nTopology 5.34.3

Parameter Optimization Ribbon
- Parameter optimization, also known as inverse design, determines the set of parameters that yield an optimal outcome for a specified objective function, subject to defined constraints.
- This new feature is available in its own Parameter Optimization ribbon, which gathers all the necessary blocks to set up and run an optimization study.


Parameter Optimization
- The Parameter Optimization block runs an optimization algorithm to find the optimal collection of Design Parameters satisfying the Objective while satisfying all Constraints.
- Note: We recommend not making Parameter Optimization a variable to avoid a known issue of handling parameter group inputs while contained in a variable
- Location: Parameter Optimization > Optimization
- Inputs:
. Objective: The objective function for the optimization process.
. Constraint Type: The constraints to satisfy during the optimization process.
. Design Parameters: The collection of Parameters the algorithm will change in its search for the optimal solution.
. Algorithm: The specific optimization method to be used. This dictates how the block will search for the optimal solution.
. Max Iterations: The maximum number of iterations allowed during the optimization process.
. Tracked Parameters: Additional Parameter to be tracked and logged during the optimization process. This is useful for debugging and detailed analysis.
- Output: Parameter Optimization Result


Parameter Objective
- The Parameter Objective block specifies a goal for Parameter Optimization. The objective guides the optimizer on the desired outcome.
- Location: Parameter Optimization > Composition
- Inputs:
. Goal: Optimization goal (Minimize or Maximize).
. Parameter: Parameter used to define the optimization objective.
- Output: Parameter Objective


Parameter Constraint
- The Parameter Constraint block defines a condition that a Parameter must satisfy. Parameter Constraints can be used to enforce limits on the Parameter to determine a valid solution space.
- Location: Parameter Optimization > Composition
- Inputs:
. Parameter: Parameter to be constrained.
. Constraint Type: Constraint operator (Less Than or Greater Than).
. Value: Minimum or Maximum value for constraint.
- Output: Parameter Constraint Parameter Constraint
- Note: Use the Parameter Constraint Group block to combine more than one constraint.


Constant Parameter
- The Constant Parameter block creates a parameter with a fixed value. It is used to define a constant design variable that does not change during optimization, serving as a placeholder for a specific, unchanging value in an optimization problem.
- Location: Parameter Optimization > Composition
- Inputs:
. Label: Unique identifier of the parameter.
. Value: Default value of the parameter.
- Output: Parameter Constant Parameter
- Note: Use the Parameter Group block to combine more than one parameter.


Independent Parameter
- The Independent Parameter block creates an independent optimization parameter that can vary within optional bounds during optimization.
- Location: Parameter Optimization > Composition
- Inputs:
. Label: Unique identifier of the parameter.
. Default Value: Default value of the parameter.
. Lower Bound: Minimum allowed value of the parameter.
. Upper Bound: Maximum allowed value of the parameter.
- Output: Parameter Independent Parameter
- Note: Use the Parameter Group block to combine more than one parameter.


Dependent Parameter
- The Dependent Parameter block creates a Dependent Parameter whose value is computed from other Parameters using a specified Function.
- Location: Parameter Optimization > Composition
- Inputs:
. Label: Unique identifier of the parameter.
. Function Object: The Function Object evaluated to produce the output Parameter's value.
. Input Parameters: Parameters defining Function Object inputs. The group elements correspond to the Function Object inputs by Index.
- Output: Parameter Dependent Parameter
- Note: Use the Parameter Group block to combine more than one parameter.


Design Parameters from Optimization
- The Design Parameters from Optimization block extracts all design parameters at a specific iteration from a Parameter Optimization result.
- Location: Parameter Optimization > Utilities
- Inputs:
. Result: Result to extract the design parameters from.
. Iteration: The Iteration number from which the Parameter Group will be extracted. If not provided, the last iteration will be used.
- Output: Parameter Group


Evaluate Parameter
- The Evaluate Parameter block computes the value of a Dependent Parameter for a specified collection of input parameters with constant or known values.
- Location: Parameter Optimization > Utilities
- Inputs:
. Dependent Parameter: Dependent Parameter to be evaluated.
. Input Parameters: Inputs used to evaluate the Dependent Parameter. If no inputs are provided, default values assigned during composition will be used.
- Output: Parameterizable


Tracked Parameters from Optimization
- The Tracked Parameters from Optimization block extracts all Tracked parameters at a specific iteration from a Parameter Optimization result.
- Location: Parameter Optimization > Utilities
- Inputs:
. Result: Result to extract the tracked parameters from.
. Iteration: The Iteration number from which the Parameter Group will be extracted. If not provided, the last iteration will be used.
- Output: Parameter Group


Usage Improvements
- You can now add a Comment that is not associated with a block by right-clicking in the notebook and selecting "Add Comment". You can also right-click an existing comment and select "Detach Comment" to remove its association from a block.


Previously, a placeholder variable was required to add comments in the notebook.

- You can now visualize Table data as a 2D plot. To access this feature, click the View Data button in the Display tab of the Right Panel and select the Show Plot button. Use the dropdown menus for the X-axis and Y-axis, which correspond to the table headers, to choose your data.


Block Update
- We have updated the Extract Optimization History block to support Parameter Optimization as well.