- pandapower.grid_equivalents.get_equivalent.get_equivalent(net, eq_type, boundary_buses, internal_buses, return_internal=True, show_computing_time=False, ward_type='ward_injection', adapt_va_degree=False, calculate_voltage_angles=True, allow_net_change_for_convergence=False, retain_original_internal_indices=False, runpp_fct=<function _runpp_except_voltage_angles>, **kwargs)
This function calculates and implements the rei or ward/xward network equivalents.
Power flow results in the given pandapower net are mandatory.
Known REI equivalents problems:
shift_degree != 0 of trafos and trafo3ws lead to errors or wrong results
despite ‘adapt_va_degree’, va_degree at the slack bus showed deviations within large grids
with large, real grids small deviations in the power flow results occured, in small grids the results fit perfectly
net - The pandapower network including power flow results will not be changed during this function.
eq_type (string) - type of the eqvalent network
The following methods are available:
“rei”: The idea of rei equivalent is to aggregate the power and current injection of the external buses to one or more fictitious radial, equivalent and independent (rei) nodes. There are three types of the rei-node in this routine, i.e. the reserved load, the reserved sgen and the reserved gen (also ext_grid). According to the demand, these elements (load, sgen and gen) are considered in the function “_create_net_zpbn” integrated or separately.
“ward”: The ward-type equivalent represents the external network with some equivalent impedance, shunts and power injections at boundary buses. The equivalent power injections represent the power of the elements (load, sgen and gen), which are removed by the equivalent calculation.
“xward”: The xward equivalent is an extended variation of the ward equivalent. Addition to the ward-representation, a fictitious PV node (generator) is added with zero active power injection at every boundary bus. The voltage of the PV node is set according to the boundary bus voltage.
ward/xward has two mothods to develop an equivalent circuits, i.e. the injection method and the admittance method. In the admittance method, all the bus power injections in the external networks are converted into shunt admittance before network equivalent. That is the only difference between these two methods. With the help of the function “adapt_net”, these methods are configurable.
boundary_buses (iterable) - list of boundary bus indices, by which the original network are divided into two networks, i.e. an internal network and an external network.
internal_buses (iterable) - list of bus indices, which are within the internal network. The program will add buses which are connected to this list of internal buses without passing boundary buses. If ‘internal_buses’ is an empty list or None, the whole grid is treated as external network.
return_internal (bool, True) - Reservation of the internal network
If True, the internal network is reserved in the final equivalent network; otherwise only the external network is the output.
show_computing_time (bool, False) - show computing time of each step
ward_type (str, “ward_injection”) - Type of ward and xward
ward and xward proposed two mothods called the Ward Injection method and the Ward Admittance method to develop equivalent circuits. The only difference between these methods is that in the Ward Admittance method, all bus power in the external networks are converted into shunt admittances before network equivalent.
adapt_va_degree (bool, None) - if True, in add_ext_grids_to_boundaries(), the va_degree of the additional ext_grids (if needed) at the boundary buses will be increased or decreased to values that minimize the difference to the given res_bus.va_degree values.
allow_net_change_for_convergence (bool, False) - if the net doesn’t converge at the first internal power flow, which is in add_ext_grids_to_boundaries(), and this parameter is True, the code tests if changes to unusual impedance values solve the divergence issue.
calculate_voltage_angles (bool, True) - parameter passed to internal runpp() runs.
retain_original_internal_indices (bool, False) - if True, the element indices in the internal net are retained; otherwise the indices will be reordered from 0.
**kwargs - key word arguments, such as sgen_separate, load_separate, gen_separate, group_name.
net_eq - The equivalent network in pandapower format