Real-time monitoring and risk visualization early warning system for power transmission line based on multi-source data

Abstract

The application discloses a power transmission line real-time monitoring and risk visualization early warning system based on multi-source data, and relates to the technical field of intelligent monitoring of power transmission lines; firstly, trend prediction is completed based on an initial digital twin model and meteorological prediction data; then, trend measured information collected by intelligent monitoring equipment is called, and the change trends of risk factors in prediction and measurement are compared; when the deviation between the two exceeds a threshold value, the latest line state data and meteorological prediction data are called to complete model updating, so that the digital twin model always fits the real working conditions and external environment changes of the power transmission line; the application can accurately adapt to dynamic fluctuations of complex weather, completely solves the prediction inaccuracy problem caused by meteorological prediction deviation and environmental mutation of a traditional fixed model, significantly improves the reliability and timeliness of trend prediction information, and provides accurate data support for risk identification and disposal.

Description

Technical Field

[0001] This invention belongs to the field of intelligent monitoring technology for power transmission lines, specifically a real-time monitoring and risk visualization early warning system for power transmission lines based on multi-source data. Background Technology

[0002] As the core carrier of power transmission, transmission lines are susceptible to the combined effects of multiple factors, including meteorological environment, topography, and line aging, placing high demands on real-time monitoring and risk early warning. Existing transmission line monitoring and early warning technologies mostly rely on a single data source to collect line status information and use predictive models with fixed parameters and structures to extrapolate status trends, enabling only simple line status monitoring and routine risk alerts under basic operating conditions. Current technologies typically build predictive models directly based on historical data and initial meteorological information, and these models remain static for extended periods, making it difficult to adapt to the dynamic changes in the actual operating environment.

[0003] This invention provides a real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data to solve the above-mentioned technical problems. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data.

[0005] To achieve the above objectives, a first aspect of the present invention provides a real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data, comprising:

[0006] Monitoring and early warning module: used to predict the trend of transmission lines based on digital twin models;

[0007] Used to verify trend prediction information using measured trend information, and update the digital twin model based on the verification results; using the updated digital twin model to predict the trend information of transmission lines; and,

[0008] It is used to extract risk information from trend forecast information, formulate disposal measures based on risk information, and complete the scheduling of maintenance personnel.

[0009] In one possible implementation, trend prediction information for transmission lines is predicted based on a digital twin model, including:

[0010] Retrieve line status data and meteorological forecast data of transmission lines; combine line status data with digital twin technology to construct a digital twin model of the transmission lines;

[0011] Based on digital twin models and meteorological forecast data, the state of transmission lines is simulated, and trend prediction information of transmission lines is extracted.

[0012] In one possible implementation, trend prediction information is verified using measured trend information, including:

[0013] Retrieve measured trend information of transmission lines; the measured trend information is collected through intelligent monitoring equipment.

[0014] Determine whether the trends of risk factors in the measured trend information and the predicted trend information are consistent; if yes, do not update the digital twin model; otherwise, update the digital twin model.

[0015] In one possible implementation, determining whether the trends of risk factor changes in the measured trend information and the predicted trend information are consistent includes:

[0016] The real-time value gap and trend gap of risk factors are calculated based on the measured trend information and the predicted trend information.

[0017] The trend assessment score is obtained by weighted summation of the real-time value difference and the trend difference; the weighting coefficients of the real-time value difference and the trend difference are obtained through historical data statistics.

[0018] When the trend assessment score is less than the preset assessment score threshold, the risk factors are judged to have consistent trends; otherwise, the risk factors are judged to have inconsistent trends.

[0019] In one possible implementation, the real-time value gap and trend gap of the risk factor are calculated based on measured trend information and predicted trend information, including:

[0020] Extract the latest measured values ​​of risk factors from the trend measurement information, and calculate the numerical difference between the measured values ​​and the predicted values ​​in the trend prediction information as the real-time value gap.

[0021] Extract several measured values ​​of risk factors from the trend measurement information, calculate the numerical difference between several measured values ​​and the corresponding predicted values, and obtain several historical value differences;

[0022] The mean squared difference of several historical values ​​is used as the trend gap.

[0023] In one possible implementation, the weighting coefficients for the real-time value difference and the trend difference are dynamically adjusted, including:

[0024] Extract the time range corresponding to the risk factors from the actual trend measurement information;

[0025] Determine if the time range exceeds the preset time range threshold; if yes, do not adjust the weighting coefficients of the real-time value difference and the trend difference; otherwise, fix the weighting coefficient of the real-time value difference to 1.

[0026] In one possible implementation, before updating the digital twin model based on the verification result, a determination is made based on the verification result to determine whether to update the digital twin model, including:

[0027] Based on the real-time value gap and trend gap, a change prediction model is used to predict the probability of change of risk factors; the change prediction model is built on an artificial intelligence model.

[0028] Determine if the probability of change is greater than a preset probability threshold; if yes, update the digital twin model; otherwise, calculate the time difference between the real-time value acquisition time and the last digital twin model update time.

[0029] If the time difference is less than the preset time difference threshold, the digital twin model will not be updated; otherwise, the digital twin model will be updated.

[0030] In one possible implementation, the updated digital twin model is used to predict trend forecasting information for transmission lines, including:

[0031] Retrieve the updated digital twin model; the update of the digital twin model is based on the updated line status data of the transmission line.

[0032] By combining updated meteorological forecast data and digital twin models, the trend forecast information of transmission lines is re-predicted.

[0033] In one possible implementation, risk information is extracted based on trend forecasting information, including:

[0034] Based on trend forecasting information, determine whether risk factors affect the safe operation of transmission lines; if so, extract the degree and duration of the impact of the risk factors.

[0035] The impact level and duration of each risk factor are integrated into risk information.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] 1. This invention verifies trend prediction information using measured trend information, dynamically updates the digital twin model based on the verification results, and re-predicts transmission line trend information using the updated model. It integrates line status data, meteorological forecast data, and real-time monitoring data to form a closed-loop working mechanism of prediction-verification-update-repreneurial prediction. This invention first completes trend prediction based on the initial digital twin model and meteorological forecast data, then retrieves measured trend information collected by intelligent monitoring equipment, compares the predicted and measured risk factor change trends, and immediately retrieves the latest line status data and meteorological forecast data to update the model when the deviation exceeds a threshold. This ensures that the digital twin model always closely matches the actual operating conditions of the transmission line and changes in the external environment. This invention can accurately adapt to the dynamic fluctuations of complex weather, completely solving the problem of inaccurate predictions caused by meteorological forecast deviations and sudden environmental changes in traditional fixed models, significantly improving the reliability and timeliness of trend prediction information, and providing accurate data support for risk identification and management.

[0038] 2. This invention, when verifying trend prediction information using measured trend information, employs a comprehensive judgment method that weights and sums the real-time value difference and the trend difference, replacing the single-dimensional judgment logic, and accurately quantifies the consistency between risk factor prediction and actual measurement. This invention first extracts the numerical difference between the latest measured value and the predicted value of the risk factor as the real-time value difference, then extracts the numerical differences between multiple sets of historical measured values ​​and predicted values, calculates the mean square error to obtain the trend difference, and combines this with weighted summations based on historical data statistics to obtain a trend evaluation score. The score is then compared with a threshold to determine whether the trend is consistent. This invention considers both instantaneous numerical deviation and long-term trend... The system comprehensively assesses prediction accuracy from two dimensions: trend deviation and periodicity deviation. This avoids misjudging trend deviation due to single-point numerical fluctuations and prevents prediction failure due to unidentified trend deviations. It significantly improves the accuracy and stability of trend consistency judgment, effectively addresses the limitations of traditional single judgment methods, reduces the frequency of invalid updates to the digital twin model, lowers the system's computational load, and ensures that the predicted information closely matches the actual operating conditions. This makes risk factor identification and risk information extraction more credible, providing a scientific basis for the formulation of subsequent disposal measures and the scheduling of maintenance personnel, and comprehensively improving the precision and intelligence of transmission line monitoring and early warning. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram illustrating the working steps of the real-time monitoring and risk visualization early warning system for power transmission lines in an embodiment of the present invention;

[0041] Figure 2 This is a schematic diagram illustrating the working steps of verifying trend prediction information using measured trend information in an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram illustrating the steps involved in extracting risk information based on trend prediction information in an embodiment of the present invention. Detailed Implementation

[0043] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] When monitoring transmission lines, their operational status is easily affected by a combination of factors, including fluctuations in external meteorological conditions, differences in topography, and line aging. Existing solutions that rely on a single data source or a fixed model for trend extrapolation are ill-suited to adapting in real time to the dynamic changes in complex weather conditions such as strong winds, heavy rainfall, frost, ice accumulation, and sudden temperature fluctuations. This can easily lead to problems such as large deviations in trend predictions and inaccurate identification of potential anomalies.

[0045] To ensure the accuracy of transmission line condition prediction, this invention verifies the prediction data using measured data of the transmission line. If the deviation between the two is large, updated multi-source data is retrieved to update the digital twin model, ensuring that the digital twin model fits reality and improving the accuracy of transmission line condition prediction.

[0046] Please see Figure 1 The first aspect of this invention provides a real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data, including a monitoring and early warning module. This module can be connected to a data storage module and intelligent monitoring equipment to obtain the data required for transmission line monitoring and early warning. The workflow of the monitoring and early warning module is as follows:

[0047] A01: Trend prediction information for transmission lines based on digital twin models;

[0048] Trend forecast information includes the changing trends of risk factors affecting the stable operation of transmission lines. These risk factors are mainly caused by changes in the external meteorological environment, including icing thickness, sag value, conductor temperature, and conductor galloping amplitude.

[0049] This invention utilizes digital twin technology to construct a digital twin model of a power transmission line, which can simulate the impact of external meteorological conditions on the transmission line. By combining this model with meteorological forecast data of the area where the transmission line is located from a third-party meteorological forecasting platform, trend prediction information for the transmission line can be obtained.

[0050] This invention achieves high-fidelity virtual-real mapping of transmission lines by constructing a digital twin model of the transmission line, which can accurately replicate the real operating conditions of the transmission line. It also has the ability to simulate multiple meteorological conditions and, combined with meteorological forecast data, can realize the state prediction of the transmission line, thus ensuring the mapping reliability of the digital twin model.

[0051] A02: Verify trend prediction information using measured trend information, update the digital twin model based on the verification results, and use the updated digital twin model to predict the trend prediction information of transmission lines.

[0052] Trend-based measured information is data collected by intelligent monitoring devices deployed on power transmission lines. It reflects the true condition of the transmission lines and is considered more reliable. Intelligent monitoring devices are primarily used to collect measured data; for example, icing sensors collect icing thickness, sag sensors collect sag values, and temperature and humidity sensors collect temperature and humidity data. The selection and setup of intelligent monitoring devices are aimed at verifying trend prediction information and are not limited to the examples mentioned above.

[0053] This invention verifies trend prediction information using measured trend data. If the trend prediction information does not match the actual state of the transmission line as expected, the risk information identified and extracted based on the trend prediction information is also unreliable, and the suitability of the targeted response measures for the actual anomaly cannot be guaranteed. In this case, it is necessary to update the digital twin model in a timely manner to obtain more reliable trend prediction information.

[0054] A03: Extract risk information from trend forecasting information, formulate response measures based on the risk information, and complete the scheduling of maintenance personnel.

[0055] Risk information includes the degree and duration of the impact of risk factors on transmission lines. The degree of impact on transmission lines should reach the level that affects the safe operation of the transmission lines. For example, if the predicted icing thickness has exceeded the preset icing thickness threshold, then it is necessary to formulate targeted measures.

[0056] This invention identifies risk information affecting the safe operation of transmission lines based on trend prediction information, formulates countermeasures to eliminate the impact of risk information on transmission lines, and dispatches maintenance personnel to implement countermeasures to eliminate hidden dangers; at the same time, the trend prediction information is visualized to facilitate timely adjustment of countermeasures by staff.

[0057] In order to establish a reliable virtual mapping of transmission lines to simulate the complex operating conditions of transmission lines, this invention is based on digital twin technology. It constructs a digital twin model by combining the line status data of the transmission lines, and then introduces meteorological forecast data to simulate the complex operating conditions of the transmission lines. Through simulation and prediction, trend prediction information of the transmission lines is obtained.

[0058] The implementation process of predicting the trend information of transmission lines based on the digital twin model in this invention is as follows:

[0059] B01: Retrieve line status data and meteorological forecast data for power transmission lines;

[0060] Line status data includes line structural data and real-time status data. Line structural data includes parameter information of the structure of the transmission line, such as towers, conductors, and insulators. Real-time status data includes data on various forces, deformations, and loads on the transmission line; force data includes ice thickness, sag value, and conductor galloping amplitude, while deformation data includes tower tilt angle, foundation settlement, and guy wire status.

[0061] Meteorological forecast data is time-series meteorological data obtained through forecasting. It is used to simulate the impact of external meteorological environments on power transmission lines, thereby predicting potential safety hazards. Meteorological forecast data can be obtained through third-party meteorological platforms or by installing meteorological forecasting equipment on the power transmission lines. The time range corresponding to the meteorological forecast data determines the time range corresponding to the trend forecast information. Taking into account maintenance preparation time and the accuracy of meteorological data forecasts, the time range of meteorological forecast data can be one week, fifteen days, or one month.

[0062] It should be noted that the line status data of transmission lines and meteorological forecast data need to be spatiotemporally registered and normalized in order to achieve multi-source data fusion.

[0063] B02: Combining line status data with digital twin technology to construct a digital twin model of transmission lines;

[0064] This invention utilizes digital twin technology to create a high-precision model of the transmission line structure, fully reproducing the structural parameters, design thresholds, and other inherent attributes of the transmission line. It simulates the current real-world state of the transmission line using line status data. The digital twin model can also interface with meteorological forecast data, using meteorological time-series data as output parameters to dynamically deduce the evolution of the line state along a time axis, realistically reproducing the changes in sag, tension, tilt, and icing under different meteorological conditions. The method for constructing the digital twin model has been disclosed in existing solutions and will not be elaborated upon here.

[0065] B03: Based on digital twin models and meteorological forecast data, the state of transmission lines is simulated to extract trend prediction information of transmission lines.

[0066] After constructing a digital twin model, state simulation of the transmission line is performed based on meteorological forecast data. This allows us to obtain the state changes of the transmission line under the influence of external meteorological environment. The changing trends of each preset risk factor are extracted from the simulation process and integrated to obtain trend prediction information. The trend prediction information may contain multiple risk factors that jointly affect the transmission line, such as simultaneous anomalies in icing thickness and sag.

[0067] This invention uses a combination of digital twin models and meteorological forecast data to predict the trend of transmission lines, enabling accurate simulation of the operating conditions of transmission lines and laying a reliable data foundation for accurately identifying the risk information of transmission lines.

[0068] Meteorological forecast data is, by its very nature, forecast data and is not absolutely reliable; moreover, its reliability decreases the longer the forecast period. When inputting meteorological forecast data into a digital twin model to predict the trend of transmission lines, it is necessary to predict the trend over a longer period of time to allow sufficient time for maintenance personnel to prepare. However, the trend prediction information may deviate from the actual situation due to the reliability of the meteorological forecast data. This can lead to inaccuracies in the risk information identified based on the trend prediction information, making it impossible for maintenance personnel to promptly avoid abnormal situations actually encountered by the transmission lines.

[0069] This invention retrieves measured trend information of transmission lines, uses this measured information to verify the reliability of trend prediction information, and if the trend prediction information is unreliable, updates the line status data and meteorological forecast data to update the digital twin model, thereby re-predicting and obtaining trend prediction information. Please refer to [link / reference]. Figure 2 The implementation process of this invention to verify trend prediction information using measured trend information is as follows:

[0070] C01: Retrieve measured trend information of transmission lines; the measured trend information is collected through the set intelligent monitoring equipment;

[0071] After collecting measured trend information, the intelligent monitoring equipment on the transmission line stores it in the storage module, which can be directly retrieved when trend prediction information needs to be verified. The measured trend information includes the historical trend changes corresponding to each risk factor and is generated based on the measured data collected by the intelligent monitoring equipment.

[0072] As mentioned above, the measured trend information is used to verify the trend prediction information. Therefore, after obtaining the trend prediction information, a verification time can be set according to a preset time interval. At the verification time, the measured trend information can be retrieved and the trend prediction information can be verified.

[0073] C02: Determine whether the trends of risk factors in the measured trend information and the predicted trend information are consistent; if yes, do not update the digital twin model; if no, update the digital twin model.

[0074] Since measured trend information is more reliable, if the changing trends of risk factors in the trend prediction information are consistent with the measured trend information, it indicates that the trend prediction information is accurate and reliable. In this case, there is no need to update the digital twin model, and the risk information identified based on the trend prediction information is more credible. Conversely, if the changing trends of risk factors are inconsistent, it indicates that there is a deviation between the trend prediction information and the actual situation. In this case, it is necessary to update the digital twin model and re-acquire the trend prediction information to ensure its reliability.

[0075] For example, after obtaining the trend prediction information of the transmission line, a verification time is determined every hour. As time progresses, at each verification time, the measured trend information of the transmission line is retrieved from the data storage module. This measured trend information should contain the latest measured data, and the trend prediction information is verified using this measured trend information.

[0076] When determining whether the changing trends of risk factors in measured and predicted trend information are consistent, if the changing trends are completely consistent, it naturally indicates that the trend prediction information closely matches the actual situation and has high reliability. However, under normal circumstances, the changing trends of risk factors are rarely completely consistent. Therefore, this invention, without affecting the accurate identification of risk information, combines the difference in real-time values ​​and the difference in trends to comprehensively determine whether the changing trends of risk factors are consistent, thus avoiding frequent updates to the digital twin model due to slight differences in changing trends.

[0077] The implementation process of this invention for determining whether the changing trends of risk factors in measured trend information and predicted trend information are consistent is as follows:

[0078] D01: Calculate the real-time value gap and trend gap of risk factors based on actual trend information and trend forecast information;

[0079] The real-time value gap is the numerical difference between the latest measured value and the predicted value of a risk factor, while the trend gap is the trend difference between the historical trend and the predicted trend of a risk factor. Both the real-time value gap and the trend gap are calculated by extracting data from measured trend information and predicted trend information, respectively.

[0080] D02: The trend assessment score is obtained by weighted summation of the real-time value difference and the trend difference; the weighting coefficients of the real-time value difference and the trend difference are obtained through historical data statistics.

[0081] The weighting coefficients for real-time value differences and trend differences are obtained through historical data statistics. The greater the correlation with the influence of risk factors, the greater their weight and the larger their weighting coefficient. These weighting coefficients can be obtained using principal component analysis.

[0082] The trend assessment score is used to comprehensively judge whether the trend of the predicted value of the risk factor is consistent with the trend of the measured value. The smaller the trend assessment score, the higher the degree of consistency, and the larger the trend assessment score, the lower the degree of consistency.

[0083] D03: When the trend assessment score is less than the preset assessment score threshold, the risk factors are judged to have consistent trends; otherwise, the risk factors are judged to have inconsistent trends.

[0084] The assessment score threshold is set based on historical experience. If the trend assessment score is less than the assessment score threshold, the risk factors are considered to have consistent trends; otherwise, the risk factors are considered to have inconsistent trends.

[0085] When determining whether the changing trends of risk factors are consistent, this invention introduces real-time value differences and trend differences for comprehensive judgment. It makes a consistency judgment on the changing trends from both instantaneous and historical perspectives, avoiding the limitations of a single perspective and improving the accuracy of consistency judgment, thereby enabling accurate updates to the digital twin model.

[0086] If the external meteorological environment changes drastically compared to the meteorological forecast data, it may directly affect the measured values ​​of risk factors, resulting in a significant difference between the measured and forecast values. However, changes in a single value may not necessarily affect the overall trend of risk factors. Therefore, when assessing consistency, it is necessary to conduct a comprehensive evaluation from both instantaneous and historical perspectives.

[0087] The implementation process for calculating the real-time value difference and trend difference of risk factors based on measured trend information and predicted trend information in this invention is as follows:

[0088] E01: Extract the latest measured value of the risk factor from the trend measurement information, and calculate the difference between the measured value and the predicted value in the trend prediction information as the real-time value gap.

[0089] The latest measured value represents the current state of the risk factor. The predicted value for the corresponding moment can be extracted from the trend prediction information. The difference between the measured value and the predicted value is called the real-time value gap. The larger the real-time value gap, the lower the reliability of the risk factor prediction data.

[0090] E02: Extract several measured values ​​of risk factors from the actual trend information, calculate the numerical difference between several measured values ​​and the corresponding predicted values, and obtain several historical value differences.

[0091] The trend measurement information includes historical measured values ​​of risk factors. The time range corresponding to the risk factors in the trend measurement information can be divided into several historical periods according to a preset time interval, and the measured values ​​corresponding to these historical periods can be extracted. It should be noted that the measured values ​​corresponding to these historical periods may include the latest measured values ​​of the risk factors.

[0092] E03: The mean squared difference of the differences between several historical values ​​is used as the trend difference.

[0093] If the time range corresponding to the risk factor in the trend measurement information is short and the number of historical moments divided according to the preset time interval is small, it is difficult to quantify the impact of the trend gap on the risk factor. Therefore, the weight coefficient of the real-time value gap should be increased and the weight coefficient of the trend gap should be decreased.

[0094] The implementation process of dynamically adjusting the weighting coefficients of real-time value difference and trend difference in this invention is as follows:

[0095] F01: Extract the time range corresponding to the risk factors in the actual trend information;

[0096] The time range refers to the time period corresponding to the measured data of risk factors collected by intelligent monitoring equipment after trend prediction information is obtained. This time range continuously expands as the measured data increases, therefore adjusting the weighting coefficients based on the time range is a dynamic process.

[0097] F02: Determine if the time range is greater than the preset time range threshold; if yes, do not adjust the weight coefficients of the real-time value difference and the trend difference; otherwise, fix the weight coefficient of the real-time value difference to 1.

[0098] The time range threshold is used to determine whether the trend gap of the risk factor should be calculated, and this time range threshold is set based on experience.

[0099] For example, suppose the trend measurement information includes several measured values ​​of risk factor A, and these measured values ​​correspond to a time range of three days. The preset time range threshold is one day. Then it is necessary to calculate the trend gap of risk factor A. Therefore, the weight coefficients of the real-time value gap and the trend gap are set to 0.5 and 0.5, respectively.

[0100] Several historical moments are determined from a time range at preset hourly intervals. Interpolation matching is used to determine the measured values ​​for these historical moments from among the measured values. The data difference between the measured values ​​and the predicted values ​​in the trend prediction information is calculated as the historical value difference. The mean squared error of these historical value differences is used as the trend gap. Simultaneously, the data difference between the latest measured value of risk factor A and its corresponding predicted value is used as the real-time value gap.

[0101] The real-time value difference and trend difference are weighted and summed according to the set weight coefficients to obtain the trend assessment score. If the trend assessment score is less than the preset assessment score threshold, it is determined that the risk factor change trend is consistent.

[0102] It should be noted that the measured and predicted values ​​were normalized before calculation.

[0103] Please see Figure 3 After determining that the changing trends of risk factors are inconsistent, these inconsistent trends do not necessarily lead to changes in risk information. Directly updating the digital twin model could result in an ineffective update. This invention assesses the probability that real-time value discrepancies and trend discrepancies might cause changes in risk information before updating the digital twin model. Based on this probability, it further determines whether an update to the digital twin model is necessary. The implementation process is as follows:

[0104] G01: Based on real-time value gaps and trend gaps, a change prediction model is used to predict the probability of changes in risk factors; whereby the change prediction model is built on an artificial intelligence model.

[0105] The change prediction model is used to predict the impact of real-time value gaps and trend gaps on the changing trends of risk factors. The change prediction model is obtained by training an artificial intelligence model (AI model, including a BP neural network or RBF neural network model) using a training sample set. The training sample set includes several sample data points, including training input data and training output data. The training input data includes risk factor identifiers, real-time value gaps, and trend gaps. The training output data includes an indicator of whether the risk factor has changed; if the degree or duration of the risk factor's influence changes under the influence of the training input data, the training output data is 1; otherwise, it is 0. The training sample set can be extracted from historical data or simulated using a digital twin model. Simulated sample data has been verified in practice. The AI ​​model uses an existing structure without modification; its training process is already publicly available in existing solutions and will not be elaborated here.

[0106] The real-time value difference and trend difference of the risk factors are input into the change prediction model to obtain the change indicator of the risk factors and their confidence level. If the risk factor changes, its confidence level is used as the change probability. If the risk factor does not change, the change probability is set to 0.

[0107] G02: Determine whether the probability of change is greater than the preset probability threshold; if yes, update the digital twin model; if no, calculate the time difference between the real-time value acquisition time and the last digital twin model update time.

[0108] Compare the probability of change with a preset probability threshold (e.g., set to 0.85). If the probability of change is greater than the preset probability threshold, it is determined that the risk factor is highly likely to change. In this case, the digital twin model needs to be updated immediately to obtain more reliable trend prediction information.

[0109] G03: If the time difference is less than the preset time difference threshold, the digital twin model will not be updated; otherwise, the digital twin model will be updated.

[0110] If the probability of change is not greater than a preset probability threshold, since both the transmission line status data and meteorological forecast data may have changed, it is necessary to determine whether the digital twin model needs to be updated based on the time difference between the last update and the current time. The time threshold can be set to three days or one week.

[0111] When updating a digital twin model, a comprehensive update is required, rather than updating only a known change in data. This ensures that the digital twin model always reflects the actual situation of the transmission line, thereby ensuring the reliability of trend prediction information.

[0112] The implementation process of this invention, which uses an updated digital twin model to predict the trend information of transmission lines, is as follows:

[0113] H01: Retrieve the updated digital twin model; the update of the digital twin model is based on the updated line status data of the transmission line.

[0114] As mentioned above, the line status data includes structural parameters, stress conditions, etc. All the latest data is retrieved during the update process. The digital twin model is updated using this latest status data of the transmission line to ensure a high-fidelity virtual mapping of the transmission line.

[0115] H02: Re-predicting the trend forecast information of transmission lines by combining updated meteorological forecast data and digital twin models.

[0116] After the digital twin model is updated, it also uses the latest meteorological forecast data to simulate the operating status of transmission lines, thereby predicting trend information. This trend prediction information is more reliable than the trend prediction information before the digital twin model was updated.

[0117] As mentioned earlier, trend forecasting information includes the changing trends of various risk factors, but not all risk factors will necessarily affect the safe operation of transmission lines. This invention identifies the risk factors affecting the safe operation of transmission lines, their degree of impact, and the duration of their impact based on trend forecasting information, integrates the identification results into risk information, and then formulates more reliable response measures.

[0118] The implementation process of extracting risk information based on trend prediction information in this invention is as follows:

[0119] I01: Determine whether risk factors affect the safe operation of transmission lines based on trend forecast information; if yes, extract the degree and duration of the impact of the risk factors.

[0120] To determine whether a risk factor affects the safe operation of a transmission line, a threshold parameter can be set. For example, if the risk factor is icing thickness, an icing thickness threshold affecting the safe operation of the transmission line can be pre-set. Once the trend of the risk factor exceeds the icing thickness threshold, the corresponding icing thickness (which serves as the degree of impact for developing response measures) and its duration of impact can be extracted.

[0121] I02: Integrate the degree and duration of the impact of each risk factor into risk information.

[0122] If multiple risk factors affect the safe operation of transmission lines, the timing and severity of their impacts should be integrated to obtain risk information. The timing of each risk factor's impact may overlap, which can be considered when developing appropriate countermeasures.

[0123] Trend forecasting information and digital twin models are visualized. After obtaining risk information, response measures can be formulated based on the degree and duration of the impact of the risk factors, thereby dispatching maintenance personnel to eliminate potential hazards according to the response measures.

[0124] It should be noted that once the digital twin model corresponding to the transmission line is updated, the trend prediction information of the transmission line is immediately obtained through this digital twin model. Once the trend prediction information is obtained, the risk information of the transmission line is immediately extracted based on the trend prediction information.

[0125] Since the formulation of response measures and the dispatch of maintenance personnel both rely on the extracted risk information, once the risk information changes, the response measures and maintenance personnel are updated simultaneously, thereby achieving the precise elimination of safety hazards in transmission lines through efficient resource allocation.

[0126] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments.

[0127] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any other combination thereof. When implemented using a software program, it can be implemented entirely or partially in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0128] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data, characterized in that, include: Monitoring and early warning module: used to predict the trend of transmission lines based on digital twin models; Used to verify trend prediction information through trend measurement information, update the digital twin model based on the verification results, and use the updated digital twin model to predict the trend prediction information of transmission lines; as well as, It is used to extract risk information from trend forecast information, formulate disposal measures based on risk information, and complete the scheduling of maintenance personnel.

2. The real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data according to claim 1, characterized in that, Trend prediction information for transmission lines based on digital twin models includes: Retrieve line status data and meteorological forecast data of transmission lines; combine line status data with digital twin technology to construct a digital twin model of the transmission lines; Based on digital twin models and meteorological forecast data, the state of transmission lines is simulated, and trend prediction information of transmission lines is extracted.

3. The real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data according to claim 1, characterized in that, Verify trend prediction information using actual trend measurement data, including: Retrieve measured trend information of transmission lines; the measured trend information is collected through intelligent monitoring equipment. Determine whether the trends of risk factors in the measured trend information and the predicted trend information are consistent; if yes, do not update the digital twin model; otherwise, update the digital twin model.

4. The real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data according to claim 3, characterized in that, Determine whether the trends of risk factors in the actual trend information and the trend forecast information are consistent, including: The real-time value gap and trend gap of risk factors are calculated based on the measured trend information and the predicted trend information. The trend assessment score is obtained by weighted summation of the real-time value difference and the trend difference; the weighting coefficients of the real-time value difference and the trend difference are obtained through historical data statistics. When the trend assessment score is less than the preset assessment score threshold, the risk factors are judged to have consistent trends; otherwise, the risk factors are judged to have inconsistent trends.

5. The real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data according to claim 4, characterized in that, The real-time value gap and trend gap of risk factors are calculated based on actual trend information and trend forecast information, including: Extract the latest measured values ​​of risk factors from the trend measurement information, and calculate the numerical difference between the measured values ​​and the predicted values ​​in the trend prediction information as the real-time value gap. Extract several measured values ​​of risk factors from the trend measurement information, calculate the numerical difference between several measured values ​​and the corresponding predicted values, and obtain several historical value differences; The mean squared difference of several historical values ​​is used as the trend gap.

6. The real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data according to claim 4, characterized in that, The weighting coefficients for real-time value differences and trend differences are dynamically adjusted, including: Extract the time range corresponding to the risk factors from the actual trend measurement information; Determine if the time range exceeds the preset time range threshold; if yes, do not adjust the weighting coefficients of the real-time value difference and the trend difference; otherwise, fix the weighting coefficient of the real-time value difference to 1.

7. The real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data according to claim 1, characterized in that, Before updating the digital twin model based on the verification results, determine whether to update the digital twin model based on the verification results, including: Based on the real-time value gap and trend gap, a change prediction model is used to predict the probability of change of risk factors; the change prediction model is built on an artificial intelligence model. Determine if the probability of change is greater than a preset probability threshold; if yes, update the digital twin model; otherwise, calculate the time difference between the real-time value acquisition time and the last digital twin model update time. If the time difference is less than the preset time difference threshold, the digital twin model will not be updated; otherwise, the digital twin model will be updated.

8. The real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data according to claim 1, characterized in that, The updated digital twin model is used to predict trend forecasting information for transmission lines, including: Retrieve the updated digital twin model; the update of the digital twin model is based on the updated line status data of the transmission line. By combining updated meteorological forecast data and digital twin models, the trend forecast information of transmission lines is re-predicted.

9. The real-time monitoring and risk visualization early warning system for transmission lines based on multi-source data according to claim 1, characterized in that, Risk information is extracted based on trend forecasting information, including: Based on trend forecasting information, determine whether risk factors affect the safe operation of transmission lines; if so, extract the degree and duration of the impact of the risk factors. The impact level and duration of each risk factor are integrated into risk information.

Images