CN113507128B - A method for optimal configuration of near-field reactive power in UHVDC converter stations - Google Patents

Abstract

The invention discloses a near-field reactive power optimal configuration method of an extra-high voltage direct current converter station, which comprises the following steps: 1) Determining a target ultra-high voltage direct current end power grid and network operation parameters thereof, and determining a node range to be connected into reactive compensation equipment; 2) Establishing a nonlinear optimization model by taking the alternating current bus voltage of the extra-high voltage converter station and the bus voltage of a key node of a certain power grid as optimization targets; 3) The voltage change of each node during reactive power change is evaluated, the sensitivity is calculated, and a reactive power configuration node primary selection set is formed according to the sensitivity; 4) And carrying out reactive power compensation calculation on the reactive power configuration node primary selection set to form a final node compensation configuration scheme.

Description

A method for optimal configuration of near-field reactive power in UHVDC converter stations

technical field

The invention belongs to the field of optimal configuration of energy resources, and in particular relates to a method for optimal configuration of near-field reactive power of an UHV DC converter station.

Background technique

With the continuous advancement of my country's energy structure transformation, the penetration rate of clean energy represented by wind power, photovoltaic and other power sources in the power grid has increased year by year. However, new energy power generation has inherent shortcomings such as random fluctuation of output, poor anti-interference ability, and weak support ability of large-scale new energy power plants to the power grid. Moreover, the capacity of conventional generating units in the western region is small, and the system’s frequency regulation and peak regulation capabilities are seriously insufficient. Under the condition of weak synchronous support, the sending-end power grid is prone to safety and stability problems such as low-voltage and high-voltage chain disconnection of new energy power stations, and the safe and stable operation of the system encounters huge challenges. Therefore, how to effectively and accurately make full use of the reactive power of the new energy unit itself, and how to coordinate it with the dynamic reactive power compensation equipment is an urgent problem to be solved. The key lies in how to optimize the configuration of reactive power.

For this reason, the present invention proposes a near-field reactive power optimal configuration method for UHV DC converter stations, which can speed up the solution efficiency of the objective function, improve the convergence of the equation, realize the reactive power stability of the UHV sending end grid, and provide guidance for the reactive power configuration of the clean energy UHV sending end.

Contents of the invention

For realizing the purpose of the present invention, adopt following technical scheme to realize:

A method for optimal configuration of near-field reactive power in an UHV DC converter station, comprising the following steps:

1) Determine the target UHVDC sending-end power grid and its network operating parameters, and determine the range of nodes to be connected to reactive power compensation equipment;

2) Establish a nonlinear optimization model with the AC bus voltage of the UHV converter station and the bus voltage of a key node of the power grid as the optimization objectives;

3) Evaluate the variation of the voltage of each node when the reactive power changes, calculate its sensitivity, and form a primary selection collection of reactive power configuration nodes according to the sensitivity;

4) Perform reactive power compensation calculation on the primary selection set of reactive power configuration nodes to form the final node compensation configuration scheme.

The near-field reactive power optimization configuration method of the UHV DC converter station, wherein: determining the range of nodes to be connected to reactive power compensation equipment in the step 1 includes selecting weak converter stations or weak power generation nodes as nodes to be connected to reactive power compensation equipment, and the above weak converter stations and weak power generation nodes are called key nodes.

In the near-field reactive power optimization configuration method of the UHV DC converter station, the nonlinear optimization model is shown in formula (1-b):

Among them, V _L is the AC bus voltage of the converter station; V _L0 is the initial value of the AC bus voltage of the converter station; V _Lmax is the maximum limit of the AC bus voltage of the converter station; V _Bi is the bus voltage _of a key node of the power grid; V _Bi0 is the initial value of the bus voltage _of a key node of the power grid;

In the method for optimal configuration of near-field reactive power of UHVDC converter station, the node sensitivity is calculated according to the following formula:

F(X,T,C)=0 (2-a)

Among them, F is the power balance equation of the power grid; X is the state vector of the power grid; T is the control variable of the power grid; C is a constant parameter of the power grid;

The method for optimal configuration of near-field reactive power of the UHV DC converter station, wherein step 4) includes:

4-1) Set artificial bee colony algorithm simulation parameters: the number of honey sources S _n , which is the number of key nodes determined in step 1); the number of algorithm iterations M _n , the solution dimension D, the initial number of iterations k = 1, and the discarding parameter limit = 20; the initial solution _Xi is randomly generated in the primary selection set of reactive power configuration, that is, a node is randomly selected, and one or more reactive power compensation devices are configured for it. The selection of this node and the configuration of reactive power compensation devices are called the initial solution _Xi

Calculate its fitness by formula (3-a):

Where G(X _i ) is the objective function corresponding to the solution to Xi _i , that is, G in formula 1-b;

4-2) Carry out M _n iterative simulations, and compare the fitness of the new solution and the old solution in the neighborhood nectar source each time, obtain the local optimum through the fitness comparison, and finally obtain the optimal nectar source position through the comparison between different nectar source positions;

4-3) Judging whether the set value of the maximum number of cycles is reached, if not reached, go to step 4-1); if reached, the algorithm ends, and the current record is the global optimal solution.

The method for optimal configuration of near-field reactive power of the UHV DC converter station, wherein step 4-2) includes:

4-2-1) At the beginning of the search process, a new solution is generated by formula (4-a), that is, a new food source

V _ij ＝X _ij +Φ _ij (X _ij -X _kj ) (4-a)

Among them, k=1,2,...,S _n , and k≠i, j=1,2,...,D, Φ _ij is a random number between [-1,1], x _ij and x _kj are the jth dimension position of solution i and solution k, then calculate the fitness fit _v of the new solution and evaluate it, if the fit _v of the new solution is better than the old solution, it will lead the bees to remember the new solution and forget the old solution, otherwise, keep the old solution;

5-2) After all the leading bees complete the search process, the leading bees dance in the recruiting area to share the solution information with the following bees, and the following bees calculate the selection probability of each solution according to the formula (4-b),

Then randomly generate a number in the interval [-1, 1]. If the probability value of the solution is smaller than the random number, keep the solution and make the discarding parameter limit=limit+1; if the probability value of the solution is greater than the random number, follow the bee to generate a new solution according to the formula (4-a), and check the fitness fit _v of the new solution. If it is better than before, the follower will remember the new solution and forget the old solution; otherwise, keep the old solution. , the role of the leading bee will be transformed into a scout bee, that is, the local optimal solution will be discarded, a new solution will be regenerated, and a new solution will be generated to enter the cycle again.

The method for optimal configuration of near-field reactive power of UHVDC converter station, wherein determining the range of nodes to be connected to reactive power compensation equipment in the step 1 includes: using the linearized probabilistic power flow model based on the analytical method shown in formula 1-a to perform power flow calculation on the target sending end power grid,

Among them, J ₀ and G ₀ are the coefficient matrices obtained by linearizing the reference point, which are the first-order partial derivatives of the injected power of each node and the power of each branch with respect to the voltage of each node; _X and _Z are the column vectors of the voltage of each node and the power of each branch; the subscript 0 indicates the state of the reference point;

Solve the formula 1-a to calculate the energy-saving voltage X. If the energy-saving voltage is lower than the predetermined value, it is judged as a weak converter station or a weak power generation energy-saving.

Description of drawings

Figure 1 is a flow chart of the near-field reactive power optimal configuration method for UHVDC converter stations;

Figure 2 shows the main topology of the power grid in the example;

Figure 3 is a dynamic characteristic diagram of the AC bus voltage of No. 1 converter station;

Figure 4 is a dynamic characteristic diagram of busbar voltage of No. 3 new energy station.

Detailed ways

Specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings.

As shown in Figure 1, the near-field reactive power optimal configuration method of UHVDC converter station includes the following steps:

1) Determine the target UHVDC sending-end power grid and its network operating parameters, such as the power of each power source and load point, state vector, etc., and perform initial power flow calculation on the target sending-end power grid. The specific steps of initial power flow calculation are as follows:

Using the linearized probabilistic power flow model based on the analytical method, the model is shown in formula (1-a), to calculate the power flow of the target sending end power grid, explore the weak points of reactive power in the system, and determine the range of nodes to be connected to reactive power compensation equipment, that is, to select weak converter stations or weak power generation nodes as nodes to be connected to reactive power compensation equipment. The above weak converter stations and weak power generation nodes are also called key nodes;

Among them, J ₀ and G ₀ are coefficient matrices obtained by linearization of the reference point, which are the first-order partial derivatives of the injected power of each node and the power of each branch with respect to the voltage of each node; T ₀ and S ₀ represent elements of the matrix; _X and _Z are column vectors of the voltage of each node and the power of each branch; the subscript 0 represents the state of the reference point;

Solving Formula 1-a, the voltage X of each node can be calculated. If the node voltage is lower than the predetermined value, it is judged as a weak converter station or a weak power generation node.

2) The nonlinear optimization model is established with the AC bus voltage of the UHV converter station and the bus voltage of a key node of the power grid as the optimization objectives, and the optimization objective function minG is established, as shown in formula (1-b)

Among them, V _L is the AC bus voltage of the converter station; V _L0 is the initial value of the AC bus voltage of the converter station; V _Lmax is the maximum limit of the _AC bus voltage of the converter station; V _Bi is the bus voltage _of a key node of the power grid; V _Bi0 is the initial value of the bus voltage of a key node of the power grid;

3) Considering the specific situation of each node to be connected to reactive power compensation equipment, evaluate the change of the voltage of each node when the reactive power changes, establish the sensitivity index as shown in the formulas (2-a) and (2-b), and form the primary selection set of reactive power configuration nodes. The principle of forming the primary selection set of reactive power configuration is: compare the node sensitivities calculated by each node.

The formula of step 3) is as follows:

F(X,T,C)=0(2-a)

Among them, F is the power balance equation of the power grid; X is the state vector of the power grid; T is the control variable of the power grid; C is a constant parameter of the power grid;

4) Set artificial bee colony algorithm simulation parameters: the number of honey sources S _n , which is the number of key nodes determined in step 1); the number of algorithm iterations M _n , the solution dimension D, the initial number of iterations k = 1, and the discarding parameter limit = 20; the initial solution X _i is randomly generated in the primary selection set of reactive power configuration, that is, a node is randomly selected, and one or more reactive power compensation devices are configured for it. The selection of this node and the configuration of reactive power compensation devices are called the initial solution _Xi .

Calculate its fitness by formula (3-a):

Among them, G(X _i ) is the objective function corresponding to the solution of Xi _i , that is, the optimization objective function in formula 1-b;

5) Carry out M _n times of iterative simulation, and compare the fitness of the new solution and the old solution in the neighborhood nectar source each time, obtain the local optimum through fitness comparison, and finally obtain the optimal nectar source location through the comparison between different nectar source locations.

5-1) At the beginning of the search process, a new solution is generated by formula (4-a), that is, a new food source

V _ij ＝X _ij +Φ _ij (X _ij -X _kj ) (4-a)

Among them, k=1,2,…,S _n and k≠i, j=1,2,…,D, Φ _ij is a random number between [-1,1], V _ij is a new solution, x _ij and x _kj are the jth dimension positions of solution i and solution k. Next, calculate the fitness fit _v of the new solution and evaluate it. If the fit _v of the new solution is better than the old solution, it will guide the bees to remember the new solution and forget the old solution, otherwise, keep the old solution;

5-2) After all the leading bees complete the search process, the leading bees will dance in the recruiting area to share the solution information with the following bees. Follow the bee to calculate the selection probability of each solution according to the formula (4-b),

Then randomly generate a number in the interval [-1,1], if the probability value of the solution is less than or equal to the random number, then keep the solution and make the discarding parameter limit=limit+1; if the probability value of the solution is greater than the random number, then follow the bee to generate a new solution according to the formula (4-a), and check the fitness fit _v of the new solution, if the fitness of the new solution is greater than the fitness of the old solution, the follower will remember the new solution and forget the old solution; otherwise, if the fitness is less than or equal to the fitness of the old solution, keep the old solution . If the number of times the old solution is retained is greater than the set abandonment parameter limit, the local optimal solution will be recorded, and at the same time, the role of the lead bee will be transformed into a scout bee (that is, the local optimal solution will be discarded and a new solution will be regenerated), and a new solution will be generated to enter the cycle again.

6) Judging whether the set value M _n of the maximum number of iterations is reached, if not, go to step 4); if it is reached, the algorithm is terminated, and the current record is the global optimal solution.

Enumerate the specific example of using the invention to create and solve practical technical problems below:

This example takes an actual UHV DC transmission system in a certain area as an example to illustrate, and its main topology is shown in Figure 2.

1) The main topology of the power grid in the calculation example is shown in Figure 2, the line type is 6×JL/G1A-500/45 and LGJ-400; determine the nodes that can be connected to reactive power compensation equipment (main converter stations and typical new energy stations); set artificial bee colony simulation parameters: total iterations M _n = 10000, discarding parameter limit = 20, solution dimension D = 2;

2) Evaluate the access point of the dynamic reactive power compensation device (converter). Considering the AC bus voltage of the UHV converter station and the bus voltage of a key node in the power grid, the sensitivity index is established, and the sensitivity evaluation is carried out on the AC bus of the important converter station in the power grid and the bus of some new energy stations. The results are shown in Table 1.

Table 1 Sensitivity of main access points

It can be seen that when performing reactive power optimization configuration, priority should be given to configuring reactive power compensation devices from access points with high sensitivity, so that the optimal reactive power optimization effect can be achieved more.

3) Considering sensitivity and economical considerations, the initial number of camera configurations is 4.

It can be concluded that the optimal reactive equipment configuration scheme is shown in Table 2.

Table 2 Optimal reactive equipment configuration scheme

Under the condition of this kind of reactive power compensation equipment configuration, the situation after the bipolar blocking fault occurs in the UHVDC sending-end power grid system is simulated. The voltage dynamic characteristic curves of the converter station bus and the typical new energy station bus are shown in Figures 3 and 4.

It can be seen that by optimizing the position of the access point of the condenser, the transient voltage rise of the UHVDC transmission grid can be significantly suppressed, and the results are in line with the actual operation conditions, which can effectively improve the stability of the grid operation.

Claims (2)

1. A near-field reactive power optimal configuration method for an extra-high voltage direct current converter station is characterized by comprising the following steps of:

1) Determining a target ultra-high voltage direct current end power grid and network operation parameters thereof, and determining a node range to be connected into reactive compensation equipment;

2) Establishing a nonlinear optimization model by taking the alternating current bus voltage of the extra-high voltage converter station and the bus voltage of a key node of a certain power grid as optimization targets;

3) The voltage change of each node during reactive power change is evaluated, the sensitivity is calculated, a reactive power configuration node primary selection set is formed according to the sensitivity, and the principle of forming the reactive power configuration primary selection set is as follows: comparing the node sensitivities calculated by the nodes, if the node sensitivities are larger than a preset value, classifying the node into a reactive configuration primary selection set, otherwise, not classifying the node into the reactive configuration primary selection set;

the formula of step 3) is as follows:

F(X,T,C)＝0 (2-a)

wherein F is a power balance equation of the power grid; x is a state vector of the power grid; t is a control variable of the power grid; c is a constant parameter of the power grid; v is the bus voltage to be solved for a certain sensitivity; q is reactive power injection of a certain node bus, S is node sensitivity;

4) Reactive power compensation calculation is carried out on the primary selection set of the reactive configuration nodes, and a final node compensation configuration scheme is formed;

the step 4) comprises the following steps:

4-1) setting simulation parameters of a manual bee colony algorithm: number of honey sources S _n I.e. the number of key nodes determined in step 1); algorithm iteration number M _n Solving dimension D, iterating the initial number k=1, discarding parameter limit=20; randomly generating an initial solution X in a reactive configuration initial selection set _i I.e. randomly selecting a node to which one or more reactive compensation devices are to be arranged, the node selection and the arrangement of the reactive compensation devices being referred to as the initial solution X _i ，

The fitness is calculated by the formula (3-a):

wherein G (X) _i ) To solve X _i A corresponding objective function;

4-2) M _n Performing iterative simulation, comparing the fitness of the new solution and the old solution in the neighborhood honey source each time, obtaining local optimum through the fitness comparison, and finally obtaining the optimum honey source position through the comparison between different honey source positions;

4-3) judging whether the set value of the maximum cycle number is reached or not, and turning to the step 4-1) is not reached; ending the algorithm when the current record is reached, wherein the current record is the global optimal solution;

the near-field reactive power optimal configuration method of the extra-high voltage direct current converter station comprises the following steps of:

4-2-1) at the beginning of the search process, a new solution, a new food source, is generated from equation (4-a)

V _ij ＝X _ij +Φ _ij (X _ij -X _kj ) (4-a)

Wherein k=1, 2, …, S _n And k+.i, j=1, 2, …, D, Φ _ij Is [ -1,1]Random number, X between _ij And X _kj For the j-th dimensional position of the solution i and the solution k, the fitness fit of the new solution is calculated next _v And evaluate it, if newly solved fit _v If the solution is superior to the old solution, leading the bee to memorize the new solution and forget the old solution, otherwise, reserving the old solution;

5-2) after all the lead bees complete the search process, the lead bees dance in the recruitment area to share the information of the solutions with the following bees, the following bees calculate the selection probability of each solution according to formula (4-b),

then in the interval [ -1,1]Randomly generating a number in the solution, if the probability value of the solution is smaller than the random number, reserving the solution and enabling a discard parameter limit=limit+1; if the probability value of the solution is greater than the random number, then following the bee, a new solution is generated from equation (4-a), and the fitness fit of the new solution is checked _v If the result is better than the previous result, the following bees will memorize the new solution and forget the old solution; otherwise, reserving the old solution, if the reservation times of the old solution are greater than the set discard parameter limit, recording the local optimal solution, and simultaneously converting the leading bee role into a reconnaissance bee, namely discarding the local optimal solution, regenerating a new solution, and generating the new solution to enter a cycle again;

the determining the node range to be accessed to the reactive compensation equipment in the step 1 comprises the following steps: carrying out load flow calculation on the target power grid at the transmitting end by using a linearization probability load flow model based on an analytic method as shown in a formula (1-a),

wherein J is ₀ 、G ₀ Respectively linearizing the reference points to obtain coefficient matrixes, and respectively solving first-order partial derivatives of the injection power of each node and the power of each branch to the voltage of each node; x, Z is the voltage of each node and the power column vector of each branch; subscript 0 indicates a reference point state; ΔS _S 、ΔS _D The random fluctuation amounts of the load power and the power injection power of each node relative to the reference point are respectively shown;

jie Gong (1-a), each of the power saving voltages X is calculated, and if the power saving voltage is lower than a predetermined value, it is judged that it is a weak converter station or weak power generation power saving.

2. The near-field reactive power optimal configuration method of the extra-high voltage direct current converter station according to claim 1, wherein the method comprises the following steps of: the determining the node range to be connected to the reactive power compensation device in the step 1 includes selecting a weak converter station or a weak power generation node as the node to be connected to the reactive power compensation device, where the weak converter station and the weak power generation node are called key nodes.