A data fusion-based converter station slope safety risk dynamic early warning method

Abstract

The application provides a data fusion-based dynamic pre-warning method for slope safety risk of a converter station, comprising the following steps: uniformly setting a number of N detection points in the slope area around the converter station, collecting structure aging interference parameters at each detection point; inputting the structure aging interference parameters into a weighted summation function for calculation to obtain a structure aging index; setting a structure aging threshold, calculating the structure aging mean value of each detection point; comparing the structure aging index and the structure aging mean value of each detection point with the structure aging threshold respectively. The application can capture local sudden cracks or single-point abnormalities such as corrosion, and can also find the overall aging trend of a large area, thereby distinguishing three different risk levels, avoiding false positives caused by accidental fluctuations or missing reports caused by regional degradation being ignored. The range of the area where problems have occurred is narrowed to more sensitively capture abnormalities, and the range of the stable area is expanded to smooth environmental interference.

Description

Technical Field

[0001] This invention relates to the field of risk analysis and early warning technology, specifically to a dynamic early warning method for safety risks of converter station slopes based on data fusion. Background Technology

[0002] In the dynamic early warning method for slope safety risks in converter stations, accurately determining whether there are risks to the slope structure is crucial. This is because converter stations are usually located in areas with complex geological conditions. Once any part of the slope structure experiences aging, crack expansion, or protective layer failure, it is highly susceptible to serious geological disasters such as landslides and collapses induced by environmental factors such as rainfall, sunshine, or earthquakes. This can not only directly damage the slope itself but also potentially destroy critical facilities within the converter station, such as transmission tower foundations, cable trenches, and grounding grids, leading to power transmission interruptions, large-scale equipment damage, and even casualties. Furthermore, as a continuous but non-uniform structure, the slope often exhibits significant differences in geological conditions, construction quality, stress state, and environmental exposure in different sections. By independently calculating the structural aging index at each monitoring point and comparing it with threshold values, local weak points or early damage can be accurately located, preventing the masking of anomalies in critical areas due to overall averaging.

[0003] In the prior art, CN116433008A discloses a database-based method and system for risk early warning of highway slopes. This technology includes: generating a slope information database based on a pre-established slope information archive; the slope information archive includes basic information about the highway slope; the basic information about the highway slope includes: basic geological conditions, protection conditions, and the attributes of the highway's disaster-bearing body; determining the risk level of the highway slope under natural conditions based on the slope information database, and identifying the target highway slope to be monitored and warned; determining the early warning level of the target highway slope based on the risk level and corresponding natural factor parameters; natural factors include natural environmental factors affecting the stability of the highway slope. This solution can alleviate the technical problems of limited applicability and low early warning accuracy in the prior art, achieving the effect of improving the accuracy of early warning for geological disasters on highway slopes.

[0004] However, in the aforementioned existing technologies, the early warning scheme only focuses on the instantaneous value of a single detection point, ignoring the overall aging trend of the area composed of multiple surrounding points. This makes it impossible to distinguish between local accidental anomalies and large-scale systematic degradation, making it difficult to set up multi-level risk warnings. At the same time, it cannot be dynamically adjusted according to the historical risk level of the slope or real-time feedback. It may be smoothed out in small-scale anomalies, or it may fail to reflect the overall changes in large-scale trends due to the small window. Moreover, the processing of boundary detection points is often directly truncated or missing, affecting the completeness of the mean calculation. Ultimately, this results in low accuracy, timeliness, and reliability of the early warning.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a dynamic early warning method for safety risks of converter station slopes based on data fusion, in order to solve the problems mentioned in the background art. This invention can capture both localized sudden anomalies such as cracks or corrosion, and large-area overall aging trends, thereby distinguishing three different risk levels and avoiding false alarms due to accidental fluctuations or missed alarms due to overlooked regional degradation. The scope of areas where problems have occurred is narrowed to more sensitively capture anomalies, while the scope of stable areas is expanded to smooth out environmental disturbances.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A dynamic early warning method for slope safety risks in converter stations based on data fusion includes the following steps:

[0009] S1: N test points are uniformly set in the slope area around the converter station. Each test point is distributed at the same horizontal height and the distance between adjacent test points is S. The structural aging interference parameters at each test point are collected. The structural aging interference parameters include carbonization depth, protective layer thickness, crack width value and annual sunshine hours. A converter station slope construction database is established to obtain the design value of crack width.

[0010] S2: Preprocess the collected structural aging interference parameters. The preprocessing includes outlier removal, missing value interpolation and normalization transformation. Construct a weighted summation function for slope structural aging. Input the preprocessed structural aging interference parameters into the weighted summation function for calculation to obtain the structural aging index. The structural aging index is used to characterize the local aging degree of each detection point in the slope structure.

[0011] S3: Set a structural aging threshold, obtain the structural aging index calculated independently for each detection point, and calculate the average structural aging value between each detection point and the M adjacent detection points on the left and right. The average structural aging value is used to characterize the overall aging degree around each detection point in the slope structure.

[0012] S4: Compare the structural aging index and the average structural aging value of each detection point with the structural aging threshold. When the structural aging index is greater than or equal to the structural aging threshold, or when the average structural aging value is greater than or equal to the structural aging threshold, the current detection point is determined to be in an abnormal state and an early warning is triggered. When both the structural aging index and the average structural aging value are less than the structural aging threshold, the current detection point is determined to be in a normal state and no early warning is triggered.

[0013] Furthermore, in S1, the process of uniformly setting N detection points in the slope area around the converter station is as follows: based on the perimeter of the slope area, the distance S between adjacent detection points is calculated by dividing the perimeter by N through mathematical modeling; all detection points are set at the same horizontal height, and a benchmark elevation reference point is introduced to unify the height standard and eliminate the influence of vertical differences on parameter acquisition.

[0014] Structural aging interference parameters are collected at each detection point. Each structural aging interference parameter is obtained through timed and fixed-point observation. A converter station slope construction database is established, integrating historical construction records, design specifications, and real-time collected data. The design value of crack width is extracted through data fusion technology.

[0015] Furthermore, during the outlier removal process, any collected structural aging interference parameter is considered an outlier if it meets the following conditions:

[0016]

[0017] in:

[0018] x represents the value of a single data point, referring to a measured value in the structural aging disturbance parameter;

[0019] μ is the average value of the parameter dataset in which the data point is located, and the parameter dataset includes carbonization depth dataset, protective layer thickness dataset, crack width value dataset, and annual sunshine hours dataset.

[0020] k is a constant threshold coefficient used to define the boundary range of outliers;

[0021] σ is the sample standard deviation of the parameter dataset in which the data point is located, used to quantify the dispersion of the data.

[0022] Furthermore, the formula for obtaining the sample mean μ is:

[0023]

[0024] Where n is the total number of data points used to provide the sample. This represents the value of a single data point with index i.

[0025] w represents the parameter type. When w=1, it represents the carbonization depth; when w=2, it represents the protective layer thickness; when w=3, it represents the crack width; and when w=4, it represents the annual sunshine hours.

[0026] Furthermore, the weighted summation function used to obtain the structural aging index is as follows:

[0027]

[0028] in:

[0029] SAI stands for Structural Aging Index;

[0030] Depth of carbonization;

[0031] The thickness of the protective layer;

[0032] This represents the crack width value.

[0033] This is the design value for the crack width;

[0034] Annual sunshine hours;

[0035] These are the weighting coefficients for the ratio of carbonization depth to protective layer thickness, the deviation of crack width from the design crack width, and annual sunshine hours. These weighting coefficients are used to adjust the relative contribution of each parameter. Each weighting coefficient is calibrated based on historical data or expert experience and meets the following requirements: ;

[0036] This indicates the deviation between the actual crack width and the design value; it only applies when the actual value exceeds the design value. Only positive values ​​are counted; otherwise, they are 0, ensuring that the aging index is non-negative.

[0037] Furthermore, the carbonization depth, protective layer thickness, crack width, and annual sunshine hours are used to support the dynamic early warning model for slope safety risks through a quantitative aging mechanism.

[0038] Carbonation depth is used to characterize the degree of surface deterioration of concrete caused by environmental carbon dioxide erosion. The larger the depth value, the higher the risk of failure of the steel reinforcement cover and the faster the structural corrosion.

[0039] The thickness of the protective layer is used to characterize the integrity of the physical barrier of concrete covering the reinforcing steel. Insufficient thickness will directly reduce the resistance to penetration and corrosion, leading to accelerated oxidation of the internal reinforcing steel.

[0040] Crack width is a physical scale used to characterize surface damage of a structure. Increased width indicates stress concentration or increased material fatigue, which can lead to water seepage, freeze-thaw damage, or a decrease in load-bearing capacity.

[0041] Annual sunshine hours are used to characterize the cumulative effects of environmental exposure. High sunshine hours promote ultraviolet aging, thermal expansion and contraction cycles, and material weathering, accelerating the overall structural deterioration process.

[0042] Furthermore, the structural aging threshold is set as follows: The formula used to calculate the average structural aging value between each detection point and its M adjacent detection points is as follows:

[0043]

[0044] in:

[0045] Let be the average structural aging value at the i-th detection point, representing the overall aging degree of the surrounding 2M+1 detection points. It reflects the cumulative risk in a region and is used to identify systemic aging.

[0046] M is the range of the number of adjacent detection points, and M is a positive integer;

[0047] Let j be the structural aging index of the j-th detection point. It reflects immediate risks at a single point and is used to capture local anomalies;

[0048] 2M+1 represents the total number of detection points involved in the calculation;

[0049] To the arrive The structural aging index of all test points is summed.

[0050] Furthermore, it also includes the process for determining the value of the range M of the number of adjacent detection points:

[0051] The initial value of M is selected based on the perimeter L of the slope area and the total number of monitoring points N. M is then dynamically adjusted based on the historical risk level of the slope, reducing the number of monitoring points that have experienced displacement or cracking exceeding the standard. to To enhance the ability to detect local anomalies and expand the low-risk area. to To smooth out environmental noise; through real-time early warning feedback closed-loop optimization, and using historical data to train regression models for iterative calibration to the optimal solution; and to use a mirror filling strategy for boundary detection points to ensure the integrity of mean calculation.

[0052] Furthermore, in step S4, the structural aging index calculated independently at the i-th detection point is labeled as... The logic for comparing the structural aging index and the average structural aging value at each detection point with the structural aging threshold is as follows:

[0053] When the structural aging index exceeds the standard: However, when the average structural aging value is normal: The location is determined to be an anomaly at a local point. At this point, the risk is low, the area is generally stable, and a yellow alert is triggered.

[0054] When the average structural aging exceeds the standard: However, when the structural aging index is normal: The region was identified as having an abnormal trend, reflecting a moderate systemic risk, thus triggering an orange alert.

[0055] Furthermore, the logic of comparing the structural aging index and the average structural aging value at each detection point with the structural aging threshold also includes:

[0056] When the average structural aging exceeds the standard: When the structural aging index also exceeds the standard: The detection point and the detection area are both at risk, indicating a high probability of slope instability, which triggers a red alert.

[0057] When the average structural aging value is normal: When the structural aging index is also normal: If the condition is determined to be normal, with no immediate risk and no warning signal triggered, continuous long-term monitoring will be conducted.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] This invention no longer relies solely on data from a single indicator or a single detection point. Instead, it calculates the structural aging index for each point based on multiple parameters and through weighted fusion. Furthermore, it considers the average aging level of multiple adjacent points around each point. This approach can capture both localized anomalies such as sudden cracks or corrosion and the overall trend of large-area aging, thereby distinguishing three different risk levels and avoiding false alarms due to accidental fluctuations or missed alarms due to neglected regional degradation. Simultaneously, this invention dynamically adjusts the range of adjacent points involved in the calculation based on the historical risk situation of the slope. It narrows the range for areas that have previously experienced problems to more sensitively capture anomalies, expands the range for stable areas to smooth out environmental disturbances, and continuously optimizes parameters through real-time monitoring data feedback, ultimately making the early warning more accurate, timely, and reliable. Attached Figure Description

[0060] Figure 1 This is a flowchart of a dynamic early warning method for slope safety risks of converter stations based on data fusion, according to the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0062] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0063] Example:

[0064] Please see Figure 1 The present invention provides the following technical solutions:

[0065] A dynamic early warning method for safety risks of converter station slopes based on data fusion is proposed. This method first uniformly marks multiple detection points around the converter station slope, and then calculates the structural aging status of each detection point using structural aging interference parameters. The main steps include:

[0066] S1: For the slope area surrounding the converter station, N monitoring points need to be uniformly set. These monitoring points are all distributed at the same horizontal level to ensure the consistency of the measurement benchmark. The spacing between adjacent monitoring points is uniformly set to S, which is usually calculated based on the actual perimeter of the slope area and the total number of monitoring points. At each monitoring point, the system collects various structural aging interference parameters, including carbonization depth, protective layer thickness, crack width, and annual sunshine hours. These parameters can comprehensively reflect the degradation status of the slope structure under different environmental and load conditions. Based on this, a converter station slope construction database is established, integrating historical construction records, design specifications, and data collected on-site to extract and determine the design value of crack width.

[0067] The specific process for uniformly setting N detection points in the slope area around the converter station is as follows: First, based on the actual perimeter of the slope area, the perimeter is precisely divided by the total number of detection points N using mathematical modeling methods. This is used to calculate and determine the reasonable spacing S between adjacent detection points, ensuring that the detection points are evenly distributed across the entire slope perimeter.

[0068] All detection points were set at the same horizontal level. Therefore, a fixed benchmark elevation was introduced as a unified height standard to effectively eliminate the adverse effects of vertical differences caused by terrain undulations on subsequent parameter acquisition. At each detection point, the system collected multiple structural aging interference parameters. Each parameter was acquired through timed and fixed-point observations, with the observation time interval and location kept fixed to ensure data consistency and comparability.

[0069] A database for converter station slope construction was established, comprehensively integrating historical construction records, relevant design specifications, and real-time on-site data. Data fusion technology was then used to collaboratively process multi-source information, accurately extracting the design value for crack width. Based on this, and leveraging industrial big data technology, a sensor network was used to continuously collect and transmit structural aging interference parameters, constructing an industrial data acquisition system for slope safety monitoring. This provides a high-quality, high-frequency fundamental data source for subsequent analysis.

[0070] S2: The collected structural aging interference parameters are systematically preprocessed. This preprocessing includes outlier removal, missing value imputation, and normalization transformation. This process fully utilizes industrial data processing and big data processing methods to clean and standardize the massive monitoring data, ensuring data quality meets the reliability requirements for subsequent aging index calculations. Based on this, a weighted summation function is constructed to describe the degree of slope structural aging. The preprocessed structural aging interference parameters are used as input variables in this weighted summation function for calculation. The structural aging index is obtained from the weighted summation result. This structural aging index can effectively characterize the local aging degree corresponding to each detection point in the slope structure.

[0071] During the outlier removal process, any collected structural aging interference parameter that meets the following conditions is considered an outlier:

[0072]

[0073] in:

[0074] x represents the value of a single data point, referring to a measured value in the structural aging disturbance parameter;

[0075] μ is the sample mean of the parameter dataset in which the data point is located. The parameter dataset includes carbonization depth dataset, protective layer thickness dataset, crack width value dataset, and annual sunshine duration dataset. μ is related to each detection point. The difference between them reflects the degree of deviation of that point from the overall average state; the greater the deviation, the higher the probability of an anomaly. There is no direct causal relationship with the risk of structural aging;

[0076] k is a constant threshold coefficient used to define the boundary range of outliers. The smaller, the boundary The easier it is to satisfy the criteria, the more outliers will be detected, resulting in high sensitivity but potentially introducing false alarms. The larger the value, the looser the boundary. Only points that deviate extremely much will be judged as abnormal. It has high specificity but may miss early abnormalities.

[0077] σ represents the sample standard deviation of the parameter dataset in which the data point is located. It is used to quantify the dispersion of the data. The larger the value, the greater the difference in parameter values ​​between the detection points, and the stronger the data volatility. At this point, the anomaly detection boundary... The width will increase accordingly, and only points with extreme deviations will be marked as anomalies; conversely, The smaller the value, the more concentrated the data becomes, the narrower the boundaries, and even small deviations may be judged as anomalies. It is related to the overall uniformity of the slope.

[0078] The formula for obtaining the sample mean μ is:

[0079]

[0080] Where n is the total number of data points. This represents the value of a single data point with index i.

[0081] w represents the parameter type. When w=1, it represents the carbonization depth; when w=2, it represents the protective layer thickness; when w=3, it represents the crack width; and when w=4, it represents the annual sunshine hours.

[0082] The weighted summation function used to obtain the structural aging index is as follows:

[0083]

[0084] in:

[0085] SAI stands for Structural Aging Index;

[0086] Depth of carbonization; For protective layer thickness; ratio This ratio comprehensively reflects the relative depth of carbonation erosion. When the carbonation depth is close to or exceeds the thickness of the protective layer, the ratio is close to or greater than 1, indicating that the reinforcement protection has seriously failed.

[0087] This represents the crack width value. Design value for crack width; Deviation item Only the portion exceeding the design limit is retained; this item is zero when the limit is not exceeded. This reflects the engineering logic that risk is only considered when the value exceeds the design allowable value.

[0088] Annual sunshine hours; It is positively correlated with SAI. The longer the sunshine duration, the faster the material deteriorates and the higher the structural aging index. However, compared with carbonization and cracking, sunshine is an environmental background factor and its effect is relatively slow.

[0089] These are the weighting coefficients for the ratio of carbonization depth to protective layer thickness, the deviation of crack width from the design crack width, and annual sunshine hours. The weighting coefficients are used to adjust the relative contribution of each parameter, and each weighting coefficient is calibrated based on historical data or expert experience.

[0090] Meanwhile, crack width deviation is given the highest weight because cracks directly reflect the stress and fatigue state of the structure and are an important precursor to slope instability. The weighting coefficient... This is used to adjust the contribution of this factor to the total SAI. Since carbonization is a long-term, gradual process, Set to above average, but less than Because exceeding the crack width limit poses a greater risk of sudden occurrence; while the weighting coefficient : Adjusting for the contribution of environmental exposure factors. Since the effects of solar radiation are long-term cumulative and regionally consistent, The weights are set to the minimum of the three to avoid environmental noise masking local structural abrupt changes. Therefore, the weights satisfy the following conditions: ;

[0091] This indicates the deviation between the actual crack width and the design value; it only applies when the actual value exceeds the design value. Only positive values ​​are counted; otherwise, they are 0, ensuring that the aging index is non-negative.

[0092] Carbonation depth is primarily used to characterize the degree of gradual deterioration of the surface layer of a concrete structure under long-term exposure to carbon dioxide erosion from the environment. A higher carbonation depth value indicates a deeper degree of concrete neutralization, which significantly increases the risk of failure of the reinforcing steel cover, further accelerating the corrosion process of the internal reinforcing steel.

[0093] The thickness of the protective layer characterizes the integrity and effectiveness of the physical barrier formed by the concrete covering the steel reinforcement. When the protective layer is insufficient, it directly weakens the concrete's impermeability and corrosion resistance, making it easier for external moisture and harmful substances to come into contact with the steel reinforcement, thereby exacerbating the oxidation reaction of the internal steel reinforcement.

[0094] Crack width is used to characterize the physical scale of damage to a structural surface. An increase in crack width usually indicates stress concentration or accelerated material fatigue in a localized area of ​​the structure. This type of damage can easily lead to a series of secondary problems such as water seepage, freeze-thaw damage, and reduced load-bearing capacity.

[0095] Annual sunshine hours are used to characterize the cumulative effects of environmental exposure conditions on slope structures over time. Higher annual sunshine hours indicate that the structure is subjected to ultraviolet radiation and thermal expansion and contraction cycles more frequently, which accelerates the weathering rate of materials and thus speeds up the overall structural deterioration process.

[0096] S3: Set a structural aging threshold as a benchmark for judging the safety status of the slope. For each detection point, obtain its independently calculated structural aging index, which reflects the local aging status of that detection point. Simultaneously, calculate the average structural aging value between each detection point and its M adjacent detection points. This average value, by comprehensively considering the aging levels of multiple surrounding detection points, characterizes the overall aging degree of the area surrounding each detection point in the slope structure. By combining the single-point aging index with the regional aging average, a more comprehensive basis is provided for subsequent early warning judgments.

[0097] Set the structural aging threshold as The formula used to calculate the average structural aging value between each detection point and its M adjacent detection points is as follows:

[0098] .

[0099] in:

[0100] Let be the average structural aging value at the i-th detection point, representing the overall aging degree of the surrounding 2M+1 detection points. It reflects the cumulative risk in a region and is used to identify systemic aging. It revolves around the first The average aging level of a local area of ​​a point, therefore This determines the spatial center location for mean calculation. Different corresponding The results may vary due to differences in the distribution of localized aging, but the formula itself is accurate. There is no direct positive or negative correlation; instead, spatial heterogeneity is reflected through neighborhood selection.

[0101] M represents the range of the number of adjacent detection points, and M is a positive integer; it indicates the number of points from the first... Each detection point is taken to the left and right. Each adjacent point is used in the mean calculation. The size determines the width of the space smoothing window.

[0102] Let j be the structural aging index of the j-th detection point. It reflects immediate risks at a single point and is used to capture local anomalies;

[0103] 2M+1 represents the total number of detection points involved in the calculation; the denominator is... and A negative correlation is observed (when the molecule remains unchanged). When the molecule (all molecules within its neighborhood)... When the sum of the terms is fixed, the larger the denominator, the smaller the mean.

[0104] To the arrive The structural aging index of all test points is summed. The summation result, with the denominator fixed, is then compared with... They exhibit a strict positive correlation. The summation range covers the values ​​starting from... Centered on the continuity The number of points ensures spatial continuity. When the aging index is generally high in the neighborhood, the summation result is large, and the mean is also large.

[0105] It also includes the process for determining the value of the range M of the number of adjacent detection points:

[0106] When determining the range M of adjacent monitoring points, a suitable initial value of M is first selected based on the actual perimeter L of the slope area and the total number N of monitoring points. Subsequently, the value of M is dynamically adjusted based on the historical risk level of the slope: for monitoring points where displacement or cracking has previously exceeded limits, the value of M should be reduced to a preset minimum. To enhance the ability to detect local anomalies, the M value is increased to a preset maximum value for detection points that have been in low-risk areas for a long time. To smooth out data fluctuations caused by environmental noise, a real-time early warning and feedback closed-loop mechanism is established to continuously optimize the M-value. A regression model is trained using accumulated historical data, and through multiple iterations and calibrations, the M-value gradually approaches the optimal solution. Furthermore, knowledge graphs and expert systems are introduced to construct a semantic network and reasoning model for slope aging mechanisms. This ensures that the dynamic adjustment of the M-value not only relies on data-driven approaches but also integrates domain knowledge and historical risk cases, enhancing the intelligence level of decision-making. In addition, for detection points located at the boundaries of slope areas, a mirror-fill strategy is used to handle boundary effects in mean calculation.

[0107] S4: After calculating the structural aging index and the average structural aging value, the structural aging index and the corresponding average structural aging value for each detection point are compared with the pre-set structural aging threshold. If the structural aging index of a detection point is greater than or equal to the structural aging threshold, or if the average structural aging value of that detection point is greater than or equal to the structural aging threshold, the current detection point is determined to be in an abnormal state, and the corresponding early warning mechanism is immediately triggered. Conversely, only when both the structural aging index and the average structural aging value of the detection point are less than the structural aging threshold can the current detection point be determined to be in a normal state. In this case, no early warning operation is required, and routine monitoring continues.

[0108] The structural aging index calculated independently at the i-th detection point is labeled as... The logic for comparing the structural aging index and the average structural aging value at each detection point with the structural aging threshold is as follows:

[0109] When the structural aging index exceeds the standard: However, when the average structural aging value is normal: If a local anomaly is detected, the risk is low and the overall area is stable, triggering a yellow alert. At this point, more frequent monitoring of the anomaly detection point is implemented, increasing the data collection frequency. On-site inspection personnel are assigned to manually verify the point, focusing on checking for crack expansion or localized deformation, and recording the trend of change. Meanwhile, routine monitoring of other areas remains unchanged.

[0110] When the average structural aging exceeds the standard: However, when the structural aging index is normal: The area was identified as having an abnormal trend, reflecting a moderate systemic risk, triggering an orange alert. Immediately, a comprehensive investigation was conducted on the area containing the monitoring point and its adjacent M monitoring points to the left and right, analyzing whether there were common environmental or structural causes, such as poor drainage or foundation settlement. Simultaneously, technical personnel were organized to conduct data consultations to assess whether local reinforcement or repair measures were necessary. The average aging value of the area was continuously tracked in subsequent monitoring cycles, and the M value was adjusted as needed to enhance risk detection capabilities.

[0111] The logic of comparing the structural aging index and the average structural aging value at each detection point with the structural aging threshold also includes:

[0112] When the average structural aging exceeds the standard: When the structural aging index also exceeds the standard: The situation was determined to be severely abnormal. At this point, both the monitoring point and the monitoring area posed a dual risk, indicating a high probability of slope instability, the highest risk, triggering a red alert. The emergency plan should be activated immediately: all work activities near the slope should be stopped, and relevant personnel should be evacuated to a safe area; an expert team should be quickly organized to conduct on-site investigation and structural safety assessment, and temporary support, unloading, or emergency reinforcement measures should be taken if necessary; simultaneously, the design, construction, and operation and maintenance units should be notified for joint handling, and continuous monitoring should be maintained until the risk drops below the safety threshold.

[0113] When the average structural aging value is normal: When the structural aging index is also normal: If the slope is deemed to be in a normal state with no immediate risk and no warning signal is triggered, continuous long-term monitoring will be conducted, with data collected and the database updated regularly to track the aging trend of the structure. This early warning process integrates pre-diagnosis and health management methods into the dynamic early warning of slope safety risks, achieving full lifecycle health management from data collection, status identification, risk classification to treatment recommendations, effectively supporting proactive operation and maintenance of converter station slopes.

[0114] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0115] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0116] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0117] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A dynamic early warning method for slope safety risks in converter stations based on data fusion, characterized in that, Includes the following steps: S1: N test points are uniformly set in the slope area around the converter station. Each test point is distributed at the same horizontal height and the distance between adjacent test points is S. The structural aging interference parameters at each test point are collected. The structural aging interference parameters include carbonization depth, protective layer thickness, crack width value and annual sunshine hours. A converter station slope construction database is established to obtain the design value of crack width. S2: Preprocess the collected structural aging interference parameters. The preprocessing includes outlier removal, missing value interpolation and normalization transformation. Construct a weighted summation function for slope structural aging. Input the preprocessed structural aging interference parameters into the weighted summation function for calculation to obtain the structural aging index. The structural aging index is used to characterize the local aging degree of each detection point in the slope structure. S3: Set a structural aging threshold, obtain the structural aging index calculated independently for each detection point, and calculate the average structural aging value between each detection point and the M adjacent detection points on the left and right. The average structural aging value is used to characterize the overall aging degree around each detection point in the slope structure. S4: Compare the structural aging index and the average structural aging value of each detection point with the structural aging threshold. When the structural aging index is greater than or equal to the structural aging threshold, or the average structural aging value is greater than or equal to the structural aging threshold, determine that the current detection point is in an abnormal state and trigger an alarm. When both the structural aging index and the average structural aging value are less than the structural aging threshold, the current detection point is determined to be in a normal state, and no warning is issued.

2. The method for dynamic early warning of safety risks of converter station slopes based on data fusion according to claim 1, characterized in that: In S1, the process of uniformly setting N detection points in the slope area around the converter station is as follows: based on the perimeter of the slope area, the distance S between adjacent detection points is calculated by dividing the perimeter by N through mathematical modeling. All detection points are set at the same horizontal level. A benchmark elevation reference point is introduced to unify the height standard and eliminate the influence of vertical differences on parameter acquisition. Structural aging interference parameters are collected at each detection point. Each structural aging interference parameter is obtained through timed and fixed-point observation. A converter station slope construction database is established, integrating historical construction records, design specifications, and real-time collected data. The design value of crack width is extracted through data fusion technology.

3. The method for dynamic early warning of safety risks of converter station slopes based on data fusion according to claim 2, characterized in that: During the outlier removal process, any collected structural aging interference parameter is considered an outlier if it meets the following conditions: ； in: x represents the value of a single data point, referring to a measured value in the structural aging disturbance parameter; μ is the average value of the parameter dataset in which the data point is located, and the parameter dataset includes carbonization depth dataset, protective layer thickness dataset, crack width value dataset, and annual sunshine hours dataset. k is a constant threshold coefficient used to define the boundary range of outliers; σ is the sample standard deviation of the parameter dataset in which the data point is located, used to quantify the dispersion of the data.

4. The method for dynamic early warning of safety risks of converter station slopes based on data fusion according to claim 3, characterized in that: The formula for obtaining the sample mean μ is: ； Where n is the total number of data points used to provide the sample. This represents the value of a single data point with index i. w represents the parameter type. When w=1, it represents the carbonization depth; when w=2, it represents the protective layer thickness; when w=3, it represents the crack width value. When w=4, it represents the annual sunshine hours.

5. The method for dynamic early warning of safety risks of converter station slopes based on data fusion according to claim 1, characterized in that: The weighted summation function used to obtain the structural aging index is: ； in: SAI stands for Structural Aging Index; Depth of carbonization; The thickness of the protective layer; This represents the crack width value. This is the design value for the crack width; Annual sunshine hours; These are the weighting coefficients for the ratio of carbonization depth to protective layer thickness, the deviation of crack width from the design crack width, and annual sunshine hours. These weighting coefficients are used to adjust the relative contribution of each parameter. Each weighting coefficient is calibrated based on historical data or expert experience and meets the following requirements: ; This indicates the deviation between the actual crack width and the design value; it only applies when the actual value exceeds the design value. Only positive values ​​are counted; otherwise, they are 0, ensuring that the aging index is non-negative.

6. The method for dynamic early warning of safety risks of converter station slopes based on data fusion according to claim 5, characterized in that: The carbonization depth, protective layer thickness, crack width, and annual sunshine hours are used to support the dynamic early warning model for slope safety risks through a quantitative aging mechanism. Carbonation depth is used to characterize the degree of surface deterioration of concrete caused by environmental carbon dioxide erosion. The larger the depth value, the higher the risk of failure of the steel reinforcement cover and the faster the structural corrosion. The thickness of the protective layer is used to characterize the integrity of the physical barrier of concrete covering the reinforcing steel. Insufficient thickness will directly reduce the resistance to penetration and corrosion, leading to accelerated oxidation of the internal reinforcing steel. Crack width is a physical scale used to characterize surface damage of a structure. Increased width indicates stress concentration or increased material fatigue, which can lead to water seepage, freeze-thaw damage, or a decrease in load-bearing capacity. Annual sunshine hours are used to characterize the cumulative effects of environmental exposure. High sunshine hours promote ultraviolet aging, thermal expansion and contraction cycles, and material weathering, accelerating the overall structural deterioration process.

7. The method for dynamic early warning of safety risks of converter station slopes based on data fusion according to claim 1, characterized in that: Set the structural aging threshold as The formula used to calculate the average structural aging value between each detection point and its M adjacent detection points is as follows: ； in: Let be the average structural aging value at the i-th detection point, representing the overall aging degree of the surrounding 2M+1 detection points. It reflects the cumulative risk in a region and is used to identify systemic aging. M is the range of the number of adjacent detection points, and M is a positive integer; Let j be the structural aging index of the j-th detection point. It reflects immediate risks at a single point and is used to capture local anomalies; 2M+1 represents the total number of detection points involved in the calculation; To the arrive The structural aging index of all test points is summed.

8. A dynamic early warning method for slope safety risks of converter stations based on data fusion according to claim 7, characterized in that, It also includes the process for determining the value of the range M of the number of adjacent detection points: The initial value of M is selected based on the perimeter L of the slope area and the total number of monitoring points N. M is then dynamically adjusted based on the historical risk level of the slope, reducing the number of monitoring points that have experienced displacement or cracking exceeding the standard. to To enhance the ability to detect local anomalies and expand the low-risk area. to To smooth out environmental noise; through real-time early warning feedback closed-loop optimization, and using historical data to train regression models for iterative calibration to the optimal solution; and to use a mirror filling strategy for boundary detection points to ensure the integrity of mean calculation.

9. A dynamic early warning method for slope safety risks of converter stations based on data fusion as described in claim 7, characterized in that: In step S4, the structural aging index calculated independently at the i-th detection point is labeled as... The logic for comparing the structural aging index and the average structural aging value at each detection point with the structural aging threshold is as follows: When the structural aging index exceeds the standard: However, when the average structural aging value is normal: The location is determined to be an anomaly at a local point. At this point, the risk is low, the area is generally stable, and a yellow alert is triggered. When the average structural aging exceeds the standard: However, when the structural aging index is normal: The region was identified as having an abnormal trend, reflecting a moderate systemic risk, thus triggering an orange alert.

10. A dynamic early warning method for safety risks of converter station slopes based on data fusion according to claim 9, characterized in that: The logic of comparing the structural aging index and the average structural aging value at each detection point with the structural aging threshold also includes: When the average structural aging exceeds the standard: When the structural aging index also exceeds the standard: The detection point and the detection area are both at risk, indicating a high probability of slope instability, which triggers a red alert. When the average structural aging value is normal: When the structural aging index is also normal: If the condition is determined to be normal, with no immediate risk and no warning signal triggered, continuous long-term monitoring will be conducted.

Images