Power grid cascading failure blocking method and system based on coordination of VDCOL and current-limiting protection

Abstract

A VDCOL and current-limiting protection coordination-based power grid cascading failure blocking method comprises the steps of S1, constructing an improved CSEE-FS simulation model containing multi-loop direct current and new energy in MATLAB / SIMULINK, adopting a double-fed fan wind power plant as the model, accessing current-limiting protection to the wind power plant at a grid-connected point, adopting a conventional high-voltage direct-current power transmission system as a model direct-current module, and adopting a traditional VDCOL control strategy; s2, based on the improved CSEE-FS simulation model containing multi-loop direct current and new energy, a multi-inflection-point scheme is adopted, a traditional VDCOL control strategy is subjected to multi-section processing, and the parameter limitation range of the improved VDCOL is analyzed and set by considering the wind power plant low-voltage off-network protection limitation, whether current limiter protection is input or not and the wind power plant low-voltage ride-through limitation; and S3, based on the power grid cascading failure simulation data and the parameter limitation range of the improved VDCOL, solving through a DCS algorithm to obtain an improved VDCOL parameter. According to the design, the cascading failure of the alternating-current and direct-current hybrid power grid containing new energy can be effectively evaluated and blocked.

Description

Technical Field

[0001] The present invention relates to a method for blocking a cascading fault of a power grid, and in particular to a method and a system for blocking a cascading fault of a power grid based on coordination between VDCOL and current limiting protection. Background Art

[0002] With the development of social economy and industry, the power grid is moving towards low carbonization and AC / DC hybrid. While multiple DC circuits are being built, large-scale new energy sources are also being connected to the power grid. The high proportion of new energy access and the construction of multiple DC circuits have brought new operating characteristics to the power grid. The highly complex power grid structure has also led to an increase in power grid security risks. In recent years, a number of major power outages caused by cascading failures around the world have posed a great threat to the safe and stable operation of the power grid. How to efficiently and accurately prevent, control, and even block cascading failures is of great research significance. At present, the cascading failure blocking method mainly focuses on traditional AC power grids. Scholars analyze the evolution law of cascading failures, extract key events of cascading failures, and control them.

[0003] Traditional methods are less applicable to cascading failures in AC / DC hybrid power grids with a high proportion of renewable energy. In the evolution of cascading failures in such new power grids, three key events interact: AC line short circuit failure, DC commutation failure failure, and renewable energy disconnection failure. Therefore, a cascading failure blocking method that takes into account the interaction and suppression methods of DC commutation failure in new power grids and renewable energy disconnection events is urgently needed to solve the above problems in the existing technology. Summary of the invention

[0004] The purpose of the present invention is to overcome the problem that the traditional cascading fault blocking method in the prior art cannot effectively block the cascading faults of AC / DC hybrid power grids containing a high proportion of new energy, and provide a power grid cascading fault blocking method, system, device and storable medium based on the coordination of VDCOL and current limiting protection.

[0005] To achieve the above objectives, the technical solution of the present invention is:

[0006] In one aspect, the present invention provides a method for blocking power grid cascading faults based on coordination between VDCOL and current limiting protection, comprising the following steps:

[0007] S1 built an improved CSEE-FS simulation model containing multiple DC and new energy in the simulation software. The model used a double-fed wind turbine wind farm. The wind farm was connected to the grid connection point with current limiting protection. The DC module of the model was a conventional high-voltage DC transmission system, and the traditional VDCOL control strategy was adopted.

[0008] S2 is based on the improved CSEE-FS simulation model with multiple DC and renewable energy sources. It adopts a multi-inflection point solution to multi-stage the traditional VDCOL control strategy. By considering the low-voltage grid-trip protection limit of the wind farm, whether to put the current limiter into protection, and the low-voltage ride-through limit analysis of the wind farm, the parameter limit range of the improved VDCOL is set.

[0009] S3 obtains the improved VDCOL parameters through DCS algorithm based on the grid cascading failure simulation data and the parameter limit range of the improved VDCOL.

[0010] In S2, the piecewise function expression of VDCOL improved by multiple inflection points is as follows:

[0011]

[0012] In the formula, I ord is the DC current command value; I n is the rated value of DC current; U dc is the DC voltage; U max The upper voltage limit for starting the VDCOL strategy, U min is the lower limit of the VDCOL strategy startup voltage; k is the minimum DC current coefficient; U1 to U5 are the newly added 1 to 5 inflection points; k1 to k6 are the slope constants of 1 to 6 segments;

[0013] The objective function expression is as follows:

[0014] minf=∫|U L (t)-U N ∣dt

[0015] Where U L (t) is the commutation bus voltage at time t, U N is the rated value of the commutation bus voltage;

[0016] Considering the reactive power characteristics of the reactive compensation device and the converter station, the maximum value range constraints of the DC voltage and DC current of the improved VDCOL control parameters are obtained as follows:

[0017]

[0018] Among them, U max The upper voltage limit for starting the VDCOL strategy, I ord.max The upper limit of the startup current for the VDCOL strategy;

[0019] Therefore, the minimum value range of the DC voltage and DC current of the improved VDCOL control parameters is constrained to be:

[0020]

[0021] Among them, Umin The lower voltage limit for starting the VDCOL strategy; I ord.min The lower current limit for the VDCOL strategy startup.

[0022] Consider the wind farm low voltage disconnection protection limit:

[0023] The DC voltage and the voltage at the wind farm grid connection point affect each other during a fault:

[0024]

[0025] Where P f ne , Q f ne Respectively represent the active power and reactive power of the wind farm after the fault; P ne , Q ne Respectively represent the active power and reactive power of the wind farm; ΔP f ne , ΔQ f ne They represent the changes in active power and reactive power of the wind farm caused by the fault respectively; P f ac , Q f ac Respectively represent the active power and reactive power of the AC system after the fault; P ac , Q ac Respectively represent the active power and reactive power of the AC system; ΔP d , ΔQ d They represent the changes in active power and reactive power of the DC converter station caused by the fault; ΔQ f ac They represent the reactive power change of the AC system caused by the fault respectively;

[0026] DC voltage after fault U f dc And the voltage U of the wind farm grid connection point f ne It can be expressed as:

[0027]

[0028] Where U f dc Represents the DC voltage after the fault; U f ac Represents the voltage of the AC system after the fault; They represent the vertical component and the horizontal component of the voltage drop of the AC system after the fault respectively; U f ne Represents the voltage at the wind farm grid connection point after the fault; They represent the longitudinal component and transverse component of the voltage drop at the wind farm grid connection point after the fault respectively;

[0029] At this time, if the current limiter protection is not put into use, the bus voltage of the wind farm is the voltage of the wind farm grid connection point; after the current limiter protection is put into use, the bus voltage of the wind farm is:

[0030]

[0031] Where U rf ne is the bus voltage of the wind farm after the current limiting protection is put into use, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the transformation ratio of the grid-connected transformer of the wind farm;

[0032] Consider the wind farm low voltage ride-through limit as:

[0033] U ne =I s R FCL / k t 2 +U s ≥U ne.min

[0034] Where U ne is the bus voltage of the wind farm, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the wind farm grid-connected transformer ratio, U s is the terminal voltage of wind farm units, U ne.min It is the minimum voltage of the wind farm.

[0035] S301, randomly generate a solution matrix X within the parameter restriction range of the improved VDCOL, each solution in the solution matrix X represents a set of improved VDCOL model parameters, then enter step S302 to start iteration;

[0036] S302, sorting the solutions in descending order of the objective function values ​​according to the objective function values ​​of the solutions in the solution matrix X, and then proceeding to step S303;

[0037] S303, calculating the differentiated knowledge acquisition index of each solution according to the ranking result of the solutions, and then proceeding to step S304;

[0038] According to the ranking results of the solutions, the calculation of the differentiated knowledge acquisition index of the i-th solution includes:

[0039]

[0040] Where η i,t is the differential knowledge acquisition index of the i-th solution; U(0,1) is a random number uniformly distributed in the range [0,1]; is the knowledge incompleteness of the ith solution; the symbol [] indicates the rounding down operation; the symbol (f) indicates whether the formula f is valid. If it is valid, the value of (f) is 0, and if it is not valid, the value is 1; R i,t is the ranking of the i-th solution among all solutions; N is the total number of solutions in the solution matrix X;

[0041] S304, updating each solution according to the differentiated knowledge acquisition index of each solution, and then proceeding to step S305;

[0042] Based on the differentiated knowledge acquisition index of each solution, updates to each solution include:

[0043] According to the differentiated knowledge acquisition index of each solution, the solutions are ranked in order from low to high differentiated knowledge acquisition index;

[0044] For the solutions ranked in the bottom P%, the solutions are updated according to the following formula:

[0045] v i,d =ωx best,d +λ t (x r2,d -x i,d )+ω i,t (x r1,d -x i,d )

[0046] In the formula, v i,d is the updated ith solution, which is the solution ranked in the bottom P% by differential knowledge acquisition index; ω is the cognitive weight of the optimal solution; ω i,t is the cognitive weight of the ith solution; x best,d is the optimal solution, which is the solution with the smallest objective function value in the solution matrix X; t is the update index, x r2,d is the random solution ranked in the top (1-P)% by differential knowledge acquisition index, x r1,d For differential knowledge acquisition, the index ranking is in the last P% of random solutions, x i,d is the i-th solution before updating;

[0047] For the top (1-P)% of the solutions, the solutions are updated according to the following formula:

[0048] v j,d =x s1,d +LK(α,σ)

[0049] In the formula, v j,d is the jth solution after update, the jth solution is the solution ranked in the top (1-P)% of the order, x s1,d is the random solution ranked in the top (1-P)%; LK(α,σ) is a random number that satisfies the Linnik distribution;

[0050] S305, judging whether the iteration time reaches the preset time threshold, if so, stopping the iteration; after stopping the iteration, obtaining the optimal solution objective function as the fitness, if the fitness is less than 1, determining that the optimization converges, and entering step S306, if the fitness is greater than 1, the convergence fails, and returning to step S301; if the preset time threshold is not reached, returning to step S302 and continuing the iteration;

[0051] S306. Select a solution with the smallest corresponding objective function value from the results of all previous iterations.

[0052] In another aspect, the present invention provides a power grid cascading fault blocking system based on the coordination of VDCOL and current limiting protection, comprising a simulation model building module, an improved VDCOL building module and a parameter solving module;

[0053] Simulation model building module: used to build a CSEE-FS simulation model containing multiple DC and new energy in the simulation software. The model adopts a double-fed wind farm. The wind farm is connected to the current limiting protection at the grid connection point. The DC module of the model is a conventional high-voltage DC transmission system, which adopts the traditional VDCOL control strategy.

[0054] Improved VDCOL construction module: Based on the improved CSEE-FS simulation model with multiple DC and new energy sources, the traditional VDCOL control strategy is multi-staged by adopting a multi-inflection point solution. The parameter limit range of the improved VDCOL is set by considering the low-voltage grid-trip protection limit of the wind farm, whether to put the current limiter into protection, and the low-voltage ride-through limit analysis of the wind farm;

[0055] Parameter solving module: It is used to solve the improved VDCOL parameters through DCS algorithm based on the grid cascading failure simulation data and the parameter limit range of the improved VDCOL.

[0056] The piecewise function expression of VDCOL improved by multiple inflection points is as follows:

[0057]

[0058] In the formula, I ord is the DC current command value; In is the rated value of DC current; U dc is the DC voltage; U max The upper voltage limit for starting the VDCOL strategy, U min is the lower limit of the VDCOL strategy startup voltage; k is the minimum DC current coefficient; U1 to U5 are the newly added 1 to 5 inflection points; k1 to k6 are the slope constants of 1 to 6 segments;

[0059] The objective function expression is as follows:

[0060] minf=||U L (t)-U N ∣dt

[0061] Where U L (t) is the commutation bus voltage at time t, U N is the rated value of the commutation bus voltage;

[0062] Considering the reactive power characteristics of the reactive compensation device and the converter station, the maximum value range constraints of the DC voltage and DC current of the improved VDCOL control parameters are obtained as follows:

[0063]

[0064] Among them, U max The upper voltage limit for starting the VDCOL strategy, I ord.max The upper limit of the startup current for the VDCOL strategy;

[0065] Therefore, the minimum value range of the DC voltage and DC current of the improved VDCOL control parameters is constrained to be:

[0066]

[0067] Among them, U min The lower voltage limit for starting the VDCOL strategy; I ord.min The lower current limit for the VDCOL strategy startup.

[0068] Consider the wind farm low voltage disconnection protection limit:

[0069] The DC voltage and the voltage at the wind farm grid connection point affect each other during a fault:

[0070]

[0071] Where P f ne , Q f ne Respectively represent the active power and reactive power of the wind farm after the fault; P ne , Q neRespectively represent the active power and reactive power of the wind farm; ΔP f ne , ΔQ f ne They represent the changes in active power and reactive power of the wind farm caused by the fault respectively; P f ac , Q f ac Respectively represent the active power and reactive power of the AC system after the fault; P ac , Q ac Respectively represent the active power and reactive power of the AC system; ΔP d , ΔQ d They represent the changes in active power and reactive power of the DC converter station caused by the fault; ΔQ f ac They represent the reactive power change of the AC system caused by the fault respectively;

[0072] DC voltage after fault U f dc And the voltage U of the wind farm grid connection point f ne It can be expressed as:

[0073]

[0074] Where U f dc Represents the DC voltage after the fault; U f ac Represents the voltage of the AC system after the fault; They represent the vertical component and the horizontal component of the voltage drop of the AC system after the fault respectively; U f ne Represents the voltage at the wind farm grid connection point after the fault; They represent the longitudinal component and transverse component of the voltage drop at the wind farm grid connection point after the fault respectively;

[0075] At this time, if the current limiter protection is not put into use, the bus voltage of the wind farm is the voltage of the wind farm grid connection point; after the current limiter protection is put into use, the bus voltage of the wind farm is:

[0076]

[0077] Where U rf ne is the bus voltage of the wind farm after the current limiting protection is put into use, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the transformation ratio of the grid-connected transformer of the wind farm;

[0078] Consider the wind farm low voltage ride-through limit as:

[0079] U ne =I s R FCL / k t 2 +U s ≥U ne.min

[0080] Where U ne is the bus voltage of the wind farm, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the wind farm grid-connected transformer ratio, U s is the terminal voltage of wind farm units, U ne.min It is the minimum voltage of the wind farm.

[0081] S301, randomly generate a solution matrix X within the parameter restriction range of the improved VDCOL, each solution in the solution matrix X represents a set of improved VDCOL model parameters, then enter step S302 to start iteration;

[0082] S302, sorting the solutions in descending order of the objective function values ​​according to the objective function values ​​of the solutions in the solution matrix X, and then proceeding to step S303;

[0083] S303, calculating the differentiated knowledge acquisition index of each solution according to the ranking result of the solutions, and then proceeding to step S304;

[0084] According to the ranking results of the solutions, the calculation of the differentiated knowledge acquisition index of the i-th solution includes:

[0085]

[0086] Where η i,t is the differential knowledge acquisition index of the i-th solution; U(0,1) is a random number uniformly distributed in the range [0,1]; is the knowledge incompleteness of the ith solution; the symbol [] indicates the rounding down operation; the symbol (f) indicates whether the formula f is valid. If it is valid, the value of (f) is 0, and if it is not valid, the value is 1; R i,t is the ranking of the i-th solution among all solutions; N is the total number of solutions in the solution matrix X;

[0087] S304, updating each solution according to the differentiated knowledge acquisition index of each solution, and then proceeding to step S305;

[0088] Based on the differentiated knowledge acquisition index of each solution, updates to each solution include:

[0089] According to the differentiated knowledge acquisition index of each solution, the solutions are ranked in order from low to high differentiated knowledge acquisition index;

[0090] For the solutions ranked in the bottom P%, the solutions are updated according to the following formula:

[0091] v i,d =ωx best,d +λ t (x r2,d -x i,d )+ω i,t (x r1,d -x i,d )

[0092] In the formula, v i,d is the updated ith solution, which is the solution ranked in the bottom P% by differential knowledge acquisition index; ω is the cognitive weight of the optimal solution; ω i,t is the cognitive weight of the ith solution; x best,d is the optimal solution, which is the solution with the smallest objective function value in the solution matrix X; t is the update index, x r2,d is the random solution ranked in the top (1-P)% by differential knowledge acquisition index, x r1,d For differential knowledge acquisition, the index ranking is in the last P% of random solutions, x i,d is the i-th solution before updating;

[0093] For the top (1-P)% of the solutions, the solutions are updated according to the following formula:

[0094] v j,d =x s1,d +LK(α,σ)

[0095] In the formula, v j,d is the jth solution after update, the jth solution is the solution ranked in the top (1-P)% of the order, x s1,d is the random solution ranked in the top (1-P)%; LK(α,σ) is a random number that satisfies the Linnik distribution;

[0096] S305, judging whether the iteration time reaches the preset time threshold, if so, stopping the iteration; after stopping the iteration, obtaining the optimal solution objective function as the fitness, if the fitness is less than 1, determining that the optimization converges, and entering step S306, if the fitness is greater than 1, the convergence fails, and returning to step S301; if the preset time threshold is not reached, returning to step S302 and continuing the iteration;

[0097] S306. Select a solution with the smallest corresponding objective function value from the results of all previous iterations.

[0098] In another aspect, the present invention provides a power grid cascading fault blocking device based on VDCOL and current limiting protection coordination, including a processor and a memory;

[0099] The memory is used to store computer program code and transmit the computer program code to the processor;

[0100] The processor is used to execute the aforementioned power grid cascading fault blocking method based on the coordination of VDCOL and current limiting protection according to the instructions in the computer program code.

[0101] On the other hand, the present invention provides a computer storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned method for blocking cascading faults of a power grid based on coordination of VDCOL and current limiting protection.

[0102] Compared with the prior art, the present invention has the following beneficial effects:

[0103] 1. The present invention provides a power grid cascading fault blocking method based on the coordination of VDCOL and current limiting protection. The conventional linear VDCOL control link is multi-staged through multiple inflection points, and the slope regulation of different stages is used to improve the flexibility of the control link and enhance the fault blocking effect.

[0104] 2. The present invention provides a grid chain fault blocking method based on the coordination of VDCOL and current limiting protection. It comprehensively considers the influence of VDCOL parameter limitation and control strategy based on reactive characteristics analysis of converter station and reactive compensation device and current limiter on bus voltage of wind farm as constraint conditions. It optimizes the recovery and regulation ability of DC and peripheral voltage during fault process through reactive characteristics analysis optimization strategy. It further enhances the low voltage ride-through capability of wind turbine by considering the coordinated influence of current limiter protection, so as to better obtain the optimal parameter combination.

[0105] 3. The present invention provides a method for blocking power grid chain faults based on the coordination of VDCOL and current limiting protection, and uses a DCS algorithm to optimize the VDCOL multi-terminal control parameters. The algorithm is based on the proposed dual-strategy method to improve decision-making efficiency, while simulating the iterative cycle of balanced divergent and convergent thinking within a team framework, supported by a differentiated knowledge acquisition process and retrospective evaluation. The concepts of differentiated knowledge acquisition and creative realism are used to further accelerate the search efficiency and accuracy of the algorithm. By setting the upper and lower limits of each piecewise function of VDCOL as optimization variables, the present invention further improves the adaptability of the control strategy to faults, and can effectively improve the effect of the control strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figure 1 It is a schematic diagram of the multi-inflection point improved VDCOL characteristics constructed by the present invention.

[0107] Figure 2 It is a schematic diagram of the DCS algorithm adopted by the present invention.

[0108] Figure 3 It is a schematic diagram of the system topology of Example 1 of the present invention.

[0109] Figure 4 It is a schematic diagram of the installation of the current limiter of Example 1 of the present invention.

[0110] Figure 5 It is a comparison diagram of the convergence curves of the DCS algorithm and the PSO algorithm in Example 1 of the present invention.

[0111] Figure 6 It is a schematic diagram of blocking the disconnection link of a wind farm due to a cascading fault in Example 1 of the present invention.

[0112] Figure 7 Schematic diagram of the blocking of the DC interlocking link of the chain fault of the embodiment 1 of the present invention

[0113] Figure 8 It is a schematic diagram of the device of the present invention. DETAILED DESCRIPTION

[0114] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0115] Embodiment 1:

[0116] A method for blocking power grid cascading faults based on coordination between VDCOL and current limiting protection comprises the following steps:

[0117] See also Figure 1-Figure 4S1 built a simulation model of the improved frequency stability standard case (Chinese Society for Electrical Engineering-Frequency Stability, CSEE-FS) with multiple DC and new energy in MATLAB / SIMULINK. The model used a double-fed wind farm as a representative of new energy. The wind farm grid connection point was connected to current limiting protection. The DC module of the model was a conventional high-voltage DC transmission system (Line-commuted Converter-High Voltage Direct Current Transmission, LCC-HVDC), and the traditional VDCOL control strategy was adopted.

[0118] S2 is based on the improved CSEE-FS simulation model with multiple DC and new energy sources. It adopts a multi-inflection point solution to multi-stage the traditional VDCOL control strategy. The parameter range of each inflection point is set by analyzing the reactive characteristics of the converter station and the reactive compensation device. At the same time, the low-voltage grid-trip protection limit of the wind farm is considered, whether to put into current limiter protection and the low-voltage ride-through limit of the wind farm.

[0119] The piecewise function expression of traditional VDCOL is as follows:

[0120]

[0121] In the formula, I ord is the DC current command value; I n is the rated value of DC current; U dc is the DC voltage; U max The upper voltage limit for starting the VDCOL strategy, U min is the lower limit of the VDCOL strategy startup voltage; k is the minimum DC current coefficient; k I is the slope constant;

[0122] The piecewise function expression of VDCOL after multiple inflection points improvement is as follows:

[0123]

[0124] Wherein, U1 to U5 are the newly added 1 to 5 DC voltage inflection points; k1 to k6 are the slope constants of 1 to 6 segments;

[0125] The objective function expression is as follows:

[0126] minf=∫|U L (t)-U N ∣dt (3)

[0127] Where U L (t) is the commutation bus voltage at time t, UN is the rated value of the commutation bus voltage;

[0128] Considering the reactive power characteristics of reactive compensation devices and converter stations:

[0129] Q F =ωC R U acR (4)

[0130]

[0131]

[0132] B=U max I ord.min -U min I ord.max

[0133] C=(U max -U min )+(I ord.max -I ord.min )6X C / π

[0134] In the formula, Q F is the reactive power provided by the reactive power compensation device, ω is the angular frequency, C R is the equivalent capacitance of reactive power consumption, U acR is the AC system voltage, Q R Reactive power consumed by the converter station, N P is the number of poles, k is the transformer ratio, U L is the commutation bus voltage, α is the leading trigger angle, X C is the commutation reactance, I ord.max is the upper limit of the VDCOL strategy startup current, I ord.min is the lower limit of VDCOL strategy startup current, α min is the minimum firing angle, U max The upper voltage limit for starting the VDCOL strategy, U min The lower voltage limit for starting the VDCOL strategy;

[0135] Therefore, in order to minimize the reactive power absorbed by the converter station from the AC power grid, the maximum value range of the DC voltage and DC current of the improved VDCOL control parameters is constrained to be:

[0136]

[0137] Among them, U max The upper voltage limit for starting the VDCOL strategy, I ord.max The upper limit of the startup current for the VDCOL strategy;

[0138] Therefore, the minimum value range of the DC voltage and DC current of the improved VDCOL control parameters is constrained to be:

[0139]

[0140] Among them, U min The lower voltage limit for VDCOL strategy startup; I ord.min The lower limit of the starting current for the VDCOL strategy;

[0141] Consider the wind farm low voltage disconnection protection limit:

[0142] The DC transmission system encounters a commutation failure, and the original reactive power compensation supply is unbalanced; the DC side absorbs reactive power from the adjacent power grid, causing its voltage to drop, making the voltage U ne The voltage is further reduced, thus entering the low voltage ride through (LVRT) stage. Therefore, the mutual influence between the DC voltage and the voltage at the wind farm grid connection point during the fault period is further considered:

[0143]

[0144] Where P f ne , Q f ne Respectively represent the active power and reactive power of the wind farm after the fault; P ne , Q ne Respectively represent the active power and reactive power of the wind farm; ΔP f ne , ΔQ f ne They represent the changes in active power and reactive power of the wind farm caused by the fault respectively; P f ac , Q f ac Respectively represent the active power and reactive power of the AC system after the fault; P ac , Q ac Respectively represent the active power and reactive power of the AC system; ΔP d , ΔQ d They represent the changes in active power and reactive power of the DC converter station caused by the fault; ΔQ f ac They represent the reactive power change of the AC system caused by the fault respectively;

[0145] DC voltage after fault U f dc And the voltage U of the wind farm grid connection point f ne It can be expressed as:

[0146]

[0147] Where U f dc Represents the DC voltage after the fault; U f ac Represents the voltage of the AC system after the fault; They represent the vertical component and the horizontal component of the voltage drop of the AC system after the fault respectively; U f ne Represents the voltage at the wind farm grid connection point after the fault; They represent the longitudinal component and transverse component of the voltage drop at the wind farm grid connection point after the fault respectively;

[0148] At this time, if the current limiter protection is not put into use, the bus voltage of the wind farm is the voltage of the wind farm grid connection point; after the current limiter protection is put into use, the bus voltage of the wind farm is:

[0149]

[0150] Where U rf ne is the bus voltage of the wind farm after the current limiting protection is put into use, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the transformation ratio of the grid-connected transformer of the wind farm;

[0151] Consider the wind farm low voltage ride-through limit as:

[0152] U ne =I s R FCL / k t 2 +U s ≥U ne.min (11)

[0153] Where U ne is the bus voltage of the wind farm, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the wind farm grid-connected transformer ratio, U s is the terminal voltage of wind farm units, U ne.min is the minimum voltage of the wind farm;

[0154] S3 obtains improved VDCOL parameters through DCS algorithm based on power grid cascading failure simulation data and improved VDCOL parameter limit range.

[0155] S301, randomly generate a solution matrix X within the parameter restriction range of the improved VDCOL, each solution in the solution matrix X represents a set of improved VDCOL model parameters, then enter step S302 to start iteration;

[0156] S302, sorting the solutions in descending order of the objective function values ​​according to the objective function values ​​of the solutions in the solution matrix X, and then proceeding to step S303;

[0157] S303, calculating the differentiated knowledge acquisition index of each solution according to the ranking result of the solutions, and then proceeding to step S304;

[0158] According to the ranking results of the solutions, the calculation of the differentiated knowledge acquisition index of the i-th solution includes:

[0159]

[0160]

[0161] Where η i,t is the differential knowledge acquisition index of the i-th solution; U(0,1) is a random number uniformly distributed in the range [0,1]; is the knowledge incompleteness of the ith solution; the symbol [] indicates the rounding down operation; the symbol (f) indicates whether the formula f is valid. If it is valid, the value of (f) is 0, and if it is not valid, the value is 1; R i,t is the ranking of the i-th solution among all solutions; N is the total number of solutions in the solution matrix X;

[0162] S304, updating each solution according to the differentiated knowledge acquisition index of each solution, and then proceeding to step S305;

[0163] Based on the differentiated knowledge acquisition index of each solution, updates to each solution include:

[0164] According to the differentiated knowledge acquisition index of each solution, the solutions are ranked in order from low to high differentiated knowledge acquisition index;

[0165] For the solutions ranked in the bottom P%, the solutions are updated according to the following formula:

[0166] v i,d =ωx best,d +λ t (x r2,d -x i,d )+ω i,t (x r1,d -x i,d) (14)

[0167] In the formula, v i,d is the updated ith solution, which is the solution ranked in the bottom P% by differential knowledge acquisition index; ω is the cognitive weight of the optimal solution; ω i,t is the cognitive weight of the ith solution; x best,d is the optimal solution, which is the solution with the smallest objective function value in the solution matrix X; t is the update index, x r2,d is the random solution ranked in the top (1-P)% by differential knowledge acquisition index, x r1,d For differential knowledge acquisition, the index ranking is in the last P% of random solutions, x i,d is the i-th solution before updating;

[0168] For the top (1-P)% of the solutions, the solutions are updated according to the following formula:

[0169] v j,d =x s1,d +LK(α,σ) (15)

[0170] In the formula, v j,d is the jth solution after update, the jth solution is the solution ranked in the top (1-P)% of the order, x s1,d is the random solution ranked in the top (1-P)%; LK(α,σ) is a random number that satisfies the Linnik distribution;

[0171] S305, judging whether the iteration time reaches the preset time threshold, if so, stopping the iteration; after stopping the iteration, obtaining the optimal solution objective function as the fitness, if the fitness is less than 1, determining that the optimization converges, and entering step S306, if the fitness is greater than 1, the convergence fails, and returning to step S301; if the preset time threshold is not reached, returning to step S302 and continuing the iteration;

[0172] S306. Select a solution with the smallest corresponding objective function value from the results of all previous iterations.

[0173] Determine whether the algorithm has converged; if not, that is, the loop termination condition cannot be met, then repeat step S302 to continue iterating; if converged, that is, the loop termination condition is met, then output the iteration result, obtain the improved VDCOL parameter, and complete the optimization. The optimization result is shown below:

[0174]

[0175] In order to illustrate the effectiveness of the DCS algorithm used in the present invention, the PSO algorithm is used to optimize the parameters for comparison. The parameter optimization fitness convergence curves of the two algorithms are shown in Figure 5 As shown in the figure, it can be seen that the DCS algorithm converges faster than the PSO algorithm, and the final convergence effect is also better.

[0176] The cascading failure simulation verification is carried out based on the following four schemes:

[0177] Solution 1: The wind farm has no unlimited current protection and adopts traditional VDCOL control.

[0178] Option 2: The wind farm applies current limiting protection and adopts traditional VDCOL control.

[0179] Solution 3: The wind farm has no unlimited current protection and adopts optimized VDCOL control.

[0180] Scheme 4: Apply current limiting protection to the wind farm, adopt optimized VDCOL control, and coordinate optimization based on current limiting protection (method in this paper)

[0181] In scheme 1, at t = 2s, a three-phase permanent metal wire grounding fault occurs in line 3-6, causing a serious overload of the power flow in line 13-14, resulting in chain fault chains 3-6, 13-15, 12-13, and 9-10: schemes 2, 3, and 4 are used under the same fault to verify the blocking effect of each scheme on chain faults. In order to judge the chain fault blocking effect of each scheme, the wind farm voltage under the four schemes is compared. Figure 6 , DC turn-off angle see Figure 7 . Figure 6 In Scheme 3, due to the lack of wind farm current limiting protection like Scheme 1, the wind farm bus voltage dropped below 0.25pu after line 13-14 tripped, and could not meet the low voltage ride-through operation requirements and was disconnected from the grid. In Scheme 2 and the method in this paper, the voltage of the wind farm during the fault period increased to 0.39pu and 0.52pu respectively, and a certain power output was still maintained, indicating that the wind farm disconnection event was effectively blocked. Figure 7 In Scheme 1, the DC transmission system converter station shutdown angle was lower than the minimum shutdown angle three times in a row, resulting in continuous commutation failure and triggering a lockout fault, and the DC current dropped to zero. In contrast, Scheme 3 adopted optimized VDCOL control, and the DC transmission system only experienced one commutation failure, and the minimum shutdown angle was zero degrees. Although there was a certain fluctuation in the DC current, it did not cause a lockout, thus successfully blocking the development of cascading faults in the DC lockout link. In Scheme 2 and the method in this paper, the DC transmission system also only experienced one commutation failure, and the minimum shutdown angles were zero degrees and 8 degrees, respectively, and the DC current fluctuation was small. Figure 6 and Figure 7It can be seen that the present invention has a good blocking effect on wind farm grid-tripping faults and DC blocking faults in the evolution of cascading faults.

[0182] Embodiment 2:

[0183] A power grid cascading fault blocking system based on the coordination of VDCOL and current limiting protection, comprising a simulation model building module, an improved VDCOL building module and a parameter solving module;

[0184] Simulation model building module: used to build a simulation model of the improved frequency stability standard case (Chinese Society for Electrical Engineering-Frequency Stability, CSEE-FS) containing multiple DC and new energy in MATLAB / SIMULINK. The model uses a doubly fed wind turbine wind farm as a representative of new energy. The wind farm grid connection point is connected to current limiting protection. The DC module of the model is a conventional high-voltage DC transmission system (Line-commuted Converter-High Voltage Direct Current Transmission, LCC-HVDC), which adopts the traditional VDCOL control strategy;

[0185] Improved VDCOL construction module: Based on the improved CSEE-FS simulation model with multiple DC and new energy sources, the traditional VDCOL control strategy is multi-staged by adopting a multi-inflection point solution. The parameter limit range of the improved VDCOL is set by considering the low-voltage grid-trip protection limit of the wind farm, whether to put the current limiter into protection, and the low-voltage ride-through limit analysis of the wind farm;

[0186] Parameter solving module: It is used to solve the improved VDCOL parameters through DCS algorithm based on the grid cascading failure simulation data and the parameter limit range of the improved VDCOL.

[0187] The piecewise function expression of VDCOL improved by multiple inflection points is as follows:

[0188]

[0189] In the formula, I ord is the DC current command value; I n is the rated value of DC current; U dc is the DC voltage; U max The upper voltage limit for starting the VDCOL strategy, U min is the lower limit of the VDCOL strategy startup voltage; k is the minimum DC current coefficient; U1 to U5 are the newly added 1 to 5 inflection points; k1 to k6 are the slope constants of 1 to 6 segments;

[0190] The objective function expression is as follows:

[0191] minf=∫|U L (t)-U N ∣dt

[0192] Where U L (t) is the commutation bus voltage at time t, U N is the rated value of the commutation bus voltage;

[0193] Considering the reactive power characteristics of the reactive compensation device and the converter station, the maximum value range constraints of the DC voltage and DC current of the VDCOL control parameters are obtained as follows:

[0194]

[0195] Among them, U max The upper voltage limit for starting the VDCOL strategy, I ord.max The upper limit of the startup current for the VDCOL strategy;

[0196] Therefore, the minimum value range of the DC voltage and DC current of the VDCOL control parameter is constrained to be:

[0197]

[0198] Among them, U min The lower voltage limit for starting the VDCOL strategy; I ord.min The lower current limit for the VDCOL strategy startup.

[0199] Consider the wind farm low voltage disconnection protection limit:

[0200] The DC voltage and the voltage at the wind farm grid connection point affect each other during a fault:

[0201]

[0202] Where P f ne , Q f ne Respectively represent the active power and reactive power of the wind farm after the fault; P ne , Q ne Respectively represent the active power and reactive power of the wind farm; ΔP f ne , ΔQ f ne They represent the changes in active power and reactive power of the wind farm caused by the fault respectively; P f ac , Q f ac Respectively represent the active power and reactive power of the AC system after the fault; P ac , Qac Respectively represent the active power and reactive power of the AC system; ΔP d , ΔQ d They represent the changes in active power and reactive power of the DC converter station caused by the fault; ΔQ f ac They represent the reactive power change of the AC system caused by the fault respectively;

[0203] DC voltage after fault U f dc And the voltage U of the wind farm grid connection point f ne It can be expressed as:

[0204]

[0205] Where U f dc Represents the DC voltage after the fault; U f ac Represents the voltage of the AC system after the fault; They represent the vertical component and the horizontal component of the voltage drop of the AC system after the fault respectively; U f ne Represents the voltage at the wind farm grid connection point after the fault; They represent the longitudinal component and transverse component of the voltage drop at the wind farm grid connection point after the fault respectively;

[0206] At this time, if the current limiter protection is not put into use, the bus voltage of the wind farm is the voltage of the wind farm grid connection point; after the current limiter protection is put into use, the bus voltage of the wind farm is:

[0207]

[0208] Where U rf ne is the bus voltage of the wind farm after the current limiting protection is put into use, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the transformation ratio of the grid-connected transformer of the wind farm;

[0209] Consider the wind farm low voltage ride-through limit as:

[0210] U ne =I s R FCL / k t 2 +U s ≥U ne.min

[0211] Where U ne is the bus voltage of the wind farm, I sis the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the wind farm grid-connected transformer ratio, U s is the terminal voltage of wind farm units, U ne.min It is the minimum voltage of the wind farm.

[0212] S301, randomly generate a solution matrix X within the parameter restriction range of S2, each solution in the solution matrix X represents a set of improved VDCOL model parameters, then enter step S302 to start iteration;

[0213] S302, sorting the solutions in descending order of the objective function values ​​according to the objective function values ​​of the solutions in the solution matrix X, and then proceeding to step S303;

[0214] S303, calculating the differentiated knowledge acquisition index of each solution according to the ranking result of the solutions, and then proceeding to step S304;

[0215] According to the ranking results of the solutions, the calculation of the differentiated knowledge acquisition index of the i-th solution includes:

[0216]

[0217] Where η i,t is the differential knowledge acquisition index of the i-th solution; U(0,1) is a random number uniformly distributed in the range [0,1]; is the knowledge incompleteness of the ith solution; the symbol [] indicates the rounding down operation; the symbol (f) indicates whether the formula f is valid. If it is valid, the value of (f) is 0, and if it is not valid, the value is 1; R i,t is the ranking of the i-th solution among all solutions; N is the total number of solutions in the solution matrix X;

[0218] S304, updating each solution according to the differentiated knowledge acquisition index of each solution, and then proceeding to step S305;

[0219] Based on the differentiated knowledge acquisition index of each solution, updates to each solution include:

[0220] According to the differentiated knowledge acquisition index of each solution, the solutions are ranked in order from low to high differentiated knowledge acquisition index;

[0221] For the solutions ranked in the bottom P%, the solutions are updated according to the following formula:

[0222] v i,d =ωx best,d+λ t (x r2,d -x i,d )+ω i,t (x r1,d -x i,d )

[0223] In the formula, v i,d is the updated ith solution, which is the solution ranked in the bottom P% by differential knowledge acquisition index; ω is the cognitive weight of the optimal solution; ω i,t is the cognitive weight of the ith solution; x best,d is the optimal solution, which is the solution with the smallest objective function value in the solution matrix X; t is the update index, x r2,d is the random solution ranked in the top (1-P)% by differential knowledge acquisition index, x r1,d For differential knowledge acquisition, the index ranking is in the last P% of random solutions, x i,d is the i-th solution before updating;

[0224] For the top (1-P)% of the solutions, the solutions are updated according to the following formula:

[0225] v j,d =x s1,d +LK(α,σ)

[0226] In the formula, v j,d is the jth solution after update, the jth solution is the solution ranked in the top (1-P)% of the order, x s1,d is the random solution ranked in the top (1-P)%; LK(α,σ) is a random number that satisfies the Linnik distribution;

[0227] S305, judging whether the iteration time reaches the preset time threshold, if so, stopping the iteration; after stopping the iteration, obtaining the optimal solution objective function as the fitness, if the fitness is less than 1, determining that the optimization converges, and entering step S306, if the fitness is greater than 1, the convergence fails, and returning to step S301; if the preset time threshold is not reached, returning to step S302 and continuing the iteration;

[0228] S306. Select a solution with the smallest corresponding objective function value from the results of all previous iterations.

[0229] Embodiment 3:

[0230] See also Figure 8 , a power grid cascading fault blocking device based on the coordination of VDCOL and current limiting protection, comprising a processor and a memory;

[0231] The memory is used to store computer program code and transmit the computer program code to the processor;

[0232] The processor is used to execute the power grid cascading fault blocking method based on the coordination of VDCOL and current limiting protection according to the instructions in the computer program code.

[0233] Embodiment 4:

[0234] A computer storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for blocking power grid cascading faults based on coordination between VDCOL and current limiting protection.

Claims

1. A method for blocking power grid cascading faults based on coordination between VDCOL and current limiting protection, characterized in that: S1 built an improved CSEE-FS simulation model containing multiple DC and new energy in the simulation software. The model used a double-fed wind turbine wind farm. The wind farm was connected to the grid connection point with current limiting protection. The DC module of the model was a conventional high-voltage DC transmission system, and the traditional VDCOL control strategy was adopted. S2 is based on the improved CSEE-FS simulation model with multiple DC and renewable energy sources. It adopts a multi-inflection point solution to multi-stage the traditional VDCOL control strategy. By considering the low-voltage grid-trip protection limit of the wind farm, whether to put the current limiter into protection, and the low-voltage ride-through limit analysis of the wind farm, the parameter limit range of the improved VDCOL is set. S3 obtains the improved VDCOL parameters through DCS algorithm based on the grid cascading failure simulation data and the parameter limit range of the improved VDCOL.

2. A method for blocking power grid cascading failures based on coordination between VDCOL and current limiting protection according to claim 1, characterized in that: In S2, the piecewise function expression of VDCOL improved by multiple inflection points is as follows: In the formula, I ord is the DC current command value; I n is the rated value of DC current; U dc is the DC voltage; U max The upper voltage limit for starting the VDCOL strategy, U min is the lower limit of the VDCOL strategy startup voltage; k is the minimum DC current coefficient; U1 to U5 are the newly added 1 to 5 inflection points; k1 to k6 are the slope constants of 1 to 6 segments; The objective function expression is as follows: minf=∫∣U L (t)-U N ∣dt Where U L (t) is the commutation bus voltage at time t, U N is the rated value of the commutation bus voltage; Considering the reactive power characteristics of the reactive compensation device and the converter station, the maximum value range constraints of the DC voltage and DC current of the improved VDCOL control parameters are obtained as follows: Among them, U max The upper voltage limit for starting the VDCOL strategy, I ord.max The upper limit of the startup current for the VDCOL strategy; Therefore, the minimum value range of the DC voltage and DC current of the improved VDCOL control parameters is constrained to be: Among them, U min The lower voltage limit for VDCOL strategy startup; I ord.min The lower current limit for the VDCOL strategy startup.

3. A method for blocking power grid cascading failures based on coordination between VDCOL and current limiting protection according to claim 2, characterized in that: Consider the wind farm low voltage disconnection protection limit: The DC voltage and the voltage at the wind farm grid connection point affect each other during a fault: In the formula, Respectively represent the active power and reactive power of the wind farm after the fault; P ne , Q ne Respectively represent the active power and reactive power of the wind farm; They represent the changes in active power and reactive power of the wind farm caused by the fault respectively; Respectively represent the active power and reactive power of the AC system after the fault; P ac , Q ac Respectively represent the active power and reactive power of the AC system; ΔP d , ΔQ d They represent the changes in active power and reactive power of the DC converter station caused by the fault respectively; They represent the reactive power change of the AC system caused by the fault respectively; DC voltage after fault And the voltage of the wind farm grid connection point It can be expressed as: In the formula, Represents the DC voltage after the fault; Represents the voltage of the AC system after the fault; They represent the longitudinal component and transverse component of the voltage drop of the AC system after a fault respectively; Represents the voltage at the wind farm grid connection point after the fault; They represent the longitudinal component and transverse component of the voltage drop at the wind farm grid connection point after the fault respectively; At this time, if the current limiter protection is not put into use, the bus voltage of the wind farm is the voltage of the wind farm grid connection point; after the current limiter protection is put into use, the bus voltage of the wind farm is: In the formula, is the bus voltage of the wind farm after the current limiting protection is put into use, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the transformation ratio of the grid-connected transformer of the wind farm; Consider the wind farm low voltage ride-through limit as: U ne =I s R FCL / k t 2 +U s ≥U ne.min Where U ne is the bus voltage of the wind farm, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the wind farm grid-connected transformer ratio, U s is the terminal voltage of wind farm units, U ne.min It is the minimum voltage of the wind farm.

4. The method for blocking power grid cascading failures based on coordination between VDCOL and current limiting protection according to claim 1, characterized in that: In S3, S301, randomly generate a solution matrix X within the parameter restriction range of the improved VDCOL, each solution in the solution matrix X represents a set of improved VDCOL model parameters, then enter step S302 to start iteration; S302, sorting the solutions in descending order of the objective function values ​​according to the objective function values ​​of the solutions in the solution matrix X, and then proceeding to step S303; S303, calculating the differentiated knowledge acquisition index of each solution according to the ranking result of the solutions, and then proceeding to step S304; According to the ranking results of the solutions, the calculation of the differentiated knowledge acquisition index of the i-th solution includes: Where η i,t is the differential knowledge acquisition index of the ith solution; U(0,1) is a random number uniformly distributed in the range [0,1]; is the knowledge incompleteness of the ith solution; the symbol [] indicates the rounding down operation; the symbol (f) indicates whether the formula f is valid. If it is valid, the value of (f) is 0, and if it is not valid, the value is 1; R i,t is the ranking of the i-th solution among all solutions; N is the total number of solutions in the solution matrix X; S304, updating each solution according to the differentiated knowledge acquisition index of each solution, and then proceeding to step S305; Based on the differentiated knowledge acquisition index of each solution, updates to each solution include: According to the differentiated knowledge acquisition index of each solution, the solutions are ranked in order from low to high differentiated knowledge acquisition index; For the solutions ranked in the bottom P%, the solutions are updated according to the following formula: v i,d =ωx best,d +λ t (x r2,d -x i,d )+ω i,t (x r1,d -x i,d ) In the formula, v i,d is the updated ith solution, which is the solution ranked in the bottom P% by differential knowledge acquisition index; ω is the cognitive weight of the optimal solution; ω i,t is the cognitive weight of the ith solution; x best,d is the optimal solution, which is the solution with the smallest objective function value in the solution matrix X; t is the update index, x r2,d is the random solution ranked in the top (1-P)% by differential knowledge acquisition index, x r1,d For differential knowledge acquisition, the index ranking is the last P% of random solutions, x i,d is the i-th solution before updating; For the top (1-P)% of the solutions, the solutions are updated according to the following formula: v j,d =x s1,d +LK(α,σ) In the formula, v j,d is the jth solution after update, the jth solution is the solution ranked in the top (1-P)% of the order, x s1,d is the random solution ranked in the top (1-P)%; LK(α,σ) is a random number that satisfies the Linnik distribution; S305, judging whether the iteration time reaches the preset time threshold, if so, stopping the iteration; after stopping the iteration, obtaining the optimal solution objective function as the fitness, if the fitness is less than 1, determining that the optimization converges, and entering step S306, if the fitness is greater than 1, the convergence fails, and returning to step S301; if the preset time threshold is not reached, returning to step S302 and continuing the iteration; S306. Select a solution with the smallest corresponding objective function value from the results of all previous iterations.

5. A power grid cascading fault blocking system based on coordination between VDCOL and current limiting protection, characterized in that: Including simulation model building module, improved VDCOL building module and parameter solving module; Simulation model construction module: used to construct an improved CSEE-FS simulation model containing multiple DC and new energy in the simulation software. The model adopts a double-fed wind turbine wind farm. The wind farm is connected to the grid connection point with current limiting protection. The DC module of the model is a conventional high-voltage DC transmission system, which adopts the traditional VDCOL control strategy. Improved VDCOL construction module: Based on the improved CSEE-FS simulation model with multiple DC and new energy sources, the traditional VDCOL control strategy is multi-staged by adopting a multi-inflection point solution. The parameter limit range of the improved VDCOL is set by considering the low-voltage grid-trip protection limit of the wind farm, whether to put the current limiter into protection, and the low-voltage ride-through limit analysis of the wind farm; Parameter solving module: It is used to solve the improved VDCOL parameters through DCS algorithm based on the grid cascading failure simulation data and the parameter limit range of the improved VDCOL.

6. A power grid cascading fault blocking system based on coordination between VDCOL and current limiting protection according to claim 5, characterized in that: The piecewise function expression of VDCOL improved by multiple inflection points is as follows: In the formula, I ord is the DC current command value; I n is the rated value of DC current; U dc is the DC voltage; U max The upper voltage limit for starting the VDCOL strategy, U min is the lower limit of the VDCOL strategy startup voltage; k is the minimum DC current coefficient; U1 to U5 are the newly added 1 to 5 inflection points; k1 to k6 are the slope constants of 1 to 6 segments; The objective function expression is as follows: minf=∫∣U L (t)-U N ∣dt Where U L (t) is the commutation bus voltage at time t, U N is the rated value of the commutation bus voltage; Considering the reactive power characteristics of the reactive compensation device and the converter station, the maximum value range constraints of the DC voltage and DC current of the improved VDCOL control parameters are obtained as follows: Among them, U max The upper voltage limit for starting the VDCOL strategy, I ord.max The upper limit of the startup current for the VDCOL strategy; Therefore, the minimum value range of the DC voltage and DC current of the improved VDCOL control parameters is constrained to be: Among them, U min The lower voltage limit for VDCOL strategy startup; I ord.min The lower current limit for the VDCOL strategy startup.

7. A power grid cascading fault blocking system based on coordination between VDCOL and current limiting protection according to claim 6, characterized in that: Consider the wind farm low voltage disconnection protection limit: The DC voltage and the voltage at the wind farm grid connection point affect each other during a fault: In the formula, Respectively represent the active power and reactive power of the wind farm after the fault; P ne , Q ne Respectively represent the active power and reactive power of the wind farm; They represent the changes in active power and reactive power of the wind farm caused by the fault respectively; Respectively represent the active power and reactive power of the AC system after the fault; P ac , Q ac Respectively represent the active power and reactive power of the AC system; ΔP d , ΔQ d They represent the changes in active power and reactive power of the DC converter station caused by the fault respectively; They represent the reactive power change of the AC system caused by the fault respectively; DC voltage after fault And the voltage of the wind farm grid connection point It can be expressed as: In the formula, Represents the DC voltage after the fault; Represents the voltage of the AC system after the fault; They represent the longitudinal component and transverse component of the voltage drop of the AC system after a fault respectively; Represents the voltage at the wind farm grid connection point after the fault; They represent the longitudinal component and transverse component of the voltage drop at the wind farm grid connection point after the fault respectively; At this time, if the current limiter protection is not put into use, the bus voltage of the wind farm is the voltage of the wind farm grid connection point; after the current limiter protection is put into use, the bus voltage of the wind farm is: In the formula, is the bus voltage of the wind farm after the current limiting protection is put into use, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the transformation ratio of the grid-connected transformer of the wind farm; Consider the wind farm low voltage ride-through limit as: U ne =I s R FCL / k t 2 +U s ≥U ne.min Where U ne is the bus voltage of the wind farm, I s is the stator side current of the wind farm unit, R FCL is the resistance value of the current limiter, k t is the wind farm grid-connected transformer ratio, U s is the terminal voltage of wind farm units, U ne.min It is the minimum voltage of the wind farm.

8. The grid cascading fault blocking system based on the coordination of VDCOL and current limiting protection according to claim 5, characterized in that: In S3, S301, randomly generate a solution matrix X within the parameter restriction range of S2, each solution in the solution matrix X represents a set of improved VDCOL model parameters, then enter step S302 to start iteration; S302, sorting the solutions in descending order of the objective function values ​​according to the objective function values ​​of the solutions in the solution matrix X, and then proceeding to step S303; S303, calculating the differentiated knowledge acquisition index of each solution according to the ranking result of the solutions, and then proceeding to step S304; According to the ranking results of the solutions, the calculation of the differentiated knowledge acquisition index of the i-th solution includes: Where η i,t is the differential knowledge acquisition index of the ith solution; U(0,1) is a random number uniformly distributed in the range [0,1]; is the knowledge incompleteness of the ith solution; the symbol [] indicates the rounding down operation; the symbol (f) indicates whether the formula f is valid. If it is valid, the value of (f) is 0, and if it is not valid, the value is 1; R i,t is the ranking of the i-th solution among all solutions; N is the total number of solutions in the solution matrix X; S304, updating each solution according to the differentiated knowledge acquisition index of each solution, and then proceeding to step S305; Based on the differentiated knowledge acquisition index of each solution, updates to each solution include: According to the differentiated knowledge acquisition index of each solution, the solutions are ranked in order from low to high differentiated knowledge acquisition index; For the solutions ranked in the bottom P%, the solutions are updated according to the following formula: v i,d =ωx best,d +λ t (x r2,d -x i,d )+ω i,t (x r1,d -x i,d ) In the formula, v i,d is the updated ith solution, which is the solution ranked in the bottom P% by differential knowledge acquisition index; ω is the cognitive weight of the optimal solution; ω i,t is the cognitive weight of the ith solution; x best,d is the optimal solution, which is the solution with the smallest objective function value in the solution matrix X; t is the update index, x r2,d is the random solution ranked in the top (1-P)% by differential knowledge acquisition index, x r1,d For differential knowledge acquisition, the index ranking is the last P% of random solutions, x i,d is the i-th solution before updating; For the top (1-P)% of the solutions, the solutions are updated according to the following formula: v j,d =x s1,d +LK(α,σ) In the formula, v j,d is the jth solution after update, the jth solution is the solution ranked in the top (1-P)% of the order, x s1,d is the random solution ranked in the top (1-P)%; LK(α,σ) is a random number that satisfies the Linnik distribution; S305: Determine whether the iteration time reaches a preset time threshold. If so, stop the iteration and proceed to step S306. If not, return to step S302 and continue the iteration. S306. Select a solution with the smallest corresponding objective function value from the results of all previous iterations.

9. A power grid cascading fault blocking device based on the coordination of VDCOL and current limiting protection, characterized in that: including a processor and a memory; The memory is used to store computer program code and transmit the computer program code to the processor; The processor is used to execute the power grid cascading fault blocking method based on the coordination of VDCOL and current limiting protection according to any one of claims 1-4 according to the instructions in the computer program code.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a power grid cascading fault blocking method based on coordination of VDCOL and current limiting protection according to any one of claims 1 to 4 are implemented.

Images