A power system scheduling data construction method and system based on multi-source measurement consistency calibration and running state reliable mapping

Abstract

The application discloses a power system scheduling data construction method and system based on multi-source measurement consistency calibration and operation state reliable mapping, and belongs to the technical field of power system data processing and intelligent scheduling. The method comprises the following steps: acquiring power system multi-source operation data, wherein the multi-source operation data is acquired through a power system multi-source measurement access system, the multi-source measurement access system comprises a data acquisition interface layer, an edge convergence node and a unified data adaptation interface, and the layers interact through a standardized data interface protocol; performing time synchronization processing and power grid topology alignment processing on the multi-source operation data to construct a unified space-time data set; performing consistency checking on the unified space-time data set based on power system physical constraints; performing abnormality detection and correction or state reconstruction on data that does not satisfy the consistency constraints; mapping and converting the corrected data based on a state estimation constraint space, wherein the state estimation constraint space is constructed based on a power system state equation and an observation equation, is used for unified constraint expression and projection mapping of different measurement sources in a state space, and generates a standardized operation state vector that can be used for scheduling; and outputting the operation state vector to a scheduling optimization module or an intelligent agent module as input data. The application realizes unified fusion, consistency calibration and operation state reliable mapping of multi-source power operation data, and provides a high-quality data basis for power system scheduling optimization and intelligent decision-making.

Description

Technical Field

[0001] This invention belongs to the field of power system data processing and intelligent dispatching technology, specifically relating to a method and system for constructing power system dispatching data based on multi-source measurement consistency calibration and operational state reliable mapping. Background Technology

[0002] With the digitalization and intelligentization of power systems, power systems aggregate power operation data from different system configurations and operating environments through a multi-source measurement access system. In centralized dispatch systems, data sources include steady-state operation data at the second to minute level provided by SCADA systems and high-frequency dynamic measurement data at the millisecond level provided by PMU synchronization phasor devices; in distributed edge access mode, data is aggregated from local measurement devices through edge computing nodes; in park-based local operation access mode, data is directly collected from smart meters, energy storage PCS, EMS systems, and local gateways.

[0003] The aforementioned multi-source data exhibit significant differences in time scale, spatial granularity, and data quality, leading to the following problems: (1) The sampling periods of multi-source data are inconsistent, making it difficult to directly integrate them for scheduling decisions; (2) The power grid topology changes dynamically with the switching state, and the data collected at different times correspond to different power grid topologies; (3) There are communication delays, data loss and measurement errors. If abnormal data is directly input into the scheduling system, it will lead to scheduling decision deviations. (4) The lack of a data consistency verification mechanism based on the physical constraints of the power system leads to the propagation of state estimation errors to the scheduling stage.

[0004] Therefore, there is an urgent need for a data construction method that is independent of system form and can achieve unified alignment of multi-source data, physical consistency verification, and reliable mapping in runtime. Summary of the Invention

[0005] (a) Purpose of the invention The purpose of this invention is to provide a method and system for constructing power system dispatch data based on multi-source measurement consistency calibration and operational state reliable mapping, so as to solve the problems of low quality of multi-source data fusion, abnormal data interference with dispatch decisions, and lack of physical consistency verification mechanism in the prior art.

[0006] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution:

[0007] S1: Acquire multi-source operation data of the power system. The multi-source operation data is acquired through the multi-source measurement access system of the power system. The multi-source measurement access system includes a data acquisition interface layer, an edge aggregation node, and a unified data adaptation interface. Each layer interacts with the others through a standardized data interface protocol. Each component collaborates to complete data aggregation according to a predefined process. This system is used to aggregate power operation data from different system environments, including centralized dispatch system access mode, distributed edge access mode, and local operation access mode in the park.

[0008] In the centralized dispatch system access mode, the data sources include steady-state operation data of the SCADA system (node ​​voltage amplitude, line active and reactive power, generator output, load power, switch status, etc.) and high-frequency dynamic measurement data of the PMU synchronous phasor device (node ​​voltage phasor and amplitude, line current phasor and amplitude).

[0009] In the distributed edge access mode, data is aggregated from local measurement devices through edge computing nodes deployed in substations or power distribution rooms.

[0010] In the local operation access mode within the park, data is collected directly from smart meters, energy storage PCS, EMS system, and local gateway, and uniformly encapsulated through the local EMS edge controller. Data from each device is uploaded after unified timestamp calibration and format conversion through the local gateway.

[0011] S2: Time synchronization and grid topology alignment are performed on multi-source operational data to construct a unified spatiotemporal data set. Time synchronization is based on a unified timestamp alignment mechanism, resampling and fusing data from different sampling periods. Topology alignment constructs a topology mapping relationship based on grid switch states and line connection relationships, determining the corresponding grid topology structure according to the switch states at each time point, and mapping the measurement data to the correct topology nodes.

[0012] S3: Perform consistency verification on a unified spatiotemporal dataset based on power system physical constraints, including power balance constraints, node voltage constraints, and topological connectivity constraints. The consistency verification is based on the residuals of the power flow equations. Data is considered inconsistent when node power injection and line power flow do not meet the power physical constraints.

[0013] S4: Perform anomaly detection on data that does not meet consistency constraints, and execute data correction or state reconstruction. Anomaly detection includes one or more of the following: communication delay anomaly detection, data missing detection, and measurement drift detection. Data correction or state reconstruction employs one or more of the following methods: Kalman filtering, interpolation, or estimation based on historical operating trajectories.

[0014] S5: The corrected data is mapped and transformed based on the state estimation constraint space to generate a standardized operating state vector usable for scheduling. The state estimation constraint space is constructed based on the power system state equations and observation equations, and is used to unify the constraint expressions and projection mappings of different measurement sources in the state space. It performs unified constraint expressions and consistent projection mappings on measurement data from different sources, converting multi-source measurement data into a standardized operating state space representation that satisfies the power system state estimation constraints. The operating state vector includes node voltage vectors, line power flow vectors, generator output vectors, and load power vectors.

[0015] S6: Output the operating state vector to the power system dispatch optimization module or the intelligent agent module as input data.

[0016] (III) Beneficial Effects Compared with the prior art, the present invention has the following beneficial effects: (1) Unified data construction method independent of system form: The differences in the underlying system form are shielded by the multi-source measurement access system. The multi-source measurement access system includes a data acquisition interface layer, an edge aggregation node and a unified data adaptation interface. Each layer interacts with each other through a standardized data interface protocol. Each component completes data aggregation in collaboration according to a predefined process. It supports three deployment modes: centralized scheduling system, distributed edge computing and local operation in the park, which improves the applicability and scalability of the method. (2) Introduce physical consistency constraint verification of power system, and determine inconsistent data based on power flow equation residuals to improve data credibility; (3) Anomalies are located by detecting communication delay anomalies, missing data, and measurement drift, thereby reducing the impact of abnormal data on scheduling decisions; (4) Construct a reliable mapping mechanism for the operating state based on the state estimation constraint space. The state estimation constraint space is constructed based on the power system state equation and observation equation. It is used to unify the constraint expression and projection mapping of different measurement sources in the state space, perform unified constraint expression and consistent projection mapping on measurement data from different sources, and map the corrected multi-source data into a standardized operating state vector to provide high-quality input for scheduling optimization and intelligent decision-making. (5) Improve the reliability and stability of scheduling input data to ensure the quality of scheduling decisions from the data source. Attached Figure Description

[0017] Figure 1 Flowchart of multi-source data fusion and runtime trusted mapping.

[0018] Figure 2 : Power system consistency verification structure diagram.

[0019] Figure 3: Schematic diagram of the generation and output of the running state vector. Detailed Implementation The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] Example 1: Centralized Dispatch System Access Mode Taking a regional power grid as an example, the power grid includes 3 generator sets, 1 energy storage system, 12 load nodes and 8 transmission lines, and is equipped with a SCADA system (sampling period of 3 seconds) and a PMU device (sampling period of 20 milliseconds).

[0021] S1: The data acquisition module obtains multi-source operational data through a centralized scheduling system access mode. The data acquisition interface layer acquires node voltage amplitude, line power, generator output, load power, and switch status data from the SCADA system, and node voltage phasor and line current phasor data from the PMU device; the edge aggregation node performs preliminary time alignment and anomaly marking on the two types of data; the unified data adaptation interface standardizes and encapsulates the aggregated data. All components collaborate to complete data aggregation according to a predefined process.

[0022] S2: The time synchronization module uses the SCADA data timestamp as a reference and downsamples the PMU data. The topology alignment module constructs a topology mapping relationship based on the switch status data at each time point.

[0023] S3: The consistency verification module verifies the unified spatiotemporal data set. For node 3, its generator injection power is 250MW, and the line outflow power is 120MW and 130MW, with a total outflow power of 250MW, so the power balance verification passes. For node 6, the voltage amplitude is 1.05pu, exceeding the upper limit of the allowable range of 0.9-1.1pu, so the node voltage constraint verification fails.

[0024] S4: The voltage over-limit problem at node 6 was analyzed to be caused by measurement drift. Kalman filtering was used for state estimation, and the corrected voltage value of 1.02 pu was used to replace the original abnormal value.

[0025] S5: The operating state mapping module maps and transforms the corrected data based on the state estimation constraint space. The state estimation constraint space is constructed based on the power system state equation and observation equation. It performs unified constraint expression and consistent projection mapping on measurement data from different sources to generate a standardized operating state vector.

[0026] S6: The data output module outputs the running status vector to the economic scheduling module and the safety constraint scheduling module.

[0027] Example 2: Local Operation Access Mode in the Park Taking a zero-carbon industrial park as an example, the park has 5MW of photovoltaic power, 2MW of wind power, 2MW / 4MWh of energy storage system and multiple industrial and commercial users, but no data platform or centralized dispatch system has been deployed.

[0028] S1: The data acquisition module obtains multi-source operational data through the park's local operation access mode. It directly collects electricity consumption data from smart meters, SOC and charge / discharge power data from the energy storage PCS, photovoltaic and wind power output data from the EMS system, and equipment operating status data from the local gateway. Data from each device is uploaded after unified timestamp calibration and format conversion through the local gateway. The data is uniformly encapsulated through the local EMS edge controller and provided to the upper layer via a unified data adaptation interface, ensuring a standardized data interface. All components collaborate to complete data aggregation according to a predefined process.

[0029] S2: The time synchronization module uses the smart meter data timestamp as a reference to align the energy storage PCS data and EMS data in time. The topology alignment module constructs the topology mapping relationship of the park's distribution network based on the switch status information provided by the local gateway.

[0030] S3: The consistency verification module verifies the park's data based on power balance constraints. Taking a specific moment as an example, the photovoltaic output is 300kW, wind power output is 150kW, energy storage discharge power is 80kW, and the grid power purchase is 200kW, for a total power supply of 730kW; the park's total load is 720kW, and the power balance verification passes. However, the energy storage SOC data shows a jump, suddenly dropping from 45% to 20%, while the charging and discharging power does not change accordingly, indicating a data anomaly.

[0031] S4: Regarding the abnormal SOC data issue, analysis revealed that communication latency caused the SOC data to arrive late. An interpolation method based on historical operating trajectories was used to correct the SOC data, which was then corrected to 42%.

[0032] S5: The operational state mapping module maps and transforms the corrected park data based on the state estimation constraint space, performs unified constraint expression and consistent projection mapping on measurement data from different sources, and generates a standardized operational state vector.

[0033] S6: The data output module outputs the running status vector to the local scheduling and optimization module in the park.

[0034] The above embodiments are only used to illustrate the present invention and do not constitute a limitation on the scope of protection of the present invention.

Claims

1. A method for constructing power system dispatch data based on multi-source measurement consistency calibration and operational state reliable mapping, characterized in that, Includes the following steps: S1: Acquire multi-source operation data of the power system. The multi-source operation data is acquired through the multi-source measurement access system of the power system. The multi-source measurement access system includes a data acquisition interface layer, an edge aggregation node, and a unified data adaptation interface. Each layer interacts with the other through a standardized data interface protocol. Each component works together to complete data aggregation according to a predefined process. This system is used to aggregate power operation data from different system environments, including centralized dispatch system access mode, distributed edge access mode, and local operation access mode in the park. S2: Perform time synchronization processing and power grid topology alignment processing on the multi-source operation data to construct a unified spatiotemporal data set; S3: Perform consistency verification on the unified spatiotemporal dataset based on power system physical constraints, including power balance constraints, node voltage constraints, and topology connectivity constraints. S4: Perform anomaly detection on data that does not meet consistency constraints, and perform data correction or state reconstruction. S5: The corrected data is mapped and transformed based on the state estimation constraint space to generate a standardized operating state vector that can be used for scheduling; the state estimation constraint space is constructed based on the power system state equation and observation equation, and is used to unify the constraint expression and projection mapping of different measurement sources in the state space. S6: Output the operating state vector to the power system dispatch optimization module or the intelligent agent module as input data.

2. The method according to claim 1, characterized in that, The time synchronization process is based on a unified timestamp alignment mechanism, which resamples and fuses data from different sampling periods.

3. The method according to claim 1, characterized in that, The topology alignment process is based on the relationship between grid switch states and line connections to construct a topology mapping relationship.

4. The method according to claim 1, characterized in that, The consistency check is based on the residual of the power flow equation. When the node power injection and the line power flow do not meet the electrical physical constraints, the data is determined to be inconsistent.

5. The method according to claim 1, characterized in that, The anomaly detection in step S4 includes one or more of the following: communication delay anomaly detection, data missing detection, and measurement drift detection.

6. The method according to claim 1, characterized in that, The data correction or state reconstruction in step S4 employs one or more of the following methods: Kalman filtering, interpolation, or estimation based on historical running trajectories.

7. The method according to claim 1, characterized in that, The operating state vector includes node voltage vector, line power flow vector, generator output vector, and load power vector.

8. The method according to claim 1, characterized in that, The mapping transformation based on the state estimation constraint space is used to convert multi-source measurement data into a standardized operating state space representation that satisfies the power system state estimation constraints.

9. A power system dispatch data construction system based on multi-source measurement consistency calibration and operational state reliable mapping, characterized in that, include: The data acquisition module is used to acquire multi-source operation data through the multi-source measurement access system of the power system; The time synchronization module is used to perform time synchronization processing on multi-source running data; The topology alignment module is used to perform power grid topology alignment processing on multi-source operating data; The consistency verification module is used to perform consistency verification on a unified spatiotemporal data set based on the physical constraints of the power system. The anomaly detection and reconstruction module is used to detect anomalies in data that does not meet consistency constraints and perform data correction or state reconstruction. The runtime mapping module is used to map and transform the corrected data based on the state estimation constraint space to generate a standardized runtime state vector that can be used for scheduling. The data output module is used to output the operating state vector to the power system dispatch optimization module or the intelligent agent module.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 8.

Images