6 novembre 2018

Optional add-ons

Saber Solutions software is composed of add-on modules that can be added to the basic package in order to improve software performance (RunTime Modules), to increase the choice of components (components library) and to make more specific analysis (inspecs)

 

 

Runtime module

Saber Runtime parallelizes and distributes iterative simulations across computing resources to dramatically improve throughput and reduce simulation time.
The Saber Runtime library works with Saber solutions, Saber Functional Safety Add-on, and Saber Inspecs Add-on to minimize the time spent running valuable performance and safety simulations. With Saber Runtime, engineering and safety teams can simulate thousands of scenarios that would otherwise have been too costly to perform. 

Robust design methodologies require advanced sensitivity and statistical analyses to verify the reliability of complex power networks. These analyses are recursive simulations requiring hundreds or thousands of runs which is impractical to support on a single CPU.

The Saber environment solves this problem by distributing iterated simulations across a compute grid allowing multiple CPUs to perform the analyses in much less time. When a simulation is complete, results are gathered into a single data file for easy processing.

Once the statistical setup is complete, the number of simulation runs required to produce statistically meaningful results must be determined. The number of runs depends on the complexity and characteristics of the design. Simulation runs on the order of several hundred to a few thousand are typically required. This represents a significant decrease in the time it takes to complete the subsystem design.

In conclusion, the RunTime module allows to improve throughput and reduce time with distributed computing. It parallelizes 1000s of simulations for Robust Design, Functional Safety or Inspecs for a maximum efficiency and a considerable savings in development resources and a faster time to market.

Components library

SaberRD has many predefined but you can improve them with a complementary library to enhance your work and performance with an extented manufacturer library of around 30,000 more components than the basic package. Aircraft and automotive design requires accurate model libraries and advanced modeling tools. Saber has the largest library of mixed-technology models in the industry. Models are available at various levels of abstraction, from high-level idealized transfer functions to precise, physics-based devices. These models are optimized for accuracy and performance.

Inspecs

Optional Saber InSpecs to optimize the design for component variations and shifts in operating conditions.

Monte Carlo

  • General Description:Monte Carlo (MC) analysis randomly varies the parameters, within user-defined tolerance ranges, and executes the specified SaberRD analysis at each parameter value. By specifying the tolerance ranges on parts in the design, SaberRD can randomly vary specified parts allowing you to evaluate how the variance of part values in the production environment affects the performance of the design.
    The Monte Carlo command executes a set of SaberRD commands in a statistical environment, which has the effect of performing a series of Monte Carlo runs on the commands.

Mont Carlo specifications

  • Comments:
    • Saber changes component values every time the loop is executed
    • Large or small-signal analyses allowed
    • The user needs also to specify the tolerance of parameters in the design: for example : 10kohms resistor with 10% tolerance: rnom=normal(10k,0.1)

Normal corresponds to the probability density functions.

Monte Carlo results with Resistor 10% tolerance

    • MC Analysis results are useful for spotting trends or correlation between a given measure and a component parameter.

 

As there is a large amount of data, user should use Pareto analysis which allows to rank order the parameters that have the biggest effect on the variance of the design performance measure. 

    • Pareto results provide correlation data about the impact of tolerance on the design performance measure

Pareto results

Sensitivity (Identify parameters that impact performance)

  • General Description:

The sensitivity analysis is used to determine how sensitive a specified design performance measurement varies in design parameters. A performance measurement is a single numeric characteristic of the circuit (e.g. DC voltage, risetime, bandwidth, etc.)

  • Required Parameters:
    • Parameter List – specifies the design parameters to be varied (perturbed)
    • Add Analyses/Measures – specifies which analyses/measures to use after each perturbation

  • Comments:
    • Saber changes each design parameter by a small amount and calculates the effect on the performance measure
    • Tolerance data not used or needed
  • Sensitivity calculations:

Where, p is the parameter value and m is the performance measure.

  • Sensitivity Report: 

After running a sensitivity analysis a report is generated to help identifying the parameters that have the highest influence on the measurements (output ripple voltage for example) : rank parameters by their impact on design performance measure.

Sensitivity Report

Sensibility process

Stress

  • General Description:

A stress analysis determines the part stress of selected components in the design. Part stress is based on the concept of safe operating areas (SOAs), each of which is a boundary for a particular property of the component. As an example, as the power dissipated in a component cannot exceed a certain rated value, the stress analysis determines the percentage of the actual power dissipation compared to the power rating.

Designers use stress to understand the part stress in their design. It is done based on large-signal analysis. The results of the stress report provide information about how to select parts used in the design.

                    

  • Comments:

Users need to specify the stress ratings for each component into the design and also enable the Stress options from Transient simulation settings.

After adding stress measures and successful completion of Transient/Stress analysis, Stress Report is generated as shown below. Stress Report specify the list of components in the ascending order of stress observed.

Stress Report

From the Stress Report, it is observed that:

  • All devices except boost diodes are stressed below 50%.
  • The voltage across boost diodes are 62% of its rated voltage 650V.

Therefore, Schottky diodes with higher PIV (peak inverse voltage), for example 800V with low forward voltage drop are recommended for this type of application.

Worst-case

  • General Description:

The SaberRD Worst-Case Analysis (WCA) tool is used to predict system reliability by determining the combination of allowable operating conditions under which an electronic or a mechtronic design will reach its worst-case limits of performance. The tool provides the user the ability to interactively set up, execute, and monitor tests that search for worst-case extremes, visualize results, annotate results back to the design, and output analysis data suitable for reporting and documentation.

Moreover, WCA is essential in fault-intolerant and safety-critical systems.

Also, sub-systems are complex, so the question is: how do you determine the minimums and maximums of important parameter values?

The Worst-Case Analysis tool is a framework providing functions to analyze the effects of design parameters defined over a permissible domain. It helps to identify the most critical design-behavior and design parameters by applying optimization algorithms

These functions are useful to:

  • Validate the performance of a design.
  • Identify the range attained by design performances of interest when stochastic parameters are varied within their tolerance range (worst-case analysis).
  • Identify combinations of design parameters leading to best design performance (design optimization).
  • Characterize parameters in order to best fit measured system component behavior (model characterization)

The SaberRD Worst-Case Analysis (WCA) Tool is fully integrated with the SaberRD design and verification environment.

Worst-Case Analysis Tool WCA Tool is displayed below :

 

As shown in the picture above :

  • WCA tool calculates WC limits
  • Significant deviation from nominal behavior (27.9V)
  • Parameters could be extracted automatically
  • Comments:

Purpose is to overcome the limitations of traditional methods

  • Build confidence in worst case results
  • Easy-to-use
  • Traditional methods have no guarantee to uncover worst-case behavior and Extreme cases may not be touched for certain designs contrary to WCA.

Salient features

  • Flexible definition of WCA objectives (e.g. Min, Max)
  • Re-use of existing design set up (e.g. MC tolerances)
  • Intuitive graphical user interface

Search Algorithms

  • Powerful local & global algorithms
  • Combined search possible to leverage synergy of different methods
  • Customized calibration possible

Test definition

  • Intuitive drag & drop solution
  • Analysis Definition
  • Measurements & objectives
  • Multi objective definition

 

Functional Safety

Fault analysis

Fault => Investigation of faults in context to function safety (support of ISO 26262).

  • General Description:

Fault analysis allows the user to define various fault conditions that a design could meet and to utilize the simulator to automatically step through them and analyze the behavior of the system for each fault condition.

You first define one or more faults for the design and then specify the analysis to perform for each fault. The specification of these analyses is similar to the body of other loop analysis, such as Monte Carlo and parameter sweep.

The benefits of this funcitonal safety analysis are :

  • Reduce costly test-bench fault experiments
  • Improve time to compliance
  • Simulate fault scenario that could not be tested before

Supported fault types

  • Comments:

This kind of analysis help to understand the impact of faults on system behavior and safety. You will be able to apply multiple faults to a system and test for on going operation in a safe range.

  • Process:

Fault analysis can be run using the command line or the GUI.

Fault Analysis Setup

After running the Fault analysis, The Pass/Fail report generated by the Experiment assists designers to interpret impact of faults on the design and take necessary actions on components affecting the design robustness.

Fault Analysis results

Back to top