Active power online emergency decision-making method of anti-transient frequency drop
By establishing the target trajectory and inertia evaluation of transient frequency drop in the power system, combined with the cooperation of controllable resources, the problem that the frequency emergency control decision-making method in the existing technology cannot quickly and accurately calculate the power requirements, and the power between multiple nodes is realized The accurate allocation and effective support of system frequency improve the stability of the power grid.
Patent Information
- Application Number
- CN202410843054.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-27
- Publication Date
- 2024-11-26
AI Technical Summary
The existing frequency emergency control decision-making method fails to effectively consider the inertia of the power system, resulting in the inability to quickly and accurately calculate the power demand of the power control node during high-power disturbances, which is prone to overcut or undercut, and cannot quickly maintain the stability of the system frequency. .
A method of online emergency decision-making for active power with anti-transient frequency drop is proposed. By establishing the target trajectory of transient frequency drop, using inertia evaluation results, combining the degree of fall of transient frequency trajectory, controlling resources are coordinated to enhance system power balance. Ability to avoid the occurrence of transient low-frequency load-down events. Specific steps include: establishing the target trajectory of the transient frequency drop, setting the node support power demand, offline setting and online setting of the node power support volume switching ratio coefficient, and allocation of the node controllable resource switching volume.
It realizes the accurate allocation of support power between multiple nodes in emergency situations, effectively supports the system frequency, avoids rapid frequency drops, and improves the reliability and operation stability of the power grid.
Smart Images

Figure CN119029918A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of power system regulation and control, and in particular relates to an online emergency decision method for active power against transient frequency drop. Background Art
[0002] For new power systems with a high proportion of renewable energy access, the system's rotational inertia continues to decrease, the frequency change rate increases, and the frequency stability problem is significant. It is urgent to improve the system's support capacity after large disturbances. When the system is subject to high-power disturbances, the rapid drop in frequency will lead to transient low-frequency load shedding. In order to improve the reliability of the power grid and reduce the economic losses caused by load shedding, a certain amount of active power can be injected into the power grid by quickly and accurately investing in new direct-controlled energy storage and cutting off adjustable loads.
[0003] Automatic Power Control (APC) is the extension of Automatic Generation Control (AGC) to the adjustable load side based on traditional generator sets. It covers the original AGC technical connotation of generator sets and expands it to the grid-connected main control units in all links of source, grid, load and storage (such as generator sets, new energy, directly controlled new energy storage and adjustable loads, etc.).
[0004] However, the existing frequency emergency control decision-making methods rarely consider the impact of power system inertia, and almost never calculate and quantify the power shortage of the system after disturbance from the perspective of system inertia assessment. It is impossible to quickly and accurately obtain the power demand of each power control node. At the same time, if the load is cut off according to the fixed value of low-frequency load reduction, over-cutting or under-cutting may occur under different faults. Moreover, the decision-making plan for power control quantity needs to be associated with the complex control strategies within a variety of new energy storage systems. Therefore, it is necessary to optimize the existing decision-making methods for automatic power control. Summary of the invention
[0005] In view of the above problems, the present invention proposes an online emergency decision-making method for active power against transient frequency drop, which is used for new power systems with a high proportion of renewable energy access. In the early stage of high-power disturbance, based on the transient frequency target trajectory and with the help of the inertial evaluation results of the transient frequency trajectory bad behavior, it is achieved that in an emergency, only the drop degree of the node transient frequency trajectory is used to effectively coordinate the controllable resources, thereby enhancing the system power balancing capability and operation stability, and avoiding the occurrence of transient low-frequency load reduction events.
[0006] The technical solution adopted by the present invention is as follows: an online emergency decision method for active power against transient frequency drop, the steps are as follows:
[0007] Step 1: Establishment of transient frequency drop target trajectory:
[0008] In order to maintain the system frequency within the limit, it is necessary to establish a transient frequency drop target trajectory at the initial stage of the disturbance under the premise of ensuring sufficient system inertia. The generated reference transient frequency drop trajectory and the reference inertia time constant information implicit in the trajectory are used as a unified scale for power system inertia evaluation, thereby completing the generation of the transient frequency drop target trajectory and setting the inertia time coefficients of different system nodes during the frequency drop process.
[0009] H sys =H sys0 (1-k) (1)
[0010] In the formula, H sys0 is the inertia constant of the synchronous machine system only; H sys is the system inertia constant when the proportion of new energy is k;
[0011] Inertia time constant and inertia constant H sys The following relationship exists:
[0012] M=2H sys (2)
[0013] Where, M is the reference inertia time constant;
[0014] Based on the reference inertia time constant obtained by tuning, the calculation relationship of the equivalent inertia time coefficient is obtained from the first-order time-varying inertia system dynamics equation:
[0015]
[0016] Where M A (t) is the equivalent inertia time coefficient of the actual monitored process A; ρ is the monitored object relative to the reference capacity S base The capacity multiple of f ref (t) is the frequency of the reference transient frequency drop process, which is a known value; is the frequency change rate of the benchmark transient frequency drop process, which is a known value; M CR is the reference inertia time constant, which is 5.051s; f A (t) is the frequency of process A actually monitored; is the frequency change rate of the process A actually monitored;
[0017] Set the monitored process A as the target trajectory of transient frequency drop, and set the capacity of the transient frequency drop target trajectory setting process as the system's benchmark capacity, then ρ in equation (3) is set to 1, and the equivalent inertia time constant M of the transient frequency drop target trajectory is obtained: T The calculation relationship is:
[0018]
[0019] In the formula, f T (t) is the frequency of the transient frequency drop target trajectory; is the frequency change rate of the transient frequency drop target trajectory;
[0020] Set the disturbance time to 0s, the frequency drops, and generate transient frequency drop target trajectories corresponding to different inertia time constants;
[0021] Step 2: Setting the node support power demand, the steps are as follows:
[0022] S21, determine the location of each observation node and monitor its frequency trajectory in real time;
[0023] S22. When the system is disturbed, the node frequency drops. When the node frequency value drops to the set start threshold, i.e. 49.967Hz, calculate the frequency change rate of the node at that moment. Go to step S23 and continue; if the node frequency value does not drop to the set start threshold, continue to monitor and observe the node frequency;
[0024] S23, when the observed node frequency drop rate is greater than the target frequency trajectory drop rate, that is, Then start the online setting of the power demand supported by the node, and go to step S24; if the node frequency drops slowly and does not exceed the target trajectory, continue to monitor and observe the node frequency;
[0025] S24, setting the power demand supported by the observation node: After the setting is started, based on the actual frequency drop trajectory signal collected by monitoring, the setting calculation is performed to obtain the power demand supported by the observation node ΔP at time t after the setting is started, if the actual frequency trajectory is adjusted to the target trajectory. T (t), the calculation relationship is shown in formula (5):
[0026]
[0027] Where: f m (t) is the frequency of the actual frequency drop process collected by monitoring; To monitor the frequency change rate of the actual frequency drop process collected;
[0028] Step 3: Offline setting of the switching ratio coefficient of the node power support:
[0029] In the offline state, determine the configuration of the controllable resources of the system, that is, judge whether each node can be used as a frequency measurement and control node for online decision-making of the APC system. If there is a controllable resource somewhere, select this point as the measurement and control node, otherwise it will not be selected. If there are i controllable resources in the system, set i measurement and control nodes in the system.
[0030] The starting conditions for supporting power setting are met, and the supporting power demand ΔP of the measurement and control node i is calculated according to step 2 Ti (t), the calculation relationship is shown in formula (6):
[0031]
[0032] In the formula, f i (t) is the frequency of the actual frequency drop process of the measurement and control node i; is the frequency change rate of the actual frequency drop process of the measurement and control node i;
[0033] However, the actual switching still needs to meet the switching start conditions, and the starting threshold f0 of the supporting power switching in the automatic power control system is set to 49.8Hz, that is, at time t0, the frequency of the measurement and control node drops to the threshold, and the supporting power switching link starts;
[0034] When making a decision on the control quantity of the measurement and control node i, it is necessary to consider the power support provided by other nodes near the measurement and control node to it, and convert the power support correlation between the measurement and control nodes into the ratio of the actual power support quantity of the measurement and control node i corresponding to different disturbances to the support power demand of the measurement and control node i, which is defined as the power support quantity switching ratio coefficient of the measurement and control node i;
[0035] Under a certain disturbance, for the measurement and control node i, due to the correlation of power support between the measurement and control nodes, the actual switching power support of the measurement and control node i is ΔP Ri (t), its supporting power demand is ΔP Ti (t), then the power support switching ratio coefficient k of the measurement and control node i i , the calculation relationship is shown in formula (7):
[0036]
[0037] In the online decision-making of local control of each node, after the power support switching ratio coefficient of the measurement and control node i is known, according to ΔP Ti (t), calculate the actual switching power support ΔP of the measurement and control node i Ri (t), the calculation relationship is shown in formula (8):
[0038] ΔP Ri (t) = k i ΔP Ti (t) (8)
[0039] It is necessary to establish the relationship between the transient frequency drop trajectory of the measurement and control node i and its power support switching ratio coefficient in an offline state. In an emergency, only the actual trajectory of the transient frequency drop of each measurement and control node i can be used to quickly and accurately determine the power support switching ratio coefficient of each measurement and control node, and then obtain the actual power support switching of each measurement and control node;
[0040] In the online decision-making process of the APC system, the transient frequency drop trajectory of each measurement and control node i is monitored in real time. Based on the set transient frequency drop target trajectory information, the equivalent inertia time coefficient absolute change ΔM between the transient frequency drop target trajectory and the transient frequency drop actual trajectory of the measurement and control node i is obtained. i (t), the calculation relationship is shown in formula (9):
[0041]
[0042] In the formula, ρ i is the adjustable capacity of the measurement and control node i relative to the system benchmark capacity S base multiples of;
[0043] Then, the relationship between the actual trajectory information of transient frequency drop of each measurement and control node i and the switching ratio coefficient of the power support of each measurement and control node i is adjusted to: establish the absolute change of the equivalent inertia time coefficient ΔM i (t) and the power support switching ratio coefficient k i the relationship between;
[0044] After determining the set of measurement and control nodes in the system, the power support switching ratio coefficient of the measurement and control node i is adjusted offline. The steps are as follows:
[0045] S31. Set the system to be disturbed to a certain extent, monitor the actual trajectory of transient frequency drop of each measurement and control node in real time, and calculate a set of absolute changes of equivalent inertia time coefficient ΔM of each measurement and control node i respectively. i1 as follows:
[0046] ΔM i1 =(ΔM 11 ΔM 21 ... ΔM i1 ... ΔM n1 ) T (10)
[0047] And the power demand ΔP supported by each measurement and control node i Ti (t);
[0048] S32, according to the drop degree of the transient frequency drop actual trajectory of each measurement and control node i, different switching ratio coefficients are set for the node, and a set of power support switching ratio coefficients k of each measurement and control node is obtained.i1 as follows:
[0049] k i1 =(k 11 k 21 ... k i1 ... k n1 ) T (11)
[0050] According to formula (8), the coefficient k i1 Calculate the actual switching power support ΔP of each measurement and control node Ri (t);
[0051] S33, providing ΔP for each measurement and control node Ri (t), and monitor whether the actual trajectory of transient frequency drop of each measurement and control node after power support is close to the target trajectory of transient frequency drop. If they are approximately coincident, then the group k i1 With ΔM i1 Correspondingly, if the trajectory is significantly different, it is necessary to reset the switching ratio coefficient for the node and repeat the above process until the trajectory is approximately coincident with the target trajectory, and obtain k i1 ;
[0052] By changing the disturbance size and type of the system, we can obtain multiple switching ratio coefficients k of each measurement and control node. i1 , k i2 , k i3 …, with ΔM i1 , ΔM i2 , ΔM i3 …one-to-one correspondence;
[0053] Based on this, the corresponding relationship between the switching ratio coefficient of the power support amount of each measurement and control node and the absolute change of the equivalent inertia time coefficient is obtained as follows:
[0054] k i =a i ×ΔM i +b i (12)
[0055] k i =(k1 k2 ... k i ) T (13)
[0056] ΔM i =(ΔM1 ΔM2 ... ΔM i ) T (14)
[0057] a i =(a1 a2 ... a i ) (15)
[0058] b i =(b1 b2 ... b i ) T (16)
[0059] In the formula, k i is the set of switching ratio coefficients of the power support quantity of each measurement and control node; ΔM i is the absolute change set of the equivalent inertia time coefficient of each measurement and control node; a i It is k i With ΔM i The set of slopes of the relationship, b i It is k i With ΔM i The intercept set of the relationship;
[0060] From the above relationship, when the absolute change of the equivalent inertia time coefficient of a certain measurement and control node i is ΔM i When the corresponding support power switching ratio coefficient k is calculated i for a i ΔM i +b i , used as a prerequisite for subsequent online decision-making to support power;
[0061] Step 4: Online setting of node support power
[0062] After completing the offline setting of the switching ratio coefficient of each measurement and control node, the online decision-making process of the power control amount of the APC system needs to monitor the actual trajectory of the transient frequency drop of each measurement and control node in real time. When the system is disturbed, under the starting conditions that meet the setting of the node power support amount, the power support shortage of each measurement and control node is calculated according to the actual trajectory of the transient frequency drop, and the power support amount is allocated to complete the mutual coordination between different measurement and control nodes in the sub-area.
[0063] If there is only one controllable resource in the sub-area, that is, the controllable resources are limited and there is only one measurement and control node, then the measurement and control node provides all the power support. The support power demand ΔP of the measurement and control node is calculated based on the actual trajectory of the transient frequency drop of the measurement and control node. Ti (t), the result is corrected according to the controllable resources of the measurement and control node; if the calculated ΔP Ti (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , it means that the actual adjustable resources of the measurement and control node cannot meet the support power demand of this point; accordingly, in order to maximize the support effect, the actual power support amount of the measurement and control node should be corrected to the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max;
[0064] If there are two or more controllable resources in a sub-area, it is necessary to consider the mutual cooperation between the measurement and control nodes in the sub-area. The support power demand ΔP of each measurement and control node is calculated based on the actual trajectory of transient frequency drop of the measurement and control node monitored in real time. Ti (t), according to k in formula (12) i With ΔM i The corresponding relationship between the power support amount of each measurement and control node is obtained, and the power support amount ΔP of each measurement and control node is calculated by formula (8): Ri (t), and modify the result according to the controllable resources of the measurement and control node; if the calculated ΔP Ri (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , it means that the actual adjustable resources of the measurement and control node cannot meet the support power demand of the point; accordingly, in order to maximize the support effect, the actual power support amount of the measurement and control node should be corrected to the upper limit of the node adjustable resource capacity ΔP i max ;
[0065] Based on this, the actual switching power support amount ΔP of each measurement and control node in the sub-area is obtained Ri (t) are as follows:
[0066]
[0067] The number of measurement and control nodes is i=1.2.3...n. When n=1, that is, there is only one measurement and control node, the node does not need to consider coordination and directly provides all power support;
[0068] Step 5: Allocation of node controllable resource switching
[0069] For any measurement and control node i in the sub-area, if the measurement and control node has only one controllable resource, namely, an adjustable load, then the power support is entirely provided by cutting off the adjustable load, and the response time of the controllable resource switching action is negligible, and the required power support can be provided instantaneously; if the measurement and control node has only one controllable resource, namely, a directly controlled new energy storage, then the power support is entirely provided by the energy storage discharge, and the response time of the controllable resource switching action needs to be considered;
[0070] If the measurement and control node has two controllable resources, namely, the direct-controlled new energy storage and the adjustable load, it is necessary to consider the mutual coordination of the two controllable resources, that is, the allocation of the switching ratio of the two resources; among them, the adjustable load action has no delay, which ensures the rapidity of the decision-making action; the direct-controlled new energy storage provides power support to avoid load shedding affecting the power supply to users, thus ensuring the reliability of power supply;
[0071] Based on the response characteristics of the direct-controlled new energy storage and the adjustable load providing power support, at the initial stage of the switching action, a power support command is simultaneously issued to the adjustable load and the direct-controlled new energy storage of the measurement and control node. After receiving the command, the adjustable load is cut off as required, and its action time is negligible. Since the direct-controlled new energy storage needs response time to reach the maximum discharge power, after a short delay, the energy storage reaches its maximum discharge power and provides power support for the node with this discharge power;
[0072] The allocation of the switching amount of the two controllable resources depends on the relationship between the actual switching power support of the measurement and control node and the adjustable resource capacity of the measurement and control node. If the actual switching power support amount of the measurement and control node i is ΔP Ri (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , then the actual adjustable resources of the measurement and control node need to be fully put into use, without considering the allocation ratio of the two controllable resources, and at the same time send instructions to the two controllable resources to switch the two adjustable resources;
[0073] If the adjustable resource capacity of the measurement and control node i is sufficient, the actual power support required by the two controllable resources is adjusted according to the priority of the APC system decision target;
[0074] If the priority of the decision-making goal is rapidity, the switching of controllable resources is mainly based on adjustable loads. When the adjustable load capacity is insufficient, the remaining power support is provided by the direct-controlled new energy storage. That is, the power support required by the two controllable resources is:
[0075]
[0076] In the formula, ΔP Li (t) is the power support required for the adjustable load; ΔP Li max is the upper limit of the capacity of the adjustable load; ΔP Si (t) The amount of power support required for direct control of new energy storage;
[0077] If the priority of the decision-making target is reliability, the switching of controllable resources is mainly based on direct-controlled new energy storage. When the capacity of direct-controlled new energy storage is insufficient, the remaining power support is provided by the adjustable load, that is, the power support required by the two controllable resources is:
[0078]
[0079] In the formula, ΔP Si max To directly control the capacity limit of new energy storage;
[0080] If there is no clear priority decision target, the actual power support amount ΔP of the measured and controlled node i will beRi (t) and the upper limit of the node's adjustable resource capacity ΔP i max The ratio of the two controllable resources is taken as the switching ratio m i , calculate the switching ratio m of the two controllable resources i for:
[0081]
[0082] Then, the power support required by the two controllable resources is obtained from the switching ratio:
[0083]
[0084] Based on the power support amount obtained by setting, instructions are issued to the two controllable resources at the same time to switch the two adjustable resources.
[0085] Another object of the present invention is to be achieved through an online emergency decision-making system for active power against transient frequency drops, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, an online emergency decision-making method for active power against transient frequency drops as described above is implemented.
[0086] Another object of the present invention is to be achieved through a computer-readable storage medium, on which a program for implementing information transmission is stored, and when the program is executed by a processor, the steps of the above-mentioned method for online emergency decision-making of active power against transient frequency drop are implemented.
[0087] Advantages and beneficial effects of the present invention: The present invention realizes accurate allocation of the supporting power switching amount between multiple nodes under the premise of only collecting the frequency drop trajectory signal, and the switching of the power support amount has an effective supporting effect on the system frequency after the disturbance. In the case of emergency control, the method of the present invention does not need to judge the location of the system disturbance center and the type of disturbance. It only needs to monitor the frequency drop trajectory of each controllable node in real time, and realize the local control of the power support amount of each node while ensuring the mutual cooperation between the controllable resources of each node, thereby maintaining the stability of the system frequency. BRIEF DESCRIPTION OF THE DRAWINGS
[0088] Figure 1 It is a schematic diagram of the actual trajectory and target trajectory of transient frequency drop;
[0089] Figure 2 The starting process diagram of the module for setting the power required by the node;
[0090] Figure 3 Offline setting diagram for the switching ratio coefficient of each controllable node;
[0091] Figure 4 This is the single line diagram of the CEPRI 8-machine 36-bus system;
[0092] Figure 5 It is a schematic diagram of the target trajectory of transient frequency drop;
[0093] Figure 6 It is the frequency trajectory diagram of the measurement and control node 8 (Bus 23) within 2s after the disturbance;
[0094] Figure 7 The frequency trajectory diagram of the measurement and control node 8 (Bus 23) within 2s after the disturbance (power support has been provided);
[0095] Figure 8 It is the frequency trajectory diagram of the measurement and control node 8 (Bus 23) within 2s after the disturbance (after the switching coefficient is corrected);
[0096] Fig. 9 The frequency trajectory diagram of the measurement and control node 8 (Bus 23) before and after the power injection support within 2 seconds after the disturbance. DETAILED DESCRIPTION
[0097] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in combination with the embodiments according to the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0098] Example 1
[0099] An online emergency decision method for active power against transient frequency drop, the steps are as follows:
[0100] Step 1: Establishment of transient frequency drop target trajectory:
[0101] In order to maintain the system frequency within the limit, it is necessary to establish a transient frequency drop target trajectory at the initial stage of the disturbance under the premise of ensuring sufficient system inertia. The generated reference transient frequency drop trajectory and the reference inertia time constant information implicit in the trajectory are used as a unified scale for power system inertia evaluation, thereby completing the generation of the transient frequency drop target trajectory and setting the inertia time coefficients of different system nodes during the frequency drop process.
[0102] H sys =H sys0 (1-k) (1)
[0103] In the formula, H sys0 is the inertia constant of the synchronous machine system only; H sys is the system inertia constant when the proportion of new energy is k;
[0104] Inertia time constant and inertia constant Hsys The following relationship exists:
[0105] M=2H sys (2)
[0106] Where, M is the reference inertia time constant;
[0107] Therefore, under the premise of ensuring sufficient system inertia, the inertia time constant of the transient frequency drop target trajectory is set, that is, when the system with this inertia time constant is disturbed, the transient frequency drop trajectory obtained is the target trajectory. Changing the inertia time constant setting value of the target trajectory will generate different transient frequency drop target trajectories. The inertia time constant can be set to different values depending on different power grid conditions.
[0108] The reference transient frequency drop trajectory used and the reference inertia time constant information M implied in the trajectory CR ≈5.051s, which is used as a unified scale for inertia evaluation of the power system, and then the generation of the transient frequency drop target trajectory and the subsequent setting of the inertia time coefficients of different nodes in the system during the frequency drop process are completed.
[0109] Based on the reference inertia time constant obtained by tuning, the calculation relationship of the equivalent inertia time coefficient is obtained from the first-order time-varying inertia system dynamics equation:
[0110]
[0111] Where M A (t) is the equivalent inertia time coefficient of the actual monitored process A; ρ is the monitored object relative to the reference capacity S base The capacity multiple of f ref (t) is the frequency of the reference transient frequency drop process, which is a known value; is the frequency change rate of the benchmark transient frequency drop process, which is a known value; M CR is the reference inertia time constant, which is 5.051s; f A (t) is the frequency of process A actually monitored; is the frequency change rate of the process A actually monitored;
[0112] Set the monitored process A as the target trajectory of transient frequency drop, and set the capacity of the transient frequency drop target trajectory setting process as the system's benchmark capacity, then ρ in equation (3) is set to 1, and the equivalent inertia time constant M of the transient frequency drop target trajectory is obtained: T The calculation relationship is:
[0113]
[0114] In the formula, f T(t) is the frequency of the transient frequency drop target trajectory; is the frequency change rate of the transient frequency drop target trajectory;
[0115] Set the disturbance time to 0s, the frequency drops, and generate transient frequency drop target trajectories corresponding to different inertia time constants;
[0116] Step 2: Setting the node support power demand:
[0117] The actual trajectory of the simulated node frequency drop and the target trajectory of the frequency drop are as follows: Figure 1 shown.
[0118] The system is disturbed, and the frequency of each node drops to varying degrees. If the frequency modulation system (rapid power support measures) cannot act in time, the node frequency will continue to drop rapidly, such as Figure 1 As shown, at t e At this moment, the frequency of the node will drop to 0.98pu, reaching the action frequency limit of the first round of system-level low-frequency load shedding, starting low-frequency load shedding, and reducing the reliability of system power supply. If effective frequency regulation measures can be taken in time to provide power support for the node at the initial stage of the disturbance, that is, at time t0, the required support power ΔP(t0) is provided to the node, and the actual trajectory of the transient frequency drop of the node is improved to the target trajectory, the time for the frequency at this point to drop to the critical frequency of low-frequency load shedding (49Hz, 0.98pu) can be delayed, thereby maintaining the safe and stable operation of the system.
[0119] like Figure 2 As shown, the steps are as follows:
[0120] S21, determine the location of each observation node and monitor its frequency trajectory in real time;
[0121] S22. When the system is disturbed, the node frequency drops. When the node frequency value drops to the set start threshold, i.e. 49.967Hz, calculate the frequency change rate of the node at that moment. Go to step S23 and continue; if the node frequency value does not drop to the set start threshold, continue to monitor and observe the node frequency;
[0122] S23, when the observed node frequency drop rate is greater than the target frequency trajectory drop rate, that is, Then start the online setting of the power demand supported by the node, and go to step S24; if the node frequency drops slowly and does not exceed the target trajectory, continue to monitor and observe the node frequency;
[0123] S24, setting the power demand supported by the observation node: After the setting is started, based on the actual frequency drop trajectory signal collected by monitoring, the setting calculation is performed to obtain the power demand supported by the observation node ΔP at time t after the setting is started, if the actual frequency trajectory is adjusted to the target trajectory. T (t), the calculation relationship is shown in formula (5):
[0124]
[0125] Where: f m (t) is the frequency of the actual frequency drop process collected by monitoring; To monitor the frequency change rate of the actual frequency drop process collected;
[0126] Step 3: Offline setting of the switching ratio coefficient of the node power support:
[0127] In the offline state, determine the configuration of the controllable resources of the system, that is, judge whether each node can be used as a frequency measurement and control node for online decision-making of the APC system. If there is a controllable resource somewhere, select this point as the measurement and control node, otherwise it will not be selected. If there are i controllable resources in the system, set i measurement and control nodes in the system.
[0128] The starting conditions for supporting power setting are met, and the supporting power demand ΔP of the measurement and control node i is calculated according to step 2 Ti (t), the calculation relationship is shown in formula (6):
[0129]
[0130] In the formula, f i (t) is the frequency of the actual frequency drop process of the measurement and control node i; is the frequency change rate of the actual frequency drop process of the measurement and control node i;
[0131] However, the actual switching still needs to meet the starting conditions for switching. At present, the existing technical specifications and safe and stable operation guidelines at home and abroad have clear provisions for the frequency dynamic indicators of the system. For example, the "Permissible Deviation of Power Quality Power System Frequency" (GB / T 15945-2008) and the "National Electricity Supply and Use Rules" stipulate that the safe operation frequency deviation of large power grids shall not exceed ±0.2Hz, and the transient frequency difference of small systems shall not exceed ±0.5Hz. Therefore, the starting threshold f0 of the support power switching in the automatic power control system is set to 49.8Hz, that is, at time t0, the frequency of the measurement and control node drops to the threshold, and the switching link of the support power is started;
[0132] After the support power is injected into the power grid, due to the influence of the power flow diffusion effect, the support power injected into a certain node i will spread to other AC nodes associated with it, providing corresponding power support for the node, but due to the different correlations between the nodes, there are differences in power support. Therefore, when making decisions on the control quantity of the measurement and control node i, it is necessary to consider the power support provided by other nodes near the measurement and control node to it, and convert the power support correlation between the measurement and control nodes into the ratio of the actual power support quantity of the measurement and control node i corresponding to different disturbances to the support power demand of the measurement and control node i, which is defined as the power support quantity switching ratio coefficient of the measurement and control node i;
[0133] Under a certain disturbance, for the measurement and control node i, due to the correlation of power support between the measurement and control nodes, the actual switching power support of the measurement and control node i is ΔP Ri (t), its supporting power demand is ΔP Ti (t), then the power support switching ratio coefficient k of the measurement and control node i i , the calculation relationship is shown in formula (7):
[0134]
[0135] In the online decision-making of local control of each node, after the power support switching ratio coefficient of the measurement and control node i is known, according to ΔP Ti (t), calculate the actual switching power support ΔP of the measurement and control node i Ri (t), the calculation relationship is shown in formula (8):
[0136] ΔP Ri (t) = k i ΔP Ti (t) (8)
[0137] However, under different disturbances, the power support switching ratio coefficients corresponding to each measurement and control node are also different. Since the subsequent online decision-making process of the APC system is an emergency, it is necessary to quickly and accurately complete the control of the power support switching. However, in the actual decision-making, it is impossible to determine the disturbance location and type of the system and the relationship between the disturbance center and the location of the controllable resources in a short time. The data obtained by real-time monitoring is only the transient frequency drop trajectory of each measurement and control node.
[0138] Therefore, it is necessary to establish the relationship between the transient frequency drop trajectory of the measurement and control node i and its power support switching ratio coefficient in an offline state. In an emergency, only the actual trajectory of the transient frequency drop of each measurement and control node i can be used to quickly and accurately determine the power support switching ratio coefficient of each measurement and control node, and then obtain the actual power support of each measurement and control node.
[0139] In the online decision-making process of the APC system, the transient frequency drop trajectory of each measurement and control node i is monitored in real time. Based on the set transient frequency drop target trajectory information, the equivalent inertia time coefficient absolute change ΔM between the transient frequency drop target trajectory and the transient frequency drop actual trajectory of the measurement and control node i is obtained. i (t), the calculation relationship is shown in formula (9):
[0140]
[0141] In the formula, ρ i is the adjustable capacity of the measurement and control node i relative to the system benchmark capacity S base multiples of;
[0142] Then, the relationship between the actual trajectory information of transient frequency drop of each measurement and control node i and the switching ratio coefficient of the power support of each measurement and control node i is adjusted to: establish the absolute change of the equivalent inertia time coefficient ΔM i (t) and the power support switching ratio coefficient k i the relationship between;
[0143] like Figure 3 As shown, after determining the set of measurement and control nodes in the system, the power support switching ratio coefficient of the measurement and control node i is adjusted offline, and the steps are as follows:
[0144] S31. Set the system to be disturbed to a certain extent, monitor the actual trajectory of transient frequency drop of each measurement and control node in real time, and calculate a set of absolute changes of equivalent inertia time coefficient ΔM of each measurement and control node i respectively. i1 as follows:
[0145] ΔM i1 =(ΔM 11 ΔM 21 ... ΔM i1 ... ΔM n1 ) T (10)
[0146] And the power demand ΔP supported by each measurement and control node i Ti (t);
[0147] S32, according to the drop degree of the transient frequency drop actual trajectory of each measurement and control node i, different switching ratio coefficients are set for the node, and a set of power support switching ratio coefficients k of each measurement and control node is obtained. i1 as follows:
[0148] k i1 =(k 11 k 21 ... k i1 ... k n1 )T (11)
[0149] According to formula (8), the coefficient k i1 Calculate the actual switching power support ΔP of each measurement and control node Ri (t);
[0150] S33, providing ΔP for each measurement and control node Ri (t), and monitor whether the actual trajectory of transient frequency drop of each measurement and control node after power support is close to the target trajectory of transient frequency drop. If they are approximately coincident, then the group k i1 With ΔM i1 Correspondingly, if the trajectory is significantly different, it is necessary to reset the switching ratio coefficient for the node and repeat the above process until the trajectory is approximately coincident with the target trajectory, and obtain k i1 ;
[0151] By changing the disturbance size and type of the system, we can obtain multiple switching ratio coefficients k of each measurement and control node. i1 , k i2 , k i3 …, with ΔM i1 , ΔM i2 , ΔM i3 …one-to-one correspondence;
[0152] Based on this, the corresponding relationship between the switching ratio coefficient of the power support amount of each measurement and control node and the absolute change of the equivalent inertia time coefficient is obtained as follows:
[0153] k i =a i ×ΔM i +b i (12)
[0154] k i =(k1 k2 ... k i ) T (13)
[0155] ΔM i =(ΔM1 ΔM2 ... ΔM i ) T (14)
[0156] a i =(a1 a2 ... a i ) (15)
[0157] b i =(b1 b2 ... b i ) T (16)
[0158] In the formula, k iis the set of switching ratio coefficients of the power support quantity of each measurement and control node; ΔM i is the absolute change set of the equivalent inertia time coefficient of each measurement and control node; a i It is k i With ΔM i The set of slopes of the relationship, b i It is k i With ΔM i The intercept set of the relationship;
[0159] From the above relationship, when the absolute change of the equivalent inertia time coefficient of a certain measurement and control node i is ΔM i When the corresponding support power switching ratio coefficient k is calculated i for a i ΔM i +b i , used as a prerequisite for subsequent online decision-making to support power;
[0160] Step 4: Online setting of node support power
[0161] After completing the offline setting of the switching ratio coefficient of each measurement and control node, the online decision-making process of the power control amount of the APC system needs to monitor the actual trajectory of the transient frequency drop of each measurement and control node in real time. When the system is disturbed, under the starting conditions that meet the setting of the node power support amount, the power support shortage of each measurement and control node is calculated according to the actual trajectory of the transient frequency drop, and the power support amount is allocated to complete the mutual coordination between different measurement and control nodes in the sub-area.
[0162] If there is only one controllable resource in the sub-area, that is, the controllable resources are limited and there is only one measurement and control node, then the measurement and control node provides all the power support. The support power demand ΔP of the measurement and control node is calculated based on the actual trajectory of the transient frequency drop of the measurement and control node. Ti (t), the result is corrected according to the controllable resources of the measurement and control node; if the calculated ΔP Ti (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , it means that the actual adjustable resources of the measurement and control node cannot meet the support power demand of this point; accordingly, in order to maximize the support effect, the actual power support amount of the measurement and control node should be corrected to the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max ;
[0163] If there are two or more controllable resources in a sub-area, it is necessary to consider the mutual cooperation between the measurement and control nodes in the sub-area. The support power demand ΔP of each measurement and control node is calculated based on the actual trajectory of transient frequency drop of the measurement and control node monitored in real time. Ti(t), according to k in formula (12) i With ΔM i The corresponding relationship between the power support amount of each measurement and control node is obtained, and the power support amount ΔP of each measurement and control node is calculated by formula (8): Ri (t), and modify the result according to the controllable resources of the measurement and control node; if the calculated ΔP Ri (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , it means that the actual adjustable resources of the measurement and control node cannot meet the support power demand of the point; accordingly, in order to maximize the support effect, the actual power support amount of the measurement and control node should be corrected to the upper limit of the node adjustable resource capacity ΔP i max ;
[0164] Based on this, the actual switching power support amount ΔP of each measurement and control node in the sub-area is obtained Ri (t) are as follows:
[0165]
[0166] Among them, the number of measurement and control nodes i=1.2.3...n, when n=1, that is, when there is only one measurement and control node, this node does not need to consider coordination and directly provides all power support.
[0167] Step 5: Allocation of node controllable resource switching
[0168] From step 4, we can get that, under the cooperation of each measurement and control node in the sub-area, the actual power support amount of the measurement and control node is ΔP Ri (t). For any measurement and control node i in the sub-area, if the measurement and control node has only one controllable resource, namely, an adjustable load, then the power support is entirely provided by cutting off the adjustable load, and the response time of the controllable resource switching action is negligible, and the required power support can be provided instantaneously; if the measurement and control node has only one controllable resource, namely, a directly controlled new energy storage, then the power support is entirely provided by the energy storage discharge, and the response time of the controllable resource switching action needs to be considered;
[0169] If the measurement and control node has two controllable resources, namely, the direct-controlled new energy storage and the adjustable load, it is necessary to consider the mutual coordination of the two controllable resources, that is, the allocation of the switching ratio of the two resources; among them, the adjustable load action has no delay, which ensures the rapidity of the decision-making action; the direct-controlled new energy storage provides power support to avoid load shedding affecting the power supply to users, thus ensuring the reliability of power supply;
[0170] Based on the response characteristics of the direct-controlled new energy storage and the adjustable load providing power support, at the initial stage of the switching action, a power support command is simultaneously issued to the adjustable load and the direct-controlled new energy storage of the measurement and control node. After receiving the command, the adjustable load is cut off as required, and its action time is negligible. Since the direct-controlled new energy storage needs response time to reach the maximum discharge power, after a short delay, the energy storage reaches its maximum discharge power and provides power support for the node with this discharge power;
[0171] The allocation of the switching amount of the two controllable resources depends on the relationship between the actual switching power support of the measurement and control node and the adjustable resource capacity of the measurement and control node. If the actual switching power support amount of the measurement and control node i is ΔP Ri (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , then the actual adjustable resources of the measurement and control node need to be fully put into use, without considering the allocation ratio of the two controllable resources, and at the same time send instructions to the two controllable resources to switch the two adjustable resources;
[0172] If the adjustable resource capacity of the measurement and control node i is sufficient, the actual power support required by the two controllable resources is adjusted according to the priority of the APC system decision target;
[0173] If the priority of the decision-making goal is rapidity, the switching of controllable resources is mainly based on adjustable loads. When the adjustable load capacity is insufficient, the remaining power support is provided by the direct-controlled new energy storage. That is, the power support required by the two controllable resources is:
[0174]
[0175] In the formula, ΔP Li (t) is the power support required for the adjustable load; ΔP Li max is the upper limit of the capacity of the adjustable load; ΔP Si (t) The amount of power support required for direct control of new energy storage;
[0176] If the priority of the decision-making target is reliability, the switching of controllable resources is mainly based on direct-controlled new energy storage. When the capacity of direct-controlled new energy storage is insufficient, the remaining power support is provided by the adjustable load, that is, the power support required by the two controllable resources is:
[0177]
[0178] In the formula, ΔP Si max To directly control the capacity limit of new energy storage;
[0179] If there is no clear priority decision target, the actual power support amount ΔP of the measured and controlled node i will beRi (t) and the upper limit of the node's adjustable resource capacity ΔP i max The ratio of the two controllable resources is taken as the switching ratio m i , calculate the switching ratio m of the two controllable resources i for:
[0180]
[0181] Then, the power support required by the two controllable resources is obtained from the switching ratio:
[0182]
[0183] Based on the power support amount obtained by setting, instructions are issued to the two controllable resources at the same time to switch the two adjustable resources.
[0184] Example 2
[0185] The 8-machine 36-bus system of CEPRI power system analysis and synthesis program is selected as the simulation object. The single-line diagram is as follows: Figure 4 As shown. The main voltage level of the system is 220kV, and the rated frequency is f n The frequency is 50Hz, and the reference capacity is 100MVA. The power types, rated capacities and inherent (rated) inertia time constants of each power source are listed in Table 1. During the simulation, when the expected disturbance event is set to 0s, the No. 1 (G1) and No. 8 (G8) units in the system experience power shortage disturbances of varying degrees, and all synchronous units except G1 are put into primary frequency regulation.
[0186] In this system, there are 9 load concentrated access nodes (Bus 9, Bus 16, Bus 18-20, Bus 21-23, Bus 29). These nodes are selected as the AC measurement and control nodes in this example, and are numbered from node 1 to node 9 in sequence. The actual adjustable capacity P of each measurement and control node is i Listed in Table 2.
[0187] Table 1 Unit setting conditions
[0188]
[0189] Table 2 Actual adjustable capacity of each measurement and control node
[0190]
[0191] (1) Establishment of transient frequency drop target trajectory
[0192] Set the inertia time constant M of the transient frequency drop target trajectory Tis 10s, set the disturbance time (usually set to 0s), the frequency drops, and the generated transient frequency drops to the target trajectory f T (t) Schematic diagram Figure 5 shown.
[0193] (2) Setting the power support required by the node
[0194] Collect the frequency trajectory signal of each measurement and control node within 2s after the disturbance, and drop the frequency of each measurement and control node to the threshold f s (49.967Hz, 0.99934pu) time t s And the starting time t that meets the starting conditions of supporting power setting s0 Listed in Table 3.
[0195] Table 3 Actual frequency trajectory drop of each measurement and control point
[0196]
[0197] It can be seen from Table 3 that when the frequency drops to the threshold value f s At this time, the RoCoF corresponding to almost all the measurement and control nodes is not faster than the target trajectory, and the starting conditions for supporting power setting are not met. Compare the time t of each node to reach the starting condition for supporting power setting s0 It can be seen that the frequency of node 8 (Bus 23) first meets the conditions for supporting power setting start, and the time t for this node to reach setting start s0 The shortest frequency drop is the most serious. The frequency trajectory signal within 2s after the disturbance is as follows: Figure 6 As shown, the measurement and control node 8 (Bus 23) can be listed as a node to be monitored emphatically after power switching control.
[0198] Taking the measurement and control node 8 (Bus 23) as an example, after meeting the starting conditions for supporting power setting, the required power support capacity of the node can be calculated to be 33.4MW by applying formula (7).
[0199] (3) Offline setting of the switching ratio coefficient of the node power support
[0200] First, the time t0 when each measurement and control node falls to the support power switching start threshold f0 (49.8Hz, 0.996pu) and the absolute change ΔM of the equivalent inertia time coefficient of the measurement and control node at the switching start time calculated based on the target frequency trajectory are calculated. i1 (t0) and the required support power ΔP Ti (t0) are listed in Table 4.
[0201] Table 4 The required support power of each measurement and control node
[0202]
[0203] According to the frequency trajectory drop degree of each measurement and control node, different switching ratio coefficients are set for the nodes, and the actual switching power support amount ΔP of each node is calculated according to this coefficient by formula (9): Ri (t0), the switching ratio coefficient k i1 The actual switching power support ΔP Ri (t0) are listed in Table 5.
[0204] Table 5 Control parameters of actual switching support power of each measurement and control node
[0205]
[0206] Provide power support ΔP for each node Ri (t0), and monitor the frequency trajectory of the power support node 8 (Bus 23) as follows Figure 7 shown.
[0207] Depend on Figure 7 It can be seen that after the power input support is provided to the node according to the above switching ratio, the frequency of the measurement and control node 8 (Bus23) at 2s still does not reach the target trajectory, and the switching ratio coefficient of each node needs to be corrected.
[0208] The switching ratio coefficient is continuously corrected according to the frequency trajectory of each node, and the actual switching power support amount ΔP of each node is calculated by formula (9): Ri (t0), until the frequency trajectory after providing power support approaches the transient drop target trajectory Among them, the frequency trajectory of the measurement and control node 8 (Bus 23) is as follows Figure 8 shown.
[0209] Depend on Figure 8 It can be seen that according to the corrected switching ratio as the node input power support, the frequency of each measurement and control node at 2s has reached the target trajectory, and the corrected switching ratio coefficient k of each measurement and control node is i1 , and the absolute change of the equivalent inertia time coefficient of the corresponding measurement and control node ΔM i1 (t0), the actual power support of the node switching ΔP Ri (t0) are listed in Table 6.
[0210] Table 6 Actual switching support power control parameters of each measurement and control node (after correction)
[0211]
[0212] Adjust the setting of the expected disturbance event, repeat the above setting process of the switching ratio coefficient, and obtain the absolute change of the inertia time coefficient of each measurement and control node ΔM i(t) The corresponding switching ratio coefficient k i Listed in Table 7.
[0213] Table 7 Corresponding relationship between the absolute change of the equivalent inertia time coefficient of each measurement and control node and the switching ratio coefficient
[0214]
[0215]
[0216] Based on Table 7, the corresponding relationship between the power switching ratio coefficient of each measurement and control node and the absolute change of the equivalent inertia time coefficient (k i -ΔM i )as follows:
[0217] k i =a i ×ΔM i +b i
[0218] in,
[0219] a i =(a1 a2...a9) T
[0220] b i =(b1 b2...b9) T
[0221] a i 、b i are the changing parameters of the above corresponding relationship, which are listed in Table 8.
[0222] Table 8 Each measurement and control node k i With ΔM i Corresponding relationship change parameters
[0223]
[0224] (4) Online setting of sub-area node support power
[0225] The controllable resource (direct-controlled new energy storage, adjustable load) capacity is configured for each measurement and control node as shown in Table 9, ignoring the switching delay and the discharge delay of the directly controllable new energy storage, and there is no clear priority decision-making target.
[0226] Table 9 Controllable resource capacity of each measurement and control node
[0227]
[0228]
[0229] Based on the frequency trajectory signals of each measurement and control node 1 to 9 (Bus 9, Bus 16, Bus 18 to 20, Bus 21 to 23, Bus 29) collected within 2 seconds after the disturbance, after the starting conditions for supporting power setting are met, the supporting power demand ΔP of each measurement and control node is calculated Ti The time t0 when each measurement and control node falls to the support power switching start threshold f0 (49.8Hz, 0.996pu) and the absolute change ΔM of the equivalent inertia time coefficient of the measurement and control node at the switching start time calculated based on the target frequency trajectory are calculated. i (t0) and the required support power ΔP Ti (t0) are listed in Table 10.
[0230] Table 10 The required support power of each measurement and control node
[0231]
[0232] Since there are two or more controllable resources in a sub-area, it is necessary to consider the coordination between the measurement and control nodes in the sub-area. According to the actual measured absolute change of the equivalent inertia time coefficient of each measurement and control node ΔM i (t0), refer to the corresponding relationship in Table 8 (k i -ΔM i ) is used to calculate the switching ratio coefficient k corresponding to each measurement and control node using formula (14): i , the actual switching power support amount ΔP of each measurement and control node is obtained by setting Ri (t0) are listed in Table 11.
[0233] Table 11 Control parameters of actual switching support power of each measurement and control node
[0234]
[0235] It can be seen from Table 11 that the actual switching power support calculated by each measurement and control node does not exceed the upper limit of the controllable resource capacity equipped by the node, which means that the controllable resources of the node meet the actual required support power of the point.
[0236] (5) Allocation of node controllable resource switching
[0237] Since the measurement and control node is equipped with two controllable resources, the coordination between the two controllable resources needs to be considered. The switching ratio m of the two resources of the measurement and control node is calculated according to formula (22): i , and then the power support amount ΔP required for each node to directly control the new energy storage and adjustable load can be obtained Li (t), ΔP Si (t), which are listed in Table 12.
[0238] Table 12 Control parameters of controllable resources switching of each measurement and control node
[0239]
[0240]
[0241] Since the switching delay and the discharge delay of the directly controllable new energy storage are ignored, and there is no clear priority decision target, power support instructions are issued to the switchable load and the directly controllable new energy storage at the same time, and the two controllable resources can provide corresponding power support for the node at the same time. Based on the power support switching amount in Table 12, power support is injected into each measurement and control node, and the frequency trajectory of the measurement and control node 8 (Bus23) before and after the injection of power support is obtained as follows Fig. 9 shown.
[0242] Depend on Figure 3-8 It can be seen that at 2s, the instantaneous frequency of the measurement and control node 8 (Bus 23) is significantly higher than the target trajectory. This shows that the support power injected into the node based on the switching amount obtained in the above process has delayed the drop of the node frequency, played a certain power support role, and provided time for subsequent frequency recovery.
[0243] The results of this example show that this active power online emergency decision-making method can achieve accurate allocation of supporting power switching between multiple nodes under the premise of only collecting frequency drop trajectory signals, and the switching of power support has an effective supporting effect on the system frequency after disturbance.
Claims
1. An online emergency decision method for active power against transient frequency drop, characterized in that: The steps are as follows: Step 1: Establishment of transient frequency drop target trajectory: In order to maintain the system frequency within the limit, it is necessary to establish a transient frequency drop target trajectory at the initial stage of the disturbance under the premise of ensuring sufficient system inertia. The generated reference transient frequency drop trajectory and the reference inertia time constant information implicit in the trajectory are used as a unified scale for power system inertia evaluation, thereby completing the generation of the transient frequency drop target trajectory and setting the inertia time coefficients of different system nodes during the frequency drop process. H sys =H sys0 (1-k)(1) In the formula, H sys0 is the inertia constant of the synchronous machine system only; H sys is the system inertia constant when the proportion of new energy is k; Inertia time constant and inertia constant H sys The following relationship exists: M=2H sys (2) Where, M is the reference inertia time constant; Based on the reference inertia time constant obtained by tuning, the calculation relationship of the equivalent inertia time coefficient is obtained from the first-order time-varying inertia system dynamics equation: Where M A (t) is the equivalent inertia time coefficient of the process A actually monitored; ρ is the monitoring object relative to the benchmark capacity S base The capacity multiple of f ref (t) is the frequency of the reference transient frequency drop process, which is a known value; is the frequency change rate of the benchmark transient frequency drop process, which is a known value; M CR is the reference inertia time constant, which is 5.051s; f A (t) is the frequency of process A actually monitored; is the frequency change rate of the process A actually monitored; Set the monitored process A as the target trajectory of transient frequency drop, and set the capacity of the transient frequency drop target trajectory setting process as the system's benchmark capacity, then ρ in equation (3) is set to 1, and the equivalent inertia time constant M of the transient frequency drop target trajectory is obtained: T The calculation relationship is: In the formula, f T (t) is the frequency of the transient frequency drop target trajectory; is the frequency change rate of the transient frequency drop target trajectory; Set the disturbance time to 0s, the frequency drops, and generate transient frequency drop target trajectories corresponding to different inertia time constants; Step 2: Setting the node support power demand, the steps are as follows: S21, determine the location of each observation node and monitor its frequency trajectory in real time; S22. When the system is disturbed, the node frequency drops. When the node frequency value drops to the set start threshold, i.e. 49.967Hz, calculate the frequency change rate of the node at that moment. Go to step S23 and continue; if the node frequency value does not drop to the set start threshold, continue to monitor and observe the node frequency; S23, when the observed node frequency drop rate is greater than the target frequency trajectory drop rate, that is, Then start the online setting of the power demand supported by the node, and go to step S24; if the node frequency drops slowly and does not exceed the target trajectory, continue to monitor and observe the node frequency; S24, setting the power demand supported by the observation node: After the setting is started, based on the actual frequency drop trajectory signal collected by monitoring, the setting calculation is performed to obtain the power demand supported by the observation node ΔP at time t after the setting is started, if the actual frequency trajectory is adjusted to the target trajectory. T (t), the calculation relationship is shown in formula (5): Where: f m (t) is the frequency of the actual frequency drop process collected by monitoring; To monitor the frequency change rate of the actual frequency drop process collected; Step 3: Offline setting of the switching ratio coefficient of the node power support: In the offline state, determine the configuration of the controllable resources of the system, that is, judge whether each node can be used as a frequency measurement and control node for online decision-making of the APC system. If there is a controllable resource somewhere, select this point as the measurement and control node, otherwise it will not be selected. If there are i controllable resources in the system, set i measurement and control nodes in the system. The starting conditions for supporting power setting are met, and the supporting power demand ΔP of the measurement and control node i is calculated according to step 2 Ti (t), the calculation relationship is shown in formula (6): In the formula, f i (t) is the frequency of the actual frequency drop process of the measurement and control node i; is the frequency change rate of the actual frequency drop process of the measurement and control node i; However, the actual switching still needs to meet the switching start conditions, and the starting threshold f0 of the supporting power switching in the automatic power control system is set to 49.8Hz, that is, at time t0, the frequency of the measurement and control node drops to the threshold, and the supporting power switching link starts; When making a decision on the control quantity of the measurement and control node i, it is necessary to consider the power support provided by other nodes near the measurement and control node to it, and convert the power support correlation between the measurement and control nodes into the ratio of the actual power support quantity of the measurement and control node i corresponding to different disturbances to the support power demand of the measurement and control node i, which is defined as the power support quantity switching ratio coefficient of the measurement and control node i; Under a certain disturbance, for the measurement and control node i, due to the correlation of power support between the measurement and control nodes, the actual switching power support of the measurement and control node i is ΔP Ri (t), its supporting power demand is ΔP Ti (t), then the power support switching ratio coefficient k of the measurement and control node i i , the calculation relationship is shown in formula (7): In the online decision-making of local control of each node, after the power support switching ratio coefficient of the measurement and control node i is known, according to ΔP Ti (t), calculate the actual switching power support ΔP of the measurement and control node i Ri (t), the calculation relationship is shown in formula (8): ΔP Ri (t)=k i ·ΔP Ti (t) (8) It is necessary to establish the relationship between the transient frequency drop trajectory of the measurement and control node i and its power support switching ratio coefficient in an offline state. In an emergency, only the actual trajectory of the transient frequency drop of each measurement and control node i can be used to quickly and accurately determine the power support switching ratio coefficient of each measurement and control node, and then obtain the actual power support switching of each measurement and control node; In the online decision-making process of the APC system, the transient frequency drop trajectory of each measurement and control node i is monitored in real time. Based on the set transient frequency drop target trajectory information, the equivalent inertia time coefficient absolute change ΔM between the transient frequency drop target trajectory and the transient frequency drop actual trajectory of the measurement and control node i is obtained. i (t), the calculation relationship is shown in formula (9): In the formula, ρ i is the adjustable capacity of the measurement and control node i relative to the system benchmark capacity S base multiples of; Then, the relationship between the actual trajectory information of transient frequency drop of each measurement and control node i and the switching ratio coefficient of the power support of each measurement and control node i is adjusted to: establish the absolute change of the equivalent inertia time coefficient ΔM i (t) and the power support switching ratio coefficient k i the relationship between; After determining the set of measurement and control nodes in the system, the power support switching ratio coefficient of the measurement and control node i is adjusted offline. The steps are as follows: S31. Set the system to be disturbed to a certain extent, monitor the actual trajectory of transient frequency drop of each measurement and control node in real time, and calculate a set of absolute changes of equivalent inertia time coefficient ΔM of each measurement and control node i respectively. i1 as follows: ΔM i1 =(ΔM 11 ΔM 21 ... ΔM i1 ... ΔM n1 ) T (10) And the power demand ΔP supported by each measurement and control node i Ti (t); S32, according to the drop degree of the transient frequency drop actual trajectory of each measurement and control node i, different switching ratio coefficients are set for the node, and a set of power support switching ratio coefficients k of each measurement and control node is obtained. i1 as follows: k i1 =(k 11 k 21 ... k i1 ... k n1 ) T (11) According to formula (8), the coefficient k i1 Calculate the actual switching power support ΔP of each measurement and control node Ri (t); S33, providing ΔP for each measurement and control node Ri (t), and monitor whether the actual trajectory of transient frequency drop of each measurement and control node after power support is close to the target trajectory of transient frequency drop. If they are approximately coincident, then the group k i1 With ΔM i1 Correspondingly, if the trajectory is significantly different, it is necessary to reset the switching ratio coefficient for the node and repeat the above process until the trajectory is approximately coincident with the target trajectory, and obtain k i1 ; By changing the disturbance size and type of the system, we can obtain multiple switching ratio coefficients k of each measurement and control node. i1 , k i2 , k i3 …, with ΔM i1 , ΔM i2 , ΔM i3 …one-to-one correspondence; Based on this, the corresponding relationship between the switching ratio coefficient of the power support amount of each measurement and control node and the absolute change of the equivalent inertia time coefficient is obtained as follows: k i =a i ×ΔM i +b i (12) k i =(k1k2 ... k i ) T (13) ΔM i =(ΔM1 ΔM2 ... ΔM i ) T (14) a i =(a1a2 ... a i ) (15) b i =(b1b2 ... b i ) T (16) In the formula, k i is the set of switching ratio coefficients of the power support quantity of each measurement and control node; ΔM i is the absolute change set of the equivalent inertia time coefficient of each measurement and control node; a i It is k i With ΔM i The set of slopes of the relationship, b i It is k i With ΔM i The intercept set of the relationship; From the above relationship, when the absolute change of the equivalent inertia time coefficient of a certain measurement and control node i is ΔM i When the corresponding support power switching ratio coefficient k is calculated i for a i ΔM i +b i , used as a prerequisite for subsequent support power online decision-making; Step 4: Online setting of node support power After completing the offline setting of the switching ratio coefficient of each measurement and control node, the online decision-making process of the power control amount of the APC system needs to monitor the actual trajectory of the transient frequency drop of each measurement and control node in real time. When the system is disturbed, under the starting conditions that meet the setting of the node power support amount, the power support shortage of each measurement and control node is calculated according to the actual trajectory of the transient frequency drop, and the power support amount is allocated to complete the mutual coordination between different measurement and control nodes in the sub-area. If there is only one controllable resource in the sub-area, that is, the controllable resources are limited and there is only one measurement and control node, then the measurement and control node provides all the power support. The support power demand ΔP of the measurement and control node is calculated based on the actual trajectory of the transient frequency drop of the measurement and control node. Ti (t), the result is corrected according to the controllable resources of the measurement and control node; if the calculated ΔP Ti (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , it means that the actual adjustable resources of the measurement and control node cannot meet the support power demand of this point; accordingly, in order to maximize the support effect, the actual power support amount of the measurement and control node should be corrected to the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max ; If there are two or more controllable resources in a sub-area, it is necessary to consider the mutual cooperation between the measurement and control nodes in the sub-area. The support power demand ΔP of each measurement and control node is calculated based on the actual trajectory of transient frequency drop of the measurement and control node monitored in real time. Ti (t), according to k in formula (12) i With ΔM i The corresponding relationship between the power support amount of each measurement and control node is obtained, and the power support amount ΔP of each measurement and control node is calculated by formula (8): Ri (t), and modify the result according to the controllable resources of the measurement and control node; if the calculated ΔP Ri (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , it means that the actual adjustable resources of the measurement and control node cannot meet the support power demand of the point; accordingly, in order to maximize the support effect, the actual power support amount of the measurement and control node should be corrected to the upper limit of the node adjustable resource capacity ΔP i max ; Based on this, the actual switching power support amount ΔP of each measurement and control node in the sub-area is obtained Ri (t) are as follows: The number of measurement and control nodes is i=1.2.3...n. When n=1, that is, there is only one measurement and control node, the node does not need to consider coordination and directly provides all power support; Step 5: Allocation of node controllable resource switching For any measurement and control node i in the sub-area, if the measurement and control node has only one controllable resource, namely, an adjustable load, then the power support is entirely provided by cutting off the adjustable load, and the response time of the controllable resource switching action is negligible, and the required power support can be provided instantaneously; if the measurement and control node has only one controllable resource, namely, a directly controlled new energy storage, then the power support is entirely provided by the energy storage discharge, and the response time of the controllable resource switching action needs to be considered; If the measurement and control node has two controllable resources, namely, the direct-controlled new energy storage and the adjustable load, it is necessary to consider the mutual coordination of the two controllable resources, that is, the allocation of the switching ratio of the two resources; among them, the adjustable load action has no delay, which ensures the rapidity of the decision-making action; the direct-controlled new energy storage provides power support to avoid load shedding affecting the power supply to users, thus ensuring the reliability of power supply; Based on the response characteristics of the direct-controlled new energy storage and the adjustable load providing power support, at the initial stage of the switching action, a power support command is simultaneously issued to the adjustable load and the direct-controlled new energy storage of the measurement and control node. After receiving the command, the adjustable load is cut off as required, and its action time is negligible. Since the direct-controlled new energy storage needs response time to reach the maximum discharge power, after a short delay, the energy storage reaches its maximum discharge power and provides power support for the node with this discharge power; The allocation of the switching amount of the two controllable resources depends on the relationship between the actual switching power support of the measurement and control node and the adjustable resource capacity of the measurement and control node. If the actual switching power support amount of the measurement and control node i is ΔP Ri (t) Exceeds the upper limit of the adjustable resource capacity of the measurement and control node ΔP i max , then the actual adjustable resources of the measurement and control node need to be fully put into use, without considering the allocation ratio of the two controllable resources, and at the same time send instructions to the two controllable resources to switch the two adjustable resources; If the adjustable resource capacity of the measurement and control node i is sufficient, the actual power support required by the two controllable resources is adjusted according to the priority of the APC system decision target; If the priority of the decision-making target is rapidity, the switching of controllable resources is mainly based on adjustable loads. When the adjustable load capacity is insufficient, the remaining power support is provided by the direct-controlled new energy storage. That is, the power support required by the two controllable resources is: Where ΔP Li (t) is the power support required for the adjustable load; ΔP Li max is the upper limit of the capacity of the adjustable load; ΔP Si (t) The amount of power support required for direct control of new energy storage; If the priority of the decision-making target is reliability, the switching of controllable resources is mainly based on direct-controlled new energy storage. When the capacity of direct-controlled new energy storage is insufficient, the remaining power support is provided by the adjustable load, that is, the power support required by the two controllable resources is: Where ΔP Si max To directly control the capacity limit of new energy storage; If there is no clear priority decision target, the actual power support amount ΔP of the measured and controlled node i will be Ri (t) and the upper limit of the node's adjustable resource capacity ΔP i max The ratio of the two controllable resources is taken as the switching ratio m i , calculate the switching ratio m of the two controllable resources i for: Then, the power support required by the two controllable resources is obtained from the switching ratio: Based on the power support amount obtained by setting, instructions are issued to the two controllable resources at the same time to switch the two adjustable resources.
2. An active power online emergency decision-making system for counteracting transient frequency drop, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, an online decision method for power control quantity of an automatic power control system as claimed in claim 1 is implemented.
3. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an implementation program for information transmission, and when the program is executed by the processor, the steps of the online emergency decision-making method for active power against transient frequency drop as described in claim 1 are implemented.
