A Power Grid Early Warning and Response Method and System Based on Data Fusion

Abstract

This application provides a power grid early warning response method and system based on data fusion, relating to the field of power grid risk management technology. The method includes: acquiring the current cycle operating data of a first power generation system and establishing a first input feature; acquiring the current cycle operating data of a second power generation system and constructing a second input feature; determining the mutual influence factors between the first and second power generation systems; modifying the first input feature based on the mutual influence factors to obtain a first modified input feature; modifying the second input feature based on the mutual influence factors to obtain a second modified input feature; fusing the first and second modified input features using multi-source features to obtain a target input feature; and inputting the target input feature into a pre-trained target power grid early warning model to obtain a power grid risk early warning result. This application aims to improve the poor performance of existing methods for power grid risk early warning when multiple power generation systems are simultaneously coupled.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of power grid risk management, and in particular to a power grid early warning response method and system based on data fusion. BACKGROUND

[0002] As an important infrastructure supporting the operation of national economy and social life, the operation state of the power grid is affected by many factors such as load fluctuation, equipment aging, weather change and new energy grid connection. Once an abnormality occurs, it may cause voltage collapse, frequency deviation or cascading failure, resulting in large-scale power outages, serious economic losses and social impact. Through power grid early warning, potential risks can be discovered and measures can be taken in advance before the risks evolve into accidents, which helps to ensure the safe and stable operation of the power system, prolong the service life of equipment, reduce the cost of accident disposal, and improve the adaptability to new energy fluctuations, ensuring the continuity and reliability of power supply.

[0003] However, for a power grid system simultaneously connected to multiple power generation systems, the current early warning method has low adaptability and prediction accuracy, making early warning difficult. SUMMARY

[0004] Embodiments of the present application provide a power grid early warning response method based on data fusion. Embodiments of the present application adopt the following technical solutions:

[0005] In a first aspect, the present application provides a power grid early warning response method based on data fusion, the method comprising:

[0006] obtaining operation data of a first power generation system in a current period and constructing a first input feature according to the operation data of the first power generation system in the current period, and obtaining operation data of a second power generation system in the current period and constructing a second input feature according to the operation data of the first power generation system in the current period;

[0007] determining a mutual influence factor of the first power generation system and the second power generation system;

[0008] correcting the first input feature according to the mutual influence factor to obtain a first corrected input feature, and correcting the second input feature according to the mutual influence factor to obtain a second corrected input feature;

[0009] performing multi-source feature fusion on the first corrected input feature and the second corrected input feature to obtain a target input feature, and inputting the target input feature into a pre-trained target power grid early warning model to obtain a power grid risk early warning result.

[0010] The application provides a power grid early warning response method based on data fusion. First, the operation data of a first power generation system and a second power generation system in a current period is acquired, and corresponding first input features and second input features are constructed to realize preliminary quantitative description of the operation states of the power generation systems. Then, the mutual influence factors between the two types of power generation systems are determined, which comprehensively reflect the coupling relationship of the power generation systems in terms of power output characteristics and operation stability, thereby revealing the dynamic interaction between different types of power generation systems. Based on the mutual influence factors, the first input features and the second input features are corrected, and first corrected input features and second corrected input features are obtained, so that the input features can more truly and accurately reflect the mutual restriction and synergistic effect between the power generation systems, and eliminate the feature isolation problem under a single system perspective. Then, the multi-source feature fusion technology is used to deeply fuse the corrected input features to form target input features containing multi-dimensional and multi-level operation information, thereby improving the richness and relevance of feature expression. Finally, the target input features are input into a pre-trained target power grid early warning model, and the learning ability of the model for historical operation data and fault cases is relied on to realize accurate identification and early warning of potential risks of the power grid. Through the method, the capturing ability of the early warning model for the complex coupling relationship of multiple types of power generation systems is improved, the accuracy and timeliness of power grid risk identification are enhanced, the accident risks caused by new energy fluctuations or equipment abnormalities are effectively reduced, the safety, stability and economy of power grid operation are improved, and the efficient management and sustainable development of smart grid are promoted.

[0011] In an optional implementation, the mutual influence factors of the first power generation system and the second power generation system are determined, including:

[0012] According to the coupling relationship of the first power generation system and the second power generation system in terms of power output characteristics, a first mutual influence factor of the first power generation system and the second power generation system is determined;

[0013] According to the coupling relationship of the first power generation system and the second power generation system in terms of power output characteristics, a first mutual influence factor of the first power generation system and the second power generation system is determined;

[0014] In an optional implementation, according to the coupling relationship of the first power generation system and the second power generation system in terms of power output characteristics, a first mutual influence factor of the first power generation system and the second power generation system is determined, including:

[0015] The output power data of the first power generation system in the Nth period and the output power data of the second power generation system in the Nth period are acquired, and according to the output power data of the first power generation system in the Nth period and the output power data of the second power generation system in the Nth period, a power influence factor of the first power generation system and the second power generation system in the time synchronization dimension is determined;

[0016] obtaining output power data of the first power generation system in the Nth period and output power data of the second power generation system in the N+1th period, and determining the power influence factor of the first power generation system and the second power generation system in the time lag dimension according to the output power data of the first power generation system in the Nth period and the output power data of the second power generation system in the N+1th period;

[0017] determining the first mutual influence factor according to the power influence factor in the time synchronization dimension and the power influence factor in the time lag dimension.

[0018] In an optional implementation, the second mutual influence factor of the first power generation system and the second power generation system is determined according to the correlation between the first power generation system and the second power generation system in the operation stability, and the second mutual influence factor comprises:

[0019] obtaining operation stability data of the first power generation system in the Nth period and operation stability data of the second power generation system in the Nth period, and determining the stability influence factor of the first power generation system and the second power generation system in the time synchronization dimension according to the operation stability data of the first power generation system in the Nth period and the operation stability data of the second power generation system in the Nth period;

[0020] obtaining operation stability data of the first power generation system in the Nth period and operation stability data of the second power generation system in the N+1th period, and determining the stability influence factor of the first power generation system and the second power generation system in the time lag dimension according to the operation stability data of the first power generation system in the Nth period and the operation stability data of the second power generation system in the N+1th period;

[0021] determining the second mutual influence factor according to the stability influence factor in the time synchronization dimension and the stability influence factor in the time lag dimension.

[0022] In an optional implementation, the first input feature is corrected according to the mutual influence factor to obtain a first corrected input feature, and the second input feature is corrected according to the mutual influence factor to obtain a second corrected input feature, and the method comprises:

[0023] correcting the first input feature according to the first mutual influence factor and the second mutual influence factor to obtain a first corrected input feature;

[0024] correcting the second input feature according to the first mutual influence factor and the second mutual influence factor to obtain a second corrected input feature.

[0025] In an optional implementation, the power grid early warning model is obtained by the following steps:

[0026] obtaining operation data of the first power generation system in a historical period, and determining the power grid early warning model of the first power generation system according to the operation data of the first power generation system in the historical period;

[0027] Obtain historical cycle operating data of the second power generation system, and determine the power grid early warning model of the second power generation system based on the historical cycle operating data of the second power generation system;

[0028] The influence weight ratios of the first power generation system and the second power generation system are determined. Based on the influence weight ratios, the power grid early warning models of the first power generation system and the second power grid early warning models are weighted and fused to obtain the target power grid early warning model.

[0029] In one optional implementation, determining the influence weight ratio of the first power generation system and the second power generation system includes:

[0030] Obtain the impact weight assessment indicators of the first and second power generation systems within the historical period, and determine the initial comprehensive weight based on the impact weight assessment indicators. The impact weight assessment indicators include at least the average output power, the operational stability score, and historical failure rate data.

[0031] Obtain the power generation contribution of the first power generation system and the second power generation system, and adjust the initial comprehensive weight according to the power generation contribution to obtain the adjusted comprehensive weight;

[0032] The influence weight ratio of the first power generation system and the second power generation system is calculated based on the revised comprehensive weight.

[0033] Secondly, embodiments of this application provide a power grid early warning and response system based on data fusion, the system comprising:

[0034] The acquisition module is used to acquire the operating data of the first power generation system in the current cycle, and construct a first input feature based on the operating data of the first power generation system in the current cycle; acquire the operating data of the second power generation system in the current cycle, and construct a second input feature based on the operating data of the first power generation system in the current cycle.

[0035] The determination module is used to determine the mutual influence factors between the first power generation system and the second power generation system.

[0036] The correction module is used to correct the first input feature according to the mutual influence factor to obtain the first corrected input feature, and to correct the second input feature according to the mutual influence factor to obtain the second corrected input feature.

[0037] The early warning module is used to perform multi-source feature fusion on the first and second corrected input features to obtain the target input features, and then input the target input features into the pre-trained target power grid early warning model to obtain the power grid risk early warning result.

[0038] In one alternative implementation, the determining module includes:

[0039] The first determining submodule is used to determine the first mutual influence factor of the first power generation system and the second power generation system based on the coupling relationship between the first power generation system and the second power generation system in terms of power output characteristics.

[0040] The second determining submodule is used to determine the second mutual influence factor of the first power generation system and the second power generation system based on the correlation between the first power generation system and the second power generation system in terms of operational stability.

[0041] In one alternative implementation, the first determining submodule includes:

[0042] The first influence factor determination unit is used to obtain the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the Nth cycle, and determine the mutual influence factor of the first power generation system and the second power generation system in the time synchronization dimension based on the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the Nth cycle.

[0043] The second influencing factor determination unit is used to obtain the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the N+1th cycle, and to determine the mutual influence factor of the first power generation system and the second power generation system in the time lag dimension based on the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the N+1th cycle.

[0044] The third influence factor determination unit is used to determine the first mutual influence factor based on the mutual influence factors in the time synchronization dimension and the mutual influence factors in the time lag dimension.

[0045] Thirdly, this application also provides an electronic device, which includes: a memory and one or more processors, the memory being coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, which, when executed by the processor, cause the electronic device to perform the method in any of the possible design embodiments of the first aspect described above.

[0046] Fourthly, this application provides a computer-readable storage medium including computer instructions; when the computer instructions are executed on an electronic device, they cause the electronic device to perform the method described in the first aspect above and any possible design of the above.

[0047] Fifthly, this application provides a computer program product that, when run on an electronic device, causes the electronic device to perform the method described in the first aspect above and any possible design of the above.

[0048] The technical effects of the second to fifth aspects refer to the technical effects of the first aspect and any of its embodiments, and will not be repeated here. Attached Figure Description

[0049] Figure 1 A flowchart illustrating the steps of a power grid early warning and response method based on data fusion, as provided in this application embodiment;

[0050] Figure 2 This is a schematic diagram of the structure of a power grid early warning and response system based on data fusion, provided as an embodiment of this application. Detailed Implementation

[0051] The terminology used in the following embodiments is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” and “this” are intended to also include expressions such as “one or more,” unless the context clearly indicates otherwise. It should also be understood that in the following embodiments of this application, “at least one” and “one or more” refer to one or more (including two). The character “ / ” generally indicates that the preceding and following objects are in an “or” relationship.

[0052] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0053] Hereinafter, the terms "first," "second," etc., are used for descriptive convenience only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "multiple" means two or more. For example, multiple processing units refer to two or more processing units.

[0054] Furthermore, in the embodiments of this application, "upper," "lower," "left," and "right" are not limited to the orientation of the components schematically placed in the accompanying drawings. It should be understood that these directional terms can be relative concepts, used for relative description and clarification, and can change accordingly depending on the orientation of the components in the accompanying drawings. In the accompanying drawings, for clarity, the thickness of layers and regions is exaggerated, and the dimensional proportions between the parts in the drawings do not reflect the actual dimensional proportions.

[0055] In the embodiments of this application, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral part; it can be a direct connection or an indirect connection through an intermediate medium. In addition, the term "electrical connection" can be a direct electrical connection or an indirect electrical connection through an intermediate medium.

[0056] In this application, the term "module" typically refers to a logically divided functional structure. A "module" can be implemented purely in hardware, or a combination of hardware and software. In this application, "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, B existing alone, or both A and B existing simultaneously.

[0057] In this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being better or more advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0058] As a vital infrastructure supporting the national economy and social life, the power grid's operation is affected by various factors such as load fluctuations, equipment aging, weather changes, and the integration of new energy sources. Anomalies can trigger voltage collapse, frequency deviations, or cascading failures, causing widespread power outages and resulting in severe economic losses and social impacts. Power grid early warning systems allow for the detection and intervention of potential risks before they escalate into accidents, helping to ensure the safe and stable operation of the power system, extend equipment lifespan, reduce accident response costs, and enhance adaptability to fluctuations in new energy sources, ensuring the continuity and reliability of power supply.

[0059] However, for power grids that simultaneously connect to multiple different power generation systems, such as those that combine thermal power, hydropower, wind power, and photovoltaic power generation systems, there are certain difficulties in fusing multi-source heterogeneous data due to differences in operating characteristics, data formats, time resolution, and sampling accuracy among the subsystems. Furthermore, different types of power generation systems can influence each other in terms of power output fluctuations, regulation response speed, fault propagation paths, and contributions to grid stability. This interaction further amplifies the uncertainty and complexity of power grid operation. Therefore, existing power grid early warning methods designed for single-generation systems with coupling are difficult to meet the requirements in terms of adaptability and prediction accuracy in such multi-generation system coupling scenarios.

[0060] To address the aforementioned issues, this application proposes the following inventive concept: by quantifying the coupling relationships between different power generation systems and performing feature correction and multi-source fusion on this basis, an overall input feature that can comprehensively reflect the power grid operating status is constructed, thereby improving the adaptability and accuracy of power grid early warning under conditions of multiple power generation system coupling.

[0061] Reference Figure 1 The embodiments of this application provide a power grid early warning response method based on data fusion, which can be applied to power grids simultaneously coupled with multiple different types of power generation systems. Specifically, it may include the following steps:

[0062] S101: Obtain the operating data of the first power generation system in the current cycle, and construct a first input feature based on the operating data of the first power generation system in the current cycle; obtain the operating data of the second power generation system in the current cycle, and construct a second input feature based on the operating data of the first power generation system in the current cycle.

[0063] In this embodiment, the first power generation system can be any of a hydropower system, a thermal power system, a wind power system, or other renewable energy power generation systems. Similarly, the second power generation system can also be any of the aforementioned systems. Furthermore, the first and second power generation systems can be of the same type or different types. After obtaining the current cycle operating data of the first and second power generation systems, the system synchronously collects and preprocesses this data, including data cleaning, anomaly detection, and normalization operations, to ensure the accuracy and consistency of subsequent analysis. Subsequently, based on the processed operating data, feature extraction is performed to obtain the first and second input features. At this point, the first and second input features only reflect the independent operating status and performance indicators of their respective power generation systems within the current cycle, without considering the interaction and coupling relationships between different types of power generation systems. Therefore, these input features are somewhat isolated and cannot comprehensively reflect the overall dynamic operating characteristics and potential risks of the power grid.

[0064] S102: Determine the mutual influence factors of the first power generation system and the second power generation system.

[0065] In this embodiment, the mutual influence factor between the first and second power generation systems refers to a parameter that quantifies the degree of interaction and characteristic differences between the two types of power generation systems during grid-connected operation. The mutual influence factor reflects the coupling relationship between the first and second power generation systems in terms of power output fluctuation patterns, regulation response speed, fault propagation characteristics, power quality impact, and contribution to grid stability. By determining the mutual influence factor, the input feature weights of various power generation systems can be adjusted in a targeted manner during subsequent feature correction, weakening irrelevant or noise effects and enhancing feature components that reflect actual coupling effects. This improves the accuracy of multi-source feature fusion and the adaptability of the grid early warning model to complex coupling scenarios. The specific steps for determining the mutual influence factor between the first and second power generation systems may include:

[0066] S1021: Based on the coupling relationship between the first power generation system and the second power generation system in terms of power output characteristics, determine the first mutual influence factor between the first power generation system and the second power generation system.

[0067] In this embodiment, the first dimension of the mutual influence factor is a first mutual influence factor characterizing the degree of coupling between the first power generation system and the second power generation system in terms of power output characteristics. It is used to reflect the correlation between the two types of power generation systems in terms of power change amplitude, power fluctuation trend, and power regulation response. The specific steps for determining the first mutual influence factor of the first power generation system and the second power generation system may include:

[0068] S10211: Obtain the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the Nth cycle. Based on the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the Nth cycle, determine the power influence factor of the first power generation system and the second power generation system in the time synchronization dimension.

[0069] S10212: Obtain the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the N+1th cycle. Based on the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the N+1th cycle, determine the power influence factor of the first power generation system and the second power generation system in the time lag dimension.

[0070] S10213: Determine the first mutual influence factor based on the power influence factor of the time synchronization dimension and the power influence factor of the time lag dimension.

[0071] In the implementations of S10211 to S10213, firstly, the output power data of the first and second power generation systems in the same statistical period (i.e., the Nth period) are acquired, and their power changes within that period are compared and analyzed. For example, the power change trends, fluctuation amplitudes, and directions of the two power generation systems within that period can be matched. If the peak and trough positions of their changes in the time series are relatively consistent, it indicates that they have a strong power correlation in the time synchronization dimension; otherwise, the correlation is weak. Then, to examine the impact of the power change of one power generation system on the other power generation system in subsequent periods, it is necessary to acquire the output power data of the first power generation system in the Nth period and the output power data of the second power generation system in the N+1th period. By comparing their change trends and response characteristics, it can be reflected whether the power adjustment of the first power generation system in the current period will have a lagging effect on the power of the second power generation system in the next period. For example, if the first power generation system experiences a significant power decrease in the Nth period, and the second power generation system also shows a similar decreasing trend in the N+1th period, it indicates that there is a strong coupling relationship between the two in the time lag dimension. Finally, the analysis results of the time synchronization and time lag dimensions are comprehensively evaluated to obtain the first mutual influence factor, which can simultaneously reflect the instantaneous coupling degree and delayed response relationship of the two types of power generation systems. This approach allows for a more comprehensive characterization of the interaction between the two types of power generation systems in terms of power characteristics, thus providing a basis for optimizing dispatch strategies. Specifically, the information from the two dimensions can be transformed into the first mutual influence factor through weighted fusion, trend fitting, or multidimensional feature mapping. As an example, taking wind power as the first power generation system and thermal power as the second, if the power fluctuations of both systems are highly consistent in the time synchronization dimension, while still exhibiting a significant response correlation in the time lag dimension, the value of the first mutual influence factor will be high, indicating a significant bidirectional coupling relationship between their power outputs. Conversely, if the correlation between the two systems is weak in either dimension, the value of the first mutual influence factor will be relatively low, reflecting a stronger independence in power characteristics.

[0072] S1022: Based on the correlation between the first power generation system and the second power generation system in terms of operational stability, determine the second mutual influence factor between the first power generation system and the second power generation system.

[0073] In this embodiment, the second dimension of the mutual influence factor is a second mutual influence factor that characterizes the degree of correlation between the first power generation system and the second power generation system in terms of operational stability characteristics. It is used to reflect the correlation between the two types of power generation systems in terms of frequency stability, voltage deviation and ramp rate changes. By combining the first mutual influence factor and the second mutual influence factor, the overall operational coupling relationship between the two types of power generation systems can be quantified.

[0074] S10221: Obtain the operational stability data of the first power generation system in the Nth cycle and the operational stability data of the second power generation system in the Nth cycle. Based on the operational stability data of the first power generation system in the Nth cycle and the operational stability data of the second power generation system in the Nth cycle, determine the stability influence factor of the first power generation system and the second power generation system in the time synchronization dimension.

[0075] S10222: Obtain the operational stability data of the first power generation system in the Nth cycle and the operational stability data of the second power generation system in the N+1th cycle. Based on the operational stability data of the first power generation system in the Nth cycle and the operational stability data of the second power generation system in the N+1th cycle, determine the stability influence factors of the first power generation system and the second power generation system in the time lag dimension.

[0076] S10223: Determine the second mutual influence factor based on the stability influence factors in the time synchronization dimension and the time lag dimension.

[0077] In the implementations of S10221 to S10223, key operational stability index data of the first and second power generation systems are first acquired within the same statistical period (i.e., the Nth period), such as frequency deviation, bus voltage offset, and phase angle fluctuation amplitude. The stability change trends of the two systems within this period are then compared and analyzed. If the fluctuation patterns, amplitude changes, and anomaly distributions of the two systems in the time series are highly consistent, it indicates a strong correlation in their operational stability in the time synchronization dimension; otherwise, the correlation is weak. Subsequently, to analyze the impact of the stability changes of one system on the other system in subsequent periods, stability index data of the first power generation system in the Nth period and stability index data of the second power generation system in the N+1th period are acquired, and their change trends are analyzed for correlation. For example, if the first power generation system exhibits significant frequency fluctuations in the Nth period, and the second power generation system also shows similar fluctuation characteristics in frequency or voltage in the N+1th period, it indicates a strong stability coupling between them in the time lag dimension. Finally, the analysis results from the time synchronization and time lag dimensions are combined to obtain a second mutual influence factor that can simultaneously reflect the instantaneous stability correlation and delayed response relationship of the two types of power generation systems. In this way, the interactive characteristics of the two types of power generation systems in terms of operational stability can be more comprehensively characterized, thereby providing data support for the stable operation of the power grid and the optimization of dynamic regulation strategies.

[0078] By identifying the first and second mutual influence factors, the interaction relationships between different power generation systems can be comprehensively characterized across two dimensions: power characteristics and operational stability. The first mutual influence factor reflects the degree of coupling and interdependence of each power generation system at the power output level, revealing their synergistic patterns in power fluctuations, load changes, and complementary characteristics. The second mutual influence factor embodies the linkage effect of each power generation system in terms of operational stability, reflecting the impact of equipment operating status, frequency stability, and response characteristics on other systems. Through joint analysis of these two mutual influence factors, a multi-dimensional quantitative assessment of the coupling relationships between multi-source power generation systems can be achieved, providing more accurate and comprehensive input features for power grid early warning models and improving the adaptability and accuracy of risk prediction.

[0079] S103: Based on the mutual influence factor, the first input feature is modified to obtain the first modified input feature; based on the mutual influence factor, the second input feature is modified to obtain the second modified input feature.

[0080] In this embodiment, after obtaining the mutual influence factors of the first and second power generation systems, the parameters related to power generation output, equipment status, and operational fluctuations in the original input features are weighted and adjusted. This allows the corrected input features to more realistically reflect the mutual constraints and synergistic effects of each power generation system in actual operation, improving the accuracy and adaptability of the features in characterizing grid operation risks. Specific steps may include:

[0081] S1031: Based on the first mutual influence factor and the second mutual influence factor, the first input feature is modified to obtain the first modified input feature;

[0082] S1032: Based on the first mutual influence factor and the second mutual influence factor, the second input feature is modified to obtain the second modified input feature.

[0083] In the implementations of S1031 and S1032, the above-mentioned correction process can employ a feature recalibration method corresponding to the weights of the mutual influence factors, differentially adjusting the key operating parameters in the input features according to the importance of power characteristics and stability features. For example, for feature parameters that are significantly affected by another system in terms of power fluctuations, a higher adjustment coefficient can be assigned to amplify their impact; while for parameters that are less affected, a smaller adjustment range is maintained to avoid noise amplification. In this way, the corrected input features not only retain the original operating information but also integrate the dynamic coupling relationship between the two types of power generation systems, providing a more accurate and physically correlated feature foundation for subsequent multi-source data fusion and power grid risk early warning model input.

[0084] S104: Perform multi-source feature fusion on the first and second corrected input features to obtain the target input features, and input the target input features into the pre-trained target power grid early warning model to obtain power grid risk early warning results.

[0085] In this embodiment, the multi-source feature fusion process comprehensively considers the correlation characteristics between different power generation systems and their impact on power grid operation. Specifically, the modified input features from different power generation systems can be processed in a unified format, including normalization, dimensionality reduction, and feature selection. Then, weighted fusion is used to integrate feature information from different sources to form target input features containing multi-dimensional and multi-level information. Subsequently, the target input features are input into a pre-trained power grid early warning model, which can accurately identify potential risk factors, achieve early warning of abnormal power grid states, and effectively improve the safety and stability of power grid operation. The obtained risk warning results include, but are not limited to, the classification of power grid operation risk levels (such as low, medium, and high risk), prediction of potential fault types (such as voltage anomalies, frequency fluctuations, equipment overload, etc.), early warning time windows, and risk trend analysis. These early warning results can provide a scientific basis for power grid dispatching and maintenance, helping operation and maintenance personnel to identify risks in advance and handle them in a timely manner.

[0086] In one feasible implementation, the power grid early warning model is obtained through the following steps:

[0087] Obtain the historical cycle operation data of the first power generation system, and determine the power grid early warning model of the first power generation system based on the historical cycle operation data of the first power generation system;

[0088] Obtain historical cycle operating data of the second power generation system, and determine the power grid early warning model of the second power generation system based on the historical cycle operating data of the second power generation system;

[0089] The influence weight ratios of the first power generation system and the second power generation system are determined. Based on the influence weight ratios, the power grid early warning models of the first power generation system and the second power grid early warning models are weighted and fused to obtain the target power grid early warning model.

[0090] In this embodiment, historical periodic operational data of the first power generation system is first collected. This data includes, but is not limited to, power output curves, frequency variations, equipment operating status, and stability indicators. By preprocessing this historical data, such as denoising, normalization, and feature engineering, key features reflecting power output characteristics and operational stability are extracted, such as power fluctuation amplitude, ramp rate, frequency deviation, and voltage stability. Based on these features, machine learning is used to train a pre-defined model, constructing a power grid early warning model suitable for the first power generation system. This model can identify and predict potential power grid risk states. Similarly, the historical operational data of the second power generation system undergoes the same processing and feature extraction to train a power grid early warning model tailored to the characteristics of this system, ensuring that the model can accurately capture the operational patterns and risk characteristics of the second power generation system. Subsequently, the contribution of the two types of power generation systems to the power grid, the previously determined mutual influence factors, and their respective operational stability performance are comprehensively analyzed to determine the influence weight ratio of the two systems. The influence weight ratio reflects the relative influence and importance of each system on the overall power grid risk. After the weights are determined, model fusion techniques, such as weighted average, model stacking, or ensemble learning frameworks, are used to weight and fuse the two independent early warning models to obtain the target power grid early warning model. The fusion process retains the sensitivity of each model to the characteristics of its own system and reflects the coupling relationship between the systems through the weights.

[0091] In one feasible implementation, determining the influence weight ratio of the first power generation system and the second power generation system includes:

[0092] Obtain the impact weight assessment indicators of the first and second power generation systems within the historical period, and determine the initial comprehensive weight based on the impact weight assessment indicators. The impact weight assessment indicators include at least the average output power, the operational stability score, and historical failure rate data.

[0093] Obtain the power generation contribution of the first power generation system and the second power generation system, and adjust the initial comprehensive weight according to the power generation contribution to obtain the adjusted comprehensive weight;

[0094] The influence weight ratio of the first power generation system and the second power generation system is calculated based on the revised comprehensive weight.

[0095] In this embodiment, the purpose of determining the weight ratio for fusion is to scientifically and rationally integrate early warning models or feature information from different power generation systems, thereby improving the accuracy and reliability of overall power grid risk early warning. Different types of power generation systems exhibit significant differences in their power generation capacity, operating characteristics, and impact on power grid risks. The weight ratio reflects the actual contribution and importance of each system, allowing the fusion process to focus more on systems with a greater impact on power grid risks, avoiding information confusion caused by simple aggregation. Furthermore, the weight ratio also reflects the strength of coupling relationships between systems. By adjusting the fusion ratio, the interactions between different systems can be better captured, thereby improving the robustness and generalization ability of the model and reducing potential biases and misjudgments from a single model. Through this weighted fusion, not only is the effective integration of information from multiple sources and multiple power generation systems achieved, but it also provides a more accurate and differentiated basis for power grid dispatching and risk early warning, helping to formulate more efficient early warning response strategies and ensure the safe and stable operation of the power grid.

[0096] The specific implementation can include first obtaining the impact weight assessment indicators of the first and second power generation systems over their historical operating cycles. These indicators mainly include the average output power, operational stability score, and historical failure rate data. The average output power reflects the basic power generation capacity of each power generation system in the grid, the operational stability score measures the system's stable performance in key parameters such as frequency and voltage, and the historical failure rate data reflects the frequency and severity of past anomalies or failures. By comprehensively analyzing these indicators, the initial comprehensive weight of each power generation system is determined, which reflects the system's fundamental influence on the overall operation of the grid. Next, the power generation contribution of the two types of power generation systems is obtained, i.e., the proportion of each system in the total power generation of the entire grid. This contribution reflects the magnitude of its actual impact on the grid's power supply. Based on the power generation contribution, the initial comprehensive weight is corrected. The specific correction method can be weighted adjustment or normalization to ensure that the weight allocation is more in line with the actual operating environment and system importance. Finally, by normalizing the corrected comprehensive weights, the final impact weight ratio of the first and second power generation systems is obtained.

[0097] This application provides a power grid early warning and response method based on data fusion. First, it acquires the current-cycle operating data of a first power generation system and a second power generation system, constructing corresponding first and second input features respectively to achieve a preliminary quantitative description of the operating status of each power generation system. Then, it determines the mutual influence factors between the two types of power generation systems. These factors comprehensively reflect the coupling relationships between the power generation systems in terms of power output characteristics and operational stability, thereby revealing the dynamic interactive influences between different types of power generation systems. Based on these mutual influence factors, the first and second input features are specifically modified to obtain first and second modified input features, enabling the input features to more realistically and accurately reflect the mutual constraints and synergistic effects between power generation systems, eliminating the problem of feature isolation from a single system perspective. Subsequently, multi-source feature fusion technology is used to deeply fuse the modified input features, forming target input features containing multi-dimensional and multi-level operational information, improving the richness and relevance of feature expression. Finally, the target input features are input into a pre-trained target power grid early warning model. Relying on the model's learning ability from historical operating data and fault cases, it achieves accurate identification and early warning of potential power grid risks. This method not only improves the early warning model's ability to capture complex coupling relationships among multiple types of power generation systems, but also enhances the accuracy and timeliness of power grid risk identification. It effectively reduces the accident risks caused by new energy fluctuations or equipment anomalies, improves the safety, stability, and economy of power grid operation, and promotes the efficient management and sustainable development of smart grids.

[0098] This application also provides a power grid early warning and response system based on data fusion, referring to... Figure 2 The diagram shows a functional block diagram of a power grid early warning and response system 200 based on data fusion, which may include the following modules:

[0099] The acquisition module 201 is used to acquire the operating data of the first power generation system in the current cycle, and construct a first input feature based on the operating data of the first power generation system in the current cycle; acquire the operating data of the second power generation system in the current cycle, and construct a second input feature based on the operating data of the first power generation system in the current cycle.

[0100] Module 202 is used to determine the mutual influence factors between the first power generation system and the second power generation system;

[0101] The correction module 203 is used to correct the first input feature according to the mutual influence factor to obtain the first corrected input feature, and to correct the second input feature according to the mutual influence factor to obtain the second corrected input feature.

[0102] The early warning module 204 is used to perform multi-source feature fusion on the first and second corrected input features to obtain the target input features, and input the target input features into the pre-trained target power grid early warning model to obtain the power grid risk early warning result.

[0103] In one alternative implementation, the determining module includes:

[0104] The first determining submodule is used to determine the first mutual influence factor of the first power generation system and the second power generation system based on the coupling relationship between the first power generation system and the second power generation system in terms of power output characteristics.

[0105] The second determining submodule is used to determine the second mutual influence factor of the first power generation system and the second power generation system based on the correlation between the first power generation system and the second power generation system in terms of operational stability.

[0106] In one alternative implementation, the first determining submodule includes:

[0107] The first influence factor determination unit is used to obtain the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the Nth cycle, and determine the mutual influence factor of the first power generation system and the second power generation system in the time synchronization dimension based on the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the Nth cycle.

[0108] The second influencing factor determination unit is used to obtain the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the N+1th cycle, and to determine the mutual influence factor of the first power generation system and the second power generation system in the time lag dimension based on the output power data of the first power generation system in the Nth cycle and the output power data of the second power generation system in the N+1th cycle.

[0109] The third influence factor determination unit is used to determine the first mutual influence factor based on the mutual influence factors in the time synchronization dimension and the mutual influence factors in the time lag dimension.

[0110] In this embodiment, the present application also provides an electronic device, which may include a memory and one or more processors. The memory and processors are coupled. The memory is used to store computer program code, which includes computer instructions. When the processor executes the computer instructions, the electronic device can perform various functions or steps in the above method embodiments.

[0111] This embodiment also provides a computer-readable storage medium storing computer instructions that, when executed on an electronic device, cause the electronic device to perform the various functions or steps described in the above method embodiments.

[0112] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the various functions or steps in the above method embodiments.

[0113] In this embodiment, the electronic device, computer-readable storage medium, and computer program product are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0114] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable system. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, a computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0115] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0116] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0117] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, units, and processes described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0118] In the embodiments provided in this application, it should be understood that the disclosed systems, methods, and approaches can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0119] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0120] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0121] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0122] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A data fusion based power grid early warning response method, characterized in that, The power grid is coupled with a plurality of different types of power generation systems, and the method comprises: obtaining the operation data of the first power generation system in the current period, and constructing the first input feature according to the operation data of the first power generation system in the current period, obtaining the operation data of the second power generation system in the current period, and constructing the second input feature according to the operation data of the first power generation system in the current period; determining the mutual influence factor of the first power generation system and the second power generation system; According to the mutual influence factor, the first input feature is corrected to obtain the first correction input feature, and the second input feature is corrected according to the mutual influence factor to obtain the second correction input feature; Multi-source feature fusion is performed on the first correction input feature and the second correction input feature to obtain a target input feature, and the target input feature is input into a pre-trained target power grid early warning model to obtain a power grid risk early warning result; The determination of the mutual influence factor of the first power generation system and the second power generation system comprises: According to the coupling relationship between the first power generation system and the second power generation system in the power output characteristic, the first mutual influence factor of the first power generation system and the second power generation system is determined, and the first mutual influence factor is used to represent the coupling degree of the first power generation system and the second power generation system in the power output characteristic; According to the correlation between the first power generation system and the second power generation system in the operation stability, the second mutual influence factor of the first power generation system and the second power generation system is determined, and the second mutual influence factor is used to represent the correlation degree of the first power generation system and the second power generation system in the operation stability characteristic; The power grid early warning model is obtained by the following steps: obtaining the operation data of the first power generation system in the historical period, and determining the power grid early warning model of the first power generation system according to the operation data of the first power generation system in the historical period; obtaining the operation data of the second power generation system in the historical period, and determining the power grid early warning model of the second power generation system according to the operation data of the second power generation system in the historical period; determining the influence weight proportion of the first power generation system and the second power generation system, and weighting and fusing the power grid early warning model of the first power generation system and the power grid early warning model of the second power generation system according to the influence weight proportion to obtain the target power grid early warning model.

2. The data fusion based power grid early warning response method of claim 1, wherein, According to the coupling relationship between the first power generation system and the second power generation system in the power output characteristic, the first mutual influence factor of the first power generation system and the second power generation system is determined, and the first mutual influence factor is used to represent the coupling degree of the first power generation system and the second power generation system in the power output characteristic; obtaining the output power data of the first power generation system in the Nth period and the output power data of the second power generation system in the Nth period, and determining the power influence factor of the first power generation system and the second power generation system in the time synchronization dimension according to the output power data of the first power generation system in the Nth period and the output power data of the second power generation system in the Nth period; acquire output power data of the first power generation system in the Nth period and output power data of the second power generation system in the N+1th period, and determine a power influence factor of the first power generation system and the second power generation system in a time lag dimension according to the output power data of the first power generation system in the Nth period and the output power data of the second power generation system in the N+1th period; determine the first mutual influence factor according to the power influence factor in the time synchronization dimension and the power influence factor in the time lag dimension.

3. The data fusion based power grid early warning response method of claim 1, wherein, The determining the second mutual influence factor of the first power generation system and the second power generation system according to the correlation between the first power generation system and the second power generation system in operation stability comprises: acquire operation stability data of the first power generation system in the Nth period and operation stability data of the second power generation system in the Nth period, and determine a stability influence factor of the first power generation system and the second power generation system in a time synchronization dimension according to the operation stability data of the first power generation system in the Nth period and the operation stability data of the second power generation system in the Nth period; acquire operation stability data of the first power generation system in the Nth period and operation stability data of the second power generation system in the N+1th period, and determine a stability influence factor of the first power generation system and the second power generation system in a time lag dimension according to the operation stability data of the first power generation system in the Nth period and the operation stability data of the second power generation system in the N+1th period; determine the second mutual influence factor according to the stability influence factor in the time synchronization dimension and the stability influence factor in the time lag dimension.

4. The data fusion based power grid early warning response method of claim 1, wherein, The correcting the first input feature according to the mutual influence factor to obtain a first corrected input feature and correcting the second input feature according to the mutual influence factor to obtain a second corrected input feature comprises: correcting the first input feature according to the first mutual influence factor and the second mutual influence factor to obtain a first corrected input feature; correcting the second input feature according to the first mutual influence factor and the second mutual influence factor to obtain a second corrected input feature.

5. The data fusion based power grid early warning response method of claim 1, wherein, The determining the influence weight proportion of the first power generation system and the second power generation system comprises: acquire an influence weight evaluation index of the first power generation system and the second power generation system in a historical period, and determine an initial comprehensive weight according to the influence weight evaluation index, wherein the influence weight evaluation index at least comprises an output power mean value, an operation stability score and historical failure rate data; acquire a power generation contribution degree of the first power generation system and the second power generation system, and correct the initial comprehensive weight according to the power generation contribution degree to obtain a corrected comprehensive weight; calculate the influence weight proportion of the first power generation system and the second power generation system according to the corrected comprehensive weight.

6. A data fusion based power grid early warning response system characterized in that, The system for implementing the method of any one of claims 1-5 comprises: The acquisition module is configured to acquire operation data of a current period of the first power generation system, and construct a first input feature according to the operation data of the current period of the first power generation system; acquire operation data of a current period of the second power generation system, and construct a second input feature according to the operation data of the current period of the second power generation system; The determination module is configured to determine a mutual influence factor of the first power generation system and the second power generation system; The correction module is configured to correct the first input feature according to the mutual influence factor to obtain a first corrected input feature, and correct the second input feature according to the mutual influence factor to obtain a second corrected input feature; The early warning module is configured to perform multi-source feature fusion on the first corrected input feature and the second corrected input feature to obtain a target input feature, and input the target input feature into a pre-trained target power grid early warning model to obtain a power grid risk early warning result.

7. The data fusion based power grid early warning response system of claim 6, wherein, The determination module comprises: A first determination sub-module is configured to determine a first mutual influence factor of the first power generation system and the second power generation system according to a coupling relationship of the first power generation system and the second power generation system in power output characteristics; A second determination sub-module is configured to determine a second mutual influence factor of the first power generation system and the second power generation system according to a correlation relationship of the first power generation system and the second power generation system in operation stability.

8. The data fusion based power grid early warning response system of claim 7, wherein, The first determination sub-module comprises: A first influence factor determination unit is configured to acquire output power data of an Nth period of the first power generation system and output power data of an Nth period of the second power generation system, and determine a mutual influence factor of the first power generation system and the second power generation system in a time synchronization dimension according to the output power data of the Nth period of the first power generation system and the output power data of the Nth period of the second power generation system; A second influence factor determination unit is configured to acquire output power data of the Nth period of the first power generation system and output power data of an (N+1) th period of the second power generation system, and determine a mutual influence factor of the first power generation system and the second power generation system in a time lag dimension according to the output power data of the Nth period of the first power generation system and the output power data of the (N+1) th period of the second power generation system; A third influence factor determination unit is configured to determine the first mutual influence factor according to the mutual influence factor in the time synchronization dimension and the mutual influence factor in the time lag dimension.

Images