CN111860950A - Probability assessment method for power adjustability of virtual power plant - Google Patents

Abstract

The invention provides a probability evaluation method for the power adjustability of a virtual power plant, and belongs to the technical field of operation and control of power systems. The method comprises the following steps: decoupling linear approximation is carried out on the power flow model of the virtual power plant, and a power flow equation in a matrix form is constructed; carrying out linearization processing on active-reactive power ranges output by various types of power generation equipment and energy storage equipment; constructing a Gaussian mixture model to perform probability distribution function fitting on uncertainty factors such as photovoltaic factors, fans, loads and the like; calculating the power adjustable range of the virtual power plant at the grid-connected point under equal confidence by using an opportunity constraint method; and calculating a scalar field function of an adjustable range and confidence coefficient in an active-reactive two-dimensional plane to obtain the power adjustability of the virtual power plant under different confidence coefficients. The method fully considers the influence of uncertainty factors on the power adjustability of the virtual power plant, effectively improves the calculation speed, can obtain more accurate assessment results, and promotes the optimal scheduling of the power system.

Description

Probability assessment method for power adjustability of virtual power plant

Technical Field

The invention belongs to the technical field of operation control of power systems, and particularly relates to a probability evaluation method for power adjustability of a virtual power plant.

Background

The virtual power plant organically integrates the distributed generator set, the controllable load and the distributed energy storage facility, and realizes the integrated regulation and control of various distributed energy sources by utilizing an advanced regulation and control technology and a communication technology. In a region, a virtual power plant can participate in optimal scheduling of a power system as a special power plant. However, since the distributed power generation resources are large in number and small in scale, the distributed power generation resources are distributed in the power distribution network and have large characteristic differences. Therefore, there are difficulties with the system in directly controlling all distributed generation resources. The virtual power plant can integrate the distributed power supplies, but due to different equipment characteristics, network constraint conditions exist, and uncertain factors such as a large amount of loads, new energy output and the like exist, so that the power adjustability of the virtual power plant is difficult to accurately evaluate at present.

The existing method for evaluating the power adjustability of the virtual power plant adopts a deterministic evaluation method and does not consider the influence caused by the uncertainty of renewable energy sources and loads in the virtual power plant. In some methods, a plurality of scenes are generated through a Monte Carlo method, and then a power adjustability model of the virtual power plant under uncertain factors is constructed, but the method is huge in calculated amount and low in calculating speed, and effective evaluation results are difficult to obtain quickly.

Disclosure of Invention

The invention aims to overcome the defects in the prior art and provides a probability evaluation method for the power adjustability of a virtual power plant. The method fully considers the influence of uncertain factors on the power adjustability of the virtual power plant, effectively improves the calculation speed, can obtain more accurate assessment results, promotes the optimal scheduling of the power system, and has higher value in practical application.

The invention provides a probability evaluation method for the power adjustability of a virtual power plant, which is characterized by comprising the following steps of:

1) constructing a network model of a virtual power plant;

based on a decoupled linearized power flow model, defining each node inside a virtual power plant as a PQ node, defining a grid-connected node as a balance node, and establishing a network power flow model of the virtual power plant as follows:

V_L＝L^-1Q-L^-1G_LLB′_LL ^-1P (1)

θ_L＝H^-1P+H^-1G_LLB_LL ^-1Q (2)

wherein, V_LA vector representing the magnitude of the voltage at the PQ node; theta_LA vector representing the PQ nodal phase angle; g_LLA self-conductance matrix representing a PQ node; b'_LLRepresents the PQ node from the na matrix without considering the parallel capacitors; b is_LLRepresents the PQ node from the capacitance matrix under consideration of the parallel capacitor; p represents an injection active power vector formed by the balance node and the PQ type node; q represents the injected active power vector of the PQ-type node; l, L,

Is a system parameter;

the active and reactive power of the branch ij power flow in the virtual power plant are respectively expressed as follows:

P_ij＝g_ij(V_i-V_j)-b_ij(θ_i-θ_j) (3)

Q_ij＝-b_ij(V_i-V_j)-g_ij(θ_i-θ_j) (4)

wherein, P_ijAnd Q_ijRespectively representing active power and reactive power passing through the branch ij; v_iAnd V_jRespectively representing the voltage amplitudes of the node i and the node j; theta_iAnd theta_jRespectively representing voltage phase angles of a node i and a node j; g_ijAnd b_ijRespectively representing the conductance and susceptance of branch ij;

2) constructing an equipment model of a virtual power plant;

distributed power generation equipment of a virtual power plant is divided into five types, namely synchronous machine type equipment, photovoltaic power generation type equipment, doubly-fed fan type equipment, energy storage type equipment and load, and the operation constraint conditions of each type of equipment are respectively expressed as follows:

2.1) active power constraint of synchronous machine type equipment:

wherein the content of the first and second substances,

representing the active output power of the synchronous machine type equipment at a node i at the time t;

and

respectively representing the minimum value and the maximum value of the active output power of the synchronous machine type equipment at the node i;

reactive power constraint of synchronous machine type equipment:

wherein the content of the first and second substances,

representing the reactive output power of the synchronous machine type equipment at a node i at the time t;

and

respectively representing the minimum value and the maximum value of the reactive output power of the synchronous machine type equipment at the node i;

2.2) active power constraint of photovoltaic power generation type devices:

Wherein the content of the first and second substances,

representing the active output power of the photovoltaic power generation type equipment at a node i at the moment t;

representing the minimum value of the active output power of the photovoltaic power generation type equipment at the node i;

representing the maximum value of the active output power of the photovoltaic power generation type equipment at the node i;

capacity constraint of photovoltaic power generation type device:

wherein the content of the first and second substances,

representing the reactive output power of a photovoltaic-type installation at node i at time t；

Represents the maximum capacity of the photovoltaic power generation type device at node i;

the approximation of equation (8) is reduced to the linear constraint of the following octagon:

photovoltaic power generation type equipment prediction error constraint:

wherein the content of the first and second substances,

respectively representing an actual value and a predicted value of active power output of the photovoltaic power generation type equipment at a node i at the time t;

representing the light abandonment quantity of the photovoltaic power generation type equipment at a node i at the time t;

the prediction error of the output active power of the photovoltaic power generation type equipment at the node i at the time t is represented;

2.3) active power constraint of doubly-fed fan type equipment:

wherein the content of the first and second substances,

representing the active output power of the doubly-fed fan-type equipment at a node i at the moment t;

representing the minimum value of the active output power of the doubly-fed fan equipment at the node i;

representing the maximum value of the active output power of the doubly-fed fan type equipment at the node i;

Reactive power constraint of doubly-fed fan type equipment:

wherein the content of the first and second substances,

representing the reactive output power of the doubly-fed fan-type equipment at a node i at the time t;

and

respectively representing the minimum value and the maximum value of the reactive output power of the doubly-fed fan type equipment at the node i;

capacity constraint of doubly-fed fan-type equipment:

wherein the content of the first and second substances,

representing the maximum capacity of doubly-fed fan-type equipment at node i;

the approximation of equation (18) is reduced to the linear constraint of the following octagon:

the prediction error constraints of doubly-fed fan-type equipment are as follows:

wherein the content of the first and second substances,

respectively representing an actual value and a predicted value of active output power of the doubly-fed fan type equipment at a node i at the time t;

representing the air abandoning amount of the doubly-fed fan type equipment at a node i at the time t;

representing the prediction error of the active output power of the doubly-fed fan type equipment at the node i at the time t;

2.4) active power constraints of energy storage devices:

wherein the content of the first and second substances,

representing the net active output power of the energy storage type device at node i at time t;

representing the maximum value of the discharge active power of the energy storage type equipment at the node i;

representing the maximum value of the charging active power of the energy storage type equipment at the node i;

capacity constraints of energy storage devices:

wherein the content of the first and second substances,

representing the reactive output power of the energy storage type equipment at a node i at a time t;

Represents the maximum capacity of the energy storage type device at node i;

the approximation of equation (27) is reduced to the linear constraint of the following octagon:

2.5) load power constraints;

wherein the content of the first and second substances,

and

respectively representing the real load with work and the real load without work at a node i at the moment t;

representing an active load predicted value at a node i at a time t;

representing the active load prediction error at the node i at the time t;

power factor of the load at node i;

2.6) calculating net injection power of internal equipment of the virtual power plant at each node:

wherein the content of the first and second substances,

and

respectively representing the active net injection power and the reactive net injection power at a node i at the time t;

3) modeling uncertainty variables in the virtual power plant by using a Gaussian mixture model to obtain opportunity constraint conditions;

setting random variables

For the prediction error of the photovoltaic power generation device at the time t node i

Prediction error of doubly-fed fan type equipment

Active load prediction error at node i at time t

The composed vector is as follows:

fitting with N Gaussian distribution functions using a Gaussian mixture model

Distribution function of (d):

wherein the content of the first and second substances,

is shown in

Is the mean value of the jth Gaussian distribution at the time t, and

the distribution is a high-dimensional normal distribution of the variance, a subscript j represents a jth normal distribution function, and a superscript t represents a tth moment;

Representing the weight of the jth normal distribution function, and N representing the number of Gaussian distribution functions;

using the model formula Error created in step 1)! Reference source not found, -Error! Determining opportunity constraint conditions met by branch power flow and node voltage:

wherein the content of the first and second substances,

representing the voltage amplitude of the node i at the time t;

representing the active power flow of the branch ij at the moment t;

representing the reactive power flow of the branch ij at the moment t; 1-represents the confidence with which the constraint is satisfied;

4) the chance constraint formula Error!obtained in the step 3) is! Reference source not found, -Error! Reference source not found. is transformed into an equivalent deterministic constraint, the expression is as follows:

wherein the content of the first and second substances,

representing a node i voltage constraint margin;

representing the constraint margin of the active power of the branch ij tide;

representing branch ij reactive power constraint margin;

representing the sum of the active power and the reactive power of the branch ij as a constraint margin;

representing the difference constraint margin between the active power and the reactive power of the branch ij; lambda [ alpha ]^PRepresenting an output active power constraint margin of the renewable energy power generation equipment; each constraint margin calculation expression is as follows:

wherein the content of the first and second substances,

representing random variables

1-quantile of;

representing random variables

1-quantile of;

representing random variables

1-quantile of;

representing random variables

Quantile division;

representing random variables

1-quantile of;

representing random variables

1-quantile of (a) to (b),

is defined as follows:

5) establishing a probability evaluation model of the adjustable capacity of the power of the virtual power plant, and evaluating the adjustable capacity of the power of the virtual power plant under a set confidence coefficient; the method comprises the following specific steps:

5.1) arranging output active and reactive variables of various types of equipment of the virtual power plant into a vector form, and expressing the output active and reactive variables as follows:

wherein the content of the first and second substances,

respectively representing vectors formed by output active and reactive variables of synchronous machine type equipment at the moment t;

respectively representing vectors formed by output active and reactive variables of the photovoltaic power generation equipment at the moment t;

respectively representing vectors formed by output active and reactive variables of the doubly-fed fan type equipment at the moment t;

respectively representing the output active and non-active of the energy storage type equipment at the moment tA vector of work variables;

5.2) construction of decision variable vector X^tAs follows:

5.3) definition of

And

respectively establishing a probability evaluation model of the power adjustability of the virtual power plant for active power and reactive power output to a main network by the virtual power plant at a grid-connected point at the confidence 1-lower moment and the t moment;

the model objective function is as follows:

The constraint of the model is the formula Error! Reference source not found, -Error! ReferencesSource not found, and formula Error! Reference source not found;

wherein, alpha is a power factor alpha epsilon [0,2 pi);

5.4) solving the probability evaluation model of the power adjustability of the virtual power plant for any given power factor alpha to obtain the corresponding optimal solution under the set confidence 1 < - >, and recording the optimal solution as

5.5) expressing the solving result of the step 5.4) under different confidence coefficients into a piecewise linear function form by utilizing a convex function piecewise linear fitting algorithm, and forming a probability function analytic expression of the power adjustability of the virtual power plant as follows:

wherein, ReLU (. cndot.) is) Representing a rectified linear unit function; y is^tRepresenting two-dimensional independent variable vector formed by virtual power plant output active power and output reactive power at moment t

Representing a virtual plant power tunability; a is_jAnd b_jRespectively representing coefficients of jth piecewise function at time t; m represents the number of piecewise linear functions;

using equation (52), given a confidence level of 1-, let Conf_t(Y^t) Not less than 1-, solving to obtain Y of corresponding virtual power plant power adjustability^tRange, evaluation ends.

The invention has the characteristics and beneficial effects that:

the invention provides a probability evaluation method for the power adjustability of a virtual power plant, which is characterized in that a mixed Gaussian model is utilized to accurately fit the probability distribution of uncertainty factor prediction errors according to historical data; the problem is changed into a convex optimization model by utilizing linear approximation and converting the opportunity constraint condition into an equivalent deterministic constraint condition, so that the calculation difficulty is simplified, and the calculation speed is increased. By the method, a more accurate evaluation result of the power adjustability of the virtual power plant can be obtained, and optimal scheduling of the power system is promoted.

Drawings

FIG. 1 is an overall flow diagram of the method of the present invention.

FIG. 2 is a schematic diagram of a process for solving the power adjustable range under the confidence of a virtual power plant and the like.

Detailed Description

The invention provides a probability evaluation method for power adjustability of a virtual power plant, which is further described in detail in the following by combining the accompanying drawings and specific embodiments.

The invention provides a probability evaluation method for the power adjustability of a virtual power plant, the whole flow is shown in figure 1, and the method comprises the following steps:

1) constructing a network model of a virtual power plant;

based on a decoupled linearized power flow model, defining each node inside a virtual power plant as a PQ node, defining a grid-connected node as a balance node, and establishing a network power flow model of the virtual power plant as follows:

V_L＝L^-1Q-L^-1G_LLB′_LL ^-1P (1)

θ_L＝H^-1P+H^-1G_LLB_LL ^-1Q (2)

wherein, V_LA vector representing the magnitude of the voltage at the PQ node; theta_LA vector representing the PQ nodal phase angle; g_LLA self-conductance matrix representing a PQ node; b'_LLRepresents the self-capacitance matrix at the PQ node without considering the parallel capacitor; b is_LLRepresents the PQ node from the capacitance matrix under consideration of the parallel capacitor; p represents an injection active power vector formed by the balance node and the PQ type node; q represents the injected active power vector of the PQ-type node; l, L,

All the parameters are system parameters, and the values of the parameters can be solved by utilizing a decoupled linearized power flow model.

The active and reactive power of the branch ij power flow in the virtual power plant are respectively expressed as follows:

P_ij＝g_ij(V_i-V_j)-b_ij(θ_i-θ_j) (3)

Q_ij＝-b_ij(V_i-V_j)-g_ij(θ_i-θ_j) (4)

wherein, P_ijAnd Q_ijRespectively representing active power and reactive power passing through the branch ij; v_iAnd V_jRespectively representing the voltage amplitudes of the node i and the node j; theta_iAnd theta_jRespectively representing voltage phase angles of a node i and a node j; g_ijAnd b_ijRespectively, the conductance and susceptance of branch ij.

2) Constructing an equipment model of a virtual power plant;

distributed power generation equipment existing in a virtual power plant can be divided into five types, namely synchronous machine type equipment, photovoltaic power generation type equipment, doubly-fed fan type equipment, energy storage type equipment and load, and the operation constraint conditions of each type of equipment can be respectively expressed as follows:

2.1) active power constraint of synchronous machine type equipment:

wherein the content of the first and second substances,

representing the active output power of the synchronous machine type equipment at a node i at the time t;

and

and respectively representing the minimum value and the maximum value of the active output power of the synchronous machine type equipment at the node i.

Reactive power constraint of synchronous machine type equipment:

wherein the content of the first and second substances,

representing the reactive output power of the synchronous machine type equipment at a node i at the time t;

and

respectively representing the minimum and maximum reactive output power values of the synchronous machine type equipment at node i.

2.2) active power constraint of photovoltaic power generation type devices:

wherein the content of the first and second substances,

representing the active output power of the photovoltaic power generation type equipment at a node i at the moment t;

representing the minimum value of the active output power of the photovoltaic power generation type equipment at the node i;

and (4) representing the maximum value of the active output power of the photovoltaic power generation type equipment at the node i.

Capacity constraint of photovoltaic power generation type device:

wherein the content of the first and second substances,

representing the reactive output power of the photovoltaic power generation type equipment at the node i at the time t;

representing the maximum capacity of the photovoltaic power generation type device at node i. To simplify the complexity of the optimization problem, the capacity constraint of the photovoltaic power generation type device of equation (8) is generally reduced approximately to the following octagonal linear constraint:

the photovoltaic power generation type equipment prediction error constraint representing the relation among the actual value, the predicted value and the prediction error of the photovoltaic power generation type equipment output active power is as follows:

wherein the content of the first and second substances,

respectively representing an actual value and a predicted value of active power output of the photovoltaic power generation type equipment at a node i at the time t;

representing the light abandonment quantity of the photovoltaic power generation type equipment at a node i at the time t;

and (4) representing the prediction error of the output active power of the photovoltaic power generation type equipment at the node i at the time t.

2.3) active power constraint of doubly-fed fan type equipment:

Wherein the content of the first and second substances,

representing the active output power of the doubly-fed fan-type equipment at a node i at the moment t;

representing the minimum value of the active output power of the doubly-fed fan equipment at the node i;

and the maximum value of the active output power of the doubly-fed fan type equipment at the node i is represented.

Reactive power constraint of doubly-fed fan type equipment:

wherein the content of the first and second substances,

representing the reactive output power of the doubly-fed fan-type equipment at a node i at the time t;

and

and respectively representing the minimum value and the maximum value of the reactive output power of the doubly-fed fan type equipment at the node i.

Capacity constraint of doubly-fed fan-type equipment:

wherein the content of the first and second substances,

representing the maximum capacity of the doubly-fed fan-type equipment at node i.

The capacity constraint of the doubly-fed fan-type plant of equation (18) is reduced approximately to the linear constraint of the following octagon:

the prediction error constraint of the doubly-fed fan type equipment representing the relationship among the actual value, the predicted value and the prediction error of the doubly-fed fan type equipment output active power is as follows:

wherein the content of the first and second substances,

respectively representing an actual value and a predicted value of active output power of the doubly-fed fan type equipment at a node i at the time t;

representing the air abandoning amount of the doubly-fed fan type equipment at a node i at the time t;

and (4) representing the prediction error of the active output power of the doubly-fed fan type equipment at the node i at the time t.

2.4) the energy storage type renewable energy device has two states of charging and discharging, and the active power of the energy storage type device is restricted as follows:

wherein the content of the first and second substances,

representing the net active output power of the energy storage type device at node i at time t;

representing the maximum value of the discharge active power of the energy storage type equipment at the node i;

representing the maximum value of the charging active power of the energy storage type equipment at the node i.

Capacity constraints of energy storage devices:

wherein the content of the first and second substances,

representing the reactive output power of the energy storage type equipment at a node i at a time t;

representing the maximum capacity of the energy storage type device at node i.

The capacity constraint problem of the energy storage device of equation (27) is reduced approximately to the following linear constraint condition of octagon:

2.5) load power constraints;

wherein the content of the first and second substances,

and

respectively representing the real load with work and the real load without work at a node i at the moment t;

representing an active load predicted value at a node i at a time t;

representing the active load prediction error at the node i at the time t;

power factor of the load at node i.

2.6) according to the positions of the various types of equipment in the nodes, the net injection power of the internal equipment of the virtual power plant at each node can be obtained:

wherein the content of the first and second substances,

and

respectively representing the active and reactive net injected power at node i at time t.

3) Modeling uncertainty variables in the virtual power plant by using a Gaussian mixture model to obtain opportunity constraint conditions;

setting random variables

Predicting error for photovoltaic power generation type equipment at time t node i

Prediction error of doubly-fed fan type equipment

Active load prediction error at node i at time t

The composed vector, namely:

fitting with N Gaussian distribution functions using a Gaussian mixture model

I.e.:

wherein the content of the first and second substances,

is shown in

Is the mean value of the jth Gaussian distribution at the time t, and

the distribution is a high-dimensional normal distribution of the variance, a subscript j represents a jth normal distribution function, and a superscript t represents a tth moment;

and the weight of the jth normal distribution function is expressed, N represents the number of Gaussian distribution functions, the specific value of the number is determined according to the precision requirement of the model, and the higher the precision requirement is, the larger the value of N is obtained.

By using the linearized power flow model expressions (1) - (4) established in the step 1), the voltage amplitude of the node i and the active reactive power of the branch ij can be expressed as a linear combination of the node injection active power and the node injection reactive power. Because uncertain factors such as renewable energy power generation equipment, load and the like exist in the virtual power plant, the predicted maximum output point of the renewable energy and the node load value are in a probability distribution form, so that the constraint condition met by the branch flow and the node voltage is an opportunity constraint, namely the constraint condition is met under certain confidence coefficient:

Wherein the content of the first and second substances,

representing the voltage amplitude of the node i at the time t;

active power of branch ij at tTidal current;

representing the reactive power flow of the branch ij at the moment t; 1-represents the confidence with which the constraint is satisfied.

4) Converting the opportunity constraint conditions obtained in the step 3) into equivalent certainty constraint conditions;

for the opportunity constraint problem, the uncertainty variable in the opportunity constraint problem can be changed into a constraint margin, and then the opportunity constraint conditions (38) - (40) obtained in the step 3) are converted into a certainty constraint condition.

Namely:

wherein the content of the first and second substances,

representing a node i voltage constraint margin;

representing the constraint margin of the active power of the branch ij tide;

representing branch ij reactive power constraint margin;

representing the sum of the active power and the reactive power of the branch ij as a constraint margin;

representing the difference constraint margin between the active power and the reactive power of the branch ij; lambda [ alpha ]^PAnd representing the output active constraint margin of the renewable energy power generation device. The constraint margin may be represented by the following equation:

wherein the content of the first and second substances,

representing random variables

1-quantile of;

representing random variables

1-quantile of;

representing random variables

1-quantile of;

representing random variables

Quantile division;

representing random variables

1-quantile of;

representing random variables

1-quantile of (a) to (b),

is defined as follows:

5) Establishing a probability evaluation model of the adjustable capacity of the power of the virtual power plant, and evaluating the adjustable capacity of the power of the virtual power plant under certain confidence;

the output active and reactive variables of various types of equipment in the virtual power plant are arranged into a vector form and are expressed as follows:

wherein the content of the first and second substances,

respectively representing vectors formed by output active and reactive variables of synchronous machine type equipment at the moment t;

respectively representing vectors formed by output active and reactive variables of the photovoltaic power generation type equipment at the moment t;

respectively representing vectors formed by output active and reactive variables of the doubly-fed fan type equipment at the moment t;

and respectively representing vectors formed by the active variables and the reactive variables of the output of the energy storage type equipment at the moment t.

Further arranging vectors formed by active and reactive variables of various types of equipment into decision variable vector X^tAs follows:

definition of

And

respectively is the active power and the reactive power output to the main network by the virtual power plant at the confidence level of 1-and the grid-connected point at the time t. And (3) establishing a probability evaluation model of the power adjustability of the virtual power plant, and solving the adjustable power range of the virtual power plant at the time t under the level of the confidence coefficient 1-, wherein the equations (5) - (35) and (41) are integrally used as constraint conditions. The objective function of the probability assessment model of the power tunability of the virtual power plant is to solve the power output limit reached by the virtual power plant at each given constant power factor α, i.e. to solve the following problem in the range of α ∈ [0,2 π):

For any given power factor alpha, solving the optimization problem equation (51) to obtain the corresponding optimal solution with the set confidence 1 —, which is recorded as

And solving results under different alpha to obtain scattered points formed by multiple groups of optimal solutions. Schematic of the solution processFIG. 2 is a graph with horizontal and vertical coordinates representing the real and reactive power output by the grid-connected point of the virtual power plant, respectively; the arrow indicates the direction in which the optimal solution is sought in the optimization problem; the scatter plot represents a solution obtained by solving the optimization problem through a variation parameter alpha; the outer boundary range represented by a plurality of discrete points represents the solved virtual plant output power regulation range.

5.5) constructing an analytic expression form for expressing the adjustable power capacity of the virtual power plant;

in step 5.4) the power tunability range of the virtual power plant at constant power factor at the level of confidence 1-at time t has been solved. In a three-dimensional space formed by the output active power, the output reactive power and the confidence level of the virtual power plant, for each confidence 1-lower step 5.4), obtaining a solution in the model

A contour on the curved surface is formed. And (4) forming a three-dimensional curved surface representing the output active power and output reactive power range and the confidence level based on the result obtained by solving the confidence 1. And expressing the solving result in a piecewise linear function form by using a convex function piecewise linear fitting algorithm to form a probability function analytic expression of the power adjustability of the virtual power plant as follows:

Wherein ReLU (·) represents a rectifying linear unit function; y is^tRepresenting two-dimensional independent variable vector formed by virtual power plant output active power and output reactive power at moment t

Representing a virtual plant power tunability; a is_jAnd b_jRespectively representing coefficients of jth piecewise function at time t; m represents the number of piecewise linear functions. Using equation (52), a confidence level of 1-is given, let Conf_t(Y^t) Not less than 1-, the corresponding virtual power plant power adjustable energy can be solvedY of force^tRange, evaluation ends.

In addition, the model established in the step (2) of the invention can be used for calculating the cost function of the virtual power plant, and the specific steps are as follows:

6) constructing a cost model of the virtual power plant, and piecewise linearly fitting a cost function of the virtual power plant;

aiming at various devices in the virtual power plant, constructing corresponding cost models:

6.1) the cost of electricity generation of a synchronous type of equipment is in the form of a quadratic function, namely:

wherein the function

The output active power of the synchronous machine type equipment at the node i at the time t is represented as

The cost of electricity generation; a is_i,GEN，b_i,GENAnd c_i,GENRespectively a secondary cost parameter, a primary cost parameter and a cost constant parameter of the synchronous machine type equipment at the node i.

6.2) photovoltaic power generation type equipment mainly considers the cost of abandoning light, namely:

Wherein the function

Representing the photovoltaic power generation type equipment at node i outputs active power at time t of

The cost of electricity generation;

and represents the power grid price at the moment t.

6.3) doubly-fed fan-type equipment mainly considers the cost of abandoning the wind, promptly:

wherein the function

The doubly-fed fan type equipment at the node i outputs active power at the moment t

The cost of electricity generation.

6.4) cost of energy storage type equipment considers charge and discharge cost, namely:

wherein the function

Indicating that the energy storage type equipment at the node i outputs active power at the moment t

The cost of electricity generation;

and

respectively representing the cost coefficients of discharging and charging of the energy storage type equipment at the node i; Δ t represents the time interval between two adjacent decision time points.

6.5) constructing a virtual power plant cost optimization model, and solving sample points of a virtual power plant cost fitting function

According to the cost functions of the various types of equipment, a virtual power plant cost optimization model is constructed, constraint conditions of the virtual power plant cost optimization model are represented by the constraint conditions of (5) - (35) and (41), and the total running cost of the internal equipment of the virtual power plant is represented by an objective function, namely:

wherein the content of the first and second substances,

representing the total cost of all equipment operations of the virtual power plant at time t.

Based on the result of step 5), the range of the power output of the virtual power plant to the outside at the time t can be determined

In the interval, K samples are randomly extracted, the specific value of the K samples is determined according to the precision requirement of the model, and the higher the precision requirement is, the larger the value of K is obtained. Let the kth sample be

Make the output power of the virtual power plant at the moment

To minimize virtual plant operating costs

For the purpose, an optimization problem is solved to obtain a minimum operating cost of

Therefore, the output power-power generation cost point corresponding to the kth sample can be obtained

And repeating the sampling and optimizing processes to obtain K output power-power generation cost points in total to form a fitting data set of a virtual power plant power generation cost curve.

6.6) piecewise linear fitting of the cost function of the virtual power plant;

and (3) fitting an output power-power generation cost curve formed by K sample points in r sections (the r value is usually 3-7 according to different fitting precision requirements) by using a convex function piecewise linear fitting algorithm. The form is as follows:

wherein, c_l，d_lThe coefficients of the first order term and the constant coefficient of the ith affine function. By using the formula (58), any active power value of the virtual power plant at the grid-connected point is input, and the corresponding operation cost can be obtained.

Claims (1)

1. A probability assessment method for power adjustability of a virtual power plant is characterized by comprising the following steps:

1) Constructing a network model of a virtual power plant;

based on a decoupled linearized power flow model, defining each node inside a virtual power plant as a PQ node, defining a grid-connected node as a balance node, and establishing a network power flow model of the virtual power plant as follows:

θ_L＝H^-1P+H^-1G_LLB_LL ^-1Q (2)

wherein, V_LA vector representing the magnitude of the voltage at the PQ node; theta_LA vector representing the PQ nodal phase angle; g_LLA self-conductance matrix representing a PQ node; b'_LLRepresents the PQ node from the na matrix without considering the parallel capacitors; b is_LLRepresents the PQ node from the capacitance matrix under consideration of the parallel capacitor; p represents an injection active power vector formed by the balance node and the PQ type node; q represents the injected active power vector of the PQ-type node; l, L,

Is a system parameter;

the active and reactive power of the branch ij power flow in the virtual power plant are respectively expressed as follows:

P_ij＝g_ij(V_i-V_j)-b_ij(θ_i-θ_j) (3)

Q_ij＝-b_ij(V_i-V_j)-g_ij(θ_i-θ_j) (4)

wherein, P_ijAnd Q_ijRespectively representing active power and reactive power passing through the branch ij; v_iAnd V_jRespectively representing the voltage amplitudes of the node i and the node j; theta_iAnd theta_jRespectively representing voltage phase angles of a node i and a node j; g_ijAnd b_ijRespectively representing the conductance and susceptance of branch ij;

2) constructing an equipment model of a virtual power plant;

distributed power generation equipment of a virtual power plant is divided into five types, namely synchronous machine type equipment, photovoltaic power generation type equipment, doubly-fed fan type equipment, energy storage type equipment and load, and the operation constraint conditions of each type of equipment are respectively expressed as follows:

2.1) active power constraint of synchronous machine type equipment:

wherein the content of the first and second substances,

representing the active output power of the synchronous machine type equipment at a node i at the time t;

and

respectively representing the minimum value and the maximum value of the active output power of the synchronous machine type equipment at the node i;

reactive power constraint of synchronous machine type equipment:

wherein the content of the first and second substances,

representing the reactive output power of the synchronous machine type equipment at a node i at the time t;

and

respectively representing the minimum value and the maximum value of the reactive output power of the synchronous machine type equipment at the node i;

2.2) active power constraint of photovoltaic power generation type devices:

wherein the content of the first and second substances,

representing the active output power of the photovoltaic power generation type equipment at a node i at the moment t;

representing the minimum value of the active output power of the photovoltaic power generation type equipment at the node i;

representing the maximum value of the active output power of the photovoltaic power generation type equipment at the node i;

capacity constraint of photovoltaic power generation type device:

wherein the content of the first and second substances,

representing the reactive output power of the photovoltaic power generation type equipment at the node i at the time t;

represents the maximum capacity of the photovoltaic power generation type device at node i;

the approximation of equation (8) is reduced to the linear constraint of the following octagon:

photovoltaic power generation type equipment prediction error constraint:

wherein the content of the first and second substances,

respectively representing an actual value and a predicted value of active power output of the photovoltaic power generation type equipment at a node i at the time t;

Representing the light abandonment quantity of the photovoltaic power generation type equipment at a node i at the time t;

the prediction error of the output active power of the photovoltaic power generation type equipment at the node i at the time t is represented;

2.3) active power constraint of doubly-fed fan type equipment:

wherein the content of the first and second substances,

representing the active output power of the doubly-fed fan-type equipment at a node i at the moment t;

representing the minimum value of the active output power of the doubly-fed fan equipment at the node i;

representing the maximum value of the active output power of the doubly-fed fan type equipment at the node i;

reactive power constraint of doubly-fed fan type equipment:

wherein the content of the first and second substances,

representing the reactive output power of the doubly-fed fan-type equipment at a node i at the time t;

and

respectively representing the minimum value and the maximum value of the reactive output power of the doubly-fed fan type equipment at the node i;

capacity constraint of doubly-fed fan-type equipment:

wherein the content of the first and second substances,

representing the maximum capacity of doubly-fed fan-type equipment at node i;

the approximation of equation (18) is reduced to the linear constraint of the following octagon:

the prediction error constraints of doubly-fed fan-type equipment are as follows:

wherein the content of the first and second substances,

respectively representing an actual value and a predicted value of active output power of the doubly-fed fan type equipment at a node i at the time t;

representing the air abandoning amount of the doubly-fed fan type equipment at a node i at the time t;

Representing the prediction error of the active output power of the doubly-fed fan type equipment at the node i at the time t;

2.4) active power constraints of energy storage devices:

wherein the content of the first and second substances,

representing the net active output power of the energy storage type device at node i at time t;

representing the maximum value of the discharge active power of the energy storage type equipment at the node i;

representing the maximum value of the charging active power of the energy storage type equipment at the node i;

capacity constraints of energy storage devices:

wherein the content of the first and second substances,

representing the reactive output power of the energy storage type equipment at a node i at a time t;

represents the maximum capacity of the energy storage type device at node i;

the approximation of equation (27) is reduced to the linear constraint of the following octagon:

2.5) load power constraints;

wherein the content of the first and second substances,

and

respectively representing the real load with work and the real load without work at a node i at the moment t;

representing an active load predicted value at a node i at a time t;

representing the active load prediction error at the node i at the time t;

power factor of the load at node i;

2.6) calculating net injection power of internal equipment of the virtual power plant at each node:

wherein the content of the first and second substances,

and

respectively representing the active net injection power and the reactive net injection power at a node i at the time t;

3) modeling uncertainty variables in the virtual power plant by using a Gaussian mixture model to obtain opportunity constraint conditions;

Setting random variables

For the prediction error of the photovoltaic power generation device at the time t node i

Prediction error of doubly-fed fan type equipment

Active load prediction error at node i at time t

The composed vector is as follows:

fitting with N Gaussian distribution functions using a Gaussian mixture model

Distribution function of (d):

wherein the content of the first and second substances,

is shown in

Is the mean value of the jth Gaussian distribution at the time t, and

the distribution is a high-dimensional normal distribution of the variance, a subscript j represents a jth normal distribution function, and a superscript t represents a tth moment;

representing the weight of the jth normal distribution function, and N representing the number of Gaussian distribution functions;

determining opportunity constraint conditions met by branch power flow and node voltage by using model expressions (1) - (4) established in the step 1):

wherein the content of the first and second substances,

representing the voltage amplitude of the node i at the time t;

representing the active power flow of the branch ij at the moment t;

representing the reactive power flow of the branch ij at the moment t; 1-represents the confidence with which the constraint is satisfied;

4) converting the chance constraint conditions (38) - (40) obtained in the step 3) into equivalent deterministic constraint conditions, wherein the expression is as follows:

wherein the content of the first and second substances,

representing a node i voltage constraint margin;

representing the constraint margin of the active power of the branch ij tide;

representing branch ij reactive power constraint margin;

representing the sum of the active power and the reactive power of the branch ij as a constraint margin;

Representing the difference constraint margin between the active power and the reactive power of the branch ij; lambda [ alpha ]^PRepresenting an output active power constraint margin of the renewable energy power generation equipment; each constraint margin calculation expression is as follows:

wherein the content of the first and second substances,

representing random variables

1-quantile of;

representing random variables

1-quantile of;

representing random variables

1-quantile of;

representing random variables

1-quantile of;

representing random variables

1-quantile of;

representing random variables

1-quantile of (a) to (b),

is defined as follows:

5) establishing a probability evaluation model of the adjustable capacity of the power of the virtual power plant, and evaluating the adjustable capacity of the power of the virtual power plant under a set confidence coefficient; the method comprises the following specific steps:

5.1) arranging output active and reactive variables of various types of equipment of the virtual power plant into a vector form, and expressing the output active and reactive variables as follows:

wherein the content of the first and second substances,

respectively representing vectors formed by output active and reactive variables of synchronous machine type equipment at the moment t;

respectively representing vectors formed by output active and reactive variables of the photovoltaic power generation equipment at the moment t;

respectively representing vectors formed by output active and reactive variables of the doubly-fed fan type equipment at the moment t;

respectively representing vectors formed by output active and reactive variables of the energy storage type equipment at the moment t;

5.2) construction of decision variable vector X^tAs follows:

5.3) definition of

And

respectively output to the main network by the virtual power plant at the grid-connected point at the confidence 1-lower moment and the time tEstablishing a probability evaluation model of the power adjustability of the virtual power plant for active power and reactive power;

the model objective function is as follows:

the constraint conditions of the model are equations (5) - (35) and (41);

wherein, alpha is a power factor alpha epsilon [0,2 pi);

5.4) solving the probability evaluation model of the power adjustability of the virtual power plant for any given power factor alpha to obtain the corresponding optimal solution under the set confidence 1 < - >, and recording the optimal solution as

5.5) expressing the solving result of the step 5.4) under different confidence coefficients into a piecewise linear function form by utilizing a convex function piecewise linear fitting algorithm, and forming a probability function analytic expression of the power adjustability of the virtual power plant as follows:

wherein ReLU (·) represents a rectifying linear unit function; y is^tRepresenting two-dimensional independent variable vector formed by virtual power plant output active power and output reactive power at moment t

Representing a virtual plant power tunability; a is_jAnd b_jRespectively representing coefficients of jth piecewise function at time t; m represents the number of piecewise linear functions;

using equation (52), given a confidence level of 1-, let Conf _t(Y^t) Not less than 1-, solving to obtain Y of corresponding virtual power plant power adjustability^tRange, evaluation ends.

Images