Cutter sensitivity calculation method and system for transient stability control

Abstract

The invention provides a generator tripping sensitivity calculation method and system for transient stability control. The method comprises the steps that power angle trajectory data of an instability scene of a power system are combined with a composite power angle algorithm to carry out unit grouping; calculating a comprehensive difference value of the accumulated unbalanced energy and the decelerable area of the equivalent unit before and after generator tripping to form an estimated value of the unbalanced energy directly taken away by the generator tripping; calculating the electromagnetic power transfer amount of a cut unit by utilizing power flow information at the cutting moment and the network topology of the power system, and calculating the variable amount of a slowable area by combining with an equivalent unit power angle curve translation change hypothesis; finally, unbalance energy directly taken away by generator tripping and slowable area variation corresponding to power transfer are added to estimate stability margin variation, and the stability margin variation is divided by the capacity of the generator tripping unit to calculate generator tripping sensitivity. Based on the method, the invention further provides a generator tripping sensitivity calculation system for transient stability control. According to the method, an application prospect is provided for optimization decision-making of an online or real-time transient stability control generator tripping scheme.

Description

Technical Field

[0001] The present application relates to the technical field of novel power system emergency control, and in particular to a method and system for calculating generator trip sensitivity for transient stability control. Background Art

[0002] Severe disturbances in the power system can cause generators to lose synchronism, leading to power angle instability and, in turn, loss of transient stability. Generator shedding is a crucial control measure for maintaining transient stability during severe power system failures, due to its rapid implementation, significant effectiveness in preventing power angle instability, and minimal impact on user power supply. However, generator shedding inevitably results in active power loss and changes to the grid structure. Excessive generator shedding can lead to subsequent problems such as insufficient hot standby capacity and frequency drop. Therefore, optimizing the generator shedding strategy while minimizing the total generator shedding capacity while maintaining system stability is a key consideration in generator shedding control.

[0003] The key to optimizing a generator shedding scheme lies in quickly identifying the units most prone to instability and the most sensitive. By removing the units with the greatest impact on system stability, a multiplier effect can be achieved, minimizing the total generator shedding capacity. Currently, two main approaches are used to optimize generator shedding schemes: one is to first calculate the total generator shedding capacity required for the system based on the energy function method or EEAC theory, then incorporate engineering experience to characterize the generator shedding sensitivity based on indicators such as the electrical distance between the generator and the fault location, power angle offset, or acceleration trend. Generator shedding locations are then selected based on these indicators until the total generator shedding capacity meets the requirements. The other is to first construct a generator shedding control optimization decision model, then use the generator shedding sensitivity to quantify the impact of removing different generators on the power system stability margin, thereby linearizing the model. The optimal generator shedding location and capacity are then directly determined by solving the linear optimization decision model. The first approach uses a total-first-and-then-decomposition approach to calculate the total number of generators to be removed without simulation. This approach is relatively rapid and can be used for optimizing generator removal decisions for online or real-time transient stability control. However, this approach simplifies the total number of generators to be removed by only considering the change in equivalent system mechanical power. This approach fails to account for changes in parameters such as the equivalent electromagnetic power of the critical generator group after the generator is removed and the system inertia time constant, resulting in inaccurate total generator removal. Furthermore, the method lacks theoretical support for determining generator removal locations based on generator output, electrical distance, or power angle lead, making it difficult to obtain an optimal generator removal solution. The second approach is more theoretically comprehensive, simultaneously determining both the number and location of generators to be removed. However, obtaining the generator removal sensitivity is more difficult, and determining this sensitivity is crucial for its application.

[0004] To obtain the generator trip sensitivity, one approach is to use ultra-real-time simulation to obtain the state trajectories of various system electrical variables. This approach then combines stability margin quantification theory to calculate the stability margin change and calculate the sensitivity using numerical difference methods. However, this approach, which relies on ultra-real-time simulation to calculate sensitivity, places high demands on simulation performance and struggles to balance model detail with rapid sensitivity calculation. For online stability control, optimized generator trip plans must be generated for a large number of potential faults in several typical scenarios within minutes (typically 15 minutes), resulting in high hardware requirements and limited model refinement. For real-time stability control, the optimized generator trip plan must be generated for a single fault in a single operating scenario within hundreds of milliseconds (typically 100 milliseconds), also posing computational resource and accuracy challenges. Another approach involves generating a large number of generator trip sensitivity samples for typical scenarios and faults, embedding the data patterns into a neural network through offline training, and then, during online or real-time stability control, using online mapping to obtain the generator trip sensitivity for a given scenario and fault within 1 millisecond. However, this neural network-based sensitivity acquisition approach suffers from generalization difficulties. Changes in operating mode or instability mode can lead to increased sensitivity calculation errors, thus compromising the generator trip control effectiveness. Therefore, how to quickly and accurately calculate the sensitivity of the machine cutting at different positions and then support the subsequent optimization of the machine cutting plan has become an important technical point of transient stability control. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a method and system for calculating the tripping sensitivity for transient stability control, which can accurately and quickly calculate the stability margin change of different tripping schemes based on the real-time status information of the power grid, and further divide the stability margin change by the tripping amount of the tripped unit to calculate the tripping sensitivity.

[0006] In a first aspect, the present invention provides a method for calculating a trip sensitivity for transient stability control, comprising the following steps: Obtaining unstable scenario trajectory data according to the stable control method of the unstable scenario; calculating the composite power angle difference using the unstable scenario trajectory data, and dividing the units into leading units and lagging units based on the composite power angle difference; Using the system trajectory data from the time of fault to the time of generator shedding as input, the extended equal-area criterion is used to determine the electromagnetic power angle curve and the mechanical power angle curve. Using these electromagnetic power angle curves and the mechanical power angle curves, the acceleration area, deceleration area, and remaining deceleration area of ​​the equivalent units before and after generator shedding are calculated based on the grouping results. The unbalanced energy mutation caused by generator shedding is then calculated. Taking the power flow information at the time of generator shedding and the power system network topology as input, the electromagnetic power transfer of the shedding unit is calculated. The change in the retarded area is calculated by assuming that the generator shedding causes a shift in the power angle curve of the equivalent unit. The stability margin change is calculated by combining the sudden change in the unbalanced energy of the cutting unit with the change in the decelerable area corresponding to the power transfer, and the cutting sensitivity is calculated by dividing it by the cutting amount of the cut-off unit.

[0007] In a second aspect, the present invention provides a machine-cut sensitivity calculation system for transient stability control, comprising: a grouping module, a first calculation module, a second calculation module, and a third calculation module; The grouping module is used to obtain unstable scenario trajectory data according to the stable control method of the unstable scenario; calculate the composite power angle difference using the unstable scenario trajectory data, and divide the units into leading units and lagging units based on the composite power angle difference; The first calculation module is configured to use the system trajectory data from the time of fault to the time of generator shedding as input, and determine the electromagnetic power angle curve and the mechanical power angle curve using the extended equal area criterion; use the electromagnetic power angle curve and the mechanical power angle curve to calculate the acceleration area, deceleration area, and remaining deceleration area of ​​the equivalent units before and after generator shedding based on the grouping results; and then calculate the unbalanced energy mutation amount caused by generator shedding; The second calculation module is used to calculate the electromagnetic power transfer of the disconnected unit based on the power flow information at the time of disconnection and the power system network topology as input, and calculate the change in the deceleration area by assuming that the disconnection causes the equivalent unit power angle curve to shift; The third calculation module is used to calculate the stability margin change amount by combining the sudden change amount of the unbalanced energy of the switching generator and the change amount of the decelerable area corresponding to the power transfer, and divide the result by the switching amount of the switched generator to calculate the switching sensitivity.

[0008] The effects provided in the summary of the invention are only the effects of the embodiments, not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects: The present invention provides a method and system for calculating generator shedding sensitivity for transient stability control. The method comprises the following steps: obtaining instability scenario trajectory data according to a stable control mode of the instability scenario; calculating a composite power angle difference using the instability scenario trajectory data, and dividing the generators into leading generator groups and lagging generator groups based on the composite power angle difference; using system trajectory data from the time of a fault to the generator shedding time as input, determining an electromagnetic power angle curve and a mechanical power angle curve using an extended equal area criterion; using the electromagnetic power angle curve and the mechanical power angle curve, calculating the acceleration area, decelerated area, and remaining decelerable area of ​​equivalent generators before and after the generator shedding according to the grouping results; then calculating the unbalanced energy mutation amount caused by the generator shedding; using power flow information at the generator shedding time and the power system network topology as input, calculating the electromagnetic power transfer amount of the generator shedding, and calculating the decelerable area change assuming that the generator shedding causes a translational change in the power angle curve of the equivalent generator shedding; calculating the stability margin change amount by dividing the unbalanced energy mutation amount of the generator shedding with the decelerable area change corresponding to the power transfer, and calculating the generator shedding sensitivity by dividing the amount of the generator shedding by the generator shedding amount. Based on a method for calculating the sensitivity of generator shedding for transient stability control, a generator shedding sensitivity calculation system for transient stability control is also proposed. This method utilizes the power system power angle, speed, electromagnetic and mechanical power, unit terminal voltage, and active and reactive output power information at and before the generator shedding control moment in a basic instability scenario to calculate the generator shedding sensitivity of each removable unit through algebraic operations. This method does not require the time series trajectory of the electrical quantities in the power system after the generator shedding, resulting in a time-saving and simple calculation method. It can provide basic information for sensitivity-based control sensitive point screening and optimal generator shedding scheme generation, and has certain application prospects in optimizing generator shedding scheme decisions for online or real-time transient stability control. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 This is a flow chart of a method for rapidly calculating the trip sensitivity for transient stability control proposed in Example 1 of the present invention; Figure 2 This is a flow chart of generator grouping based on the composite power angle algorithm proposed in Example 1 of the present invention; Figure 3 This is a relationship diagram of various time nodes proposed in Example 1 of the present invention; Figure 4 This is a flow chart for calculating the sudden change in unbalanced energy of the cutting machine based on EEAC proposed in Example 1 of the present invention; Figure 5 The equivalent power angle curves of different stable control modes proposed in Example 1 of the present invention; Figure 6 This is the equivalent power angle characteristic curve before and after removing the cut-off unit proposed in Example 1 of the present invention; Figure 7 This is a flow chart for calculating the change in the decelerable area considering the continuous effect of power transfer proposed in Example 1 of the present invention; Figure 8 This is a flow chart of calculating the switching sensitivity based on the stability margin variation proposed in Example 1 of the present invention; Figure 9 This is a structural diagram of a power system; Figure 10 This is the equivalent power angle curve diagram of the 109 unit; Figure 11 This is a schematic diagram of a system for rapidly calculating trip sensitivity for transient stability control proposed in Example 2 of the present invention. DETAILED DESCRIPTION

[0010] Example 1 Embodiment 1 of the present invention proposes a method for quickly calculating the tripping sensitivity for transient stability control, which is used to solve the technical problem that the tripping sensitivity information used in the transient stability control of the power system depends on the electromechanical transient simulation trajectory, resulting in a long time consumption and difficulty in meeting online control requirements.

[0011] The implementation process of the present invention includes: first, using the power angle trajectory data of the basic instability scenario of the power system in combination with the composite power angle algorithm to group the units; then, based on the Extended Equal Area Criterion (EEAC), the cumulative unbalanced energy and the comprehensive difference of the retarded area of ​​the equivalent units before and after the generator shedding are calculated to form an estimated value of the unbalanced energy directly removed by the generator shedding; further, using the power flow information at the time of generator shedding and the power system network topology to calculate the electromagnetic power transfer of the cut units, and combining the EEAC equivalent unit power angle curve translation change hypothesis to calculate the change in the retarded area; finally, the unbalanced energy directly removed by the generator shedding is added to the change in the retarded area corresponding to the power transfer to estimate the change in the stability margin, and the generator shedding sensitivity is calculated by dividing it by the capacity of the cut units.

[0012] Figure 1 This is a flow chart of a method for rapidly calculating the trip sensitivity for transient stability control proposed in Example 1 of the present invention; In step 1, the unstable scenario trajectory data is obtained according to the stable control method of the unstable scenario; the composite power angle difference is calculated using the unstable scenario trajectory data, and the units are divided into leading units and lagging units based on the composite power angle difference; In the basic instability scenario, firstly, according to different stability control methods, the generator inertia parameters and generator power angle, angular velocity (the difference between the actual angular velocity per unit value and the synchronous angular velocity per unit value), electromagnetic power, and mechanical power trajectory data are obtained. Then, based on the trajectory data, the composite power angle index is used to group the units. Figure 2 This is a flow chart of generator grouping based on the composite power angle algorithm proposed in Example 1 of the present invention.

[0013] The process of obtaining the trajectory data of the unstable scenario according to the stable control method of the unstable scenario includes: If the stabilization control mode is online stabilization control, the generator inertia parameter data in the simulation file is read to form the generator inertia vector; the generator inertia vector is expressed as: ; (1) in, is the moment of inertia of all operating generators, For the The moment of inertia of the generator; If the stabilization control mode is real-time stabilization control, the inertia parameter data of the operating generator is obtained through acquisition devices such as PMU and WAMS to form the generator inertia vector: ; (2) If the stability control mode is online stability control, perform electromechanical transient simulation on the instability scenario and record the time of fault occurrence with a step size of 0.01s. Until the system reaches the point of no return The state trajectory data of each generator is used to form the state trajectory of each generator; the state trajectory of each generator is expressed as: ; (3) in, For the Generator power angle; For the Generator speed; For the Electromagnetic power of generators; For the Mechanical power state trajectory of each generator; , is the number of generators; The time when the fault occurs; is the moment when the system reaches the point of no return.

[0014] set up is the fault clearing moment, To control the delay, To control the cutting time, Figure 3This is a relationship diagram of various time nodes proposed in Example 1 of the present invention; If the stability control mode is real-time stability control, the PMU and WAMS devices will be started to collect the generator status trajectory after the fault is detected. The sampling frequency is 100Hz, and the time from the fault occurrence to the generator trip control moment will be recorded. The data between is stored in the state trajectory vector, which is expressed in the same way as formula (3), but the data length is shorter.

[0015] Using the compound power angle algorithm to divide the generators into leading groups and lagging fleets .

[0016] If the stabilization control mode is online stabilization control, the composite power angle of each generator is expressed as: ; (4) in, For the The composite power angle of the generator; is the power angle at the time when the system does not return; is the speed at the point of no return of the system; is the angular velocity of the center of inertia of the system; is the tuning coefficient, which is a positive real number. ; ; (5) If the stabilization control mode is real-time stabilization control, the composite power angle of each generator is expressed as: ; (6) in, The power angle at the system switching control moment; The speed of the system at the moment of cutting machine control; is the angular velocity of the center of inertia of the system; ; (7) Press all generators Sort from large to small to form an ordered sequence , calculate the composite power angle difference between adjacent generators in sequence: ; (8) By difference The largest position , as the boundary of the generator group; the first Leading Group , the remaining generator groups are divided into lagging groups , among which the advanced cluster It is also a group of cutting machine solutions.

[0017] In step 2, the system trajectory data from the time of fault to the time of generator shedding is used as input, and the extended equal-area criterion is used to determine the electromagnetic power angle curve and the mechanical power angle curve. The acceleration area, deceleration area, and remaining deceleration area of ​​the equivalent units before and after generator shedding are calculated based on the clustering results using these electromagnetic power angle curves and mechanical power angle curves. The unbalanced energy mutation caused by generator shedding is then calculated.

[0018] The EEAC theory is used to calculate the acceleration area, deceleration area and remaining deceleration area of ​​the equivalent system in two cases: whether the cut-off unit is eliminated or not. The difference between the two cases is compared to calculate the unbalanced energy mutation caused by the cutting of the unit. Figure 4 This is a flow chart for calculating the sudden change in unbalanced energy of the cutting machine based on EEAC proposed in Example 1 of the present invention.

[0019] Taking the grouping results and the trajectory data of each generator as input, the equivalent system power angle, speed, electromagnetic power and mechanical power trajectory are calculated according to the EEAC theory to form the electromagnetic power-power angle curve and the mechanical power-power angle curve.

[0020] If the stabilization control mode is online stabilization control, the equivalent process includes: using to The trajectory data at each moment is used to calculate the power angle, angular velocity, electromagnetic power and mechanical power trajectory data of the inertia center; ; (9) in, for Group equivalent moment of inertia; for Group equivalent moment of inertia; for The equivalent power angle of the group; for The speed of the group, for The electromagnetic power of the group; for Mechanical power trajectory data of the group; for The equivalent power angle of the group; for The rotation speed of the group; for The electromagnetic power of the group; for Mechanical power trajectory data of the group; The power angle, angular velocity, electromagnetic power and mechanical power trajectory data of the equivalent system from the time the fault occurs to the point where the system does not return are calculated as follows: ; (10) in, is the inertia of the equivalent system; is the power angle of the equivalent system; is the rotational speed of the equivalent system; is the mechanical power of the equivalent system; is the electromagnetic power trajectory data of the equivalent system; The equivalent system state trajectory is expressed as: ; (11) If the stabilization control mode is real-time stabilization control, the equivalent process includes: to Substitute the trajectory data at the moment into equations (9) and (10) to calculate the power angle, angular velocity, electromagnetic power and mechanical power trajectory data of the equivalent system from the time of fault occurrence to the time of generator switching control; Model the relationship between electromagnetic power, mechanical power, and power angle: ; (12) ; (13) in, is the first parameter to be fitted; is the second parameter to be fitted; is the third parameter to be fitted; is the fourth parameter to be fitted; is the fifth parameter to be fitted; is the sixth parameter to be fitted; is the equivalent system power angle, is the electromagnetic power angle curve of the equivalent system, is the mechanical power-angle curve of the equivalent system; by 、 and , To map the sequence, the nonlinear and least squares parameter identification methods are used to calculate the parameters to be identified. ; The power angle at the time of cutting control As a starting point, For the end, Generate extrapolated power angle prediction sequence with equal intervals of 0.01rad in the interval , substitute the prediction sequence into the established power angle curve equation to obtain the corresponding electromagnetic power and mechanical power prediction sequence , ; Search from front to back to meet The sequence elements form the electromagnetic power prediction sequence corresponding to the period from the moment of machine cutting control to the moment of the system no return point. and mechanical power prediction series , the corresponding power angle prediction sequence is ; The equivalent system state quantity trajectory data and the state quantity prediction sequence data are stored in the state quantity trajectory vector to obtain the equivalent system state trajectory in the same form as the previous formula (11). Figure 5 These are equivalent power angle curves of different stable control modes proposed in Example 1 of the present invention; the equivalent power angle curve of the online stable control mode is shown in FIG5(a), and the equivalent power angle curve of the real-time stable control mode is shown in FIG5(b).

[0021] Using the electromagnetic power angle curve and the mechanical power angle curve, the acceleration area, deceleration area, and remaining deceleration area of ​​the equivalent units before and after the generator shedding are calculated based on the grouping results. The process of calculating the unbalanced energy mutation caused by the generator shedding includes the following steps: Based on the trajectory data during the fault period, the rectangular method is used to calculate the acceleration area during the fault period, with the power angle as the independent variable and the mechanical power and electromagnetic power as the dependent variables; ; (14) in, is the acceleration area of ​​the equivalent system during the fault period, i.e., the accumulated unbalanced energy during the fault period; Based on the trajectory data after fault removal, the rectangular method is used to calculate the deceleration area from the time of fault removal to the time of generator control, with power angle as the independent variable and mechanical power and electromagnetic power as the dependent variables. ; (15) in, It is the deceleration area of ​​the equivalent system from the time when the fault is cleared to the time when the generator is cut off, that is, the calculation of the unbalanced energy from the time when the fault is cleared to the time when the generator is cut off; The accumulated unbalanced energy is calculated by subtracting the calculated acceleration area from the deceleration area after the fault is removed: ; (16) Where, It is the accumulated unbalanced energy from the occurrence of system failure to the moment of generator trip control; Based on the trajectory data after the machine shedding control, the power angle is the independent variable, and the mechanical power and electromagnetic power are the dependent variables. The rectangular method is used to calculate the deceleration area from the machine shedding control to the point of no return: ; (17) in, It is the deceleration area of ​​the equivalent system from the moment of cutting machine control to the point of no return; From the leading fleet Choose any unit Perform removal to form the remaining leading group , get the remaining leading group With lagging cluster The clustering result. Taking the basic instability scene trajectory data as input, executing the above step 2 operation, we can get: According to the new advanced fleet and lagging fleets , perform equivalent calculation operations to calculate the power angle of the equivalent system after removing the unit , speed , mechanical power , electromagnetic power The trajectory data of the generator set is shown in Figure 6. Figure 6 (a) is the equivalent power angle characteristic curve before the unit is eliminated; Figure 6 (b) is the equivalent power angle characteristic curve after removing the unit.

[0022] Calculation of accumulated unbalanced energy during fault period: ; (18) Where, is the acceleration area of ​​the equivalent system during the fault period, i.e., the accumulated unbalanced energy during the fault period Calculation of unbalanced energy from fault removal to generator trip control: ; (19) Where, It is the deceleration area of ​​the equivalent system from the time when the fault is cleared to the time when the generator is cut off, that is, the calculation of the unbalanced energy from the time when the fault is cleared to the time when the generator is cut off; Calculation of accumulated unbalanced energy during machine switching control: ; (20) Where, It is the accumulated unbalanced energy from the occurrence of system failure to the moment of generator trip control; Calculation of deceleration area: ;(twenty one) Where, It is the deceleration area of ​​the equivalent system from the moment of machine cutting control to the point of no return; Calculate the change in the difference between the accumulated unbalanced energy and the deceleration area before and after the machine is removed, which is the sudden change in the unbalanced energy of the machine being cut: ;(twenty two) Where, is the unbalanced energy mutation of the cutting machine Check whether all generators in the leading group have been evaluated. If so, proceed to step 3. Otherwise, recalculate the accumulated unbalanced energy and deceleration area after removing the cut-off generators.

[0023] In step 3, the power flow information at the time of generator shedding and the power system network topology are used as input to calculate the electromagnetic power transfer of the shedding unit. The change in the deceleration area is calculated by assuming that the generator shedding causes the equivalent unit power angle curve to shift. Figure 7 This is a flow chart for calculating the change in the decelerable area considering the continuous effect of power transfer proposed in Example 1 of the present invention; Taking the power flow information at the time of generator shedding and the power system network topology as input, the transient internal potential of the generator and the phase angle under the synchronous coordinate are calculated.

[0024] Taking the power flow information at the time of generator shedding and the power system network topology as input, the transient internal potential of the generator and the phase angle under the synchronous coordinate are calculated; the nodes within the generator are expanded and the non-generator nodes are eliminated to form a simplified admittance matrix; A unit is randomly selected from the leading group for removal, and the change in electromagnetic power of each generator and the change in electromagnetic power of the equivalent system are calculated. Assuming that the machine removal causes the overall shift of the EEAC equivalent unit power angle curve, the change in the retarded area is calculated.

[0025] Based on the power flow information at the time of power outage and the power system network topology data, including the generator Active power and reactive power ,dynamo Amplitude of terminal voltage and phase angle ,dynamo Subtransient reactance , calculate the transient internal potential of the generator Phase angle with synchronous coordinates : ;(twenty three) ;(twenty four) Expand the nodes in the generator and eliminate the non-generator nodes to form a simplified admittance matrix. Based on the power system network topology, establish the original node admittance matrix of the power system generator ; ; (25) Calculate the equivalent admittance parameters of the generator: ; (26) in, For the Equivalent admittance parameters of the generator; For the Equivalent resistance of a generator; For the Equivalent reactance of each generator; The generator internal impedance equivalent admittance is merged into the original node admittance matrix Y. The correction method is to select any generator from the generator set, and its original node number is , the newly added internal node number is , expand the admittance matrix: ; (27) in, is the original generator node self-admittance; is the newly added generator node self-admittance; is the mutual admittance between the original generator node and the newly added generator node; is the mutual admittance between the newly added generator node and the original generator node; Check whether the impedance equivalent admittances of all generator sets are merged into the original node admittance matrix. If so, proceed to the next step. If not, select another generator set to execute the process. Choose any load , model the load as a parallel admittance and calculate the parallel admittance: ; (28) in, is the parallel admittance; To control the active load at the previous moment; To control the reactive load at the previous moment; is the node bus voltage amplitude; Merge the load equivalent admittance into the original node admittance matrix ; ; (29) in, is the original load node self-admittance; Check whether all load equivalent admittances are merged into the original node admittance matrix. If so, proceed to the next step. If not, select another load to execute the process.

[0026] The modified node admittance matrix is ​​rearranged in the following way: the non-generator internal nodes are placed at the front of the admittance matrix, and the generator internal nodes are placed at the back. The rearranged admittance matrix is: ; (30) in, represents the rearranged admittance matrix; represents the self-admittance submatrix between non-generator nodes, represents the node admittance matrix within the generator, represents the mutual admittance between non-generator internal nodes and generator internal nodes; represents the mutual admittance between the nodes within the generator and the nodes outside the generator; Perform the Kron reduction operation to eliminate non-generator internal nodes and obtain a simplified equivalent admittance matrix containing only generator internal nodes: ; (31) in, To simplify the equivalent admittance matrix; Separate the simplified equivalent admittance matrix according to the real and imaginary parts of the admittance to form a simplified equivalent conductance matrix and the simplified equivalent susceptance matrix .

[0027] Select any unit from the leading group The process of performing the removal and calculating the electromagnetic power variation of each generator and the electromagnetic power variation of the equivalent system includes: Calculate the change in electromagnetic power of the remaining units caused by generator shedding, specifically: ; (32) ; in, For generator nodes Node with the removed generator set Synchronous torque coefficient between; For generator nodes The internal potential of is the generator terminal voltage before the cut-off unit is cut off. To simplify the equivalent susceptance matrix OK column values, For generators Voltage phase angle difference before disturbance, Node of the removed generator set Active power; Substitute equation (32) into equation (10) to calculate the cut-off unit The change of electromagnetic power of the equivalent system is: ; (33) in, To control the cutting time ahead of the first The change in electromagnetic power output of each generator; The first unit in the group with the delayed switching control time The change in electromagnetic power output of each generator; For the generator set to be removed Inertia; . is the sum of the inertia of the leading group and the inertia of the lagging group; Assuming that the generator shedding causes the EEAC equivalent unit power angle curve to shift as a whole, the process of calculating the change in the retarded area includes: If the stable control mode is online stable control, the method for determining the equivalent system power angle curve after the machine is cut off is: the equivalent system electromagnetic power angle curve after the machine is cut off is Pan Up , and obtain the equivalent system electromagnetic power angle curve after translation : ; (34) in, It is the equivalent system power angle after removing the cut-off generator set; Mechanical power is the equivalent system mechanical power angle curve after the machine is cut off. ; If the stable control mode is real-time stable control, the method for determining the equivalent system power angle curve after the machine is cut off is: the equivalent system electromagnetic power angle curve after the machine is cut off is Pan Up , and obtain the equivalent system electromagnetic power angle curve after translation : ; (35) Mechanical power is the equivalent system mechanical power angle curve after the machine is cut off. : ; (36) The power angle at the time of cutting control As a starting point, For the end, Generate extrapolated power angle sequence with equal intervals of 0.01rad in the interval , substitute the power angle sequence into the established power angle curve, and search from front to back for the first one that satisfies The power angle value is the power angle value at the point of no return of the power angle curve of the equivalent system after translation. ; Use the trapezoidal calculation formula to calculate the change in the deceleration area after the cutting machine control: ; (37) in, To eliminate the power angle at the system's non-return point after the disconnected generator set is removed; Check whether all units in the leading group have been evaluated. If so, go to step 4. Otherwise, recalculate the electromagnetic power transfer amount of the disconnected units.

[0028] In step 4, the stability margin change is calculated by combining the sudden change in the unbalanced energy of the cutting generator with the change in the decelerable area corresponding to the power transfer, and the cutting sensitivity is calculated by dividing it by the cutting amount of the cut generator.

[0029] The stability margin change is calculated by combining the sudden change in the unbalanced energy of the generator shedding with the change in the decelerable area corresponding to the power transfer. This change is then divided by the shedding amount of the shedding unit to calculate the shedding sensitivity. The specific calculation process is shown in Figure 8. Figure 8 This is a flow chart of calculating the switching sensitivity based on the stability margin variation proposed in Example 1 of the present invention; Based on the unbalanced energy mutation and the change in the deceleration area of ​​all units in the leading group calculated in step 2 and step 3, add the two together to obtain the change in stability margin: ; (38) Divide the change in stability margin corresponding to each cut-off unit by the cut-off amount of the cut-off unit to calculate the cut-off sensitivity of all units in the leading group: ; (39) The present invention proposes a method for rapid calculation of the tripping sensitivity for transient stability control, which is suitable for rapid sensitivity calculation when generating a strategy table for online stability control or a tripping plan for real-time stability control. The rapid calculation of the tripping sensitivity is achieved through the coordination of four parts: grouping of generator sets based on composite power angle, calculation of the sudden change in unbalanced energy due to tripping based on EEAC, calculation of the change in the decelerable area considering the continuous impact of power transfer, and calculation of the tripping sensitivity based on the change in stability margin. In step 1, real-time status information from power system measurement equipment such as PMUs / WAMS or simulation data of basic instability scenarios is used as input. A composite power angle algorithm is used to group the units based on the power angle trajectory and angular velocity deviation of each unit, and the set of units for which the generator trip sensitivity calculation is to be performed is determined. In step 2, the system trajectory data from the time of the fault to the time of generator trip is used as input. Based on the grouping results, the difference between the accumulated unbalanced energy and the deceleration area of ​​the equivalent units before and after the generator trip is calculated based on the grouping results, forming the energy directly removed by the generator trip. In step 3, the power flow information at the time of generator trip and the power system network topology are used as input to calculate the electromagnetic power transfer of the switched units. The change in the deceleration area is calculated assuming that the generator trip causes a shift in the power angle curve of the equivalent units. In step 4, the stability margin change is calculated by combining the results of steps 2 and 3, and divided by the trip amount of the switched units to obtain the generator trip sensitivity.

[0030] A method for quickly calculating the tripping sensitivity for transient stability control proposed in Example 1 of the present invention can accurately and quickly calculate the stability margin change of different tripping schemes based on the real-time status information of the power grid, and further divide the stability margin change by the tripping amount of the tripped unit to calculate the tripping sensitivity.

[0031] In order to verify the technical effects achieved by the present invention, a power supply system is taken as an example to illustrate the effectiveness of Example 1 of the present invention. Figure 9This is a diagram of a power system structure; it includes 11 1000 kV nodes forming the AC main grid, 122 500 kV nodes, 199 AC lines, and 6 DC lines, forming a complex power network. A three-phase short circuit fault is set at 1 second on the first section of lines 178-152 and 124-110. After the fault lasts for 0.15 seconds, the relay protection trips the faulty line. 0.1 second after the tripping occurs, the generator is tripped.

[0032] First, according to step 1, the generator inertia parameter data and the state trajectory data from the fault occurrence to the generator tripping control moment are recorded to form a state trajectory vector with a data time step of 0.01s.

[0033] The power angle and speed at the time of system trip control are read from the generator trajectory data. The composite power angle of each generator is calculated and sorted from largest to smallest. The composite power angle differences between adjacent generators are calculated in sequence. The position with the largest difference is the boundary of the generator group, and the generator groups are divided into leading and lagging groups. Based on the generator grouping results and the state trajectory vector, the equivalent calculation is performed according to formula (9)-formula (10) to obtain the equivalent system state trajectory from the fault occurrence to the generator trip control moment; Based on the equivalent system state trajectory during the fault period, the system acceleration area is calculated to be 11.37; According to the equivalent system state trajectory after fault removal, the decelerated area is calculated to be -8.66; Subtracting the decelerated area from the accelerated area yields a cumulative unbalanced energy of 20.03; Based on the equivalent system state trajectory after fault removal, the system power angle curve is obtained by curve fitting, and the equivalent system power angle characteristic prediction curve of the system no-return point is extrapolated; Based on the prediction results of electromagnetic power and mechanical power based on the power angle curve, the deceleration area is calculated to be 4.09; After removing the G109, G149, G179, G190, G192, and G219 related variables, the generator trajectory data was re-grouped, revalued, and curve-fitted and predicted. The cumulative unbalanced energy and retarded area after removing different units were calculated. The change in the cumulative unbalanced energy and the change in the retarded area were added to obtain the sudden change in the unbalanced energy after the generator was cut off. Simplify the system admittance matrix and calculate the electromagnetic power change of each generator at the time of power off, and further calculate the equivalent electromagnetic power change; The cutting of the generator causes the equivalent power angle curve to shift, and the change in the decelerable area is calculated. Taking the cutting of G109 as an example, the equivalent power angle curve is shown in Figure 10. Figure 10 This is the equivalent power angle curve diagram of the 109 unit; The change in stability margin is obtained by adding the sudden change in unbalanced energy of the cutting unit to the change in the decelerable area, and then divided by the capacity of the cut-off unit to obtain the cutting sensitivity. The calculation results are shown in Table 1.

[0034] Table 1: Calculation results of cutting machine sensitivity

[0035] According to the analysis of the sensitivity calculation results, some machine cutting schemes have large errors, but the calculated sensitivity ranking is basically consistent with the actual sensitivity ranking, and the control effects of different machine cutting control measures can still be compared. Among them, the average error between the calculated sensitivity value and the actual value of the node with higher sensitivity is about 10%, and the error is controlled within the allowable range, indicating that the sensitivity calculation of highly sensitive nodes is accurate, verifying the effectiveness of the fast calculation method of machine cutting sensitivity for transient stability control proposed in Example 1 of the present invention.

[0036] Example 2 Based on the method for calculating the trip sensitivity for transient stability control proposed in Example 1 of the present invention, Example 2 of the present invention further proposes a system for calculating the trip sensitivity for transient stability control. Figure 11 This is a schematic diagram of a system for rapidly calculating the trip sensitivity for transient stability control proposed in Example 2 of the present invention, the system comprising: a grouping module, a first calculation module, a second calculation module, and a third calculation module; The grouping module is used to obtain unstable scenario trajectory data according to the stable control method of the unstable scenario; the composite power angle difference is calculated using the unstable scenario trajectory data, and the units are divided into leading units and lagging units based on the composite power angle difference; The first calculation module is configured to use the system trajectory data from the time of fault to the time of generator shedding as input, and determine the electromagnetic power angle curve and the mechanical power angle curve using the extended equal area criterion; use the electromagnetic power angle curve and the mechanical power angle curve to calculate the acceleration area, deceleration area, and remaining deceleration area of ​​the equivalent units before and after generator shedding based on the grouping results; and then calculate the unbalanced energy mutation amount caused by generator shedding; The second calculation module is used to calculate the electromagnetic power transfer of the disconnected unit based on the power flow information at the time of disconnection and the power system network topology as input, and calculate the change in the deceleration area by assuming that the disconnection causes the equivalent unit power angle curve to shift; The third calculation module is used to calculate the stability margin change by combining the sudden change in the unbalanced energy of the cutting generator with the change in the decelerable area corresponding to the power transfer, and divide the result by the cutting amount of the cut generator to calculate the cutting sensitivity.

[0037] A rapid calculation system for generator tripping sensitivity for transient stability control proposed in Example 2 of the present invention can accurately and rapidly calculate the stability margin change of different generator tripping schemes based on real-time grid status information, and further divide the stability margin change by the generator tripping amount of the tripped unit to calculate the generator tripping sensitivity.

[0038] The description of the relevant parts of the machine-cutting sensitivity calculation system for transient stability control provided in Example 2 of the present application can be found in the detailed description of the corresponding parts of the machine-cutting sensitivity calculation method for transient stability control provided in Example 1 of the present application, and will not be repeated here. The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A method for calculating the trip sensitivity for transient stability control, characterized in that: The following steps are involved: Obtaining unstable scenario trajectory data according to the stable control method of the unstable scenario; calculating the composite power angle difference using the unstable scenario trajectory data, and dividing the units into leading units and lagging units based on the composite power angle difference; Using the system trajectory data from the time of fault to the time of generator shedding as input, the extended equal-area criterion is used to determine the electromagnetic power angle curve and the mechanical power angle curve. Using these electromagnetic power angle curves and the mechanical power angle curves, the acceleration area, deceleration area, and remaining deceleration area of ​​the equivalent units before and after generator shedding are calculated based on the grouping results. The unbalanced energy mutation caused by generator shedding is then calculated. Taking the power flow information at the time of generator shedding and the power system network topology as input, the electromagnetic power transfer of the shedding unit is calculated. The change in the retarded area is calculated by assuming that the generator shedding causes a shift in the power angle curve of the equivalent unit. The stability margin change is calculated by combining the sudden change in the unbalanced energy of the cutting unit with the change in the decelerable area corresponding to the power transfer, and the cutting sensitivity is calculated by dividing it by the cutting amount of the cut-off unit.

2. The method according to claim 1, wherein The process of obtaining the trajectory data of the unstable scenario according to the stable control method of the unstable scenario includes: If the stabilization control mode is online stabilization control, the generator inertia parameter data in the simulation file is read to form the generator inertia vector; if the stabilization control mode is real-time stabilization control, the inertia parameter data of the operating generator is obtained through the acquisition device to form the generator inertia vector: the generator inertia vector is expressed as: ； in, is the moment of inertia of all operating generators, For the The moment of inertia of the generator; If the stable control mode is online stable control, the state trajectory data of each generator is obtained by performing electromechanical transient simulation on the unstable scene; if the stable control mode is real-time stable control, the state trajectory data of each generator is collected by the acquisition device to form the state trajectory of each generator; the state trajectory of each generator is expressed as: ； in, For the Generator power angle; For the Generator speed; For the Electromagnetic power of generators; For the Mechanical power state trajectory of each generator; , is the number of generators; The time when the fault occurs; is the moment when the system reaches the point of no return.

3. The method according to claim 2, wherein The process of calculating the composite power angle difference using the instability scenario trajectory data and dividing the units into leading units and lagging units based on the composite power angle difference includes: If the stabilization control mode is online stabilization control, the composite power angle of each generator is expressed as: ； in, For the The composite power angle of the generator; is the power angle at the time when the system does not return; is the speed at the point of no return of the system; is the angular velocity of the center of inertia of the system; is the setting coefficient; ； If the stabilization control mode is real-time stabilization control, the composite power angle of each generator is expressed as: ； in, The power angle at the system switching control moment; The speed of the system at the moment of cutting machine control; is the angular velocity of the center of inertia of the system; ； Press all generators Sort from large to small to form an ordered sequence , calculate the composite power angle difference between adjacent generators in sequence: ； By difference The largest position , as the boundary of the generator group; the first Leading Group , the remaining generator groups are divided into lagging groups , among which the advanced cluster It is also a group of cutting machine solutions.

4. The method according to claim 3, wherein The process of using the system trajectory data from the time of fault to the time of generator trip as input and using the extended equal area criterion to determine the electromagnetic power angle curve and the mechanical power angle curve includes: If the stabilization control mode is online stabilization control, the equivalent process includes: using to The trajectory data at each moment is used to calculate the power angle, angular velocity, electromagnetic power and mechanical power trajectory data of the inertia center; ； in, for Group equivalent moment of inertia; for Group equivalent moment of inertia; for The equivalent power angle of the group; for The speed of the group, for The electromagnetic power of the group; for Mechanical power trajectory data of the group; for The equivalent power angle of the group; for The rotation speed of the group; for The electromagnetic power of the group; for Mechanical power trajectory data of the group; The power angle, angular velocity, electromagnetic power and mechanical power trajectory data of the equivalent system from the time the fault occurs to the point where the system does not return are calculated as follows: ； in, is the inertia of the equivalent system; is the power angle of the equivalent system; is the rotational speed of the equivalent system; is the mechanical power of the equivalent system; is the electromagnetic power trajectory data of the equivalent system; The equivalent system state trajectory is expressed as: ； If the stabilization control mode is real-time stabilization control, the equivalent process includes: using to The trajectory data at each moment is used to calculate the power angle, angular velocity, electromagnetic power and mechanical power trajectory data of the equivalent system from the time the fault occurs to the time of the generator switching control; Model the relationship between electromagnetic power, mechanical power, and power angle: ； ； in, is the first parameter to be fitted; is the second parameter to be fitted; is the third parameter to be fitted; is the fourth parameter to be fitted; is the fifth parameter to be fitted; is the sixth parameter to be fitted; is the equivalent system power angle, is the electromagnetic power angle curve of the equivalent system, is the mechanical power-angle curve of the equivalent system; by 、 and , To map the sequence, the nonlinear and least squares parameter identification methods are used to calculate the parameters to be identified. ; The power angle at the time of cutting control As a starting point, For the end, Generate extrapolated power angle prediction sequence with equal intervals of 0.01rad in the interval , substitute the prediction sequence into the established power angle curve equation to obtain the corresponding electromagnetic power and mechanical power prediction sequence , ; Search from front to back to meet The sequence elements form the electromagnetic power prediction sequence corresponding to the period from the moment of machine cutting control to the moment of the system no return point. and mechanical power prediction series , the corresponding power angle prediction sequence is ; The state quantity trajectory data of the equivalent system and the state quantity prediction sequence data are stored in the state quantity trajectory vector to obtain the state trajectory of the equivalent system.

5. The method according to claim 4, wherein Utilizing the electromagnetic power angle curve and the mechanical power angle curve, the acceleration area, the deceleration area, and the remaining deceleration area of ​​the equivalent units before and after the cutting of the generator are calculated according to the grouping results; Then the process of calculating the unbalanced energy mutation caused by cutting the machine includes: Based on the trajectory data during the fault period, the rectangular method is used to calculate the acceleration area during the fault period, with the power angle as the independent variable and the mechanical power and electromagnetic power as the dependent variables; ； in, is the acceleration area of ​​the equivalent system during the fault period, i.e., the accumulated unbalanced energy during the fault period; Based on the trajectory data after fault removal, the rectangular method is used to calculate the deceleration area from the time of fault removal to the time of generator control, with power angle as the independent variable and mechanical power and electromagnetic power as the dependent variables. ； in, It is the deceleration area of ​​the equivalent system from the time when the fault is cleared to the time when the generator is cut off, that is, the calculation of the unbalanced energy from the time when the fault is cleared to the time when the generator is cut off; The accumulated unbalanced energy is calculated by subtracting the calculated acceleration area from the deceleration area after the fault is removed: ； Where, It is the accumulated unbalanced energy from the occurrence of system failure to the moment of generator trip control; Based on the trajectory data after the machine shedding control, the power angle is the independent variable, and the mechanical power and electromagnetic power are the dependent variables. The rectangular method is used to calculate the deceleration area from the machine shedding control to the point of no return: ； in, It is the deceleration area of ​​the equivalent system from the moment of cutting machine control to the point of no return; Calculate the accumulated unbalanced energy and the deceleration area after removing the cut generator; and calculate the change in the difference between the accumulated unbalanced energy and the deceleration area before and after removing the cut generator, that is, the sudden change in the unbalanced energy of the cut generator.

6. The method according to claim 5, wherein The process of calculating the electromagnetic power transfer of the disconnected unit using the power flow information at the time of generator disconnection and the power system network topology as input and assuming that the generator disconnection causes a shift in the power angle curve of the equivalent unit to calculate the change in the retarded area includes the following steps: Taking the power flow information at the time of generator shedding and the power system network topology as input, the transient internal potential of the generator and the phase angle under the synchronous coordinate are calculated; the nodes within the generator are expanded and the non-generator nodes are eliminated to form a simplified admittance matrix; A unit is randomly selected from the leading group for removal, and the change in electromagnetic power of each generator and the change in electromagnetic power of the equivalent system are calculated. Assuming that the machine removal causes the overall shift of the EEAC equivalent unit power angle curve, the change in the retarded area is calculated.

7. The method according to claim 6, wherein Taking the power flow information at the time of generator shedding and the power system network topology as input, the transient internal potential of the generator and the phase angle under the synchronous coordinate are calculated; The process of expanding the nodes within the generator and eliminating the non-generator nodes to form a simplified admittance matrix includes: Based on the power flow information at the time of power outage and the power system network topology data, including the generator Active power and reactive power ,dynamo Amplitude of terminal voltage and phase angle ,dynamo Subtransient reactance , calculate the transient internal potential of the generator Phase angle with synchronous coordinates : ； ； Based on the power system network topology, the original node admittance matrix of the power system generator is established ; ； Calculate the equivalent admittance parameters of the generator: ； in, For the Equivalent admittance parameters of the generator; For the Equivalent resistance of a generator; For the Equivalent reactance of each generator; Expand the admittance matrix: ； in, is the original generator node self-admittance; is the newly added generator node self-admittance; is the mutual admittance between the original generator node and the newly added generator node; is the mutual admittance between the newly added generator node and the original generator node; Until all the impedance equivalent admittances in the generator sets are merged into the original node admittance matrix; Choose any load , model the load as a parallel admittance and calculate the parallel admittance: ； in, is the parallel admittance; To control the active load at the previous moment; To control the reactive load at the previous moment; is the node bus voltage amplitude; Merge the load equivalent admittance into the original node admittance matrix ; ； in, is the original load node self-admittance; Until all load equivalent admittances are merged into the original node admittance matrix; Arrange the non-generator internal nodes to the front of the admittance matrix and the generator internal nodes to the back. The rearranged admittance matrix is: ； in, represents the rearranged admittance matrix; represents the self-admittance submatrix between non-generator nodes, represents the node admittance matrix within the generator, represents the mutual admittance between non-generator internal nodes and generator internal nodes; represents the mutual admittance between the nodes within the generator and the nodes outside the generator; Eliminate non-generator internal nodes and obtain a simplified equivalent admittance matrix containing only generator internal nodes: ； in, To simplify the equivalent admittance matrix; Separate the simplified equivalent admittance matrix according to the real and imaginary parts of the admittance to form a simplified equivalent conductance matrix and the simplified equivalent susceptance matrix .

8. The method according to claim 7, wherein The process of randomly selecting a unit from the leading group and removing it and calculating the change in electromagnetic power of each generator and the change in electromagnetic power of the equivalent system includes: Calculate the change in electromagnetic power of the remaining units caused by generator shedding, specifically: ； ； in, For generator nodes Node with the removed generator set Synchronous torque coefficient between; For generator nodes The internal potential of is the generator terminal voltage before the cut-off unit is cut off. To simplify the equivalent susceptance matrix OK column values, For generators Voltage phase angle difference before disturbance, Node of the removed generator set Active power; Then calculate the node of the removed generator set The change of electromagnetic power of the equivalent system is: ； in, To control the cutting time ahead of the first The change in electromagnetic power output of each generator; The first unit in the group with the delayed switching control time The change in electromagnetic power output of each generator; For the generator set to be removed Inertia; . is the sum of the inertia of the leading group and the inertia of the lagging group; Assuming that the generator shedding causes the EEAC equivalent unit power angle curve to shift as a whole, the process of calculating the change in the retarded area includes: If the stable control mode is online stable control, the method for determining the equivalent system power angle curve after the machine is cut off is: the equivalent system electromagnetic power angle curve after the machine is cut off is Pan Up , and obtain the equivalent system electromagnetic power angle curve after translation : ； in, It is the equivalent system power angle after removing the cut-off generator set; Mechanical power is the equivalent system mechanical power angle curve after the machine is cut off. ; If the stable control mode is real-time stable control, the method for determining the equivalent system power angle curve after the machine is cut off is: the equivalent system electromagnetic power angle curve after the machine is cut off is Pan Up , and obtain the equivalent system electromagnetic power angle curve after translation : ； Mechanical power is the equivalent system mechanical power angle curve after the machine is cut off. : ； The power angle at the time of cutting control As a starting point, For the end, Generate extrapolated power angle sequence with equal intervals of 0.01rad in the interval , substitute the power angle sequence into the established power angle curve, and search from front to back for the first one that satisfies The power angle value is the power angle value at the point of no return of the power angle curve of the equivalent system after translation. ; Calculate the change in deceleration area after cutting machine control: ； in, To eliminate the power angle at the system's non-return point after the disconnected generator set is removed; This continues until all units in all leading fleets have completed their assessments.

9. The method according to claim 8, wherein The process of calculating the stability margin change by combining the sudden change in the unbalanced energy of the cutting unit with the change in the decelerable area corresponding to the power transfer, and dividing the result by the cutting amount of the cut-off unit to calculate the cutting sensitivity includes: The stability margin change is obtained by adding the unbalanced energy mutation of all units in the leading group to the change of the decelerable area; ； is the change of stability margin; Divide the change in stability margin corresponding to each cut-off unit by the cut-off amount of the cut-off unit to calculate the cut-off sensitivity of all units in the leading group: ； in, It is the tripping sensitivity of all units in the leading group.

10. A trip sensitivity calculation system for transient stability control, characterized in that: comprising a grouping module, a first calculation module, a second calculation module and a third calculation module; The grouping module is used to obtain unstable scenario trajectory data according to the stable control method of the unstable scenario; calculate the composite power angle difference using the unstable scenario trajectory data, and divide the units into leading units and lagging units based on the composite power angle difference; The first calculation module is configured to use the system trajectory data from the time of fault to the time of generator shedding as input, and determine the electromagnetic power angle curve and the mechanical power angle curve using the extended equal area criterion; use the electromagnetic power angle curve and the mechanical power angle curve to calculate the acceleration area, deceleration area, and remaining deceleration area of ​​the equivalent units before and after generator shedding based on the grouping results; and then calculate the unbalanced energy mutation amount caused by generator shedding; The second calculation module is used to calculate the electromagnetic power transfer of the disconnected unit based on the power flow information at the time of disconnection and the power system network topology as input, and calculate the change in the deceleration area by assuming that the disconnection causes the equivalent unit power angle curve to shift; The third calculation module is used to calculate the stability margin change amount by combining the sudden change amount of the unbalanced energy of the switching generator and the change amount of the decelerable area corresponding to the power transfer, and divide the result by the switching amount of the switched generator to calculate the switching sensitivity.

Images