Merge branch 'develop' of gitlab:pp/pandapower into rbolgaryn

nomorestory · Apr 23, 2018 · 68c13c7 · 68c13c7
2 parents 239587c + 0c1cfc2
commit 68c13c7
Show file tree

Hide file tree

Showing 9 changed files with 140 additions and 78 deletions.
diff --git a/README.rst b/README.rst
@@ -79,18 +79,20 @@ There is a project to develop a GUI for pandapower: https://github.com/johaack/p
 Citing pandapower
 ==================
 
-A paper describing pandapower has been accepted for publication in IEEE Transaction on Power Systems, a preprint of this paper is available on `arXiv <https://arxiv.org/abs/1709.06743>`_ Please acknowledge the usage of pandapower by citing the Paper as follows:
+A paper describing pandapower has been accepted for publication in IEEE Transaction on Power Systems, a preprint of this paper is available on `arXiv <https://arxiv.org/abs/1709.06743>`_. Please acknowledge the usage of pandapower by citing the Paper as follows:
 
-- **L. Thurner, A. Scheidler, F. Schäfer et al.**, “pandapower - an Open Source Python Tool for Convenient Modeling, Analysis and Optimization of Electric Power Systems,” IEEE Transaction on Power Systems (to be published), 2018.
+- **L. Thurner, A. Scheidler, F. Schäfer et al**, `pandapower - an Open Source Python Tool for Convenient Modeling, Analysis and Optimization of Electric Power Systems <https://arxiv.org/abs/1709.06743>`_, IEEE Transactions on Power Systems, `DOI:10.1109/TPWRS.2018.2829021 <https://doi.org/10.1109/TPWRS.2018.2829021>`_, 2018.
 
 You can use the following BibTex entry: ::
 
-   @article{pandapower,
-	author = {{Thurner}, L. and {Scheidler}, A. and {Sch{\"a}fer}, F. and {Menke}, J.-H. and {Dollichon}, J. and {Meier}, F. and {Meinecke}, S. and {Braun}, M.},
-	title = "{pandapower - an Open Source Python Tool for Convenient Modeling, Analysis and Optimization of Electric Power Systems}",
-	year = 2018,
-	journal = {IEEE Transaction on Power Systems (to be published)},
-	url = {https://arxiv.org/abs/1709.06743}
+	@ARTICLE{pandapower.2018,
+	author={L. Thurner and A. Scheidler and F. Schafer and J. H. Menke and J. Dollichon and F. Meier and S. Meinecke and M. Braun},
+	journal={IEEE Transactions on Power Systems},
+	title={pandapower - an Open Source Python Tool for Convenient Modeling, Analysis and Optimization of Electric Power Systems},
+	year={2018},
+	doi={10.1109/TPWRS.2018.2829021},
+	url={https://arxiv.org/abs/1709.06743},
+	ISSN={0885-8950}
 	}
 
 Minimal Example

diff --git a/doc/about/citing.rst b/doc/about/citing.rst
@@ -1,13 +1,18 @@
 Citing pandapower
 ==================
 
-We are currently working on publishing a paper about pandapower, which is available as a `preprint on arxiv <https://arxiv.org/abs/1709.06743>`_.
-Until the paper has passed peer-review, please acknowledge the usage of pandapower in your paper as follows: ::
-
-    @online{pandapower,
-       author = {{Thurner}, L. and {Scheidler}, A. and {Sch{\"a}fer}, F. and {Menke}, J.-H. and {Dollichon}, J. and {Meier}, F. and {Meinecke}, S. and {Braun}, M.},
-       title = "{pandapower - an Open Source Python Tool for Convenient Modeling, Analysis and Optimization of Electric Power Systems}",
-       year = 2017,
-       url = {https://arxiv.org/abs/1709.06743},
-       note = {preprint}
-    }
+A paper describing pandapower has been accepted for publication in IEEE Transaction on Power Systems, a preprint of this paper is available on `arXiv <https://arxiv.org/abs/1709.06743>`_. Please acknowledge the usage of pandapower by citing the Paper as follows:
+
+- **L. Thurner, A. Scheidler, F. Schäfer et al**, `pandapower - an Open Source Python Tool for Convenient Modeling, Analysis and Optimization of Electric Power Systems <https://arxiv.org/abs/1709.06743>`_, IEEE Transactions on Power Systems, `DOI:10.1109/TPWRS.2018.2829021 <https://doi.org/10.1109/TPWRS.2018.2829021>`_, 2018.
+
+You can use the following BibTex entry: ::
+
+	@ARTICLE{pandapower.2018,
+	author={L. Thurner and A. Scheidler and F. Schafer and J. H. Menke and J. Dollichon and F. Meier and S. Meinecke and M. Braun},
+	journal={IEEE Transactions on Power Systems},
+	title={pandapower - an Open Source Python Tool for Convenient Modeling, Analysis and Optimization of Electric Power Systems},
+	year={2018},
+	doi={10.1109/TPWRS.2018.2829021},
+	url={https://arxiv.org/abs/1709.06743},
+	ISSN={0885-8950}
+	}
diff --git a/doc/elements/storage.png b/doc/elements/storage.png
diff --git a/doc/elements/storage.psd b/doc/elements/storage.psd
diff --git a/doc/elements/storage.rst b/doc/elements/storage.rst
@@ -16,33 +16,33 @@ Input Parameters
 *net.storage*
 
 .. tabularcolumns:: |p{0.10\linewidth}|p{0.10\linewidth}|p{0.25\linewidth}|p{0.40\linewidth}|
-.. csv-table:: 
+.. csv-table::
    :file: storage_par.csv
    :delim: ;
    :widths: 10, 10, 25, 40
 
 .. |br| raw:: html
 
    <br />
-   
+
 \*necessary for executing a power flow calculation |br| \*\*optimal power flow parameter
 
-   
+
 Electric Model
 =================
 
 Storages are modelled as PQ-buses in the power flow calculation:
 
-.. image:: pq.png
+.. image:: storage.png
 	:width: 8em
 	:alt: alternate Text
 	:align: center
-    
+
 The PQ-Values are calculated from the parameter table values as:
 
 .. math::
    :nowrap:
-   
+
    \begin{align*}
     P_{storage} &= p\_kw \cdot scaling \\
     Q_{storage} &= q\_kvar \cdot scaling \\
@@ -53,15 +53,15 @@ The PQ-Values are calculated from the parameter table values as:
    Since all power values are given in the consumer system, negative power models charging and positive activee power models discharging.
 
 .. note::
-    
+
     The apparent power value sn_kva, state of charge soc and storage capacity max_e_kwh are provided as additional information for usage in controller or other applications based on panadapower. It is not considered in the power flow!
 
 Result Parameters
 ==========================
 *net.res_storage*
 
 .. tabularcolumns:: |p{0.10\linewidth}|p{0.10\linewidth}|p{0.50\linewidth}|
-.. csv-table:: 
+.. csv-table::
    :file: storage_res.csv
    :delim: ;
    :widths: 10, 10, 50

diff --git a/pandapower/create.py b/pandapower/create.py
@@ -61,7 +61,7 @@ def create_empty_network(name="", f_hz=50., sn_kva=1e3):
                     ("p_kw", "f8"),
                     ("q_kvar", "f8"),
                     ("sn_kva", "f8"),
-                    ("soc", "f8"),
+                    ("soc_percent", "f8"),
                     ("min_e_kwh", "f8"),
                     ("max_e_kwh", "f8"),
                     ("scaling", "f8"),
@@ -267,7 +267,7 @@ def create_empty_network(name="", f_hz=50., sn_kva=1e3):
                             ("q_kvar", "f8")],
         "_empty_res_storage": [("p_kw", "f8"),
                                ("q_kvar", "f8"),
-                               ("soc", "f8")],
+                               ("soc_percent", "f8")],
         "_empty_res_gen": [("p_kw", "f8"),
                            ("q_kvar", "f8"),
                            ("va_degree", "f8"),
@@ -792,10 +792,10 @@ def create_sgen_from_cosphi(net, bus, sn_kva, cos_phi, mode, **kwargs):
     return create_sgen(net, bus, sn_kva=sn_kva, p_kw=p_kw, q_kvar=q_kvar, **kwargs)
 
 
-def create_storage(net, bus, p_kw, max_e_kwh, q_kvar=0, sn_kva=nan, soc=nan, min_e_kwh=0.0,
+def create_storage(net, bus, p_kw, max_e_kwh, q_kvar=0, sn_kva=nan, soc_percent=nan, min_e_kwh=0.0,
                    name=None, index=None, scaling=1., type=None, in_service=True, max_p_kw=nan,
                    min_p_kw=nan, max_q_kvar=nan, min_q_kvar=nan, controllable = nan):
-    """create_storage(net, bus, p_kw, max_e_kwh, q_kvar=0, sn_kva=nan, soc=nan, min_e_kwh=0.0, \
+    """create_storage(net, bus, p_kw, max_e_kwh, q_kvar=0, sn_kva=nan, soc_percent=nan, min_e_kwh=0.0, \
                    name=None, index=None, scaling=1., type=None, in_service=True, max_p_kw=nan, \
                    min_p_kw=nan, max_q_kvar=nan, min_q_kvar=nan, controllable = nan)
     Adds a storage to the network.
@@ -828,7 +828,7 @@ def create_storage(net, bus, p_kw, max_e_kwh, q_kvar=0, sn_kva=nan, soc=nan, min
 
         **sn_kva** (float, default None) - Nominal power of the storage
 
-        **soc** (float, NaN) - The state of charge of the storage
+        **soc_percent** (float, NaN) - The state of charge of the storage
 
         **min_e_kwh** (float, 0) - The minimum energy content of the storage \
             (minimum charge level)
@@ -863,7 +863,7 @@ def create_storage(net, bus, p_kw, max_e_kwh, q_kvar=0, sn_kva=nan, soc=nan, min
         **index** (int) - The unique ID of the created storage
 
     EXAMPLE:
-        create_storage(net, 1, p_kw = -30, max_e_kwh = 60, soc = 1.0, min_e_kwh = 5)
+        create_storage(net, 1, p_kw = -30, max_e_kwh = 60, soc_percent = 1.0, min_e_kwh = 5)
 
     """
     if bus not in net["bus"].index.values:
@@ -879,9 +879,9 @@ def create_storage(net, bus, p_kw, max_e_kwh, q_kvar=0, sn_kva=nan, soc=nan, min
     dtypes = net.storage.dtypes
 
     net.storage.loc[index, ["name", "bus", "p_kw", "q_kvar", "sn_kva", "scaling",
-                            "soc", "min_e_kwh", "max_e_kwh", "in_service", "type"]] = \
+                            "soc_percent", "min_e_kwh", "max_e_kwh", "in_service", "type"]] = \
         [name, bus, p_kw, q_kvar, sn_kva, scaling,
-         soc, min_e_kwh, max_e_kwh, bool(in_service), type]
+         soc_percent, min_e_kwh, max_e_kwh, bool(in_service), type]
 
     # and preserve dtypes
     _preserve_dtypes(net.storage, dtypes)

diff --git a/pandapower/opf/make_objective.py b/pandapower/opf/make_objective.py
@@ -8,6 +8,7 @@
 # and Energy System Technology (IEE), Kassel. All rights reserved.
 
 
+import pandas as pd
 from numpy import zeros, array, concatenate, power
 from pandapower.idx_cost import MODEL, NCOST, COST
 
@@ -77,50 +78,55 @@ def _make_objective(ppci, net):
         ppci["gencost"] = zeros((len_gencost, 4 + n_coefficients), dtype=float)
         ppci["gencost"][:, MODEL:COST] = array([2, 0, 0, n_coefficients])
 
-        for cost_type in ["p", "q"]:
-
-            if cost_type == "q":
-                shift_idx = ng
-                sign_corr = -1
-            else:
-                shift_idx = 0
-                sign_corr = 1
-
-            for el in ["gen", "sgen", "ext_grid", "load", "storage", "dcline"]:
-
-                if el == "gen":
-                    idx = gen_idx
-                elif el == "sgen":
-                    idx = sgen_idx
-                elif el == "ext_grid":
-                    idx = eg_idx
-                elif el == "load":
-                    idx = load_idx
-                elif el == "storage":
-                    idx = stor_idx
-                elif el == "dcline":
-                    idx = dcline_idx
+        if len(net.piecewise_linear_cost):
+            for cost_type in ["p", "q"]:
+                if (net.piecewise_linear_cost.type == cost_type).any():
+                    costs = net.piecewise_linear_cost[net.piecewise_linear_cost.type == cost_type].reset_index(
+                        drop=True)
 
-                if len(net.piecewise_linear_cost):
+                    if cost_type == "q":
+                        shift_idx = ng
+                        sign_corr = -1
+                    else:
+                        shift_idx = 0
+                        sign_corr = 1
 
-                    if (net.piecewise_linear_cost.type == cost_type).any():
-                        costs = net.piecewise_linear_cost[
-                                net.piecewise_linear_cost.type == cost_type].reset_index(drop=True)
+                    # for element types with costs defined
+                    for el in pd.unique(costs.element_type):
 
                         if not costs.element[costs.element_type == el].empty:
 
+                            if el == "gen":
+                                idx = gen_idx
+                            elif el == "sgen":
+                                idx = sgen_idx
+                            elif el == "ext_grid":
+                                idx = eg_idx
+                            elif el == "load":
+                                idx = load_idx
+                            elif el == "storage":
+                                idx = stor_idx
+                            elif el == "dcline":
+                                idx = dcline_idx
+
                             # cost data to write into gencost
-                            el_is = net[el].loc[(net[el].in_service) & net[el].index.isin(
+                            # (only write cost data of controllable and in service elements)
+                            if el == "ext_grid" or el == "dcline":
+                                el_is = net[el].loc[net[el].in_service & net[el].index.isin(
+                                    costs.loc[costs.element_type == el].element)].index
+                            else:
+                                el_is = net[el].loc[net[el].controllable & net[el].in_service & net[el].index.isin(
                                     costs.loc[costs.element_type == el].element)].index
+
                             p = costs.loc[(costs.element_type == el) & (
                                     costs.element.isin(el_is))].p.reset_index(drop=True)
                             f = costs.loc[(costs.element_type == el) & (
                                     costs.element.isin(el_is))].f.reset_index(drop=True)
+
                             if len(p) > 0:
                                 p = concatenate(p)
                                 f = concatenate(f)
                                 # gencost indices
-
                                 elements = idx[el_is] + shift_idx
 
                                 ppci["gencost"][elements, COST:COST+n_piece_lin_coefficients:2] = p
@@ -140,8 +146,7 @@ def _make_objective(ppci, net):
         if len(net.polynomial_cost):
             for cost_type in ["p", "q"]:
                 if (net.polynomial_cost.type == cost_type).any():
-                    costs = net.polynomial_cost[net.polynomial_cost.type == cost_type].reset_index(
-                        drop=True)
+                    costs = net.polynomial_cost[net.polynomial_cost.type == cost_type].reset_index(drop=True)
 
                     if cost_type == "q":
                         shift_idx = ng
@@ -150,7 +155,8 @@ def _make_objective(ppci, net):
                         shift_idx = 0
                         sign_corr = 1
 
-                    for el in ["gen", "sgen", "ext_grid", "load", "storage", "dcline"]:
+                    # for element types with costs defined
+                    for el in pd.unique(costs.element_type):
 
                         if not costs.element[costs.element_type == el].empty:
                             if el == "gen":
@@ -166,8 +172,15 @@ def _make_objective(ppci, net):
                             if el == "dcline":
                                 idx = dcline_idx
 
-                            el_is = net[el].loc[(net[el].in_service) & net[el].index.isin(
-                                costs.loc[costs.element_type == el].element)].index
+                            # cost data to write into gencost
+                            # (only write cost data of controllable and in service elements)
+                            if el == "ext_grid" or el == "dcline":
+                                el_is = net[el].loc[net[el].in_service & net[el].index.isin(
+                                    costs.loc[costs.element_type == el].element)].index
+                            else:
+                                el_is = net[el].loc[net[el].controllable & net[el].in_service & net[el].index.isin(
+                                    costs.loc[costs.element_type == el].element)].index
+
                             c = costs.loc[(costs.element_type == el) &
                                           (costs.element.isin(el_is))].c.reset_index(drop=True)
 

diff --git a/pandapower/test/opf/test_basic.py b/pandapower/test/opf/test_basic.py
@@ -578,17 +578,11 @@ def test_storage_opf():
                       min_p_kw=-25, max_q_kvar=25, min_q_kvar=-25)
     pp.create_sgen(net, b1, p_kw=-25, controllable=True, max_p_kw=0, min_p_kw=-25,
                    max_q_kvar=25, min_q_kvar=-25)
-    #pp.create_load(net, b1, p_kw=2.5, controllable=True, max_p_kw=2.5, min_p_kw=0,
-    #               max_q_kvar=2.5, min_q_kvar=-2.5)
     pp.create_load(net, b1, p_kw=25, controllable=True, max_p_kw=25, min_p_kw=0,
                    max_q_kvar=25, min_q_kvar=-25)
-
-
+
     # costs
     pp.create_polynomial_cost(net, 0, "ext_grid", np.array([0, 3, 0]))
-    #pp.create_polynomial_cost(net, 0, "load", np.array([0, -1, 0]))
-    # TODO - Gitlab Issue #27 - OPF error wenn Kosten für non-controllable PQ-Element zugewiesen
-    #   vgl. make_objective.py, Z.47ff.
     pp.create_polynomial_cost(net, 0, "sgen", np.array([0, 2, 0]))
     pp.create_polynomial_cost(net, 0, "storage", np.array([0, 1, 0]))
     pp.create_polynomial_cost(net, 1, "sgen", np.array([0, 1, 0]))
@@ -662,9 +656,57 @@ def test_storage_opf():
     assert np.isclose(res_cost_stor, res_cost_load)
 
 
+def test_opf_no_controllables_vs_pf():
+    """ Comparing the calculation results of PF and OPF in a simple network with non-controllable
+     elements """
+
+    # boundaries
+    vm_max = 1.3
+    vm_min = 0.9
+    max_line_loading_percent = 100
+
+    # create network
+    net = pp.create_empty_network()
+
+    b1 = pp.create_bus(net, vn_kv=0.4, max_vm_pu=vm_max, min_vm_pu=vm_min)
+    b2 = pp.create_bus(net, vn_kv=0.4, max_vm_pu=vm_max, min_vm_pu=vm_min)
+
+    pp.create_line(net, b1, b2, length_km=5, std_type="NAYY 4x50 SE",
+                   max_loading_percent=max_line_loading_percent)
+
+    # test elements static
+    pp.create_ext_grid(net, b2)
+    pp.create_load(net, b1, p_kw=7.5, controllable=False)
+    pp.create_sgen(net, b1, p_kw=-25, controllable=False, max_p_kw=-10, min_p_kw=-25,
+                   max_q_kvar=25, min_q_kvar=-25)
+
+    # testing cost assignment (for non-controllable elements - see Gitlab Issue #27)
+    pp.create_polynomial_cost(net, 0, "ext_grid", np.array([0, 3, 0]))
+    pp.create_polynomial_cost(net, 0, "load", np.array([0, -3, 0]))
+    pp.create_polynomial_cost(net, 0, "sgen", np.array([0, 2, 0]))
+
+    # do calculations
+    pp.runopp(net, verbose=False)
+    assert net["OPF_converged"]
+
+    res_opf_line_loading = net.res_line.loading_percent
+    res_opf_bus_voltages = net.res_bus.vm_pu
+
+    pp.runpp(net, verbose=False)
+    assert net["converged"]
+
+    res_pf_line_loading = net.res_line.loading_percent
+    res_pf_bus_voltages = net.res_bus.vm_pu
+
+    # assert calculation behaviour
+    assert np.isclose(res_opf_line_loading, res_pf_line_loading).all()
+    assert np.isclose(res_opf_bus_voltages, res_pf_bus_voltages).all()
+
+
 if __name__ == "__main__":
-    pytest.main(['-s', __file__])
+    #pytest.main(['-s', __file__])
     #test_storage_opf()
+    test_opf_no_controllables_vs_pf()
     #test_opf_varying_max_line_loading()
      # pytest.main(["test_basic.py", "-s"])
     # test_simplest_dispatch()