CloudPF® User manual
This document is the user’s manual for CloudPF®. It is also available in Italian, French and German.
CloudPF® (Cloud Process Flowsheeting) is a toolkit to publish the process models built with LIBPF® (the LIBrary for Process Flowsheeting) as web apps for cloud-based process simulation.
With CloudPF® you can easily create simple-to-use, customized web applications, that make process models more accessible so that non-experts can prioritize and optimize decisions in the proposal and design phases of new plants.
CloudPF® can be tested on simevo SaaS (Software-as-a-Service). CloudPF® is presented below in order to show its features. Here the Electrolysis demo (Elysim) was taken as an example, Elysim is a variant of CloudPF® to demonstrate the cloud-based simulation of electrolysis systems for H₂ production. The other demos are available here.
Homepage
The homepage (Cases list) looks like this:

Here you can create new cases or view those already created. Clicking on the button “New case” opens a modal where is possible to configure the plant and set certain presets (as set forth by the model developer):

Case detail page
Clicking on the case name in the case list leads to the detail page where you can see all the information on the case.
Here there are some tabs that you can choose to see different aspect of the plant.
The most useful and significant one is the “PFD” (Process Flowsheet Diagram) tab:

From here you can select the units and streams by clicking on the different parts of the flowsheet:

An alternative way of browsing units and streams is via the tree view panel on the left side.
By clicking on “Inputs” tab you can change different variables of the currently selected unit or stream:

Once all the values have been changed (optionally switching units of measurement) the case must be recalculated (button with the Pocket Calculator icon), after that the new results will be available on the “Outputs” tabs:

Certain models also offer graphical display of results such as Pie charts, Bar charts and Sankey diagrams.
Sensitivity studies
Very often in the course of a project sensitivity studies need to be performed. These sensitivity studies are very repetitive to perform manually: you must change one or more parameters (controlled variables) in a certain range for a certain number of points, run a simulation for every point and record in a table a certain number of results (monitored variables).
To automatize these tasks CloudPF® has a multi-dimensional sensitivity analysis feature which allows you to carry out several simulations by controlling one or more variables while maintaining all the other unchanged, and monitoring the effect on one or more results. With this tool you can find out how the model reacts to the variation of the parameters, and how sensitive it is; it also provides assistance for the selection of the so-called “critical” variables, that is those whose deviation from the nominal value influences the performance indicators the most; with this tool you can acquire a lot of information about the model and the process in a short time.
To start, click on the Sensitivity button in the toolbar: a drop-down menu appears with the list of saved sensitivities for the current model type (if any), and an entry “New sensitivity”.
Clicking on New sensitivity opens a modal dialog which proposes a name for the study based on the current timestamp (change it if you desire), and three tabs:
-
On the Control tab by clicking on the + Add controlled variable button at the bottom left it is possible to add one or more controlled variables, choosing from the list of all the variables that can be specified for the current case (these are the variables that appear in the “Input” tab of the units and streams), and setting:
-
the start and end values for the range to scan
-
the number of points the range must be divided into
-
optionally: a custom label which will appear in the results table column header
You can remove a previously set controlled variable by clicking on the X button on the right side of each row.
Once at least one controlled variable has been set, you can press Next to proceed to the second tab.
-
-
On the Monitor tab by clicking on the + Add monitored variable button at the bottom left it is possible to add one or more variables (results) to be monitored, choosing from the list of the results for the current case (these are the variables that appear in the “Output" tab of the units and streams), and optionally setting a custom label which will appear in the results table column header:
You can remove a previously set monitored variable by clicking on the X button on the right side of each row.
Once at least one monitored variable has been set, the Save button becomes active and you can press Next to proceed to the third tab.
-
On the Results tab you can start the sensitivity study by clicking on the Start button. The modal stays open and is all frozen while CloudPF® repeatedly calculates the case. During the execution of the sensitivity study the rows relative to each value of the controlled variables are color-coded and the results of the monitored variables are tabulated as soon as they become available. You can at any moment copy the table to the clipboard(by pressing the Copy results button), stop (by pressing the Stop button) and then resume (by pressing the Resume button).
At the end of the execution the sensitivity study configuration is saved so that you can recall it later; the sensitivities are saved in the browser so beware that if you change browser or device you will not find the studies you had saved on the other one.
About page
The About page provides an opportunity to explore in detail the demo, explaining the process involved and what CloudPF® can do in practice:

Support link
The Support link opens in another browser tab the LIBPF® / CloudPF® support community https://support.libpf.com right at the CloudPF cloud-based simulation category.
This is an open forum where users help and provide support to each other on a voluntary basis; it runs Discourse, the open-source discussion platform software system.
To post your questions and answers you must register with your email address or sign in with your Google or Github accounts.