Direct-current embedded power grid partitioning method and system considering new energy short-circuit current

Abstract

The invention relates to the technical field of power system planning, in particular to a DC embedded power grid partitioning method and system considering new energy short-circuit current, and the method comprises the steps: improving a multi-infeed short-circuit ratio, evaluating the voltage supporting capability of a system, and analyzing the cascading failure risk of the system in combination with a multi-infeed interaction factor; a controlled current source and impedance are combined, a sequence impedance model considering new energy characteristics is constructed, and the short-circuit current level of the system is evaluated; a line switching state is used as an optimization variable, a multi-target partition optimization model considering voltage support, cascading failure risks and system short-circuit characteristics is constructed, and collaborative optimization of power grid partitions is realized under power balance, power flow and voltage constraint conditions; according to the direct-current embedded power grid partitioning method, an adaptive inertia weight mechanism is introduced to improve a particle swarm algorithm, a partitioning model is solved in combination with Pareto sorting, an optimal power grid partitioning scheme is obtained, and the stability and the risk resistance of the direct-current embedded power grid partitioning scheme are improved through the technical means of quantitative evaluation and collaborative optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system planning, and particularly relates to a DC embedded power grid partitioning method and system considering short-circuit current of new energy. BACKGROUND

[0002] With the continuous promotion of China's power system towards energy green and low carbon, the power grid gradually transforms and upgrades to a new type of power system with new energy as the main body. In order to improve the transmission section capacity and alleviate the problem of uneven power flow distribution, embedded DC is gradually promoted and applied. However, the introduction of embedded DC makes the power grid structure gradually present a new pattern of AC-DC hybrid, multi-source coexistence and improved operation complexity. The interaction between AC and DC leads to significant changes in the voltage support capability, cascading failure characteristics and short-circuit current distribution of the system. Especially in the case of coexistence of embedded DC and high proportion of new energy, the AC power grid and embedded DC are highly coupled in operation and technology, and the traditional power grid planning method cannot accurately reflect the real coupling degree and stability characteristics of the system. Therefore, under the background of AC-DC hybrid system, it is of great significance to build a power grid partition optimization model by fully considering the dynamic characteristics of embedded DC and considering the short-circuit current safety constraint under new energy for realizing the coordinated planning of embedded DC and AC power grid and the safe and stable operation of the system.

[0003] At present, the research on power grid structure planning and optimization at home and abroad mainly focuses on ultra-high voltage transmission and other aspects, aiming to improve the stability of the power grid under the condition of new energy access. Most researches take the traditional long-distance transmission type DC system as the object, emphasize the expansion of transmission channel and optimization of power grid interconnection across regions, and some scholars propose that strengthening the network structure and reasonable hierarchical partition layout can effectively improve the stability of the power grid. However, existing researches are mostly based on AC systems or conventional DC, and the characteristics of embedded DC are not considered. Embedded DC usually realizes "AC to DC" within the same AC system, and there is close electrical coupling and power flow interaction between it and the AC power grid, which has a significant impact on system voltage support, short-circuit current constraint and dynamic recovery characteristics after fault. In addition, existing planning researches mostly focus on improving the connectivity and power supply capacity of the system, ignoring the coordinated optimization problem between AC and DC under the operation constraint of embedded DC.

[0004] In addition, how to effectively suppress the short-circuit current level has become a problem that cannot be ignored in the planning and operation of the receiving end power grid. Some scholars achieve global coordinated control of short-circuit current by establishing a power grid bi-level expansion planning model considering short-circuit current limitation, and some researchers achieve this purpose by establishing a network structure optimization model considering N-1 safety and short-circuit current of the power grid. However, the short-circuit current calculation in the above methods is still based on the synchronous generator model and the equivalent reactance method, and the influence of new energy units on the system short-circuit current characteristics is not considered. With the continuous increase of new energy installed capacity, the deviation between the traditional short-circuit current estimation result and the actual operation characteristics makes it difficult for the planning result to accurately reflect the real safety margin of the system.

[0005] On the other hand, the existing research on power grid partitioning problem is mostly based on clustering algorithms such as hierarchical clustering and spectral clustering: some scholars first determine a predetermined scheme that meets the requirements of transformer power supply and short-circuit current, and then use the K-Medoids algorithm to cluster nodes, and some scholars map the nodes to be clustered to the spatial coordinate system defined by the reactive power source, and convert the power grid partitioning problem into a clustering problem using spatial coordinate information. The above methods treat power grid partitioning as a static analysis problem, ignoring the coupling relationship between the partitioning result and the system operation constraints and planning objectives.

[0006] In summary, the above researches have considered the influencing factors of power system partitioning planning method to some extent, but these partitioning methods do not comprehensively consider the embedded DC characteristics, and also ignore the difference in short-circuit current calculation under the condition of high proportion of new energy, which is one-sided, and the rationality of the final power grid partitioning result still has room for further improvement. SUMMARY

[0007] To solve the above problems, the DC embedded power grid partitioning method considering new energy short-circuit current is provided, comprising:

[0008] Step S1, improve the voltage support capability of the multi-infeed short-circuit ratio evaluation system, and analyze the system cascading failure risk in combination with the multi-infeed interaction factor;

[0009] Step S2, adopt the form of controlled current source combined with impedance to construct a sequence impedance model considering the characteristics of new energy, and evaluate the system short-circuit current level;

[0010] Step S3, taking the line switching state as the optimization variable, a multi-objective partitioning optimization model considering voltage support, cascading failure risk and system short-circuit characteristics is constructed, and the coordinated optimization of power grid partitioning is realized under the conditions of power balance, power flow and voltage constraints;

[0011] Step S4, introduce an adaptive inertia weight mechanism to improve the particle swarm algorithm, and solve the partitioning model combined with Pareto sorting to obtain the optimal power grid partitioning scheme.

[0012] Further, in step S1, the improved multi-infeed short-circuit ratio evaluation system voltage support capability includes:

[0013] Based on the multi-infeed short-circuit ratio and the multi-port Thevenin equivalent method, the traditional multi-infeed short-circuit ratio is improved by introducing a DC power weighting mechanism to evaluate the voltage support strength in multiple partition states, and the expression is:

[0014] ;

[0015] In the formula, is the improved multi-infeed short-circuit ratio, n is the number of DC transmission systems, is the active power of the i-th DC transmission system, is the total active power, i is the DC loop number, is the short-circuit capacity of the DC feeder bus, is the equivalent DC power considering the influence of other DC loops, is the rated voltage of the converter bus, and are the equivalent node self-impedance and mutual impedance from each DC converter bus.

[0016] Further, in step S1, the system combined with the multi-infeed interaction factor analyzes the risk of cascading faults, which includes:

[0017] Based on the multi-infeed interaction factor and the node impedance method, the voltage coupling relationship between the multi-loop embedded DC systems is considered to evaluate the voltage coupling type cascading failure risk in multiple partition states, and the expression is:

[0018] ;

[0019] In the formula, is the multi-infeed interaction factor between two DC systems, is the voltage deviation on bus i, is the original voltage of bus j, is the mutual impedance between bus i and bus j, is the self-impedance of bus j.

[0020] Further, in step S2, the construction of the sequence impedance model considering the characteristics of new energy includes:

[0021] Based on the symmetrical component method, a short-circuit current calculation model of new energy units considering wind power and photovoltaic is established, and the positive sequence model adopts a controlled current source model:

[0022] ;

[0023] In the formula, is the increment of output current before and after the fault, is the grid-connected point voltage, and is the active and reactive current during the fault, and is the rated active and reactive power before the fault, and is the rated capacity and current;

[0024] Based on the symmetrical component method, a short-circuit current calculation model of a new energy unit considering wind power and photovoltaic power is established, and the negative sequence model is an impedance model:

[0025] ;

[0026] In the formula, is the negative sequence impedance, and is the rated capacity and rated voltage, is the negative sequence impedance correction coefficient;

[0027] Based on the symmetrical component method, a short-circuit current calculation model of a new energy unit considering wind power and photovoltaic power is established, and the zero sequence model is an impedance model determined by the neutral grounding mode of the unit transformer.

[0028] Further, in step S3, the multi-objective partition optimization model considering voltage support, cascading failure risk and system short-circuit characteristics comprises:

[0029] Based on the variability of the network topology, the regional tie line and the key section are selected as the optimization object, the opening decision of the line is selected as the optimization variable, and the equivalent impedance of the system is adjusted by changing the on or off state;

[0030] Based on the calculated voltage support capability, cascading failure risk and short-circuit current level as the core evaluation index, a multi-objective optimization function is constructed, and its expression is:

[0031] ;

[0032] In the formula, X is the decision variable, is the number of buses, is the short-circuit current of the i-th bus when a fault occurs, is the average value of the multi-infeed interaction factor, is the improved multi-infeed short-circuit ratio.

[0033] Further, in step S3, the constraint conditions of the multi-objective partition optimization model include power balance constraint, power flow constraint and voltage constraint, and its expression is:

[0034] ;

[0035] In the formula, This represents the power output of the i-th generator. This represents the power demand of the j-th load. Indicates total transmission loss. and These are the active power and reactive power of the i-th bus, respectively. This represents the voltage of the i-th bus. and These represent the susceptance and phase angle difference between busbars i and j, respectively. , and Representing nodes respectively The actual voltage amplitude, minimum voltage amplitude, and maximum voltage amplitude.

[0036] Furthermore, in step S4, the improvement of the particle swarm algorithm by introducing an adaptive inertia weight mechanism includes:

[0037] ;

[0038] in, and These are the upper and lower limits of the inertia weight, respectively. is the number of iterations, and is the maximum number of iterations.

[0039] Further, in step S4, the step of solving the partitioning model using Pareto sort includes:

[0040] Based on the improved particle swarm optimization algorithm and Pareto non-dominated sorting, the multi-objective optimization model of power grid partitioning is solved to obtain a non-dominated solution set that satisfies multiple constraints.

[0041] Further, in step S4, obtaining the optimal power grid partitioning scheme includes:

[0042] Based on typical operating scenarios and simulation analysis, the obtained Pareto front solutions are verified and compared, and the scheme with the best comprehensive performance among multiple objective indicators is selected to determine the final power grid zoning scheme.

[0043] A system for applying the DC embedded grid partitioning method that takes into account the short-circuit current of new energy sources includes:

[0044] The operation characteristic analysis module is used to improve the voltage support capability of the multi-infeed short-circuit ratio assessment system and analyze the risk of system cascading failures by combining the multi-infeed interaction factor.

[0045] The short-circuit current modeling module is used to construct a sequence impedance model that takes into account the characteristics of new energy sources by combining controlled current sources with impedance, so as to realize the quantitative assessment of the system's short-circuit current level.

[0046] The partition model construction module is connected to the operation characteristic analysis module and the short-circuit current modeling module respectively. It is used to construct a multi-objective partition optimization model that considers voltage support, cascading fault risk and system short-circuit characteristics with the line switching state as the optimization variable, so as to achieve coordinated optimization of power grid partitions under the conditions of power balance, power flow and voltage constraints.

[0047] The optimization solution module is connected to the operating characteristic analysis module, the short-circuit current modeling module, and the partition model construction module, respectively. It is used to introduce an adaptive inertia weight mechanism to improve the particle swarm algorithm, and combine Pareto sorting to solve the partition model to obtain the optimal power grid partition scheme.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a DC embedded grid partitioning method and system that takes into account the short-circuit current of new energy sources. It combines the operating mechanism and stability influencing factors of AC-DC hybrid systems, improves the voltage support capability of the multi-infeed short-circuit ratio evaluation system, and combines the multi-infeed interaction factor analysis system cascading failure risk to achieve quantitative evaluation of the characteristics of embedded DC grids. It fully considers the effect of embedded DC operating characteristics on system stability and avoids the one-sidedness of existing planning methods that are only based on traditional AC or conventional DC.

[0049] Furthermore, this invention considers the impact of renewable energy access on short-circuit current and establishes a calculation model for renewable energy short-circuit current determined by positive-sequence controlled current source, negative-sequence impedance, and zero-sequence grounding method. This avoids the calculation deviations caused by existing methods based on synchronous generator and equivalent reactance models, and can more accurately assess the short-circuit constraint level and safety margin of the system under high-proportion renewable energy access.

[0050] Furthermore, this invention focuses on solving the problem of power grid zoning with embedded DC, and can serve as an important technical support for the planning and operation decision-making of AC / DC hybrid systems, thereby realizing the coordinated planning of embedded DC and AC power grids and the safe and stable operation of the system.

[0051] Furthermore, by introducing a DC power weighting mechanism and multi-port Thevenin equivalence, the improved multi-infeed short-circuit ratio index not only reflects the impact of changes in grid structure (node ​​impedance) but also quantifies the differences in power transmission between different DC systems. This enables a more accurate assessment of the voltage support strength of the grid to the DC system under various zoning operation conditions, overcoming the limitations of traditional methods in inaccurate assessments in complex multi-DC systems. By calculating the multi-infeed interaction factor based on the node impedance method, this index clearly reveals the voltage coupling strength between different DC converter buses. This allows for a quantitative assessment of the risk of cascading failures in adjacent DC systems caused by a fault in one DC line under different zoning schemes, providing a direct and quantitative decision-making basis for selecting low-coupling, high-safety zoning schemes.

[0052] Furthermore, this invention employs a hybrid modeling method combining positive-sequence controlled current sources and negative-sequence / zero-sequence impedances. This model accurately reflects the dominant role of the controlled source characteristics (positive sequence) of new energy units during faults and reasonably describes their response characteristics to negative-sequence and zero-sequence components, thereby comprehensively and accurately calculating the actual impact of new energy grid connection on the system's short-circuit current level. Based on the mature symmetrical component method, this model provides clear and easily obtainable mathematical models for positive-sequence, negative-sequence, and zero-sequence components. In particular, it clarifies the key factor that the zero-sequence path depends on the transformer grounding method, providing a standardized and operable solution for large-scale short-circuit current calculation and analysis in power grids with a high proportion of new energy. This sequence model can effectively analyze the situation when the system experiences asymmetrical faults. By accurately calculating each sequence component, it provides a crucial data foundation for evaluating the coordination of protection devices and the safety and stability of the system under asymmetrical faults.

[0053] Furthermore, this invention integrates three key stability indicators—voltage support capability, cascading failure risk, and system short-circuit characteristics—into a unified multi-objective optimization model, providing a comprehensive and quantitative comparison benchmark and optimization direction for different partitioning schemes. By changing the decision variable of line switching status, the network topology adjustment is directly linked to the system's stability, safety, and other operational characteristics. This enables the optimization process to automatically find the Pareto optimal partitioning scheme that simultaneously improves voltage support strength, reduces cascading failure risk, and controls short-circuit current levels, while meeting rigid safety constraints such as power balance, power flow, and voltage. By organically combining core indicators reflecting the overall stability level of the system with constraints that ensure local safety, the final partitioning scheme can both improve the overall stability of the system macroscopically and meet all necessary engineering safety operation requirements microscopically.

[0054] Furthermore, this invention employs an adaptive inertia weighting mechanism, enabling the algorithm to maintain strong global exploration capabilities in the early stages of iteration to broadly search the solution space, while enhancing local development capabilities in the later stages of iteration for refined convergence. This effectively balances the contradiction between exploration and development, overcoming the shortcomings of the basic particle swarm optimization algorithm, which is prone to premature convergence or getting trapped in local optima, thus ensuring a more reliable search for high-quality partitioning schemes. By combining Pareto non-dominated sorting, the algorithm can directly solve multi-objective optimization models, outputting a Pareto optimal solution set composed of numerous non-dominated solutions. This solution set intuitively demonstrates the competitive and non-simultaneous optimal trade-offs among the three objectives of voltage support capability, cascading fault risk, and short-circuit current level, providing decision-makers with a rich pool of candidate schemes with different emphases. Through simulation verification and comprehensive comparison of Pareto front solutions based on typical operating scenarios, this process can screen out schemes that are robust under various actual operating conditions and achieve the best balance among multiple key indicators. This ensures that the final selected power grid partitioning scheme is not only theoretically optimal but also possesses high engineering practical value, adaptability, and risk resistance. Attached Figure Description

[0055] Figure 1 This is a flowchart illustrating the DC embedded power grid partitioning method of the present invention that takes into account the short-circuit current of new energy sources.

[0056] Figure 2 This is a structural diagram of an actual receiving-end system containing embedded DC in an embodiment of the DC embedded grid partitioning method for considering the short-circuit current of new energy sources according to the present invention;

[0057] Figure 3 This is a schematic diagram of a specific receiving-end grid partitioning plan containing embedded DC power in an embodiment of the DC embedded grid partitioning method for considering the short-circuit current of new energy sources according to the present invention.

[0058] Figure 4 The simulation curves are for embodiments of the DC embedded grid partitioning method of the present invention that take into account the short-circuit current of new energy sources, but do not adopt the method of the present invention.

[0059] Figure 5 The simulation curves are shown in the embodiment of the DC embedded grid partitioning method of the present invention, which takes into account the short-circuit current of new energy sources, after adopting the method of the present invention.

[0060] Figure 6 This is a schematic diagram of the DC embedded power grid partitioning system that takes into account the short-circuit current of new energy sources, as described in this invention. Detailed Implementation

[0061] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0062] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0063] Please see Figures 1-5 As shown, Figure 1 This is a flowchart illustrating the DC embedded power grid partitioning method of the present invention that takes into account the short-circuit current of new energy sources. Figure 2 This is a structural diagram of an actual receiving-end system containing embedded DC in an embodiment of the DC embedded grid partitioning method for considering the short-circuit current of new energy sources according to the present invention; Figure 3 This is a schematic diagram of a specific receiving-end grid partitioning plan containing embedded DC power in an embodiment of the DC embedded grid partitioning method for considering the short-circuit current of new energy sources according to the present invention. Figure 4 The simulation curves are for embodiments of the DC embedded grid partitioning method of the present invention that take into account the short-circuit current of new energy sources, but do not adopt the method of the present invention. Figure 5 This is a simulation curve of the DC embedded grid partitioning method of the present invention after adopting the method of the present invention in an embodiment of the method for partitioning DC grids that takes into account the short-circuit current of new energy sources.

[0064] The present invention provides a DC embedded grid partitioning method that takes into account the short-circuit current of new energy sources, comprising:

[0065] Step S1: Improve the evaluation of the system voltage support capability by multi-infeed short-circuit ratio, and analyze the risk of system cascading failures by combining multi-infeed interaction factors.

[0066] Step S2: Using a combination of controlled current sources and impedance, a sequence impedance model considering the characteristics of new energy sources is constructed to evaluate the system short-circuit current level.

[0067] Step S3: Using the line switching status as the optimization variable, construct a multi-objective zonal optimization model that considers voltage support, cascading fault risk and system short-circuit characteristics, and achieve coordinated optimization of power grid zonals under power balance, power flow and voltage constraints.

[0068] Step S4: An adaptive inertia weighting mechanism is introduced to improve the particle swarm algorithm. Combined with Pareto sorting, the partitioning model is solved to obtain the optimal power grid partitioning scheme.

[0069] This invention provides a DC embedded grid partitioning method and system that takes into account the short-circuit current of new energy sources. It combines the operating mechanism and stability influencing factors of AC-DC hybrid systems, improves the evaluation of the voltage support capability of the system by multi-infeed short-circuit ratio, and combines the analysis of the system cascading failure risk by multi-infeed interaction factors. It realizes the quantitative evaluation of the characteristics of embedded DC grids, fully considers the effect of embedded DC operating characteristics on system stability, and avoids the one-sidedness of existing planning methods that are only based on traditional AC or conventional DC.

[0070] This invention considers the impact of renewable energy access on short-circuit current and establishes a calculation model for renewable energy short-circuit current determined by positive-sequence controlled current source, negative-sequence impedance, and zero-sequence grounding method. This avoids the calculation deviations caused by existing methods based on synchronous generator and equivalent reactance models, and can more accurately assess the short-circuit constraint level and safety margin of the system under high-proportion renewable energy access.

[0071] This invention addresses the grid zoning problem involving embedded DC power and can serve as a crucial technical support for the planning and operation decisions of AC / DC hybrid systems, thereby enabling coordinated planning of embedded DC and AC power grids and the safe and stable operation of the system.

[0072] Specifically, in step S1, the improved multi-infeed short-circuit ratio evaluation system voltage support capability includes:

[0073] Based on the multi-infeed short-circuit ratio and multi-port Thevenin equivalent method, a DC power weighting mechanism is introduced to improve the traditional multi-infeed short-circuit ratio, and the voltage support strength under various partition conditions is evaluated. Its expression is:

[0074] ;

[0075] In the formula, The improved multi-infeed short-circuit ratio, where n is the number of DC transmission systems. It is the active power of the i-th DC transmission system. The sum of active power, where i is the DC circuit number. This refers to the short-circuit capacity of the DC-fed converter bus. To account for the equivalent DC power after considering the influence of other DC circuits, This is the rated voltage of the converter bus. and The equivalent node self-impedance and mutual impedance are seen from each DC converter bus.

[0076] Specifically, in step S1, the combined multi-feed interaction factor analysis of system cascading failure risk includes:

[0077] Based on the multi-feed interaction factor and nodal impedance method, considering the voltage coupling relationship between multi-circuit embedded DC systems, the risk of voltage-coupled cascading faults under various partition conditions is evaluated, and its expression is:

[0078] ;

[0079] In the formula, This refers to the multi-feed interaction factor between two DC systems. The voltage deviation on bus i. Let be the original voltage of bus j. Let be the mutual impedance between bus i and bus j. Let J be the self-impedance of bus j.

[0080] In this embodiment of the invention, after embedded DC access, the voltage support capability of the power grid and the voltage coupling relationship between nodes change significantly. The system stability depends not only on the short-circuit capacity of the AC network itself, but also on the interaction between multiple DC inlets. Based on the equivalent impedance relationship of the system, the voltage of the node where the DC converter station is located depends on the short-circuit capacity of the node and the degree of electrical coupling between it and other DC landing points. When the system operates under different line switching states, the equivalent impedance of the nodes changes, causing a redistribution of short-circuit capacity and mutual impedance, which leads to dynamic changes in voltage support strength. Therefore, based on the multi-inlet short-circuit ratio and the multi-port Thevenin equivalent method, a DC power weighting mechanism is introduced to improve the traditional multi-inlet short-circuit ratio, which can evaluate the voltage support strength under various partition states.

[0081] This invention introduces a DC power weighting mechanism and multi-port Thevenin equivalence. The improved multi-infeed short-circuit ratio index not only reflects the impact of changes in grid structure (node ​​impedance) but also quantifies the differences in power transmission between different DC systems. This enables a more accurate assessment of the voltage support strength of the grid to the DC system under various zoning operation conditions, overcoming the limitations of traditional methods in inaccurate assessments in complex multi-DC systems. By calculating the multi-infeed interaction factor based on the node impedance method, this index clearly reveals the voltage coupling strength between different DC converter buses. This allows for a quantitative assessment of the risk of cascading failures in adjacent DC systems caused by a fault in one DC line under different zoning schemes, providing a direct and quantitative decision-making basis for selecting low-coupling, high-safety zoning schemes.

[0082] Specifically, in step S2, constructing the sequence impedance model that takes into account the characteristics of new energy sources includes:

[0083] Based on the symmetric component method, a short-circuit current calculation model for new energy units considering wind power and photovoltaic power is established. The positive sequence model adopts a controlled current source model.

[0084] ;

[0085] In the formula, This represents the increase in output current before and after the fault. The voltage at the grid connection point. and The active and reactive currents during the fault period and The rated active and reactive power before the fault. and Rated capacity and current;

[0086] Based on the symmetrical component method, a short-circuit current calculation model for new energy units considering wind power and photovoltaic power is established. The negative sequence model adopts the impedance model.

[0087] ;

[0088] In the formula, It is a negative sequence impedance. and For rated capacity and rated voltage, This is the negative sequence impedance correction factor;

[0089] Based on the symmetrical component method, a short-circuit current calculation model for new energy units that take into account wind power and photovoltaics is established. The zero-sequence model is the impedance model determined by the neutral point grounding method of the unit transformer.

[0090] In this embodiment of the invention, the zero-sequence model is mainly determined by the connection method of the grid-connected transformer and the neutral point grounding condition. When a grounded connection method is adopted, the zero-sequence current can form a path through the grounded branch, affecting the zero-sequence short-circuit current of the system; if an ungrounded connection method is adopted, the zero-sequence component is approximately in an open-circuit state.

[0091] This invention employs a hybrid modeling method combining positive-sequence controlled current sources and negative-sequence / zero-sequence impedances. This model accurately reflects the dominant role of controlled source characteristics (positive sequence) of new energy units during faults and reasonably describes their response characteristics to negative-sequence and zero-sequence components, thus comprehensively and accurately calculating the actual impact of new energy grid connection on the system's short-circuit current level. Based on the mature symmetrical component method, this model provides clear and easily obtainable mathematical models for positive-sequence, negative-sequence, and zero-sequence components. In particular, it clarifies the key factor that the zero-sequence path depends on the transformer grounding method, providing a standardized and operable solution for large-scale short-circuit current calculation and analysis in power grids with a high proportion of new energy. This sequence model can effectively analyze the situation when the system experiences asymmetrical faults. By accurately calculating each sequence component, it provides a crucial data foundation for evaluating the coordination of protection devices and the safety and stability of the system under asymmetrical faults.

[0092] Specifically, in step S3, the construction of a multi-objective partitioned optimization model that considers voltage support, cascading failure risk, and system short-circuit characteristics includes:

[0093] Based on the variability of network topology, regional tie lines and key sections are selected as optimization objects, and the opening and closing decisions of the lines are used as optimization variables. The equivalent impedance of the system is adjusted by changing their conduction or disconnection states.

[0094] Based on the calculated voltage support capability, cascading fault risk, and short-circuit current level as core evaluation indicators, a multi-objective optimization function is constructed, the expression of which is:

[0095] ;

[0096] In the formula, X is the decision variable. Number of busbars It is the short-circuit current of the i-th bus when a fault occurs. This represents the average value of the multi-feed interaction factors. This is the improved multi-infeed short-circuit ratio.

[0097] Specifically, in step S3, the constraints of the multi-objective partitioning optimization model include power balance constraints, power flow constraints, and voltage constraints, and their expressions are as follows:

[0098] ;

[0099] In the formula, This represents the power output of the i-th generator. This represents the power demand of the j-th load. Indicates total transmission loss. and These are the active power and reactive power of the i-th bus, respectively. This represents the voltage of the i-th bus. and These represent the susceptance and phase angle difference between busbars i and j, respectively. , and Representing nodes respectively The actual voltage amplitude, minimum voltage amplitude, and maximum voltage amplitude.

[0100] In this embodiment of the invention, the decision variables of the model are the switching status of the network topology and the opening and closing decisions of critical tie lines, with values ​​of 0 (disconnected) or 1 (conductive), to reflect the operational and disconnected states of different lines. By adjusting the network connection relationship, the equivalent impedance and power flow distribution of the system can be changed, thereby affecting the voltage support strength and the coupling relationship between nodes. To quantify the impact of different partition structures on system stability, the model uses voltage support capability, cascading fault risk, and short-circuit current level as optimization objectives. Voltage support capability is measured by improving the multi-infeed short-circuit ratio, reflecting the voltage support level of the AC system for the DC converter station under different partitions. Cascading fault risk is indicated by the multi-infeed interaction factor, used to characterize the voltage coupling strength between DC channels. The short-circuit current level is constrained by the maximum short-circuit current of the system.

[0101] This invention integrates three key stability indicators—voltage support capability, cascading failure risk, and system short-circuit characteristics—into a unified multi-objective optimization model, providing a comprehensive and quantitative comparison benchmark and optimization direction for different partitioning schemes. By changing the decision variable of line switching status, it directly links network topology adjustment with operational characteristics such as system stability and security. This enables the optimization process to automatically find the Pareto optimal partitioning scheme that simultaneously improves voltage support strength, reduces cascading failure risk, and controls short-circuit current level, while meeting rigid safety constraints such as power balance, power flow, and voltage. It organically combines core indicators reflecting the overall stability level of the system with constraints that ensure local safety, ensuring that the final partitioning scheme can both improve the overall stability of the system macroscopically and meet all necessary engineering safety operation requirements microscopically.

[0102] Specifically, in step S4, the improvement of the particle swarm algorithm by introducing an adaptive inertia weight mechanism includes:

[0103] ;

[0104] in, and These are the upper and lower limits of the inertia weight, respectively. is the number of iterations, and is the maximum number of iterations.

[0105] Specifically, in step S4, the solution of the partitioning model using Pareto sort includes:

[0106] Based on the improved particle swarm optimization algorithm and Pareto non-dominated sorting, the multi-objective optimization model of power grid partitioning is solved to obtain a non-dominated solution set that satisfies multiple constraints.

[0107] Specifically, in step S4, obtaining the optimal power grid partitioning scheme includes:

[0108] Based on typical operating scenarios and simulation analysis, the obtained Pareto front solutions are verified and compared, and the scheme with the best comprehensive performance among multiple objective indicators is selected to determine the final power grid zoning scheme.

[0109] This invention employs an adaptive inertia weighting mechanism, enabling the algorithm to maintain strong global exploration capabilities in the early stages of iteration to broadly search the solution space, while enhancing local development capabilities in the later stages of iteration for refined convergence. This effectively balances the contradiction between exploration and development, overcoming the shortcomings of the basic particle swarm optimization algorithm, which is prone to premature convergence or getting trapped in local optima, thus ensuring a more reliable search for high-quality partitioning schemes. By combining Pareto non-dominated sorting, the algorithm can directly solve multi-objective optimization models, outputting a Pareto optimal solution set composed of numerous non-dominated solutions. This solution set intuitively demonstrates the competitive and non-simultaneous trade-offs among the three objectives of voltage support capability, cascading fault risk, and short-circuit current level, providing decision-makers with a rich set of candidate schemes with different emphases. Through simulation verification and comprehensive comparison of Pareto front solutions based on typical operating scenarios, this process can screen out schemes that are robust under various actual operating conditions and achieve the best balance among multiple key indicators. This ensures that the final selected power grid partitioning scheme is not only theoretically optimal but also possesses high engineering practical value, adaptability, and risk resistance.

[0110] Example: Taking a regional power grid as an example, the effectiveness of the proposed DC embedded power grid zoning method considering the short-circuit current of new energy sources is verified. This regional power grid contains 11 DC systems (H1-H11). Only the 1000kV and 500kV voltage levels of the AC system are considered. The system structure is as follows: Figure 2 As shown.

[0111] Based on the established multi-objective optimization model for power grid partitioning, an improved particle swarm optimization algorithm and Pareto non-dominated sorting are used to solve the problem, obtaining a Pareto front solution set that considers voltage support capability, cascading fault risk, and short-circuit current level. The solution set is analyzed in conjunction with typical operating scenarios, and representative partitioning scheme 1 is selected, with the topology results shown below. Figure 3 As shown in the diagram, the boundary between different filling backgrounds is the partition section. Without partitioning in the 500kV power grid, the maximum short-circuit current level of the lines in the system is 63.13kA. When partitioning scheme 1 is adopted, the maximum short-circuit current level decreases to 53.26kA.

[0112] To verify the effectiveness of the zoning scheme in reducing the risk of DC cascading faults, time-domain simulations were performed in PSD-BPA. A short-circuit fault lasting 0.1 seconds was applied near DC system H9. Figure 4 The waveforms of DC arc extinction angle, converter bus voltage, and DC active power transmission in the original system are shown. Figure 5 The waveform curves of the corresponding running variables after adopting the partitioning scheme 1 designed using the strategy of this invention.

[0113] Depend on Figure 4It can be seen that when the system does not adopt the partitioning scheme, after a short-circuit fault occurs near H9, the lowest voltage on the H7 converter bus is 0.364 pu. Eight DC lines immediately experience commutation failure after the initial fault, with a total commutation failure time of 0.36 seconds. Furthermore, H8 experiences continuous commutation failures during the recovery process. Regarding the active power transmitted by the DC system, from the occurrence of commutation failures to full recovery, the total power transmitted by the DC system is reduced by 2511.89 kW·h. After adopting the partitioning scheme 1 determined in this embodiment, as... Figure 5 As shown, the lowest voltage on converter bus H7 after the fault was 0.586 pu. The voltage recovery speed on each converter bus after the fault was cleared was faster than before any measures were taken. Furthermore, in the system with zoning measures, only 3 DC circuits experienced commutation failure faults, the duration of commutation failures was reduced by 0.23 seconds, and H8 did not experience any commutation failure faults during the recovery process. In terms of DC transmission active power, the reduction in transmitted power during the fault period was 935.13 kW·h.

[0114] Simulation analysis verified the effectiveness of the proposed DC embedded grid partitioning method, which takes into account the short-circuit current of new energy sources, in reducing the short-circuit current level and improving the safety and stability of the system for receiving-end grids with embedded DC.

[0115] Please see Figure 6 As shown, Figure 6 This is a schematic diagram of the DC embedded power grid partitioning system that takes into account the short-circuit current of new energy sources, as described in this invention.

[0116] Specifically, a system applied to the DC embedded grid partitioning method considering the short-circuit current of new energy sources includes:

[0117] The operation characteristic analysis module is used to improve the voltage support capability of the multi-infeed short-circuit ratio assessment system and analyze the risk of system cascading failures by combining the multi-infeed interaction factor.

[0118] The short-circuit current modeling module is used to construct a sequence impedance model that takes into account the characteristics of new energy sources by combining controlled current sources with impedance, so as to realize the quantitative assessment of the system's short-circuit current level.

[0119] The partition model construction module is connected to the operation characteristic analysis module and the short-circuit current modeling module respectively. It is used to construct a multi-objective partition optimization model that considers voltage support, cascading fault risk and system short-circuit characteristics with the line switching state as the optimization variable, so as to achieve coordinated optimization of power grid partitions under the conditions of power balance, power flow and voltage constraints.

[0120] The optimization solution module is connected to the operating characteristic analysis module, the short-circuit current modeling module, and the partition model construction module, respectively. It is used to introduce an adaptive inertia weight mechanism to improve the particle swarm algorithm, and combine Pareto sorting to solve the partition model to obtain the optimal power grid partition scheme.

[0121] The above embodiments are merely illustrative examples and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this application.

Claims

1. A DC embedded power grid partitioning method considering the short-circuit current of new energy sources, characterized in that, include: Step S1: Improve the evaluation of the system voltage support capability by multi-infeed short-circuit ratio, and analyze the risk of system cascading failures by combining multi-infeed interaction factors. Step S2: Using a combination of controlled current sources and impedance, a sequence impedance model considering the characteristics of new energy sources is constructed to evaluate the system short-circuit current level. Step S3: Using the line switching status as the optimization variable, construct a multi-objective zonal optimization model that considers voltage support, cascading fault risk and system short-circuit characteristics, and achieve coordinated optimization of power grid zonals under power balance, power flow and voltage constraints. Step S4: An adaptive inertia weighting mechanism is introduced to improve the particle swarm algorithm. Combined with Pareto sorting, the partitioning model is solved to obtain the optimal power grid partitioning scheme.

2. The DC embedded grid partitioning method considering the short-circuit current of new energy sources according to claim 1, characterized in that, In step S1, the improved multi-infeed short-circuit ratio evaluation system voltage support capability includes: Based on the multi-infeed short-circuit ratio and multi-port Thevenin equivalent method, a DC power weighting mechanism is introduced to improve the traditional multi-infeed short-circuit ratio, and the voltage support strength under various partition conditions is evaluated. Its expression is: ； In the formula, The improved multi-infeed short-circuit ratio, where n is the number of DC transmission systems. It is the active power of the i-th DC transmission system. The sum of active power, where i is the DC circuit number. This refers to the short-circuit capacity of the DC-fed converter bus. To account for the equivalent DC power after considering the influence of other DC circuits, This is the rated voltage of the converter bus. and The equivalent node self-impedance and mutual impedance are seen from each DC converter bus.

3. The DC embedded grid partitioning method considering the short-circuit current of new energy sources according to claim 1, characterized in that, In step S1, the combined multi-feed interaction factor analysis of system cascading failure risk includes: Based on the multi-feed interaction factor and nodal impedance method, considering the voltage coupling relationship between multi-circuit embedded DC systems, the risk of voltage-coupled cascading faults under various partition conditions is evaluated, and its expression is: ； In the formula, This refers to the multi-feed interaction factor between two DC systems. The voltage deviation on bus i. Let be the original voltage of bus j. Let be the mutual impedance between bus i and bus j. Let J be the self-impedance of bus j.

4. The DC embedded grid partitioning method considering the short-circuit current of new energy sources according to claim 1, characterized in that, In step S2, constructing the sequence impedance model that takes into account the characteristics of new energy sources includes: Based on the symmetric component method, a short-circuit current calculation model for new energy units considering wind power and photovoltaic power is established. The positive sequence model adopts a controlled current source model. ； In the formula, This represents the increase in output current before and after the fault. The voltage at the grid connection point. and The active and reactive currents during the fault period and The rated active and reactive power before the fault. and Rated capacity and current; Based on the symmetrical component method, a short-circuit current calculation model for new energy units considering wind power and photovoltaic power is established. The negative sequence model adopts the impedance model. ； In the formula, It is a negative sequence impedance. and For rated capacity and rated voltage, This is the negative sequence impedance correction factor; Based on the symmetrical component method, a short-circuit current calculation model for new energy units that take into account wind power and photovoltaics is established. The zero-sequence model is the impedance model determined by the neutral point grounding method of the unit transformer.

5. The DC embedded grid partitioning method considering the short-circuit current of new energy sources according to claim 1, characterized in that, In step S3, constructing a multi-objective partitioned optimization model that considers voltage support, cascading failure risk, and system short-circuit characteristics includes: Based on the variability of network topology, regional tie lines and key sections are selected as optimization objects, and the opening and closing decisions of the lines are used as optimization variables. The equivalent impedance of the system is adjusted by changing their conduction or disconnection states. Based on the calculated voltage support capability, cascading fault risk, and short-circuit current level as core evaluation indicators, a multi-objective optimization function is constructed, the expression of which is: ； In the formula, X is the decision variable. Number of busbars It is the short-circuit current of the i-th bus when a fault occurs. This represents the average value of the multi-feed interaction factors. This is the improved multi-infeed short-circuit ratio.

6. The DC embedded grid partitioning method considering the short-circuit current of new energy sources according to claim 5, characterized in that, In step S3, the constraints of the multi-objective partitioning optimization model include power balance constraints, power flow constraints, and voltage constraints, and their expressions are as follows: ； In the formula, This represents the power output of the i-th generator. This represents the power demand of the j-th load. Indicates total transmission loss. and These are the active power and reactive power of the i-th bus, respectively. This represents the voltage of the i-th bus. and These represent the susceptance and phase angle difference between busbars i and j, respectively. , and Representing nodes respectively The actual voltage amplitude, minimum voltage amplitude, and maximum voltage amplitude.

7. The DC embedded grid partitioning method considering the short-circuit current of new energy sources according to claim 1, characterized in that, In step S4, the improvement of the particle swarm algorithm by introducing an adaptive inertia weight mechanism includes: ； in, and These are the upper and lower limits of the inertia weight, respectively. is the number of iterations, and is the maximum number of iterations.

8. The DC embedded grid partitioning method considering the short-circuit current of new energy sources according to claim 1, characterized in that, In step S4, the solution of the partitioning model using Pareto sort includes: Based on the improved particle swarm optimization algorithm and Pareto non-dominated sorting, the multi-objective optimization model of power grid partitioning is solved to obtain a non-dominated solution set that satisfies multiple constraints.

9. The DC embedded grid partitioning method considering the short-circuit current of new energy sources according to claim 1, characterized in that, In step S4, obtaining the optimal power grid partitioning scheme includes: Based on typical operating scenarios and simulation analysis, the obtained Pareto front solutions are verified and compared, and the scheme with the best comprehensive performance among multiple objective indicators is selected to determine the final power grid zoning scheme.

10. A system applied to the DC embedded grid partitioning method considering the short-circuit current of new energy sources as described in any one of claims 1-9, characterized in that, include: The operation characteristic analysis module is used to improve the voltage support capability of the multi-infeed short-circuit ratio assessment system and analyze the risk of system cascading failures by combining the multi-infeed interaction factor. The short-circuit current modeling module is used to construct a sequence impedance model that takes into account the characteristics of new energy sources by combining controlled current sources with impedance, so as to realize the quantitative assessment of the system's short-circuit current level. The partition model construction module is connected to the operation characteristic analysis module and the short-circuit current modeling module respectively. It is used to construct a multi-objective partition optimization model that considers voltage support, cascading fault risk and system short-circuit characteristics with the line switching state as the optimization variable, so as to achieve coordinated optimization of power grid partitions under the conditions of power balance, power flow and voltage constraints. The optimization solution module is connected to the operating characteristic analysis module, the short-circuit current modeling module, and the partition model construction module, respectively. It is used to introduce an adaptive inertia weight mechanism to improve the particle swarm algorithm, and combine Pareto sorting to solve the partition model to obtain the optimal power grid partition scheme.

Images