emergency Relief Effluent Handling System Design
Our team has decades of experience performing PRFS analysis and design and methods to maximize existing flare structures.
Pressure Relief and Flare System Design
Our risk-based approach helps mitigate near-unventable scenarios to a tolerable level of risk and develop economical designs for more credible events.
Process Simulation
Better and more accurately evaluate hazards in your oil, chemical, pharmaceutical, or LNG facility with an accurate process simulation.
Relief Header and Flare Analysis Systems
Delivering properly designed pressure relief systems for refineries and chemical plants that save both money and time.
To learn more, visit our website > https://bit.ly/3eKzTJA
#EmergencyReliefSystem #OverpressureProtection #ReliefDesign #FaultTree
Copyright ? 2024 ioMosaic Corporation. All rights reserved