Merge pull request e2nIEE#2324 from KS-HTK/issues/2321

solution for merge_nets issue 2321

CHANGELOG.rst

@@ -37,6 +37,7 @@ Change Log
 - [FIXED] PowerFactory converter: fix trafo3w tap dependent impedance
 - [ADDED] PowerFactory converter: support load types (constI, constZ) and the setting whether to consider voltage dependency of loads
 - [FIXED] deprecation of matplotlib.cm.get_cmap(name) -> matplotlib.colormaps[name]
+- [FIXED] merge_nets failing if net2 has custom DataFrame that is not present in net1
 
 [2.14.7] - 2024-06-14
 -------------------------------

pandapower/toolbox/grid_modification.py

@@ -224,6 +224,8 @@ def _merge_nets(net1, net2, validate=True, merge_results=True, tol=1e-9,
     # reindex net2 elements if some indices already exist in net
     reindex_lookup = dict()
     for elm_type in elm_types:
+        if elm_type not in net:
+            continue
         is_dupl = pd.Series(net2[elm_type].index).isin(net[elm_type].index)
         if any(is_dupl):
             start = max(net1[elm_type].index.max(), net2[elm_type].index[~is_dupl].max()) + 1
@@ -240,9 +242,12 @@ def _merge_nets(net1, net2, validate=True, merge_results=True, tol=1e-9,
 
     # copy dataframes from net2 to net (output)
     for elm_type in elm_types:
-        dtypes = net[elm_type].dtypes
-        net[elm_type] = pd.concat([net[elm_type], net2[elm_type]])
-        _preserve_dtypes(net[elm_type], dtypes)
+        if elm_type in net:
+            dtypes = net[elm_type].dtypes
+            net[elm_type] = pd.concat([net[elm_type], net2[elm_type]])
+            _preserve_dtypes(net[elm_type], dtypes)
+        else:
+            net[elm_type] = net2[elm_type].copy()
 
     # copy standard types of net by data of net2
     for type_ in net.std_types.keys():