Multi-level coordinated voltage control method and system for power distribution network based on three-tier priority objectives

Abstract

The present application provides a multi-level coordinated voltage control method and system for a power distribution network based on three-tier priority objectives. The method comprises: acquiring operation parameters of a power distribution network; on the basis of the operation parameters of the power distribution network and a pre-constructed three-tier optimization objective planning model, using an affine theory, a duality theory, and power circle linearization and absolute value linearization methods to solve the three-tier optimization objective planning model to obtain maximization of an admissible net-load disturbance domain at each node, minimization of the total operation cost of the power distribution network and minimization of an expected voltage deviation; and performing coordinated control on various reactive devices of the power distribution network by means of the operation parameters of the power distribution network corresponding to the maximization of the admissible net-load disturbance domain at each node, the minimization of the total operation cost of the power distribution network and the minimization of the expected voltage deviation. The present application can more clearly characterize the uncertainty of distributed generation power output and the impact of the distributed generation power output on system reserve capacity, and mitigate the problems of ambiguous characterization of distributed generation power output characteristics, and multi-level voltage violation and voltage fluctuation of the power distribution network caused by large-scale distributed generation integration.

Description

Multi-level voltage coordination control method and system for distribution network based on three-layer priority targets

[0001] Cross-reference to Related Applications

[0002] The present application is based on the Chinese patent application No. 202410264658.7, filed on March 8, 2024, entitled "Multi-level voltage coordination control method and system for distribution network based on three-layer priority targets", and claims priority to the Chinese patent application No. 202410264658.7, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD

[0003] The present application relates to the field of power system distribution network optimization scheduling, in particular to a multi-level voltage coordination control method and system for distribution network based on three-layer priority targets. BACKGROUND

[0004] With the increasing proportion of distributed energy such as wind power and photovoltaic in the distribution network, the fluctuation of distributed power output causes the power quality of the distribution network to deteriorate. To solve the uncertainty brought by the access of distributed energy and improve the safety, economy and reliability of the operation of the distribution network, coordinated control and optimization of multiple voltage levels are needed for the distribution network.

[0005] At present, most studies achieve voltage control of the distribution network by minimizing network loss or voltage fluctuation, and only consider single-level voltage. There are several situations in the prior art: (1) theoretically analyzing the impact of distributed photovoltaic access on the power quality of the distribution network and performing simulation analysis. (2) Using an improved primal-dual interior point method to study a multi-level coordinated distribution network voltage quality optimization method. (3) Studying a multi-level coordinated control strategy for distribution network voltage based on distributed photovoltaic clusters. (4) Taking the minimization of network loss cost as the objective function, and solving the multi-level control model of reactive power and voltage by particle swarm optimization.

[0006] In addition, the traditional way of describing the fluctuation of distributed power output by interval or probability distribution cannot accurately express the uncertainty of distributed power output caused by weather changes. SUMMARY

[0007] In order to solve the problems that the prior art achieves voltage control of the distribution network by minimizing network loss or voltage fluctuation, and only considers single-level voltage, and the traditional way of describing the fluctuation of distributed power output by interval or probability distribution cannot accurately express the uncertainty of distributed power output caused by weather changes, the present application proposes a multi-level voltage coordination control method for distribution network based on three-layer priority targets, which comprises:

[0008] obtaining distribution network operation parameters;

[0009] based on the power distribution network operation parameters and a pre-constructed three-layer optimization objective programming model, the three-layer optimization objective programming model is solved by using affine theory, duality theory, power circle linearization and absolute value linearization method, to obtain the maximum node acceptable net load disturbance domain, the minimum total cost of power distribution network operation and the minimum expected voltage deviation;

[0010] The power distribution network operation parameters corresponding to the maximum node acceptable net load disturbance domain, the minimum total cost of power distribution network operation and the minimum expected voltage deviation are used to cooperatively control various reactive power equipment of the power distribution network.

[0011] The three-layer optimization objective programming model is constructed by taking the maximum node acceptable net load disturbance domain, the minimum total cost of power distribution network operation and the minimum expected voltage deviation as three-layer objective functions, and setting constraint conditions for the three-layer objective functions.

[0012] In some embodiments, the construction of the three-layer optimization objective programming model includes:

[0013] The first layer optimization objective is constructed by maximizing the node acceptable net load disturbance domain.

[0014] The second layer optimization objective is constructed by minimizing the total cost of power distribution network operation.

[0015] The third layer optimization objective is constructed by minimizing the expected voltage deviation.

[0016] Constraint conditions are set for the first layer optimization objective, the second layer optimization objective and the third layer optimization objective.

[0017] The constraint conditions include constraints under net load prediction value and constraints under net load disturbance.

[0018] In some embodiments, the constraints under net load prediction value include total power balance constraints under net load prediction value, line flow constraints, branch capacity constraints, node voltage upper and lower limit constraints, unit output constraints, energy storage constraints, upper-level power grid power supply constraints, light and wind curtailment, load shedding constraints, grouping switched capacitor constraints, static var compensator SVC constraints.

[0019] The constraints under net load disturbance include total power balance constraints under net load disturbance, net load acceptable domain constraints, line flow constraints, line capacity constraints, node voltage upper and lower limit constraints, generator related constraints, energy storage constraints, upper-level power grid power supply constraints, light and wind curtailment, load shedding constraints, grouping switched capacitor constraints, static var compensator constraints.

[0020] In some embodiments, the first layer optimization objective is as follows:

[0021] wherein Z1 is a first layer objective function, is an upward node admissible net load disturbance domain, is a downward node admissible net load disturbance domain, NT represents a set of optimization periods; NI represents a set of nodes; i is a node number; t is an optimization period;

[0022] The second layer optimization objective is shown in the following formula: Z2 = min(C oper + C cut + C ess )

[0023] wherein Z2 is a second layer objective function, C oper is an operation cost, C cut is a curtailment cost, and C ess is an energy storage cost;

[0024] The third layer optimization objective is shown in the following formula:

[0025] wherein Z3 is a third layer objective function, is a node voltage prediction value, V i,t is a node actual voltage.

[0026] In some embodiments, based on the power distribution network operation parameters and the pre-constructed three-layer optimization objective planning model, the three-layer optimization objective planning model is solved by using affine theory, dual theory, power circle linearization and absolute value linearization method to obtain maximized node admissible net load disturbance domain, minimized power distribution network operation total cost and minimized expected voltage deviation, including:

[0027] The uncertainty formula in the three-layer optimization objective planning model is converted into a deterministic formula by using affine theory and dual theory;

[0028] The nonlinear formula in the three-layer optimization objective planning model is converted into a linear formula by using power circle linearization and absolute value linearization method;

[0029] The converted three-layer optimization objective planning model is solved to obtain maximized node admissible net load disturbance domain, minimized power distribution network operation total cost and minimized expected voltage deviation.

[0030] In some embodiments, the uncertainty formula in the three-layer optimization objective planning model is converted into a deterministic formula by using affine theory and dual theory, including:

[0031] The node voltage uncertainty in the three-layer optimization objective planning model is converted by introducing an auxiliary variable by using affine theory to obtain a deterministic variable;

[0032] The uncertainty of the unit regulating disturbance output is transformed by using the duality theory to obtain a node admissible net load disturbance domain.

[0033] In some embodiments, the power circle linearization and absolute value linearization method is used to transform the nonlinear formula in the three-layer optimization objective programming model into a linear formula, including:

[0034] The uncertainty formula in the three-layer optimization objective programming model that can be segmented linearized is linearized by using the power circle linearization method;

[0035] The absolute value in the uncertainty formula is linearized by using the absolute value linearization method.

[0036] In some embodiments, the power circle linearization method is used to linearize the uncertainty formula in the three-layer optimization objective programming model that can be segmented linearized, including:

[0037] The feasible region of the uncertainty formula is the interior of a circle, and a regular polygon inscribed in the circle is approximated as the circle corresponding to the feasible region of the uncertainty formula.

[0038] The area surrounded by the regular polygon is approximated instead of the area surrounded by the circle, and the uncertainty formula is transformed into an area calculation formula of the regular polygon.

[0039] In some embodiments, the distribution network operation parameters include: distribution network topology, distribution network line capacity and resistance reactance value, time-of-use electricity price, controllable generator data, upper-level grid data, and load data.

[0040] In another aspect, the application also provides a distribution network multi-level voltage cooperative control system based on a three-layer priority objective, including:

[0041] A parameter acquisition module is configured to acquire distribution network operation parameters.

[0042] A target solving module is configured to solve the three-layer optimization objective programming model by using the affine theory, the duality theory, the power circle linearization method, and the absolute value linearization method based on the distribution network operation parameters and the pre-constructed three-layer optimization objective programming model, to obtain a maximum node admissible net load disturbance domain, a minimum distribution network total operation cost, and a minimum expected voltage deviation.

[0043] A control module is configured to cooperatively control various reactive power devices of the distribution network by using the distribution network operation parameters corresponding to the maximum node admissible net load disturbance domain, the minimum distribution network total operation cost, and the minimum expected voltage deviation.

[0044] The three-layer optimization target programming model is constructed by taking maximization of node admissible net load disturbance domain, minimization of total distribution network operation cost, and minimization of expected voltage deviation as three-layer objective functions, and setting constraint conditions for the three-layer objective functions.

[0045] In some embodiments, further comprising a model construction module configured to:

[0046] constructing a first-layer optimization target by maximizing node admissible net load disturbance domain;

[0047] constructing a second-layer optimization target by minimizing total distribution network operation cost;

[0048] constructing a third-layer optimization target by minimizing expected voltage deviation;

[0049] setting constraint conditions for the first-layer optimization target, the second-layer optimization target, and the third-layer optimization target;

[0050] The constraint conditions comprise constraint under net load prediction value and constraint under net load disturbance.

[0051] In some embodiments, the constraint under net load prediction value comprises total power balance constraint under net load prediction value, line power flow constraint, branch capacity constraint, node voltage upper and lower limit constraint, unit output constraint, energy storage constraint, upper-level power grid power supply constraint, light and wind curtailment, load shedding constraint, grouping switched capacitor constraint, static var compensator (SVC) constraint.

[0052] The constraint under net load disturbance comprises total power balance constraint under net load disturbance, net load admissible domain constraint, line power flow constraint, line capacity constraint, node voltage upper and lower limit constraint, generator related constraint, energy storage constraint, upper-level power grid power supply constraint, light and wind curtailment, load shedding constraint, grouping switched capacitor constraint, static var compensator (SVC) constraint.

[0053] In some embodiments, the first-layer optimization target is shown in the following formula:

[0054] In the formula, Z1 is the first-layer objective function, is upward node admissible net load disturbance domain, is downward node admissible net load disturbance domain, NT represents a set of optimization time periods, NI represents a set of nodes, i is a node number, and t is an optimization time period.

[0055] The second-layer optimization target is shown in the following formula: Z2=min(C oper +C cut +C ess )

[0056] wherein Z2 is a second layer objective function, C oper is an operation cost, C cut is a curtailment cost, C ess is an energy storage cost;

[0057] The third layer optimization objective is shown in the following formula:

[0058] wherein Z3 is a third layer objective function, is a node voltage prediction value, V i,t is a node actual voltage.

[0059] In some embodiments, the target solving module comprises:

[0060] A deterministic transformation submodule is configured to transform the uncertainty in the three-layer optimization objective programming model into a deterministic formula by using affine theory and duality theory;

[0061] A linearization submodule is configured to transform the non-linear formula in the three-layer optimization objective programming model into a linear formula by using power circle linearization and absolute value linearization methods;

[0062] A solving submodule is configured to solve the transformed three-layer optimization objective programming model to obtain the maximum node admissible net load disturbance domain, the minimum total distribution network operation cost, and the minimum expected voltage deviation.

[0063] In some embodiments, the deterministic transformation submodule is specifically configured to:

[0064] Transform the node voltage uncertainty in the three-layer optimization objective programming model by introducing an auxiliary variable to obtain a deterministic variable by using affine theory;

[0065] Transform the unit regulation disturbance output uncertainty to obtain a node admissible net load disturbance domain by using duality theory.

[0066] In some embodiments, the linearization submodule is specifically configured to:

[0067] Linearize the uncertainty formula in the three-layer optimization objective programming model that can be segmented and linearized by using a power circle linearization method;

[0068] Linearize the absolute value in the uncertainty formula by using an absolute value linearization method.

[0069] In some embodiments, the specific implementation steps of the linearization submodule for linearizing the uncertainty formula in the three-layer optimization objective programming model that can be segmented and linearized by using a power circle linearization method comprise:

[0070] The feasible region of the uncertainty formula is the interior of a circle, and a regular polygon inscribed in the circle is approximated as the feasible region of the uncertainty formula corresponding to the circle.

[0071] The area surrounded by the regular polygon is approximated instead of the area surrounded by the circle, and the uncertainty formula is converted into an area calculation formula of the regular polygon.

[0072] In still another aspect, the application further provides a computing device, comprising at least one processor and a memory;

[0073] The memory is configured to store one or more programs.

[0074] When the one or more programs are executed by the at least one processor, the three-layer priority target-based multi-level voltage coordinated control method for a power distribution network is implemented.

[0075] In still another aspect, the application further provides a computer readable storage medium having a computer program stored thereon, and the computer program is executed to implement the three-layer priority target-based multi-level voltage coordinated control method for a power distribution network.

[0076] Compared with the prior art, the application has the following beneficial effects:

[0077] The application provides a three-layer priority target-based multi-level voltage coordinated control method for a power distribution network, comprising: obtaining power distribution network operation parameters; based on the power distribution network operation parameters and a pre-constructed three-layer optimization target planning model, using affine theory, duality theory, power circle linearization and absolute value linearization method to solve the three-layer optimization target planning model to obtain a maximum node admissible net load disturbance domain, a minimum power distribution network operation total cost and a minimum expected voltage deviation; and using the power distribution network operation parameters corresponding to the maximum node admissible net load disturbance domain, the minimum power distribution network operation total cost and the minimum expected voltage deviation to cooperatively control various reactive power equipment of the power distribution network; wherein the three-layer optimization target planning model is constructed by taking the maximum node admissible net load disturbance domain, the minimum power distribution network operation total cost and the minimum expected voltage deviation as three-layer objective functions, and setting constraint conditions for the three-layer objective functions. The application takes the maximum node admissible net load disturbance domain, the minimum power distribution network operation total cost and the minimum expected voltage deviation as three-layer objective functions, which can more clearly depict the uncertainty of distributed power output and the influence of distributed power output on system reserve capacity, and improve the problem of fuzzy description of distributed power output characteristics; the power distribution network voltage deviation is optimized in a targeted manner, and the multi-level voltage out-of-limit and voltage fluctuation problems of the power distribution network caused by large-scale access of distributed power are effectively improved; the safety, economy and reliability of the power distribution network operation are considered as a whole, so that the optimization result is more practical in engineering. BRIEF DESCRIPTION OF DRAWINGS

[0078] Fig. 1 is a flow chart of the three-layer priority target-based multi-level voltage collaborative control method of the power distribution network of the present application;

[0079] Fig. 2 is a flow chart of the multi-level voltage collaborative control method of the embodiment of the present application;

[0080] Fig. 3 is a schematic diagram of the range of the node-admissible net load disturbance domain of the present application;

[0081] Fig. 4 is a schematic diagram of the piecewise linearization of the unit operation cost of the present application;

[0082] Fig. 5 is a schematic diagram of the linearization of the branch capacity constraint and the unit capacity constraint of the present application. DETAILED DESCRIPTION

[0083] The present application proposes a three-layer priority target-based multi-level voltage collaborative control method and system of the power distribution network, constructs a three-layer priority target planning model, the first layer optimization target is to maximize the node-admissible net load disturbance domain, to more flexibly describe the uncertainty of the distributed power output in the form of interval endpoints that can be optimized, to reduce the impact of distributed power disturbance on the system voltage, to maximize the consumption of distributed power output, to reduce the cost of traditional energy output, and to meet the double carbon target; the second layer optimization target is to minimize the total operation cost of the power distribution network, to meet the economic requirements of the power system operation; the third layer optimization target is to minimize the expected voltage deviation, when the conventional control technology is used to increase the distributed power generation, the voltage change of each node will become a problem, in order to make the voltage change not harm the user's equipment and other devices of the power system, the node voltage should be kept within a reasonable range, by adjusting various reactive power devices in the system, the voltage of each node of the power distribution network is optimized. At the same time, the generation load transfer factor of decoupled linearized power flow is used to construct the power flow expression, to more accurately describe the change of the voltage of the power distribution network. Through the three-layer objective function, the safety, economy and reliability of the power distribution network are considered as a whole, to realize the multi-level voltage collaborative control of the power distribution network in a more reasonable way.

[0084] The present application uses affine theory and duality theory to convert the uncertainty model into a deterministic model, uses power circle linearization and absolute value linearization method to convert the nonlinear model into a linear model, improves the calculation efficiency and solution accuracy.

[0085] Embodiment 1:

[0086] The three-layer priority target-based multi-level voltage collaborative control method of the power distribution network, as shown in Fig. 1, includes:

[0087] Step 1: Obtain the operation parameters of the power distribution network;

[0088] Step 2: based on the power distribution network operation parameters and the pre-constructed three-layer optimization objective programming model, using affine theory, duality theory, power circle linearization and absolute value linearization method to solve the three-layer optimization objective programming model, to obtain the maximum node acceptable net load disturbance domain, the minimum total cost of power distribution network operation and the minimum expected voltage deviation;

[0089] Step 3: the power distribution network operation parameters corresponding to the maximum node acceptable net load disturbance domain, the minimum total cost of power distribution network operation and the minimum expected voltage deviation are used to cooperatively control various reactive power equipment of the power distribution network.

[0090] The three-layer optimization objective programming model is constructed by taking the maximum node acceptable net load disturbance domain, the minimum total cost of power distribution network operation and the minimum expected voltage deviation as the three-layer objective functions, and setting constraint conditions for the three-layer objective functions.

[0091] The three-layer priority objective-based power distribution network multi-level voltage cooperative control method and system proposed in the application adopts three-layer objective programming theory, reasonably considers the safety, economy and reliability of the power distribution network according to the priority of actual demand, introduces the node acceptable net load disturbance domain into the power distribution network operation optimization, can realize good mutual aid between the load side and the power supply side, uses the generation load transfer factor based on decoupled linearized power flow to construct system operation constraints, realizes the joint optimization of active power flow, reactive power flow and voltage, and linearizes the constraints using power circle linearization and 0 / 1 variable linearization, converts the large-scale MINLP (Mixed Integer Nonlinear Programming) problem of the power distribution network into a MIP (Mixed Integer Linear Programming) problem, uses affine strategy and duality theory to eliminate the uncertainty in the model, and realizes efficient solution of the model.

[0092] Before step 1, the three-layer optimization objective programming model is constructed, and the construction process of the three-layer optimization objective programming model is as follows:

[0093] 1.1 Objective function: Z2=min(C oper +C cut +C ess ) (2)

[0094] The model has three-layer objective functions, the first-layer objective function Z1 is to maximize the node acceptable net load disturbance domain, as shown in formula (1), including the upward node acceptable net load disturbance domain and the downward node acceptable net load disturbance domain The second layer objective function Z2 is to minimize the total cost of distribution network operation, as shown in equation (2), including operation cost C oper , curtailment cost C cut , energy storage cost C ess ; the third layer objective function Z3 is to minimize the expected voltage deviation, as shown in equation (3), which is the sum of the absolute values of the differences between the actual voltage V i,t of all nodes of the distribution network and the predicted voltage value of each node, i is the node number, and t is the optimization period. The second layer objective function cost can be explicitly expressed as follows:

[0095] In the formula, is the purchase cost of the distribution network from the upper-level power grid, is the unit operation cost, is the load shedding cost, is the wind curtailment cost, is the light curtailment cost; NT represents the set of optimization periods; NI represents the set of nodes; NF represents the set of upper-level power stations; NG represents the set of generator units; NW represents the set of wind turbines; NPV represents the set of photovoltaics; and NE represents the set of energy storage devices. and are the cost coefficients of the active power and the reactive power provided by the upper-level power station f in the t period; and represent the active power and the reactive power generated by the upper-level power grid f in the t period when the net load is the predicted value; represents the active power output of the generator g in the t period when the net load is the predicted value, i.e., the active operation base point; a g , b g , c g are the operation cost coefficients of the generator g; and represent the upward and downward reserve that the generator g can provide in the t period; represent the cost of the upward and downward reserve provided by the generator; σ L , σ w , and σ pv are the penalty cost coefficients of load shedding, wind curtailment, and light curtailment, respectively; ΔP i,t and ΔQ i,t are the active and reactive load shedding amounts of the node i in the t period; ΔP w,t and ΔP pv,t are the wind curtailment amount of the wind turbine w and the light curtailment amount of the photovoltaic pv in the t period; ρ E is the energy storage charging and discharging cost coefficient; and​ Pcharge (e, t) and Pdischarge (e, t) are the charging and discharging power of the energy storage e in time period t, e is the index of the energy storage, and f is the index of the upper power station or power grid.

[0096] 1.2 Constraints

[0097] The constraints in the model of the multi-level voltage collaborative control method of the distribution network based on three levels of priority targets include constraints under the predicted value of the net load and constraints under the disturbance of the net load.

[0098] Constraints under the predicted value of the net load:

[0099] (1) Total power balance equation

[0100] In the equation, Pgen (g, t) represents the active power output of the generator g in time period t, i.e., the active operation base point when the net load is the predicted value; Pload (i, t) is the predicted active value of the net load at node i in time period t, Qgen (g, t) represents the reactive power output of the generator g in time period t, i.e., the reactive operation base point when the net load is the predicted value; Qload (i, t) is the predicted reactive value of the net load at node i in time period t. NSVC represents the set of static var compensators svc; Qsvc (svc, t) represents the reactive power provided by the static var compensator svc in time period t. NCB represents the set of capacitor banks cb; Qcb (cb, t) represents the reactive power provided by the capacitor bank cb in time period t. g is the index of the generator;

[0101] Equation (7) ensures that the sum of the changes in the total generator output (active or reactive) and the upper grid output (active or reactive) is equal to the sum of the changes in the total predicted net load (active or reactive) in any time period.

[0102] (2) Line flow calculation equation

[0103] In the equation, ref is the parameter and variable under the predicted value of the net load, and rer is the reference node, Pij (t) and Qij (t) represent the active and reactive power flows of the line ij in time period t under the predicted value of the net load, respectively; loc (g) = k represents the generator g connected to node k; Pjk (k) and Qjk (k) represent the active and reactive power load transfer factors of the active power flow with respect to the active and reactive power injection at node k, respectively; Pjk (k) and Qjk (k) represent the active and reactive power load transfer factors of the active power flow with respect to the active and reactive power injection at node k, respectively; b fis a binary variable, equal to 1 if the upper grid is directly connected to node k, and equal to 0 if the upper grid is not directly connected to node k; loc(svc) = k indicates that a static var compensator is connected at node k; loc(cb) = k indicates that a capacitor bank is connected at node k; is the active power output of generator g connected at node k in time period t, is the reactive power output of generator g in time period t, is the reactive power output of generator g connected at node k in time period t, is the reactive power output of static var compensator svc in time period t, is the reactive power output of static var compensator svc connected at node k in time period t, is the reactive power output of capacitor bank b in time period t, is the reactive power output of capacitor bank cb connected at node k in time period t, g is the generator set number, i and j are both node numbers, NE is the set of energy storage devices, e is the number of energy storage devices, and t is the optimization period, are the active and reactive parameters, respectively, satisfying equation (9):

[0104] In the formula, NL = {NL none ,NL end ,NL start}; NL none represents the set of branches not directly connected to the reference node; NL start represents the set of branches with the reference node as the starting node; NL end represents the set of branches with the reference node as the terminal node; is the active power load transfer factor of active power flow with respect to the active power injection at node k, G k(ref) is the predicted value of the conductance of the line connected to node k, V ref is the predicted value of the voltage amplitude, B k(ref) is the predicted value of the susceptance of the line connected to node k, θ ref is the predicted value of the voltage phase angle, is the active power load transfer factor of active power flow with respect to the reactive power injection at node k, is the active power load transfer factor of reactive power flow with respect to the active power injection at node k, is the active power load transfer factor of reactive power flow with respect to the reactive power injection at node k, b ij is the susceptance of line ij, g ij is the conductance of line ij.

[0105] (3) Branch capacity constraints:

[0106] wherein, denotes the upper limit of the capacity of branch ij.

[0107] Equation (10) denotes that the total flow on the branch cannot exceed the branch capacity limit.

[0108] (4) Node voltage upper and lower limit constraints

[0109] wherein, V i max and V i min denote the upper limit of the voltage magnitude and the lower limit of the voltage magnitude of node i, respectively; θ i max , θ i min denote the upper limit of the voltage phase angle and the lower limit of the voltage phase angle of node i, respectively; wherein, denote the generation load shift factor of the voltage magnitude of node i with respect to the active power and the reactive power injection at node k, respectively; denote the generation load shift factor of the voltage phase angle of node i with respect to the active power and the reactive power injection at node k, respectively; and are the voltage magnitude related parameter and the voltage phase angle related parameter, respectively, satisfying equation (13):

[0110] wherein, denote the generation load shift factor of the voltage magnitude of node i with respect to the active power and the reactive power injection at node k, respectively; G k(ref) is the conductance prediction value of the line connected to node k, V ref is the voltage magnitude prediction value, θ ref is the voltage phase angle prediction value, B k(ref) is the susceptance prediction value of the line connected to node k.

[0111] The intermediate term of equation (11) and the intermediate term of equation (12) are the node voltage magnitude and the node voltage phase angle calculated based on the decoupled linearized load, respectively, both of which need to be within the specified upper and lower limit range.

[0112] (5) Unit output constraints

[0113] wherein, and denote the minimum active power output value and the maximum active power output value of unit g, respectively; denotes the upper limit of the output capacity of unit g; and are the upward ramping capacity and downward ramping capacity of unit g; At represents the duration of time period t; UR g is the upward ramping rate of unit g, DR g is the downward ramping rate of unit g; represents the active power output of generator g in time period t when the net load is predicted, i.e., the active operating base point; represents the active power output of generator g in time period t-1 when the net load is predicted.

[0114] The output of a generator unit is limited by its physical characteristics. The first equation of equation (14) indicates that the active power output of generator g should be within the specified upper and lower limits, and the second equation indicates that the total output of the generator should be less than the upper limit of the generator capacity. Equation (15) provides the upper and lower limit range constraints for the upward and downward reserve of generator g. Equation (16) provides the upward ramping rate and downward ramping rate constraints for unit g.

[0115] (6) Energy storage constraint

[0116] In the equation, and represent the maximum charging power and the maximum discharging power of energy storage e, respectively; E e,t represents the energy stored by energy storage e in time period t; E e,t-1 represents the energy stored by energy storage e in time period t-1; η c and η d represent the charging and discharging efficiencies of the energy storage, respectively; and represent the upper limit and the lower limit of the energy stored by energy storage e in time period t, respectively; E start and E end represent the initial energy and the final energy in a charging cycle, respectively. This constraint limits the charging and discharging power and the change of stored energy of the energy storage device.

[0117] (7) Upper-level power grid power supply constraint

[0118] In the equation, and represent the maximum active power and reactive power provided by the upper-level power grid f, respectively. This equation constrains the output of the upper-level power grid to be within the upper and lower limits.

[0119] (8) Light curtailment, wind curtailment, and load shedding constraint

[0120] In the equation, and are the active power output prediction values of photovoltaic pv and wind turbine w in time period t, respectively; and are the active and reactive load of node i at time period t, respectively; ΔP w,t is the curtailment of wind farm w at time period t, ΔP pv,t is the curtailment of photovoltaic pv at time period t, is the active curtailment of node i at time period t, is the reactive curtailment of node i at time period t.

[0121] (9) Grouped switched capacitor (CB) constraints

[0122] where Q cb,t is the reactive power output of capacitor bank cb at time period t, h is the gear position, NC cb,h,t is the number of capacitor banks switched in when capacitor bank cb is at gear position h, is a parameter; x cb,h,t is a 0 / 1 variable indicating whether capacitor bank cb is at gear position h at time t, and the value of 1 means that capacitor bank cb is at gear position h at time t, and vice versa; this equation indicates that capacitor bank cb can only be at one gear position at time period t.

[0123] where B cb,t is a 0 / 1 variable, and the value of 0 means that capacitor bank cb maintains the original state without action, and the value of 1 means that capacitor bank cb adjusts; Q cb,t-1 is the reactive power output of capacitor bank cb at time period t-1; is the unit reactive power output of capacitor bank cb, is the maximum number of gear positions that capacitor bank cb can adjust at time t, is the maximum number of actions of capacitor bank cb in the optimization period.

[0124] (12) Static var compensator (SVC) constraints

[0125] where, and are the upper and lower limits of the SVC output, respectively.

[0126] In order to ensure that the reserve can be successfully and effectively transmitted when the net load fluctuates randomly, the model needs to satisfy the constraints under the net load disturbance, which are similar to the constraints under the net load prediction value:

[0127] 1) Total power balance equation

[0128] where, and P ij,t and Q ij,t denote the random active and reactive power output of the generator g connected at node k in the time period t, respectively, and denote the random active and reactive power output of the generator g connected at node k in the time period t, respectively, denote the random reactive power output of the static var compensator svc in the time period t, denote the random reactive power output of the switched capacitor in the time period t, and denote the random active and reactive power output of the generator g connected at node k in the time period t, respectively,

[0129] 2) Net load admissible region constraint

[0130] where, denotes the admissible downward net load disturbance region of node i at time t, denotes the admissible upward net load disturbance region of node i at time t, and denote the upper and lower limits of the net load disturbance at node i in the time period t, respectively, which are determined by the load characteristics and the natural disturbance law of the distributed energy output, and are parameters that can be obtained by probabilistic prediction methods.

[0131] Equation (24) indicates that the size of the admissible net load disturbance region is within the upper and lower limits of the net load disturbance (as shown in FIG. 2).

[0132] 3) Line power flow calculation equation

[0133] where, P ij,t and Q ij,t denote the active and reactive power flow of line ij in the time period t under the net load disturbance; P g,t denotes the active power output of generator g in the time period t in the disturbance state, P f,t denotes the active power output of the upper-level power grid f in the time period t in the disturbance state, denotes the active power output of the generator g connected at node k in the time period t in the disturbance state, Q g,t denotes the reactive power output of generator g in the time period t in the disturbance state, denotes the reactive power output of the generator g connected at node k in the time period t in the disturbance state, Q svc,t denotes the reactive power output of the static var compensator svc in the time period t in the disturbance state, Q svck,tQ represents the reactive power output of the static var compensator (SVC) connected to node k during time interval t under disturbance conditions. cb,t Q represents the reactive power output of capacitor bank b during time interval t under disturbance conditions. cbk,t This represents the reactive power output of capacitor bank cb connected to node k during time interval t under disturbance conditions. According to equation (25), when the net load fluctuates within the acceptable net load disturbance domain, the power flow on the line also fluctuates randomly.

[0134] 4) Line capacity constraints

[0135] In the formula, This represents the line capacity under disturbance conditions.

[0136] According to equation (26), when the net load fluctuates within the acceptable net load disturbance range, the total power flow on the line cannot exceed the line's capacity limit.

[0137] 5) Node voltage upper and lower limit constraints

[0138] In the formula, and These are the voltage amplitude related parameters and the voltage phase angle related parameters, respectively, which satisfy equation (13).

[0139] Equations (27) and (28) indicate that when the net load fluctuates within the acceptable net load disturbance range, the node voltage amplitude and phase angle must be within the specified upper and lower limits.

[0140] 6) Generator-related constraints

[0141] In the formula, t represents the upper limit of the output capacity of generator set g, and t represents the optimization period.

[0142] Equation (29) indicates that when the net load fluctuates within the acceptable net load disturbance range, the active power output of the generator needs to be within the upper and lower limits of active power, and the total output needs to be less than the upper limit of capacity. Equation (30) indicates that when the net load fluctuates within the acceptable net load disturbance range, the reserve provided by the generator should be greater than the reserve required by the system.

[0143] The constraints on energy storage, upstream power grid supply, curtailment of solar and wind power load shedding, group switching of capacitors, and static var compensator are the same as the relevant constraints under the net load forecast, and will not be repeated here.

[0144] It can be seen that when considering net load disturbance, some parameters and variables contained in constraints (23)-(30) are random, and therefore contain an infinite number of constraints.

[0145] Step 1: Obtain power distribution network operation parameters;

[0146] The power distribution network operation parameters include:

[0147] The power distribution network topology, the power distribution network line capacity and resistance reactance value, time-of-use electricity price, controllable generator data, superior grid data, and load data.

[0148] Step 2: Based on the power distribution network operation parameters and the pre-constructed three-layer optimization objective planning model, the three-layer optimization objective planning model is solved by using affine theory, duality theory, power circle linearization, and absolute value linearization method to obtain the maximum node acceptable net load disturbance domain, the minimum total power distribution network operation cost, and the minimum expected voltage deviation, including:

[0149] The uncertainty formula in the three-layer optimization objective planning model is converted into a deterministic formula by using affine theory and duality theory;

[0150] The nonlinear formula in the three-layer optimization objective planning model is converted into a linear formula by using power circle linearization and absolute value linearization method;

[0151] The converted three-layer optimization objective planning model is solved to obtain the maximum node acceptable net load disturbance domain, the minimum total power distribution network operation cost, and the minimum expected voltage deviation.

[0152] Further, the uncertainty formula in the three-layer optimization objective planning model is converted into a deterministic formula by using affine theory and duality theory, including:

[0153] The node voltage uncertainty in the three-layer optimization objective planning model is converted by introducing an auxiliary variable by using affine theory to obtain a deterministic variable;

[0154] The unit regulation disturbance output uncertainty is converted by using duality theory to obtain a representation of the node acceptable net load disturbance domain.

[0155] Further, the nonlinear formula in the three-layer optimization objective planning model is converted into a linear formula by using power circle linearization and absolute value linearization method, including:

[0156] The uncertainty formula in the three-layer optimization objective planning model that can be segmented and linearized is linearized by using power circle linearization method;

[0157] The absolute value in the uncertainty formula is linearized by using absolute value linearization method.

[0158] Further, the uncertainty formula in the three-layer optimization target planning model which can be segmented linearized adopts a power circle linearization method for linearization, including:

[0159] The feasible region of the uncertainty formula is inside a circle, and a regular polygon inscribed in the circle is approximated as the feasible region corresponding to the circle of the uncertainty formula.

[0160] The area surrounded by the regular polygon is approximated instead of the area surrounded by the circle, and the uncertainty formula is converted into an area calculation formula of the regular polygon.

[0161] The specific steps of step 2 are as follows:

[0162] (1) Linearization of the model

[0163] The operating cost part of the model objective function can be segmented linearized, as shown in FIG. 3; the second formula of equations (10), (14), the second formula of (26), and (29) are quadratic constraints, and the above constraints are linearized by using a power circle linearization method for effective solution.

[0164] Taking the second formula of equation (29) as an example, the feasible region of the formula is inside a circle, and a regular polygon inscribed in the circle is approximated as the circle. In view of the accuracy and calculation efficiency of the model, the area surrounded by a dodecagon inscribed in the circle is approximated instead of the area surrounded by the circle, as shown in FIG. 4, and finally equation (29) can be converted into equation (31):

[0165] In the formula, and are coefficients corresponding to the linearized power circle constraint, which changes with the number of sides of the segmented regular polygon; α and β respectively represent the radian angles of two adjacent vertices of the regular polygon inscribed in the circle; u is the number of sides of the regular polygon inscribed in the circle; cb adjusts the number of linearization, and the intermediate variable ε cb,t = |Q cb,t -Q cb,t-1 |, Q cb,t represents the reactive power output of the capacitor bank b in the t time period under the disturbance state, and ε cb,t is the reactive power change value of the capacitor bank cb from t-1 time to t time, and Q cb,t-1 is the reactive power output of the capacitor bank b in the t-1 time period under the disturbance state. The absolute value in equation (21) is linearized to obtain:

[0166] In the formula, M is a penalty coefficient, which is taken as a number infinitely large in principle, and δ cb,t is an auxiliary 0\1 variable, B cb,t is a 0\1 variable representing whether the capacitor bank cb acts or not, for the unit reactive power output of the capacitor bank cb, for the maximum number of gears that the capacitor bank cb can adjust at time t.

[0167] (2) Deterministic transformation of the model

[0168] The model takes into account the uncertainty of the net load disturbance and the resulting uncertainty of the unit output adjustment and the node voltage change, contains countless uncertain equality constraints and inequality constraints, and is difficult to solve directly. To solve this problem, the present disclosure uses an affine strategy to deal with uncertain variables in the constraints. The affine strategy is also known as affine theory.

[0169] The affine strategy is shown in equation (35). Under the state of net load disturbance, the total output of the generator, the upper-level power grid, the grouped switched capacitor, and the static var compensator is the operating base point plus the random output adjustment. The random output adjustment is the product of the allocated participation factor and the total net load disturbance of the system. Since the sum of the output adjustments of the above-mentioned devices needs to be equal to the sum of the net load disturbance values, equation (36) ensures that the total participation factor of the above-mentioned devices is 1.

[0170] In the formula, respectively represent the active participation factor and the reactive participation factor of the unit g at time period t; respectively represent the active participation factor and the reactive participation factor of the upper-level power grid f at time period t; represents the reactive participation factor of the static var compensator svc at time period t; represents the reactive participation factor of the grouped switched capacitor cb at time period t; respectively represent the active output adjustment and the reactive output adjustment of the unit g at time period t; respectively represent the active output adjustment and the reactive output adjustment of the upper-level power grid f at time period t; represents the random active change at node i in time period t, represents the random reactive change at node i in time period t. respectively represent the reactive output adjustment of the static var compensator and the grouped switched capacitor at time period t.

[0171] Taking the reserve provided by the generator as an example, the uncertain model is transformed into a deterministic model.

[0172] The reserve provided by each unit is determined by its participation factor and the total net load acceptable range of the system, so the upward and downward reserves required by the unit g are respectively represented as:

[0173] In the formula, The upward net load admissible disturbance domain of node i at time t, The downward net load admissible disturbance domain of node i at time t.

[0174] The same reasoning can be applied to the reserve provided by the upper grid and the reactive power compensation equipment.

[0175] After adopting the affine strategy, the first formula of the total power balance equation (7), (23) and (30) are automatically satisfied, so they can be omitted. By introducing an auxiliary variable z i,t , the uncertain model is converted into a deterministic model. The random net load admissible disturbance domain can be explicitly expressed as:

[0176] In the formula, p is a proportional coefficient, assuming that the active net load disturbance of the node and the reactive net load disturbance of the node are in a proportional relationship.

[0177] Substitute equation (35) into the constraint based on the net load disturbance. At this time, the random adjustment amount of the generator, the random adjustment amount of the upper grid, the random adjustment amount of the reactive power compensation equipment, the random node voltage amplitude and phase angle, and the random line flow can be expressed by the random admissible net load disturbance value . Then for the transformed equation (31), , the equation should hold within its range. Since this equation holds for the worst-case realization of the random variable, it holds for any realization of the uncertain variable, so equation (31) can be transformed into:

[0178] In the formula, η i,g,t,u is the dual variable, is the upper limit of the output capacity of generator g.

[0179] After substituting equation (38) into equation (39), using the duality theory, equation (39) can be transformed into:

[0180] In the formula, is the active participation factor of unit g at time period t, is the reactive participation factor of unit g at time period t, η i,g,t,u is the dual variable, and p is a proportional coefficient, assuming that the active net load disturbance of the node and the reactive net load disturbance of the node are in a proportional relationship.

[0181] Using the same method, equations (26)-(28) can be similarly transformed, which will not be repeated here. Through the above transformation, the model is transformed into a deterministic linear programming model, which can be solved using mature commercial solvers

[0182] Since the security of the system is more important than the economy in the operation scheduling of the distribution network, the model is converted into a three-layer model by using the priority goal programming method, so that the model has a clear hierarchical order. The first layer model has priority, and the second layer model has one more constraint condition than the first layer model, which ensures that the value of the total net load admissible region obtained after optimization is not less than the optimization scheduling result of the first layer model, as shown in equation (41).

[0183] In the formula, λ is a conservativeness control factor, which is used to control the conservativeness of the optimization result. A larger value indicates that the optimization result of the system tends to be more secure for the power system, and a smaller value indicates that the optimization result of the system tends to be more economical for the power system. When λ = 1, the total net load admissible region obtained by the second layer optimization is the same as the result obtained by the first layer optimization. When λ = 0, the model is simplified into a deterministic economic dispatching problem; is the optimization result of the objective function of the first layer model.

[0184] Since the access of distributed power sources brings severe challenges to the voltage of the distribution network, the third layer model specifically takes the minimum expected voltage deviation as the objective function to solve the voltage out-of-limit problem.

[0185] In the formula, V i,t is the actual voltage of each node of the system at node i at time t, is the ideal voltage of each node of the system at node i at time t.

[0186] The absolute value can be linearized, and equation (42) is converted into:

[0187] In the formula, ZP is an auxiliary variable.

[0188] The third layer model has one more constraint condition than the second layer model, which ensures that the value of the total distribution network operation cost obtained after optimization is not less than the optimization scheduling result of the second layer model, as shown in equation (44).

[0189] In the formula, C oper is the distribution network operation cost, C cut is the distribution network wind and light curtailment load shedding cost, C ess is the energy storage operation cost, and is the distribution network operation cost after optimization of the second layer model, and π is a conservativeness control factor, which is used to control the conservativeness of the optimization result. Its function is the same as the conservativeness control factor λ, and will not be described again.

[0190] Step 3: Cooperatively control various reactive power equipment of the distribution network by using the distribution network operation parameters corresponding to the maximized node admissible net load disturbance domain, the minimized total distribution network operation cost, and the minimized expected voltage deviation.

[0191] Compared with the prior art, the application has the beneficial effects that:

[0192] (1) The application constructs a multi-level voltage collaborative control method model of a power distribution network based on three layers of priority targets, adopts a node admissible net load disturbance domain to quantitatively describe the ability of the node to admit distributed energy and the ability of the system to resist load disturbance, comprehensively considers the collaborative interaction ability among distributed energy output, energy storage systems and load disturbance, and excavates the comprehensive regulation potential of a multi-target multi-agent new type power system.

[0193] (2) The application constructs a three-layer model based on priority target programming theory, the first layer model is to maximize the node admissible net load disturbance domain, the second layer model is to minimize the operation cost of the power distribution network, and the third layer is to minimize the expected voltage deviation. Meanwhile, the optimal result of the target function decided by the first layer model is introduced into the second layer model, and is reasonably relaxed as a restrictive constraint. The targets decided by the first layer and the second layer model are both introduced into the third layer model as restrictive constraints, and the third layer model is solved to obtain the optimal operation strategy of the power distribution network. The model ensures that the power distribution network system comprehensively considers safety, economy and reliability, and meets the requirements of safe and economic operation of the system.

[0194] (3) The application adopts a generation load transfer factor based on decoupled linearized power flow that is more suitable for the operation of the power distribution network to construct the related constraints of the system, finely considers the size relationship between the line resistance and the line reactance of the power distribution network, finely considers the influence of the voltage amplitude change and the reactive power of the power distribution network, can realize collaborative optimization of active power flow, reactive power flow and node voltage, obviously reduces the power flow calculation error, increases the reliability of the optimization result of the power distribution network, and improves the practical engineering application value of the optimization of the power distribution network.

[0195] (4) The application adopts an affine strategy to process the uncertain variables in the model, and further quantitatively represents the node admissible net load disturbance domain by using the duality theory, converts the model of the disclosure into a deterministic model, linearizes the nonlinear part of the model by using the power circle linearization and the big M method, and converts the model into a deterministic linear model, thereby improving the calculation efficiency and practical application value of the model.

[0196] Embodiment 2:

[0197] According to some embodiments, the application adopts the following technical scheme:

[0198] In a first aspect, the application provides a multi-level voltage collaborative control method of a power distribution network based on three layers of priority targets, comprising:

[0199] The maximum node admissible net load disturbance domain is taken as a first layer target, the minimum distribution network operation cost is taken as a second layer target, the minimum distribution network expected voltage deviation is taken as a third layer target, a storage system model and a plurality of reactive power regulation device models are introduced into the dispatching model, a generation load transfer factor based on decoupled linearization power flow is used to construct system operation constraints, and a distribution network multi-level voltage collaborative control method model based on three layer priority target is constructed. The distribution network multi-level voltage collaborative control method model is also referred to as a three layer optimization target planning model.

[0200] The affine strategy and the duality theory are used to convert the node voltage uncertainty and the unit regulation disturbance output uncertainty into determinacy, the model is converted into a determinacy model, the power circle linearization method is used to linearize the branch capacity constraint expression, the large M method is used to linearly express the capacitor bank and the on-load regulation transformer regulation times constraint, the model is converted into a linear model, and finally the distribution network multi-level voltage collaborative control method model based on three layer priority target is solved by a linear programming algorithm to obtain the operation and dispatching scheme of each subject of the power system.

[0201] As an optional implementation, the node admissible net load disturbance domain includes an upward node admissible net load disturbance domain and a downward node admissible net load disturbance domain; the distribution network operation cost includes a unit operation cost, a superior grid power purchase cost, a wind power and light power abandonment and load shedding cost, and a storage dispatching cost; and the expected voltage deviation is the sum of the difference between the actual voltage of all nodes of the 24-hour distribution network 35kv-10kv-0.38kv and the voltage prediction value of each node.

[0202] As an optional implementation, the constraint condition of the distribution network multi-level voltage collaborative control method model based on three layer priority target includes a total power balance constraint, a node admissible net load disturbance domain capacity constraint, an active power flow constraint, a reactive power flow constraint, a node voltage constraint, a unit output constraint, a superior grid power supply constraint, a photovoltaic wind turbine output constraint, a wind power and light power abandonment and load shedding constraint, a capacitor bank constraint, a storage operation constraint, and a static reactive power compensator constraint.

[0203] As an optional implementation, the non-linear terms in the objective function and the non-linear constraints such as the unit operation cost, the branch capacity constraint, and the unit capacity constraint are linearized by a linearization method, the uncertainty variables in the constraints are processed based on the affine strategy, and the model is further converted into a determinacy model based on the duality theory, so that the entire model is converted into a determinacy linear programming model. The mature commercial solver can be used for solving to obtain the distribution network optimization dispatching scheme.

[0204] As an alternative implementation, the maximum node-admissible net load disturbance domain includes an upward-admissible disturbance domain and a downward-admissible disturbance domain; the total power distribution network operation cost includes an operation cost, a wind and light curtailment and load shedding cost, and an energy storage operation cost; the minimum expected deviation includes a total expected deviation of the power distribution network 35kv-10kv-0.38kv three-level voltage.

[0205] As an alternative implementation, the specific process of solving includes:

[0206] (1) The amplitude and phase angle of branch flow and node voltage are quantitatively represented using a power generation load transfer factor based on decoupled linearized power flow, a calculation formula for representing the amplitude and phase angle of branch flow and node voltage using the power generation load transfer factor is obtained, and the requirement of accurately considering the voltage amplitude change and reactive power change of the power distribution network is met.

[0207] (2) Affine strategy and duality theory are used to eliminate uncertain variables in the model, and the node-admissible net load disturbance domain is quantitatively represented, so that the model of the present disclosure is converted into a deterministic model.

[0208] (3) The square term in the branch capacity constraint and the unit capacity constraint is linearized using the power circle linearization method, and the regulation times constraint of the capacitor bank is linearized using the large M method, to obtain the corresponding linear expression. Thus, the optimal scheduling model is rewritten as a linear programming model, and a linear programming solver is used for solving, wherein the linearization diagram of the branch capacity constraint and the unit capacity constraint is shown in FIG. 5.

[0209] The present application takes maximizing the node-admissible net load disturbance domain, minimizing the total operation cost of the power distribution network, and minimizing the expected voltage deviation as three-layer objective functions, which can more clearly depict the uncertainty of distributed power output and the influence of distributed power output on system reserve capacity, and improve the problem of fuzzy description of distributed power output characteristics; the voltage deviation of the power distribution network is optimized, and the problems of multi-level voltage out-of-limit and voltage fluctuation of the power distribution network caused by large-scale access of distributed power are effectively improved; the safety, economy and reliability of the power distribution network operation are considered as a whole, so that the optimization result is more practical in engineering.

[0210] Embodiment 3:

[0211] Based on the same inventive concept, the present application also provides a power distribution network multi-level voltage coordinated control system based on three-layer priority objectives, which includes:

[0212] The parameter acquisition module is configured to acquire power distribution network operation parameters.

[0213] a target solving module, configured to solve a three-layer optimization target programming model based on the power distribution network operation parameters and the three-layer optimization target programming model, and to obtain a maximum node admissible net load disturbance domain, a minimum total power distribution network operation cost, and a minimum expected voltage deviation by using affine theory, dual theory, power circle linearization, and absolute value linearization methods;

[0214] a control module, configured to perform coordinated control on various reactive power devices of the power distribution network by using the power distribution network operation parameters corresponding to the maximum node admissible net load disturbance domain, the minimum total power distribution network operation cost, and the minimum expected voltage deviation.

[0215] The three-layer optimization target programming model is constructed by taking the maximum node admissible net load disturbance domain, the minimum total power distribution network operation cost, and the minimum expected voltage deviation as three-layer objective functions, and setting constraint conditions for the three-layer objective functions.

[0216] Further, the model construction module is further configured to:

[0217] construct a first-layer optimization target by taking the maximum node admissible net load disturbance domain;

[0218] construct a second-layer optimization target by taking the minimum total power distribution network operation cost;

[0219] construct a third-layer optimization target by taking the minimum expected voltage deviation;

[0220] set constraint conditions for the first-layer optimization target, the second-layer optimization target, and the third-layer optimization target.

[0221] The constraint conditions include constraint conditions under a net load prediction value and constraint conditions under a net load disturbance.

[0222] Further, the constraint conditions under the net load prediction value include total power balance constraint conditions under the net load prediction value, line flow constraint conditions, branch capacity constraint conditions, node voltage upper and lower limit constraint conditions, unit output constraint conditions, energy storage constraint conditions, upper-level power grid power supply constraint conditions, light and wind curtailment constraint conditions, load shedding constraint conditions, grouping switched capacitor constraint conditions, static var compensator (SVC) constraint conditions.

[0223] The constraint conditions under the net load disturbance include total power balance constraint conditions under the net load disturbance, net load admissible domain constraint conditions, line flow constraint conditions, line capacity constraint conditions, node voltage upper and lower limit constraint conditions, generator-related constraint conditions, energy storage constraint conditions, upper-level power grid power supply constraint conditions, light and wind curtailment constraint conditions, load shedding constraint conditions, grouping switched capacitor constraint conditions, and static var compensator (SVC) constraint conditions.

[0224] Further, the first-layer optimization target is as shown in the following formula:

[0225] Z1 = min (C is the upward node admissible net load disturbance domain, is the downward node admissible net load disturbance domain, NT represents a set of optimization periods; NI represents a set of nodes; i is a node number; t is an optimization period;

[0226] The second layer optimization objective is as follows: Z2 = min (C oper + C cut + C ess )

[0227] Z2 is a second layer objective function, C oper is an operation cost, C cut is a wind and light curtailment load cost, C ess is an energy storage cost;

[0228] The third layer optimization objective is as follows:

[0229] Z3 is a third layer objective function, is a node voltage prediction value, V i,t is a node actual voltage.

[0230] Further, the target solving module comprises:

[0231] A deterministic transformation submodule is configured to transform the uncertainty formula in the three-layer optimization objective programming model into a deterministic formula by using affine theory and duality theory;

[0232] A linearization submodule is configured to transform the nonlinear formula in the three-layer optimization objective programming model into a linear formula by using power circle linearization and absolute value linearization methods;

[0233] A solving submodule is configured to solve the transformed three-layer optimization objective programming model to obtain the maximum node admissible net load disturbance domain, the minimum total distribution network operation cost, and the minimum expected voltage deviation.

[0234] Further, the deterministic transformation submodule is specifically configured to:

[0235] The affine theory is used to transform the node voltage uncertainty in the three-layer optimization objective programming model by introducing an auxiliary variable to obtain a deterministic variable;

[0236] The duality theory is used to transform the unit regulation disturbance output uncertainty to obtain a node admissible net load disturbance domain.

[0237] Further, the linearization submodule is specifically configured to:

[0238] The uncertainty formula in the three-layer optimization objective programming model which can be segmented linearized is linearized by using a power circle linearization method.

[0239] An absolute value linearization method is used to linearize the absolute value in the uncertainty formula.

[0240] Further, the specific implementation steps of linearizing the uncertainty formula in the three-layer optimization objective programming model which can be segmented linearized by using a power circle linearization method in the linearization submodule include:

[0241] The feasible region of the uncertainty formula is the interior of a circle, and a circle-inscribed regular polygon is approximated as the circle corresponding to the feasible region of the uncertainty formula.

[0242] The area surrounded by the regular polygon is approximated instead of the area surrounded by the circle, and the uncertainty formula is converted into an area calculation formula of the regular polygon.

[0243] Implementation: 4:

[0244] Based on the same inventive concept, the application further provides a computer device, which comprises a processor and a memory, the memory is used to store a computer program, the computer program comprises program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to realize the corresponding method flow or corresponding function, so as to realize the steps of the power distribution network multi-level voltage coordination control method based on three-layer priority objectives in the above-mentioned embodiments.

[0245] Embodiment 5:

[0246] Based on the same inventive concept, the application further provides a storage medium, specifically a computer readable storage medium (Memory), which is a memory device in a computer device and is used to store programs and data. It can be understood that the computer readable storage medium herein can include an internal storage medium in the computer device, and of course can also include an extended storage medium supported by the computer device. The computer readable storage medium provides a storage space, which stores an operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that the computer readable storage medium herein can be a high-speed RAM memory or a non-volatile memory such as at least one disk memory. One or more instructions stored in the computer readable storage medium can be loaded and executed by the processor to implement the steps of the power distribution network multi-level voltage coordination control method based on the three-layer priority target in the above embodiment.

[0247] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0248] The present application is described with reference to flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a device that implements the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0249] These computer program instructions can also be stored in a computer readable memory that can guide the computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including instruction devices that implement the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0250] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable devices provide steps for implementing the functions specified in one or more flows in the flowchart and / or one or more blocks in the block diagram.

[0251] The above is only an embodiment of the present application and is not used to limit the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the scope of the claims of the present application. Industrial applicability

[0252] The application provides a power distribution network multi-level voltage collaborative control method and system based on three-layer priority targets, and the method comprises the following steps: acquiring power distribution network operation parameters; based on the power distribution network operation parameters and a pre-constructed three-layer optimization target planning model, the three-layer optimization target planning model is solved by using an affine theory, a dual theory, a power circle linearization method and an absolute value linearization method, to obtain a maximum node acceptable net load disturbance domain, a minimum power distribution network operation total cost and a minimum expected voltage deviation; and various reactive power equipment of the power distribution network are collaboratively controlled by the power distribution network operation parameters corresponding to the maximum node acceptable net load disturbance domain, the minimum power distribution network operation total cost and the minimum expected voltage deviation. The application can more clearly depict the uncertainty of distributed power output and the influence of the distributed power output on system reserve capacity, and improve the problems of multi-level voltage out-of-limit and voltage fluctuation of the power distribution network caused by the fuzzy description of distributed power output characteristics and large-scale access.

Claims

1. A multi-level voltage coordinated control method for a power distribution network based on three-layer priority objectives, comprising: obtaining power distribution network operation parameters; based on the power distribution network operation parameters and a pre-constructed three-layer optimization objective planning model, using affine theory, duality theory, power circle linearization and absolute value linearization method to solve the three-layer optimization objective planning model to obtain a maximum node admissible net load disturbance domain, a minimum total power distribution network operation cost and a minimum expected voltage deviation; coordinately controlling various reactive power equipment of the power distribution network by the power distribution network operation parameters corresponding to the maximum node admissible net load disturbance domain, the minimum total power distribution network operation cost and the minimum expected voltage deviation; wherein the three-layer optimization objective planning model is constructed by taking the maximum node admissible net load disturbance domain, the minimum total power distribution network operation cost and the minimum expected voltage deviation as three-layer objective functions, and setting constraint conditions for the three-layer objective functions. 2.The method of claim 1, wherein the construction of the three-layer optimization objective planning model comprises: constructing a first-layer optimization objective with the maximum node admissible net load disturbance domain; constructing a second-layer optimization objective with the minimum total power distribution network operation cost; constructing a third-layer optimization objective with the minimum expected voltage deviation; setting constraint conditions for the first-layer optimization objective, the second-layer optimization objective and the third-layer optimization objective; the constraint conditions include: constraints under a net load prediction value and constraints under a net load disturbance.

3. The method of claim 2, the constraint at the net load forecast value comprising: total power balance constraints, line flow constraints, branch capacity constraints, node voltage upper and lower limit constraints, unit output constraints, energy storage constraints, upper-level power grid power supply constraints, light and wind curtailment, load shedding constraints, grouped switched capacitor constraints, static var compensator (SVC) constraints under the net load prediction value; the constraints under the net load disturbance include: total power balance constraints, net load admissible domain constraints, line flow constraints, line capacity constraints, node voltage upper and lower limit constraints, generator-related constraints, energy storage constraints, upper-level power grid power supply constraints, light and wind curtailment, load shedding constraints, grouped switched capacitor constraints, static var compensator (SVC) constraints under the net load disturbance.

4. The method as described in claim 2, wherein the first-layer optimization objective is as follows: wherein Z1 is a first layer objective function, to an upper node admissible net load perturbation domain, for a downward node admissible net load disturbance domain, NT represents a set of optimization periods; NI represents a set of nodes, i is a node number; t is an optimization period; the second-layer optimization objective is shown in the following formula: Z2 = min(C oper + C cut + C ess ) In the formula, Z2 is a second layer target function, C oper is a running cost, C cut is a wind and light abandoned load cutting cost, C ess is an energy storage cost; The third layer optimization objective is shown in the following formula: In the formula, Z3 is a third layer objective function, Vpred is the node voltage prediction value i,t Vact is the node actual voltage. 5.The method of claim 1, wherein based on the power distribution network operation parameters and the pre-constructed three-layer optimization objective planning model, using affine theory, duality theory, power circle linearization and absolute value linearization method to solve the three-layer optimization objective planning model to obtain the maximum node admissible net load disturbance domain, the minimum total power distribution network operation cost and the minimum expected voltage deviation, comprises: using affine theory and duality theory to convert uncertainty formulas in the three-layer optimization objective planning model into deterministic formulas; using power circle linearization and absolute value linearization method to convert nonlinear formulas in the three-layer optimization objective planning model into linear formulas; The three-layer optimization objective programming model after transformation is solved to obtain the maximum node admissible net load disturbance domain, the minimum distribution network operation total cost and the minimum expected voltage deviation.

6. The method of claim 5, wherein the transformation of the uncertainty in the three-layer optimization objective programming model into a deterministic formula using affine theory and duality theory comprises: transforming the node voltage uncertainty in the three-layer optimization objective programming model into a deterministic variable by introducing an auxiliary variable using affine theory; transforming the unit regulation disturbance output uncertainty into a node admissible net load disturbance domain using duality theory.

7. The method of claim 5, wherein the transformation of the non-linear formula in the three-layer optimization objective programming model into a linear formula using power circle linearization and absolute value linearization comprises: linearizing the uncertainty formula in the three-layer optimization objective programming model that can be segmented and linearized using power circle linearization; linearizing the absolute value in the uncertainty formula using absolute value linearization.

8. The method of claim 7, wherein the linearization of the uncertainty formula in the three-layer optimization objective programming model that can be segmented and linearized using power circle linearization comprises: the feasible region of the uncertainty formula is the interior of a circle, and a regular polygon inscribed in the circle is approximated as the circle corresponding to the feasible region of the uncertainty formula; the area surrounded by the regular polygon is approximated to replace the area surrounded by the circle, and the uncertainty formula is transformed into an area calculation formula of the regular polygon.

9. A distribution network multi-level voltage collaborative control system based on three-layer priority objectives, comprising: a parameter acquisition module configured to acquire distribution network operation parameters; a target solving module configured to solve the three-layer optimization objective programming model based on the distribution network operation parameters and a pre-constructed three-layer optimization objective programming model, and to obtain the maximum node admissible net load disturbance domain, the minimum distribution network operation total cost and the minimum expected voltage deviation by using affine theory, duality theory, power circle linearization and absolute value linearization; a control module configured to collaboratively control various reactive power devices of the distribution network according to the distribution network operation parameters corresponding to the maximum node admissible net load disturbance domain, the minimum distribution network operation total cost and the minimum expected voltage deviation; wherein the three-layer optimization objective programming model is constructed by taking the maximum node admissible net load disturbance domain, the minimum distribution network operation total cost and the minimum expected voltage deviation as three-layer objective functions, and setting constraint conditions for the three-layer objective functions.

10. The system of claim 9, further comprising a model construction module configured to: construct a first-layer optimization objective by taking the maximum node admissible net load disturbance domain; construct a second-layer optimization objective by taking the minimum distribution network operation total cost; construct a third-layer optimization objective by taking the minimum expected voltage deviation; set constraint conditions for the first-layer optimization objective, the second-layer optimization objective and the third-layer optimization objective; the constraint conditions comprise constraints under the net load prediction value and constraints under the net load disturbance.

11. A computer device comprising: at least one processor and a memory; The memory, configured to store one or more programs; When the one or more programs are executed by the at least one processor, the three-layer priority target-based power distribution network multi-level voltage collaborative control method in any one of claims 1 to 8 is implemented.

12. A computer readable storage medium, having stored thereon a computer program, which when executed by a computer, implements the three-layer priority target-based power distribution network multi-level voltage collaborative control method in any one of claims 1 to 8.

Images