Archived OpenModelica forums. Posting is disabled.

Alternative forums include GitHub discussions or StackOverflow (make sure to read the Stack Overflow rules; you need to have well-formed questions)


Forgot password? | Forgot username? | Register
Page Start Prev 1 Next End

Exported FMU returns wrong results

#2663  Permalink Top

Exported FMU returns wrong results

Sep-17-12 11:29:17

Hello,

I'm testing a simulation tool for FMI Models. I created multiple Models via OpenModelica to test basic functionality.
Somehow the exported FMU returns invalid results

The Model calculates the sum and product of the current time.

Code:


model OpenModelica_Simple_Add_Mult_Auto

  output Real sum(start = 0);

  output Real prod(start = 0);

algorithm

  sum:=time + time;

  prod:=time * time;

end OpenModelica_Simple_Add_Mult_Auto;

The Results (FMU exported by OpenModelica, version 1.8.1 r11643 AND 1.9.0 r12942)

Code:


Step 0000: sum: 0.000000; prod: 0.000000

Step 0001: sum: 0.000000; prod: 0.000000

Step 0002: sum: 0.000000; prod: 0.000000

Step 0003: sum: 0.020000; prod: 0.000100

Step 0004: sum: 0.040000; prod: 0.000400

Step 0005: sum: 0.060000; prod: 0.000900

Step 0006: sum: 0.080000; prod: 0.001600

Step 0007: sum: 0.100000; prod: 0.002500

Step 0008: sum: 0.120000; prod: 0.003600

Step 0009: sum: 0.140000; prod: 0.004900

Step 0010: sum: 0.160000; prod: 0.006400

The results using Dymola 2013 to export

Code:


Step 0000: sum: 0.000000; prod: 0.000000

Step 0001: sum: 0.020000; prod: 0.000100

Step 0002: sum: 0.040000; prod: 0.000400

Step 0003: sum: 0.060000; prod: 0.000900

Step 0004: sum: 0.080000; prod: 0.001600

Step 0005: sum: 0.100000; prod: 0.002500

Step 0006: sum: 0.120000; prod: 0.003600

Step 0007: sum: 0.140000; prod: 0.004900

Step 0008: sum: 0.160000; prod: 0.006400

Step 0009: sum: 0.180000; prod: 0.008100

Step 0010: sum: 0.200000; prod: 0.010000

The results using OpenModelica somehow lags behind.

Does anyone have the same problem?

#2668  Permalink Top

Re: Exported FMU returns wrong results

Sep-18-12 18:56:39

Is this really an error?

Counting the number of evaluations of the system, which you seem to be doing, is not they way to extract your data from the FMU. You'd rather want to match the time series with the variable time.

For more complex FMUs you'll notice that Dymola also will evaluate the system at "intermediate" steps.

/patrik

Page Start Prev 1 Next End
Previous topic
Next topic
There are 0 guests and 0 other users also viewing this topic
You are here: