Since the release of version 3.3.0, Source has undergone several functional and visual changes. These have resulted in significant performance improvements and a more robust user interface. Key features that have been introduced in Source 3.5.0 are outlined below. Note that further technical details about these changes are available in the Source 3.5.0 Technical Release Notes. A list of all issues can be viewed here: Full Jira list.
Performance improvements
The removal of some save points and how they are handled in Source 3.5.0 has resulted in a considerable improvement in editing performance. In particular, this has been implemented on almost all node and link feature editors and the standard “Single analysis” configuration. Groundwater functionality (used in storage routing and storages) still needs save points, it is now disabled by default to speed up editing for most users (choose Edit » Enable Groundwater to enable).
Additionally, expressions, now called functions, are executed more efficiently. Projects containing a large number of functions in particular will have considerable runtime performance improvements.
Function manager
A new Function Manager has superseded the Expression Editor, and as its name suggests, allows users to manage all functions and expressions at a central location. Its new user interface and text editor allow you to:
- View all the functions and expressions that have been configured in the scenario including where they are used; and
- Work with and edit all defined expressions and functions using the Function Editor, containing improved features such as auto-completion, syntax highlighting and a drag-and-drop capability.
A video demonstrating use of the Function Manager is available here: /wiki/spaces/SC/pages/51643405.
Moreover, a new Metaparameter Explorer user interface makes the process of setting up and working with metaparameters more efficient and easy to use.
Error reporting
Source 3.5.0 has an improved error reporting capability, with the following changes implemented in the new user interface:
- Changes in the log reporter provide a more informative explanation of errors, along with the ability to filter on each column;
- Identification and reporting of errors to include which component was involved;
- Implementation of a notes system, where you can add a comment in feature editors (right click on main node choose Add Note), data sources, individual values and functions. This becomes visible in the Log Reporter, Schematic Editor and the Geographic Editor once the scenario has been run;
- Changes in the way functions are processed:
- Time of evaluation log (choose Tools » Function Execution Order); and
- Enable a log of all evaluations of an expression at a nominated time-step (right hand click on a Function and select Add Log Date);
- Improved error reporting and validation on the controlled splitter node.
Insight: multi-objective optimisation
The goal of optimisation is to optimise a set of defined objectives based on decision variables that act as management levers. In Source, this is implemented in Insight, which is a multi-objective optimisation decision support framework that allows you to input decision variables and objectives you wish to optimise. The output consists of a set of optimised decision variables, values of the objective functions, and also values of any other functions you wish to monitor.
Key improvements to Insight include:
- The availability of a wide range of statistical functions for defining objective functions;
- The reporting of non-optimised (or secondary) objective functions, resulting in a broader scope of results, making the decision-making process more valid;
- The ability to re-use optimisation results in Source, eliminating the time and effort required to obtain a dataset directly from a Source project; and
- The ability to optimise over a range of scenarios, including climatic, operating rules, and project configurations.
Details on using Insight are available here: Insight: Objective optimisation.
Other important changes in Source 3.5.0:
- Removal of the Ordered Water Supplied recording parameter (for the water user node) to remove errors for certain configurations. In cases where model variables use this recorder, you will generally need to replace the recorder with the Supply Point's Extracted Volume parameter;
- Improvements to Lumped constituent routing, including implementation of link decay. Lumped routing constituent models without decay will show small changes in constituent concentrations and timings of changes in concentrations. The total constituent load over time, however, will not change. The change to constituent routing involved changing the order in which outflows, inflows and mixing were processed at link storages. Any projects using lumped routing should either not upgrade to Source v3.5.0, or re-run all analyses to re-establish baseline and scenario results.
- Reporting time of execution changes:
- Modifications in simulation phases (3.5.0 Simulation phases) resulting in a change of the recording phase (after the End Of Time Step execution) for the Function manager. Current Day can now be set to Previous Day and the recorded value will be the one calculated at the end of the Previous Day; and
- A new time of evaluation option in the Function Manager for resource assessment purposes: Post Flow phase.
- Implementation of the bulk licensing node, which allows representation of separate bulk licensing for irrigation districts and companies;
- Minimum flow requirements can now be input as rule curves, providing users the ability to manage releases;
- The ability to load command-line plugins directly, along with improved error reporting;
- A variety of changes in resource assessment, specifically undertaken for MDBA and Victoria;
- The ability to reload data sets for scenarios, allowing increased control on data sources;
New community plugins
There are more community plugins available with this release. In particular:
Insight
To run Insight you also need to install the 32bit version of Microsoft Visual C++ 2010 Redistributable Package.
This is for both 32bit machines and 64bit machines since the library we use is 32bit: Microsoft Visual C++ 2010 Redistributable Package (x86).
Existing Issues
- Copy & paste functionality:
- does not copy expressions and
- copied scenarios fail to run when using marker routing for constituents
- Can now add constituents to a catchments model. Save and re-load project - it will not be able to run straight away.