A dc modeling system and method based on ac-dc hybrid system response

Abstract

The application relates to a DC modeling system and method based on AC-DC hybrid system response, and belongs to the technical field of power system modeling, which comprises an information acquisition module, which is used for acquiring data information of a preset power grid operation state, and constructing an AC-DC hybrid power impedance model based on the data information; meanwhile, the data information is divided into stages; an LSTM network data acquisition module is used for acquiring characteristic data in the data information, dividing the characteristic data into three groups, and adopting an incremental combination mode to analyze the change rule of the preset power grid operation state in the three groups of characteristic data in the adjacent data. In the application, a fault text instruction different from the preset power grid historical operation data is input in the process of power modeling, the operation trend and the change rule of the power grid under different fault conditions in a future period can be monitored and known in the process of modeling, and the sensitivity analysis and reliability evaluation of the power grid fault prediction result are enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to a DC modeling system and method based on AC-DC hybrid system response, belonging to the technical field of power system modeling. BACKGROUND

[0002] At present, with the development of DC power transmission technology, the domestic power grid gradually develops into a large-scale power grid containing multiple return and multiple types of DC, which has been built to include conventional two-terminal DC, conventional three-terminal DC, conventional-flexible hybrid three-terminal DC, and conventional / flexible back-to-back DC, which is the most complex large-scale AC-DC hybrid power grid in the world, and contains complex stability characteristics. Based on the current development status and future planning of State Grid, China's power system will continue to show the characteristics and development trend of AC-DC hybrid transmission in the future. In order to realize accurate analysis and rapid evaluation of the operating state of the power system, and effectively verify the related control strategy, it is necessary to accurately model the actual physical power grid and restore the real operating state of the power system to the greatest extent possible.

[0003] Power system modeling is an important basis and effective tool for power system operation characteristic analysis and control strategy verification. The main idea of traditional power system simulation modeling is to derive the differential algebraic equations of the power system based on the basic physical principles of artificial systems, and to obtain a knowledge-driven mathematical model through digital modeling methods. However, there are limitations in describing unmodeled dynamics and uncertain information of the system, especially in the case of AC-DC hybrid response. The hybrid operation of power grid current can cause more uncertainty to the load. Traditional techniques mostly use reference power grid fixed operating parameters and / or statistical methods (such as time series analysis), machine learning algorithms (such as neural networks, support vector machines) for system modeling. However, considering the uncertainty of AC-DC hybrid power grid load, the sensitivity analysis and reliability evaluation of traditional techniques for power grid fault prediction results are poor, making it difficult to provide a basis for power system scheduling and planning according to the predicted load demand at different times. SUMMARY

[0004] In order to solve the problems in the prior art, the present application provides a DC modeling system and method based on AC-DC hybrid system response, which can enhance the sensitivity analysis and reliability evaluation of power grid fault prediction results.

[0005] In order to achieve the above purpose, the technical solution of the present application is as follows: a DC modeling system based on AC-DC hybrid system response, comprising:

[0006] An information acquisition module is used to acquire data information of a preset power grid operating state, and to construct an AC-DC hybrid power impedance model based on the data information, while dividing the data information into stages;

[0007] An LSTM network data acquisition module is configured to acquire feature data in the data information, divide the feature data into three groups, and each of the three groups of feature data is adjacent data. Meanwhile, the three groups of feature data are analyzed in an incremental combination manner to obtain a change rule of a preset power grid operating state, and the change rule is uploaded to the impedance model, and the impedance model is updated.

[0008] A clustering data analysis module is configured to analyze and calculate a transition time of a preset power grid operating state under different groups of data features, and analyze a change feature of the preset power grid operating state during different transition times.

[0009] A warning evaluation module is configured to mark different risk warning values for the change feature, and preset a safety threshold based on the maximum risk warning value. When the change feature of the preset power grid during the transition time exceeds the safety threshold, the system determines that the preset power grid operating state is abnormal, and issues a warning. Otherwise, it is not determined, and the warning information is uploaded to the impedance model.

[0010] A monitoring terminal is configured to record a communication base station and / or a communication base station operator matched with the preset power grid. When the change feature of the preset power grid during the transition time exceeds the safety threshold, the monitoring terminal initiates an information acquisition request for a preset power grid warning state to the communication base station and / or the communication base station operator corresponding to the preset power grid.

[0011] Further design of the above technical solution is that in the information acquisition module, the stage division includes dividing the data information according to a fault simulation state, wherein the fault simulation state is divided into a low fault state, a medium fault state and a high fault state, and the operating features of the preset power grid under different fault states are acquired, the operating features are uploaded to the impedance model, and the impedance model is updated.

[0012] In different fault states, a plurality of fault instructions and / or text data are inputted, the number of the plurality of fault instructions and / or text data is at least 3, and the fault instructions and / or text data are divided into ▽1, ▽2,..., ▽ n Wherein, n represents the nth fault instruction and / or text data, and the operating state and / or trend of the preset power grid under the input of different fault instructions and / or text data in different fault states is analyzed.

[0013] After the clustering data analysis module obtains the change feature, the change feature is enhanced by using a Q-learning algorithm.

[0014] The clustering data analysis module comprises a first clustering unit and a second clustering unit, the first clustering unit is used for reducing the transition time of the preset power grid operation state to the transition time in summer and autumn, and obtaining the change characteristics of the preset power grid operation state during the transition time in summer and autumn, the second clustering unit is used for reducing the transition time of the preset power grid operation state to the transition time in winter and spring, wherein the change characteristics of the preset power grid operation state are collected based on the average ambient temperature of the region where the preset power grid is located in winter and summer, the change characteristics of the operation state include the fluctuation change characteristics of the power load of the preset power grid, and the fluctuation change characteristics of the power load are uploaded to the impedance model.

[0015] The collected average ambient temperatures in winter and summer are divided into steps based on a difference of 3-5℃ per phase, and the divided data are at least 3, and the fault instructions and / or text data are input in the divided data, and the change of the preset power grid operation state is obtained.

[0016] The input fault instructions and / or text data in the divided data include current and / or voltage fault instructions and / or text data, wherein the fault instructions and / or text data include current and / or voltage sudden drop and sudden rise fault instructions and / or text data, and a preset observation period is set, the influence of current and / or voltage sudden drop and sudden rise on the preset power grid operation state in this period is calculated, and the following formula is used for calculation:

[0017] Wherein, E i represents the i-th preset observation period, the preset observation period is at least 6-10 minutes, and X represents the fault instruction and / or text data parameter;

[0018] In the formula, K represents the fluctuation data of the power load of the collected preset power grid, T represents the collection time corresponding to the preset observation period, and P represents the current and / or voltage sudden drop and / or sudden rise difference value;

[0019] Within the preset length of the observation period, the maximum sudden drop data and / or sudden drop data are used as the basis to preset the risk threshold of the fluctuation of the power load of the preset power grid, when the fluctuation value of the power load of the preset power grid in the future period exceeds the risk threshold, the system determines that the preset power grid operation state is abnormal, and issues a warning, and the warning information is uploaded to the impedance model, otherwise, it is not determined.

[0020] The data of the current and / or voltage sudden drop and sudden rise are generated into a data set [p a *ρ b *ρ c *ρ nWherein, n represents the nth data set, and in any one of the data sets, the data values of the current and / or voltage step-down and step-up are respectively stepped down and / or stepped up by 10 numerical units per phase, and the preset power grid power load fluctuation value in the measurement process is obtained.

[0021] The system further comprises a fusion data monitoring module, which is responsive to the grouping data of the LSTM network data acquisition module, and is used for monitoring the correlation law between the grouping data change of the preset power grid in the future period and the change of the operating state of the preset power grid.

[0022] A DC modeling method based on AC-DC hybrid system response, applied to the DC modeling system based on AC-DC hybrid system response, comprising the following steps:

[0023] Obtain data information of the operating state of the preset power grid, and construct an AC-DC hybrid power impedance model based on the data information, and divide the data information into stages;

[0024] Acquire feature data in the data information, and divide the feature data into at least three groups, and the three groups of feature data are adjacent data, and in the adjacent data, the change law of the operating state of the preset power grid in three groups of feature data is analyzed by using an incremental combination method, and the change law is uploaded to the impedance model, and the impedance model is updated;

[0025] Analyze and calculate the transition time of the operating state of the preset power grid under different groups of data features, and analyze the change characteristics of the operating state of the preset power grid during different transition times;

[0026] Mark the change characteristics with different risk alert values, and preset a safety threshold based on the maximum risk alert value, and when the change characteristics of the preset power grid during the transition time exceed the safety threshold, the system determines that the operating state of the preset power grid is abnormal, and issues a warning, otherwise, it is not determined, and the warning information is uploaded to the impedance model;

[0027] Based on grouping data, for monitoring the correlation law between the grouping data change of the preset power grid in the future period and the change of the operating state of the preset power grid;

[0028] Record the communication base station and / or communication base station operator matched with the preset power grid, and when the change characteristics of the preset power grid during the transition time exceed the safety threshold, initiate an information acquisition request of the preset power grid warning state to the communication base station and / or communication base station operator corresponding to the preset power grid.

[0029] The beneficial effects of the present application are:

[0030] The application can monitor the operation trend and change rule of the power grid under different fault conditions in the future period during the modeling process by inputting fault text instructions different from the preset power grid historical operation data, and then obtain the influence of different change rules on the power load of the power grid, and also can analyze the influence of the mutation of different power indexes in the preset time in different fault text instructions on the operation of the power grid according to the data difference of seasons and environmental temperature, thereby enhancing the sensitivity analysis and reliability evaluation of the power grid fault prediction result, and then providing a basis for power grid system scheduling and planning according to the load demand at different times. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 It is a modular structure schematic diagram of the DC modeling system based on the AC-DC hybrid system response of the embodiment of the application.

[0032] Figure 2 It is a method flow schematic diagram of the embodiment of the application.

[0033] Figure 3 It is a flow structure schematic diagram of the embodiment of the application.

[0034] In the figure, the number 110 is a comprehensive information acquisition module; the number 120 is an LSTM network data acquisition module; the number 130 is a clustering data analysis module; the number 1301 is a first clustering unit; the number 1302 is a second clustering unit; the number 140 is a comprehensive early warning evaluation module; the number 150 is a monitoring terminal; and the number 160 is a fusion data monitoring module. DETAILED DESCRIPTION

[0035] The application will be described in detail below in combination with the drawings and specific embodiments.

[0036] Embodiment one

[0037] Referring to FIGS. 1-4, the embodiment provides a DC modeling system based on the AC-DC hybrid system response, which comprises: Figure 1 Figure 3 The comprehensive information acquisition module 110 is used to acquire data information of a preset power grid operation state, and construct an AC-DC hybrid power impedance model based on the data information, and simultaneously divide the data information into stages.

[0038] The data information comprises current and voltage of the preset power grid, and the number of electricity users in the region where the preset power grid is located, and daily electricity consumption data.

[0039] The data information comprises current and voltage of the preset power grid, and the number of electricity users in the region where the preset power grid is located, and daily electricity consumption data.

[0040] ​The embodiment needs to be explained that the stage division includes dividing the data information according to the fault simulation state, wherein the fault simulation state is divided into low fault state, medium fault state and high fault state, and the running characteristics of the preset power grid under different fault states are collected, the running characteristics are uploaded to the impedance model, and the impedance model is updated.

[0041] The embodiment further inputs a plurality of fault instructions and / or text data in different fault states, wherein the number of the plurality of fault instructions and / or text data is at least 3, and the fault instructions and / or text data are divided into ∇1, ∇2,..., ∇ n Wherein, n represents the nth fault instruction and / or text data, the running state and / or trend of the preset power grid under the input of different fault instructions and / or text data in different fault states are analyzed, and a prediction model is generated; the running state and / or trend of the preset power grid under the input of different fault instructions and / or text data are predicted.

[0042] It needs to be emphasized that the fault instructions and / or text data include current and / or voltage fault instructions and / or text data;

[0043] The LSTM network data acquisition module 120 is used to collect feature data in the data information, and divide the feature data into three groups, and the three groups of feature data are adjacent data, and in the adjacent data, the change law of the running state of the preset power grid in the three groups of feature data is analyzed by using incremental combination, and the change law is uploaded to the impedance model, and the impedance model is updated.

[0044] The clustering data analysis module 130 is used to analyze and calculate the transition time of the running state of the preset power grid under different groups of data characteristics, analyze the change characteristics of the running state of the preset power grid during different transition times, and use Q-learning algorithm for reinforcement learning, and the clustering data analysis module includes a first clustering unit 1301 and a second clustering unit 1302.

[0045] In the embodiment, the first clustering unit 1301 is used to summarize the transition time of the running state of the preset power grid into the transition time in summer and autumn, and obtain the change characteristics of the running state of the preset power grid during the transition time in summer and autumn.

[0046] The second clustering unit 1302 is used to summarize the transition time of the running state of the preset power grid into the transition time in winter and spring, wherein the change characteristics of the running state of the preset power grid are collected based on the average environmental temperature of the region where the preset power grid is located in winter and summer, the change characteristics of the running state include the fluctuation change characteristics of the power load of the preset power grid, and the fluctuation change characteristics of the power load are uploaded to the impedance model.

[0047] The embodiment further illustrates that the real data shows that the change characteristics of the power grid under different environmental temperature conditions are first reflected in the fluctuation of the power load, and the fluctuation has obvious seasonality. In the high-temperature weather in summer, the power load peak value often appears in July, and the power load fluctuation in this period is increased by about 20% than in normal periods, because the high-temperature weather causes the frequency of using air conditioners and other electrical appliances to increase, thereby increasing the power load. In the low-temperature weather in winter, the power load peak value often appears in January, and is increased by about 38% than in normal periods, because the low-temperature weather causes the frequency of using electric heaters and other electrical appliances to increase, thereby also increasing the power load. In spring and autumn, the temperature is relatively moderate, and the power load is relatively stable, so the average environmental temperature in winter and summer is used as a reference to collect the change characteristics of the preset power grid operating state, which has more reference value.

[0048] Further, the preset power grid daily power load curve is obtained, and the daily power load curve includes the power load fluctuation change of the preset power grid in the morning and evening peak periods. The morning peak includes 7:00-9:00, and the evening peak includes 18:00-22:00.

[0049] In real life, the morning peak appears when the resident and commercial power consumption begins, between 7:00 and 9:00 in the morning, and the evening peak appears from the evening to about 10:00 at night, because the power demand of residents increases after work, including lighting, cooking and entertainment electrical appliances, and the load curve is different on weekdays and weekends. The morning and evening peaks are not as obvious on weekends, but the overall power consumption distribution is more average, so the power load fluctuation change in the above two peak periods is used as a reference, which has more practical significance.

[0050] Further, the average environmental temperature in winter and summer is divided into steps based on a difference of 3-5°C per phase, and the divided data is at least 3, and the fault instruction and / or text data are input into the divided data, and the change of the preset power grid operating state is obtained.

[0051] Based on the above, the fault instruction and / or text data input into the divided data include the fault instruction and / or text data of current and / or voltage, wherein the fault instruction and / or text data include the fault instruction and / or text data of current and / or voltage drop and surge, and a preset observation period is set, the influence of the current and / or voltage drop and surge on the preset power grid operating state in the observation period is calculated, and the calculation is obtained according to the following formula:

[0052] wherein E i represents the i-th preset observation period, the observation period is preset to be at least 6-10 minutes, and X represents the fault instruction and / or text data parameter.

[0053] In the formula, K represents the collected power load fluctuation data of the preset power grid, T represents the collection time corresponding to the preset observation period, and P represents the current and / or voltage step-down and / or step-up difference value.

[0054] And within the preset length range of the observation period, the preset power grid power load fluctuation preset risk threshold is based on the maximum step-down data and / or step-down data. When the power load fluctuation value of the preset power grid in the future period exceeds the risk threshold, the system determines that the operation state of the preset power grid is abnormal, and issues a warning, and uploads the warning information to the impedance model, otherwise, it is not determined.

[0055] Wherein, the data of current and / or voltage step-down and step-up are generated data set [p a *ρ b *ρ c *ρ n ], wherein n represents the nth data set, and in any one data set, the data values of current and / or voltage step-down and step-up are respectively measured by step-down and / or step-up every 10 numerical units, the power load fluctuation value of the preset power grid in the measurement process is obtained, and the training set, the validation set and the test set are divided according to the data set.

[0056] Wherein, the embodiment measures step-down and / or step-up every 10 numerical units, including current and / or voltage every 10A and / or 10V.

[0057] The comprehensive warning evaluation module 140 responds to the analysis of the change characteristics of the operation state of the preset power grid during the transition time by the clustering data analysis module, and is used for marking the change characteristics with different risk warning values, and presetting a safety threshold based on the maximum risk warning value. When the change characteristics of the preset power grid during the transition time exceed the safety threshold, the system determines that the operation state of the preset power grid is abnormal, and issues a warning, otherwise, it is not determined.

[0058] The monitoring terminal 150 is used to record the communication base station and / or communication base station operator matched with the preset power grid, and receive the field data uploaded by the monitoring point, and update the background data at the same time; when the change characteristics of the preset power grid during the transition time exceed the safety threshold, the information acquisition request of the preset power grid warning state is initiated to the communication base station and / or communication base station operator corresponding to the preset power grid.

[0059] The fusion data monitoring module 160 responds to the grouping data of the LSTM network data collection module, and is used for monitoring the association rule between the grouping data change of the preset power grid in the future period and the change of the operation state of the preset power grid.

[0060] Based on the above, the application can monitor the running trend and change rule of the power grid under different fault conditions in the future period during modeling by inputting fault text instructions different from the preset power grid historical operation data, and then obtain the influence of different change rules on the power grid power load, thereby enhancing the sensitivity analysis and reliability evaluation of the power grid fault prediction result, and providing a basis for power grid system scheduling and planning according to the predicted load demand at different times.

[0061] Embodiment two

[0062] This embodiment combines the DC modeling system based on the response of the AC-DC hybrid system of embodiment one, and proposes a DC modeling method based on the response of the AC-DC hybrid system, as shown in Figure 2 , specifically as follows:

[0063] Obtain data information of the preset power grid running state, and construct an AC-DC hybrid power impedance model based on the data information, and divide the data information into stages.

[0064] Collect feature data in the data information, and divide the feature data into at least three groups, and the three groups of feature data are adjacent data, and in the adjacent data, the change rule of the preset power grid running state in the three groups of feature data is analyzed by using an incremental combination method, and the change rule is uploaded to the impedance model, and the impedance model is updated.

[0065] Analyze and calculate the transition time of the preset power grid running state under different groups of data characteristics, analyze the change characteristics of the preset power grid running state during different transition times, and use a Q-learning algorithm for enhanced learning.

[0066] Mark the change characteristics with different risk alert values, and preset a safety threshold based on the largest risk alert value, and when the change characteristics of the preset power grid during the transition time exceed the safety threshold, the system determines that the preset power grid running state is abnormal, and issues a warning, otherwise, it is not determined, and the warning information is uploaded to the impedance model.

[0067] Based on the grouped data, the association rule between the grouped data change of the preset power grid in the future period and the change of the preset power grid running state is monitored.

[0068] Record the communication base station and / or communication base station operator matched with the preset power grid, and receive the on-site data uploaded by the monitoring point, and when the change characteristics of the preset power grid during the transition time exceed the safety threshold, initiate an information acquisition request of the preset power grid warning state to the communication base station and / or communication base station operator corresponding to the preset power grid.

[0069] To sum up, the embodiment constructs an AC-DC hybrid power impedance model, and by inputting fault text instructions different from preset power grid historical operation data in the power modeling process, the operation trend and change law of the power grid under different fault conditions in the future period can be monitored and known in the modeling process, and the optimized impedance model is updated, and then the influence of different change laws on the power load of the power grid is obtained.

[0070] Meanwhile, the influence of the mutation of different power indicators in the preset time on the operation of the power grid in different fault text instructions can also be analyzed according to the data difference of seasons and environmental temperature, and the optimized impedance model is also used for updating, thereby enhancing the sensitivity analysis and reliability evaluation of the power grid fault prediction result.

[0071] The technical solutions of the present application are not limited to the above embodiments, and any technical solution obtained by equivalent replacement falls within the scope of the present application.

Claims

1. A DC modeling system based on the response of an AC / DC hybrid system, characterized in that, include: Information acquisition module; This is used to obtain data information on the preset power grid operating status, and to construct an AC / DC hybrid power impedance model based on the data information, while dividing the data information into stages. LSTM network data acquisition module; This is used to collect feature data from the data information, divide the feature data into three groups, and the three groups of feature data are all adjacent data. At the same time, in the adjacent data, the change law of the preset power grid operation state in the three groups of feature data is analyzed by an incremental combination method, and the change law is uploaded to the impedance model, and the impedance model is updated at the same time. Clustering data analysis module; Used to analyze and calculate the transition time of the preset power grid operating state under different sets of data characteristics, and to analyze the change characteristics of the preset power grid operating state during different transition times; The early warning assessment module is used to mark the change characteristics with different risk warning values, and to preset a safety threshold based on the largest risk warning value. When the change characteristics of the preset power grid exceed the safety threshold during the transition period, the system determines that the preset power grid is in an abnormal operating state and issues an early warning. Otherwise, no judgment is made, and the early warning information is uploaded to the impedance model. Monitoring terminal; This is used to record the communication base stations and / or communication base station operators that are matched with the preset power grid. When the changes in the preset power grid during the transition period exceed the safety threshold, a request for information acquisition of the preset power grid early warning status is initiated to the communication base stations and / or communication base station operators corresponding to the preset power grid.

2. The DC modeling system based on the response of an AC / DC hybrid system according to claim 1, characterized in that: In the information acquisition module, the stage division includes dividing the data information according to the fault simulation state, wherein the fault simulation state is divided into low fault state, medium fault state and high fault state, and the operating characteristics of the preset power grid under different fault states are collected, the operating characteristics are uploaded to the impedance model, and the impedance model is updated.

3. The DC modeling system based on the response of an AC / DC hybrid system according to claim 1, characterized in that: Input several fault commands and / or text data under different fault states, with a minimum of 3 fault commands and / or text data entries, and classify the fault commands and / or text data into ▽1, ▽2, ..., ▽ n Where n represents the nth fault command and / or text data, and the operating status and / or trend of the preset power grid are analyzed under different fault conditions and different fault commands and / or text data are input.

4. The DC modeling system based on the response of an AC / DC hybrid system according to claim 1, characterized in that: After obtaining the changing features, the clustering data analysis module uses the Q-learning algorithm to perform reinforcement learning on the changing features.

5. The DC modeling system based on the response of an AC / DC hybrid system according to claim 4, characterized in that: The clustering data analysis module includes a first clustering unit and a second clustering unit. The first clustering unit is used to summarize the transition time of the preset power grid operating state into the transition time in summer and autumn, and to obtain the change characteristics of the preset power grid operating state during the transition time in summer and autumn. The second clustering unit is used to summarize the transition time of the preset power grid operating state into the transition time in winter and spring. The change characteristics of the preset power grid operating state are collected based on the average ambient temperature of the region where the preset power grid is located in winter and summer. The change characteristics of the operating state include the fluctuation characteristics of the preset power grid power load, and the fluctuation characteristics of the power load are uploaded to the impedance model.

6. The DC modeling system based on the response of an AC / DC hybrid system according to claim 5, characterized in that: The average ambient temperatures collected in winter and summer are divided into steps based on a difference of 3 to 5°C, with each step containing at least three data points. Fault commands and / or text data are input into the divided data, while the preset power grid operating status changes are obtained.

7. The DC modeling system based on the response of an AC / DC hybrid system according to claim 6, characterized in that: The fault commands and / or text data input into the segmented data include fault commands and / or text data related to current and / or voltage, wherein the fault commands and / or text data include fault commands and / or text data related to sudden drops and rises in current and / or voltage. A preset observation period is established, and the impact of sudden drops and rises in current and / or voltage on the preset power grid operating state during this period is calculated using the following formula: Among them, E i The i-th preset observation period is indicated, and the preset duration of the observation period is at least 6 to 10 minutes. X represents the fault command and / or text data parameters. In the formula, K represents the collected power load fluctuation data of the preset power grid, T represents the collection time corresponding to the preset observation period, and P represents the difference between current and / or voltage sudden drop and / or sudden rise. Within a preset duration of the observation period, a risk threshold for the power load fluctuation of the preset power grid is preset based on the maximum drop data and / or the drop data. When the power load fluctuation value of the preset power grid in a future period exceeds the risk threshold, the system determines that the preset power grid is in an abnormal operating state and issues an early warning, while uploading the early warning information to the impedance model. Otherwise, no determination is made.

8. The DC modeling system based on the response of an AC / DC hybrid system according to claim 7, characterized in that: Generate a dataset from the data of sudden drops and rises in current and / or voltage [ρ] a *ρ b *ρ c *ρ n ], where n represents the nth dataset, and in any of the datasets, the data values ​​of the current and / or voltage sag and swell are respectively calculated by sag and / or swell for every 10 numerical units of phase difference, and the preset power grid load fluctuation value is obtained during the calculation process.

9. The DC modeling system based on the response of an AC / DC hybrid system according to claim 1, characterized in that: It also includes a fusion data monitoring module, which responds to the packet data from the LSTM network data acquisition module to monitor the correlation between changes in packet data and changes in the operating status of the preset power grid in future time periods.

10. A DC modeling method based on the response of an AC / DC hybrid system, applied to the DC modeling system based on the response of an AC / DC hybrid system as described in claim 1, characterized in that, Includes the following steps: Acquire data information on the preset power grid operating status, and construct an AC / DC hybrid power impedance model based on the data information, while dividing the data information into stages; The feature data in the data information is collected and divided into at least three groups, and the three groups of feature data are all adjacent data. At the same time, the change law of the preset power grid operation state in the three groups of feature data is analyzed by using an incremental combination method in the adjacent data, and the change law is uploaded to the impedance model and the impedance model is updated. The transition time of the preset power grid operating state under different data characteristics is analyzed and calculated, and the change characteristics of the preset power grid operating state during different transition times are analyzed. The changing characteristics are marked with different risk warning values, and a safety threshold is preset based on the largest risk warning value. When the changing characteristics of the preset power grid exceed the safety threshold during the transition period, the system determines that the preset power grid is in an abnormal operating state and issues a warning. Otherwise, no judgment is made, and the warning information is uploaded to the impedance model. Based on grouped data, it is used to monitor the correlation between changes in grouped data of the preset power grid in future time periods and changes in the preset power grid operating status; Record the communication base stations and / or communication base station operators that are matched with the preset power grid. When the changes in the preset power grid during the transition period exceed the safety threshold, a request for information acquisition of the preset power grid early warning status is initiated to the communication base stations and / or communication base station operators corresponding to the preset power grid.

Images