Early warning method and device for instability situation of tailing dam

Abstract

The invention relates to a tailing dam instability situation early warning method and device. The method comprises the following steps: determining a plurality of characteristic parameters of the tailing dam based on monitoring data of the tailing dam; inputting the plurality of characteristic parameters into a trained long short-term memory (LSTM) network model to obtain respective weight values of the plurality of characteristic parameters output by the LSTM model; based on the current environment variable of the tailing dam, utilizing a Bayesian probability model to determine an environment correction factor; based on the multiple characteristic parameters, the respective weight values of the multiple characteristic parameters and the environment correction factors, determining an instability risk comprehensive score of the tailing dam; and outputting early warning information based on the instability risk comprehensive score. According to the scheme, the accuracy of tailing dam instability early warning is improved.

Description

Technical Field

[0001] This disclosure relates to the field of tailings dam safety monitoring technology, and in particular to a method and device for early warning of tailings dam instability. Background Technology

[0002] In related technologies, tailings dam stability is a crucial issue in mine safety production, especially in non-coal mines where the stability of tailings dam slopes can be affected by internal and external factors at any time. Therefore, monitoring the safety and stability of tailings dams is becoming increasingly important to ensure production safety in non-coal mines.

[0003] Currently, traditional manual monitoring methods are commonly used to monitor the safety status of mine tailings dams. This method suffers from low monitoring frequency, limited monitoring indicators, poor early warning timeliness, and high labor costs. Relying on manual monitoring makes it difficult to identify safety hazards and provide early warnings of disasters at tailings dams. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this disclosure provides a method and device for early warning of tailings dam instability.

[0005] According to a first aspect of the present disclosure, a method for early warning of tailings dam instability is provided, comprising: Acquire monitoring data of the tailings dam, and determine multiple characteristic parameters of the tailings dam based on the monitoring data; the multiple characteristic parameters are used to reflect the stability state of the tailings dam. Multiple feature parameters are input into the trained Long Short-Term Memory (LSTM) network model to obtain the weight values ​​of each feature parameter output by the LSTM model. Obtain the current environmental variables of the tailings dam, and determine the environmental correction factor based on the current environmental variables using a Bayesian probability model; Based on multiple characteristic parameters, their respective weights, and the environmental correction factor, a comprehensive score for the instability risk of the tailings dam is determined. Early warning information is output based on the comprehensive score of instability risk.

[0006] According to a second aspect of the present disclosure, a tailings dam instability early warning device is provided, comprising: The first determining unit is used to acquire monitoring data of the tailings dam and determine multiple characteristic parameters of the tailings dam based on the monitoring data; the multiple characteristic parameters are used to reflect the stability state of the tailings dam. The second determining unit is used to input multiple feature parameters into the trained Long Short-Term Memory (LSTM) network model to obtain the weight values ​​of each of the multiple feature parameters output by the LSTM model. The third determining unit is used to obtain the current environmental variables of the tailings dam and, based on the current environmental variables, determine the environmental correction factor using a Bayesian probability model. The fourth determining unit is used to determine the comprehensive score of the instability risk of the tailings dam based on multiple feature parameters, the weight values ​​of each feature parameter, and the environmental correction factor. The early warning unit is used to output early warning information based on the comprehensive score of instability risk.

[0007] According to a third aspect of the present disclosure, an electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as described in any one of the first aspects.

[0008] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of the first aspects.

[0009] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method as described in any one of the first aspects.

[0010] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: determining multiple characteristic parameters of a tailings dam based on monitoring data; inputting the multiple characteristic parameters into a trained Long Short-Term Memory (LSTM) network model to obtain the weight values ​​of each characteristic parameter output by the LSTM model; determining an environmental correction factor based on the current environmental variables of the tailings dam using a Bayesian probability model; determining a comprehensive instability risk score of the tailings dam based on the multiple characteristic parameters, their respective weight values, and the environmental correction factor; and outputting early warning information based on the comprehensive instability risk score. By dynamically allocating the weight values ​​of the characteristic parameters and combining them with the environmental correction factor, adaptive fusion of multi-source monitoring data is achieved, thereby significantly improving the accuracy of tailings dam instability early warning and reducing the risk of false alarms and missed alarms.

[0011] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0012] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0013] Figure 1 This is a flowchart illustrating an early warning method for tailings dam instability according to an exemplary embodiment.

[0014] Figure 2 This is a block diagram illustrating an early warning device for tailings dam instability according to an exemplary embodiment.

[0015] Figure 3 This is a block diagram illustrating an apparatus for an early warning method for tailings dam instability, according to an exemplary embodiment.

[0016] Figure Labels 201-First determining unit; 202-Second determining unit; 203-Third determining unit; 204-Fourth determining unit; 205-Early warning unit; 300-Device; 302-Processing component; 304-Memory; 306-Power component; 308-Multimedia component; 310-Audio component; 312-I / O interface; 316-Communication component; 320-Processor. Detailed Implementation

[0017] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0018] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the present disclosure. The singular forms “a” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0019] It should be understood that although the terms first, second, third, etc., may be used to describe various information in embodiments of this disclosure, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of embodiments of this disclosure, and similarly, second information may also be referred to as first information. Depending on the context, the words “if” and “suppose” as used herein may be interpreted as “when”, “when”, or “in response to a determination”.

[0020] Furthermore, various forms of processes shown in the embodiments of this disclosure can be used to reorder, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and no limitation is imposed herein.

[0021] It should be noted that the collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0022] Figure 1 This is a flowchart illustrating an early warning method for tailings dam instability according to an exemplary embodiment, such as... Figure 1 As shown, it should be noted that the tailings dam instability early warning method of this disclosure embodiment is applied to the tailings dam instability early warning device. For example... Figure 1 As shown, the method may include the following steps: Step 101: Obtain monitoring data of the tailings dam and determine multiple characteristic parameters of the tailings dam based on the monitoring data.

[0023] Among them, several characteristic parameters are used to reflect the stability of the tailings dam.

[0024] In one embodiment, multiple types of raw monitoring data can be collected in real time through a sensor network (including triaxial accelerometers, tilt sensors, electronic compasses, vibration sensors, etc.) deployed at key locations of the tailings dam. The raw data is then preprocessed (including outlier removal, noise filtering, and spatiotemporal alignment), and the monitoring data is converted into characteristic parameters that can directly characterize the stability of the dam body through a preset physical model and algorithm.

[0025] For example, the surface displacement can be obtained by performing a second integral and drift correction on the filtered acceleration signal, the tilt angle and azimuth angle can be obtained by directly reading the calibrated sensor signal, and the dominant frequency and amplitude can be extracted by performing spectral analysis on the vibration waveform. These characteristic parameters together constitute a set of indicators reflecting the dynamic behavior and safety status of the dam structure, providing standardized input for subsequent integrated evaluation.

[0026] In one embodiment, the characteristic parameters may include the surface displacement, tilt angle, magnetic declination, acceleration, and vibration parameters (e.g., vibration frequency and amplitude) of the tailings dam.

[0027] Step 102: Input multiple feature parameters into the trained LSTM (Long Short-Term Memory) model to obtain the weight values ​​of each feature parameter output by the LSTM model.

[0028] In one embodiment, the trained LSTM model has learned from historical data (including long-term feature parameter sequences and corresponding dam state labels), thereby grasping the deep correlation between the temporal variation patterns of different parameters and the eventual instability risk. Real-time, time-ordered feature parameter sequences (e.g., data collected every 5 minutes over the past 72 hours) can be input into the LSTM model. The LSTM model analyzes the evolution trends of these parameters over a past period, their temporal dependencies, and the strength of their correlation with the current risk state.

[0029] In this embodiment of the disclosure, the output of the LSTM model can be a set of normalized weight values, each corresponding to an input feature parameter. These weight values ​​quantify the reliability and importance of the feature parameter at the current time and in a specific environmental context.

[0030] For example, during periods of continuous rainfall, the LSTM model can automatically assign higher weights to parameters such as "seepage pressure" or "displacement," while in the post-earthquake phase, it may increase the weights of parameters such as "vibration frequency" or "acceleration." This dynamic weight allocation mechanism realizes the transformation from static, experience-based fixed weights to dynamic, data-driven intelligent weights, which is a key technical link in improving the accuracy and targeted nature of subsequent DSI scoring.

[0031] As an example of a possible implementation, the training process of an LSTM neural network is as follows: Historical multi-source monitoring data of tailings dams over a long period (including time series of various feature parameters) and corresponding dam status labels (such as "stable," "early warning," and "instability") calibrated by experts or actual events are collected to form a training sample set. Then, an LSTM network structure is constructed with time-series features as input and weights reflecting the contribution or risk state of each feature parameter as output. The backpropagation algorithm is used to iteratively adjust the internal weights and bias parameters of the network, and the forget gate, input gate, and output gate are trained to enable the network to automatically identify and memorize the complex nonlinear mapping relationship between the long-term evolution patterns of each feature parameter and the final risk state under different operating conditions. Finally, the trained LSTM model has the ability to dynamically infer and output a set of normalized weight values ​​reflecting the current importance of each parameter based on the real-time input feature parameter sequence.

[0032] Step 103: Obtain the current environmental variables of the tailings dam, and determine the environmental correction factor based on the current environmental variables using a Bayesian probability model.

[0033] In this embodiment of the disclosure, a Bayesian probability model is used to quantitatively integrate dynamically changing, unstructured external environmental information (such as rainfall, earthquakes, and loads) to generate a correction coefficient (i.e., an environmental correction factor) that can reflect the actual impact of current environmental conditions on the risk of dam instability in real time. This transforms the fuzzy concept of "environmental risk" into a mathematical parameter that can be accurately calculated and dynamically adjusted, in order to correct and optimize the risk assessment results based solely on structural monitoring data, thereby enabling the early warning system to have the ability to adapt to changes in external conditions.

[0034] In some embodiments of this disclosure, the current environmental variables include any one or more of rainfall intensity, seismic intensity, and dam load change rate. Step 103 may specifically include the following steps: Obtain the weight coefficient for each current environment variable; the weight coefficient is obtained by regression processing using historical environment variable data; Use Bayes' theorem to determine the variable correction factor for each current environmental variable; The environmental correction factor is obtained by weighting and summing the variable correction factors for each current environmental variable based on the weighting coefficients.

[0035] In this embodiment of the disclosure, the aforementioned weighting coefficients can be obtained by performing regression processing using historical environmental variable data: Historical environmental variable data (including time series of rainfall intensity, seismic intensity, and dam load change rate) and the corresponding observed or error values ​​ΔDSI of the comprehensive risk score (DSI) for dam instability risk are collected. For each historical moment, variable correction factors for each environmental variable are calculated based on Bayes' theorem. Using the above variable correction factors as independent variables and the risk score error ΔDSI (or risk state change) at that moment as the dependent variable, a multiple linear regression model is established. The regression coefficients of each variable are obtained by fitting using the least squares method. These coefficients characterize the influence of each environmental correction term on the risk error. The regression coefficients are normalized (so that the sum of all regression coefficients is 1), which yields the weight coefficients a, b, and c used for weighted summation. In addition, to ensure the timeliness of the weight coefficients, a rolling time window can be used to periodically re-regress and update them.

[0036] In some embodiments of this disclosure, environmental correction factors The following formula can be used to calculate it:

[0037] in, For the correction factor of rainfall intensity variable, For earthquake intensity variable correction factor, This is a correction factor for the rate of change of dam load. for The weighting coefficients, b for The weighting coefficients, c for The weighting coefficients.

[0038] In this embodiment, by determining the weight coefficients based on historical data regression and combining the correction factors of each environmental variable dynamically calculated by Bayes' theorem, a weighted fusion is performed. This achieves data-driven quantification, real-time dynamic correction, and multi-factor interpretability fusion of the impact of environmental risks. As a result, external environmental factors such as rainfall and earthquakes are transformed from vague background interference into accurately calculable risk adjustment variables. This significantly improves the scientific perception and adaptive capability of tailings dam instability early warning to changes in external conditions, and avoids the problem of insufficient or excessive early warning sensitivity caused by relying on fixed experience values ​​in traditional methods.

[0039] In some embodiments of this disclosure, step 103, which uses Bayes' theorem to determine the variable correction factor for each current environmental variable, may specifically include the following steps: For each current environmental variable, obtain the prior probability distribution corresponding to the current environmental variable; the prior probability distribution is determined based on the correspondence between historical environmental data and tailings dam instability events; Using current environmental variables as evidence, the prior probability distribution is updated through Bayes' theorem to obtain the posterior probability reflecting the current environmental risk of the tailings dam. The ratio of the posterior probability to the prior probability is calculated to obtain the variable correction factor.

[0040] In this embodiment of the disclosure, Bayes' theorem formula is:

[0041] in, This is the prior probability, that is, the initial probability estimate of the event that "a certain environmental condition E will significantly increase the risk of dam instability" based only on historical statistical experience without considering the current real-time monitoring data D. For example, based on data from the past 10 years, it is statistically determined that the probability of the dam entering the warning state when "the daily rainfall exceeds 50mm" is 15%, then P(E=rainstorm) is 0.15. The posterior probability, which is the total probability of pattern D occurring in the currently observed monitoring data without considering environmental factors, is a normalization factor that ensures the calculated posterior probability is a valid probability value. It equals the sum of probabilities of observing the current monitoring data D under all possible environmental conditions (including safety conditions); P(D|E) is the likelihood probability, which is the conditional probability of actually observing the current monitoring data D (such as a sudden increase in displacement or abnormal seepage pressure) under the assumption that a certain environmental risk condition E (such as a rainstorm) actually exists. For example, if a similar displacement increase pattern (D) is observed 80% of the time when a rainstorm (E) occurs in history, then the value of P(D|E) is relatively large (such as 0.8). The posterior probability is the updated probability estimate of the event "environmental condition E is significantly increasing the risk of the dam body" given the current real-time monitoring data D. For example, even if the probability of a historically heavy rainstorm is not high (P(E) is small), if P(D|E) is large, then the calculated P(E|D) will be significantly higher than the prior probability.

[0042] As an example of a possible implementation, for each environmental variable (such as rainfall intensity), a prior probability distribution is first established based on the correspondence between historical environmental data and dam instability events. For example, statistical analysis is used to determine the historical frequency of dam instability signs within different rainfall intensity ranges, forming an initial relationship model of "rainfall intensity - instability probability". Once the measured value (evidence) of the current environmental variable is obtained, Bayes' theorem is used to update the prior distribution: the current environmental data D is taken as new evidence, combined with the likelihood function (i.e., the probability of observing this environmental evidence under known instability conditions). ) and prior probability The posterior probability reflecting the risk of dam instability under the current specific environmental conditions is calculated. This update process essentially integrates static historical experience with dynamic real-time observation, enabling risk assessments to reflect the latest environmental conditions. Finally, the ratio of the calculated posterior probability to the prior probability is used as a correction factor for the environmental variable. A ratio greater than 1 indicates that the current environment is more risky than historical norms, requiring increased early warning sensitivity; a ratio less than 1 indicates that the current environment is relatively safe. This scheme, through probabilistic modeling, transforms environmental risk assessment from qualitative, empirical judgment to quantitative, dynamic calculation, thereby enabling a scientific response to real-time changes in environmental conditions.

[0043] Step 104: Determine the comprehensive score of the tailings dam's instability risk based on multiple characteristic parameters, their respective weight values, and environmental correction factors.

[0044] In one embodiment, the normalized deviation of each feature parameter can be calculated based on its corresponding preset safety or critical threshold. This deviation is then multiplied by the dynamic weight value output by the LSTM model to reflect the relative importance of different parameters under the current operating conditions. Finally, the weighted results for all feature parameters are summed and multiplied by an environmental correction factor determined by a Bayesian model to obtain the final instability risk comprehensive score (DSI). The DSI not only integrates the dam's structural state but also incorporates the real-time impact of external environments (such as rainfall and earthquakes), achieving a precise and adaptive mapping from multi-dimensional monitoring data to an intuitive risk level, providing a scientific and quantitative core basis for subsequent early warning decisions.

[0045] In some embodiments of this disclosure, step 104 may specifically include the following steps: Based on multiple feature parameters, weight vectors, and environmental correction factors, the comprehensive score for the instability risk of tailings dams is calculated using the following formula:

[0046] in, Let be the weight value of the i-th feature parameter. Let be the value of the i-th feature parameter. The security threshold corresponding to the i-th feature parameter is... This represents the critical threshold corresponding to the i-th feature parameter. It is an environmental correction factor.

[0047] In some embodiments of this disclosure, the weight value of the i-th feature parameter The following steps were used to calculate the result:

[0048]

[0049] in, This is the risk scoring error value. The predicted risk score is determined based on historical data. The overall score for the risk of instability is calculated. The historical weight value of the i-th feature parameter obtained in the previous calculation. Error when given the i-th feature parameter The probability of occurrence, for The total probability of occurrence.

[0050] In this embodiment of the disclosure, the current instability risk comprehensive score (DSI) is calculated and compared with the score predicted based on historical models. Error between This is used to measure the accuracy of the model's predictions. For each feature parameter corresponding to a sensor, the analysis is performed on the historical data of when the sensor data appeared and the current error. The conditional probability P(of the occurrence) |sensor i ), and compare it with the total probability P(error occurs). Compared to the previous time step, the correlation strength ratio between the sensor data and the model error is obtained. This ratio is then used as an adjustment factor and multiplied by the old weight value of the feature parameter from the previous time step. This enables dynamic updates to the weights, achieving continuous self-calibration capabilities. It can automatically identify and reduce the weights of sensors whose readings are frequently correlated with model errors (i.e., reliability decline), while increasing the influence weights of stable and reliable sensors. This achieves adaptive optimization of sensor importance, thereby significantly improving the long-term robustness and prediction accuracy of the risk assessment model in complex scenarios such as sensor performance drift and environmental changes.

[0051] Step 105: Output early warning information based on the comprehensive score of instability risk.

[0052] In one embodiment, an early warning message is output when the comprehensive score for instability risk is greater than a preset score.

[0053] As an example of a possible implementation, a paradigm shift from parameter threshold early warning to quantitative risk assessment can be achieved through a comprehensive dam instability risk scoring model. Traditional systems only issue alarms when a single parameter exceeds a threshold, while this disclosure uses multi-parameter coupling analysis to accurately quantify the overall instability probability of the dam and supports dynamic adjustment of early warning thresholds. For example, the seepage pressure weight can be automatically reduced under heavy rain conditions, and the vibration frequency weight can be increased within 48 hours after a geological disaster, so that the early warning sensitivity matches the disaster scenario. The different sensor parameters under different conditions are all updated and calculated in real time using an LSTM neural network. The boundary conditions of the dam break simulation are dynamically adjusted through the DSI scoring results, and the DSI scoring results are calibrated by the model to optimize the sensor priority and environmental correction factors in the weight allocation algorithm. When the DSI triggers a red alert, a hazard distribution matrix is ​​formed based on the dam's three-dimensional model and sensor data distributed on the dam. The current personnel location, hazard distribution matrix, road network map, and evacuation target points are analyzed, and finally, a dam break risk zoning map and emergency evacuation routes are automatically generated in the digital twin model and simultaneously pushed to the monitoring platform and emergency control system.

[0054] Based on drill core data, ground-penetrating radar scan data, and UAV aerial imagery, a high-fidelity three-dimensional geological model (i.e., a three-dimensional model of the dam body) is constructed for the tailings dam. This model includes parameters such as the thickness of the soil and rock layers, permeability coefficient, and porosity. By accessing real-time monitoring data of dam surface displacement, seepage pressure, and pore water pressure via the Internet of Things, the model parameters can be dynamically corrected using a finite element inversion algorithm, ensuring that the model is synchronized with the physical entity's state.

[0055] In some embodiments of this disclosure, the method further includes: The contribution of the characteristic parameters is calculated using the following formula:

[0056] in, The contribution of the i-th feature parameter is used to evaluate the importance priority of the i-th feature parameter, where k is the sequence number of the historical instability event. Let $\frac{k}{k}$ be the risk score error value for the k-th historical instability event. Let be the weight value of the i-th feature parameter.

[0057] In this embodiment of the disclosure, for the i-th feature parameter, the ratio of the absolute value of the risk score error of the i-th feature parameter to the sum of the risk score errors of all historical instability events of the tailings dam is multiplied by the dynamic weight value of the i-th feature parameter. To obtain the overall contribution This method quantitatively assesses the cumulative impact of each monitoring parameter on historical risk events, thereby determining the importance ranking of sensors. It enables an objective and long-term assessment of sensor importance, providing data support for system maintenance optimization (such as prioritizing the maintenance of high-contribution sensors), key resource allocation (such as strengthening monitoring of critical locations), and priority processing of critical data in emergency situations. This enhances the intelligent management level and overall reliability of the monitoring system.

[0058] The tailings dam instability early warning method proposed in this disclosure involves determining multiple characteristic parameters of the tailings dam based on monitoring data; inputting these parameters into a trained Long Short-Term Memory (LSTM) network model to obtain the weight values ​​of each parameter output by the LSTM model; determining an environmental correction factor based on the current environmental variables of the tailings dam using a Bayesian probability model; determining a comprehensive instability risk score for the tailings dam based on the multiple characteristic parameters, their respective weight values, and the environmental correction factor; and outputting early warning information based on the comprehensive instability risk score. By dynamically allocating the weight values ​​of the characteristic parameters and combining them with the environmental correction factor, adaptive fusion of multi-source monitoring data is achieved, thereby significantly improving the accuracy of tailings dam instability early warning and reducing the risk of false alarms and missed alarms.

[0059] Figure 2This is a block diagram illustrating an early warning device for tailings dam instability according to an exemplary embodiment. (Refer to...) Figure 2 The device includes a first determining unit 201, a second determining unit 202, a third determining unit 203, a fourth determining unit 204, and an early warning unit 205.

[0060] The first determining unit 201 is used to acquire monitoring data of the tailings dam and determine multiple characteristic parameters of the tailings dam based on the monitoring data; the multiple characteristic parameters are used to reflect the stability of the tailings dam. The second determining unit 202 is used to input multiple feature parameters into the trained Long Short-Term Memory (LSTM) network model to obtain the weight values ​​of each feature parameter output by the LSTM model. The third determining unit 203 is used to obtain the current environmental variables of the tailings dam and, based on the current environmental variables, to determine the environmental correction factor using a Bayesian probability model. The fourth determining unit 204 is used to determine the comprehensive score of the tailings dam's instability risk based on multiple characteristic parameters, the weight values ​​of each characteristic parameter, and environmental correction factors. The early warning unit 205 is used to output early warning information based on the comprehensive score of instability risk.

[0061] In some embodiments of this disclosure, the fourth determining unit 204 may specifically be used for: Based on multiple feature parameters, weight vectors, and environmental correction factors, the comprehensive score for the instability risk of tailings dams is calculated using the following formula:

[0062] in, Let be the weight value of the i-th feature parameter. Let be the value of the i-th feature parameter. The security threshold corresponding to the i-th feature parameter is... This represents the critical threshold corresponding to the i-th feature parameter. It is an environmental correction factor.

[0063] In some embodiments of this disclosure, the current environmental variables include any one or more of rainfall intensity, seismic intensity, and dam load change rate. The third determining unit 203 can specifically be used for: Based on current environmental variables, environmental correction factors are determined using a Bayesian probability model, including: Obtain the weight coefficient for each current environment variable; the weight coefficient is obtained by regression processing using historical environment variable data; Use Bayes' theorem to determine the variable correction factor for each current environmental variable; The environmental correction factor is obtained by weighting and summing the variable correction factors for each current environmental variable based on the weighting coefficients.

[0064] In some embodiments of this disclosure, the third determining unit 203 may specifically be used for: For each current environmental variable, obtain the prior probability distribution corresponding to the current environmental variable; the prior probability distribution is determined based on the correspondence between historical environmental data and tailings dam instability events; Using current environmental variables as evidence, the prior probability distribution is updated through Bayes' theorem to obtain the posterior probability reflecting the current environmental risk of the tailings dam. The ratio of the posterior probability to the prior probability is calculated to obtain the variable correction factor.

[0065] In some embodiments of this disclosure, the weight value of the i-th feature parameter The following steps were used to calculate the result:

[0066]

[0067] in, This is the risk scoring error value. The predicted risk score is determined based on historical data. The overall score for the risk of instability is calculated. The historical weight value of the i-th feature parameter obtained in the previous calculation. Error when given the i-th feature parameter The probability of occurrence, for The total probability of occurrence.

[0068] In some embodiments of this disclosure, the apparatus may further include a contribution calculation unit, which may specifically be used for: In some embodiments of this disclosure, the fourth determining unit 204 may specifically be used for: The contribution of the characteristic parameters is calculated using the following formula:

[0069] in, The contribution of the i-th feature parameter is used to evaluate the importance priority of the i-th feature parameter, where k is the sequence number of the historical instability event. Let $\frac{k}{k}$ be the risk score error value for the k-th historical instability event. Let be the weight value of the i-th feature parameter.

[0070] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0071] The tailings dam instability early warning device proposed in this embodiment determines multiple characteristic parameters of the tailings dam based on monitoring data. These parameters are then input into a trained Long Short-Term Memory (LSTM) network model to obtain the weight values ​​of each parameter output by the LSTM model. An environmental correction factor is determined using a Bayesian probability model based on the current environmental variables of the tailings dam. A comprehensive instability risk score for the tailings dam is determined based on the multiple characteristic parameters, their respective weight values, and the environmental correction factor. An early warning message is output based on the comprehensive instability risk score. By dynamically allocating the weight values ​​of the characteristic parameters and combining them with the environmental correction factor, adaptive fusion of multi-source monitoring data is achieved, significantly improving the accuracy of tailings dam instability early warning and reducing the risk of false alarms and missed alarms.

[0072] Figure 3 This is a block diagram illustrating an apparatus for an early warning method for tailings dam instability, according to an exemplary embodiment. For example, apparatus 300 may be an electronic device, such as a mobile phone, computer, digital broadcasting terminal, messaging device, tablet device, personal digital assistant, etc.

[0073] Reference Figure 3 The device 300 may include one or more of the following components: processing component 302, memory 304, power component 306, multimedia component 308, audio component 310, input / output (I / O) interface 312, sensor component 314, and communication component 316.

[0074] Processing component 302 typically controls the overall operation of device 300, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 302 may include one or more processors 320 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 302 may include one or more modules to facilitate interaction between processing component 302 and other components. For example, processing component 302 may include a multimedia module to facilitate interaction between multimedia component 308 and processing component 302.

[0075] Memory 304 is configured to store various types of data to support the operation of device 300. Examples of such data include instructions for any application or method operating on device 300, contact data, phonebook data, messages, pictures, videos, etc. Memory 304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0076] The power supply component 306 provides power to the various components of the device 300. The power supply component 306 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device 300.

[0077] Multimedia component 308 includes a screen that provides an output interface between the device 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 308 includes a front-facing camera and / or a rear-facing camera. When the device 300 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0078] Audio component 310 is configured to output and / or input audio signals. For example, audio component 310 includes a microphone (MIC) configured to receive external audio signals when device 300 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 304 or transmitted via communication component 316. In some embodiments, audio component 310 also includes a speaker for outputting audio signals.

[0079] I / O interface 312 provides an interface between processing component 302 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, start buttons, and lock buttons.

[0080] Sensor assembly 314 includes one or more sensors for providing status assessments of various aspects of device 300. For example, sensor assembly 314 may detect the on / off state of device 300, the relative positioning of components such as the display and keypad of device 300, changes in the position of device 300 or a component of device 300, the presence or absence of user contact with device 300, the orientation or acceleration / deceleration of device 300, and temperature changes of device 300. Sensor assembly 314 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 314 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 314 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0081] Communication component 316 is configured to facilitate wired or wireless communication between device 300 and other devices. Device 300 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 316 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 316 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0082] In an exemplary embodiment, the apparatus 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0083] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 304 including instructions, which can be executed by a processor 320 of the device 300 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0084] In an exemplary embodiment, a computer program product is also provided, including a computer program that implements the above-described method when executed by the processor 320 of the device 300.

[0085] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0086] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for early warning of tailings dam instability, characterized in that, include: Acquire monitoring data of the tailings dam, and determine multiple characteristic parameters of the tailings dam based on the monitoring data; The Multiple characteristic parameters are used to reflect the stability of the tailings dam; Multiple feature parameters are input into the trained Long Short-Term Memory (LSTM) network model to obtain the weight values ​​of each feature parameter output by the LSTM model. Obtain the current environmental variables of the tailings dam, and determine the environmental correction factor based on the current environmental variables using a Bayesian probability model; Based on multiple characteristic parameters, their respective weights, and the environmental correction factor, a comprehensive score for the instability risk of the tailings dam is determined. Early warning information is output based on the comprehensive score of instability risk.

2. The tailings dam instability early warning method according to claim 1, characterized in that, The determination of the comprehensive instability risk score of the tailings dam based on multiple characteristic parameters, their respective weights, and the environmental correction factor includes: Based on multiple feature parameters, the weight vector, and the environmental correction factor, the comprehensive score for the instability risk of the tailings dam is calculated using the following formula: in, Let be the weight value of the i-th feature parameter. Let be the value of the i-th feature parameter. The security threshold corresponding to the i-th feature parameter is... This represents the critical threshold corresponding to the i-th feature parameter. It is an environmental correction factor.

3. The tailings dam instability early warning method according to claim 1, characterized in that, The current environmental variables include any one or more of the following: rainfall intensity, seismic intensity, and dam load change rate. The step of determining the environmental correction factor based on the current environmental variables using a Bayesian probability model includes: Obtain the weight coefficient for each current environment variable; the weight coefficient is obtained by regression processing using historical environment variable data; Use Bayes' theorem to determine the variable correction factor for each current environmental variable; The environmental correction factor is obtained by weighting and summing the variable correction factors for each current environmental variable based on the weighting coefficients.

4. The tailings dam instability early warning method according to claim 3, characterized in that, The process of determining the variable correction factor for each current environmental variable using Bayes' theorem includes: For each current environmental variable, obtain the prior probability distribution corresponding to the current environmental variable; the prior probability distribution is determined based on the correspondence between historical environmental data and the tailings dam instability events; Using the current environmental variables as evidence, the prior probability distribution is updated using Bayes' theorem to obtain the posterior probability reflecting the current environmental risk of the tailings dam. The ratio of the posterior probability to the prior probability is calculated to obtain the variable correction factor.

5. The tailings dam instability early warning method according to claim 2, characterized in that, The weight value of the i-th feature parameter The following steps were used to calculate the result: in, This is the risk scoring error value. The predicted risk score is determined based on historical data. The overall score for the aforementioned instability risk is as follows: The historical weight value of the i-th feature parameter obtained in the previous calculation. Error when given the i-th feature parameter The probability of occurrence, for The total probability of occurrence.

6. The tailings dam instability early warning method according to claim 1, characterized in that, The method also includes: The contribution of the characteristic parameters is calculated using the following formula: in, The contribution of the i-th feature parameter is used to evaluate the importance priority of the i-th feature parameter, where k is the sequence number of the historical instability event. Let $\frac{k}{k}$ be the risk score error value for the k-th historical instability event. Let be the weight value of the i-th feature parameter.

7. A tailings dam instability early warning device, characterized in that, include: The first determining unit is used to acquire monitoring data of the tailings dam and determine multiple characteristic parameters of the tailings dam based on the monitoring data. The Multiple characteristic parameters are used to reflect the stability of the tailings dam; The second determining unit is used to input multiple feature parameters into the trained Long Short-Term Memory (LSTM) network model to obtain the weight values ​​of each of the multiple feature parameters output by the LSTM model. The third determining unit is used to obtain the current environmental variables of the tailings dam and, based on the current environmental variables, determine the environmental correction factor using a Bayesian probability model. The fourth determining unit is used to determine the comprehensive score of the instability risk of the tailings dam based on multiple feature parameters, the weight values ​​of each feature parameter, and the environmental correction factor. The early warning unit is used to output early warning information based on the comprehensive score of instability risk.

8. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, The computer program, when executed by a processor, implements the method as described in any one of claims 1 to 6.

Images