Control center equipment state early warning system and method based on digital twinning

Abstract

The application relates to the technical field of centralized control centers, and discloses a centralized control center equipment state early warning system and method based on digital twinning, which comprises a construction module, a collection model, an evaluation module and an early warning module.The construction module is used for constructing a digital twinning model of each centralized control center equipment; the collection model is used for setting a plurality of key state indexes, determining key monitoring data of each equipment, generating a data monitoring strategy of the corresponding equipment, collecting data according to the data monitoring strategy, and generating a real-time key data set; the evaluation module is used for mapping the real-time correlation data set to the digital twinning model of the corresponding equipment, obtaining simulation simulation results, performing equipment state evaluation, and obtaining a state value of the corresponding equipment; and the early warning module is used for judging whether to generate an early warning instruction of the corresponding equipment according to the state value, integrating all equipment information of the early warning instruction to generate a comprehensive early warning report, realizing accurate and efficient early warning of the state of the centralized control center equipment, and improving the intelligent level of equipment operation and maintenance and the timeliness of fault processing.

Description

Technical Field

[0001] This application relates to the field of centralized control center technology, and in particular to a centralized control center equipment status early warning system and method based on digital twins. Background Technology

[0002] With the rapid development of industrial automation and intelligence, the centralized control center, as the core hub in industrial production, energy dispatch, urban management and other fields, has its equipment's stable operation directly related to the safety and efficiency of the entire system.

[0003] In existing technologies, traditional equipment status monitoring and early warning methods in centralized control centers largely rely on manual inspections, single-sensor data acquisition, and simple threshold judgments, which have many limitations. Manual inspections not only consume significant manpower and resources but also struggle to achieve real-time, continuous monitoring of equipment status, making it easy to overlook potential faults due to human error. Single-sensor data often only reflects certain operational parameters of the equipment, making it difficult to comprehensively and accurately assess the overall health status. Simple threshold judgment methods often only issue warnings after a fault has occurred or caused some impact, delaying the crucial time for fault handling. Therefore, the aforementioned early warning methods are no longer sufficient to meet the needs of modern centralized control centers for refined management and intelligent operation and maintenance in terms of data monitoring capabilities, fault prediction accuracy, and early warning response speed. Summary of the Invention

[0004] To address the aforementioned technical challenges, this application provides a digital twin-based early warning system and method for the status of equipment in a centralized control center. By constructing a digital twin model, the system determines the key monitoring data and corresponding data monitoring strategies for each device, generates a real-time key dataset, and inputs it into the digital twin model. Simulation results are obtained, along with the status values ​​of the corresponding devices. Based on these status values, the system determines whether to issue an early warning command and integrates the data to generate a comprehensive early warning report. This enables accurate and efficient early warning of the status of equipment in the centralized control center, improving the intelligence level of equipment operation and maintenance and the timeliness of fault handling.

[0005] In some embodiments of this application, a digital twin-based centralized control center equipment status early warning system is provided, including: Modules are used to build digital twin models of each control center device; The data acquisition model is used to set several key status indicators, determine the key monitoring data of each device based on all key status indicators, generate the corresponding data monitoring strategy for the device, collect data according to the data monitoring strategy, and generate real-time key datasets for each device. The evaluation module is used to map the real-time correlated dataset to the digital twin model of the corresponding device, obtain the simulation results and evaluate the device status to obtain the status value of the corresponding device. The early warning module is used to determine whether to generate an early warning command for the corresponding device based on the status value, and to integrate all device information that issued the early warning command to generate a comprehensive early warning report.

[0006] In some embodiments of this application, constructing a digital twin model of each central control center device includes: Obtain the 3D design drawings, equipment parameter manuals and historical operating data of each of the central control center devices, and construct the geometric model of the central control center devices based on 3D modeling technology; Collect the physical property parameters of each key component of the equipment, and construct the physical model of the equipment by combining it with multiphysics simulation algorithms; By connecting to the device's real-time monitoring system and historical database through data interfaces, a behavioral model of the device can be established; By integrating geometric models, physical models, and behavioral models, a digital twin model of the corresponding device is formed.

[0007] In some embodiments of this application, key monitoring data for each device are determined based on all key status indicators, and a corresponding data monitoring strategy for the device is generated, including: Establish a historical monitoring data source sequence for each device, wherein the historical monitoring data source sequence includes several historical monitoring data sources; Historical data packets for each key status indicator are generated based on historical monitoring logs. The historical data packets include several historical status assessment values ​​for the corresponding key status indicator, as well as historical values ​​for each historical monitoring data source for each device of each historical node corresponding to each historical status assessment value. Generate historical impact values ​​of each historical monitoring data source on each key status indicator based on historical data packets; The historical comprehensive impact value of the corresponding historical monitoring data source is generated based on the historical impact values ​​of the same historical monitoring data source on all key status indicators and the weight coefficients of the corresponding key status indicators. Based on historical comprehensive impact values, key monitoring data for each device are determined, and monitoring sub-strategies for each key monitoring data are generated. Generate a data monitoring strategy for the corresponding device based on all key monitoring data of the same device and the corresponding monitoring sub-strategies.

[0008] In some embodiments of this application, key monitoring data for each device are determined based on historical comprehensive impact values, and a monitoring sub-strategy is generated for each key monitoring data, including: The impact level of the corresponding historical monitoring data source is set according to the historical comprehensive impact value; The impact level threshold is preset, and the impact level threshold is mapped to a corresponding preset monitoring strategy; Set the historical monitoring data corresponding to the historical monitoring data source whose impact level is greater than the impact level threshold as the key monitoring data of the corresponding equipment; Calculate the difference between the impact level of key monitoring data and the impact level threshold. Based on the level difference-correction coefficient mapping table, match the correction coefficient of the current level difference to obtain the strategy correction coefficient of key monitoring data. Based on the preset monitoring strategy and strategy correction coefficient, a monitoring sub-strategy is generated to correspond to the key monitoring data. The monitoring sub-strategy for each key monitoring data point is generated sequentially.

[0009] In some embodiments of this application, device status assessment is performed to obtain the status value of the corresponding device, including: Based on the data monitoring strategy, several real-time key monitoring data points for each device at the current time point are generated, and a real-time key dataset is generated. The real-time key dataset is input into the digital twin model of the corresponding equipment, and the multi-physics coupling simulation algorithm inside the model is used to perform real-time simulation calculations to obtain the simulation operation data of each key monitoring data of the equipment in the future period. The simulation data is compared with the preset normal data range, and the simulation anomaly coefficient of each key monitoring data in the future period is generated based on the comparison results. Generate the actual anomaly coefficient for each key monitoring data point within the current time interval of interest; Calculate the difference between the actual anomaly coefficient and the simulated anomaly coefficient of the same key monitoring data, and set the compensation coefficient of the simulated anomaly coefficient of the corresponding key monitoring data based on the difference in anomaly coefficient. The comprehensive simulation anomaly coefficient of the corresponding equipment is obtained by weighted summation of the simulation anomaly coefficient, compensation coefficient and corresponding weight coefficient of each key monitoring data. The state value of each device is generated based on the comprehensive simulation anomaly coefficient of each device.

[0010] In some embodiments of this application, generating the actual anomaly coefficient of each key monitoring data point within the current time interval of interest includes: Generate a time reference line for the time interval of interest, and map the actual values ​​of key monitoring data within the time interval of interest monitored according to the data monitoring strategy to the time reference line to obtain the data change curve of each key monitoring data in the time interval of interest. Based on pre-defined data feature indicators, feature extraction is performed on the data change curve to generate several feature values; The data characteristic indicators include data fluctuation amplitude, data change rate, data mean deviation, and frequency of outlier occurrence; The actual anomaly coefficient of the corresponding key monitoring data within the current time interval is generated based on several characteristic values ​​and the weighting coefficients of data characteristic indicators.

[0011] In some embodiments of this application, determining whether to generate a warning command for the corresponding device based on the status value includes: Pre-set the warning status value for each device; When the status value is greater than the warning status value, no warning command is generated for the corresponding device; When the status value is not greater than the warning status value, a warning instruction is generated for the corresponding device. The warning indicators include the warning level of the corresponding device, the warning trigger time, and key abnormal data items.

[0012] In some embodiments of this application, a comprehensive early warning report is generated by integrating all device information that issued the early warning command, including: Obtain device information of the device that issued the warning command, and generate a list of warning devices; Analyze the warning instructions for each device in the warning equipment list to obtain the priority score for each device in the list; All devices in the list are prioritized according to their priority scores to obtain a warning priority sequence. The key abnormal data items of the devices are then input into the preset operation and maintenance model to obtain the corresponding operation and maintenance strategy. A comprehensive early warning report is generated based on the early warning priority sequence and all operation and maintenance strategies.

[0013] In some embodiments of this application, a priority score is obtained for each device in the list, including: Extract the warning level, equipment importance coefficient, and warning trigger time from each warning instruction, and assign a basic priority score corresponding to the warning level; Multiply the basic priority score by the equipment importance coefficient to obtain the weighted priority score; Calculate the time difference between the current time and the warning trigger time, determine the time correction coefficient based on the time difference, and multiply the weighted priority score by the time correction coefficient to obtain the priority score of the device.

[0014] In some embodiments of this application, a method for early warning of the status of central control center equipment based on digital twins is also included: Construct a digital twin model of each control center device; Several key status indicators are set, and key monitoring data for each device are determined based on all key status indicators. A corresponding data monitoring strategy for the device is generated, and data is collected according to the data monitoring strategy to generate a real-time key dataset for each device. The real-time correlated dataset is mapped to the digital twin model of the corresponding device to obtain simulation results and device status assessment, thereby obtaining the status value of the corresponding device. Based on the status value, determine whether to generate a warning command for the corresponding device, and integrate all device information that issued the warning command to generate a comprehensive warning report.

[0015] The digital twin-based centralized control center equipment status early warning system and method of this application, compared with the prior art, has the following advantages: By constructing a digital twin model, the key monitoring data and corresponding data monitoring strategies for each device are determined. A real-time key dataset is generated and input into the digital twin model to obtain simulation results and the status values ​​of the corresponding devices. Based on the status values, it is determined whether to issue an early warning command and integrate them to obtain a comprehensive early warning report. This enables accurate and efficient early warning of the status of devices in the central control center, improving the intelligence level of equipment operation and maintenance and the timeliness of fault handling. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of a digital twin-based centralized control center equipment status early warning system in an embodiment of this application; Figure 2 This is a flowchart illustrating the device status early warning method for a centralized control center based on digital twins in this application embodiment. Detailed Implementation

[0017] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0018] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] like Figure 1 As shown in the figure, the digital twin-based centralized control center equipment status early warning system of this application includes: Modules are used to build digital twin models of each control center device; The data acquisition model is used to set several key status indicators, determine the key monitoring data of each device based on all key status indicators, generate the corresponding data monitoring strategy for the device, collect data according to the data monitoring strategy, and generate real-time key datasets for each device. The evaluation module is used to map the real-time correlated dataset to the digital twin model of the corresponding device, obtain the simulation results and evaluate the device status to obtain the status value of the corresponding device. The early warning module is used to determine whether to generate an early warning command for the corresponding device based on the status value, and to integrate all device information that issued the early warning command to generate a comprehensive early warning report.

[0022] In this embodiment, key status indicators include vibration indicators, thermal imaging indicators, electrical indicators, main grid operation indicators, power quality indicators, environmental indicators, security indicators, communication indicators, and other status indicators of the central control center. These indicators can comprehensively reflect the real-time operating status and potential risks of all equipment in the central control center from different dimensions.

[0023] In some embodiments of this application, constructing a digital twin model of each central control center device includes: Obtain the 3D design drawings, equipment parameter manuals and historical operating data of each of the central control center devices, and construct the geometric model of the central control center devices based on 3D modeling technology; Collect the physical property parameters of each key component of the equipment, and construct the physical model of the equipment by combining it with multiphysics simulation algorithms; By connecting to the device's real-time monitoring system and historical database through data interfaces, a behavioral model of the device can be established; By integrating geometric models, physical models, and behavioral models, a digital twin model of the corresponding device is formed.

[0024] In this embodiment, the geometric model includes the external dimensions of the equipment, the assembly relationship of the components, and the spatial layout, which can intuitively display the physical form of the equipment; the physical model is based on the physical property parameters such as density, elastic modulus, and thermal conductivity of each key component of the equipment, combined with multi-physics simulation algorithms such as fluid mechanics, thermodynamics, and structural mechanics, to simulate the physical behavior of the equipment under different operating conditions, such as temperature field distribution, stress and strain, and vibration characteristics; the behavioral model constructs a mapping relationship between the equipment's operating state and time series through real-time equipment operating data (such as voltage, current, speed, temperature, etc.) and historical fault records, so as to realize the dynamic simulation of the equipment's start-up and shutdown, load changes, fault evolution and other behavioral processes.

[0025] In this embodiment, a digital twin model is obtained by deeply integrating these three models. The digital twin model can not only reproduce the physical entity of the device, but also predict its operating status and behavior patterns, providing a high-precision simulation basis for subsequent status assessment and early warning.

[0026] In some embodiments of this application, key monitoring data for each device are determined based on all key status indicators, and a corresponding data monitoring strategy for the device is generated, including: Establish a historical monitoring data source sequence for each device, wherein the historical monitoring data source sequence includes several historical monitoring data sources; Historical data packets for each key status indicator are generated based on historical monitoring logs. The historical data packets include several historical status assessment values ​​for the corresponding key status indicator, as well as historical values ​​for each historical monitoring data source for each device of each historical node corresponding to each historical status assessment value. Generate historical impact values ​​of each historical monitoring data source on each key status indicator based on historical data packets; The historical comprehensive impact value of the corresponding historical monitoring data source is generated based on the historical impact values ​​of the same historical monitoring data source on all key status indicators and the weight coefficients of the corresponding key status indicators. Based on historical comprehensive impact values, key monitoring data for each device are determined, and monitoring sub-strategies for each key monitoring data are generated. Generate a data monitoring strategy for the corresponding device based on all key monitoring data of the same device and the corresponding monitoring sub-strategies.

[0027] In this embodiment, the historical status assessment value of each key status indicator refers to the comprehensive status assessment result of the corresponding indicator generated by the central control center based on the historical operating data of all equipment at the corresponding historical node. This result is obtained by weighting the historical operating parameters of all equipment through a preset assessment algorithm, which can quantitatively reflect the overall health level of the key status indicator at a historical moment.

[0028] In this embodiment, for each key status indicator, historical values ​​of each historical monitoring data source under different historical status assessment values ​​of the indicator are extracted from the historical data package. The Pearson correlation coefficient method is used to calculate the linear correlation between each historical monitoring data source and the key status indicator, and the absolute value of the correlation coefficient is used as the historical impact value of the historical monitoring data source on the key status indicator.

[0029] In this embodiment, the historical impact values ​​are weighted and summed to obtain the historical comprehensive impact value. The weight coefficients are set according to the importance level of the key status indicators, and the importance level is set in advance.

[0030] In some embodiments of this application, key monitoring data for each device are determined based on historical comprehensive impact values, and a monitoring sub-strategy is generated for each key monitoring data, including: The impact level of the corresponding historical monitoring data source is set according to the historical comprehensive impact value; The impact level threshold is preset, and the impact level threshold is mapped to a corresponding preset monitoring strategy; Set the historical monitoring data corresponding to the historical monitoring data source whose impact level is greater than the impact level threshold as the key monitoring data of the corresponding equipment; Calculate the difference between the impact level of key monitoring data and the impact level threshold. Based on the level difference-correction coefficient mapping table, match the correction coefficient of the current level difference to obtain the strategy correction coefficient of key monitoring data. Based on the preset monitoring strategy and strategy correction coefficient, a monitoring sub-strategy is generated to correspond to the key monitoring data. The monitoring sub-strategy for each key monitoring data point is generated sequentially.

[0031] In this embodiment, a continuous range of comprehensive impact values ​​is preset, and each range corresponds to a preset impact level. The larger the value, the higher the impact level. The mapping relationship between comprehensive impact values ​​and preset impact levels is pre-configured based on expert experience.

[0032] In this embodiment, the impact level threshold can be adjusted according to the importance of the central control center equipment and the monitoring accuracy requirements. For example, for core control equipment, the impact level threshold can be set to level 3, and for auxiliary equipment, the impact level threshold can be set to level 2.

[0033] In this embodiment, the preset monitoring strategy includes preset monitoring time intervals for corresponding monitoring data under different operating conditions. The operating conditions include normal operation, operation under different loads, fluctuating operation, and start-stop phase. The preset monitoring time intervals under different operating conditions are set based on historical monitoring data and data acquisition accuracy.

[0034] In this embodiment, the level difference-correction coefficient mapping table includes several preset level differences, and each preset level difference is mapped to a corresponding preset strategy correction coefficient. Each strategy correction coefficient includes preset correction coefficients for several preset level differences under certain operating conditions.

[0035] In this embodiment, a preset monitoring strategy is first determined based on the impact level of key monitoring data. This preset monitoring strategy includes initial monitoring time intervals under different operating conditions, such as normal operation, operation under different loads, fluctuating operation, and start-up / shutdown phases. Then, a strategy correction coefficient matching the current level difference is obtained through a level difference-correction coefficient mapping table. This correction coefficient sets corresponding correction ratios for different operating conditions. Finally, the initial monitoring time intervals under each operating condition in the preset monitoring strategy are multiplied by the corresponding strategy correction coefficient to obtain the adjusted monitoring time intervals, thereby forming the final monitoring sub-strategy. This achieves precise dynamic control of the key monitoring data acquisition frequency.

[0036] In some embodiments of this application, device status assessment is performed to obtain the status value of the corresponding device, including: Based on the data monitoring strategy, several real-time key monitoring data points for each device at the current time point are generated, and a real-time key dataset is generated. The real-time key dataset is input into the digital twin model of the corresponding equipment, and the multi-physics coupling simulation algorithm inside the model is used to perform real-time simulation calculations to obtain the simulation operation data of each key monitoring data of the equipment in the future period. The simulation data is compared with the preset normal data range, and the simulation anomaly coefficient of each key monitoring data in the future period is generated based on the comparison results. Generate the actual anomaly coefficient for each key monitoring data point within the current time interval of interest; Calculate the difference between the actual anomaly coefficient and the simulated anomaly coefficient of the same key monitoring data, and set the compensation coefficient of the simulated anomaly coefficient of the corresponding key monitoring data based on the difference in anomaly coefficient. The comprehensive simulation anomaly coefficient of the corresponding equipment is obtained by weighted summation of the simulation anomaly coefficient, compensation coefficient and corresponding weight coefficient of each key monitoring data. The state value of each device is generated based on the comprehensive simulation anomaly coefficient of each device.

[0037] In this embodiment, the real-time key monitoring data at the current time node refers to the most recent sampled data collected according to the data acquisition strategy. This data is filtered, denoised, and standardized before being set as the real-time key monitoring data at the current time node.

[0038] In this embodiment, the preset normal data range is determined based on the equipment's factory standards, historical best operating data, and industry specifications. If the simulated operating data is within the preset normal data range, the simulation anomaly coefficient of the corresponding key monitoring data in the future period is 0. If the simulated operating data exceeds the preset normal data range, the simulation anomaly coefficient is calculated based on the extent of the exceedance. The greater the exceedance, the higher the simulation anomaly coefficient. The specific calculation formula is: Simulation anomaly coefficient = (Simulation operating data - Upper limit of normal data range) / (Historical maximum deviation value). When the simulated operating data is lower than the lower limit of the normal data range, the calculation formula is: Simulation anomaly coefficient = (Lower limit of normal data range - Simulation operating data) / (Historical maximum deviation value), where the historical maximum deviation value is the maximum absolute value of the key monitoring data exceeding the normal data range during historical operation.

[0039] In this embodiment, the current focus time interval refers to the continuous time range of the previous preset duration at the current time node. This duration can be flexibly configured according to the type and fault characteristics of the equipment in the central control center. For example, for core critical equipment that is prone to sudden failures, the current focus time interval can be set to the previous 30 minutes; for auxiliary equipment with relatively stable operating status, the current focus time interval can be set to the previous 2 hours.

[0040] In this embodiment, the anomaly coefficient difference is the difference between the actual anomaly coefficient and the simulated anomaly coefficient. The anomaly coefficient difference is used to evaluate the prediction accuracy of the digital twin model for the simulated operation data of key monitoring data in the future period. When the anomaly coefficient difference is within the preset anomaly coefficient variation range and the larger the difference, the value range of the generated compensation coefficient is [1, 1.25]. The larger the difference, the larger the compensation coefficient. When the anomaly coefficient difference is not within the preset anomaly coefficient variation range, the value range of the generated compensation coefficient is (0.75, 1) when the absolute value of the anomaly coefficient difference is larger. The preset anomaly coefficient variation range refers to a reasonable range determined based on the statistical analysis of historical operation data, used to judge whether the anomaly coefficient variation of key monitoring data is within the normal fluctuation range.

[0041] In this embodiment, the specific mapping relationship between the anomaly coefficient difference and the compensation coefficient is obtained through machine learning training on historical anomaly coefficient differences and corresponding compensation effects. The specific training process includes: collecting historical anomaly coefficient difference samples and corresponding predicted state evaluation deviation data, constructing a neural network model with the anomaly coefficient difference as input and the compensation coefficient as output, optimizing the model parameters through the gradient descent algorithm so that the compensation coefficient output by the model can minimize the state evaluation deviation, and finally forming a nonlinear mapping relationship between the anomaly coefficient difference and the compensation coefficient.

[0042] In this embodiment, the weighting coefficient of each key monitoring data is determined based on its historical comprehensive impact value. The higher the historical comprehensive impact value, the larger the weighting coefficient.

[0043] In this embodiment, the formula for calculating the comprehensive simulation anomaly coefficient is: Comprehensive simulation anomaly coefficient = Σ (simulation anomaly coefficient × compensation coefficient × weight coefficient).

[0044] In this embodiment, the comprehensive simulation anomaly coefficient is converted into a state value of 0-100. When the comprehensive simulation anomaly coefficient is 0, the state value is 100, which indicates that the equipment is in excellent operating condition. The higher the comprehensive simulation anomaly coefficient, the lower the state value. When the state value is lower than the preset warning threshold, the warning mechanism is triggered.

[0045] In this embodiment, the state value conversion formula is: State value = 100 - (Comprehensive simulation anomaly coefficient × 100). When the comprehensive simulation anomaly coefficient is greater than 1, the state value is treated as 0, so as to ensure that the state value is always within the effective range of 0-100 and intuitively reflects the health status of the equipment.

[0046] In this embodiment, the method also includes a step of converting the actual anomaly coefficient and the simulated anomaly coefficient into the same dimension. Specifically, the actual anomaly coefficient and the simulated anomaly coefficient are mapped to the interval [0,1] through standardization. The standardization formula is: Standardized value = (actual value - historical minimum value) / (historical maximum value - historical minimum value), where the historical minimum value and the historical maximum value are the minimum and maximum values ​​of the actual anomaly coefficient and the simulated anomaly coefficient of the key monitoring data during the historical operation, respectively, so as to ensure the accuracy and consistency of the anomaly coefficient difference calculation.

[0047] In this embodiment, the simulation operation data of key monitoring data of each device in the future period is generated by digital twin model and the simulation anomaly coefficient is calculated. The simulation anomaly coefficient is then compensated and corrected by combining the actual anomaly coefficient. This can effectively improve the accuracy and foresight of the device status assessment and avoid misjudgment or omission caused by model prediction deviation.

[0048] In some embodiments of this application, generating the actual anomaly coefficient of each key monitoring data point within the current time interval of interest includes: Generate a time reference line for the time interval of interest, and map the actual values ​​of key monitoring data within the time interval of interest monitored according to the data monitoring strategy to the time reference line to obtain the data change curve of each key monitoring data in the time interval of interest. Based on pre-defined data feature indicators, feature extraction is performed on the data change curve to generate several feature values; The data characteristic indicators include data fluctuation amplitude, data change rate, data mean deviation, and frequency of outlier occurrence; The actual anomaly coefficient of the corresponding key monitoring data within the current time interval is generated based on several characteristic values ​​and the weighting coefficients of data characteristic indicators.

[0049] In this embodiment, data fluctuation amplitude refers to the difference between the maximum and minimum values ​​in the data change curve. This difference is compared with a preset normal fluctuation threshold to obtain the abnormal fluctuation amplitude ratio, which is the characteristic value of the data fluctuation amplitude index. Data change rate refers to the amount of data change per unit time obtained by calculating the first derivative of the data change curve. The percentage of time the change rate exceeds a preset rate threshold is statistically analyzed to obtain the abnormal rate ratio, which is the characteristic value of the data change rate index. Data mean deviation refers to the percentage deviation obtained by comparing the actual mean of the data change curve with a preset normal mean, which is the characteristic value of the data mean deviation index. The number of times abnormal values ​​appear in the data change curve that exceed the preset normal data range is statistically analyzed, and the proportion of the frequency of abnormal value occurrences to the total number of monitoring times is calculated, which is the characteristic value of the abnormal value occurrence frequency index.

[0050] In this embodiment, the actual anomaly coefficient of key monitoring data within the current monitoring time interval is obtained by weighting the percentage of abnormal fluctuation amplitude, the percentage of abnormal rate, the deviation of the mean, and the frequency of outliers, combined with the preset weights of each characteristic indicator. The weight coefficients of each characteristic indicator are determined based on the contribution of each feature to equipment anomalies in historical fault cases. In this application, the weight coefficients for data fluctuation amplitude, data change rate, data mean deviation, and the frequency of outliers are 0.3, 0.25, 0.25, and 0.2, respectively. The specific calculation formula is: Actual anomaly coefficient = (percentage of abnormal fluctuation amplitude × 0.3) + (percentage of abnormal rate × 0.25) + (deviation of mean × 0.25) + (percentage of outliers × 0.2).

[0051] In this embodiment, by analyzing the multi-dimensional features of the data change curve, it is possible to comprehensively capture the abnormal performance of key monitoring data within the current time interval of interest, providing a reliable practical basis for subsequent simulation anomaly coefficient compensation, thereby further improving the accuracy of equipment status assessment.

[0052] In some embodiments of this application, determining whether to generate a warning command for the corresponding device based on the status value includes: Pre-set the warning status value for each device; When the status value is greater than the warning status value, no warning command is generated for the corresponding device; When the status value is not greater than the warning status value, a warning instruction is generated for the corresponding device. The warning indicators include the warning level of the corresponding device, the warning trigger time, and key abnormal data items.

[0053] In this embodiment, the warning level is determined based on the difference between the status value and the warning status value. The larger the difference, the higher the warning level. The specific level classification standard is as follows: when the status value is in the range of [warning status value - 10, warning status value], the warning level is Level 1; when the status value is in the range of [warning status value - 20, warning status value - 10), the warning level is Level 2; when the status value is less than the warning status value - 20, the warning level is Level 3.

[0054] In this embodiment, the key abnormal data item is the key monitoring data in the comprehensive simulation abnormality coefficient calculation process where (simulation abnormality coefficient × compensation coefficient) is greater than the corresponding simulation abnormality coefficient threshold.

[0055] In this embodiment, the warning status value is set according to the importance of the equipment, the frequency of historical failures, and the operation and maintenance response capability. The higher the importance, the more frequent the failures, and the weaker the operation and maintenance response capability, the higher the warning status value is set.

[0056] In this embodiment, the generated early warning command will be pushed to the monitoring platform of the central control center in real time, and simultaneously notified to the operation and maintenance personnel through audible and visual alarms, pop-up prompts and SMS notifications, so as to ensure that relevant personnel can obtain equipment abnormality information in a timely manner and take countermeasures.

[0057] In some embodiments of this application, a comprehensive early warning report is generated by integrating all device information that issued the early warning command, including: Obtain device information of the device that issued the warning command, and generate a list of warning devices; Analyze the warning instructions for each device in the warning equipment list to obtain the priority score for each device in the list; All devices in the list are prioritized according to their priority scores to obtain a warning priority sequence. The key abnormal data items of the devices are then input into the preset operation and maintenance model to obtain the corresponding operation and maintenance strategy. A comprehensive early warning report is generated based on the early warning priority sequence and all operation and maintenance strategies.

[0058] In this embodiment, the equipment information includes the equipment name, equipment number, equipment type, location, and responsible department.

[0059] In this embodiment, the comprehensive early warning report includes an early warning priority sequence, the operation and maintenance strategy for each device, the early warning trigger time, and key abnormal data items. Details of the key abnormal data items also include the real-time monitoring value, simulation operation data, normal data range, and the calculation process for the abnormality coefficient.

[0060] In this embodiment, the preset operation and maintenance model is constructed based on equipment failure handling cases. It integrates a hybrid decision-making mechanism based on rule reasoning and case reasoning. When key abnormal data items are input, the system first matches the standard operation and maintenance process that matches the abnormal data characteristics through rule reasoning, and then combines case reasoning to call the handling solutions of similar failures in history for supplementation and optimization. Finally, it generates a detailed operation and maintenance strategy that includes failure cause analysis, handling steps, required tools and materials, and precautions.

[0061] In this embodiment, by constructing an early warning priority sequence, maintenance personnel can quickly identify the most urgent equipment anomalies and prioritize the handling of potential faults that may have serious consequences. The resulting comprehensive early warning report can clearly present the priorities of maintenance work and provide a scientific basis for the efficient scheduling of maintenance resources.

[0062] In some embodiments of this application, a priority score is obtained for each device in the list, including: Extract the warning level, equipment importance coefficient, and warning trigger time from each warning instruction, and assign a basic priority score corresponding to the warning level; Multiply the basic priority score by the equipment importance coefficient to obtain the weighted priority score; Calculate the time difference between the current time and the warning trigger time, determine the time correction coefficient based on the time difference, and multiply the weighted priority score by the time correction coefficient to obtain the priority score of the device.

[0063] In this embodiment, a Level 1 warning corresponds to 30 points, a Level 2 warning corresponds to 60 points, and a Level 3 warning corresponds to 90 points.

[0064] In this embodiment, the equipment importance coefficient is comprehensively evaluated based on factors such as the equipment's functional role in the centralized control center system, the scope of its failure impact, and the economic losses caused by historical failures. It is divided into core equipment (1.5), important equipment (1.2), and general equipment (1.0).

[0065] In this embodiment, the correction factor is 1.2 for time differences within 0-30 minutes, 1.0 for time differences between 30 minutes and 1 hour, and 0.8 for time differences exceeding 1 hour.

[0066] In this embodiment, by calculating the priority score of each device that issues an early warning command, the urgency and processing priority of each early warning device can be quantitatively assessed. This avoids decision-making confusion for maintenance personnel when faced with multiple devices issuing early warnings simultaneously, and ensures that limited maintenance resources are prioritized for handling the most critical and urgent device failures, thereby minimizing the impact of device anomalies on the overall operation of the control center.

[0067] In some embodiments of this application, such as Figure 2 As shown, it also includes a digital twin-based method for early warning of equipment status in the centralized control center: S201: Construct a digital twin model of each control center device; S202: Set several key status indicators, determine the key monitoring data of each device based on all key status indicators and generate the corresponding data monitoring strategy for the device, collect data according to the data monitoring strategy, and generate the real-time key dataset for each device. S203: Map the real-time correlated dataset to the digital twin model of the corresponding device, obtain the simulation results and evaluate the device status to obtain the status value of the corresponding device; S204: Determine whether to generate a warning command for the corresponding device based on the status value, and integrate all device information that issued the warning command to generate a comprehensive warning report.

[0068] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. A centralized control center equipment status early warning system based on digital twins, characterized in that, include: Modules are used to build digital twin models of each control center device; The data acquisition module is used to set several key status indicators, determine the key monitoring data of each device based on all key status indicators, generate the corresponding data monitoring strategy for the device, collect data according to the data monitoring strategy, and generate real-time key datasets for each device. The evaluation module is used to map the real-time correlated dataset to the digital twin model of the corresponding device, obtain the simulation results and evaluate the device status to obtain the status value of the corresponding device. The early warning module is used to determine whether to generate an early warning command for the corresponding device based on the status value, and to integrate all device information that issued the early warning command to generate a comprehensive early warning report.

2. The centralized control center equipment status early warning system based on digital twin as described in claim 1, characterized in that, Construct a digital twin model for each control center device, including: Obtain the 3D design drawings, equipment parameter manuals and historical operating data of each of the central control center devices, and construct the geometric model of the central control center devices based on 3D modeling technology; Collect the physical property parameters of each key component of the equipment, and construct the physical model of the equipment by combining it with multiphysics simulation algorithms; By connecting to the device's real-time monitoring system and historical database through data interfaces, a behavioral model of the device can be established; By integrating geometric models, physical models, and behavioral models, a digital twin model of the corresponding device is formed.

3. The centralized control center equipment status early warning system based on digital twin as described in claim 1, characterized in that, Based on all key status indicators, the key monitoring data for each device are determined, and a corresponding data monitoring strategy for that device is generated, including: Establish a historical monitoring data source sequence for each device, wherein the historical monitoring data source sequence includes several historical monitoring data sources; Historical data packets for each key status indicator are generated based on historical monitoring logs. The historical data packets include several historical status assessment values ​​for the corresponding key status indicator, as well as historical values ​​for each historical monitoring data source for each device of each historical node corresponding to each historical status assessment value. Generate historical impact values ​​of each historical monitoring data source on each key status indicator based on historical data packets; The historical comprehensive impact value of the corresponding historical monitoring data source is generated based on the historical impact values ​​of the same historical monitoring data source on all key status indicators and the weight coefficients of the corresponding key status indicators. Based on historical comprehensive impact values, key monitoring data for each device are determined, and monitoring sub-strategies for each key monitoring data are generated. Generate a data monitoring strategy for the corresponding device based on all key monitoring data of the same device and the corresponding monitoring sub-strategies.

4. The centralized control center equipment status early warning system based on digital twin as described in claim 3, characterized in that, Based on historical comprehensive impact values, key monitoring data for each device are determined, and monitoring sub-strategies are generated for each key monitoring data point, including: The impact level of the corresponding historical monitoring data source is set according to the historical comprehensive impact value; The impact level threshold is preset, and the impact level threshold is mapped to a corresponding preset monitoring strategy; Set the historical monitoring data corresponding to the historical monitoring data source whose impact level is greater than the impact level threshold as the key monitoring data of the corresponding equipment; Calculate the difference between the impact level of key monitoring data and the impact level threshold. Based on the level difference-correction coefficient mapping table, match the correction coefficient of the current level difference to obtain the strategy correction coefficient of key monitoring data. Based on the preset monitoring strategy and strategy correction coefficient, a monitoring sub-strategy is generated to correspond to the key monitoring data. The monitoring sub-strategy for each key monitoring data point is generated sequentially.

5. The centralized control center equipment status early warning system based on digital twin as described in claim 1, characterized in that, Perform equipment status assessment to obtain the status values ​​of the corresponding equipment, including: Based on the data monitoring strategy, several real-time key monitoring data points for each device at the current time point are generated, and a real-time key dataset is generated. The real-time key dataset is input into the digital twin model of the corresponding equipment, and the multi-physics coupling simulation algorithm inside the model is used to perform real-time simulation calculations to obtain the simulation operation data of each key monitoring data of the equipment in the future period. The simulation data is compared with the preset normal data range, and the simulation anomaly coefficient of each key monitoring data in the future period is generated based on the comparison results. Generate the actual anomaly coefficient for each key monitoring data point within the current time interval of interest; Calculate the difference between the actual anomaly coefficient and the simulated anomaly coefficient of the same key monitoring data, and set the compensation coefficient of the simulated anomaly coefficient of the corresponding key monitoring data based on the difference in anomaly coefficient. The comprehensive simulation anomaly coefficient of the corresponding equipment is obtained by weighted summation of the simulation anomaly coefficient, compensation coefficient and corresponding weight coefficient of each key monitoring data. The state value of each device is generated based on the comprehensive simulation anomaly coefficient of each device.

6. The centralized control center equipment status early warning system based on digital twin as described in claim 5, characterized in that, Generate the actual anomaly coefficient for each key monitoring data point within the current time interval of interest, including: Generate a time reference line for the time interval of interest, and map the actual values ​​of key monitoring data within the time interval of interest monitored according to the data monitoring strategy to the time reference line to obtain the data change curve of each key monitoring data in the time interval of interest. Based on pre-defined data feature indicators, feature extraction is performed on the data change curve to generate several feature values; The data characteristic indicators include data fluctuation amplitude, data change rate, data mean deviation, and frequency of outlier occurrence; The actual anomaly coefficient of the corresponding key monitoring data within the current time interval is generated based on several characteristic values ​​and the weighting coefficients of data characteristic indicators.

7. The centralized control center equipment status early warning system based on digital twin as described in claim 5, characterized in that, Based on the status value, determine whether to generate a warning command for the corresponding device, including: Pre-set the warning status value for each device; When the status value is greater than the warning status value, no warning command is generated for the corresponding device; When the status value is not greater than the warning status value, a warning instruction is generated for the corresponding device. The warning indicators include the warning level of the corresponding device, the warning trigger time, and key abnormal data items.

8. The centralized control center equipment status early warning system based on digital twin as described in claim 7, characterized in that, Integrate information from all devices that issued early warning commands to generate a comprehensive early warning report, including: Obtain device information of the device that issued the warning command, and generate a list of warning devices; Analyze the warning instructions for each device in the warning equipment list to obtain the priority score for each device in the list; All devices in the list are prioritized according to their priority scores to obtain a warning priority sequence. The key abnormal data items of the devices are then input into the preset operation and maintenance model to obtain the corresponding operation and maintenance strategy. A comprehensive early warning report is generated based on the early warning priority sequence and all operation and maintenance strategies.

9. The centralized control center equipment status early warning system based on digital twin as described in claim 8, characterized in that, Obtain the priority score for each device in the list, including: Extract the warning level, equipment importance coefficient, and warning trigger time from each warning instruction, and assign a basic priority score corresponding to the warning level; Multiply the basic priority score by the equipment importance coefficient to obtain the weighted priority score; Calculate the time difference between the current time and the warning trigger time, determine the time correction coefficient based on the time difference, and multiply the weighted priority score by the time correction coefficient to obtain the priority score of the device.

10. A method for early warning of equipment status in a centralized control center based on digital twins, characterized in that, include: Construct a digital twin model of each control center device; Several key status indicators are set, and key monitoring data for each device are determined based on all key status indicators. A corresponding data monitoring strategy for the device is generated, and data is collected according to the data monitoring strategy to generate a real-time key dataset for each device. The real-time correlated dataset is mapped to the digital twin model of the corresponding device to obtain simulation results and device status assessment, thereby obtaining the status value of the corresponding device. Based on the status value, determine whether to generate a warning command for the corresponding device, and integrate all device information that issued the warning command to generate a comprehensive warning report.

Images