Power grid multi-dimensional stability monitoring method based on dynamic graph

Abstract

The application discloses a power grid multi-dimensional stability monitoring method based on a dynamic atlas, establishes a multi-level monitoring system of a region-station-grid point-feeder-generating equipment, unifies the per-unit value benchmark of each level and each type of equipment, constructs a voltage safety domain and a PQ safety domain containing multi-level early warning boundaries and limit boundaries, draws a safety domain atlas according to the level and the type of equipment, forms a global panoramic dynamic visual monitoring graph of the station, distinguishes active adjustable margin and passive available margin, and realizes fine state perception. The running point groups of the same type of generating equipment in the PQ plane or the voltage vector plane are fitted by using a smooth curve to form an outer envelope line, the running point closest to the safety domain boundary is identified as a weak point of the same type of generating equipment, and accurate positioning of the weak point is realized. The application solves the problems of the existing system, such as non-fine early warning, insufficient equipment characteristic identification and non-comprehensive stability risk prediction, and significantly improves the safety and stability monitoring and collaborative control level of the new energy high-occupancy power grid.

Description

Technical Field

[0001] This invention relates to the field of power system safety and stability monitoring and control technology, and in particular to a multi-dimensional stability monitoring method for power grids based on dynamic graphs. Background Technology

[0002] With the continuous increase in the proportion of new energy power generation capacity, the power system's source-grid structure is becoming increasingly complex, its operational fluctuation characteristics are significantly enhanced, and the coupling of stability issues is prominent. Traditional monitoring systems are mostly dispersed at various plant and substation levels, lacking a unified view across the entire domain. The electrical quantity benchmarks for different voltage levels and different types of equipment are not uniform, making it difficult to achieve cross-level and cross-equipment comparative analysis. At the same time, existing safety domain monitoring methods lack precise early warning grading, failing to make multi-level predictions and refined adjustments before the operating state approaches its limit boundary, and lacking a precise identification mechanism for weak points in similar power generation equipment groups, resulting in incomplete perception of grid stability, untimely early warnings, and difficulties in fault location. In addition, there is a control switching boundary at the grid connection point voltage. When the grid connection point voltage approaches or exceeds this switching boundary, it will directly trigger a switch in the control strategy and operating mode of the connected power generation equipment. Traditional monitoring methods are unable to visualize and provide early warnings for such boundary effects.

[0003] Traditional wide-area measurement relies on the centralized transmission of massive amounts of raw data. This results in a huge data volume and heavy transmission pressure, making it difficult to support comprehensive, panoramic, and visual monitoring of the entire power grid. This restricts the ability to coordinate and control safety and stability across regions and levels. Moreover, existing monitoring methods mostly perform steady-state analysis based on the effective values ​​of electrical quantities or conduct transient waveform analysis based solely on instantaneous values. The steady-state and dynamic analyses are disconnected, making it difficult to combine the stability margin reflected by the effective values ​​with the waveform distortion and oscillation characteristics reflected by the instantaneous values ​​for unified analysis. This makes it impossible to achieve joint early warning of full-frequency oscillation waveforms and safety boundaries. Summary of the Invention

[0004] The purpose of this invention is to propose a multi-dimensional stability monitoring method for power grids based on dynamic graphs. This method integrates multi-level monitoring, unified per-unit measurement, global graph, hierarchical early warning, weak point identification, on-site-remote collaboration, and visualized decision-making.

[0005] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows:

[0006] According to one aspect of the present invention, a method for multidimensional stability monitoring of power grids based on dynamic graphs is provided, comprising the following steps:

[0007] S1. Collect electrical operation data of power plants, new energy stations, grid connection points, feeders, and power generation equipment within the regional power grid;

[0008] S2. Establish a unified per-unit value benchmark system for grid connection points, feeders, and power generation equipment, and standardize all electrical quantities;

[0009] S3. Based on the standardized multi-dimensional stability indicators, the voltage security domain and PQ security domain are constructed and drawn according to the vertical layering of grid connection point layer, feeder layer and power generation equipment layer, and the horizontal classification of different types of power generation equipment.

[0010] The voltage safety domain is the voltage operating range, which consists of limit boundaries and multi-level early warning boundaries; the PQ safety domain is the active-reactive operating range, which consists of limit boundaries and multi-level early warning boundaries.

[0011] S4. Perform group analysis and envelope fitting on the operating points of the same type of power generation equipment in the voltage vector plane or PQ plane to identify the weak points of the power generation equipment that are closest to the safety domain limit boundary, and use them as the key points with the lowest stability margin and the most likely to exceed the limit.

[0012] S5. Construct a multi-level monitoring system for regions, substations, grid connection points, feeders, and power generation equipment. Based on the relative positions of operating points and multi-level early warning boundaries and extreme boundaries, as well as the identification results of weak points, conduct graded alarms and safety and stability control. Remotely transmit key safety domain information to the centralized control center or dispatch center, and establish a local and remote co-management platform.

[0013] Preferably, in step S2, the unified per-unit benchmark system includes voltage benchmark, power benchmark, and current benchmark. The voltage benchmark corresponds to the grid connection point rated voltage, feeder rated voltage, and generator terminal rated voltage, respectively. The power benchmark corresponds to the rated capacity of a single generator, the total rated installed capacity of the grid connection point, and the total rated capacity of the generator connected to the feeder, respectively. The current benchmark is calculated from the corresponding power benchmark and voltage benchmark.

[0014] Preferably, in step S3, the multi-level early warning boundary is a gradient early warning boundary. The gradient early warning boundary is configured with single-level or multi-level early warning levels according to the operating scenario. The multi-level early warning levels are set in layers from the inside to the outside within the safety domain, forming a gradient early warning range that approaches the limit boundary from the normal operating range.

[0015] Preferably, the early warning method includes the following steps:

[0016] The system provides tiered early warnings based on the points of operation crossing different warning boundaries, and displays these warnings in a dynamic, panoramic monitoring map of the entire station using color differentiation and regional rendering.

[0017] Preferably, in step S5, the key security domain information includes security domain boundaries, early warning boundaries, operating point coordinates, weak point information, and limit-crossing alarm information.

[0018] Preferably, in step S5, the method for processing critical security domain information by the local and remote co-management platform includes the following steps:

[0019] On the ground side, edge computing is used to complete online calculation of the security domain, identification of weak points, early warning judgment, and violation judgment.

[0020] The key features of the safety domain boundary, early warning information, weak point information, and operation point status are compressed, and the feature information of single-point stations is spliced ​​to generate a regional full-domain monitoring view.

[0021] Multi-view display of the overall panoramic map of the station, the operation curves of individual equipment and individual feeders, and layered overlay of limit boundaries and early warning boundaries;

[0022] By integrating steady-state RMS values ​​and transient instantaneous sampling data of electrical quantities, a comprehensive analysis of wideband oscillation waveforms and spectral characteristics is performed.

[0023] Based on the positional relationship between the operating point and the multi-level early warning boundary and extreme boundary, multi-level risk early warning judgment is carried out;

[0024] By combining the results of on-site assessment with the dispatching strategies, on-site control and remote dispatching are carried out, and the control results are fed back to the on-site side.

[0025] Preferably, the method for determining the multi-level risk warning is as follows:

[0026] When the operating point successively crosses the warning boundaries of each level or approaches the limit boundary of the weak point, the corresponding level alarm is triggered, the abnormal location, weak point and its equipment type are located in the map, and the graded control command is executed.

[0027] Preferably, the method further includes the following steps:

[0028] For grid-connected and grid-linked power generation equipment, the active adjustable margin or passive available margin of voltage and power dimensions are visualized and marked in their respective security domains.

[0029] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0030] 1. This invention establishes a multi-level monitoring system encompassing the region, substation, grid connection point, feeder, and power generation equipment. It unifies the per-unit value benchmarks for each level and type of equipment, constructs a voltage safety domain and a PQ safety domain including multi-level early warning boundaries and limit boundaries, and classifies safety domain maps according to level and equipment type. This forms a global, dynamic, and visual monitoring map of the entire substation, distinguishing between active adjustable margin and passive availability margin to achieve refined state perception. For a group of operating points of the same type of power generation equipment in the PQ plane or voltage vector plane, a smooth curve is fitted to form an outer envelope. The operating point closest to the safety domain boundary is identified as a weak point of the same type of power generation equipment, achieving precise weak point location.

[0031] 2. This invention achieves remote transmission of key characteristic quantities through edge computing, data compression, redundant communication and time synchronization, and constructs a platform for local autonomy and remote collaborative management. Based on multi-level early warning boundaries, it realizes hierarchical alarm and control closed loop, effectively solving problems such as the bottleneck of massive data transmission in wide-area measurement systems, imprecise early warning, insufficient identification of equipment characteristics, and incomplete prediction of stability risks, and significantly improving the level of safety and stability monitoring and collaborative control of power grids with a high proportion of new energy.

[0032] 3. This invention adopts a unified per-unit reference system to achieve normalization of electrical quantities across multiple devices, voltage levels, and monitoring levels, eliminating dimensional differences and laying a data foundation for the unified construction of a comprehensive safety domain.

[0033] 4. This invention constructs a multi-level integrated monitoring system covering regions, power plants, grid connection points, feeders, and power generation equipment. Voltage safety domains and PQ safety domains are drawn according to levels and equipment types to form a global dynamic visualization map, enabling full-domain visibility of stable conditions, hierarchical positioning, and classification identification.

[0034] 5. This invention sets up multi-level early warning boundaries within the safety domain to achieve multi-level refined early warning. Alarms are triggered in stages before the operating point approaches the limit boundary, which significantly improves the system's safety redundancy and fault prediction capabilities.

[0035] 6. This invention clearly distinguishes between active adjustable margin and passive available margin, accurately matches the voltage regulation and reactive power regulation characteristics of different power generation equipment, realizes refined perception of equipment-level operating status, and adapts to the differentiated control needs of power grids with a high proportion of new energy.

[0036] 7. This invention uses a smooth curve to fit the outer envelope of the operating group, which intuitively reflects the overall operating range of similar equipment and automatically identifies the weak point closest to the safety domain boundary, thereby achieving quantitative assessment of the stability margin of the equipment group and precise location of weak points, providing a clear target for priority control.

[0037] 8. This invention constructs early warning and discrimination logic based on the general stability mechanism of power systems, and is uniformly applicable to synchronous generator sets, grid-connected converters, and grid-linked converters. It has a solid theoretical foundation and strong engineering applicability.

[0038] 9. This invention adopts edge computing at the site, constructs a visual map from point to surface, and only sends up key feature quantities, thereby solving the bottleneck of wide-area massive raw data transmission from the source, while ensuring the real-time performance and consistency of panoramic monitoring.

[0039] 10. This invention constructs a visual co-management platform and a control mode that combines local autonomy with remote collaboration, realizing a closed loop of remote information transmission and hierarchical control, taking into account both rapid local response and remote overall decision-making, and comprehensively improving the level of safe and stable operation of the power grid.

[0040] 11. This invention innovatively achieves joint analysis of effective and instantaneous values, and uniformly depicts steady-state stability margin, dynamic waveform distortion, full-band oscillation characteristics and limit boundaries in the same curve view, solving the problems of separation between steady-state and dynamic states and separation between boundaries and waveforms in traditional monitoring, and significantly improving the ability to identify broadband oscillations early and provide comprehensive early warning of stability risks.

[0041] 12. This invention supports the independent selection or combined display of global panoramic maps and the operating curves of single / multiple devices, single / multiple feeders, and limit boundaries and warning boundaries, clearly distinguishing the device's own capability boundaries from the power grid safety constraint boundaries, with clear hierarchy and intuitive identification; it also has online query, offline query and spectrum analysis functions, which can realize broadband oscillation feature extraction and oscillation source location, and expand the status traceability and fault analysis capabilities. Attached Figure Description

[0042] Figure 1 This is an overall flowchart of the present invention;

[0043] Figure 2 This is a diagram illustrating the security domain configuration for a single power generation device according to the present invention.

[0044] Figure 3 This is a schematic diagram of the security domain of the grid-type power generation equipment of the present invention;

[0045] Figure 4 This is a schematic diagram of the safety domain of the grid-connected power generation equipment of the present invention;

[0046] Figure 5 This is a schematic diagram illustrating the PQ operating group envelope and weak point identification of similar power generation equipment according to the present invention;

[0047] Figure 6 This invention provides a global panoramic dynamic visualization monitoring map of the site. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and preferred embodiments. However, it should be noted that many details listed in the specification are merely to provide the reader with a thorough understanding of one or more aspects of the invention, and these aspects of the invention can be achieved even without these specific details.

[0049] From the perspective of power system stability mechanism, all types of power generation equipment follow a general stability law: when the operating point approaches the safety domain boundary in the active, reactive or voltage dimension, the equipment regulation capability tends to saturate, the available stability margin narrows, the system damping level decreases, and it is easy to induce power coupling oscillation, voltage-reactive oscillation or even instability and grid disconnection.

[0050] Specifically, this manifests as follows:

[0051] Active power boundary approaching: The operating point approaches the active power limit boundary, the power regulation capability is saturated, the stable operating range shrinks, and it is easy to cause power coupling oscillations in the system;

[0052] Approaching the reactive power boundary: When the operating point approaches the reactive power limit boundary, the reactive power regulation capability becomes saturated, the voltage support capability decreases, and voltage-reactive power coupling oscillation is easily induced.

[0053] Voltage boundary approaching: When the operating point approaches the voltage limit boundary, the voltage regulation margin is exhausted, and the risk of voltage instability and equipment disconnection increases sharply.

[0054] The above mechanism applies uniformly to all types of power generation equipment, including synchronous generator sets, grid-connected converters, and grid-connected converters.

[0055] Please see Figure 1-6 This invention provides a method for multidimensional stability monitoring of power grids based on dynamic graphs, comprising the following steps:

[0056] S1, Multi-node electrical quantity acquisition

[0057] Specifically, high-frequency electrical quantity operation data of power plants, renewable energy stations, grid connection points, feeders, and various generating equipment within the regional power grid are collected synchronously to provide raw basic data sources. In this embodiment, operation data of a renewable energy station's 35kV feeder, generating equipment, and 220kV grid connection point are collected.

[0058] S2, Standardized per-unit benchmark normalization

[0059] Specifically, a unified per-unit value benchmark system is established for grid connection points, feeders, and generating equipment, standardizing all electrical quantities. This unified per-unit value benchmark system is designed for the overall construction of power plants or regional power grids, adapting to the unified calculation requirements of online dynamic monitoring and offline analysis visualization interfaces. It includes voltage benchmark values, power benchmark values, and current benchmark values. The voltage benchmark values ​​correspond to the rated voltage of the grid connection point, the rated voltage of the feeder, and the rated voltage at the generator terminals of the generating equipment, respectively. The power benchmark values ​​correspond to the rated capacity of a single generating unit, the total rated installed capacity of the grid connection point power plant, and the total rated capacity of the generating equipment connected to the feeder, respectively. The current benchmark value is calculated from the corresponding power benchmark value and voltage benchmark value.

[0060] In this embodiment, the feeder voltage reference is 35kV, the power reference is the total capacity of the connected power generation equipment, the power generation equipment reference is the rated capacity of a single unit and the rated voltage at the generator terminal, and the grid connection point reference is the total installed capacity of the power station and the rated voltage of 220kV.

[0061] S3, Voltage and PQ Multidimensional Security Domain Construction

[0062] Specifically, based on standardized multi-dimensional stability indicators, the system is vertically layered according to the grid connection point layer, feeder layer, and power generation equipment layer, and horizontally classified according to different types of power generation equipment. Voltage safety domains and PQ safety domains are constructed and drawn for each layer and equipment type. The voltage safety domain is the voltage operating range, consisting of limit boundaries and multi-level early warning boundaries. The PQ safety domain is the active-reactive power operating range, also consisting of limit boundaries and multi-level early warning boundaries, to achieve tiered early warning and form a global, dynamic, and visual monitoring map of the power station.

[0063] like Figure 2 As shown, Figure 2 A diagram showing the security domain settings for a single power generation device. Among them:

[0064] (1) The PQ point of the grid connection point, feeder or power generation equipment Warning boundaries;

[0065] (2) The PQ point of the grid connection point, feeder or power generation equipment. The limit boundary; beyond this boundary, stability will be lost.

[0066] (3) The voltage warning boundary for feeders or power generation equipment;

[0067] (4) This is the voltage limit boundary of the feeder or power generation equipment. Exceeding this boundary will result in instability.

[0068] (5) Reset the boundary for the grid connection point voltage control strategy. If the boundary is exceeded, AVC will change the control mode.

[0069] (6) is the high and low voltage switching boundary at the grid connection point. Exceeding this boundary will enter the high and low voltage override mode.

[0070] in, , , representing the vertical and horizontal axes of the power per unit value plane, respectively; , representing the horizontal phasor of the voltage per unit value at the grid connection point. For the first Voltage of generator or feeder node Relative grid connection point voltage The phase difference value, This is the phase angle characteristic amplification factor.

[0071] The multi-level early warning boundary is a gradient early warning boundary, which can be configured with single or multiple early warning levels according to the operating scenario. Each level of early warning boundary is deployed layer by layer from the inside out, forming a gradient early warning interval. Graded early warnings are achieved based on the operating point crossing different early warning boundaries, and are visually displayed in the safety domain map through color differentiation and regional rendering. In this embodiment, corresponding voltage safety domains and PQ safety domains are drawn for synchronous generator sets, grid-connected converters, and grid-linked converters, respectively, and multi-level early warning boundaries are sequentially set from the inside out within each safety domain.

[0072] like Figure 3 As shown, Figure 3 The diagram shows the adjustable margin setting for grid-connected power generation equipment. In the diagram, (1) represents the normal operating range of the grid-connected power generation equipment PQ; and (2) represents the reserved adjustable margins for active and reactive power. Figure 4 For grid-connected power generation equipment, to ensure the frequency and voltage stability of the grid-connected power generation equipment during grid disturbances, a more sufficient active and reactive power adjustment space is reserved; (3) the normal operating voltage range of the grid-connected power generation equipment is set; (4) a voltage adjustment margin is reserved: the grid-connected power generation equipment has voltage maintenance capability during grid disturbances, relatively Figure 4 For grid-connected power generation equipment, the reserved space is relatively small.

[0073] like Figure 4 As shown, Figure 4 The diagram shows the passive availability margin setting for grid-connected power generation equipment. In the diagram, (5) represents the normal operating range of the grid-connected power generation equipment PQ; (6) represents the reserved active and reactive passive availability margins to ensure that PQ has sufficient swing space during grid disturbances; (7) represents the normal operating range of the grid-connected power generation equipment voltage; and (8) represents the reserved passive availability margin of voltage to ensure that the voltage has sufficient swing space during grid disturbances, relative to... Figure 3 Grid-type power generation equipment has a large voltage margin.

[0074] S4. Running point cluster fitting and weak point identification

[0075] Specifically, group analysis and envelope fitting are performed on the operating points of the same type of power generation equipment in the PQ plane or voltage vector plane to identify the weak points closest to the safety domain limit boundary of that type of equipment. These weak points are identified as the key points with the lowest stability margin and the most likely to exceed the limit. For each set of operating points of various types of power generation equipment, a smooth closed curve is used to fit and generate the corresponding outer envelope of the operating group in the PQ plane. The envelope represents the overall operating range of the same type of equipment. Each operating point is traversed and its shortest Euclidean distance to the safety domain limit boundary is calculated. The operating point closest to the limit boundary is identified as the weak point of that type of power generation equipment and marked with a highlighted symbol in the visualization map. This is identified as the key weak point with the lowest stability margin and the most likely to trigger stability exceedance.

[0076] like Figure 5 As shown in the figure, (1) is the envelope of the PQ operating group of similar power generation equipment; (2) is the operating point closest to the PQ limit boundary marked as a weak point, with flashing color warning; (3) is the envelope of the voltage operating group of similar power generation equipment; (4) is the operating point closest to the voltage limit boundary marked as a weak point, with flashing color warning.

[0077] like Figure 6 As shown, hierarchical security domains are formed at the grid connection point layer, feeder layer, and power generation equipment layer, respectively. These are then overlaid to construct a unified security domain map for the entire site, generating a global panoramic dynamic visualization monitoring map of the station.

[0078] S5. Construct a multi-level monitoring system for regions, substations, grid connection points, feeders, and power generation equipment. Based on the relative positions of operating points and multi-level early warning boundaries and extreme boundaries, as well as the results of weak point identification, implement graded alarms and safety and stability control. Remotely transmit key safety domain information and weak point information to the centralized control center or dispatch center, and establish a joint management platform for on-site and remote operations.

[0079] Specifically, it enables on-site and remote collaborative control of power stations and regional power grids, transmitting safety domain boundaries, early warning boundaries, operating point coordinates, weak point information, and over-limit alarm information to the centralized control center or dispatch center via highly reliable communication links, constructing a visual co-management platform, and forming a control mode of on-site autonomy and remote collaboration.

[0080] On the local side, edge computing is used to perform online calculation of the security domain, identification of weak points, early warning judgment, and violation judgment. Specifically, it includes the following steps:

[0081] S501, On-site Analysis of Station Edge Calculation

[0082] On the local side, edge computing is used to complete online calculation of the security domain, identification of weak points, early warning judgment, and limit violation judgment. Key feature quantities such as security domain boundaries, early warning information, weak point information, and operational status are transmitted to the central control center and dispatch center via redundant communication links. On-site edge computing at the site completes on-site analysis of the security domain, boundaries, and weak points, only transmitting key feature quantities, and then stitching them from point to surface to form a full-domain visualization map, thereby solving the problem of massive data congestion and transmission bottlenecks in wide-area measurement systems.

[0083] S502, Feature Data Compression and Global Data Stitching

[0084] Key operational features are selected and compressed into data. The feature information of individual stations is then stitched together to generate a regional full-area monitoring view.

[0085] S503, panoramic and single-device dynamic map visualization

[0086] It supports multi-view display of global panoramic maps of the site, single-device and single-feeder operating curves, and layered overlay of limit boundaries and warning boundaries. When an operating point successively crosses the warning boundaries at each level or a weak point approaches the limit boundary, the corresponding level of alarm is triggered. The abnormal location, weak point, and corresponding equipment type are located in the map, and graded control commands are executed to achieve rapid closed-loop monitoring and control. Based on the visualized map, it achieves full-domain dynamic rendering, providing timely and intuitive visualized decision-making basis for on-site accident recollection, accident handling, and safety and stability control, thus achieving rapid closed-loop monitoring and control.

[0087] In the visualization map, smooth curves are used to mark the boundaries of the operating status areas of each power generation device and feeder, intuitively displaying the adjustable margin or passive availability margin of the operating point in terms of voltage, active power, and reactive power. For power generation devices with active voltage regulation and active reactive power adjustment capabilities, both voltage and reactive power margins are actively adjustable; for power generation devices with passive following characteristics, both voltage and reactive power margins are passively available. The safety domain map can be integrated with existing monitoring systems, protection systems, or fault recording systems, providing a unified visual support for oscillation source location and stability control.

[0088] S504, Joint Analysis of RMS and Instantaneous Values ​​Across the Entire Frequency Band

[0089] By integrating steady-state RMS values ​​of electrical quantities with transient instantaneous sampling data, a unified analysis of wideband oscillation waveforms and spectral characteristics can be achieved.

[0090] S505, Boundary Comparison and Hierarchical Early Warning Judgment

[0091] Based on the positional relationship between the operating point and the multi-level early warning boundary and extreme boundary, the system automatically realizes the judgment of multi-level risk early warning.

[0092] S506. Combining on-site assessment results with dispatching strategies, implement on-site autonomous control and remote dispatching coordinated regulation, and feed the regulation results back to the fifth process frame to form a closed-loop iterative process of monitoring, assessment, early warning and regulation.

[0093] When the operating point successively crosses the warning boundaries of each level or approaches the limit boundary of the weak point, the corresponding level alarm is triggered, the abnormal location, weak point and the type of equipment to which it belongs are located in the map, and the hierarchical control command is executed to realize the rapid closed loop of monitoring and control.

[0094] The visualized graphs are displayed using dynamic curves and color-coded zones. Active voltage regulators use active adjustable margin to represent their adjustment range, while passively following regulators use passive available margin to represent their adaptation range. The operational group is shown with a smooth envelope curve illustrating its overall distribution, with weak points highlighted, providing an intuitive basis for on-site operation and maintenance and scheduling decisions.

[0095] The platform supports the independent selection or combination of two types of views based on monitoring needs: one is a dynamic panoramic map of the entire station, and the other is the operating curves of a single or multiple devices or a single or multiple feeders, as well as their corresponding limit boundaries and warning boundaries.

[0096] It can simultaneously characterize safety limit boundaries based on operating curves, and combine effective and instantaneous electrical quantity data to achieve unified analysis of steady-state margin assessment, dynamic transient waveforms, and full-frequency oscillation characteristics, overcoming the limitations of traditional, fragmented steady-state and dynamic monitoring. The limit boundaries refer to the operating limits determined by the rated parameters, capacity limits, regulation capabilities, and physical and electrical characteristics of the generating equipment or feeder itself, reflecting the upper and lower technical limits of the adjustable range of the equipment or feeder. It also supports online real-time querying of operating data, offline historical backtracking querying, and broadband spectrum analysis, enabling oscillation feature extraction and oscillation source identification.

[0097] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for multidimensional stability monitoring of power grids based on dynamic graphs, characterized in that, Includes the following steps: S1. Collect electrical operation data of power plants, new energy stations, grid connection points, feeders, and power generation equipment within the regional power grid; S2. Establish a unified per-unit value benchmark system for grid connection points, feeders, and power generation equipment, and standardize all electrical quantities; S3. Based on the standardized multi-dimensional stability indicators, the voltage security domain and PQ security domain are constructed and drawn according to the vertical layering of grid connection point layer, feeder layer and power generation equipment layer, and the horizontal classification of different types of power generation equipment. The voltage safety domain is the voltage operating range, which consists of limit boundaries and multi-level early warning boundaries; the PQ safety domain is the active-reactive operating range, which consists of limit boundaries and multi-level early warning boundaries. S4. Perform group analysis and envelope fitting on the operating points of the same type of power generation equipment in the voltage vector plane or PQ plane to identify the weak points of the power generation equipment that are closest to the safety domain limit boundary, and use them as the key points with the lowest stability margin and the most likely to exceed the limit. S5. Construct a multi-level monitoring system for regions, substations, grid connection points, feeders, and power generation equipment. Based on the relative positions of operating points and multi-level early warning boundaries and extreme boundaries, as well as the identification results of weak points, conduct graded alarms and safety and stability control. Remotely transmit key safety domain information to the centralized control center or dispatch center, and establish a local and remote co-management platform.

2. The method for multidimensional stability monitoring of power grids based on dynamic graphs according to claim 1, characterized in that, In step S2, the unified per-unit benchmark system includes voltage benchmark, power benchmark and current benchmark. The voltage benchmark corresponds to the grid connection point rated voltage, feeder rated voltage and generator terminal rated voltage, respectively. The power benchmark corresponds to the rated capacity of a single generator, the total rated installed capacity of the grid connection point, and the total rated capacity of the generator connected to the feeder, respectively. The current benchmark is calculated from the corresponding power benchmark and voltage benchmark.

3. The method for multi-dimensional stability monitoring of power grids based on dynamic graphs according to claim 1, characterized in that, In step S3, the multi-level early warning boundary is a gradient early warning boundary. The gradient early warning boundary is configured with single-level or multi-level early warning levels according to the operating scenario. The multi-level early warning levels are set in layers from the inside to the outside within the safety domain, forming a gradient early warning range that approaches the limit boundary from the normal operating range.

4. The power grid multidimensional stability monitoring method based on dynamic graphs according to claim 3, characterized in that, The early warning method includes the following steps: The system provides tiered early warnings based on the points of operation crossing different warning boundaries, and displays these warnings in a dynamic, panoramic monitoring map of the entire station using color differentiation and regional rendering.

5. The method for multi-dimensional stability monitoring of power grids based on dynamic graphs according to claim 1, characterized in that, In step S5, the key security domain information includes security domain boundaries, early warning boundaries, operating point coordinates, weak point information, and limit overrun alarm information.

6. The method for multi-dimensional stability monitoring of power grids based on dynamic graphs according to claim 1, characterized in that, In step S5, the method for processing critical security domain information by the local and remote co-management platform includes the following steps: On the ground side, edge computing is used to complete online calculation of the security domain, identification of weak points, early warning judgment, and violation judgment. The key features of the safety domain boundary, early warning information, weak point information, and operation point status are compressed, and the feature information of single-point stations is spliced ​​to generate a regional full-domain monitoring view. Multi-view display of the overall panoramic map of the station, the operation curves of individual equipment and individual feeders, and layered overlay of limit boundaries and early warning boundaries; By integrating steady-state RMS values ​​and transient instantaneous sampling data of electrical quantities, a comprehensive analysis of wideband oscillation waveforms and spectral characteristics is performed. Based on the positional relationship between the operating point and the multi-level early warning boundary and extreme boundary, multi-level risk early warning judgment is carried out; By combining the results of on-site assessment with the dispatching strategies, on-site control and remote dispatching are carried out, and the control results are fed back to the on-site side.

7. The method for multidimensional stability monitoring of power grids based on dynamic graphs according to claim 6, characterized in that, The method for determining multi-level risk warnings is as follows: When the operating point successively crosses the warning boundaries of each level or approaches the limit boundary of the weak point, the corresponding level alarm is triggered, the abnormal location, weak point and its equipment type are located in the map, and the graded control command is executed.

8. The power grid multidimensional stability monitoring method based on dynamic graphs according to claim 6, characterized in that, It also includes the following steps: For grid-connected and grid-linked power generation equipment, the active adjustable margin or passive available margin of voltage and power dimensions are visualized and marked in their respective security domains.

Images