Monitoring and protection system and method for power grid planning based on multi-source data

Abstract

The application discloses a monitoring and protection system and method for power grid planning based on multi-source data, and relates to the technical field of data management. Firstly, the multi-source data of a target area are collected and fused in real time, a short-term prediction model is used to generate a planning output prediction distribution, and each potential high-pressure area of the power grid is identified. Then, graded measuring point enhancement protection is implemented on the high-pressure area, and a pressure distribution map is generated to visualize the risks of each area. Finally, real-time observation data and the prediction distribution are compared and analyzed to determine the actual operating state of the high-risk area, and a quick response measure is dynamically started to timely intervene in potential over-limit or abnormal fluctuations, which can realize real-time sensing and response to changes in power grid pressure caused by local rapid load growth or new energy access, improve the safety and reliability of power grid operation, shorten the response lag caused by the traditional planning update cycle, and effectively support stable operation and planning optimization of the power grid.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data management, in particular to a monitoring and protection system and method for power grid planning based on multi-source data. BACKGROUND

[0002] In regional power grids with high proportion of new energy access and rapid load growth, such as coastal photovoltaic / wind power base centralized grid connection, urban data center group and large-scale charging pile rapid operation distribution network, as well as intelligent microgrid or urban distribution pilot requiring frequent short-time frequency modulation, reactive power support and load scheduling. In these scenarios, the power grid planning must be able to real-time perceive new energy output fluctuation, local load surge, equipment health change and external environmental disturbance.

[0003] For example, the publication number is: CN116976611A, a kind of power grid planning test method, device, electronic equipment and storage medium are disclosed. Method includes: constructing virtual model of power grid, and constructing communication interaction channel between virtual model of power grid and real power grid equipment;According to the debugging instruction generated by the power grid planning scheme;The debugging instruction is input into the virtual model of power grid, and the debugging instruction is sent to the corresponding real power grid equipment through the communication interaction channel, to control the real power grid equipment to test operation according to the debugging instruction;Obtain the test running result data of real power grid equipment;According to the test running result data, in combination with the limit protection threshold value of the pre-determined device parameter of real power grid equipment, determine whether the real power grid equipment meets the requirements of the power grid planning scheme.

[0004] For example, the publication number is: CN119940965A, a kind of power grid line path planning method, device, equipment and medium are disclosed. Obtain the geographic space data of a preset area;According to the geographic space data, generate a planning layer, and set a protection area on the planning layer;Obtain the path planning scheme, generate a preset path on the planning layer according to the path planning scheme;Determine the positional relationship between the preset path and the protection area, and output the planning layer and the preset path if the preset path does not pass through the protection area.

[0005] The above-mentioned technology at least has the following technical problems:

[0006] In regional power grids with high proportion of new energy access and rapid load growth, the accessed new energy may cause local voltage instability, power ramping too fast and other abnormalities without sufficient prediction and monitoring, making it difficult for the dispatching system to take timely mitigation measures, thereby increasing power supply uncertainty and dispatching risk, increasing the pressure on the original planning transmission and distribution path, and even causing local power outage or cascading failure, which seriously affects the safety, reliability and dispatching flexibility of the power grid. SUMMARY

[0007] In order to solve the technical problems existing in the prior art, the embodiment of the present application provides a monitoring and protection system for power grid planning based on multi-source data.

[0008] The power grid planning dynamic monitoring and short-term prediction module is used for receiving and fusing the multi-source data of the target area in real time, generating the planning output prediction distribution data of the power supply side in a predetermined time window through a short-term prediction model, and identifying each potential high-pressure area of the power grid.

[0009] The stress evaluation module is used for analyzing the load data of each node of each potential high-pressure area of the power grid, obtaining the overrun probability factor of each potential high-pressure area of the power grid, performing the measurement point enhancement protection on each potential high-pressure area of the power grid, and generating the stress distribution map of each potential high-pressure area.

[0010] The quick-action resource scheduling and execution module is used for comparing and analyzing the real-time data of each potential high-pressure area of the power grid, and starting the quick-action response measure for each potential high-pressure area of the power grid.

[0011] The second aspect of the present application also provides a monitoring and protection method for power grid planning based on multi-source data, which comprises the following steps of: receiving and fusing the multi-source data of the target area in real time, generating the planning output prediction distribution data of the power supply side in a predetermined time window through a short-term prediction model, and identifying each potential high-pressure area of the power grid.

[0012] The load data of each node of each potential high-pressure area of the power grid is analyzed, the overrun probability factor of each potential high-pressure area of the power grid is obtained, the measurement point enhancement protection is performed on each potential high-pressure area of the power grid, and the stress distribution map of each potential high-pressure area is generated.

[0013] The real-time data of each potential high-pressure area of the power grid is compared and analyzed, and the quick-action response measure is started for each potential high-pressure area of the power grid.

[0014] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects:

[0015] (1) The application proposes a monitoring protection system for power grid planning based on multi-source data. Firstly, by real-time collection and fusion of multi-source data of the target area, a short-term prediction model is used to generate a planning output prediction distribution, so as to identify each potential high-pressure area of the power grid. Then, the stress evaluation module analyzes the node load data of these high-pressure areas, implements graded measurement point enhancement protection, and generates a stress distribution map to visualize the risk of each area. Finally, the quick action resource scheduling and execution module compares and analyzes the real-time observation data and the prediction distribution, judges the actual operation state of the high-risk area according to the deviation and stress index, and dynamically starts the quick response measures. Unlike traditional point prediction or single data source, the application explicitly models uncertainty and directly participates in power flow calculation, making the identification of potential high-pressure areas a probabilistic criterion. Before the occurrence of new energy or load mutation events, it can identify high tail risk time windows in advance (improve early warning lead time), reduce false positive rate caused by single point noise, and provide basis for subsequent decision-making, significantly improving the observability and controllability of short-term extreme disturbance.

[0016] (2) The application inputs multi-source data of the power grid power supply side into a short-term prediction model, generates a planning output prediction distribution in a predetermined time window, and obtains potential high-pressure factors of each regional power grid. Unlike traditional static inspection or fixed frequency telemetry, this method realizes a risk-driven sampling strategy. When the potential high-pressure factor of each regional power grid rises, the reporting frequency of key inverters, bus stations and PMUs is automatically increased and edge event identification is enabled. This enables the power grid to capture key information with high resolution at a minimum communication and calculation cost during new energy or load mutation events, reduces false negatives, ensures the bounded credibility of subsequent power flow correction and quick response input data, and reduces false actions or overly conservative operations caused by insufficient data.

[0017] (3) The application obtains the over-limit probability factor of each potential high-pressure area of the power grid, combines multi-source heterogeneous data and short-term probability prediction, and generates a dynamic multi-dimensional stress distribution map. Unlike existing two-dimensional load / power flow views, it visually helps to combine probability risk, sensitivity exposure and historical vulnerability into an interactive topological layer. On the other hand, through hierarchical trigger measurement point enhancement and quick response, the decision-making time from early warning to measure issuance is significantly shortened, and verifiable operation evidence is provided for subsequent long-term planning (such as adding energy storage, transforming lines, and adjusting access conditions).

[0018] (4) The application can compare the real-time deviation sequence with the equivalent disturbance curve mapped to the receiving end by the short-term prediction sample path in the form of similarity, which helps the system trigger the fast response according to the predefined level, and can reduce the false triggering and resource waste to the maximum extent while ensuring that the protection action has a basis, which is manifested as higher disposal hit rate, fewer false fast dynamic calls, and more controllable energy storage and inverter stress. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0020] Figure 1 is a system module schematic diagram provided by the embodiment of the present application;

[0021] Figure 2 is a method flowchart of the monitoring protection system for power grid planning based on multi-source data provided by the embodiment of the present application;

[0022] Figure 3 The method schematic diagram provided by the embodiment of the present application. DETAILED DESCRIPTION

[0023] The technical solutions in the present application will be described below with reference to the drawings.

[0024] In the embodiments of the present application, the words such as "example", "for example" are used to represent as an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0025] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0026] In the embodiments of the present application, sometimes the subscript such as W1 may be written in the form of non-subscript such as W1, and the meanings expressed thereby are consistent when the difference is not emphasized.

[0027] To make the technical problems, technical solutions and advantages to be solved by the present application clearer, specific embodiments will be described in detail below with reference to the drawings.

[0028] As shown in the Figure 1 embodiments of the present application, a monitoring and protection system for power grid planning based on multi-source data is provided, which comprises: a power grid planning dynamic monitoring and short-term prediction module, configured to receive and fuse multi-source data of a target area in real time, generate planning output prediction distribution data of a power supply side in a predetermined time window through a short-term prediction model, and identify each potential high-pressure area of the power grid.

[0029] A stress evaluation module is configured to analyze load data of each node of each potential high-pressure area of the power grid, obtain an overrun probability factor of each potential high-pressure area of the power grid, perform point enhancement protection on each potential high-pressure area of the power grid, and generate a stress distribution map of each potential high-pressure area.

[0030] A quick-response resource scheduling and execution module is configured to compare and analyze real-time data of each potential high-pressure area of the power grid, and start a quick-response measure for each potential high-pressure area of the power grid.

[0031] As shown in the Figure 2 embodiments of the present application, a monitoring and protection system for power grid planning based on multi-source data is provided, which comprises: a power grid planning dynamic monitoring and short-term prediction module, configured to receive and fuse multi-source data of a target area in real time, generate planning output prediction distribution data of a power supply side in a predetermined time window through a short-term prediction model, and identify each potential high-pressure area of the power grid. Figure 2 The process chart of the monitoring and protection system for power grid planning based on multi-source data provided by the embodiments of the present application is shown in the

[0032] embodiments of the present application, the system first receives and fuses multi-source data of a target area in real time, generates power supply side planning output prediction distribution through a short-term prediction model; then extracts monitoring quantities of each potential high-pressure area, calculates an overrun probability factor and judges a risk level, and implements point enhancement protection on different risk areas; generates a monitoring quantity deviation sequence curve in a monitoring time period, and compares the curve with a predicted output curve in terms of similarity, determines whether to start a quick-response or to maintain normal monitoring according to the curve similarity; finally, verifies the effect of the quick-response measure, including deviation reduction and recovery rate parameters, thereby forming a dynamic monitoring, prediction warning and quick-response closed-loop mechanism covering potential high-pressure areas. The power supply side planning output prediction distribution is generated through the short-term prediction model, and the specific process is as follows: a time window is preset, multi-source data of a power supply side of the power grid in the time window is input into the short-term prediction model, and planning output prediction distribution data of the power supply side in a predetermined time window is output, including predicted values of monitoring quantities of each potential high-pressure area of the power grid, mean values and standard deviations corresponding to the monitoring quantities.It should be noted that in the power grid monitoring and planning protection system, the generation of the planning output prediction distribution of the power supply side in the predetermined time window by the short-term prediction model first needs to collect and integrate multi-source data, including historical meteorological data, real-time meteorological observations, high-frequency PMU waveform data, and inverter telemetry data. These data provide a comprehensive information base for the prediction model, among which the historical meteorological data are used to establish long-term output characteristics and installed capacity models, and the real-time meteorological observations are used to capture short-term disturbances such as wind speed mutations, cloud shadow movements, and wind shear. Among them, factors such as wind speed mutations and cloud shadow movements are quantified by converting real-time meteorological observation data into numerical characteristic quantities during model input, for example, wind speed mutations can be obtained by calculating the difference or acceleration index of wind speed within a certain time window to obtain the instantaneous change rate, and cloud shadow movements can be calculated by satellite or radiation sensor data to calculate the cloud cover rate and light intensity decay curve, and then combined with geographic coordinates to map the local irradiance changes of different power generation units. These numerical indicators are standardized and input into the short-term prediction model as time series features, enabling the model to capture the short-term fluctuation trend of new energy power supply and its immediate impact on output, thereby achieving accurate generation of the planning output prediction distribution. PMU waveform and inverter data are used to reflect the instantaneous dynamic behavior of new energy power supply in actual operation. After preprocessing, these information is input into the short-term prediction model to capture the dynamic output characteristics of the power supply side within a predetermined time window (such as 5 minutes, 15 minutes, 30 minutes, or several hours) in the future. The short-term prediction model usually adopts a time series prediction model, such as the autoregressive moving average model (ARIMA), and the output is the predicted output sequence or prediction distribution of the power supply side within the specified time window in the future. The prediction distribution not only provides the predicted value of the monitoring quantity at each time point, but also calculates the corresponding uncertainty indicators, including the mean μ (the average power, average injected voltage, or average ramp rate of the node or line within the prediction time window) and standard deviation σ (the possible fluctuation range of the quantity within the prediction time window, represented by the standard deviation. The larger the standard deviation, the more uncertain the future behavior of the node or line, and the higher the risk.) of the monitoring quantity corresponding to each potential high-pressure area in the power grid. The mean reflects the expected output level, and the standard deviation reflects the short-term fluctuation range or the size of the prediction error, which is applied to the over-limit probability calculation and node pressure assessment to ensure that the short-term fluctuations of the power supply side output are quantified and controllable in subsequent high-pressure area identification.

[0033] It should be noted that the monitoring quantity (operating parameter) of each potential high pressure area of the power grid includes but is not limited to node voltage, line power flow, busbar injection power, short-time load, power ramp rate, etc., that is, the operating quantity of each node or line of each potential high pressure area of the power grid (in each identified potential high pressure area, the actual operating state data of all key nodes (such as substation busbar, important access point) and lines, including voltage, power flow, injection power, load change and power ramp rate, etc.). The mean and standard deviation corresponding to the monitoring quantity are parameters obtained for the prediction distribution of a specific operating quantity such as busbar injection active power, node voltage, line power or ramp rate in a specified prediction window: the mean represents the expected value (or window average) of the operating quantity in the time window, and the standard deviation represents the fluctuation scale (or prediction uncertainty) around the mean.

[0034] Specifically, each potential high pressure area of the power grid is identified, and the specific process is as follows: the overall voltage deviation, line utilization rate, power flow sensitivity and historical abnormal frequency of each region in the target area are obtained, weighted fusion is performed to obtain the potential high pressure factor of the power grid of each region in the target area, and the potential high pressure factor of the power grid of each region in the target area is used to quantitatively evaluate the overload risk and stability of the region in operation in a predetermined time window.

[0035] It should be noted that the division of the region should not only conform to the physical topology of the power engineering, but also take into account the actual needs of operation, planning and data acquisition: generally, the target area is divided into several initial modules according to the power grid topology as the main line (to ensure electrical connection and relatively concentrated power flow path in each region), and then combined with load and new energy access density, important user and key load distribution, historical event hot area and future construction planning to merge these initial modules; on this basis, the nodes extremely sensitive to external injection are identified by using power flow sensitivity and transmission impact (such as PTDF / LODF), and they are separately divided into sensitive sub-areas for focused monitoring; at the same time, the measurement points and communication coverage, operation and maintenance responsibility boundary and dispatching unit jurisdiction are considered to ensure that each region is manageable in data acquisition, control issuance and emergency response.

[0036] It should be noted that the weighted fusion, the specific process is to first preprocess and normalize each original index (for example, based on engineering threshold or quantile, convert voltage deviation, line utilization, power flow sensitivity, historical abnormal frequency, etc. to exposure score in 0-1 interval, and interpolate or conservatively correct abnormal values and missing data), secondly, each parameter is assigned a weight, which can be given by domain experts experience (such as setting higher weights for power flow sensitivity and line utilization to reflect conductivity and carrying limit), or automatically calibrated through historical event playback and machine learning (training model with event occurrence and non-occurrence labels and reading feature importance); Finally, aggregate into regional level potential high pressure factors of power grid in a weighted sum manner, thereby providing quantitative basis for subsequent threshold determination, measurement point enhancement and quick response.

[0037] The potential high pressure factors of each regional power grid in the target area are compared with the first threshold value of the potential high pressure factors stored in the database. If the potential high pressure factor of a regional power grid in the target area is higher than or equal to the first threshold value of the potential high pressure factor, the region is marked as a potential high pressure region, thereby obtaining each potential high pressure region of the power grid. On the contrary, if the potential high pressure factor of a regional power grid in the target area is lower than the threshold value of the potential high pressure factor, the region does not need to be marked as a potential high pressure region.

[0038] Specifically, the over-limit probability factor of each potential high pressure region of the power grid is obtained, and the specific process is as follows: the monitoring quantity of each potential high pressure region of the power grid in a predetermined time window is obtained, and the mean value and the standard deviation of the monitoring quantity of each potential high pressure region of the power grid in the planning output prediction distribution of the power supply side output by the short-time prediction model in the predetermined time window are extracted, then the over-limit probability of all monitoring quantities in each potential high pressure region of the power grid is arithmetically averaged and normalized to obtain the over-limit probability factor of each potential high pressure region of the power grid.

[0039] It should be noted that the over-limit probability factor of each potential high pressure region of the power grid is analyzed under the following conditions:

[0040] ;

[0041] In the formula, P k represents the over-limit probability factor of the kth potential high pressure region of the power grid, φ represents the cumulative distribution function of the standard normal distribution, L i represents the limit value corresponding to the ith monitoring quantity stored in the database, μ ki represents the mean value corresponding to the ith monitoring quantity of the kth potential high pressure region of the power grid, σ ki represents the standard deviation corresponding to the ith monitoring quantity of the kth potential high pressure region of the power grid, P 2ian over-limit probability factor of a potential high stress region in the database, k represents a number of the potential high stress region, k = 1, 2, 3,..., n, n represents a total number of the potential high stress region, i represents a number of the monitoring quantity, i = 1, 2, 3,..., m, and m represents a total number of the monitoring quantity.

[0042] It should be noted that the mean and the standard deviation of the monitoring quantity are a centralized level and a fluctuation amplitude of the operation state of the power grid in a future period of time, which determine a possible distribution interval of the power output prediction on the power supply side in different regions; the mean reflects a central tendency of the prediction, and the standard deviation reflects an uncertainty range; when the mean is closer to the corresponding limit value and the standard deviation is larger, the possibility of the monitoring quantity exceeding the safety limit is higher, thereby increasing the over-limit probability of the region; and when the mean limit value is far away and the standard deviation is small, the over-limit probability is naturally low. The over-limit probability factor obtained by quantization and comparison is used to intuitively evaluate and compare the operation risk levels of different potential high stress regions in a predetermined time window, and the larger the over-limit probability factor is, the more likely the monitoring quantity is to break through the safety threshold in the future, indicating that the region has a higher overload risk.

[0043] Specifically, the point enhancement protection is performed on each potential high stress region of the power grid, and the specific process is as follows: the over-limit probability factor of each potential high stress region of the power grid is compared with the over-limit probability factor threshold of the potential high stress region stored in the database; if the over-limit probability factor of a certain potential high stress region of the power grid is higher than or equal to the over-limit probability factor threshold of the potential high stress region, and the power grid potential high stress factor of the certain potential high stress region of the power grid is higher than or equal to the second threshold of the potential high stress factor stored in the database, the power grid potential high stress region is marked as a first risk region; if the over-limit probability factor of a certain potential high stress region of the power grid is higher than or equal to the over-limit probability factor threshold of the potential high stress region, and the power grid potential high stress factor of the certain potential high stress region of the power grid is lower than the second threshold of the potential high stress factor stored in the database, the power grid potential high stress region is marked as a second risk region; and if the over-limit probability factor of a certain potential high stress region of the power grid is lower than the over-limit probability factor threshold of the potential high stress region, the power grid potential high stress region is marked as a third risk region.

[0044] The point enhancement protection is performed on each first risk region, each second risk region, and each third risk region of the power grid.

[0045] It should be noted that the specific process of point enhancement protection according to each first risk area, each second risk area and the third risk area of the power grid is as follows: for the point enhancement protection of each first, second and third risk area of the power grid, the system first classifies and labels each type of risk area according to the early risk assessment results, including the calibration results and confidence estimates of the short-time prediction model on the historical playback, and each type of threshold, mapping table and response strategy (such as the over-limit probability threshold, the potential high-pressure factor threshold, the sampling-response mapping rule, etc.) saved in the database, wherein the first risk area is a high-risk node or line, the second risk area is a medium-risk node or line, and the third risk area is a low-risk node or line. When implementing the point enhancement protection, the system first obtains the over-limit probability factor of each node and line and the difference increment of the over-limit probability factor and the risk threshold in the database (i.e. the difference between the power grid potential high-pressure factor of a potential high-pressure area of the power grid and the potential high-pressure factor second threshold, and the effective increment is considered only when the difference is positive for k consecutive evaluation windows). This difference reflects the over-limit amplitude of the current node or line risk level. The system inputs the difference into the pre-set over-limit-sampling response mapping table, which is a rule base obtained based on historical operation data playback and simulation experiments. The mapping table output includes the reporting frequency increment of key inverters and bus stations, the sampling frequency adjustment suggestion of key points, the upper limit of the duration of enhanced sampling, and the trigger sensitivity of the event identification algorithm. The mapping process considers the node voltage fluctuation, current ramp rate, power fluctuation and edge computing resource availability to ensure that high-risk nodes can capture abnormal output or protection misoperation signs in a short time window, while ensuring that the data input to the power flow correction and pressure evaluation is reliable and bounded; for the second risk area and the third risk area, the regular point sampling frequency and data reporting strategy are maintained, and only the obvious deviation or long-term trend anomaly is recorded and statistically analyzed, thereby forming a point enhancement protection system covering the whole network, grading and dynamically adjustable, realizing differentiated monitoring and rapid response of high, medium and low risk areas, and effectively supporting power grid operation safety and dynamic planning protection.

[0046] Specifically, a pressure distribution map of each potential high-pressure area is generated, and the specific process is as follows: the over-limit probability factor of each potential high-pressure area of the power grid and the operating parameters of each potential high-pressure area of the power grid are extracted, the over-limit probability factor of each potential high-pressure area is mapped to the topology information of the power grid, and a multi-dimensional pressure distribution map of each potential high-pressure area is generated in combination with the operating parameters of each potential high-pressure area of the power grid.

[0047] It should be noted that in the stage of performing the measurement point enhanced protection or generating the multi-dimensional stress distribution map, all potential high stress areas need to be uniformly processed, not just the use of hierarchical labels. The hierarchical results are mainly used for risk classification and early warning decision-making, while in the further topology mapping and visualization link, different categories of areas need to be re-summarized under the overall framework of potential high stress areas for overall topology modeling and parameter analysis. In other words, classification is the purpose of the previous step, while mapping and distribution map generation are the purpose of consistent modeling based on all areas.

[0048] It should be noted that the power grid topology information is essentially a structured model of the power grid, including the connection relationship of nodes (busbars, substations, junction points, etc.), branches (lines, transformers), generators and loads, as well as their connectivity, impedance, rated capacity and other attributes. The power grid topology information is usually stored in the database of the dispatch center, including static and dynamic information of the entire network busbar, transformer, line, access point, etc., which is directly used as the basis for topology modeling; secondly, the geographic information system (GIS) of the distribution network or microgrid provides node geographic coordinates, line path and spatial layout, which can be combined with EMS information to form a spatialized topology; finally, if there is new energy access or rapid change scenarios, the newly added new energy station, charging pile group, data center load access point needs to be dynamically updated to the topology. Topology information comes from the operation database of the power grid, GIS spatial data and planning / design materials, and after integration, it forms a full network topology model that can be used to map the over-limit probability factor and operating parameters, for stress distribution visualization and protection strategy design.

[0049] It should be noted that mapping the factor to the topology information can be understood as labeling the over-limit probability factor value of each potential high stress area to the topology node or branch of the power grid: for example, mapping the risk factor values corresponding to the voltage, line flow, etc. of the main nodes in the area to the node weight or edge weight, and then forming a network graph with risk labels on the power grid topology graph, which can intuitively show which parts have higher risk values and which parts have lower risk values, and in combination with the process of operating parameters, not only the over-limit probability factor is displayed in topology mapping, but also the operating state information of the area is introduced, such as voltage deviation, line utilization rate, power flow sensitivity, historical abnormal frequency, etc. These indicators and over-limit probability factors are used as multi-dimensional features of the area, making the risk map more three-dimensional. After mapping is completed, a stress distribution map can be generated through multi-layer visualization, which can use the power grid topology as the base map, and then use color gradient, line thickness, node size or heat map overlay to display the factors and parameters of different areas, thereby forming a multi-dimensional stress distribution map. This map can not only reflect the risk level after classification of the area, but also show the stress distribution of specific operating parameters, achieving panoramic and visual monitoring of the operating stress of the power grid.

[0050] Specifically, real-time data of each potential high-pressure area of the power grid is compared and analyzed, and the specific process is as follows: a monitoring time period is preset, a difference between the monitoring quantity of each potential high-pressure area of the power grid and a predicted value of the monitoring quantity in the monitoring time period is processed to obtain a monitoring quantity deviation sequence of each potential high-pressure area of the power grid in the monitoring time period, and a monitoring quantity deviation sequence curve of each potential high-pressure area of the power grid is generated.

[0051] A predicted output sequence of the power supply side in a future specified time window is obtained, and the predicted output sequence of the power supply side in the future specified time window is plotted as a predicted output time sequence curve of the power supply side in the monitoring time period.

[0052] The monitoring quantity deviation sequence curve of each potential high-pressure area of the power grid and the predicted output time sequence curve of the power supply side are compared to obtain a similarity between the monitoring quantity deviation sequence curve of each potential high-pressure area of the power grid and the predicted output time sequence curve of the power supply side, which is recorded as a curve comparison similarity of each potential high-pressure area of the power grid.

[0053] It should be noted that the specific process of generating the monitoring quantity deviation sequence curve of each potential high-pressure area of the power grid is as follows: in the preset monitoring time period, the system continuously collects the original time sequence data of all target monitoring quantities (such as node voltage, active / reactive power injection, line power, phase angle, ramp rate, etc.) in the region according to a unified time baseline, and performs time synchronization verification and data quality processing (clock alignment, missing value interpolation, Hampel or median filter to remove transient impulse noise, and discarding low-quality samples according to measurement point confidence); at the same time, the expected value sequence of the corresponding monitoring quantity given by the short-term prediction model (or using the mean value of the prediction output as the baseline) is read, the observation value at each time is subtracted from the prediction baseline at the same time to obtain the deviation value sequence; the deviation sequence can be calculated and displayed in parallel, and the original difference curve and the curve smoothed by the sliding window (such as short window moving average and weighted exponential smoothing) are displayed, and the key event points (peak value, mutation point, continuous interval) and their meta information (timestamp, amplitude, measurement point ID, confidence) are marked on the curve, and finally the deviation sequence curve can be used as a direct shape comparison object, or can be used to further extract the kinetic characteristics such as peak value, ramp rate and duration for similarity determination and protection decision calling.

[0054] It should be noted that firstly, the predicted output data of the target power source (such as a certain wind farm or photovoltaic power station) in the monitoring time period is obtained from the short-term prediction module; the original prediction is unfolded and mapped onto the target drawing time axis according to a unified time reference (for example, a timestamp at a second or minute level), and if the prediction resolution is different from the sampling rate of the monitoring time window, resampling and interpolation (linear or spline) are performed on the prediction sequence to align the time grid; for sample paths from multiple models or members, set processing can be performed first, for example, calculating the sample mean at each time point, or extracting the median and upper and lower quantiles as representative curves and confidence bands; before drawing, the prediction sequence is checked for quality (outlier members are removed, delays or missing values are compensated), and smoothing / filtering (short-term low-pass or Kalman filtering) is optionally applied to remove unnecessary high-frequency noise, and finally the processed representative curve (and optional quantile band) is drawn as a power-side predicted output time sequence curve on the same time coordinate as the monitoring time window, and the prediction source, member number and confidence information are retained in the metadata for subsequent similarity comparison and attribution use.

[0055] It should be noted that the similarity between the monitoring quantity deviation sequence curve of each potential high-pressure area of the power grid and the predicted output time sequence curve of the power side is to preprocess the two curves according to a unified time reference, including time alignment, missing value interpolation and normalization processing, mapping the amplitude to the same numerical interval to eliminate the dimensional difference, and constructing a two-dimensional cost matrix, the rows of the matrix correspond to the time points of the monitoring quantity deviation sequence, the columns correspond to the time points of the predicted output sequence, and each matrix element stores the absolute difference or squared difference of the amplitudes of the two sequences at the corresponding time points as the local distance; on the cost matrix, apply the dynamic time warping algorithm, allow nonlinear matching of the time axis, and get the best matching total cost of the two curves in the entire time period through the cumulative minimum distance path, which is the similarity of the curve comparison.

[0056] Specifically, the rapid response measures for each potential high-pressure area of the power grid are started, and the specific process is: the curve comparison similarity of each potential high-pressure area of the power grid is extracted, and compared with the response level corresponding to each curve comparison similarity interval stored in the database, to obtain the response level of each potential high-pressure area of the power grid, and the rapid response measures are started according to the response level of each potential high-pressure area of the power grid.

[0057] It should be noted that the specific process of starting the quick response measure according to the response level of each potential high-pressure area of the power grid is that the response level of each potential high-pressure area of the power grid is sorted in descending order according to the current over-limit probability factor of the area of the power grid to obtain the final emergency ranking of the area. Then, the control module searches for available quick response resources (including energy storage SOC and instantaneous available power, grid-connected inverter power and adjustable range, local switchable capacitor / conversion equipment, callable demand response capacity and fast start-stop units) in the local and regional resource pool, and selects the candidate action package of the corresponding level in the preset response scheme library based on the response level. Each candidate contains an action sequence (source end amplitude limiting, inverter Q adjustment, BESS discharge, switching capacitor, temporary topology adjustment, DR pre-announcement, unit start-stop, etc.), expected effect estimation, required resources and upper limit of duration.

[0058] Specifically, starting the quick response measure for each potential high-pressure area of the power grid also includes verifying the effect of the quick response measure for each potential high-pressure area of the power grid. The specific process is as follows: a response measure effect monitoring window is preset, and after the starting quick response measure for each potential high-pressure area of the power grid is obtained in the response measure effect monitoring window, the curve comparison similarity of each potential high-pressure area of the power grid and the potential high-pressure factor of each potential high-pressure area of the power grid are obtained.

[0059] If the curve comparison similarity of the starting quick response measure for each potential high-pressure area of the power grid is higher than or equal to the curve comparison similarity threshold stored in the database and the potential high-pressure factor of the potential high-pressure area of the power grid is lower than the potential high-pressure factor threshold, the effect of the starting quick response measure for the potential high-pressure area is marked as successful, otherwise the effect of the starting quick response measure for the potential high-pressure area is marked as failed.

[0060] It should be noted that, yes, the curve contrast similarity is higher than the threshold value and the potential high pressure factor is higher than the threshold value; the curve contrast similarity is lower than the threshold value and the potential high pressure factor is higher than the threshold value; the curve contrast similarity is lower than the threshold value and the potential high pressure factor is lower than the threshold value. The three situations should be judged as failure, because both the deviation is indeed matched with the new energy output fluctuation (curve similarity >= threshold value) and the regional vulnerability has been effectively reduced (potential high pressure factor < threshold value) should be proved; as long as one item is missing, the response has not completed its dual goal. The specific reasons are as follows: the first kind (high similarity but high pressure factor) shows that the action hits the source of disturbance correctly, but the quick action measure issued is not enough to reduce the exposure of the region and the out-of-limit probability to the safety zone, which may be due to insufficient resource capacity, too small action amplitude or response delay, and thus needs to be upgraded or supplemented; the second kind (low similarity and high pressure factor) shows that the region is still highly vulnerable and the current deviation is not caused by the expected new energy output, indicating that neither the vulnerability is relieved nor the real root cause is hit, and the response strategy must be adjusted by returning to the diagnosis stage to find other fault sources; the third kind (low similarity and low pressure factor) shows that the response does not solve the observed deviation (either the deviation comes from other causes or the response itself does not take effect or produces side effects), so it should also be judged as failure to avoid covering up the wrong disposal or missing the hidden risks. Any situation that does not meet one of the two conditions is not a qualified disposal and should be marked as failure and trigger the corresponding escalation, backtracking or manual verification process.

[0061] It should be noted that if the start-up quick response measure of a potential high pressure region is marked as failure, the event is immediately upgraded to the senior dispatch level and the corresponding engineer is notified, and the reversible action in the previously issued start-up quick response measure is cancelled (such as cancelling the amplitude limit, stopping the discharge of the BESS) and manual intervention is attempted.

[0062] As shown in Figure 3 The second aspect of the present application also provides a method for monitoring and protecting power grid planning based on multi-source data, which comprises: receiving and fusing multi-source data of a target area in real time, generating planning output prediction distribution data of the power supply side within a predetermined time window through a short-term prediction model, and identifying each potential high pressure region of the power grid.

[0063] The load data of each node of each potential high pressure region of the power grid is analyzed to obtain an out-of-limit probability factor of each potential high pressure region of the power grid, a measurement point enhancement protection is performed on each potential high pressure region of the power grid, and a pressure distribution map of each potential high pressure region is generated.

[0064] The real-time data of each potential high pressure region of the power grid is compared and analyzed, and a start-up quick response measure of each potential high pressure region of the power grid is started.

[0065] It should also be understood that the memory in the embodiments of the present application can be volatile or nonvolatile memory, or can include both volatile and nonvolatile memory. Nonvolatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0066] The above-described embodiments can be implemented in whole or in part by software, hardware (e.g., circuitry), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The 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, the processes or functions according to the embodiments of the present application are entirely or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center through wired (e.g., infrared, wireless, microwave, etc.) or wireless means. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. 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. The semiconductor medium can be a solid-state disk.

[0067] It should be understood that the term "and / or" herein merely describes an association relationship of associated objects, which means that there can be three relationships, for example, A and / or B can represent three cases of A alone, A and B together, and B alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents an "or" relationship between the associated objects before and after it, but it can also represent an "and / or" relationship. The specific meaning can be understood according to the context before and after it.

[0068] In the present application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or similar expressions means any combination of these items, including any combination of single item or multiple items. For example, at least one of a, b, or c can represent a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0069] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined according to their functions and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0070] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed 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 the present application.

[0071] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the devices, apparatuses and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0072] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely schematic, for example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0073] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0074] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0075] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0076] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A monitoring and protection system for power grid planning based on multi-source data, characterized in that, The system includes: The power grid planning dynamic monitoring and short-term forecasting module is used to receive and integrate multi-source data of the target area in real time, generate planned output forecast distribution data of the power source side within a predetermined time window through the short-term forecasting model, and identify potential high-pressure areas of the power grid. The pressure assessment module is used to analyze the load data of each node in each potential high-pressure area of ​​the power grid, obtain the over-limit probability factor of each potential high-pressure area of ​​the power grid, perform enhanced protection of each potential high-pressure area of ​​the power grid, and generate a pressure distribution map of each potential high-pressure area. The rapid resource scheduling and execution module is used to compare and analyze real-time data of potential high-pressure areas of the power grid and initiate rapid response measures for these areas. The specific process for identifying potential high-pressure areas in the power grid is as follows: The overall voltage deviation, line utilization, power flow sensitivity and historical abnormal frequency of each area in the target region are obtained, and weighted fusion is performed to obtain the potential high pressure factor of the power grid in each area of ​​the target region. The potential high pressure factor of the power grid in each area of ​​the target region is used to quantitatively evaluate the overload risk and stability of the region during operation within a predetermined time window. The potential high-pressure factors of the power grid in each region of the target area are compared with the first threshold of potential high-pressure factors stored in the database. If the potential high-pressure factor of a certain region of the target area is higher than or equal to the first threshold of potential high-pressure factors, the region is marked as a potential high-pressure region, thus obtaining the potential high-pressure regions of the power grid. Conversely, if the potential high-pressure factor of a certain region of the target area is lower than the first threshold of potential high-pressure factors, the region does not need to be marked as a potential high-pressure region. The process for obtaining the over-limit probability factors for each potential high-pressure region of the power grid is as follows: Within a predetermined time window, the monitoring data of each potential high-pressure area of ​​the power grid is acquired. The mean and standard deviation of the monitoring data of each potential high-pressure area of ​​the power grid are extracted from the planned output prediction distribution of the power source side within the predetermined time window output by the short-term prediction model. Then, the probability of exceeding the limit of all monitoring data in each potential high-pressure area of ​​the power grid is arithmetically averaged and normalized to obtain the probability factor of exceeding the limit of each potential high-pressure area of ​​the power grid.

2. The monitoring and protection system for power grid planning based on multi-source data according to claim 1, characterized in that, The specific process of generating the planned power output prediction distribution data of the power source side within a predetermined time window through the short-time prediction model is as follows: A preset time window is set, and multi-source data from the power grid source side is imported into the short-term prediction model within the time window. The model outputs the predicted distribution data of the planned output of the power source side within the preset time window, including the predicted values ​​of the power grid monitoring quantities, the mean and standard deviation of the monitoring quantities.

3. The monitoring and protection system for power grid planning based on multi-source data according to claim 1, characterized in that, The specific process for enhancing protection at measurement points in various potentially high-pressure areas of the power grid is as follows: The over-limit probability factor of each potential high-pressure area of ​​the power grid is compared with the over-limit probability factor threshold of potential high-pressure areas stored in the database. If the over-limit probability factor of a potential high-pressure area of ​​the power grid is higher than or equal to the over-limit probability factor threshold of the potential high-pressure area, and the power grid potential high-pressure factor of a potential high-pressure area is higher than or equal to the second threshold of potential high-pressure factors stored in the database, then the potential high-pressure area of ​​the power grid is marked as a first risk area. If the over-limit probability factor of a potential high-pressure area of ​​the power grid is higher than or equal to the over-limit probability factor threshold of the potential high-pressure area, and the power grid potential high-pressure factor of a potential high-pressure area is lower than the second threshold of potential high-pressure factors stored in the database, then the potential high-pressure area of ​​the power grid is marked as a second risk area. If the over-limit probability factor of a potential high-pressure area of ​​the power grid is lower than the over-limit probability factor threshold of the potential high-pressure area, then the potential high-pressure area of ​​the power grid is marked as a third risk area. Enhanced protection measures were implemented for each of the first, second, and third risk areas of the power grid.

4. The monitoring and protection system for power grid planning based on multi-source data according to claim 1, characterized in that, The specific process for generating pressure distribution maps for each potential high-pressure region is as follows: The over-limit probability factors and operating parameters of each potential high-pressure area of ​​the power grid are extracted. The over-limit probability factors of each potential high-pressure area are mapped onto the power grid topology information. Combined with the operating parameters of each potential high-pressure area, a multi-dimensional pressure distribution map of each potential high-pressure area is generated.

5. The monitoring and protection system for power grid planning based on multi-source data according to claim 2, characterized in that, The specific process of comparing and analyzing real-time data from various potentially high-pressure areas of the power grid is as follows: A preset monitoring period is set. During the monitoring period, the monitoring quantities and predicted values ​​of the monitoring quantities of each potential high-pressure area of ​​the power grid are extracted and the difference is processed to obtain the monitoring quantity deviation sequence of each potential high-pressure area of ​​the power grid during the monitoring period, and the monitoring quantity deviation sequence curve of each potential high-pressure area of ​​the power grid is generated. Obtain the predicted power output sequence of the power supply side within a specified future time window, and plot the predicted power output sequence of the power supply side within the specified future time window as a predicted power output time series curve of the power supply side during the monitoring period. The monitoring deviation sequence curves of each potential high-pressure area of ​​the power grid are compared with the predicted output time series curves of the power source side to obtain the similarity between the monitoring deviation sequence curves of each potential high-pressure area of ​​the power grid and the predicted output time series curves of the power source side, which is denoted as the curve comparison similarity of each potential high-pressure area of ​​the power grid.

6. The monitoring and protection system for power grid planning based on multi-source data according to claim 4, characterized in that, The specific process for initiating rapid response measures in various potentially high-pressure areas of the power grid is as follows: Extract the curve similarity of each potential high-pressure area of ​​the power grid and compare it with the response level corresponding to each curve similarity interval stored in the database to obtain the response level of each potential high-pressure area of ​​the power grid. Based on the response level of each potential high-pressure area of ​​the power grid, initiate rapid response measures.

7. The monitoring and protection system for power grid planning based on multi-source data according to claim 6, characterized in that, The aforementioned initiation of rapid response measures in various potentially high-pressure areas of the power grid also includes verifying the effectiveness of these measures. The specific process is as follows: A preset response measure effect monitoring window is set up. In the response measure effect monitoring window, the curve comparison similarity of each potential high-pressure area of ​​the power grid after the start of the fast response measure and the potential high-pressure factor of each potential high-pressure area of ​​the power grid are obtained. If, after the initiation of a rapid response measure in a potentially high-pressure area of ​​the power grid, the curve comparison similarity of that potentially high-pressure area is higher than or equal to the curve comparison similarity threshold stored in the database, and the potential high-pressure factor of that potentially high-pressure area is lower than the potential high-pressure factor threshold, then the effect of the rapid response measure in that potentially high-pressure area is marked as successful; otherwise, the effect of the rapid response measure in that potentially high-pressure area is marked as failed.

8. A method for monitoring and protecting a power grid planning system based on multi-source data as described in any one of claims 1-7, characterized in that, include: Real-time reception and fusion of multi-source data from the target region; generation of planned power output forecast distribution data for the power supply side within a predetermined time window through a short-time forecast model; identification of potential high-pressure areas in the power grid. The load data of each node in each potential high-pressure area of ​​the power grid are analyzed to obtain the over-limit probability factor of each potential high-pressure area of ​​the power grid. Enhanced protection of each potential high-pressure area of ​​the power grid is carried out at the measuring point, and pressure distribution map of each potential high-pressure area is generated. Real-time data from potential high-pressure areas of the power grid are compared and analyzed, and rapid response measures are initiated for each potential high-pressure area.

Images