Method for setting dynamic threshold of civil structure monitoring data and safety warning method

Abstract

The present application relates to the technical field of civil structure monitoring, and discloses a setting method of dynamic threshold of civil structure monitoring data and a safety warning method. The setting method first collects all monitoring data in a previous preset time period of a sensor, fits the serial number corresponding to each monitoring data with the monitoring data according to the time sequence of data collection, and obtains a fitting curve. Then, the monitoring data is subtracted from the data of the corresponding serial number in the fitting curve in sequence, a group of data without trend items is obtained, the absolute value is taken, the normal distribution is calculated, the data in the upper and lower quartile ranges are selected and the mean value is taken as the accurate value, the data is fitted again through comparison and replacement, and the fitting curve under the influence of external load can be obtained. Finally, the monitoring data of a subsequent preset time period of the sensor is numbered in the same way, the monitoring data center curve of the future time period is obtained, and the dynamic threshold interval is generated accordingly. The present application can set appropriate dynamic thresholds for the civil structure monitoring data of different measuring points.

Description

Technical Field

[0001] This invention relates to the field of civil structure monitoring technology, specifically to a method for setting dynamic thresholds for civil structure monitoring data, and a safety early warning method using this setting method. Background Technology

[0002] While the technical level of civil structure management and maintenance has made significant progress, the age of some civil structures, environmental erosion, or external loads have led to a decline in material performance and component aging, posing a significant risk to the safety of civil structures and resulting in frequent civil structure collapses in recent years. The key to solving the current problem of civil structure safety protection lies in establishing a structural health monitoring system to achieve real-time acquisition and dynamic sensing of the stress state of the structure, timely and effective assessment of the structure's safety performance, and response to potential safety risks.

[0003] Civil structure health monitoring systems rely on front-end data acquisition systems to transmit massive amounts of monitoring data to the monitoring system in real time. Currently, in addition to short-term fluctuations caused by external loads, civil structure monitoring data is largely affected by the daily, quarterly, and annual cycles of environmental temperature. Therefore, setting the threshold values ​​for civil structure monitoring data is crucial. It is necessary not only to consider the coupled effects of external loads and temperature, but also to effectively display abnormal data caused by external loads on the civil structure; otherwise, the assessment of civil structure safety will be significantly affected.

[0004] Currently, static thresholds are commonly used for setting monitoring data thresholds in civil structures. However, due to the influence of temperature effects, monitoring data in civil structures undergo periodic changes, resulting in a non-constant "absolute height difference" between the threshold and the baseline monitoring data, leading to excessive false alarms. Although some monitoring indicators use dynamic thresholds, most are based on calculations using finite element method (FEM) software. Considering the different sensor installation locations, the results of a single FEM software are difficult to apply to all monitoring indicators and to set all sensors. Therefore, it is difficult to use historical monitoring data to reverse-engineer the monitoring data and use the historical variation characteristics of monitoring data from different locations on the civil structure as a standard for dynamic threshold evaluation. This is a problem that urgently needs to be solved in the field of civil structure monitoring technology. Summary of the Invention

[0005] To address the technical problem in existing technologies, namely the difficulty in setting effective and appropriate dynamic thresholds for civil structure monitoring data at different monitoring points, this invention provides a method for setting dynamic thresholds for civil structure monitoring data and a safety early warning method.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] This invention discloses a method for setting dynamic thresholds for civil structure monitoring data, including steps S1 to S6.

[0008] S1. Collect all monitoring data from the sensor before the preset time period T prior to the current time t0. i The monitoring data S is arranged in chronological order of collection time. i The corresponding serial number N i By fitting it, the fitted curve Y is obtained. n .

[0009] S2. Sequentially process each monitoring data S i With fitted curve Y n Data Y corresponding to the serial number n(i) The difference is calculated to obtain a set of data Q = {q1, q2, ..., q} after removing the trend term. K}, and take the absolute value |Q|; where q i =S i -Y n(i) , i = 1, 2, ..., K.

[0010] S3. Calculate the normal distribution of |Q|, select the data in the upper and lower quartiles of the normal distribution and calculate the mean, and use the mean as the precise value A for the difference substitution.

[0011] S4. Determine the absolute value of each data point in Q, |q|. i | and the size of A; if |q i If |≥A, then the corresponding monitoring data S i Replace with S i The average of the previous p monitoring data points yields a set of data M after removing external loads. i ; p≥2.

[0012] S5. Set the corresponding sequence number N i Data M with removal of external load i By performing fitting, the fitting curve Y under the influence of external load is obtained. m .

[0013] S6. The monitoring data not collected in the preset time period T after the current time t0 of the sensor are numbered by referring to the monitoring data in the previous preset time period T, and these numbers are substituted into the fitting curve Y. m Thus, the predicted center curve Y of the sensor's monitoring data for the preset time period T is obtained. f , with Y f The dynamic threshold range of the civil structure sensor monitoring data is generated around the center.

[0014] As a further improvement to the above scheme, in step S6, by using the center curve Y... fCentered on the data, shift upwards by a value |H1| and downwards by a value |H2| to obtain the dynamic threshold interval H of the civil structure sensor monitoring data. The calculation formula is as follows:

[0015] H = [Y] f +|H1|,Y f -|H2|]

[0016]

[0017]

[0018] In the formula, r is the preset number of days, j = 1, 2, ..., r; in the first r days, Q dxj For all data q after removing the trend term on day j i The maximum value in Q dnj For all data q after removing the trend term on day j i The minimum value in Q; dxa Q represents the maximum value of all values ​​after removing the trend term from the previous r days. dx The average value of Q; dna Q is the minimum value of all values ​​after removing the trend term from the previous r days. dn The average value; λ and β are the adjustment coefficients for the upper and lower limits of the threshold, respectively.

[0019] As a further improvement to the above scheme, in step S6, the preset number of days r is 5 to 10 days.

[0020] As a further improvement to the above scheme, in step S2, the data Q obtained each time after removing the trend term is further processed. i The data is stored and sequentially generated into daily trend-removed data Q. d And calculate Q. d The mean μ and standard deviation δ, when |S i When -μ|>3δ, the corresponding S i It is not used for calculating the maximum and minimum values ​​in step S6.

[0021] As a further improvement to the above scheme, in step S4, S i The average value S of the previous p monitoring data i * represents:

[0022]

[0023] Among them, the corresponding monitoring data S i Replacement S i *Continue participating in other steps in S4 | q i | The calculation process for determining the size of A.

[0024] As a further improvement to the above scheme, in step S4, S i *This is obtained by averaging the previous p monitoring data, where 5 ≤ p ≤ 10.

[0025] As a further improvement to the above scheme, the preset time period T has a duration of 10 to 60 minutes.

[0026] As a further improvement to the above scheme, in step S1, the collected sensor monitoring data includes temperature effect and response data caused by external load.

[0027] This invention also discloses a safety early warning method for monitoring data of civil structures, including step one and step two.

[0028] 1. Set the dynamic threshold range for a future preset time period T of the civil structure sensor monitoring data in real time; wherein, the dynamic threshold range is set using the above-mentioned method for setting dynamic thresholds for civil structure monitoring data.

[0029] 2. The monitoring data collected by the sensor during a preset time period T is compared with the dynamic threshold range. An alarm is generated when the monitoring data exceeds the upper or lower limit.

[0030] Compared with the prior art, the beneficial effects of the present invention are:

[0031] (1) The method for setting dynamic thresholds for civil structure monitoring data disclosed in this invention only transforms the sensor monitoring data using an algorithm, without considering the influence of too many isomorphic sensors, thus ensuring the independence and authenticity of the civil structure monitoring data. Furthermore, it is applicable to all civil structure monitoring points, not limited to high-frequency and low-frequency data, and can achieve real-time batch calculation of data collected from different monitoring indicators of civil structures.

[0032] (2) The present invention fits the corresponding sensor monitoring data and serial number within a certain time period to remove the influence of trend change caused by temperature effect. By fitting the monitoring data after difference replacement with the corresponding serial number again, the influence of external load and burr can be effectively eliminated, ensuring high accuracy of prediction data.

[0033] (3) After removing the trend changes caused by temperature effect, the present invention selects the historical maximum and minimum values ​​to determine the upper and lower limits of the threshold interval. Based on the characteristics of the monitoring data, it is used to predict the monitoring data, which can reflect the real stress state of the civil structure, ensure the reliability of the selected threshold data, and reduce the occurrence of false alarms. Attached Figure Description

[0034] Figure 1 This is a flowchart of a safety early warning method for monitoring civil structure data in an embodiment of the present invention.

[0035] Figure 2 This is a flowchart illustrating the method for setting dynamic thresholds for civil structure monitoring data in an embodiment of the present invention.

[0036] Figure 3 This is a graph of the original monitoring sensor for civil structures in an embodiment of the present invention.

[0037] Figure 4 This is a graph showing the relationship between the dynamic threshold and sensor monitoring data in an embodiment of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Please see Figure 1 This invention provides a safety early warning method for monitoring civil structures, comprising step one and step two.

[0040] 1. Set the dynamic threshold range for the future preset time period T of the civil structure sensor monitoring data in real time;

[0041] The dynamic threshold range is set using the following method for setting dynamic thresholds for civil structure monitoring data.

[0042] Please see Figure 2 The method for setting dynamic thresholds for civil structure monitoring data includes steps S1 to S6.

[0043] S1. Collect all monitoring data from the sensor at the current time t0, preceding a preset time period T = 10 min. i The monitoring data S is arranged in chronological order of collection time. i The corresponding serial number N i By fitting it, the fitted curve Y is obtained. n .

[0044] In this embodiment, the collected sensor monitoring data may include response data caused by temperature effects and external loads. In some embodiments, the sensor may be of electrical, optical, magnetic, or other types, and is typically installed in locations such as main beams, supports, piers, walls, roofs, and building foundations to monitor structural stress, deformation, and other data. When collecting this data, it is usually affected by temperature effects and external loads; therefore, setting dynamic thresholds can achieve more reasonable early warnings and reduce false alarm rates.

[0045] In addition, the preset time period T in this embodiment is set to a duration of 10 minutes. Of course, in other embodiments, the preset time period T can also be set to other durations, and the duration can be adaptively adjusted according to factors such as the application environment of the method and the frequency of data acquisition.

[0046] The fitted data parameters are an i×2 matrix, with the first column being the sensor monitoring data sequence number N sorted by time. i The second column contains sensor monitoring data S. i The curve Y was obtained by fitting. n .

[0047] Please see Figure 3 The figure shows the curve of the change of raw strain data monitored by the civil structure sensor in this embodiment over a day.

[0048] S2. Sequentially process each monitoring data S i With fitted curve Y n Data Y corresponding to the serial number n(i) The difference is calculated to obtain a set of data Q = {q1, q2, ..., q} after removing the trend term. K}, and take the absolute value |Q|; where q i =S i -Y n(i) , i = 1, 2, ..., K.

[0049] It should be noted that in step S2, the data Q obtained each time the trend item is removed is also stored, forming the daily trend-removed data Q sequentially. d And calculate Q. d The mean μ and standard deviation δ, when |S i When -μ|>3δ, the corresponding S i It is not used for calculating the maximum and minimum values ​​in step S6.

[0050] S3. Calculate the normal distribution of |Q|. Data within the upper and lower quartiles of the normal distribution can be selected and their mean calculated. This mean is used as the precise value A for the difference substitution. This method effectively removes the influence of external loads and puncture data, preserving the fluctuations in monitoring data of the civil structure in static conditions.

[0051] S4. Determine the absolute value of each data point in Q, |q|. i | and the size of A; if |q i If |≥A, then |q i |Corresponding monitoring data S i Replace with S i The average of the previous p monitoring data points yields a set of data M after removing external loads and burrs. i Otherwise, the |q i|Corresponding monitoring data S i No replacement is needed.

[0052] In this embodiment, p can be set to 5. i The average value S of the previous 5 monitoring data i * represents:

[0053]

[0054] Among them, the corresponding monitoring data S i Replacement S i *Continue participating in other steps in S4 | q i | The calculation process for determining the size of A.

[0055] Of course, in other embodiments, p can also be set to other values ​​not less than 2, depending on factors such as the sampling frequency of the sensor and the duration of the preset time period T.

[0056] S5. Remove the external load from each data M i With the corresponding serial number N i By performing fitting, the fitting curve Y under the influence of external load is obtained. m .

[0057] S6. The monitoring data not collected in the preset time period T after the current time t0 of the sensor are numbered by referring to the monitoring data in the previous preset time period T, and these numbers are substituted into the fitting curve Y. m Thus, the predicted center curve Y of the sensor's monitoring data for the preset time period T is obtained. f , with Y f The dynamic threshold range of the civil structure sensor monitoring data is generated around the center.

[0058] For example, if there are 100 monitoring data points in the previous preset time period T, numbered from 1 to 100, then the 100 monitoring data points in the next preset time period T will be numbered from 101 to 200. That is, 101, 102, 103...200 are substituted into Y sequentially. m In the middle, a curve Y is obtained. f .

[0059] Among them, by using the central curve Y f Centered on the data, shift upwards by a value |H1| and downwards by a value |H2| to obtain the dynamic threshold interval H of the civil structure sensor monitoring data. The calculation formula is as follows:

[0060] H = [Y] f +|H1|,Y f -|H2|]

[0061]

[0062]

[0063] In the formula, j = 1, 2, ..., 7. In the first 7 days, Q... dxj For all data q after removing the trend term on day j i The maximum value in Q dnj For all data q after removing the trend term on day j i The minimum value in.

[0064] λ and β are the adjustment coefficients for the upper and lower limits of the threshold, respectively. In some embodiments, λ and β can be calculated using the following formulas:

[0065]

[0066]

[0067] Q dxa Q represents the maximum value of all values ​​after removing the trend term from the previous 7 days. dx The average value is equal to (Q) dx1 +Q dx2 +...+Q dx7 ) / 7.

[0068] Q dna Q is the minimum value of all values ​​after removing the trend term from the previous 7 days. dn The average value is equal to (Q) dn1 +Q dn2 +...+Q dn7 ) / 7.

[0069] In this embodiment, the statistical range of the maximum and minimum values ​​is 7 days. Of course, in other embodiments, other numbers of days can be adaptively set according to the season, climate, geography, and application scenario of the civil structure where the sensor is located, so as to make the dynamic threshold setting method more universal.

[0070] 2. Compare the monitoring data collected by the sensor over the next 10 minutes with the dynamic threshold range. An alarm is triggered when the monitoring data exceeds the upper or lower limit. Specific judgment conditions are as follows:

[0071] (1) When S i >Y f +|H1| triggers the upper limit alarm;

[0072] (2) When Y f -|H2|≤S i ≤Y f +|H1| indicates that the monitoring data is within a reasonable range;

[0073] (3) When Si <Y f -|H2| triggers a lower limit alarm.

[0074] Repeat the above steps to calculate the threshold interval every 10 minutes.

[0075] Please see Figure 4 As can be seen from the figure, during a certain period (21:20 to 21:50), although some of the monitored raw strain data fluctuated, they were still within the upper and lower limits of the dynamic threshold range. That is, all the monitoring data during this period were in a reasonable range and did not trigger an alarm.

[0076] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for setting dynamic thresholds for civil structure monitoring data, characterized in that, Including the following steps: S1. Collect all monitoring data from the sensor at the current time t0 before the preset time period T. S i The monitoring data were arranged in chronological order of collection time. S i Corresponding serial number N i By fitting it, a fitting curve is obtained. Y n ; S2. Sequentially process each monitoring data. S i With fitted curve Y n Data corresponding to the serial number Y n（i） The difference is calculated to obtain a set of data after removing the trend term. Q ={ q 1, q 2,…, q K }, and take the absolute value | Q |;Among them, i=1,2,…, K ; S3. Calculation | Q | is a normal distribution. Select the data in the upper and lower quartiles of the normal distribution and calculate the mean. Use this mean as the precise value A for the difference substitution. S4. Judge sequentially Q The absolute value of each data point | q i |and the size of A; if| q i If |≥A, then the corresponding monitoring data S i Replace with S i The average of the previous p monitoring data points is used to obtain a set of data after removing external loads. M i p≥2; S i The average value of the previous p monitoring data S i Represented as: Among them, the corresponding monitoring data S i Replacement S i Continue participating in other steps in S4 | q i | The calculation process for determining the size of A; S5. Set the corresponding serial number N i Data with external load removed M i By performing fitting, a fitting curve unaffected by external loads is obtained. Y m ; S6. The monitoring data not collected in the next preset time period T after the current time t0 of the sensor are numbered by referring to the monitoring data in the previous preset time period T, and these numbers are substituted into the fitting curve. Y m Thus, the predicted center curve of the sensor's monitoring data for the preset time period T is obtained. Y f ,by Y f The dynamic threshold range of the civil structure sensor monitoring data is generated centered on the curve; Y f Centered on, shift upwards by one value | H 1|, shift down by one value| H 2|, thus obtaining the dynamic threshold interval H of the civil structure sensor monitoring data, the calculation formula is as follows: In the formula, r is the preset number of days, j=1,2,…,r; in the first r days, Q dxj For all data after removing the trend term on day j q i The maximum value in, Q dnj For all data after removing the trend term on day j q i The minimum value in; Q dxa For the previous r days, remove the trend term and select all maximum values. Q dx The average value; Q dna The minimum values ​​after removing the trend term from the previous r days. Q dn The average value; λ and β These are the adjustment coefficients for the upper and lower limits of the threshold, respectively.

2. The method for setting dynamic thresholds for civil structure monitoring data according to claim 1, characterized in that, In step S6, the preset number of days r is 5 to 10 days.

3. The method for setting dynamic thresholds for civil structure monitoring data according to claim 1, characterized in that, In step S2, the data obtained each time after removing the trend term is also processed. Q Store the data, and sequentially generate daily data with trends removed. Q d ; And calculate Q d Mean μ and standard deviation δ ,when |S i - μ |> 3 δ At that time, the corresponding S i It is not used for calculating the maximum and minimum values ​​in step S6.

4. The method for setting dynamic thresholds for civil structure monitoring data according to claim 1, characterized in that, In step S4, S i It is obtained from the average of the previous p monitoring data, where 5 ≤ p ≤ 10.

5. The method for setting dynamic thresholds for civil structure monitoring data according to claim 1, characterized in that, The preset time period T has a duration of 10~60 minutes.

6. The method for setting dynamic thresholds for civil structure monitoring data according to claim 1, characterized in that, In step S1, the collected sensor monitoring data includes temperature effects and response data caused by external loads.

7. A safety early warning method for civil structure monitoring data, characterized in that, Including the following steps:

1. Real-time setting of dynamic threshold range for future preset time period T of civil structure sensor monitoring data; wherein, the dynamic threshold range is set using the method for setting dynamic threshold of civil structure monitoring data as described in any one of claims 1 to 6; 2. The monitoring data collected by the sensor during a preset time period T is compared with the dynamic threshold range. An alarm is generated when the monitoring data exceeds the upper or lower limit.

Images