CN109524988B - Wind power collection station voltage control method based on total active power trend prediction - Google Patents

Abstract

The invention relates to a voltage control method for a wind power gathering station based on total active power output trend prediction, in particular to a 750kV/500kV wind power gathering station voltage control method based on total active power output trend prediction, and belongs to the technical field of power system control. The method includes: reading the current time and wind power forecast data for one hour in the future in each calculation cycle of the lower-level 330/220kV power grid system, and calculating the active power sensitivity of the unit to the designated line through the quasi-steady-state sensitivity, and obtaining the future of the designated line. The predicted value of active power for one hour, and then the active power predicted value of the line is added to get the total wind power active power trend of the entire 750/500kV aggregated substation in the future one hour, based on the aggregated substation in the future The total wind power transmission trend changes automatically to achieve voltage priority control , judging the priority switching of capacitors in the 750/500kV pooling substation to meet the voltage control requirements of the wind power pooling area.

Description

A voltage control method for wind power gathering stations based on trend prediction of total active power

technical field

The invention relates to a voltage control method for a wind power collecting station based on total active power trend prediction, in particular to a 750kV/500kV wind power collecting station voltage control method based on total active power output trend prediction, and belongs to the technical field of power system control.

Background technique

Automatic Voltage Control (AVC, Automatic Voltage Control) system is an important means to achieve safe (improving voltage stability margin), economical (reducing network loss), and high-quality (improving voltage qualification rate) operation of the power grid. The AVC system is based on the power grid energy management system (EMS), and can use the real-time operation data of the power grid to scientifically decide the best reactive power and voltage adjustment plan from the perspective of global optimization of the power grid, and automatically send it to power plants, substations and lower-level power grid dispatching agency implementation. AVC system is constrained by voltage safety and high quality, aiming at system operation economy, continuously and closed-loop real-time optimal control of voltage, and realizes a complete set of online generation, real-time distribution, and closed-loop automatic control of reactive power and voltage coordinated control scheme. Analysis, decision-making, control, and re-analysis, re-decision and re-control of reactive power and voltage real-time tracking control problems can effectively overcome the shortcomings of traditional grid reactive power and voltage control methods, and improve the level of grid security, stability and economic operation.

With the increasing pressure of energy shortage, the need for sustainable development, the enhancement of environmental protection awareness and the proposal of relevant policies, the development and utilization of new energy has become an inevitable trend. Wind energy is a clean and pollution-free renewable energy. With the implementation of the Renewable Energy Law, my country's wind energy utilization has entered a stage of rapid development. As of 2015, the newly installed capacity of wind power in all regions of my country has maintained a growth trend, and the northwest region is still the region with the most newly installed capacity, exceeding 11GW, accounting for 38% of the total installed capacity. The wind power gathering area in the northwest region mainly uses the 750/500kV substation as the sending channel, and the lower level is the 330/220kV gathering substation, which serves as the energy gathering point for the 330/110/35kV hydro-thermal power, wind farm and photovoltaic power station. The typical power grid structure of the 750/500kV wind power gathering area is shown in Figure 1 in the annex.

In 2011, many large-scale wind turbine chain disconnection accidents occurred in large-scale wind power grid-connected areas in Northwest my country and North China. Investigations show that such accidents have a similar development process. First, the instantaneous voltage drop caused by a sudden short-circuit fault somewhere, which in turn caused the wind turbines in a specific wind farm with close electrical connection to be partially or completely disconnected from the grid in a short period of time. Afterwards, the transmission power of the feeder line in the wind farm and the transmission line between the wind farms is reduced, and the reactive power generated by the line charging capacitor and the parallel capacitor input in the wind farm booster station and the regional gathering station is relatively excess reactive power absorbed by the line. Due to the small short-circuit capacity in this area, the excess reactive power causes the system voltage to rise sharply, and eventually causes the overvoltage protection of the fans in other adjacent wind farms. This process further increased the excess reactive power of the system, and continued to expand the area of the area where the wind turbine off-grid accident occurred.

Xu Fengda, Guo Qinglai, Sun Hongbin, etc., in "Analysis and Simulation of Multi-wind Farm Chain Off-Grid Process [J]" (Power Grid Technology, 2014, 38(6): 1425-1431), on the multi-wind farm chain off-grid caused by the above voltage Detailed analysis, simulation and settlement of an actual case of the wind farm were carried out, and the conclusion was drawn: Before the accident, most of the wind turbines in the wind farm were close to full power. more capacitors. When an accident occurs in the collection area and the fans in a wind farm are disconnected from the grid, the long line connecting the wind farm and the power grid changes from heavy load to light load. At this time, the charging reactive power is large, and the line consumption reactive power is sharply reduced. The capacitor cannot be automatically cut off, which aggravates the excess reactive power of the system, which further raises the regional voltage and causes other wind farms to be disconnected from the grid. It can be seen from the analysis that only relying on the capacitors and reactors of the collecting station to maintain the voltage level of the wind power gathering area. , reducing the service life of the equipment; on the other hand, it may also cause multiple wind farms in the collection area to be chained off the grid in the event of disturbances such as accidents. However, only relying on the reactive power regulation capacity of the wind farm in the gathering area to adjust the voltage in the gathering area, its total reactive power regulation capacity may not be enough to meet the voltage safety requirements in the larger and smaller power generation modes. Therefore, how to reasonably control the capacitors and reactors in the collection station is a difficult problem in the automatic control of AVC.

The quasi-steady-state sensitivity of the power grid operation, its physical meaning is the change of the active power of each branch in the power grid after the unit active power is added at a certain node. Sun Hongbin, Zhang Boming, Xiang Niande proposed the quasi-steady-state sensitivity method in "Quasi-steady-state sensitivity analysis method" (Chinese Journal of Electrical Engineering, April 1999 V19N4, pp.9-13), which is different from the conventional static sensitivity method. The analysis method is different. The quasi-steady-state sensitivity method considers the physical response of the quasi-steady state of the power system, and takes into account the total change between the old and the new steady state before and after system control, which effectively improves the accuracy of sensitivity analysis. The method is based on the PQ decoupling model of the power system. When the generator is equipped with an automatic voltage regulator (AVR), the generator node can be considered as a PV node; and when the generator is equipped with automatic reactive power regulation (AQR) or During automatic power factor adjustment (APFR), it can be considered that the generator node is the same as the common load node and is a PQ node. In addition, the load voltage static characteristic is considered as a linear or quadratic curve of the node voltage. The established power flow model naturally takes these quasi-steady-state physical responses into account, so the sensitivity calculated based on the power flow model is the quasi-steady-state sensitivity.

SUMMARY OF THE INVENTION

The purpose of the present invention is to propose a voltage control method for wind power gathering stations based on total active power trend prediction, especially a 750kV/500kV wind power gathering station voltage control method based on total active power trend prediction, using the existing quasi-steady-state sensitivity Method, through wind power prediction and calculation of quasi-steady-state sensitivity, the power prediction of the wind farm is converted to the line connected to the receiving end grid, and the total wind power active power trend of the entire 750/500kV pooling substation (one hour in the future) is calculated. The future active power transmission trend of the convergent substation will change to realize automatic voltage priority control, and the capacitors in the 750/500 convergent substation will be switched in priority to cope with the upcoming power changes.

The method for controlling the voltage of the wind power gathering station based on the trend prediction of the total active power proposed by the present invention includes the following steps:

(1) Set the voltage control period of the wind power gathering area;

(2) When each control cycle comes, read the power flow data of the current power grid from the power grid energy management system (EMS). The current power flow data of the power grid include node voltage, branch active power, branch reactive power, and branch current. and switch switch state, divide the power grid into multiple wind power collection zones Z _n , each wind power collection zone Z _n includes a collection substation T _n of the upper _- level power grid, and the outlet lines (L ₁ , ...,L _j ), j is the serial number of the outgoing line, and the equivalent wind farm units (G ₁ ,...,G _i ) of the lower power grid, each equivalent unit corresponds to a wind farm in the lower power grid, i is the serial number of the wind farm;

(3) Read and record the current active power value (P ₁₀ ,...,G _i ) of each wind farm equivalent unit (G ₁ ,...,G _i ) in the wind power collection zone Zn from the grid energy management system P _i0 ), that is, the active power of each wind farm corresponding to the equivalent units of each wind farm;

(4) Read the active power forecast data of the wind farm equivalent units from the power grid energy management system. The active power forecast data is 96 points of data throughout the day, one forecast data every 15 minutes, and the next one hour is read. 4 points of data, and the current active power value read in step (3), a total of 5 points of data, form an i*5-dimensional matrix, denoted as F _g , the elements P _i1 ~ P _i4 in the matrix F _g Represents the active power forecast data of the ith wind farm equivalent unit at 4 points in the next hour:

(5) According to the active power prediction data of the wind farm equivalent units obtained in step (4), the quasi-steady-state sensitivity calculation method in the power flow calculation of the power grid is used to obtain the (G ₁ ,...,G _i ) pairs of wind farm equivalent units. Active power sensitivity S _ij of lines (L ₁ ,...,L _j ), traverse all wind farm equivalent units G _i and lines L _j in the wind power collection zone _{Zn, repeat this step to obtain all active power sensitivities S ij ,} All active sensitivities S _ij form an i*j order sensitivity matrix S _gl as follows:

(6) According to the sensitivity matrix S _gl in step (5) and the active power prediction data F _g of wind farm equivalent units in step (4 _{), all outgoing lines (L 1 ,} ... , L _j ) the predicted active power in the next hour, and a j*5 order matrix F _l is obtained. In the matrix F _l , the element F _j0 represents the active power of the line L _j at the current moment, F _j1 , F _j2 , F _j3 , F _j4 in turn represents the predicted active power of line L _j every 15 minutes in the next hour:

(7) According to the predicted active power F l of the outgoing lines of the wind power gathering zone Zn in step (6), the predicted active power F _l in the next hour obtained _from all outgoing lines (L ₁ , . . . , L _{j )} is accumulated , that is, add the values of the vertical columns in the matrix F _l to obtain a 1*5 order vector L _n :

L _n =F _l ^T *I _n =[L _n,0 L _n,1 L _n,2 L _n,3 L _n,4 ]

Among them, is one, the superscript T is the matrix transpose, I _n is a j-order unit vector, I _n =[1,1,...1[ ^T , the element L _n,0 in the vector L _n is the current active power Power, L _{n, 1} L _{n, 2} L _{n, 3} L _{n, 4} in turn represent the predicted active power of the convergent substation T _n every 15 minutes in the next hour;

(8) Determine the change trend of the total wind power active power transmission of the upper-level grid collection substation in the next hour, and obtain an automatic voltage control method, including the following steps:

(8-1) Initialization, set the counter U=0 for the rising time of the active power trend, and the counter D=0 for the falling time of the active power trend;

(8-2) Judging the change trend of the total wind power active power transmission of the aggregated substation in the next 1 hour, and performing the following steps in sequence:

(8-2-1) According to step (7), obtain the active power trend prediction data L _n of the aggregated substation in the next hour, set k as the corresponding time point of the data from the current to the next hour, and the value of k is 0-3, then L _{n, k} represent the element data in the vector L _n , with k=0 as the initial value, and calculate in turn:

in,

(8-2-2) Set a threshold value S ₁ for the rate of change of the total wind power active power in the aggregated substation of the upper-level power grid, and a threshold value S ₂ for the rate of change of the load shedding. According to the set thresholds, the above ΔL ₀ is judged, if ΔL ₀ > 0, and satisfy ΔL ₀ >S ₁ , update the counter U of the active trend rising time, make U=U+1, and further judge k, if k is less than 3, return to step (8-2-1) , if k is greater than or equal to 3, go to step (8-2-3), if ΔL ₀ <0, and satisfy ΔL ₀ <S ₂ , update the counter D of the active trend down time, make D=D+1, and further Judgment on k, if k is less than 3, then make k=k+1, return to step (8-2-1), if k is greater than or equal to 3, proceed to step (8-2-3);

(8-2-3) Judging the updated counter U and counter D in step (8-2-2), if U>0, and D=0, then it is judged that the total active power of wind power in the aggregated substation is delivered in It will enter the rising stage in the next hour; if D>0 and U=0, it is determined that the total wind power active power transmission in the convergent substation will enter the descending stage in the next hour, and in other cases, the total wind power in the convergent substation will be determined. Active power delivery will be at a plateau in the next hour;

(8-3) In the upper-level power grid pooling substation, according to step (8-2-3), the judgment of the change trend of the total wind power active power transmission in the upper-level power grid pooling substation in the next 1 hour generates an automatic voltage control method for the upper-level power grid pooling substation, Perform the following steps in sequence:

(8-3-1) When the power flow value of the busbar voltage on the middle and high voltage side of the upper-level power grid convergent substation is less than the lower limit value of the busbar voltage set by the power grid:

If the total active power transmission of wind power in the collecting substation will enter the rising stage, remove the reactor in the collecting substation, or put the capacitor in the collecting substation; Then coordinate the lower power grid to increase the reactive power of the wind farm;

(8-3-2) When the power flow value of the busbar voltage on the middle and high voltage side of the upper-level power grid convergent substation is greater than the upper limit value of the busbar voltage set by the power grid:

If the total wind power active power transmission in the convergent substation will enter the rising stage, then cut off the capacitors in the convergent substation, or put in the reactor in the convergent substation; , then coordinate the lower power grid to reduce the reactive power output of the wind farm;

Realize the voltage control of wind power gathering station based on the trend prediction of total active power.

The advantages of the wind power gathering station voltage control method based on the total active power trend prediction proposed by the present invention are:

In the method of the present invention, the lower-level 330/220kV power grid system reads the wind power prediction data of the current time and the next hour in each calculation cycle, and calculates the active power sensitivity of the unit to the designated line through the quasi-steady-state sensitivity, and obtains the designated line. The predicted value of active power in the next hour, and then the active power predicted value of the line is added to get the total wind power active power trend of the entire 750/500kV aggregated substation in the future one hour, based on the aggregated substation future total wind power transmission trend change automatically to achieve voltage priority Control, judge the priority switching of capacitors in 750/500kV convergent substations. The invention calculates the active power prediction data of the tie line according to the tie line of the upper and lower power grids, the sensitivity of the unit to the tie line and the power prediction data of the unit, and then adds up to obtain the active power trend prediction of the upper-level aggregated substation. According to the active power prediction data, combined with the control strategy of automatic voltage control, the possible load fluctuations are reduced by adjusting the reactive power equipment of the substation and coordinating the reactive power processing of the lower power grid, thereby reducing the loss of equipment and ensuring the stability of the power grid. and security.

Description of drawings

Fig. 1 is a flow chart of the method of the present invention.

FIG. 2 is a schematic diagram of the connection relationship of the substation involved in the method of the present invention.

FIG. 3 is a flow chart of the automatic voltage control involved in the method of the present invention.

Detailed ways

The method for controlling the voltage of the wind power gathering station based on the total active power trend prediction proposed by the present invention has a flow chart as shown in Figure 1, and the connection relationship of the substation involved is shown in Figure 2. The method includes the following steps:

(1) Set the voltage control period of the wind power gathering area. In an embodiment of the present invention, the set period is 15 minutes (the text in the blue part is used as an explanatory term and can be reserved in the specific implementation of the description, the same below);

(2) When each control cycle comes, read the power flow data of the current power grid from the power grid energy management system (EMS). The current power flow data of the power grid include node voltage, branch active power, branch reactive power, and branch current. and switch switch state, divide the power grid into multiple wind power collection zones Z _n , each wind power collection zone Z _n includes a collection substation T _n of the upper _- level power grid, and the outlet lines (L ₁ , …,L _j ), j is the serial number of the outgoing line, and the equivalent units of the wind farm in the lower power grid (G ₁ ,…,G _i ), each equivalent unit corresponds to a wind farm in the lower power grid, and i is the wind farm the serial number;

(3) Read and record the _current active power value (P ₁₀ ,...,P _i0 ) of each wind farm equivalent unit (G ₁ ,...,G _i ) in the wind power collection zone Zn from the grid energy management system, That is, the active power of each wind farm corresponding to the equivalent units of each wind farm;

(4) Read the active power forecast data of the wind farm equivalent units from the power grid energy management system. The active power forecast data is 96 points of data throughout the day, one forecast data every 15 minutes, and the next one hour is read. 4 points of data, and the current active power value read in step (3), a total of 5 points of data, form an i*5-dimensional matrix, denoted as F _g , the elements P _i1 ~ P _i4 in the matrix F _g Represents the active power forecast data of the ith wind farm equivalent unit at 4 points in the next hour:

(5) According to the active power prediction data of the equivalent wind farm units obtained in step (4), using the quasi-steady-state sensitivity calculation method in the power flow calculation of the power grid to obtain the equivalent units (G ₁ ,...,G _i ) of the wind farm to the line ( The physical meaning of the active power sensitivity S _ij and S _ij of L ₁ ,..., L _j ) is: for each additional unit active power of the wind farm equivalent unit G _i , the active power change amount of the line L _j flowing into the upper-level collecting station, for the partition Z _n All the wind farm equivalent units and lines in the wind farm are the objects, traverse all the wind farm equivalent units G _i and lines L _j in the wind power collection zone Z _n , repeat this step to obtain all active power sensitivities S _ij , all active power sensitivities S _ij A sensitivity matrix S _gl of order i*j is formed as follows:

(6) According to the sensitivity matrix S _gl of step (5) and the active power prediction data F _g of wind farm equivalent units in step (4), calculate and obtain all outgoing lines (L ₁ , . . . , L ₁ , . _j ) The predicted active power in the next hour, obtain a j*5 order matrix F _l , in the matrix F _l , the element F _j0 represents the active power of the line L _j at the current moment, F _j1 , F _j2 , F _j3 , F _j4 in turn Represents the predicted active power of line L _j at every 15-minute point in the next hour:

(7) According to the predicted active power F l of the outgoing lines of the wind power gathering zone Zn in step (6), the predicted active power F _l in the next hour obtained _from all outgoing lines (L ₁ , . . . , L _{j )} is accumulated , that is, add the values of the vertical columns in the matrix F _l to obtain a 1*5 order vector L _n :

L _n =F _l ^T *I _n =[L _n,0 L _n,1 L _n,2 L _n,3 L _n,4 ]

Among them, is one, the superscript T is the matrix transpose, I _n is a j-order unit vector, I _n =[1,1,...1] ^T , the element L _n,0 in the vector L _n is the current active power Power, L _{n, 1} L _{n, 2} L _{n, 3} L _{n, 4} in turn represent the predicted active power of the convergent substation T _n every 15 minutes in the next hour;

(8) Determine the change trend of the total wind power active power transmission of the upper-level grid collection substation in the next hour, and obtain an automatic voltage control method, including the following steps:

(8-1) Initialization, set the counter U=0 for the rising time of the active power trend, and the counter D=0 for the falling time of the active power trend;

(8-2) Judging the change trend of the total wind power active power transmission of the aggregated substation in the next 1 hour, and performing the following steps in sequence:

(8-2-1) According to step (7), obtain the active power trend prediction data L _n of the aggregated substation in the next hour, set k as the corresponding time point of the data from the current to the next hour, and the value of k is 0-3, then L _{n, k} represent the element data in the vector L _n , with k=0 as the initial value, and calculate in turn:

in,

(8-2-2) Set a threshold value S ₁ for the rate of change of the total wind power active power in the aggregated substation of the upper-level power grid, and a threshold value S ₂ for the rate of change of the load shedding. The units of S ₁ and S ₂ are: MW/min, and can The threshold value of the increase or decrease load change rate is set according to the power generation change characteristics of the wind farm in the zone of the convergent substation. According to the set threshold, the above-mentioned ΔL ₀ is judged. If ΔL ₀ >0, and ΔL ₀ >S ₁ is satisfied, update the counter U of the rising time of the active power trend to make U=U+1, and further judge k, if If k is less than 3, go back to step (8-2-1), if k is greater than or equal to 3, go to step (8-2-3), if ΔL ₀ <0, and satisfy ΔL ₀ <S ₂ , update the active power trend For the counter D of the falling time, set D=D+1, and further judge k, if k is less than 3, set k=k+1, return to step (8-2-1), if k is greater than or equal to 3, proceed to Step (8-2-3);

(8-2-3) Judging the updated counter U and counter D in step (8-2-2), if U>0, and D=0, then it is judged that the total active power of wind power in the aggregated substation is delivered in It will enter the rising stage in the next hour; if D>0 and U=0, it is determined that the total wind power active power transmission in the convergent substation will enter the descending stage in the next hour, and in other cases, the total wind power in the convergent substation will be determined. Active power delivery will be at a plateau in the next hour;

(8-3) In the upper-level power grid pooling substation, according to step (8-2-3), the judgment of the change trend of the total wind power active power transmission in the upper-level power grid pooling substation in the next 1 hour generates an automatic voltage control method for the upper-level power grid pooling substation, Perform the following steps in sequence:

(8-3-1) When the power flow value of the busbar voltage on the middle and high voltage side of the upper-level power grid convergent substation is less than the lower limit value of the busbar voltage set by the power grid:

If the total active power transmission of wind power in the collecting substation will enter the rising stage, remove the reactor in the collecting substation, or put the capacitor in the collecting substation; Then coordinate the lower power grid to increase the reactive power of the wind farm;

(8-3-2) When the power flow value of the busbar voltage on the middle and high voltage side of the upper-level power grid convergent substation is greater than the upper limit value of the busbar voltage set by the power grid:

If the total wind power active power transmission in the convergent substation will enter the rising stage, then cut off the capacitors in the convergent substation, or put in the reactor in the convergent substation; , then coordinate the lower-level power grid to reduce the reactive power output of the wind farm; realize the voltage control of the wind power collection station based on the trend prediction of the total active power.

The working principle of the method of the invention is that, by performing quasi-steady-state sensitivity calculation on the existing lower-level provincial adjustment area model, the active power sensitivity of the wind turbine to the wind power output line is obtained, the power prediction data of the wind turbine is read, and the wind power output connection is calculated. The active power prediction data of the line is accumulated to obtain the total active power prediction trend of the 750/500kV aggregated substation of the upper power grid connected to the line. Combined with the control strategy of automatic voltage control, the reactive power equipment of the substation is adjusted preferentially and the reactive power of the lower power grid is coordinated. deal with possible load fluctuations.

An embodiment of the method of the present invention is introduced below:

This embodiment is a wind power collection substation with a 750kV collection substation. The connection relationship of the division Z _n in this embodiment is shown in Figure 2 . The _subsection includes a 750kV collection substation Tn, and the outgoing lines L1, 330kV voltage level _on the medium voltage side. L ₂ , L ₃ , L ₄ , L ₅ , L ₆ , 330kV substations T ₁ , T ₂ , T ₃ , and wind farm equivalent units G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , _G7 , _G8 , _G9 and other equipment;

(1) Set the voltage control period of the wind power gathering area, and in this embodiment, the set period is 15 minutes;

(2) When each control cycle comes, read the power flow data of the current power grid from the power grid energy management system (EMS). The current power flow data of the power grid include node voltage, branch active power, branch reactive power, and branch current. and switch switch state, divide the power grid into multiple wind power collection zones Z _n , each wind power collection zone Z _n includes a collection substation T _n of the upper _- level power grid, and the outlet lines (L ₁ , ...,L _j ), j is the serial number of the outgoing line, and the equivalent wind farm units (G ₁ ,...,G _i ) of the lower power grid, each equivalent unit corresponds to a wind farm in the lower power grid, i is the serial number of the wind farm;

_In this practical example, the zone Zn includes _{one 750kV} wind power gathering substation _Tn , and _six outgoing lines L1, L2, L3, _L4 , L5, L6 _of the medium voltage _330kV voltage level of the gathering substation _Tn , there are nine wind farm equivalent units G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ , and G ₉ .

(3) Read and record the current active power value (P ₁₀ ,...,G _i ) of each wind farm equivalent unit (G ₁ ,...,G _i ) in the wind power collection zone Zn from the grid energy management system P _i0 ), that is, the active power of each wind farm corresponding to the equivalent units of each wind farm;

_In this embodiment, the current time _{is 16:45 pm} , and the current active power values (P ₁₀ , . As shown in Table 1: (unit MW)

P₁₀ P₂₀ P₃₀ P₄₀ P₅₀ P₆₀ P₇₀ P₈₀ P₉₀ 25.32 24.98 26.01 46.02 47.11 47.98 16.11 17.67 16.68

Table 1

(4) Read the active power forecast data of the wind farm equivalent units from the power grid energy management system. The active power forecast data is 96 points of data throughout the day, one forecast data every 15 minutes, and the next one hour is read. 4 points of data, and the current active power value read in step (3), a total of 5 points of data, form an i*5-dimensional matrix, denoted as F _g , the elements P _i1 ~ P _i4 in the matrix F _g Represents the active power forecast data of the ith wind farm equivalent unit at 4 points in the next hour:

In this embodiment, when the current time is 16:45 pm, the four predicted data time points are: 17:00, 17:15, 17:30, 17:45, and 18:00 respectively, and the wind farm is obtained. The active power prediction data values of these four points of the equivalent unit, plus the active power value of the current time read in step (3), are shown in Table 2: (unit MW)

G₁ G₂ G₃ G₄ G₅ G₆ G₇ G₈ G₉ 0 (16:45) 25.32 24.98 26.01 46.02 47.11 47.98 16.11 17.67 16.68 1(17:00) 23.94 22.99 24.02 41.04 42.03 42.99 14.32 16.01 14.66 2 (17:15) 21.58 21.05 21.89 36.99 37.43 38.27 12.74 13.97 13.13 3 (17:30) 18.95 18.66 19.11 32.48 34.01 34.77 10.73 11.95 11.22 4 (17:45) 16.32 16.01 16.54 27.97 28.68 28.97 8.90 10.02 9.50

Table 2

(5) According to the active power prediction data of the wind farm equivalent units obtained in step (4), the quasi-steady-state sensitivity calculation method in the power flow calculation of the power grid is used to obtain the (G ₁ ,...,G _i ) pairs of wind farm equivalent units. The active power sensitivity S _ij of the line (L ₁ ,...,L _j ), the physical meaning of S _ij is: for each additional unit active power of the wind farm equivalent unit G _i , the active power change of the line L _j flowing into the upper gathering station, _Take all the wind farm equivalent units and lines in the zone Zn as the object, traverse all the wind farm equivalent units G _i and lines L _j in the wind power collection zone Zn _, repeat this step, and obtain all active power sensitivities S _ij , all The active sensitivity S _ij forms an i*j order sensitivity matrix S _gl as follows:

In this embodiment, the sensitivity matrix S _gl can be shown in Table 3:

G₁ G₂ G₃ G₄ G₅ G₆ G₇ G₈ G₉ L₁ 0.4351 0.4328 0.4278 0.0503 0.0188 0.0678 0.1236 0.1036 0.1134 L₂ 0.4351 0.4167 0.4389 0.0503 0.0188 0.0665 0.1278 0.1128 0.1092 L₃ 0.1247 0.1056 0.1176 0.4893 0.4782 0.4887 0.0976 0.0974 0.1031 L₄ 0.1378 0.1178 0.1072 0.4892 0.4890 0.4796 0.0954 0.0927 0.1043 L₅ 0.0523 0.0123 0.0631 0.0478 0.0165 0.0471 0.2982 0.2973 0.2842 L₆ 0.0412 0.0243 0.0512 0.0478 0.0165 0.0432 0.2981 0.2963 0.2982

table 3

(6) According to the sensitivity matrix S _gl in step (5) and the active power prediction data F _g of wind farm equivalent units in step (4 _{), all outgoing lines (L 1 ,} ... , L _j ) the predicted active power in the next hour, and a j*5 order matrix F _l is obtained. In the matrix F _l , the element F _j0 represents the active power of the line L _j at the current moment, F _j1 , F _j2 , F _j3 , F _j4 in turn represents the predicted active power of line L _j every 15 minutes in the next hour:

In this example, it can be concluded that:

(7) According to the predicted active power F l of the outgoing lines of the wind power gathering zone Zn in step (6), the predicted active power F _l in the next hour obtained _from all outgoing lines (L ₁ , . . . , L _{j )} is accumulated , that is, add the values of the vertical columns in the matrix F _l to obtain a 1*5 order vector L _n :

L _n =F _l ^T *I _n =[L _n,0 L _n,1 L _n,2 L _n,3 L _n,4 ]

Among them, the superscript T is the matrix transposition, I _n is a j-order unit vector, I _n =[1,1,...1] ^T , the element L _n,0 in the vector L _n is the current active power, L _{n, 1} L _{n, 2} L _{n, 3} L _{n, 4} in turn represent the predicted active power at each 15-minute point of the convergent substation T _n in the next hour. In this embodiment, L _n =[265.82 240.37 214.92 190.9 162.06]

(8) Determine the change trend of the total wind power active power transmission of the upper-level grid collection substation in the next hour, and obtain an automatic voltage control method, including the following steps:

(8-1) Initialization, set the counter U=0 for the rising time of the active power trend, and the counter D=0 for the falling time of the active power trend;

(8-2) Judging the change trend of the total wind power active power transmission of the aggregated substation in the next 1 hour, and performing the following steps in sequence:

(8-2-1) According to step (7), obtain the active power trend prediction data L _n of the aggregated substation in the next hour, set k as the corresponding time point of the data from the current to the next hour, and the value of k is 0-3, then L _{n, k} represent the element data in the vector L _n , with k=0 as the initial value, and calculate in turn:

(8-2-2) Set a threshold value S ₁ for the rate of change of the total wind power active power in the aggregated substation of the upper-level power grid, and a threshold value S ₂ for the rate of change of the load shedding. The units of S ₁ and S ₂ are: MW/min, and can The threshold value of the increase or decrease load change rate is set according to the power generation change characteristics of the wind farm in the zone of the convergent substation. According to the set threshold, the above-mentioned ΔL ₀ is judged. If ΔL ₀ >0, and ΔL ₀ >S ₁ is satisfied, update the counter U of the rising time of the active power trend to make U=U+1, and further judge k, if If k is less than 3, go back to step (8-2-1), if k is greater than or equal to 3, go to step (8-2-3), if ΔL ₀ <0, and satisfy ΔL ₀ <S ₂ , update the active power trend For the counter D of the falling time, set D=D+1, and further judge k, if k is less than 3, set k=k+1, return to step (8-2-1), if k is greater than or equal to 3, proceed to Step (8-2-3);

-0.0957<0并且-0.0957<-0.05，当前计数器D＝0，更新D＝D+1＝1，k＝0且k<3，则更新k＝k+1＝1，继续进行步骤(8-2-1)进行循环，直到判定k>＝3则退出循环，最后计算可以得出计数器D＝4，U＝0。In this embodiment, S ₁ =0.05 and S ₂ =-0.05 are set. Set the initial k = 0, then

-0.0957<0 and -0.0957<-0.05, current counter D=0, update D=D+1=1, k=0 and k<3, then update k=k+1=1, continue to step (8- 2-1) Execute the loop until it is determined that k>=3, then exit the loop, and finally calculate the counter D=4, U=0.

(8-2-3) Judging the updated counter U and counter D in step (8-2-2), if U>0, and D=0, then it is judged that the total active power of wind power in the aggregated substation is delivered in It will enter the rising stage in the next hour; if D>0 and U=0, it is determined that the total wind power active power transmission in the convergent substation will enter the descending stage in the next hour, and in other cases, the total wind power in the convergent substation will be determined. Active power delivery will be at a plateau in the next hour;

In this embodiment, if D>0 and U= ₀ , it can be determined that the total wind power active power transmission in the 750kV aggregated substation Tn will enter a decline phase in the next hour.

(8-3) In the upper-level power grid collection substation, according to step (8-2-3), the judgment of the change trend of the total wind power active power transmission of the upper-level power grid collection substation in the next 1 hour generates an automatic voltage control method for the upper-level power grid collection substation, Perform the following steps in sequence:

(8-3-1) When the power flow value of the busbar voltage on the middle and high voltage side of the upper-level power grid convergent substation is less than the lower limit value of the busbar voltage set by the power grid:

If the total active power transmission of wind power in the collecting substation will enter the rising stage, remove the reactor in the collecting substation, or put the capacitors in the collecting substation; Then coordinate the lower power grid to increase the reactive power of the wind farm;

(8-3-2) When the power flow value of the busbar voltage on the middle and high voltage side of the upper-level power grid convergent substation is greater than the upper limit value of the busbar voltage set by the power grid:

If the total wind power active power transmission in the convergent substation will enter the rising stage, then cut off the capacitors in the convergent substation, or put in the reactor in the convergent substation; , then coordinate the lower-level power grids to reduce the reactive power output of the wind farm; in this embodiment, since it has been determined in the previous step that the total active power transmission in the _750kV aggregated substation Tn will enter the descending stage, if the medium and high voltage in the aggregated substation _Tn If the _busbar voltage value is higher than the upper limit, it can be considered to reduce the reactive power in the station by removing the reactor in the collecting substation _Tn or putting in the capacitor. Reactive output of wind farms.

Claims (1)

1. A wind power collecting station voltage control method based on total active power trend prediction, is characterized in that, this method comprises the following steps: (1) Set the voltage control period of the wind power gathering area; (2) When each control cycle comes, read the power flow data of the current power grid from the power grid energy management system. The current power flow data of the power grid include node voltage, branch active power, branch reactive power, branch current and switch knife. In the gate state, the power grid is divided into multiple wind power collection zones Z _n , each wind power collection zone Z _n includes a collection substation T _n of the upper _- level power grid, and the outlet lines (L ₁ , . . . . , L _j ), j is the serial number of the outgoing line, and the equivalent units of the wind farm in the lower power grid (G ₁ ,...,G _i ), each equivalent unit corresponds to a wind farm in the lower power grid, and i is the wind power the serial number of the field; (3) Read and record the current active power value (P ₁₀ ,...,G _i ) of each wind farm equivalent unit (G ₁ ,...,G _i ) in the wind power collection zone Zn from the grid energy management system P _i0 ), that is, the active power of each wind farm corresponding to the equivalent units of each wind farm; (4) Read the active power forecast data of the wind farm equivalent units from the power grid energy management system. The active power forecast data is 96 points of data throughout the day, one forecast data every 15 minutes, and the next one hour is read. 4 points of data, and the current active power value read in step (3), a total of 5 points of data, form an i*5-dimensional matrix, denoted as F _g , the elements P _i1 -P _i4 in the matrix F _g Represents the active power forecast data of the ith wind farm equivalent unit at 4 points in the next hour:

(5) According to the active power prediction data of the wind farm equivalent units obtained in step (4), the quasi-steady-state sensitivity calculation method in the power flow calculation of the power grid is used to obtain the (G ₁ ,...,G _i ) pairs of wind farm equivalent units. Active power sensitivity S _ij of lines (L ₁ ,...,L _j ), traverse all wind farm equivalent units G _i and lines L _j in the wind power collection zone _{Zn, repeat this step to obtain all active power sensitivities S ij ,} All active sensitivities S _ij form an i*j order sensitivity matrix S _gl as follows:

(6) According to the sensitivity matrix S _gl in step (5) and the active power prediction data F _g of wind farm equivalent units in step (4 _{), all outgoing lines (L 1 ,} ... , L _j ) the predicted active power in the next hour, and a j*5 order matrix F _l is obtained. In the matrix F _l , the element F _j0 represents the active power of the line L _j at the current moment, F _j1 , F _j2 , F _j3 , F _j4 in turn represents the predicted active power of line L _j every 15 minutes in the next hour:

(7) According to the predicted active power F l of the outgoing lines of the wind power gathering zone Zn in step (6), the predicted active power F _l in the next hour obtained _from all outgoing lines (L ₁ , . . . , L _{j )} is accumulated , that is, add the values of the vertical columns in the matrix F _l to obtain a 1*5 order vector L _n : L _n =F _l ^T *I _n =[L _n,0 L _n,1 L _n,2 L _n,3 L _n,4 ] Among them, the superscript T is the matrix transposition, I _n is a j-order unit vector, I _n =[1,1,...1] ^T , the element L _n,0 in the vector L _n is the current active power, L _{n, 1} L _{n, 2} L _{n, 3} L _{n, 4} in turn represent the predicted active power of the convergent substation T _n every 15 minutes in the next hour; (8) Determine the change trend of the total wind power active power transmission of the upper-level grid collection substation in the next 1 hour, and obtain an automatic voltage control method, including the following steps: (8-1) Initialization, set the counter U=0 for the rising time of the active power trend, and the counter D=0 for the falling time of the active power trend; (8-2) Judging the change trend of the total wind power active power transmission of the aggregated substation in the next 1 hour, and performing the following steps in sequence: (8-2-1) According to step (7), obtain the active power trend prediction data L _n of the aggregated substation in the next hour, set k as the corresponding time point of the data from the current to the next hour, and the value of k is 0-3, then L _{n, k} represent the element data in the vector L _n , take k=0 as the initial value, and calculate in turn:

(8-2-2) Set a threshold value S ₁ for the rate of change of the total wind power active power in the aggregated substation of the upper-level power grid, and a threshold value S ₂ for the rate of change of the load shedding. According to the set thresholds, the above ΔL ₀ is judged, if ΔL ₀ > 0, and satisfy ΔL ₀ >S ₁ , update the counter U of the active trend rising time, make U=U+1, and further judge k, if k is less than 3, return to step (8-2-1) , if k is greater than or equal to 3, go to step (8-2-3), if ΔL ₀ <0, and satisfy ΔL ₀ <S ₂ , update the counter D of the active trend down time, make D=D+1, and further Judgment on k, if k is less than 3, then make k=k+1, return to step (8-2-1), if k is greater than or equal to 3, proceed to step (8-2-3); (8-2-3) Judging the updated counter U and counter D in step (8-2-2), if U>0, and D=0, then it is judged that the total active power of wind power in the aggregated substation is delivered in It will enter the rising stage in the next hour; if D>0 and U=0, it is determined that the total wind power active power transmission in the convergent substation will enter the descending stage in the next hour, and in other cases, the total wind power in the convergent substation will be determined. Active power delivery will be at a plateau in the next hour; (8-3) In the upper-level power grid pooling substation, according to step (8-2-3), the judgment of the change trend of the total wind power active power transmission in the upper-level power grid pooling substation in the next 1 hour generates an automatic voltage control method for the upper-level power grid pooling substation, Perform the following steps in sequence: (8-3-1) When the power flow value of the busbar voltage on the middle and high voltage side of the upper-level power grid convergent substation is less than the lower limit value of the busbar voltage set by the power grid: If the total active power transmission of wind power in the collecting substation will enter the rising stage, remove the reactor in the collecting substation, or put the capacitors in the collecting substation; Then coordinate the lower power grid to increase the reactive power of the wind farm; (8-3-2) When the power flow value of the busbar voltage on the middle and high voltage side of the upper-level power grid convergent substation is greater than the upper limit value of the busbar voltage set by the power grid: If the total wind power active power transmission in the convergent substation will enter the rising stage, cut off the capacitors in the convergent substation, or put in the reactor in the convergent substation; , then coordinate the lower power grid to reduce the reactive power output of the wind farm; Realize the voltage control of wind power gathering station based on the trend prediction of total active power.

Images