OpenModelica中跨模型传递全局值并求解最小步长dx的实现方法问询

Absolutely, this is totally feasible in OpenModelica! Let’s walk through how to make this work, building on the progress you’ve already made with your staggered grid pipe models and global data storage setup. We’ll use Modelica’s inner/outer mechanism (a staple in the Fluid library for global system coordination) to pass data seamlessly between pipes and your central system model.

Step 1: Refine Your Global Data Storage Model #

First, update your global system model to:

• Store an array of dx values from each pipe
• Compute the minimum dx from that array

Here’s a simplified example of what that might look like:

model GlobalFluidSystem
  // Define inner variables so pipes can access them
  inner Real dx_min "Minimum step size across all pipes";
  inner Real dx_values[:] "Array to store dx from each individual pipe";
  
  equation
    // Calculate the minimum step size using Modelica's built-in min() function
    dx_min = min(dx_values);
end GlobalFluidSystem;

Step 2: Pass Pipe dx Values to the Global Array #

In each of your custom pipe models, add a reference to the global system and assign your computed dx value to the corresponding position in the global array. You’ll need a unique index for each pipe to avoid overwriting values:

model CustomStaggeredPipe
  // Unique index to identify this pipe in the global array
  parameter Integer pipeIndex(start=1) "Index for this pipe in global dx array";
  
  // Reference the global system using outer
  outer GlobalFluidSystem globalSystem;
  
  // Your existing variable for computed step size
  Real dx_i "Step size calculated for this pipe";
  
  // --- Your existing staggered grid dx calculation code here ---
  // (e.g., based on pipe length, flow conditions, etc.)
  
  equation
    // Pass this pipe's dx value to the global array
    globalSystem.dx_values[pipeIndex] = dx_i;
end CustomStaggeredPipe;

Step 3: Initialize the Global System with Correct Array Size #

In your top-level model, instantiate the global system and initialize the dx_values array to match the number of pipes you have. Then assign unique indices to each pipe instance:

model TopLevelFluidModel
  // Initialize the global dx array with length equal to number of pipes (2 here)
  inner GlobalFluidSystem globalSystem(dx_values = fill(0.0, 2));
  
  // Instantiate your custom pipes with unique indices
  CustomStaggeredPipe pipe1(pipeIndex = 1);
  CustomStaggeredPipe pipe2(pipeIndex = 2);
  
  // --- Connect your pipes, tanks, and other components here ---
end TopLevelFluidModel;

Step 4: Retrieve the Minimum dx in Each Pipe #

Now you can access the global dx_min value directly in each pipe model for your staggered grid setup. For example, if you need to use the minimum step size to adjust your grid discretization, just reference globalSystem.dx_min wherever you need it in the pipe’s code.

Key Notes to Avoid Issues #

• Index Uniqueness: Make sure every pipe has a unique pipeIndex that matches a valid position in the global dx_values array (no out-of-bounds indices!).
• Array Initialization: The initial fill(0.0, 2) in the global system sets up an array of length 2—adjust this number if you add more pipes later.
• Calculation Order: Ensure your dx_i calculation in the pipe completes before it’s assigned to the global array. If you’re using an algorithm section for dx calculation, you can still assign the value to the global array within that section.

内容的提问来源于stack exchange，提问作者Jonas