Method, device and equipment for improving photovoltaic bearing capacity of power distribution system and medium

Abstract

The invention discloses a method, a device and equipment for improving photovoltaic bearing capacity of a power distribution system and a medium. The method comprises the following steps: constructing a deterministic optimization model based on data of the power distribution system by taking maximization of total photovoltaic yield as a target; constructing a first uncertainty set to describe the correlation between the photovoltaic efficiency coefficients and the uncertainty of the load; constructing an initial robust HC model based on the deterministic optimization model and the first uncertainty set; adopting a plurality of HC improvement schemes, and constructing a plurality of robust HC models based on the initial robust HC model; solving the plurality of robust HC models to obtain a plurality of HC results; determining an adjustable robust ANM scheme based on the plurality of HC results; and adjusting the power distribution system based on the adjustable robust ANM scheme so as to improve the photovoltaic bearing capacity of the power distribution system. According to the method, the power distribution system is adjusted by determining the adjustable robust ANM scheme in the plurality of HC improvement schemes, so that the photovoltaic bearing capacity of the power distribution system can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of distributed power generation technology, and in particular to a method, device, equipment and medium for improving the photovoltaic carrying capacity of a power distribution system. Background Art

[0002] The high penetration rate of distributed generation (DG) units may cause problems such as overvoltage or line and transformer overload in the distribution system. These problems will reduce the distribution system's ability to accommodate DG units, so it is necessary to improve the distribution system's photovoltaic carrying capacity, and it is necessary to take into account the time-varying nature of power demand and DG output. Most of the existing technologies use robust optimization (RO) methods to deal with the uncertainty in the carrying capacity (HC) problem, or implement the evaluation method of the distribution system HC based on the robust optimal operation of on-load tap changers (OLTCs) and static VAR compensators. Some also use two-stage adjustable robust optimization to deal with the uncertainty of grid demand and photovoltaic output, as well as robust optimization allocation methods for battery energy storage systems to improve the HC of the distribution system.

[0003] Existing studies usually assume that DG units cannot take recourse actions to deal with uncertainties, or assume that DG communicates with central entities in real time, but in reality, if these two conditions are lacking, the model will be difficult to establish. In addition, existing robust methods ignore uncertainty correlation and only assign an uncertainty set to each uncertainty parameter, but in reality, there is a strong spatial correlation between renewable energy sources such as photovoltaics. Some existing studies have developed a hyper-ellipsoid uncertainty set to consider uncertainty characteristics, taking into account the intrinsic correlation between PV outputs. However, the introduction of ellipsoidal uncertainty sets in linear programming leads to quadratic programming, which imposes strict restrictions on the dimension of the problem and may not be solved using a solver. In addition, ANM schemes such as on-load tap changer (OLTC) regulation, reactive power compensation, and actual power abandonment of inverters are effective means to expand the capacity of the distribution network, and these plans should not be ignored in HC improvement methods. However, the literature on HC either ignores ANM schemes or requires communication infrastructure to centrally control the equipment in real time. Summary of the invention

[0004] The present invention provides a method, device, equipment and medium for improving the photovoltaic carrying capacity of a distribution system to solve the above-mentioned technical problems and effectively improve the photovoltaic carrying capacity of the distribution system.

[0005] In order to solve the above technical problems, the present invention provides a method for improving the photovoltaic carrying capacity of a power distribution system, comprising:

[0006] Obtain distribution system data, and build a deterministic optimization model based on the distribution system data with the goal of maximizing the total photovoltaic output;

[0007] The first uncertainty set is constructed to describe the correlation between the PV efficiency coefficients and the uncertainty of the loads;

[0008] An initial robust HC model is constructed based on the deterministic optimization model and the first uncertainty set;

[0009] Several HC improvement schemes are adopted to construct several robust HC models based on the initial robust HC model;

[0010] Solve several robust HC models and obtain several HC results;

[0011] Based on several HC results, an adjustable robust ANM scheme is determined;

[0012] The distribution system is adjusted based on an adjustable robust ANM scheme to improve the photovoltaic carrying capacity of the distribution system.

[0013] The beneficial effects of the present invention are:

[0014] The present invention adopts several HC improvement schemes, and obtains several robust HC models based on the initial robust HC model respectively. By solving several HC results obtained by the several robust HC models, an adjustable robust ANM scheme is determined. Finally, the distribution system is adjusted based on the adjustable robust ANM scheme to improve the photovoltaic carrying capacity of the distribution system. Therefore, the present invention can optimize the robust operation of on-load tap changers and inverters by determining to adjust the distribution system based on the adjustable robust ANM scheme among several HC improvement schemes, and take real-time recourse actions after discovering the uncertainty of the distribution system, thereby effectively improving the photovoltaic carrying capacity of the distribution system.

[0015] As a preferred solution, the deterministic optimization model is constructed based on the distribution system data with the goal of maximizing the total photovoltaic output, including:

[0016] Based on the distribution system data, an objective function is constructed with the goal of maximizing the total photovoltaic output and a first set of constraints is determined; the objective function is specifically:

[0017]

[0018] In the above formula, i is the node number; is the set of nodes with power consumption or injection; t represents the time interval; is the set of t; is the predicted efficiency coefficient of photovoltaic power generation at node i at time t, which is defined as the ratio of actual photovoltaic power generation to installed capacity; G i represents the photovoltaic installed capacity corresponding to the i-node; Indicates the actual amount of power abandoned corresponding to node i;

[0019] A deterministic optimization model is constructed based on the objective function and the first set of constraints.

[0020] As a preferred solution, the first set of constraints includes photovoltaic energy abandonment constraints, voltage constraints, inverter output constraints, photovoltaic installed capacity constraints, grid constraints and branch thermal constraints;

[0021] The photovoltaic energy curtailment constraints are specifically:

[0022]

[0023] In the above formula, Δt is the duration of the time interval; γ is a constant, which represents the energy allowed to be reduced relative to the total power generation;

[0024] The voltage constraint is specifically:

[0025]

[0026] In the above formula, U and is the square of the acceptable upper and lower voltage limits; where U0 = |V0| 2 is the square of the equilibrium node voltage; p it is the active power consumed by node i at time t; q it is the reactive power consumed by node i at time t; U j is the square of the voltage at node j; and are the sensitivities of the square of the voltage at node j to the active power and reactive power consumed at node i, respectively; where superscript d and superscript g correspond to the demand and power generation, respectively; θ i is the power factor angle of node i;

[0027] The inverter output constraints are specifically:

[0028]

[0029] In the above formula, is the active power generation of node i in the time interval t; is the reactive power generation of node i in the time interval t;

[0030] The photovoltaic installed capacity constraints are specifically:

[0031]

[0032] In the above formula, G i and are the lower and upper bounds of the PV installed capacity of node i, respectively;

[0033] The grid constraints are specifically:

[0034]

[0035] In the above formula, U 0t represents the square of the voltage amplitude of the equilibrium node in the time interval t;

[0036] The branch thermal constraints are specifically:

[0037]

[0038] In the above formula, P ij,t and Q ij,t are the active power and reactive power of the branch between node i and node j in time interval t respectively; S ij is the power of the branch between node i and node j.

[0039] The above scheme is based on the distribution system data, constructs an objective function and determines the first set of constraints with the goal of maximizing the total photovoltaic output, and constructs a deterministic optimization model based on the objective function and the first set of constraints. Based on the deterministic optimization model, the photovoltaic carrying capacity (HC) in the distribution system can be determined.

[0040] As a preferred solution, the first uncertainty set includes a polyhedral uncertainty set and a hypercube uncertainty set;

[0041] The polyhedral uncertainty set is used to describe the correlation between photovoltaic efficiency coefficients; the polyhedral uncertainty set is specifically:

[0042]

[0043] In the above formula, ψ + and ψ - represents the wave vector in the positive and negative directions, It is the upper limit of the efficiency coefficient deviation in the positive and negative directions;

[0044] The hypercube uncertainty set is used to describe the uncertainty of the load; the hypercube uncertainty set is specifically:

[0045]

[0046] In the above formula, p d+ and p d- Represents the predicted value The up and down fluctuation vectors; is the upper bound of the fluctuation value.

[0047] In this preferred embodiment, the first uncertainty set includes a polyhedron uncertainty set and a hypercube uncertainty set, wherein the polyhedron uncertainty set is used to describe the correlation between photovoltaic efficiency coefficients, and the hypercube uncertainty set is used to describe the uncertainty of the load, so that the first uncertainty set has the uncertainty characteristics of solar power generation and load.

[0048] As a preferred solution, the constructing of the initial robust HC model based on the deterministic optimization model and the first uncertainty set includes:

[0049] The initial robust HC model is specifically:

[0050]

[0051] In the above formula, x=[G,δ cur ,q g ,U oltc ] is the set of all decision variables, where G is the photovoltaic installed capacity, δ cur is the actual amount of power abandoned, q g is the reactive power generation, U oltc is the square of the node voltage with the on-load tapchanger installed; ξ=[ψ + ,ψ - ,p d+ ,p d- ] is the set of all uncertain parameters, where ψ + and ψ - represents the wave vector in the positive and negative directions, p d+ and p d- Represents the predicted value The up and down fluctuation vectors; is the first uncertainty set; f(x,ξ) is the objective function; A1, A0, B1, b0, M and m are all constant parameters, where M and m are obtained based on the first uncertainty set; is the expected value of the corresponding vector.

[0052] This preferred solution constructs an initial robust HC model based on the deterministic optimization model and the first uncertainty set, so that the initial robust HC model takes into account the uncertainty characteristics of solar power generation and load.

[0053] As a preferred solution, the method adopts several HC improvement solutions to construct several robust HC models based on the initial robust HC model, including:

[0054] Based on the adjustable robust ANM scheme, an adjustable variable is obtained according to the initial robust HC model; the adjustable variable is represented as an affine function of the uncertain parameters;

[0055] The constraints of the worst uncertainty realization in the first uncertainty set are robustified and dualized using the maximum protection function to obtain linear inequality constraints.

[0056] Based on the initial robust HC model, adjustable variables and linear inequality constraints, a first robust HC model corresponding to the adjustable robust ANM solution is obtained; the first robust HC model is specifically:

[0057]

[0058] In the above formula, g1 and g0 are linear functions; x u is an unadjustable variable in the decision variable; S represents the sample space; s represents the sth sample in the sample space; λ is the first uncertainty set is the vector of dual variables associated with the boundary constraints; α and β are the parameters in the affine function.

[0059] This preferred scheme is based on an adjustable robust ANM scheme and an initial robust HC model to obtain a first robust HC model corresponding to the adjustable robust ANM scheme, wherein the adjustable variables are restricted to affine functions of uncertain parameters, which can make the uncertainty problem computationally easy to handle, and utilize the maximum protection function to perform robustification and dual transformation on the constraints of the worst uncertainty realization in the first uncertainty set to obtain linear inequality constraints, so as to avoid the situation where any uncertainty realization in the first uncertainty set needs to satisfy the constraints that have not been robustified and dual transformed, resulting in an infinite number of constraints.

[0060] Accordingly, in order to solve the above technical problems, the present invention also provides a device for improving photovoltaic carrying capacity of a power distribution system, comprising: a first building module, a second building module, a model solving module and a system adjustment module;

[0061] The first model building module is used to obtain distribution system data, and based on the distribution system data, build a deterministic optimization model with the goal of maximizing the total photovoltaic output; build a first uncertainty set to describe the correlation between photovoltaic efficiency coefficients and the uncertainty of loads; and build an initial robust HC model based on the deterministic optimization model and the first uncertainty set;

[0062] The second construction module is used to adopt several HC improvement schemes to construct several robust HC models based on the initial robust HC model respectively;

[0063] The model solving module is used to solve several robust HC models to obtain several HC results;

[0064] The system adjustment module is used to determine an adjustable robust ANM solution based on a number of HC results; and to adjust the distribution system based on the adjustable robust ANM solution to improve the photovoltaic carrying capacity of the distribution system.

[0065] As a preferred solution, the first building block includes: a function building unit and a first building unit;

[0066] The objective function construction unit is used to construct an objective function based on the power distribution system data and to determine a first set of constraint conditions with the goal of maximizing the total photovoltaic output; the objective function is specifically:

[0067]

[0068] In the above formula, i is the node number; is the set of nodes with power consumption or injection; t represents the time interval; is the set of t; is the predicted efficiency coefficient of photovoltaic power generation at node i at time t, which is defined as the ratio of actual photovoltaic power generation to installed capacity; G i represents the photovoltaic installed capacity corresponding to the i-node; Indicates the actual amount of power abandoned corresponding to node i;

[0069] The first model building unit is used to build a deterministic optimization model based on the objective function and the first constraint condition set.

[0070] The present invention also provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for improving the photovoltaic carrying capacity of a distribution system as described in any one of the above items are implemented.

[0071] The present invention also provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for improving the photovoltaic carrying capacity of a power distribution system as described in any one of the above items are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 : A flow chart of an embodiment of a method for improving photovoltaic carrying capacity of a power distribution system provided by the present invention;

[0073] Figure 2 : A schematic diagram of the change of HC results with the photovoltaic efficiency coefficient and the forecast deviation in different schemes under different upper bounds of the photovoltaic efficiency coefficient deviating in the positive and negative directions provided by the embodiment of the present invention;

[0074] Figure 3: A schematic diagram of net photovoltaic power generation when the actual power cutoff of the inverter is non-adjustable (RO) and adjustable (ARC) provided in an embodiment of the present invention;

[0075] Figure 4 : A schematic structural diagram of an embodiment of a device for improving photovoltaic carrying capacity of a power distribution system provided by the present invention;

[0076] Figure 5 : A structural diagram of an embodiment of the first building block provided by the present invention. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] Existing studies usually assume that DG units cannot take recourse actions to deal with uncertainties, or assume that DG communicates with central entities in real time, but in reality, if these two conditions are lacking, the model will be difficult to establish. In addition, existing robust methods ignore uncertainty correlation and only assign an uncertainty set to each uncertainty parameter, but in reality, there is a strong spatial correlation between renewable energy sources such as photovoltaics. Some existing studies have developed a hyper-ellipsoid uncertainty set to consider uncertainty characteristics, taking into account the intrinsic correlation between PV outputs. However, the introduction of ellipsoidal uncertainty sets in linear programming leads to quadratic programming, which imposes strict restrictions on the dimension of the problem and may not be solved using a solver. In addition, ANM schemes such as on-load tap changer (OLTC) regulation, reactive power compensation, and actual power abandonment of inverters are effective means to expand the capacity of the distribution network, and these plans should not be ignored in HC improvement methods. However, the literature on HC either ignores ANM schemes or requires communication infrastructure to centrally control the equipment in real time.

[0079] Embodiment 1

[0080] In order to solve the above technical problems, the present invention provides a method, device, equipment and medium for improving the photovoltaic carrying capacity of a distribution system to solve the above technical problems and effectively improve the photovoltaic carrying capacity of the distribution system.

[0081] Please refer to Figure 1 , is a flow chart of an embodiment of a method for improving photovoltaic carrying capacity of a power distribution system provided by an embodiment of the present invention. Figure 1As shown, the method for improving the photovoltaic carrying capacity of the power distribution system includes steps 101 to 107, specifically:

[0082] Step 101: obtaining distribution system data, and building a deterministic optimization model based on the distribution system data with the goal of maximizing total photovoltaic output;

[0083] Step 102: construct a first uncertainty set to describe the correlation between photovoltaic efficiency coefficients and the uncertainty of loads;

[0084] Step 103: constructing an initial robust HC model based on the deterministic optimization model and the first uncertainty set;

[0085] Step 104: adopting several HC improvement schemes to construct several robust HC models based on the initial robust HC model respectively;

[0086] Step 105: solving several robust HC models to obtain several HC results;

[0087] Step 106: Determine an adjustable robust ANM solution based on the plurality of HC results;

[0088] Step 107: Adjust the distribution system based on the adjustable robust ANM solution to improve the photovoltaic carrying capacity of the distribution system.

[0089] The present invention adopts several HC improvement schemes, and obtains several robust HC models based on the initial robust HC model respectively. By solving several HC results obtained by the several robust HC models, an adjustable robust ANM scheme is determined. Finally, the distribution system is adjusted based on the adjustable robust ANM scheme to improve the photovoltaic carrying capacity of the distribution system. Therefore, the present invention can optimize the robust operation of on-load tap changers and inverters by determining to adjust the distribution system based on the adjustable robust ANM scheme among several HC improvement schemes, and take real-time recourse actions after discovering the uncertainty of the distribution system, thereby effectively improving the photovoltaic carrying capacity of the distribution system.

[0090] In step 101, the power distribution system data is obtained, specifically:

[0091] The distribution system data of historical time is obtained, and the distribution system data includes the set of nodes of the distribution system, node voltage, node active power, node reactive power, actual photovoltaic power generation, photovoltaic installed capacity, actual abandoned power, active power generation and active power of branches between nodes, etc.

[0092] Initialization, input the parameters of the IEEE 37-bus system, the input voltage fluctuation range is [0.95, 1.05] pu, and the thermal limit of the transformer is 1.25MVA; input the historical data of the hourly photovoltaic efficiency coefficient (PV efficiency coefficient) to generate a 30-year hourly location-related training PV efficiency coefficient dataset, and select 32 locations close to their PV efficiency coefficient spatial correlation.

[0093] In this embodiment, the subsequently constructed model can be simulated on the 37-node system in initialization to test the effectiveness of the present invention, and the data input in the initialization is used to solve the model constructed based on the 37-node system.

[0094] Furthermore, in step 101, the deterministic optimization model is constructed based on the power distribution system data with the goal of maximizing the total photovoltaic output, including:

[0095] Based on the distribution system data, an objective function is constructed with the goal of maximizing the total photovoltaic output and a first set of constraints is determined; the objective function is specifically:

[0096]

[0097] In the above formula, i is the node number; is the set of nodes with power consumption or injection; t represents the time interval; is the set of t; is the predicted efficiency coefficient of photovoltaic power generation at node i at time t, which is defined as the ratio of actual photovoltaic power generation to installed capacity; G i represents the photovoltaic installed capacity corresponding to the i-node; Indicates the actual amount of power abandoned corresponding to node i;

[0098] A deterministic optimization model is constructed based on the objective function and the first set of constraints.

[0099] Furthermore, the first set of constraints includes photovoltaic energy curtailment constraints, voltage constraints, inverter output constraints, photovoltaic installed capacity constraints, grid constraints and branch thermal constraints;

[0100] The photovoltaic energy curtailment constraints are specifically:

[0101]

[0102] In the above formula, t represents the time interval; is the set of t; represents the actual amount of power abandoned corresponding to node i; Δt is the duration of the time interval; γ is a constant, which represents the energy allowed to be reduced relative to the total power generation; is the predicted efficiency coefficient of photovoltaic power generation at node i at time t, which is defined as the ratio of actual photovoltaic power generation to installed capacity; G i represents the photovoltaic installed capacity corresponding to the i-node;

[0103] The voltage constraint is specifically:

[0104]

[0105] In the above formula, U and is the square of the acceptable upper and lower voltage limits; where U0 = |V0| 2 is the square of the equilibrium node voltage; p it is the active power consumed by node i at time t; q it is the reactive power consumed by node i at time t; U j is the square of the voltage at node j; and are the sensitivities of the square of the voltage at node j to the active power and reactive power consumed at node i, respectively; where superscript d and superscript g correspond to the demand and power generation, respectively; θ i is the power factor angle of node i;

[0106] The inverter output constraint includes the actual output constraint of the photovoltaic inverter, the output power does not exceed the photovoltaic power generation power constraint and the thermal limit constraint of the inverter; the inverter output constraint is specifically:

[0107]

[0108] In the above formula, is the active power generation of node i in the time interval t; is the reactive power generation of node i in the time interval t; Indicates the actual amount of power abandoned corresponding to node i; is the predicted efficiency coefficient of photovoltaic power generation at node i at time t, which is defined as the ratio of actual photovoltaic power generation to installed capacity; G i represents the photovoltaic installed capacity corresponding to the i-node;

[0109] The photovoltaic installed capacity constraints are specifically:

[0110]

[0111] In the above formula, G i and are the lower and upper bounds of the PV installed capacity of node i, respectively;

[0112] The grid constraint is to establish an OLTC voltage control model on a transformer connecting the balancing node to the upstream network; the grid constraint is specifically:

[0113]

[0114] In the above formula, U 0t represents the square of the voltage amplitude of the equilibrium node in the time interval t;

[0115] The branch thermal constraints are specifically:

[0116]

[0117] In the above formula, P ij,t and Q ij,t are the active power and reactive power of the branch between node i and node j in time interval t respectively; S ij is the power of the branch between node i and node j.

[0118] The above scheme is based on the distribution system data, constructs an objective function and determines the first set of constraints with the goal of maximizing the total photovoltaic output, and constructs a deterministic optimization model based on the objective function and the first set of constraints. Based on the deterministic optimization model, the photovoltaic carrying capacity (HC) in the distribution system can be determined.

[0119] Further, in step 102, the first uncertainty set includes a polyhedral uncertainty set and a hypercube uncertainty set;

[0120] The polyhedral uncertainty set is used to describe the correlation between photovoltaic efficiency coefficients; the polyhedral uncertainty set is specifically:

[0121]

[0122] In the above formula, ψ + and ψ - represents the wave vector in the positive and negative directions, It is the upper limit of the efficiency coefficient deviation in the positive and negative directions;

[0123] The hypercube uncertainty set is used to describe the uncertainty of the load; the hypercube uncertainty set is specifically:

[0124]

[0125] In the above formula, p d+ and p d- Represents the predicted value The up and down fluctuation vectors; is the upper bound of the fluctuation value.

[0126] The first uncertainty set in the present invention includes a polyhedral uncertainty set and a hypercube uncertainty set, wherein the polyhedral uncertainty set is used to describe the correlation between photovoltaic efficiency coefficients (photovoltaic uncertainty), and the hypercube uncertainty set is used to describe the uncertainty of the load, so that the first uncertainty set has the uncertainty characteristics of solar power generation and load. Therefore, the present invention not only considers the spatial correlation between photovoltaic outputs in the same geographical area, but also can be effectively applied to large-scale problems of power systems, while significantly reducing the computational burden compared with other methods.

[0127] In this embodiment, the step of establishing the polyhedron uncertainty set includes:

[0128] First, the correlation between the photovoltaic efficiency coefficient (PV efficiency coefficient) of the distribution system is described by a hyperellipsoid uncertainty set as follows:

[0129]

[0130] Where: l is the number of nodes including photovoltaic panels; Δψ=ψ + -ψ - is the photovoltaic efficiency coefficient with the predicted value The change in fluctuation, ψ + and ψ - Represents the wave vector in the positive and negative directions;

[0131] represents the covariance matrix of historical data, Y v is the uncertainty budget for the predefined coverage v;

[0132] make get:

[0133]

[0134] Transformation yields:

[0135]

[0136] The uncertainty set is reconstructed using the polyhedron approximation method of the second-order cone as follows:

[0137]

[0138] In the above formula, u is the set of all auxiliary variables; H[] is the second-order cone polyhedron approximation function; h is a constant, h∈[1,∞];

[0139] Therefore, the polyhedral uncertainty set is established as follows:

[0140]

[0141] In the above formula, It is the upper limit of the deviation of the photovoltaic efficiency coefficient in the positive and negative directions.

[0142] In this embodiment, the step of establishing the hypercube uncertainty set includes:

[0143] First, a hypercubic uncertainty set is used to describe the uncertainty of the load:

[0144]

[0145] In the above formula, p d+ and p d- Represents the predicted value The up and down fluctuation vectors; is the upper bound of these fluctuation values;

[0146] Since the maximum fluctuations of all loads in the electrical system are inconsistent, the uncertainty budget is performed on the hypercube uncertainty set of the above formula, considering the compliance factor in the first formula as follows:

[0147]

[0148] In the above formula, For collection The cardinality, F c ∈[0,1] is a suitable factor.

[0149] Furthermore, in step 103, the initial robust HC model is constructed based on the deterministic optimization model and the first uncertainty set, including:

[0150] The initial robust HC model is specifically:

[0151]

[0152] In the above formula, x=[G,δ cur ,q g ,U oltc ] is the set of all decision variables, where G is the photovoltaic installed capacity, δ cur is the actual amount of power abandoned, q g is the reactive power generation, U oltc is the square of the node voltage with the on-load tapchanger installed; ξ=[ψ + ,ψ - ,p d+ ,p d- ] is the set of all uncertain parameters, where ψ + and ψ - represents the wave vector in the positive and negative directions, p d+ and p d- Represents the predicted value The up and down fluctuation vectors; is the first uncertainty set; f(x,ξ) is the objective function; A1, A0, B1, b0, M and m are all constant parameters, where M and m are obtained based on the first uncertainty set; is the expected value of the corresponding vector.

[0153] In this embodiment, before constructing the initial robust HC model based on the deterministic optimization model and the first uncertainty set, in order to prevent uncertain parameters from appearing in the objective function when establishing the initial robust HC model, Replace with The first uncertainty set and the first formula are considered as constraints of the initial robust HC model.

[0154] The present invention constructs an initial robust HC model based on a deterministic optimization model and a first uncertainty set, so that the initial robust HC model takes into account uncertainty characteristics of solar power generation and load.

[0155] Furthermore, in step 104, the plurality of HC improvement schemes are adopted to construct a plurality of robust HC models based on the initial robust HC model, including:

[0156] Based on the adjustable robust ANM scheme, an adjustable variable is obtained according to the initial robust HC model; the adjustable variable is represented as an affine function of the uncertain parameters;

[0157] The constraints of the worst uncertainty realization in the first uncertainty set are robustified and dualized using the maximum protection function to obtain linear inequality constraints.

[0158] Based on the initial robust HC model, adjustable variables and linear inequality constraints, a first robust HC model corresponding to the adjustable robust ANM solution is obtained; the first robust HC model is specifically:

[0159]

[0160] In the above formula, g1 and g0 are linear functions; x u is an unadjustable variable in the decision variable; S represents the sample space; s represents the sth sample in the sample space; λ is the first uncertainty set is the vector of dual variables associated with the boundary constraints; α and β are the parameters in the affine function.

[0161] The present invention constructs a first robust HC model corresponding to the adjustable robust ANM scheme based on an adjustable robust ANM scheme and an initial robust HC model, wherein the adjustable variables are restricted to affine functions of uncertain parameters, which can make the uncertainty problem computationally easy to handle, and utilizes the maximum protection function to perform robustification and dual transformation on the constraints of the worst uncertainty realization in the first uncertainty set to obtain linear inequality constraints, so as to avoid the situation in which an infinite number of constraints are caused by the need to satisfy the constraints that have not been robustified and dual transformed for any uncertainty realization in the first uncertainty set.

[0162] In this embodiment, an active network management (ANM) solution is added to the established initial robust HC model, specifically:

[0163] Before implementing the decision variable x, many uncertain parameters ξ must be determined. The present invention refers to these decision variables as unadjustable variables and uses x u The rest are variables that can be determined after the implementation of the uncertain parameters, which are called adjustable variables and are represented by x a It means that, thus we get:

[0164]

[0165] In the present invention, the actual amount of power wasted by the inverter δ cur and reactive power generation q g Modeled as adjustable variables; to make the problem computationally tractable, the adjustable variables are restricted to affine functions or piecewise affine functions of uncertain parameters, such as:

[0166] x a =αξ+β;

[0167] Then the initial robust HC model with adjustable ANM scheme is obtained as follows:

[0168]

[0169] In the above formula, S represents the sample space; s represents the sth sample in the sample space; g1 and g0 are linear functions. Note that for the first uncertainty set Any uncertainty realization within needs to satisfy the constraints in the above formula, which will lead to an infinite number of constraints. Therefore, in order to solve this problem, the present invention uses the maximum protection function The constraints of the worst uncertainty realization are robustified as follows:

[0170]

[0171] Then use the dual technique to replace the above equation with a finite set of linear inequality constraints:

[0172]

[0173] In the above formula, λ is the first uncertainty set The vector of dual variables associated with the boundary constraints in ;

[0174] The first robust HC model corresponding to the adjustable robust ANM solution is obtained as follows:

[0175]

[0176] Furthermore, this embodiment combines Figure 2 This is to illustrate an implementation example of solving several robust HC models and obtaining several HC results in step 105. The details are as follows:

[0177] Figure 2 The embodiments of the present invention provide different Schematic diagram of the change of HC results with the photovoltaic efficiency coefficient and the forecast deviation in different schemes under (upper limit of the deviation of photovoltaic efficiency coefficient in positive and negative directions).

[0178] like Figure 2 As shown, in this embodiment, the HC improvement scheme includes: (1) a scheme without ANM; (2) actual power limitation of the inverter, including robust optimization (RO, non-adjustable) and adjustable robust optimization (ARC, adjustable) schemes; (3) inverter reactive power compensation, including RO and ARC schemes; (4) an on-load tap changer (OLTC) scheme; (5) an adjustable robust ANM scheme of the present invention, in which the OLTC is non-adjustable and the inverter is adjustable but uncertain.

[0179] like Figure 2 As shown in the figure, when the inverter has actual power abandonment capability (limited to within 10% of the power generation), the HC of the power grid increases by 26% and 32.23% on average in the cases where the power abandonment is unadjustable and uncertain and adjustable, respectively; on average, adding OLTC or reactive power compensation function in the network can increase the HC by 20.38%; using a robust and unadjustable ANM scheme can increase the capacity of the network by 40.64%, while using an adjustable robust ANM scheme can increase the HC of the IEEE 37-bus system by 46.08%.

[0180] After obtaining the PV installed capacity and inverter control parameters in scheme (2), the net power generation (MWh) during the study period is obtained in real time using 5-minute resolution data to examine the effectiveness of the adjustable wind curtailment scheme. Figure 3 A schematic diagram of net photovoltaic power generation when the actual power cutoff of the inverter is non-adjustable (RO) and adjustable (ARC) provided in an embodiment of the present invention.

[0181] Figure 3The net generation of non-regulatory (RO) and adjustable (ARC) wind curtailment capacity is shown in Figure 1. It can be seen that when the deviation is zero, the two schemes have the same net generation, but when the deviation increases, the net generation of the adjustable scheme is higher; depending on the size of the PV efficiency coefficient deviation, this increase can vary between 0.3-1.5GWh.

[0182] Therefore, based on several HC results, it is determined that the adjustable robust ANM scheme can optimize the improvement effect of the photovoltaic carrying capacity of the distribution system and can best improve the net power generation of the distribution system.

[0183] Furthermore, the distribution system is adjusted based on the adjustable robust ANM scheme, and the active consumption and multi-level coordinated utilization of various types of renewable energy in the distribution system are achieved through active management and optimization of the use of network resources, so as to improve the photovoltaic carrying capacity of the distribution system.

[0184] Accordingly, in order to solve the above technical problems, the present invention also provides a photovoltaic load capacity improvement device for a power distribution system, please refer to Figure 4 , Figure 4 A schematic structural diagram of an embodiment of a device for improving photovoltaic carrying capacity of a power distribution system provided by the present invention.

[0185] like Figure 4 As shown, the photovoltaic carrying capacity improvement device 20 of the power distribution system includes: a first construction module 201, a second construction module 202, a model solving module 203 and a system adjustment module 204;

[0186] The first model building module 201 is used to obtain distribution system data, and based on the distribution system data, build a deterministic optimization model with the goal of maximizing the total photovoltaic output; build a first uncertainty set to describe the correlation between photovoltaic efficiency coefficients and the uncertainty of loads; and build an initial robust HC model based on the deterministic optimization model and the first uncertainty set;

[0187] The second construction module 202 is used to adopt several HC improvement schemes to construct several robust HC models based on the initial robust HC model respectively;

[0188] The model solving module 203 is used to solve several robust HC models to obtain several HC results;

[0189] The system adjustment module 204 is used to determine an adjustable robust ANM solution based on a number of HC results; and adjust the power distribution system based on the adjustable robust ANM solution to improve the photovoltaic carrying capacity of the power distribution system.

[0190] Figure 5 A schematic structural diagram of an embodiment of the first building block provided by the present invention.

[0191] Furthermore, if Figure 5 As shown, the first construction module 201 includes: a function construction unit 2011 and a first construction unit 2012;

[0192] The objective function construction unit 2011 is used to construct an objective function based on the power distribution system data and to determine a first set of constraint conditions with the goal of maximizing the total photovoltaic output; the objective function is specifically:

[0193]

[0194] In the above formula, i is the node number; is the set of nodes with power consumption or injection; t represents the time interval; is the set of t; is the predicted efficiency coefficient of photovoltaic power generation at node i at time t, which is defined as the ratio of actual photovoltaic power generation to installed capacity; G i represents the photovoltaic installed capacity corresponding to the i-node; Indicates the actual amount of power abandoned corresponding to node i;

[0195] The first model building unit 2012 is used to build a deterministic optimization model based on the objective function and the first constraint condition set.

[0196] The present invention also provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for improving the photovoltaic carrying capacity of a distribution system as described in any one of the above items are implemented.

[0197] The present invention also provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for improving the photovoltaic carrying capacity of a power distribution system as described in any one of the above items are implemented.

[0198] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0199] 1. In order to improve the distributed generation carrying capacity of the distribution system to maximize the overall DG power generation, the present invention first takes the maximization of the total photovoltaic output as the objective function, considers the constraints such as voltage constraints, inverter output constraints, photovoltaic installed capacity constraints, etc., establishes a deterministic optimization model to determine the photovoltaic carrying capacity (HC) in the distribution system, and considers the uncertainty characteristics of solar power generation and load to establish an initial robust HC model; then adopts several HC improvement schemes, respectively constructs several robust HC models based on the initial robust HC model, and determines an adjustable robust ANM scheme by solving several HC results obtained by several robust HC models. Finally, the distribution system is adjusted based on the adjustable robust ANM scheme to improve the photovoltaic carrying capacity of the distribution system. Therefore, the present invention can optimize the robust operation of on-load tap changers and inverters by determining to adjust the distribution system based on the adjustable robust ANM scheme among several HC improvement schemes, and takes real-time pursuit actions after discovering the uncertainty of the distribution system, thereby effectively improving the photovoltaic carrying capacity of the distribution system.

[0200] 2. Specifically, considering the limitations in actual situations, the present invention constructs the HC problem based on the adjustable robust optimization (ARC) method, and obtains HC while optimizing the piecewise affine function, which is capable of taking real-time recourse actions after discovering uncertainty factors; and the parameters of the piecewise affine function of the present invention are obtained globally, and can act locally without the need for further real-time communication; the present invention linearizes the ellipsoidal uncertainty set to obtain a polyhedral approximation to develop the polyhedral uncertainty set of DG output, and then in problems with highly correlated uncertain variables, in order to expand the uncertainty set to a larger dimension, an approximation is introduced, and a new polyhedron set is established, which not only considers the correlation of photovoltaic output, but also can be effectively applied to large-scale problems of power systems.

[0201] 3. The first robust HC model based on the adjustable robust ANM solution established by the present invention has coordinated dynamic characteristics, which can allow operators to consider the pursuit behavior of fast-action devices without assuming the existence of a real-time communication infrastructure.

[0202] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for improving photovoltaic carrying capacity of a power distribution system, characterized in that: include: Obtain distribution system data, and build a deterministic optimization model based on the distribution system data with the goal of maximizing the total photovoltaic output; The first uncertainty set is constructed to describe the correlation between the PV efficiency coefficients and the uncertainty of the loads; An initial robust HC model is constructed based on the deterministic optimization model and the first uncertainty set; Several HC improvement schemes are adopted to construct several robust HC models based on the initial robust HC model; Solve several robust HC models and obtain several HC results; Based on several HC results, an adjustable robust ANM scheme is determined; The distribution system is adjusted based on an adjustable robust ANM scheme to improve the photovoltaic carrying capacity of the distribution system.

2. A method for improving photovoltaic carrying capacity of a power distribution system according to claim 1, characterized in that: The deterministic optimization model is constructed based on the distribution system data with the goal of maximizing the total photovoltaic output, including: Based on the distribution system data, an objective function is constructed with the goal of maximizing the total photovoltaic output and a first set of constraints is determined; the objective function is specifically: In the above formula, i is the node number; is the set of nodes with power consumption or injection; t represents the time interval; is the set of t; is the predicted efficiency coefficient of photovoltaic power generation at node i at time t, which is defined as the ratio of actual photovoltaic power generation to installed capacity; G i represents the photovoltaic installed capacity corresponding to the i-node; Indicates the actual amount of power abandoned corresponding to node i; A deterministic optimization model is constructed based on the objective function and the first set of constraints.

3. A method for improving photovoltaic carrying capacity of a power distribution system as claimed in claim 2, characterized in that: The first set of constraints includes photovoltaic energy abandonment constraints, voltage constraints, inverter output constraints, photovoltaic installed capacity constraints, grid constraints and branch thermal constraints; The photovoltaic energy curtailment constraints are specifically: In the above formula, Δt is the duration of the time interval; γ is a constant, which represents the energy allowed to be reduced relative to the total power generation; The voltage constraint is specifically: In the above formula, U and is the square of the acceptable upper and lower voltage limits; where U0 = |V0| 2 is the square of the equilibrium node voltage; p it is the active power consumed by node i at time t; q it is the reactive power consumed by node i at time t; U j is the square of the voltage at node j; and are the sensitivities of the square of the voltage at node j to the active power and reactive power consumed at node i, respectively; where superscript d and superscript g correspond to the demand and power generation, respectively; θ i is the power factor angle of node i; The inverter output constraints are specifically: In the above formula, is the active power generation of node i in the time interval t; is the reactive power generation of node i in the time interval t; The photovoltaic installed capacity constraints are specifically: In the above formula, G i and are the lower and upper bounds of the PV installed capacity of node i, respectively; The grid constraints are specifically: In the above formula, U 0t represents the square of the voltage amplitude of the equilibrium node in the time interval t; The branch thermal constraints are specifically: In the above formula, P ij,t and Q ij,t are the active power and reactive power of the branch between node i and node j in time interval t respectively; S ij is the power of the branch between node i and node j.

4. A method for improving photovoltaic carrying capacity of a power distribution system according to claim 1, characterized in that: The first uncertainty set includes a polyhedral uncertainty set and a hypercube uncertainty set; The polyhedral uncertainty set is used to describe the correlation between photovoltaic efficiency coefficients; the polyhedral uncertainty set is specifically: In the above formula, ψ + and ψ - represents the wave vector in the positive and negative directions, It is the upper limit of the efficiency coefficient deviation in the positive and negative directions; The hypercube uncertainty set is used to describe the uncertainty of the load; the hypercube uncertainty set is specifically: In the above formula, p d+ and p d- Represents the predicted value The up and down fluctuation vectors; is the upper bound of the fluctuation value.

5. A method for improving photovoltaic carrying capacity of a power distribution system as claimed in claim 2, characterized in that: The constructing of an initial robust HC model based on the deterministic optimization model and the first uncertainty set includes: The initial robust HC model is specifically: In the above formula, x=[G,δ cur ,q g ,U oltc ] is the set of all decision variables, where G is the photovoltaic installed capacity, δ cur is the actual amount of power abandoned, q g is the reactive power generation, U oltc is the square of the node voltage with the on-load tapchanger installed; ξ=[ψ + ,ψ - ,p d+ ,p d- ] is the set of all uncertain parameters, where ψ + and ψ - represents the wave vector in the positive and negative directions, p d+ and p d- Represents the predicted value The up and down fluctuation vectors; is the first uncertainty set; f(x,ξ) is the objective function; A1, A0, B1, b0, M and m are all constant parameters, where M and m are obtained based on the first uncertainty set; is the expected value of the corresponding vector.

6. A method for improving photovoltaic carrying capacity of a power distribution system as claimed in claim 5, characterized in that: The method adopts several HC improvement schemes to construct several robust HC models based on the initial robust HC model, including: Based on the adjustable robust ANM scheme, an adjustable variable is obtained according to the initial robust HC model; the adjustable variable is represented as an affine function of the uncertain parameters; The constraints of the worst uncertainty realization in the first uncertainty set are robustified and dualized using the maximum protection function to obtain linear inequality constraints. Based on the initial robust HC model, adjustable variables and linear inequality constraints, a first robust HC model corresponding to the adjustable robust ANM solution is obtained; the first robust HC model is specifically: In the above formula, g1 and g0 are linear functions; x u is an unadjustable variable in the decision variable; S represents the sample space; s represents the sth sample in the sample space; λ is the first uncertainty set is the vector of dual variables associated with the boundary constraints; α and β are the parameters in the affine function.

7. A device for improving photovoltaic carrying capacity of a power distribution system, characterized in that: include: A first building module, a second building module, a model solving module and a system adjustment module; The first model building module is used to obtain distribution system data, and based on the distribution system data, to build a deterministic optimization model with the goal of maximizing the total photovoltaic output; to build a first uncertainty set to describe the correlation between photovoltaic efficiency coefficients and the uncertainty of loads; An initial robust HC model is constructed based on the deterministic optimization model and the first uncertainty set; The second construction module is used to adopt several HC improvement schemes to construct several robust HC models based on the initial robust HC model respectively; The model solving module is used to solve several robust HC models to obtain several HC results; The system adjustment module is used to determine an adjustable robust ANM solution based on a number of HC results; and to adjust the distribution system based on the adjustable robust ANM solution to improve the photovoltaic carrying capacity of the distribution system.

8. A device for improving photovoltaic carrying capacity of a power distribution system as claimed in claim 7, characterized in that: The first construction module includes: a function construction unit and a first construction unit; The objective function construction unit is used to construct an objective function based on the power distribution system data and to determine a first set of constraint conditions with the goal of maximizing the total photovoltaic output; the objective function is specifically: In the above formula, i is the node number; is the set of nodes with power consumption or injection; t represents the time interval; is the set of t; is the predicted efficiency coefficient of photovoltaic power generation at node i at time t, which is defined as the ratio of actual photovoltaic power generation to installed capacity; G i represents the photovoltaic installed capacity corresponding to the i-node; Indicates the actual amount of power abandoned corresponding to node i; The first model building unit is used to build a deterministic optimization model based on the objective function and the first constraint condition set.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for improving the photovoltaic carrying capacity of the distribution system as described in any one of claims 1 to 6 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for improving the photovoltaic carrying capacity of a power distribution system as claimed in any one of claims 1 to 6 are implemented.

Images