Constraints

Constraint Templates

Constraint templates help simplify data wrangling across multiple Power Flow formulations by providing an abstraction layer between the network data and network constraint definitions. The constraint template's job is to extract the required parameters from a given network data structure and pass the data as named arguments to the Power Flow formulations.

These templates should be defined over AbstractPowerModel and should not refer to model variables. For more details, see the files: core/constraint_template.jl and core/constraint.jl (core/constraint_template.jl provides higher level APIs, and pulls out index information from the data dictionaries, before calling out to methods defined in core/constraint.jl).

Voltage Constraints

PowerModels.constraint_model_voltage — Function

This constraint captures problem agnostic constraints that are used to link the model's voltage variables together, in addition to the standard problem formulation constraints.

Notable examples include the constraints linking the voltages in the ACTPowerModel, constraints linking convex relaxations of voltage variables.

do nothing, most models to not require any model-specific voltage constraints

PowerModels.constraint_model_voltage_on_off — Function

This constraint captures problem agnostic constraints that are used to link the model's voltage variables together, in addition to the standard problem formulation constraints. The on/off name indicates that the voltages in this constraint can be set to zero via an indicator variable

Notable examples include the constraints linking the voltages in the ACTPowerModel, constraints linking convex relaxations of voltage variables.

do nothing, most models to not require any model-specific on/off voltage constraints

do nothing, this model does not have complex voltage constraints

do nothing, this model does not have complex voltage variables

PowerModels.constraint_ne_model_voltage — Function

This constraint captures problem agnostic constraints that are used to link the model's voltage variables together, in addition to the standard problem formulation constraints. The network expantion name (ne) indicates that the voltages in this constraint can be set to zero via an indicator variable

Notable examples include the constraints linking the voltages in the ACTPowerModel, constraints linking convex relaxations of voltage variables.

do nothing, most models to not require any model-specific network expansion voltage constraints

do nothing, this model does not have complex voltage constraints

do nothing, this model does not have complex voltage variables

Generator Constraints

PowerModels.constraint_gen_setpoint_active — Function

pg[i] == pg

pg[i] == pg

PowerModels.constraint_gen_setpoint_reactive — Function

qq[i] == qq

qq[i] == qq

do nothing, apo models do not have reactive variables

Bus Constraints

Setpoint Constraints

PowerModels.constraint_theta_ref — Function

reference bus angle constraint

t[ref_bus] == 0

nothing to do, no voltage angle variables

t[ref_bus] == 0

Do nothing, no way to represent this in these variables

PowerModels.constraint_voltage_magnitude_setpoint — Function

v[i] == vm

v[i] == vm

do nothing, this model does not have voltage variables

Power Balance Constraints

PowerModels.constraint_power_balance — Function

PowerModels.constraint_power_balance_ls — Function

nodal power balance with constant power factor load and shunt shedding

PowerModels.constraint_ne_power_balance — Function

Branch Constraints

Ohm's Law Constraints

PowerModels.constraint_ohms_yt_from — Function

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

p[f_idx] ==  (g+g_fr)/tm*v[f_bus]^2 + (-g*tr+b*ti)/tm^2*(v[f_bus]*v[t_bus]*cos(t[f_bus]-t[t_bus])) + (-b*tr-g*ti)/tm^2*(v[f_bus]*v[t_bus]*sin(t[f_bus]-t[t_bus]))
q[f_idx] == -(b+b_fr)/tm*v[f_bus]^2 - (-b*tr-g*ti)/tm^2*(v[f_bus]*v[t_bus]*cos(t[f_bus]-t[t_bus])) + (-g*tr+b*ti)/tm^2*(v[f_bus]*v[t_bus]*sin(t[f_bus]-t[t_bus]))

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

p[f_idx] == -b*(t[f_bus] - t[t_bus])

nothing to do, no voltage angle variables

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

PowerModels.constraint_ohms_yt_to — Function

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

p[t_idx] ==  (g+g_to)*v[t_bus]^2 + (-g*tr-b*ti)/tm^2*(v[t_bus]*v[f_bus]*cos(t[t_bus]-t[f_bus])) + (-b*tr+g*ti)/tm^2*(v[t_bus]*v[f_bus]*sin(t[t_bus]-t[f_bus]))
q[t_idx] == -(b+b_to)*v[t_bus]^2 - (-b*tr+g*ti)/tm^2*(v[t_bus]*v[f_bus]*cos(t[f_bus]-t[t_bus])) + (-g*tr-b*ti)/tm^2*(v[t_bus]*v[f_bus]*sin(t[t_bus]-t[f_bus]))

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

nothing to do, this model is symetric

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

PowerModels.constraint_ohms_y_from — Function

Creates Ohms constraints for AC models (y post fix indicates that Y values are in rectangular form)

p[f_idx] ==  (g+g_fr)*(v[f_bus]/tr)^2 + -g*v[f_bus]/tr*v[t_bus]*cos(t[f_bus]-t[t_bus]-as) + -b*v[f_bus]/tr*v[t_bus]*sin(t[f_bus]-t[t_bus]-as)
q[f_idx] == -(b+b_fr)*(v[f_bus]/tr)^2 + b*v[f_bus]/tr*v[t_bus]*cos(t[f_bus]-t[t_bus]-as) + -g*v[f_bus]/tr*v[t_bus]*sin(t[f_bus]-t[t_bus]-as)

PowerModels.constraint_ohms_y_to — Function

Creates Ohms constraints for AC models (y post fix indicates that Y values are in rectangular form)

p[t_idx] ==  (g+g_to)*v[t_bus]^2 + -g*v[t_bus]*v[f_bus]/tr*cos(t[t_bus]-t[f_bus]+as) + -b*v[t_bus]*v[f_bus]/tr*sin(t[t_bus]-t[f_bus]+as)
q[t_idx] == -(b+b_to)*v[t_bus]^2 +  b*v[t_bus]*v[f_bus]/tr*cos(t[f_bus]-t[t_bus]+as) + -g*v[t_bus]*v[f_bus]/tr*sin(t[t_bus]-t[f_bus]+as)

On/Off Ohm's Law Constraints

PowerModels.constraint_ohms_yt_from_on_off — Function

p[f_idx] == z*(g/tm*v[f_bus]^2 + (-g*tr+b*ti)/tm^2*(v[f_bus]*v[t_bus]*cos(t[f_bus]-t[t_bus])) + (-b*tr-g*ti)/tm^2*(v[f_bus]*v[t_bus]*sin(t[f_bus]-t[t_bus])))
q[f_idx] == z*(-(b+c/2)/tm*v[f_bus]^2 - (-b*tr-g*ti)/tm^2*(v[f_bus]*v[t_bus]*cos(t[f_bus]-t[t_bus])) + (-g*tr+b*ti)/tm^2*(v[f_bus]*v[t_bus]*sin(t[f_bus]-t[t_bus])))

-b*(t[f_bus] - t[t_bus] + vad_min*(1-z_branch[i])) <= p[f_idx] <= -b*(t[f_bus] - t[t_bus] + vad_max*(1-z_branch[i]))

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

p[f_idx] ==        g/tm*w_fr[i] + (-g*tr+b*ti)/tm*(wr[i]) + (-b*tr-g*ti)/tm*(wi[i])
q[f_idx] == -(b+c/2)/tm*w_fr[i] - (-b*tr-g*ti)/tm*(wr[i]) + (-g*tr+b*ti)/tm*(wi[i])

PowerModels.constraint_ohms_yt_to_on_off — Function

p[t_idx] == z*(g*v[t_bus]^2 + (-g*tr-b*ti)/tm^2*(v[t_bus]*v[f_bus]*cos(t[t_bus]-t[f_bus])) + (-b*tr+g*ti)/tm^2*(v[t_bus]*v[f_bus]*sin(t[t_bus]-t[f_bus])))
q[t_idx] == z*(-(b+c/2)*v[t_bus]^2 - (-b*tr+g*ti)/tm^2*(v[t_bus]*v[f_bus]*cos(t[f_bus]-t[t_bus])) + (-g*tr-b*ti)/tm^2*(v[t_bus]*v[f_bus]*sin(t[t_bus]-t[f_bus])))

nothing to do, this model is symetric

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

p[t_idx] ==        g*w_to[i] + (-g*tr-b*ti)/tm*(wr[i]) + (-b*tr+g*ti)/tm*(-wi[i])
q[t_idx] == -(b+c/2)*w_to[i] - (-b*tr+g*ti)/tm*(wr[i]) + (-g*tr-b*ti)/tm*(-wi[i])

PowerModels.constraint_ne_ohms_yt_from — Function

p_ne[f_idx] == z*(g/tm*v[f_bus]^2 + (-g*tr+b*ti)/tm^2*(v[f_bus]*v[t_bus]*cos(t[f_bus]-t[t_bus])) + (-b*tr-g*ti)/tm^2*(v[f_bus]*v[t_bus]*sin(t[f_bus]-t[t_bus])))
q_ne[f_idx] == z*(-(b+c/2)/tm*v[f_bus]^2 - (-b*tr-g*ti)/tm^2*(v[f_bus]*v[t_bus]*cos(t[f_bus]-t[t_bus])) + (-g*tr+b*ti)/tm^2*(v[f_bus]*v[t_bus]*sin(t[f_bus]-t[t_bus])))

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

p[f_idx] == g/tm*w_fr_ne[i] + (-g*tr+b*ti)/tm*(wr_ne[i]) + (-b*tr-g*ti)/tm*(wi_ne[i])
q[f_idx] == -(b+c/2)/tm*w_fr_ne[i] - (-b*tr-g*ti)/tm*(wr_ne[i]) + (-g*tr+b*ti)/tm*(wi_ne[i])

PowerModels.constraint_ne_ohms_yt_to — Function

p_ne[t_idx] == z*(g*v[t_bus]^2 + (-g*tr-b*ti)/tm^2*(v[t_bus]*v[f_bus]*cos(t[t_bus]-t[f_bus])) + (-b*tr+g*ti)/tm^2*(v[t_bus]*v[f_bus]*sin(t[t_bus]-t[f_bus])))
q_ne[t_idx] == z*(-(b+c/2)*v[t_bus]^2 - (-b*tr+g*ti)/tm^2*(v[t_bus]*v[f_bus]*cos(t[f_bus]-t[t_bus])) + (-g*tr-b*ti)/tm^2*(v[t_bus]*v[f_bus]*sin(t[t_bus]-t[f_bus])))

nothing to do, this model is symetric

Creates Ohms constraints (yt post fix indicates that Y and T values are in rectangular form)

p[t_idx] == g*w_to_ne[i] + (-g*tr-b*ti)/tm*(wr_ne[i]) + (-b*tr+g*ti)/tm*(-wi_ne[i])
q[t_idx] == -(b+c/2)*w_to_ne[i] - (-b*tr+g*ti)/tm*(wr_ne[i]) + (-g*tr-b*ti)/tm*(-wi_ne[i])

Current

PowerModels.constraint_current_balance — Function

Kirchhoff's current law applied to buses sum(cr + im*ci) = 0

PowerModels.constraint_power_magnitude_sqr — Function

p[f_idx]^2 + q[f_idx]^2 <= w[f_bus]/tm*cm[f_bus,t_bus]

PowerModels.constraint_power_magnitude_sqr_on_off — Function

p[arc_from]^2 + q[arc_from]^2 <= w[f_bus]/tm*cm[i]

PowerModels.constraint_power_magnitude_link — Function

cm[f_bus,t_bus] == (g^2 + b^2)*(w[f_bus]/tm + w[t_bus] - 2*(tr*wr[f_bus,t_bus] + ti*wi[f_bus,t_bus])/tm) - c*q[f_idx] - ((c/2)/tm)^2*w[f_bus]

PowerModels.constraint_power_magnitude_link_on_off — Function

ccm[f_bus,t_bus] == (g^2 + b^2)*(w[f_bus]/tm + w[t_bus] - 2*(tr*wr[f_bus,t_bus] + ti*wi[f_bus,t_bus])/tm) - c*q[f_idx] - ((c/2)/tm)^2*w[f_bus]

Thermal Limit Constraints

PowerModels.constraint_thermal_limit_from — Function

constraint_thermal_limit_from(pm::AbstractPowerModel, n::Int, i::Int)

Adds the (upper and lower) thermal limit constraints for the desired branch to the PowerModel.

p[f_idx]^2 + q[f_idx]^2 <= rate_a^2

[rate_a, p[f_idx], q[f_idx]] in SecondOrderCone

p[f_idx]^2 + q[f_idx]^2 <= rate_a^2

-rate_a <= p[f_idx] <= rate_a

PowerModels.constraint_thermal_limit_to — Function

p[t_idx]^2 + q[t_idx]^2 <= rate_a^2

[rate_a, p[t_idx], q[t_idx]] in SecondOrderCone

p[t_idx]^2 + q[t_idx]^2 <= rate_a^2

nothing to do, this model is symetric

PowerModels.constraint_thermal_limit_from_on_off — Function

p[f_idx]^2 + q[f_idx]^2 <= (rate_a * z_branch[i])^2

PowerModels.constraint_thermal_limit_to_on_off — Function

p[t_idx]^2 + q[t_idx]^2 <= (rate_a * z_branch[i])^2

nothing to do, this model is symetric

PowerModels.constraint_ne_thermal_limit_from — Function

p_ne[f_idx]^2 + q_ne[f_idx]^2 <= (rate_a * branch_ne[i])^2

PowerModels.constraint_ne_thermal_limit_to — Function

p_ne[t_idx]^2 + q_ne[t_idx]^2 <= (rate_a * branch_ne[i])^2

nothing to do, this model is symetric

Current Limit Constraints

Missing docstring.

Missing docstring for constraint_current_limit. Check Documenter's build log for details.

PowerModels.constraint_current_to — Function

Defines how current distributes over series and shunt impedances of a pi-model branch

PowerModels.constraint_current_from — Function

Defines how current distributes over series and shunt impedances of a pi-model branch

Phase Angle Difference Constraints

PowerModels.constraint_voltage_angle_difference — Function

Bounds the voltage angle difference between bus pairs

branch voltage angle difference bounds

t[f_bus] - t[t_bus] <= angmax
t[f_bus] - t[t_bus] >= angmin

nothing to do, no voltage angle variables

PowerModels.constraint_voltage_angle_difference_on_off — Function

angmin <= z_branch[i]*(t[f_bus] - t[t_bus]) <= angmax

angmin*z_branch[i] + vad_min*(1-z_branch[i]) <= t[f_bus] - t[t_bus] <= angmax*z_branch[i] + vad_max*(1-z_branch[i])

angmin*z_branch[i] + vad_min*(1-z_branch[i]) <= t[f_bus] - t[t_bus] <= angmax*z_branch[i] + vad_max*(1-z_branch[i])

angmin*wr[i] <= wi[i] <= angmax*wr[i]

PowerModels.constraint_ne_voltage_angle_difference — Function

angmin <= branch_ne[i]*(t[f_bus] - t[t_bus]) <= angmax

angmin*branch_ne[i] + vad_min*(1-branch_ne[i]) <= t[f_bus] - t[t_bus] <= angmax*branch_ne[i] + vad_max*(1-branch_ne[i])

angmin*branch_ne[i] + vad_min*(1-branch_ne[i]) <= t[f_bus] - t[t_bus] <= angmax*branch_ne[i] + vad_max*(1-branch_ne[i])

angmin*wr_ne[i] <= wi_ne[i] <= angmax*wr_ne[i]

Loss Constraints

PowerModels.constraint_power_losses — Function

Defines branch flow model power flow equations

Defines linear branch flow model power flow equations

PowerModels.constraint_power_losses_lb — Function

p[f_idx] + p[t_idx] >= 0
q[f_idx] + q[t_idx] >= -c/2*(v[f_bus]^2/tr^2 + v[t_bus]^2)

PowerModels.constraint_voltage_magnitude_difference — Function

Defines voltage drop over a branch, linking from and to side voltage magnitude

Defines voltage drop over a branch, linking from and to side voltage magnitude

PowerModels.constraint_voltage_drop — Function

Defines voltage drop over a branch, linking from and to side complex voltage

Storage Constraints

PowerModels.constraint_storage_thermal_limit — Function

PowerModels.constraint_storage_current_limit — Function

PowerModels.constraint_storage_complementarity_nl — Function

PowerModels.constraint_storage_complementarity_mi — Function

PowerModels.constraint_storage_losses — Function

Neglects the active and reactive loss terms associated with the squared current magnitude.

PowerModels.constraint_storage_state_initial — Function

PowerModels.constraint_storage_state — Function

DC Line Constraints

PowerModels.constraint_dcline_power_losses — Function

Creates Line Flow constraint for DC Lines (Matpower Formulation)

p_fr + p_to == loss0 + p_fr * loss1

PowerModels.constraint_dcline_setpoint_active — Function

pf[i] == pf, pt[i] == pt

p_fr[i] == pref_fr, p_to[i] == pref_to

Switch Constraints

PowerModels.constraint_switch_state — Function

enforces static switch constraints

PowerModels.constraint_switch_on_off — Function

enforces controlable switch constraints

PowerModels.constraint_switch_thermal_limit — Function

enforces an mva limit on the power flow over a switch