Merge pull request firemodels#13285 from cxp484/master

Reduce number of timesteps to write in Cantera output;Add Cantera output csv.
mcgratta · Aug 6, 2024 · 4eb777b · 4eb777b
2 parents f51591a + 3d3da13
commit 4eb777b
Show file tree

Hide file tree

Showing 2 changed files with 249 additions and 5 deletions.
diff --git a/Utilities/Python/scripts/cantera_ignition_delay.py b/Utilities/Python/scripts/cantera_ignition_delay.py
@@ -20,9 +20,9 @@
 equivRatios = np.array([0.6, 1.0,  1.4])
 T = np.array([900, 1000, 1100, 1200])
 
-
 csvdata = pd.DataFrame()
 caseCount = 0
+writeInterval = 20
 
 for phi in equivRatios:
     estimated_ignition_delay_times = np.ones_like(T, dtype=float)
@@ -38,6 +38,7 @@
     for i, state in enumerate(ignition_delays):
         caseCount = caseCount +1
         stateArr = []
+        stateArrReduced = []
         # Setup the gas and reactor
         gas.TPX = state.TPX
         r = ct.IdealGasReactor(contents=gas, name="Batch Reactor")
@@ -54,17 +55,29 @@
             time_history.append(t)
             reference_species_history.append(gas[reference_species].X[0])
             stateArr.append([t, gas[reference_species].Y[0],gas.T-273.15 ]) 
-
         i_ign = np.array(reference_species_history).argmax()
         tau = time_history[i_ign]
         t1 = time.time()
-        stateArrDF = pd.DataFrame(stateArr)
+
+        # Reduce number of timesteps to reduce file size
+        # Before ignition delay+0.1 s write after every writeInterval variable 
+        # After that write all the times.
+        tCount = 0
+        for j in range(len(stateArr)):
+            t= stateArr[j][0]
+            if (t < tau + 0.1):
+                if (tCount%writeInterval ==0):
+                    stateArrReduced.append(stateArr[j])
+            else:
+                stateArrReduced.append(stateArr[j])
+            tCount = tCount+1
+        stateArrDF = pd.DataFrame(stateArrReduced)
 
         caseIndx = str(caseCount)
         if caseCount ==1:
             csvdata = stateArrDF
         else:    
-            # Pad the smaller DataFrame with NaNs to match the number of rows
+            #Pad the smaller DataFrame with NaNs to match the number of rows
             if csvdata.shape[0] > stateArrDF.shape[0]:
                 pad_rows = csvdata.shape[0] - stateArrDF.shape[0]
                 stateArrDF = pd.concat([stateArrDF, pd.DataFrame(np.full((pad_rows, stateArrDF.shape[1]), np.nan))], axis=0, ignore_index=True)
@@ -78,11 +91,20 @@
 
         ignition_delays[i].tau = tau
 
-# Set Dataframe column headers
+# Set Dataframe column headers and writing format
+timeFormat = '{:10.5f}'
+OHFormat = '{:.3E}'
+TMPFormat = '{:10.2f}'
 colPerCase = 3 # Time, OH, TMP
 for i in range(caseCount):
     caseIndx = str(i+1)
     csvdata = csvdata.rename(columns={i*colPerCase: 'Time'+caseIndx, i*colPerCase+1: 'OH'+caseIndx, i*colPerCase+2: 'TMP'+caseIndx})
+    csvdata['Time'+caseIndx] = csvdata['Time'+caseIndx].map(lambda x: timeFormat.format(x))
+    csvdata['OH'+caseIndx] = csvdata['OH'+caseIndx].map(lambda x: OHFormat.format(x))
+    csvdata['TMP'+caseIndx] = csvdata['TMP'+caseIndx].map(lambda x: TMPFormat.format(x))
 
+csvdata = csvdata.apply(lambda x: x.str.strip() if x.dtype == 'object' else x)
+csvdata.replace('nan','',inplace=True)
+csvdata.replace('NAN','',inplace=True)
 csvdata.to_csv(Chemistry_DIR+"cantera_ignition_delay.csv",index=False)