Modeling and Analysis
      Back to home
      1. Innoslate Help Center
      2. Walkthrough Series
      3. Modeling and Analysis

      Design of Experiment Use Case

      Learn how to create a Design of Experiment using Innoslate to analyze the change in design based on changes in variables.

      Introduction

      Design of Experiments (DOE) is a systematic method used in the engineering industry to determine the relationship between factors affecting a process and the output of that process. It involves planning, conducting, analyzing, and interpreting controlled tests to evaluate the factors that may influence a particular outcome. DOE is widely used for optimizing processes, improving product quality, and reducing variability.

      In the context of systems engineering, DOE plays a crucial role in the design, development, and optimization of complex systems. Systems engineering often involves integrating various subsystems and components, each with its own set of variables and parameters. DOE helps in systematically studying these variables to identify the optimal configurations and interactions that enhance system performance and reliability, based on the adjustments and change in variables that create optimal impacts.

      We aim to demonstrate the process through a real-world example of optimizing an electronic device's cooling system, highlighting the steps involved in setting up and analyzing experiments to achieve optimal cooling efficiency.

      To follow along and try out the features mentioned in this walkthrough, click here to instantly download the Design of Experiment Walkthrough .INNO File and import it into a new Innoslate project.

      Walkthrough Use Case

      Setup

      Within Innoslate, users have access to analytic features that can be utilized to enhance experiments and designs within your projects. In the example project shared below, we will delve into this by focusing on an electronic cooling system.

      In the design of an electronic cooling system, multiple factors such as fan speed, heat sink material, and thermal paste type can influence the overall cooling efficiency. By applying DOE, engineers can experiment with different combinations of these factors to determine the most effective setup. This approach not only leads to better-performing systems but also provides insights into the underlying mechanisms driving system behavior.

      DOEActionDiagram

      The Action Diagram as shown above, is structured to demonstrate the Design of Experiments (DOE) methodology for optimizing the cooling efficiency of an electronic device. The diagram consists of three distinct experimental setups, each with predefined values that can be changed in the script, and a final computation node to calculate the cooling efficiency.

      The Experiment Flow processes through the different variables being declared, calculates the cooling efficiency of the 1st design, and moves onto the next setup of variables to be analyzed. The following list details the defined variables in each of the action diagram entities.

      1. Design of Experiment 1 (Entity 1.1 to 1.4):

        • Node 1.1 (Fan Speed): Sets the fan speed to 2000 RPM.

        • Node 1.2 (Heat Sink Material): Sets the thermal conductivity to 150 W/mK.

        • Node 1.3 (Thermal Paste Type): Sets the thermal resistance to 0.1 K/W.

        • Node 1.4 (Final Computation DOE.1): Computes the cooling efficiency using the formula:

          Cooling Efficiency = (Fan Speed × Thermal Conductivity​) / Thermal Resistance

      2. Design of Experiment 2 (Entity 1.5 to 1.8):

        • Node 1.5 (Fan Speed): Sets the fan speed to 2500 RPM.

        • Node 1.6 (Heat Sink Material): Sets the thermal conductivity to 180 W/mK.

        • Node 1.7 (Thermal Paste Type): Sets the thermal resistance to 0.05 K/W.

        • Node 1.8 (Final Computation DOE.2): Computes the cooling efficiency using the given formula.

      3. Design of Experiment 3 (Entity 1.9 to 1.12):

        • Node 1.9 (Fan Speed): Sets the fan speed to 4000 RPM.

        • Node 1.10 (Heat Sink Material): Sets the thermal conductivity to 225 W/mK.

        • Node 1.11 (Thermal Paste Type): Sets the thermal resistance to 0.02 K/W.

        • Node 1.12 (Final Computation DOE.3): Computes the cooling efficiency using the given formula.

      The detailed set of variables were defined in the Action Diagram on their respective entity nodes. The following image shown below is the script for '1.1 Fan Speed (RPM)' entity. The other scripts that declare variables follow the similar format:

      VariableSetScript

      The script shown in the image is used to set the value of the fan speed (RPM) for the first design of the experiment. The onStart() function is executed at the beginning of the simulation to initialize the variable. It assigns a value of 2000 to the global variable Val1 using the globals.set('Val1', 2000) method. This ensures that the fan speed is consistently defined and used throughout the subsequent computations in the experiment.

      Similarly, the same format is applied to all other variable setting entities in the action diagram, ensuring consistent initialization of values for each experiment.

      Once the set of variables are set within each branch, there is a Final Computation DOE.1/2/3 entity that calculates the cooling efficiency through the simulation. The image below shows the script detail of '1.4 Final Computation DOE.1':

      DOECalculationScript

      The script for '1.4 Final Computation DOE.1' retrieves the previously set values for fan speed, heat sink material, and thermal paste type using the globals.get method. It then prints these values to the console for confirmation. Using these dimension parameters, it calculates the cooling efficiency with the formula (Val1 * Val2) / Val3 and prints the result to the console.

      Like mentioned previously, the same process happens for DOE.2 and DOE.3.

      Results

      Upon running the Action Diagram through the Discrete Event Simulator, the results of our Design of Experiments (DOE) for optimizing the cooling efficiency of an electronic device are displayed in the Global Tracker and Console panels. Each experiment's variables and computed cooling efficiency are recorded, providing clear insights into the performance outcomes.

      DOESimResults

      The results demonstrate that the second experiment, with a fan speed of 4000 RPM, a heat sink material thermal conductivity of 225 W/mK, and a thermal paste thermal resistance of 0.02 K/W, yielded the highest cooling efficiency of 45,000,000. This indicates that these parameter values are most effective in optimizing the cooling efficiency for the given setup.

      The first experiment, with lower initial values, produced a cooling efficiency of 3,000,000, while the third experiment, based on higher values, achieved a cooling efficiency of 45,000,000. This variation highlights the importance of carefully selecting the variables to achieve optimal system performance.

      Conclusion

      By utilizing the Discrete Event Simulator within Innoslate, a systematic testing of various configurations was conducted to identify the most efficient setup. This method not only enhanced the cooling system's performance but also provides valuable insights into the interactions of different variables.

      These results presented by Innoslate emphasize the significance of Design of Experiments (DOE) in engineering design. This method enables engineers to make informed decisions based on data, ultimately improving system performance and reliability

       

      To continue learning about Modeling and Simulation, Click Here.

      (Next Article: Failure Modes & Effects Analysis Use Case)