Direct-current modeling system and method based on response of alternating-current and direct-current hybrid system

Abstract

The invention 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, and the system comprises an information obtaining module which is used for obtaining data information of a preset power grid operation state, constructing an AC-DC hybrid power impedance model based on the data information, and carrying out the stage division of the data information, and the LSTM network data acquisition module is used for acquiring feature data in the data information and dividing the feature data into three groups, the three groups of feature data are adjacent data, and in the adjacent data, a change rule of a preset power grid operation state in the three groups of feature data is analyzed by adopting an incremental combination mode. According to the method, the fault text instruction different from the preset historical operation data of the power grid is input in the electric power modeling process, so that the operation trend and the change rule of the power grid in different fault conditions in the future time period can be monitored and known in the modeling process, and the sensitivity analysis and the reliability evaluation of the power grid fault prediction result are enhanced.

Description

Technical Field

[0001] The invention relates to a direct current modeling system and method based on the response of an alternating current and a direct current hybrid system, and belongs to the technical field of power system modeling. Background Art

[0002] At present, with the development of direct current transmission technology, the domestic power grid has gradually developed into a super-large power grid with multiple circuits and types of direct current. It has built conventional two-terminal direct current, conventional three-terminal direct current, conventional-flexible hybrid three-terminal direct current, and conventional / flexible back-to-back direct current. It is the most complex super-large AC / DC hybrid power grid in the world, which contains complex stability characteristics. Based on the development status and future planning of the State Grid, my country's power system will continue to show the characteristics and development trends of AC / DC hybrid transmission in the future. In order to achieve accurate analysis and rapid evaluation of the operating status of the power system and effectively verify the relevant control strategies, it is necessary to accurately model the actual physical power grid and restore the real operating status of the power system to the greatest extent possible.

[0003] Grid system modeling is an important foundation and effective tool for grid 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 obtain a knowledge-driven mathematical model through digital modeling methods. However, it has limitations in describing the unmodeled dynamics and uncertain information of the system, especially in the case of AC / DC hybrid response. The hybrid operation of the grid current is likely to cause more uncertainty in the load. Most traditional technologies model the system by referring to fixed grid operation parameters and / or statistical methods (such as time series analysis) and machine learning algorithms (such as neural networks and support vector machines). However, considering the uncertainty of AC / DC hybrid grid load, traditional technologies have poor sensitivity analysis and reliability assessment of grid fault prediction results, which makes it difficult to provide a basis for grid system scheduling and planning based on the predicted load demand at different times. Summary of the invention

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

[0005] In order to achieve the above object, the technical solution proposed by the present invention is: a DC modeling system based on the response of an AC / DC hybrid system, comprising:

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

[0007] LSTM network data acquisition module; used to collect characteristic data in the data information, and divide the characteristic data into three groups, and the three groups of characteristic data are adjacent data, and at the same time, in the adjacent data, an incremental combination method is used to analyze the change law of the preset power grid operation state in the three groups of characteristic data, and the change law is uploaded to the impedance model, and the impedance model is updated at the same time;

[0008] Cluster data analysis module; used to analyze and calculate the transition time of the preset power grid operation state under different groups of data characteristics, and analyze the change characteristics of the preset power grid operation state during different transition times;

[0009] An early warning assessment module; used to mark the change characteristics with different risk warning values, and preset 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 an early warning. Otherwise, no determination is made and the early warning information is uploaded to the impedance model at the same time.

[0010] A monitoring terminal; used to record the communication base stations and / or communication base station operators that match the preset power grid, and when the changing characteristics of the preset power grid during the transition time exceed the safety threshold, a request for obtaining information on the preset power grid warning status is initiated to the communication base stations and / or communication base station operators corresponding to the preset power grid.

[0011] A further design of the above technical solution is as follows: 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 a low fault state, a medium fault state and a 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.

[0012] Input a plurality of fault instructions and / or text data in different fault states, 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 , where n represents the nth fault instruction and / or text data, analyzes different fault states, and presets the operating state and / or trend of the power grid when different fault instructions and / or text data are input.

[0013] After obtaining the change features, the clustering data analysis module uses a Q-learning algorithm to perform enhanced learning on the change features.

[0014] 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 operation state into the transition time in summer and autumn, and obtain the change characteristics of the preset power grid operation 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 operation state into the transition time in winter and spring. The change characteristics of the preset power grid operation state are collected based on the average ambient temperature of the area 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 preset power grid power load, 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 to 5°C per phase, and the divided data are at least 3. Fault instructions and / or text data are input into the divided data, and the preset power grid operation status changes are obtained at the same time.

[0016] The fault instructions and / or text data input in the divided data include fault instructions and / or text data of current and / or voltage, wherein the fault instructions and / or text data include fault instructions and / or text data of current and / or voltage sag and swell, and an observation period is preset, and the influence of the current and / or voltage sag and swell on the preset power grid operation state during this period is calculated, and the calculation is obtained according to the following formula:

[0017] Among them, E i represents the i-th preset observation period, wherein the preset duration of the observation period is at least 6 to 10 minutes, and X represents a fault instruction and / or text data parameter;

[0018] Wherein, 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 sag and / or swell difference;

[0019] And within the preset time range of the observation period, a preset risk threshold for the power load fluctuation of the preset power grid is set based on the maximum sag data and / or sag 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 operating state of the preset power grid is abnormal, and issues an early warning, and uploads the early warning information to the impedance model at the same time. Otherwise, no determination is made.

[0020] The data of the current and / or voltage sag and swell are used to generate a data set [ρ a *ρ b *ρ c *ρ n], wherein n represents the nth data set, and in any one of the data sets, the data values ​​of the current and / or voltage sag and swell are respectively calculated for sag and / or swell with a difference of 10 numerical units, so as to obtain the preset grid power load fluctuation value during the calculation process.

[0021] The system also includes a fusion data monitoring module, which responds to the grouped data of the LSTM network data acquisition module and is used to monitor the correlation between changes in the grouped data of a preset power grid in the future period and changes in the preset power grid operating state.

[0022] A DC modeling method based on the response of an AC / DC hybrid system is applied to the DC modeling system based on the response of an AC / DC hybrid system, and comprises the following steps:

[0023] Acquire data information of a preset power grid operation state, and construct an AC / DC hybrid power impedance model based on the data information, while dividing the data information into stages;

[0024] Collect characteristic data in the data information, and divide the characteristic data into at least three groups, wherein the three groups of characteristic data are all adjacent data, and at the same time, in the adjacent data, use an incremental combination method to analyze the change law of the preset power grid operation state in the three groups of characteristic data, and upload the change law to the impedance model, and update the impedance model at the same time;

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

[0026] The change characteristics are marked with different risk warning values, and a safety threshold is preset based on the largest 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 an early warning. Otherwise, no determination is made and the early warning information is uploaded to the impedance model at the same time.

[0027] Based on the grouped data, it is used to monitor the correlation between the changes in the grouped data of the preset power grid in the future period and the changes in the operation status of the preset power grid;

[0028] The communication base stations and / or communication base station operators that match the preset power grid are recorded, and when the preset power grid exceeds the safety threshold due to the changing characteristics during the transition time, an information acquisition request for the preset power grid warning status is initiated to the communication base stations and / or communication base station operators corresponding to the preset power grid.

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

[0030] The present invention inputs fault text instructions that are different from preset historical power grid operation data during the power modeling process, so that the operation trend and change law of the power grid under different fault conditions in the future period can be monitored during the modeling process, and then the influence of different change laws on the power load of the power grid can be obtained. At the same time, the influence of sudden changes in different power indicators within a preset time in different fault text instructions on the operation of the power grid can be analyzed according to the data differences of seasons and ambient temperature, thereby enhancing the sensitivity analysis and reliability evaluation of the power grid fault prediction results, and providing a basis for the scheduling and planning of the power grid system according to the load demand at different times. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic diagram of the modular structure of a DC modeling system based on the response of an AC / DC hybrid system according to an embodiment of the present invention;

[0032] Figure 2 A schematic diagram of a method flow chart of an embodiment of the present invention;

[0033] Figure 3 A schematic diagram of the process structure of an embodiment of the present invention;

[0034] Numbers in the figure: 110 - comprehensive information acquisition module; 120 - LSTM network data acquisition module; 130 - clustering data analysis module; 1301 - first clustering unit; 1302 - second clustering unit; 140 - comprehensive early warning assessment module; 150 - monitoring terminal; 160 - fusion data monitoring module. DETAILED DESCRIPTION

[0035] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Embodiment 1

[0037] Reference Figure 1 and Figure 3 As shown, this embodiment provides a DC modeling system based on the response of an AC / DC hybrid system, including:

[0038] The comprehensive information acquisition module 110 is used to acquire data information of a preset power grid operation status, and to construct an AC / DC hybrid power impedance model based on the data information, while dividing the data information into stages.

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

[0040] It should be noted that in this embodiment, the stage division includes dividing the data information according to the 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 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.

[0041] In this embodiment, a plurality of fault instructions and / or text data are further input 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 , where n represents the nth fault instruction and / or text data, analyzes the operating status and / or trend of the preset power grid under different fault conditions when different fault instructions and / or text data are input, and generates a prediction model; predicts the operating status and / or trend of the preset power grid when different fault instructions and / or text data are input.

[0042] It should 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 characteristic data in the data information and divide the characteristic data into three groups, and the three groups of characteristic data are adjacent data. At the same time, in the adjacent data, an incremental combination method is used to analyze the change law of the preset power grid operation state in the three groups of characteristic data, and the change law is uploaded to the impedance model, and the impedance model is updated at the same time.

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

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

[0046] The second clustering unit 1302 is used to summarize the transition time of the preset power grid operation state into 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 area where the preset power grid is located in winter and summer, and 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.

[0047] This embodiment further illustrates that actual data show that the changing characteristics of the power grid under different ambient temperature conditions are first manifested in the fluctuation of power load, and this fluctuation has obvious seasonality. Among them, in the high temperature weather in summer, the peak power load mostly occurs in July, and the power load fluctuation in this period increases by about 20% compared with the normal period. This is because high temperature weather will cause people to use electrical appliances such as air conditioners more frequently, thereby increasing the power load; and in the low temperature weather in winter, the peak power load mostly occurs in January, which is about 38% higher than the normal period. This is because low temperature weather will cause people to use electrical appliances such as electric heaters more frequently, which also increases the power load. The temperature in spring and autumn is relatively mild, and the power load is relatively stable. Therefore, this application uses the average ambient temperature in winter and summer as a benchmark to collect the changing characteristics of the preset power grid operation status, which is more valuable for reference.

[0048] Furthermore, it also includes obtaining the daily power load curve of the preset power grid, which includes the fluctuation changes of power load during the morning and evening peak hours in the area where the preset power grid is located. The morning peak includes 7 am to 9 am; the evening peak includes 6 am to 10 am.

[0049] In real life, the morning peak occurs when residential and commercial electricity consumption begins, approximately between 7 and 9 in the morning; the evening peak occurs in the late afternoon to around 10 pm. This is because residents' electricity demand increases after returning home from work, including the use of lighting, cooking and entertainment appliances. The load curves on weekdays and weekends will be different. The morning and evening peaks on weekends are not as obvious as on weekdays, but the overall electricity consumption distribution throughout the day is more even. Therefore, it is more realistic for this application to use the fluctuations in power load during the above two peak periods as a reference.

[0050] Furthermore, the collected average ambient temperatures in winter and summer are divided into steps based on a difference of 3 to 5°C per phase, and the divided data are at least 3, and fault instructions and / or text data are input into the divided data, and the preset power grid operation status changes are obtained at the same time.

[0051] On the basis of the above, this embodiment inputs fault instructions and / or text data including fault instructions and / or text data of current and / or voltage into the divided data, wherein the fault instructions and / or text data include fault instructions and / or text data of current and / or voltage sag and swell, and presets an observation period, calculates the influence of the current and / or voltage sag and swell on the preset power grid operation state during this period, and calculates it according to the following formula:

[0052] Among them, E i represents the i-th preset observation period, the preset duration of the observation period is at least 6 to 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 sag and / or swell difference.

[0054] And within the preset time range of the observation period, a preset risk threshold is set for the power load fluctuation of the preset power grid based on the maximum sag data and / or sag 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 operating state of the preset power grid is abnormal, and issues an early warning. At the same time, the early warning information is uploaded to the impedance model. Otherwise, no judgment is made.

[0055] The data of current and / or voltage sag and swell are used to generate a data set [ρ a *ρ b *ρ c *ρ n ], wherein n represents the nth data set, and in any data set, the data values ​​of current and / or voltage sag and swell are calculated with a difference of 10 numerical units, respectively, to obtain the preset grid power load fluctuation value in the measurement process, and the data set is divided into training set, validation set and test set.

[0056] In this embodiment, the sag and / or swell calculation is performed with a difference of 10 numerical units per phase, including the current and / or voltage being calculated with a difference of 10A and / or 10V per phase.

[0057] The comprehensive early warning assessment module 140 responds to the clustering data analysis module's analysis of the changing characteristics of the preset power grid operation status during different transition times, and is used to mark the changing characteristics with different risk warning values, and preset a safety threshold based on the largest risk warning value. When the changing characteristics of the preset power grid during the transition time exceed the safety threshold, the system determines that the preset power grid operation status is abnormal and issues a warning, otherwise, no determination is made.

[0058] The monitoring terminal 150 is used to record the communication base stations and / or communication base station operators that match the preset power grid, receive the field data uploaded by the collection monitoring points, and update the background data at the same time; when the preset power grid exceeds the safety threshold due to the changing characteristics during the transition time, a request for obtaining information on the preset power grid warning status is initiated to the communication base stations and / or communication base station operators corresponding to the preset power grid.

[0059] The fusion data monitoring module 160 responds to the grouped data of the LSTM network data acquisition module and is used to monitor the correlation between the changes in the grouped data of the preset power grid in the future period and the changes in the preset power grid operation status.

[0060] Based on the above, it can be known that the present application can monitor and know the operation trend and change law of the power grid under different fault conditions in the future period during the modeling process by inputting fault text instructions that are different from the preset historical operation data of the power grid during the power modeling process, and then obtain the impact of different change laws on the power load of the power grid, thereby enhancing the sensitivity analysis and reliability assessment of the power grid fault prediction results, and then provide a basis for the scheduling and planning of the power grid system according to the predicted load demand at different times.

[0061] Embodiment 2

[0062] This embodiment combines the DC modeling system based on the response of the AC / DC hybrid system of the first embodiment to propose a DC modeling method based on the response of the AC / DC hybrid system. Figure 2 As shown, the details are as follows:

[0063] Obtain data information on the preset power grid operating status, and build an AC / DC hybrid power impedance model based on the data information, while dividing the data information into stages.

[0064] Characteristic data in the data information is collected, and the characteristic data is divided into at least three groups, and the three groups of characteristic data are adjacent data. At the same time, in the adjacent data, an incremental combination method is used to analyze the change law of the preset power grid operation state in the three groups of characteristic data, and the change law is uploaded to the impedance model, and the impedance model is updated at the same time.

[0065] The transition time of the preset power grid operation state under different sets of data characteristics is analyzed and calculated, the change characteristics of the preset power grid operation state during different transition times are analyzed, and the Q-learning algorithm is used for reinforcement learning.

[0066] The change characteristics are marked with different risk warning values, and the safety threshold is preset based on the maximum risk warning value. When the change characteristics of the preset power grid during the transition period exceed the safety threshold, the system determines that the operation status of the preset power grid is abnormal and issues an early warning. Otherwise, no judgment is made and the early warning information is uploaded to the impedance model.

[0067] Based on the grouped data, it is used to monitor the correlation between the changes in the grouped data of the preset power grid in the future period and the changes in the preset power grid operation status.

[0068] The communication base stations and / or communication base station operators that match the preset power grid are recorded, and the field data uploaded by the collection monitoring points are received. When the changing characteristics of the preset power grid during the transition period exceed the safety threshold, an information acquisition request for the preset power grid warning status is initiated to the communication base stations and / or communication base station operators corresponding to the preset power grid.

[0069] In summary, this embodiment constructs an AC / DC hybrid power impedance model, and by inputting fault text instructions that are different from the preset historical operation data of the power grid during 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 during the modeling process, and the impedance model can be updated and optimized based on this, thereby obtaining the impact of different change laws on the power load of the power grid.

[0070] At the same time, it can also analyze the impact of sudden changes in different power indicators within a preset time in different fault text instructions on grid operation based on the data differences of seasons and ambient temperature. It is also used to update and optimize the impedance model, enhancing the sensitivity analysis and reliability assessment of grid fault prediction results.

[0071] The technical solutions of the present invention are not limited to the above-mentioned embodiments, and any technical solutions obtained by equivalent replacement methods fall within the scope of protection required by the present invention.

Claims

1. A DC modeling system based on the response of an AC / DC hybrid system, characterized in that: include: Information acquisition module; Used to obtain data information of a preset power grid operation status, and 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; Used to collect characteristic data in the data information, and divide the characteristic data into three groups, wherein the three groups of characteristic data are adjacent data, and at the same time, in the adjacent data, use an incremental combination method to analyze the change law of the preset power grid operation state in the three groups of characteristic data, and upload the change law to the impedance model, and update the impedance model at the same time; Clustering data analysis module; Used to analyze and calculate the transition time of the preset power grid operation state under different sets of data characteristics, and analyze the change characteristics of the preset power grid operation state during different transition times; An early warning assessment module; used to mark the change characteristics with different risk warning values, and preset 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 an early warning. Otherwise, no determination is made and the early warning information is uploaded to the impedance model at the same time. Monitoring terminal; Used to record the communication base stations and / or communication base station operators that match the preset power grid. When the changing characteristics of the preset power grid during the transition period exceed the safety threshold, a request for obtaining information on the preset power grid 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 the 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 a low fault state, a medium fault state and a high fault state, and collecting the operating characteristics of the preset power grid under different fault states, uploading the operating characteristics to the impedance model, and updating the impedance model.

3. The DC modeling system based on the response of the AC / DC hybrid system according to claim 1, characterized in that: Input a plurality of fault instructions and / or text data in different fault states, 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 , where n represents the nth fault instruction and / or text data, analyzes different fault states, and presets the operating state and / or trend of the power grid when different fault instructions and / or text data are input.

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

5. The DC modeling system based on the response of the 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 operation state into the transition time in summer and autumn, and obtain the change characteristics of the preset power grid operation 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 operation state into the transition time in winter and spring. The change characteristics of the preset power grid operation state are collected based on the average ambient temperature of the area 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 preset power grid power load, and the fluctuation change characteristics of the power load are uploaded to the impedance model.

6. The DC modeling system based on the response of the AC / DC hybrid system according to claim 5, characterized in that: The collected average ambient temperatures in winter and summer are divided into steps based on a difference of 3 to 5°C per phase, and the divided data are at least 3. Fault instructions and / or text data are input into the divided data, and the preset power grid operation status changes are obtained at the same time.

7. The DC modeling system based on the response of the AC / DC hybrid system according to claim 6, characterized in that: The fault instructions and / or text data input in the divided data include fault instructions and / or text data of current and / or voltage, wherein the fault instructions and / or text data include fault instructions and / or text data of current and / or voltage sag and swell, and an observation period is preset, and the influence of the current and / or voltage sag and swell on the preset power grid operation state during this period is calculated, and the calculation is obtained according to the following formula: Among them, E i represents the i-th preset observation period, wherein the preset duration of the observation period is at least 6 to 10 minutes, and X represents a fault instruction and / or text data parameter; Wherein, 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 sag and / or swell difference; And within the preset time range of the observation period, a preset risk threshold for the power load fluctuation of the preset power grid is set based on the maximum sag data and / or sag 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 operating state of the preset power grid is abnormal, and issues an early warning, and uploads the early warning information to the impedance model at the same time. Otherwise, no determination is made.

8. The DC modeling system based on the response of the AC / DC hybrid system according to claim 7, characterized in that: The data of the current and / or voltage sag and swell are used to generate a data set [ρ a *ρ b *ρ c *ρ n ], wherein n represents the nth data set, and in any one of the data sets, the data values ​​of the current and / or voltage sag and swell are respectively calculated for sag and / or swell with a difference of 10 numerical units, so as to obtain the preset grid power load fluctuation value during the calculation process.

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

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 claimed in claim 1, characterized in that: The following steps are involved: Acquire data information of a preset power grid operation state, and construct an AC / DC hybrid power impedance model based on the data information, while dividing the data information into stages; Collect characteristic data in the data information, and divide the characteristic data into at least three groups, wherein the three groups of characteristic data are all adjacent data, and at the same time, in the adjacent data, use an incremental combination method to analyze the change law of the preset power grid operation state in the three groups of characteristic data, and upload the change law to the impedance model, and update the impedance model at the same time; Analyze and calculate the transition time of the preset power grid operation state under different sets of data characteristics, and analyze the change characteristics of the preset power grid operation state during different transition times; The change characteristics are marked with different risk warning values, and a safety threshold is preset based on the largest 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 an early warning. Otherwise, no determination is made and the early warning information is uploaded to the impedance model at the same time. Based on the grouped data, it is used to monitor the correlation between the changes in the grouped data of the preset power grid in the future period and the changes in the operation status of the preset power grid; The communication base stations and / or communication base station operators that match the preset power grid are recorded, and when the preset power grid exceeds the safety threshold due to the changing characteristics during the transition time, an information acquisition request for the preset power grid warning status is initiated to the communication base stations and / or communication base station operators corresponding to the preset power grid.

Images