A method and system for evaluating the safety state of a hydraulic structure

Abstract

This invention discloses a method and system for assessing the safety status of hydraulic structures. The safety status assessment method includes the following steps: acquiring monitoring data from a set of monitoring points during the operation of the hydraulic structure; performing single-factor curve fitting on the time-series data of the monitoring points based on the continuous mapping characteristics of the monitoring data and the mathematical laws of data fitting to obtain the overall deviation of the monitoring point set; determining the basic influence weight and corrected influence weight of the monitoring point set on the safety of the hydraulic structure based on the common knowledge corpus and proprietary knowledge corpus of the safety evaluation of hydraulic structures; and calculating the safety status value of the hydraulic structure based on the overall deviation, basic influence weight, and corrected influence weight of the monitoring point set. This invention can dynamically track the long-term evolution of the monitoring point data. The model can not only accurately reflect the characteristics of existing data, but also continuously optimize the ability of the fitted curve to analyze the continuous changes of a single variable through iterative correction using measured data.

Description

Technical Field

[0001] This invention relates to the field of smart hydropower technology, specifically to a method and system for assessing the safety status of hydraulic structures. Background Technology

[0002] Safety analysis of hydraulic structures is one of the core aspects of smart hydropower. Traditional methods for safety analysis of hydraulic structures often rely on historical extreme values ​​and fixed limits, which have significant limitations. On the one hand, they are difficult to consider the impact of long-term changes in measuring points on the safety of hydraulic structures, and the analysis perspective is relatively short-sighted, with insufficient analysis of continuous changes caused by a single variable. On the other hand, the prediction of safety issues is lagging and it is impossible to predict potential risks in advance.

[0003] Hydropower station hydraulic monitoring data can be divided into three categories: First, raw data, which consists of basic environmental and attribute parameters. This data is characterized by location-specific variations and difficulty in measurement, and its long-term effects can influence intermediate data. Second, intermediate data, which consists of intermediate variables influenced by raw data. These variables require sensor measurement, are highly abstract, and their impact mechanisms and proportions on intuitive data are difficult to summarize. Third, intuitive data, which is most closely related to the hydropower station's operational period and directly reflects the station's current state. This data is readily available, but some data can only be observed in real-time, and due to technological and algorithmic limitations, effective prediction is difficult. Traditional methods, due to these shortcomings, cannot provide a scientific and timely evaluation and prediction of the safety status of hydraulic structures. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention designs a method for assessing the safety status of hydraulic structures, comprising the following steps: acquiring monitoring data from a set of monitoring points during the operation of the hydraulic structure; performing single-factor curve fitting processing on the time-series data of the monitoring points based on the continuous mapping characteristics of the monitoring data and the mathematical laws of data fitting to obtain the overall deviation of the monitoring point set; determining the basic influence weight and corrected influence weight of the monitoring point set on the safety of the hydraulic structure based on the common knowledge corpus and proprietary knowledge corpus of the safety evaluation of hydraulic structures; calculating the safety status value of the hydraulic structure based on the overall deviation, basic influence weight, and corrected influence weight of the monitoring point set; and judging and predicting the safety status of the hydraulic structure based on the magnitude of the safety status value.

[0005] Preferably, the formula for calculating the safety state value is: S=a1×b1×▽1+a2×b2×▽2+……+an×bn×▽n; where S is the safety state value, an is the basic influence weight of the nth measuring point group on the safety of hydraulic structures, bn is the corrected influence weight of the nth measuring point group on the safety of hydraulic structures, and ▽n is the overall deviation of the nth measuring point group.

[0006] Preferably, the process of performing single-factor curve fitting on the time series data of the measurement points to obtain the overall deviation of the measurement point group includes the following steps: establishing a regression model for the time series data of a single measurement point using artificial intelligence, wherein the regression model is a fitting curve based on existing intermediate data, which has the characteristics of accurately reflecting existing data, making reasonable and understandable predictions of future time, and being able to be corrected and iterated according to measured data; predicting future changes of the measurement points according to the regression model, and using the future curve time function as the state prediction value for that period; acquiring or calculating the actual state value of the measurement point through sensors, and calculating the difference between the actual state value and the state prediction value; normalizing the difference to obtain the deviation of the measurement point, where the deviation is a value in the interval (0, 1); and summing the deviations of all measurement points in the measurement point group to obtain the overall deviation of the measurement point group.

[0007] Preferably, an artificial intelligence-based regression model for single-point time series data is established, employing one or more combinations of least squares method, neural network model, and large time series model.

[0008] Preferably, the single-factor curve fitting process for the time series data of the measurement points also includes the following steps: regression model optimization: rewrite the difference between the actual value and the fitted value in the form of a percentage, perform statistics on the data with a difference greater than 90%, ignore the data with a difference greater than 90% when the number is less than 10%, and revise the regression model again when the number is not less than 10%.

[0009] Preferably, determining the basic influence weight of the measuring point group on the safety of hydraulic structures includes the following steps: screening common knowledge corpus for the safety evaluation of hydraulic structures, and determining the measuring point group required for the safety status assessment; performing word segmentation and vectorization processing on the common knowledge corpus, and setting the initial value of the basic influence weight based on the frequency of simultaneous occurrence of the measuring point group name and the hydraulic structure name; adjusting the initial value according to the actual operation of the hydraulic structure to obtain the final basic influence weight.

[0010] Preferably, determining the corrective impact weight of the monitoring point group on the safety of hydraulic structures includes the following steps: collecting proprietary knowledge corpus of design and operation data of hydraulic structures, historical defects and potential hazards, and registration and maintenance status; performing word segmentation and vectorization processing on the proprietary knowledge corpus, and setting initial values ​​for the corrective impact weight based on the frequency of simultaneous occurrence of the monitoring point group name and the hydraulic structure name; adjusting the initial values ​​according to the actual operation of the hydraulic structure to obtain the final corrective impact weight.

[0011] Based on the same design concept, this invention designs a safety status assessment system for hydraulic structures, characterized by comprising: a data processing module, used to acquire monitoring data of a set of measuring points during the operation of the hydraulic structure, and perform single-factor curve fitting processing on the time-series data of the measuring points according to the continuous mapping characteristics of the monitoring data and the mathematical laws of data fitting, to obtain the overall deviation of the measuring point group; a weight determination module, used to determine the basic influence weight and corrected influence weight of the measuring point group on the safety of the hydraulic structure based on the common knowledge corpus and proprietary knowledge corpus of the safety evaluation of hydraulic structures; and a safety assessment module, used to calculate the safety status value of the hydraulic structure based on the overall deviation of the measuring point group, the basic influence weight, and the corrected influence weight, and to judge and predict the safety status of the hydraulic structure based on the magnitude of the safety status value.

[0012] Preferably, the formula for calculating the safety state value is: S=a1×b1×▽1+a2×b2×▽2+……+an×bn×▽n; where S is the safety state value, an is the basic influence weight of the nth measuring point group on the safety of hydraulic structures, bn is the corrected influence weight of the nth measuring point group on the safety of hydraulic structures, and ▽n is the overall deviation of the nth measuring point group.

[0013] Preferably, the data processing module includes: a model building unit, used to establish a regression model for single-point time-series data using artificial intelligence, wherein the regression model is a fitted curve based on existing intermediate data, possessing the characteristics of accurately reflecting existing data, making reasonable and understandable predictions of future time, and being able to be corrected and iterated based on measured data; a state prediction unit, used to predict future changes of the measuring point based on the regression model, using the future curve time function as the state prediction value for that period; a difference calculation unit, used to acquire or calculate the actual state value of the measuring point through sensors, and calculate the difference between the actual state value and the state prediction value; and a deviation generation unit, used to normalize the difference to obtain the deviation of the measuring point, the deviation being a value within the interval (0, 1), and to accumulate the deviations of all measuring points in the measuring point group to obtain the overall deviation of the measuring point group.

[0014] Compared with the closest prior art, the beneficial effects of the present invention are as follows: 1. Based on the continuous mapping characteristics of hydraulic structure monitoring data, this invention constructs a single-point time series data regression model through artificial intelligence, which can dynamically track the long-term evolution of the monitoring point data. The model can not only accurately reflect the characteristics of existing data, but also continuously optimize the ability of the fitted curve to analyze the continuous changes of a single variable through iterative correction of measured data. It can also capture the gradual process of monitoring point data from normal fluctuations to abnormal accumulation, avoiding the one-sidedness of safety analysis caused by ignoring long-term trends.

[0015] 2. This invention uses a regression model obtained through single-factor curve fitting to predict future state values. Combined with deviation calculation normalized by the Sigmoid function, it can transform the difference between actual and predicted values ​​into a quantitative index within the (0,1) interval, identifying the trend of measurement data approaching abnormal states in advance. This breaks the lag of traditional methods that can only respond to problems after the fact, enabling the prediction and early warning of potential safety risks. On the other hand, this invention uses intermediate data as the core fitting object, avoiding the difficulties in collecting original data due to its location-specific variations and difficulty in measurement. Through artificial intelligence model iteration and overall deviation accumulation, it can reverse the safety status by utilizing the continuous change characteristics of intermediate data without precisely summarizing the impact mechanism of intermediate data on intuitive data. At the same time, it combines the real-time observation results of intuitive data to correct the model. Finally, through the weighted calculation of basic influence weight + corrected influence weight, it achieves the synergistic utilization of three types of monitoring data, solving the problem of unscientific and untimely safety assessment caused by poor data type adaptability in traditional methods. Attached Figure Description Figure 1 This is a flowchart illustrating the safety status assessment method of the present invention.

[0016] Figure 2 This is a schematic diagram of the hydraulic structure measuring point group of the present invention. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0018] Example 1 like Figures 1-2 As shown, this invention provides a method for assessing the safety status of hydraulic structures, comprising the following steps: acquiring monitoring data from a set of monitoring points during the operation of the hydraulic structure; performing single-factor curve fitting processing on the time-series data of the monitoring points based on the continuous mapping characteristics of the monitoring data and the mathematical laws of data fitting to obtain the overall deviation of the monitoring point set; determining the basic influence weight and corrected influence weight of the monitoring point set on the safety of the hydraulic structure based on the common knowledge corpus and proprietary knowledge corpus of the safety evaluation of hydraulic structures; calculating the safety status value of the hydraulic structure based on the overall deviation, basic influence weight, and corrected influence weight of the monitoring point set; and judging and predicting the safety status of the hydraulic structure based on the magnitude of the safety status value.

[0019] In a preferred embodiment, the formula for calculating the safety state value is: S = a1 × b1 × ▽1 + a2 × b2 × ▽2 + ... + an × bn × ▽n; where S is the safety state value, an is the basic influence weight of the nth measuring point group on the safety of hydraulic structures, bn is the corrected influence weight of the nth measuring point group on the safety of hydraulic structures, and ▽n is the overall deviation of the nth measuring point group.

[0020] In a preferred embodiment, a single-factor curve fitting process is performed on the time series data of the measurement points to obtain the overall deviation of the measurement point group. This includes the following steps: establishing a regression model for the time series data of a single measurement point using artificial intelligence, wherein the regression model is a fitting curve based on existing intermediate data, possessing the characteristics of accurately reflecting existing data, making reasonable and understandable predictions of future time, and being able to be corrected and iterated based on actual measured data; predicting future changes of the measurement points based on the regression model, and using the future curve time function as the state prediction value for that period; acquiring or calculating the actual state value of the measurement point through sensors, and calculating the difference between the actual state value and the state prediction value; normalizing the difference to obtain the deviation of the measurement point, where the deviation is a value within the interval (0, 1); and summing the deviations of all measurement points in the measurement point group to obtain the overall deviation of the measurement point group.

[0021] In a preferred embodiment, a regression model for single-point time series data is established using artificial intelligence, employing one or more combinations of least squares method, neural network model, and large time series model.

[0022] In a preferred embodiment, the single-factor curve fitting process for the time series data of the measurement points further includes the following steps: regression model optimization: rewrite the difference between the actual value and the fitted value in the form of a percentage, perform statistics on the data with a difference greater than 90%, ignore the data with a difference greater than 90% when the number is less than 10%, and revise the regression model again when the number is not less than 10%.

[0023] In a preferred embodiment, determining the basic influence weight of the measuring point group on the safety of hydraulic structures includes the following steps: screening common knowledge corpus for the safety evaluation of hydraulic structures, and determining the measuring point group required for the safety status assessment; performing word segmentation and vectorization processing on the common knowledge corpus, and setting an initial value for the basic influence weight based on the frequency of simultaneous occurrence of the measuring point group name and the hydraulic structure name; adjusting the initial value according to the actual operation of the hydraulic structure to obtain the final basic influence weight.

[0024] In a preferred embodiment, determining the corrective influence weight of the measuring point group on the safety of hydraulic structures includes the following steps: collecting proprietary knowledge corpus of design and operation data of hydraulic structures, historical defects and potential hazards, and registration and scheduled inspection status; performing word segmentation and vectorization processing on the proprietary knowledge corpus, and setting an initial value for the corrective influence weight based on the frequency of simultaneous occurrence of the measuring point group name and the hydraulic structure name; adjusting the initial value according to the actual operation of the hydraulic structure to obtain the final corrective influence weight.

[0025] Based on the same design concept, this invention designs a safety status assessment system for hydraulic structures, characterized by comprising: a data processing module, used to acquire monitoring data of a set of measuring points during the operation of the hydraulic structure, and perform single-factor curve fitting processing on the time-series data of the measuring points according to the continuous mapping characteristics of the monitoring data and the mathematical laws of data fitting, to obtain the overall deviation of the measuring point group; a weight determination module, used to determine the basic influence weight and corrected influence weight of the measuring point group on the safety of the hydraulic structure based on the common knowledge corpus and proprietary knowledge corpus of the safety evaluation of hydraulic structures; and a safety assessment module, used to calculate the safety status value of the hydraulic structure based on the overall deviation of the measuring point group, the basic influence weight, and the corrected influence weight, and to judge and predict the safety status of the hydraulic structure based on the magnitude of the safety status value.

[0026] In a preferred embodiment, the formula for calculating the safety state value is: S = a1 × b1 × ▽1 + a2 × b2 × ▽2 + ... + an × bn × ▽n; where S is the safety state value, an is the basic influence weight of the nth measuring point group on the safety of hydraulic structures, bn is the corrected influence weight of the nth measuring point group on the safety of hydraulic structures, and ▽n is the overall deviation of the nth measuring point group.

[0027] In a preferred embodiment, the data processing module includes: a model building unit, used to build a regression model for single-point time-series data using artificial intelligence, wherein the regression model is a fitted curve based on existing intermediate data, and has the characteristics of accurately reflecting existing data, making reasonable and understandable predictions of future time, and being able to be corrected and iterated according to measured data; a state prediction unit, used to predict future changes of the measuring point according to the regression model, and use the future curve time function as the state prediction value for that period; a difference calculation unit, used to acquire or calculate the actual state value of the measuring point through sensors, and calculate the difference between the actual state value and the state prediction value; and a deviation generation unit, used to normalize the difference to obtain the deviation of the measuring point, wherein the deviation is a value in the interval (0, 1), and to accumulate the deviations of all measuring points in the measuring point group to obtain the overall deviation of the measuring point group.

[0028] Based on the continuous mapping characteristics of hydraulic structure monitoring data, this invention constructs a time-series data regression model for a single measuring point using artificial intelligence. This model can dynamically track the long-term evolution of the measuring point data. The model can not only accurately reflect the characteristics of existing data, but also continuously optimize the ability of the fitted curve to analyze the continuous changes of a single variable through iterative correction using measured data. It can also capture the gradual process of measuring point data from normal fluctuations to abnormal accumulation, avoiding the one-sidedness of safety analysis caused by ignoring long-term trends.

[0029] The regression model obtained by single-factor curve fitting can be extended forward to predict the state value of future periods. Combined with the deviation calculation normalized by the Sigmoid function, the difference between the actual value and the predicted value can be transformed into a quantitative index in the (0,1) interval, which can identify the trend of the measurement point data approaching the abnormal state in advance, breaking the lag of traditional methods that can only respond to problems after the fact, and realizing the prediction and early warning of potential safety risks. On the other hand, this invention uses intermediate data as the core fitting object, which not only avoids the difficulty of collecting original data due to its varying location and difficulty in measurement, but also uses artificial intelligence model iteration and overall deviation accumulation. It does not need to accurately summarize the influence mechanism of intermediate data on intuitive data. It can use the continuous change characteristics of intermediate data to infer the safety status. At the same time, it combines the real-time observation results of intuitive data to correct the model. Finally, through the weighted calculation of basic influence weight + corrected influence weight, it realizes the synergistic use of three types of monitoring data, which solves the problem of unscientific and untimely safety evaluation caused by poor data type adaptability of traditional methods. In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0030] Furthermore, the terms "upper" and "lower" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "upper" or "lower" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0031] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between components; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0032] In this application, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0033] The above are merely preferred embodiments of the present invention and are not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of the claims of the pending application of the present invention.

Claims

1. A method for evaluating a safety state of a hydraulic structure, characterized by, Includes the following steps: The monitoring data of the measuring point group during the operation of the hydraulic structure are obtained. Based on the continuous mapping characteristics of the monitoring data and the mathematical law of data fitting, the time series data of the measuring points are subjected to single-factor curve fitting processing to obtain the overall deviation of the measuring point group. Based on the common knowledge corpus and proprietary knowledge corpus of the safety assessment of hydraulic structures, the basic influence weight and modified influence weight of the measuring point group on the safety of hydraulic structures are determined. Based on the overall deviation of the measuring point group, the basic influence weight, and the corrected influence weight, the safety status value of the hydraulic structure is calculated, and the safety status of the hydraulic structure is judged and predicted based on the magnitude of the safety status value.

2. The method of safety state evaluation of a hydraulic structure according to Claim 1, wherein The formula for calculating the safety status value is: S=a1×b1×▽1+a2×b2×▽2+……+an×bn×▽n; In the formula, S is the safety state value, an is the basic influence weight of the nth measuring point group on the safety of hydraulic structures, bn is the corrected influence weight of the nth measuring point group on the safety of hydraulic structures, and ▽n is the overall deviation of the nth measuring point group.

3. The method of claim 1, wherein, The process of performing single-factor curve fitting on the time-series data of the measurement points to obtain the overall deviation of the measurement point group includes the following steps: Artificial intelligence is used to establish a regression model for time series data of a single measurement point. The regression model is a fitting curve based on existing intermediate data. It has the characteristics of accurately reflecting existing data, making reasonable and understandable predictions of future time, and being able to be corrected and iterated based on actual measurement data. The regression model is used to predict future changes in the measurement points, and the future curve time function is used as the state prediction value for that period. The actual state value of the measuring point is acquired or calculated by the sensor, and the difference between the actual state value and the predicted state value is calculated. The difference is normalized to obtain the deviation of the measuring point, and the deviation is a value in the interval (0, 1). The deviations of all measuring points in the measuring point group are summed to obtain the overall deviation of the measuring point group.

4. The method of claim 3, wherein The regression model for single-point time series data is established using artificial intelligence, employing one or more combinations of least squares method, neural network model, and large time series model.

5. The method for safety state evaluation of a hydraulic structure according to Claim 3, wherein The single-factor curve fitting process for the time series data of the measurement points also includes the following steps: Regression model optimization: The difference between the actual value and the fitted value is rewritten as a percentage. For data with a difference greater than 90%, statistics are performed. If the number is less than 10%, the data with a difference greater than 90% is ignored. If the number is not less than 10%, the regression model is revised.

6. The method for safety state evaluation of a hydraulic structure according to Claim 1, wherein The determination of the basic influence weight of the measuring point group on the safety of hydraulic structures includes the following steps: Screen common knowledge corpora for safety evaluation of hydraulic structures to determine the set of measurement points required for safety status assessment; The common knowledge corpus is segmented and vectorized, and the initial value of the basic influence weight is set according to the frequency of simultaneous occurrence of the measurement point group name and the hydraulic structure name. The initial values ​​are adjusted according to the actual operation of the hydraulic structure to obtain the final basic influence weight.

7. The method for safety state evaluation of a hydraulic structure according to Claim 1, wherein The determination of the corrected impact weight of the measuring point group on the safety of hydraulic structures includes the following steps: Collect proprietary knowledge corpus of design and operation data, historical defects and potential hazards, and registration and regular inspection information of hydraulic structures; The proprietary knowledge corpus is processed by word segmentation and vectorization. Based on the frequency of simultaneous occurrence of the names of the measurement point group and the hydraulic structure, the initial value of the correction influence weight is set. The initial values ​​are adjusted according to the actual operation of the hydraulic structures to obtain the final corrected influence weight.

8. A hydraulic structure safety state evaluation system characterized by comprising: include: The data processing module is used to acquire monitoring data of the measuring point group during the operation of hydraulic structures. Based on the continuous mapping characteristics of the monitoring data and the mathematical law of data fitting, it performs single-factor curve fitting processing on the time series data of the measuring points to obtain the overall deviation of the measuring point group. The weight determination module is used to determine the basic influence weight and modified influence weight of the measuring point group on the safety of hydraulic structures based on the common knowledge corpus and proprietary knowledge corpus of the safety evaluation of hydraulic structures. The safety assessment module is used to calculate the safety status value of hydraulic structures based on the overall deviation of the measuring point group, the basic influence weight, and the corrected influence weight, and to judge and predict the safety status of hydraulic structures based on the magnitude of the safety status value.

9. The hydraulic structure safety state evaluation system according to Claim 8, wherein The formula for calculating the safety status value is: S=a1×b1×▽1+a2×b2×▽2+……+an×bn×▽n; In the formula, S is the safety state value, an is the basic influence weight of the nth measuring point group on the safety of hydraulic structures, bn is the corrected influence weight of the nth measuring point group on the safety of hydraulic structures, and ▽n is the overall deviation of the nth measuring point group.

10. The hydraulic structure safety state evaluation system according to Claim 8, wherein The data processing module includes: The model building unit is used to build a regression model for single-point time series data using artificial intelligence. The regression model is a fitting curve based on existing intermediate data and has the characteristics of accurately reflecting existing data, making reasonable and understandable predictions of future time, and being able to be corrected and iterated based on actual measured data. The state prediction unit is used to predict future changes of the measuring points based on the regression model, and to use the future curve time function as the state prediction value for that period. The difference calculation unit is used to acquire or calculate the actual state value of the measuring point through the sensor, and calculate the difference between the actual state value and the state prediction value; The deviation generation unit is used to normalize the difference to obtain the deviation of the measuring point. The deviation is a value in the interval (0, 1). The deviation of all measuring points in the measuring point group is accumulated to obtain the overall deviation of the measuring point group.

Images