Transient voltage stability control method and terminal based on data mechanism fusion

Abstract

The present invention provides a data-mechanism-integrated transient voltage stability control method and terminal. The method includes: constructing a voltage threshold dataset; building and training a gated recurrent unit network as a threshold enhancement model; real-time monitoring of power and current criteria for triggering; and real-time assessment of transient voltage stability and emergency control. This method enables accurate, rapid, and interpretable transient voltage assessment and control, ensuring safe and efficient operation of actual power grids.

Description

Technical Field

[0001] The present invention belongs to the field of large power grid stability analysis and control applications, and more specifically, relates to a transient voltage stability control method and terminal integrating data and mechanism. Background Art

[0002] In recent years, large-scale power outages caused by voltage collapse have become increasingly frequent worldwide. On the one hand, the rapid growth in electricity demand has increased the density of load centers, pushing the power transmission capacity of the power system to its limits. On the other hand, with the increasing proportion of renewable energy and dynamic loads, the controllability of the power system has continued to decline, and it has frequently operated in the high-risk critical voltage stability zone. As a result, the transient voltage stability of the system after a disturbance has become increasingly prominent.

[0003] Currently, relatively mature transient voltage stability assessment methods include time-domain simulation, energy function methods, and PV plane methods. These methods clearly explain instability phenomena but are highly dependent on system model parameters. However, the complexity of modern power systems makes it difficult to establish detailed physical models. This makes online assessment of system transient voltage stability difficult, and there is an urgent need for an accurate, fast, and interpretable real-time assessment and control method.

[0004] With the widespread deployment of phasor measurement units, a large number of model-free methods have emerged, such as the Lyapunov exponent method and machine learning methods. Although these methods have good accuracy, their evaluation timeliness and interpretability are poor, making them difficult to promote and apply in actual power grids.

[0005] Therefore, there is an urgent need to establish an accurate, fast, and interpretable data mechanism fusion transient voltage stability control technology solution for transient voltage assessment and control to ensure the safe and efficient operation of the actual power grid. Summary of the Invention

[0006] In light of this, the present invention proposes a transient voltage stability control method and terminal that integrates data mechanisms, aiming to address technical issues related to accuracy, timeliness, and interpretability that plague existing technologies. This method, which utilizes only measurement data to achieve real-time assessment of transient voltage stability, demonstrates rigorous logic, convenience, and speed, facilitating analysis of transient voltage safety and stability characteristics in AC / DC hybrid power grids, ensuring the safe and efficient operation of actual power grids.

[0007] In a first aspect, the present application provides a data-mechanism fusion transient voltage stability control method, comprising:

[0008] Construct voltage threshold dataset;

[0009] Build and train a gated recurrent unit network as a threshold enhancement model;

[0010] Real-time monitoring of whether the power and current criteria are triggered;

[0011] Real-time assessment of transient voltage stability and emergency control.

[0012] Furthermore, the stable samples and unstable samples respectively correspond to input features;

[0013] The voltage amplitude and phase angle of the monitoring node are selected as input features;

[0014] The voltage amplitude and phase angle of the monitoring node before the fault occurs, at the time of fault occurrence, at the time of fault removal, at the start and end time of the first criterion triggering or at the start and end time of more than one criterion triggering are selected as input features.

[0015] Furthermore, when training a gated recurrent unit network as a threshold enhancement model,

[0016] Gated recurrent unit networks use reset gates and update gates to control the flow of information;

[0017] Based on the missed detection rate, false positive rate, accuracy, and average response time of the test model, the hyperparameters are repeatedly adjusted until the evaluation performance reaches the optimal level.

[0018] Furthermore, the real-time monitoring of whether the power current criterion is triggered includes:

[0019] Once a fault is detected in real time and cleared, the active power P, apparent power S, and current I are monitored to see whether the following criteria 1 and 2 are met:

[0020] I k+1 -I k >ε&S k+1 -S k >ε

[0021] I k+1 -I k >ε&P k+1 -P k >ε

[0022] The variable ε ranges from 0.00001 to 0.0002; k is the current moment, and k+1 is the next moment;

[0023] When the voltage continuous decrease process determined by the above criterion 1 or criterion 2 is greater than or equal to 0.05s, the trigger power current criterion is determined.

[0024] Furthermore, the real-time assessment of transient voltage stability and emergency control includes:

[0025] After the power and current criterion is triggered for the first time, the measured data is preprocessed to form input features and then fed into the trained gated recurrent unit network as a threshold enhancement model.

[0026] The trained gated recurrent unit network as a threshold enhancement model quickly predicts the voltage threshold based on the learned mapping relationship;

[0027] Compare the operating voltage value after the first judgment is completed with the predicted voltage threshold;

[0028] If the operating voltage value is less than the predicted voltage threshold, it is determined that the system is unstable; otherwise, it is determined that the system is not unstable.

[0029] Furthermore, the real-time transient voltage assessment and emergency control includes:

[0030] Each time the power and current criterion is triggered, the measured data is preprocessed to form input features and then fed into the trained gated recurrent unit network as a threshold enhancement model.

[0031] The trained gated recurrent unit network as a threshold enhancement model quickly predicts the voltage threshold based on the learned mapping relationship;

[0032] Compare the operating voltage value after each judgment with the predicted voltage threshold;

[0033] If the operating voltage value is less than the predicted voltage threshold, it is determined that the system is unstable; otherwise, it is determined that the system is not unstable.

[0034] Furthermore, once a fault is detected in real time and removed, the power current criterion is continuously monitored in real time to see if it is triggered within the set maximum time of 10s.

[0035] At least once, it is determined by the above criterion 1 or criterion 2 that the voltage drop process lasts longer than or equal to 0.05s and the triggering power current criterion is determined.

[0036] Furthermore, it also includes:

[0037] Once the system is identified as unstable, emergency control measures are immediately implemented to reduce the risk of instability;

[0038] The emergency control measures include: load shedding and DC power regulation.

[0039] In a second aspect, the present application provides a terminal, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0040] In a third aspect, the present application provides a computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the method described in the first aspect.

[0041] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0043] Figure 1 A flow chart of a transient voltage stability control method with data mechanism fusion according to an embodiment of the present application;

[0044] Figure 2 This is a flow chart of a transient voltage stability control method using data mechanism fusion according to another embodiment of the present application;

[0045] Figure 3 Schematic diagram of the CSEE-VC-88 system topology used in the embodiments of the present application;

[0046] Figure 4 for Figure 3 The label distribution diagram of the CSEE-VC-88 example is shown;

[0047] Figure 5 A schematic diagram of the composition of a terminal that applies the transient voltage stabilization control method using data mechanism fusion according to an embodiment of the present application. DETAILED DESCRIPTION

[0048] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other unless they conflict.

[0049] Transient voltage indicators designed based on voltage collapse phenomena can determine system stability by simply comparing the real-time calculated indicator value with a fixed threshold set by human experience. However, they lack a reasonable activation mechanism, and their accuracy is highly dependent on the selection of the prediction threshold.

[0050] Unlike other metrics, the power-current criterion uses an increase in busbar current and a decrease in power after a fault as its triggering mechanism. Once this mechanism is triggered, it indicates that the system is in an unstable region, and a fixed threshold is used to determine stability. However, the choice of fixed threshold still affects the accuracy of this criterion to a certain extent, leaving it with significant potential for improvement.

[0051] The technical solution provided in this application takes power current criterion as the leading method and uses deep learning technology to enhance the selection of its voltage threshold, thereby establishing a complementary relationship between data-driven and knowledge-driven, and forming a transient voltage assessment step that is both accurate, timely and interpretable.

[0052] The technical solution provided in this application also discloses an intelligent enhancement method for improving the voltage support strength index.

[0053] like Figure 1 As shown, the transient voltage stability control method based on data mechanism fusion according to the embodiment of the present application includes the following steps:

[0054] S100: constructing a voltage threshold data set;

[0055] S200: Build and train a gated recurrent unit network as a threshold enhancement model;

[0056] S300: Real-time monitoring of power and current criteria;

[0057] S400: Real-time assessment of transient voltage stability and emergency control.

[0058] In this way, transient voltage assessment and control can be achieved accurately, quickly and explainably to ensure the safe and efficient operation of the actual power grid.

[0059] In some embodiments, step S100 constructs a voltage threshold data set, i.e. Figure 2 For the sample set shown, the dataset consists of input features and labels. The principles for selecting input features are shown in 1A to 1B, and the label calculation process is shown in 2A to 2E.

[0060] In some embodiments, the principles for selecting input features are as follows 1A to 1B:

[0061] 1A. Considering the scalability of features, the cost of model training, and the configuration cost of the phasor measurement unit, a high feature dimension will reduce the practicality of the transient voltage assessment step. Therefore, the feature dimension cannot be larger than the feature dimension used in the criterion. The voltage amplitude and phase angle of the monitoring node are selected as input features;

[0062] 1B. Considering the response speed of the power and current criteria, the model needs to provide a predicted threshold before the first criteria trigger. Therefore, the input features consist of the node voltage amplitudes and phase angles before the fault occurs, at the moment of fault occurrence, at the moment of fault removal, and at the start and end times of the first criteria trigger. These sampling points represent the steady-state power flow level, the degree of fault impact, the degree of post-fault voltage recovery, and the degree of voltage drop after the initial entry into the voltage instability region.

[0063] In some embodiments, the tag calculation process is as follows 2A to 2E:

[0064] 2A. After the fault is cleared, the power and current change relationship of the monitoring point is started to be sliding windowed;

[0065] 2B. When a certain monitoring point triggers the judgment, record the line number and the operating voltage value at the time when the judgment ends;

[0066] 2C. When the time domain simulation (TDS) ends, if the voltage recovers to above 0.80 pu within 10 s, the system transient voltage is considered stable and the classification label is set to 1. Otherwise, it is considered unstable and the classification label is set to 0.

[0067] 2D,At the same time, the dimension of the operating voltage set in the simulation process is changed to a 3D array, where the dimensions are the number of samples, the line number, and the number of times the power and current criterion is triggered.

[0068] 2E), the minimum and maximum values ​​of the array are taken as threshold labels for stable samples and unstable samples respectively.

[0069] In some embodiments, as Figure 2 As shown, when step S200 constructs and trains the gated recurrent unit network as a threshold enhancement model,

[0070] The gated recurrent unit (GRU) uses two gated units to control the flow of information: the reset gate and the update gate. The reset gate determines the influence of the previous hidden state on the current hidden state, and the update gate determines the influence of the current hidden state on the next hidden state. At time t, the input of the GRU is x t , the hidden state of the previous input is h t-1 .

[0071] As shown in equations (1) to (3), first, GRU calculates the reset gate r t and update gate z t , then r t and h t-1 Calculate candidate hidden states Finally, according to z t 、ht-1 and Calculate the current hidden state h t , the calculation formulas are as follows:

[0072] z t =σ(W z [h t-1 ,x t ]+b z ) (1)

[0073] r t =σ(W r [h t-1 ,x t ]+b r ) (2)

[0074]

[0075] Where: W r 、W z and W h Reset gate, update gate and current hidden state respectively The corresponding weight matrix; b r 、b z and b h Reset gate, update gate and current hidden state respectively The corresponding bias matrix; σ(·) is the sigmoid activation function. Finally, the hidden state h t To transmit information between units.

[0076] When training a gated recurrent unit network as a threshold enhancement model, set hyperparameters such as learning rate, batch size, dropout, penalty factor, and number of network layers.

[0077] According to the above calculation formulas (1) to (3), the GRU is trained offline on the training set to establish the mapping relationship between the electrical quantity and the threshold. In this way, the trained gated recurrent unit network serves as a threshold enhancement model and learns the mapping relationship between the electrical quantity and the threshold. The dynamic voltage threshold dataset generated in step S100 is divided into a training set and a test set.

[0078] Finally, based on the missing alarm rate (MAR), false alarm rate (FAR), accuracy (ACC), and average response time (ART) of the test model, the hyperparameters are repeatedly adjusted until the evaluation performance reaches the optimal level.

[0079] In some embodiments, when step 300 monitors the power and current criteria in real time,

[0080] like Figure 4 As shown in Figure 2, according to the observability of the power grid, the monitoring scope can be divided into full-network measurement or local measurement.

[0081] In online applications, once a fault occurs and is cleared, the active power P, apparent power S, and current I at the monitoring point are monitored to see whether the following criteria 1 and 2 are met:

[0082] I k+1 -I k >ε&S k+1 -S k >ε (5)

[0083] I k+1 -I k >ε&P k+1 -P k >ε (6)

[0084] Where, the variable ε ranges from 0.00001 to 0.0002; k is the current moment, and k+1 is the next moment;

[0085] In some embodiments, the variable ε is a very small positive number, such as 0.0001, which is used to avoid misjudgment caused by calculation errors.

[0086] When the voltage continuously decreasing process determined by the above criterion 1 or criterion 2 is greater than or equal to 0.05s, it is determined that the power current criterion is triggered and the system is in the voltage continuously decreasing process.

[0087] In this way, once a fault is detected in real time and the fault is cleared, within the set maximum time of 10s (10s can have a maximum of 200 0.05s), it may be detected multiple times that the power current criterion is triggered and the system is in a process of continuous voltage reduction.

[0088] In the above, the voltage continuous reduction process determined by criterion 1 or criterion 2 should be greater than or equal to 0.05s, which can filter out interference points that may be caused by switching actions or data acquisition errors.

[0089] Once the voltage continues to decrease and triggers the power current criterion, the process proceeds to step S400.

[0090] In some embodiments, as Figure 2 As shown, step S400 is to evaluate transient voltage stability and emergency control in real time, including:

[0091] 3A. After the first triggering criterion, the measured data is preprocessed to form input features and fed into the model. The model quickly predicts the voltage threshold based on the learned mapping relationship.

[0092] 3B. Compare the operating voltage value after each judgment with the prediction threshold. If the operating voltage value U ei Less than the prediction threshold U r , the system is deemed to be unstable; otherwise, the system is deemed not to be unstable and the monitoring is continued to the set maximum time of 10s.

[0093] Figure 2 In, t si and t ei Respectively represent the start time and end time of the i-th criterion triggering; U ei U is the operating voltage of the monitoring point at the end of the judgment; r is the voltage threshold predicted by GRU, where i is a natural number greater than or equal to 1 and less than 300.

[0094] 3C. Once instability occurs, immediately implement emergency control measures such as load shedding and DC power regulation to reduce the risk of instability.

[0095] The data-mechanism-integrated transient voltage stability control method described above in this application embodiment provides an intelligent enhanced transient voltage criterion that integrates power and current criteria with a GRU. Compared to traditional machine learning methods, this method is a white-box model that integrates physical mechanisms and offers better interpretability. Compared to existing mechanism-based identification methods, the proposed method requires only real-time measurements of key sections to accurately identify transient voltage stability, demonstrating promising engineering application prospects.

[0096] In some embodiments, the stable samples and the unstable samples respectively correspond to input features;

[0097] The voltage amplitude and phase angle of the monitoring node are selected as input features;

[0098] The voltage amplitude and phase angle of the monitoring node before the fault occurs, at the time of fault occurrence, at the time of fault removal, at the start and end time of the first criterion triggering or at the start and end time of more than one criterion triggering are selected as input features.

[0099] In this way, the voltage amplitude and phase angle of any monitoring node before the fault occurs, at the time of fault occurrence, at the time of fault removal, at the start and end time of the first criterion triggering, or at the start and end time of more criterion triggering form a voltage amplitude array and a phase angle array.

[0100] In this way, the minimum value and / or maximum value of the voltage amplitude array of the stable sample and the unstable sample can be respectively taken as the threshold label.

[0101] Specifically, a dynamic voltage threshold sample set can be generated by using a fault set through time domain simulation, which will not be described in detail here.

[0102] In some embodiments, the real-time monitoring of whether the power current criterion is triggered includes:

[0103] Once a fault is detected in real time and cleared, the active power P, apparent power S, and current I are monitored to see whether the following criteria 1 and 2 are met:

[0104] I k+1 -I k >ε&S k+1 -S k >ε

[0105] I k+1 -I k >ε&P k+1 -P k >ε

[0106] The variable ε ranges from 0.00001 to 0.0002; k is the current moment, and k+1 is the next moment;

[0107] When the voltage continuous decrease process determined by the above criterion 1 or criterion 2 is greater than or equal to 0.05s, the trigger power current criterion is determined.

[0108] In this way, once a fault is detected in real time and the fault is cleared, within the set maximum time of 10s (10s can have a maximum of 200 0.05s), it may be detected multiple times that the power current criterion is triggered and the system is in a process of continuous voltage reduction.

[0109] In some embodiments, as Figure 2 As shown, the real-time assessment of transient voltage stability and emergency control includes:

[0110] After the power and current criterion is triggered for the first time, the measured data is preprocessed to form input features and then fed into the trained gated recurrent unit network as a threshold enhancement model.

[0111] The trained gated recurrent unit network as a threshold enhancement model quickly predicts the voltage threshold based on the learned mapping relationship;

[0112] Compare the operating voltage value after the first judgment is completed with the predicted voltage threshold;

[0113] If the operating voltage value is less than the predicted voltage threshold, it is determined that the system is unstable; otherwise, it is determined that the system is not unstable.

[0114] In some embodiments, as Figure 2 As shown, the real-time assessment of transient voltage stability and emergency control includes:

[0115] Each time (or more times as described above) the power and current criterion is triggered, the measured data is preprocessed to form input features and input into the trained gated recurrent unit network as a threshold enhancement model;

[0116] The trained gated recurrent unit network as a threshold enhancement model quickly predicts the voltage threshold based on the learned mapping relationship;

[0117] Compare the operating voltage value after each judgment with the predicted voltage threshold;

[0118] If the operating voltage value is less than the predicted voltage threshold, it is determined that the system is unstable; otherwise, it is determined that the system is not unstable.

[0119] In this way, the voltage amplitude and phase angle of the monitoring node before the fault occurs, at the time of fault occurrence, at the time of fault removal, at the start and end times of the first criterion triggering, or at the start and end times of more criterion triggering are used as input features, and the trained voltage threshold enhancement model based on the penalty loss function quickly predicts the voltage threshold according to the learned mapping relationship, thereby realizing the evaluation of transient voltage and emergency control.

[0120] In some embodiments, once a fault is detected in real time and removed, the power current criterion is continuously monitored in real time to see if it is triggered within a set maximum time of 10s. Accordingly,

[0121] At least once, it is determined by the above criterion 1 or criterion 2 that the voltage drop process lasts longer than or equal to 0.05s and the triggering power current criterion is determined.

[0122] In this way, real-time evaluation of transient voltage and emergency control are achieved.

[0123] In some embodiments, further comprising:

[0124] Once the system is identified as unstable, emergency control measures are immediately implemented to reduce the risk of instability;

[0125] The emergency control measures include: load shedding and DC power regulation.

[0126] In this way, accurate evaluation of transient voltage and emergency control are achieved.

[0127] In some embodiments, further comprising:

[0128] Once the system is deemed unstable, emergency control measures will be implemented after a preset delay (no more than the aforementioned 0.05s and less than the aforementioned 10s) to reduce the risk of instability;

[0129] The emergency control measures include: load shedding and DC power regulation.

[0130] In this way, accurate evaluation of transient voltage and emergency control are achieved.

[0131] The following is combined with Figure 3 and Figure 4 , taking the CSEE-VC-88 node system as the test system, the transient voltage stability control method of the data mechanism fusion embodiment of the present application is further explained.

[0132] Specifically, for Figure 3 For the CSEE-VC-88 bus system shown in the figure, various fault conditions were set to generate a dataset related to transient voltage stability. A total of 11,826 samples were generated, of which 7,609 were stable samples and 4,217 were unstable samples. 80% of the dataset was divided into a training set, and the remaining 20% ​​was a test set.

[0133] The monitoring scope is divided into full network measurement and local measurement. Full network measurement represents 60 525KV lines, and local measurement represents 10 key lines. Figure 4 shown.

[0134] To demonstrate the enhancement effect, the original criterion, data-driven method and transient voltage stability control method integrated with the data mechanism of the embodiment of the application are compared. Figure 4 As shown in Figure 1, the original criterion uses 0.75 pu as the power and current criterion for the voltage threshold. The data-driven method directly uses the GRU to perform binary classification of transients. Three methods were tested on the test set, and the results are shown in Table 1. The ART for the data-driven method is calculated based on the end time of the first criterion for the unstable sample.

[0135] Table 1 Comparison of intelligent enhancement effects

[0136]

[0137]

[0138] As shown in Table 1, the conservative value of 0.75 pu results in a low evaluation performance for the original criterion. When the monitoring scope is global measurement, the MAR is as high as 24.09%, and the ACC is as low as 84.38%. This indicates that in many stable scenarios, the operating voltage value at the end of the criterion falls below 0.75 pu and then returns to normal. When the monitoring scope is narrowed to local measurement, the FAR drops to 14.51%. This is because the reduction in the monitoring scope increases the overall threshold distribution of stable samples, reducing the number of false positives. However, the evaluation performance of the original criterion still fails to meet the requirements.

[0139] Compared to the original criteria, the data-driven approach achieves higher evaluation performance. However, class imbalance causes the GRU to overfit stable samples (class imbalance refers to the number of stable samples being significantly greater than the number of unstable samples), resulting in a MAR significantly higher than the FAR. However, misclassification of stable samples can lead to erroneous control actions, while missed detection of unstable samples can seriously impact the system's stable operation. Therefore, the misclassification costs of the two classes of samples differ, and the high MAR makes the data-driven approach unsatisfactory for evaluation.

[0140] Compared to the data-driven approach, the enhanced criterion achieves a MAR below 0.6% for the following reasons: 1) In the proposed method, the GRU only predicts the voltage threshold, which is unaffected by class imbalance; and 2) unstable samples have a high tolerance for threshold errors, making misclassification less likely. Compared to the original criterion, the enhanced criterion achieves a significantly lower FAR and maintains an ACC above 99%. Furthermore, the enhanced criterion is less affected by the monitoring range, maintaining high evaluation performance even when the input information is limited to local measurements. These test results demonstrate a significant enhancement effect, validating the effectiveness of the proposed method.

[0141] like Figure 5 As shown, the embodiment of the present application also provides a terminal to execute the method described above. Figure 5 FIG2 shows a schematic diagram of a terminal 8 provided in some embodiments of the present application. Figure 5 As shown, the terminal 8 includes: a processor 800, a memory 801, a bus 802 and a communication interface 803, and the processor 800, the communication interface 803 and the memory 801 are connected via the bus 802; the memory 801 stores a computer program that can be run on the processor 800, and the processor 800 executes the method described in any of the aforementioned embodiments of the present application when running the computer program.

[0142] The memory 801 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage. The communication connection between the device network element and at least one other network element is achieved through at least one communication interface 803 (which may be wired or wireless), and may use the Internet, a wide area network, a local area network, a metropolitan area network, etc.

[0143] The bus 802 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 801 is used to store programs. The processor 800 executes the programs upon receiving execution instructions. The method disclosed in any of the aforementioned embodiments of the present invention may be applied to the processor 800 or implemented by the processor 800.

[0144] The processor 800 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method may be performed by hardware integrated logic circuits or software instructions within the processor 800. The processor 800 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It may implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present invention may be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules within the decoding processor. The software modules may be located in storage media well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or the like. The storage medium is located in the memory 801 , and the processor 800 reads the information in the memory 801 and completes the steps of the above method in combination with its hardware.

[0145] The terminal provided by the embodiment of the present invention and the method provided by the embodiment of the present invention are based on the same inventive concept and have the same beneficial effects as the method adopted, operated or implemented by them.

[0146] An embodiment of the present application also provides a computer-readable storage medium corresponding to the method described in the aforementioned embodiment. The computer-readable storage medium is a CD-ROM on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the method described in any of the aforementioned embodiments.

[0147] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0148] The computer-readable storage medium provided by the above-mentioned embodiment of the present application is based on the same inventive concept as the method of the embodiment of the present invention, and has the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0149] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A transient voltage stability control method based on data mechanism fusion, characterized in that: include: Construct voltage threshold dataset; Build and train a gated recurrent unit network as a threshold enhancement model; Real-time monitoring of whether the power and current criteria are triggered; Real-time assessment of transient voltage stability and emergency control; The real-time monitoring of whether the power current criterion is triggered includes: Once a fault is detected in real time and cleared, the active power P, apparent power S, and current I are monitored to see whether the following criteria 1 and 2 are met: I k+1 -I k >e&S k+1 -S k >eI k+1 -I k >e&P k+1 -P k >e The variable ε ranges from 0.00001 to 0.0002; k is the current moment, and k+1 is the next moment; When the voltage continues to decrease for more than or equal to 0.05s as determined by the above criteria 1 or 2, the trigger power current criterion is determined; The real-time assessment of transient voltage stability and emergency control includes: After the power and current criterion is triggered for the first time, the measured data is preprocessed to form input features and then fed into the trained gated recurrent unit network as a threshold enhancement model. The trained gated recurrent unit network as a threshold enhancement model quickly predicts the voltage threshold based on the learned mapping relationship; Compare the operating voltage value after the first judgment is completed with the predicted voltage threshold; If the operating voltage value is less than the predicted voltage threshold, it is determined that the system is unstable; otherwise, it is determined that the system is not unstable.

2. The transient voltage stability control method based on data-mechanism fusion according to claim 1, characterized in that: The stable samples and unstable samples respectively correspond to input features; The voltage amplitude and phase angle of the monitoring node are selected as input features; The voltage amplitude and phase angle of the monitoring node before the fault occurs, at the time of fault occurrence, at the time of fault removal, at the start and end time of the first criterion triggering or at the start and end time of more than one criterion triggering are selected as input features.

3. The transient voltage stability control method based on data-mechanism fusion according to claim 1, characterized in that: When training a gated recurrent unit network as a threshold-enhanced model, Gated recurrent unit networks use reset gates and update gates to control the flow of information; Based on the missed detection rate, false positive rate, accuracy, and average response time of the test model, the hyperparameters are repeatedly adjusted until the evaluation performance reaches the optimal level.

4. The transient voltage stability control method based on data mechanism fusion according to claim 1, characterized in that: The real-time transient voltage assessment and emergency control include: Each time the power and current criterion is triggered, the measured data is preprocessed to form input features and then fed into the trained gated recurrent unit network as a threshold enhancement model. The trained gated recurrent unit network as a threshold enhancement model quickly predicts the voltage threshold based on the learned mapping relationship; Compare the operating voltage value after each judgment with the predicted voltage threshold; If the operating voltage value is less than the predicted voltage threshold, it is determined that the system is unstable; otherwise, it is determined that the system is not unstable.

5. The transient voltage stability control method based on data-mechanism fusion according to claim 1, characterized in that: Once a fault is detected in real time and removed, the power current criterion is continuously monitored in real time to see if it is triggered within the set maximum time of 10s. At least once, it is determined by the above criterion 1 or criterion 2 that the voltage drop process lasts longer than or equal to 0.05s and the triggering power current criterion is determined.

6. The transient voltage stability control method based on data mechanism fusion according to claim 1 or 4, characterized in that: Also includes: Once the system is identified as unstable, emergency control measures are immediately implemented to reduce the risk of instability; The emergency control measures include: load shedding and DC power regulation.

7. A terminal comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 6.

8. A computer storage medium, characterized in that Computer-executable instructions are stored, and the computer-executable instructions are used to execute the method according to any one of claims 1 to 6.

Images