Method and device for constructing multi-dimensional relationship characteristics of bridge deflection and temperature

Abstract

The application provides a bridge deflection and temperature multi-dimensional correlation relationship feature construction method and device, wherein the method comprises: obtaining environmental temperature monitoring data and bridge deflection monitoring data, preprocessing the bridge deflection monitoring data to obtain temperature-induced deflection data; time-synchronizing the environmental temperature monitoring data and the temperature-induced deflection data to obtain a synchronous temperature sequence and a synchronous temperature-induced deflection sequence; determining the multi-dimensional relationship features of the environmental temperature and the bridge deflection according to the synchronous temperature sequence and the synchronous temperature-induced deflection sequence, which include: temperature rise features and temperature drop features, historical lag features, basic temperature nonlinear features, temperature difference intensity nonlinear features, and environmental daily cycle features. The application can fully utilize the multi-dimensional relationship features, comprehensively integrate key characteristics such as temperature time sequence evolution, cross-period time lag effect and thermodynamic asymmetric response, and provide high-quality and high-robustness data support for bridge temperature effect prediction models or structure damage identification models.

Description

Technical Field

[0001] This invention relates to the field of bridge health monitoring technology, specifically to a method and apparatus for constructing multidimensional correlation features between bridge deflection and temperature. Background Technology

[0002] As the core of transportation infrastructure, the structural stability of bridges directly affects their service life and traffic safety. During long-term service, bridge structures are subjected to the coupled effects of various external loads, among which changes in bridge structural response caused by ambient temperature are dominant. For example, the daily variation in vertical deflection caused by ambient temperature far exceeds the deflection variation caused by vehicle loads. Vertical deflection is an important indicator for judging the strength and stiffness anomalies of bridge structures. Bridge structural materials exhibit shrinkage characteristics affected by ambient temperature, which is an important characteristic indicator for detecting structural damage during long-term operation. However, the constituent materials of bridge structures are variable, and the structural temperature field is non-uniform, making it difficult to accurately characterize the relationship between ambient temperature and the structural response (deflection and strain) of the main girder using a linear model.

[0003] Existing bridge temperature effect prediction models or structural damage identification models generally rely only on a single temperature time series as input features, failing to consider the evolution history of cross-sectional temperature gradients and the cumulative effect of thermal inertia.

[0004] Therefore, how to more accurately characterize the complex relationship between ambient temperature and bridge structural response, including obvious nonlinearity, asymmetry, and historical time delay (thermal inertia), is an urgent problem to be solved in order to provide reasonable characteristic parameters for temperature prediction models or structural damage models. Summary of the Invention

[0005] To address the aforementioned problems, the present invention aims to provide a method and apparatus for constructing a multidimensional correlation feature between bridge deflection and temperature.

[0006] In a first aspect, embodiments of the present invention provide a method for constructing a multidimensional correlation feature between bridge deflection and temperature, including: Ambient temperature monitoring data and bridge deflection monitoring data are acquired, and the bridge deflection monitoring data is preprocessed to obtain temperature-induced deflection data. The ambient temperature monitoring data and the temperature-induced deflection data are synchronized in time to obtain a synchronized temperature sequence and a synchronized temperature-induced deflection sequence. Based on the synchronous temperature sequence and synchronous temperature-induced deflection sequence, a multidimensional relationship feature between ambient temperature and bridge deflection is determined. The multidimensional relationship feature includes: heating and cooling features, historical lag features, nonlinear features of base temperature, nonlinear features of temperature difference intensity, and environmental diurnal cycle features. The multidimensional relationship feature is constructed as a feature vector, which is used as input to the bridge temperature effect prediction model or structural damage identification model.

[0007] In some optional implementations, the method further includes: A preset incremental time window is defined, and the temperature increment and temperature-induced deflection increment corresponding to the incremental time window are determined based on the synchronized temperature sequence and the synchronized temperature-induced deflection sequence; the temperature increment and temperature-induced deflection increment are used to determine the multidimensional relationship features.

[0008] In some optional implementations, the temperature increment and the temperature-induced deflection increment are defined as follows:

[0009]

[0010]

[0011] in, for Temperature increment over time For the synchronous temperature sequence The synchronized temperature value at any given time. The number of sampling steps corresponding to the incremental time window. for Temperature-induced deflection increment at time t, In the synchronous temperature-induced deflection sequence The synchronous temperature-induced deflection value at any given time.

[0012] In some optional implementations, the heating characteristic and the cooling characteristic are:

[0013]

[0014]

[0015] in, As a characteristic of temperature rise, As a cooling characteristic, This is the temperature increment.

[0016] In some optional implementations, the historical lag characteristics include the temperature increment characteristics of the previous time step and the temperature-induced deflection increment characteristics of the previous time step, specifically defined as:

[0017]

[0018]

[0019] This represents the temperature increment from the previous moment. The temperature-induced deflection increment at the previous moment.

[0020] In some optional implementations, the base temperature nonlinearity characteristic includes a reference temperature characteristic and a reference temperature quadratic characteristic, specifically defined as:

[0021]

[0022]

[0023] in, As a reference temperature characteristic, The characteristic is the square root of the reference temperature. The number of sampling steps corresponding to the incremental time window. For the synchronous temperature sequence The synchronized temperature value at any given time; The nonlinear characteristics of the temperature difference intensity include the absolute value characteristic of the temperature difference and the quadratic characteristic of the temperature difference, specifically defined as:

[0024]

[0025]

[0026] in, Characterized by the absolute value of the temperature difference. It exhibits a quadratic characteristic of temperature difference. This is the temperature increment.

[0027] In some optional implementations, the environmental diurnal cycle characteristic is:

[0028]

[0029]

[0030] in, for The daily periodic sine component corresponding to the time. for The daily cosine component corresponding to the time. express The hour corresponding to the given time.

[0031] Secondly, embodiments of the present invention also provide a device for constructing a multidimensional correlation feature between bridge deflection and temperature, comprising: Data acquisition and preprocessing module: used to acquire ambient temperature monitoring data and bridge deflection monitoring data, and preprocess the bridge deflection monitoring data to obtain temperature-induced deflection data; Time synchronization module: used to synchronize the ambient temperature monitoring data and the temperature-induced deflection data in time to obtain a synchronized temperature sequence and a synchronized temperature-induced deflection sequence; Feature construction module: used to determine the multidimensional relationship features between ambient temperature and bridge deflection based on the synchronous temperature sequence and synchronous temperature-induced deflection sequence. The multidimensional relationship features include: heating feature and cooling term feature, historical lag feature, base temperature nonlinearity feature, temperature difference intensity nonlinearity feature, and environmental diurnal cycle feature.

[0032] Thirdly, embodiments of the present invention also provide a computer storage medium storing computer-executable instructions for executing the bridge deflection and temperature multidimensional correlation feature construction method described in the first aspect above.

[0033] Fourthly, embodiments of the present invention also provide an electronic device, comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the bridge deflection and temperature multidimensional correlation feature construction method described in the first aspect above.

[0034] In the solution provided by the first aspect of this invention, key characteristics such as temperature time-series evolution, cross-time-period delay effects, and thermodynamic asymmetric response are comprehensively integrated through the construction of multi-dimensional relationship features. This enables a deep reconstruction of the complex physical processes of bridge temperature response, including nonlinearity, asymmetry, and historical time delay, effectively compensating for the shortcomings of traditional features being singular and lacking physical information. It significantly improves the completeness and accuracy of feature representation, effectively weakens the impact of noise interference and abnormal fluctuations, enhances the stability and reliability of model input, and provides high-quality, robust data support for downstream temperature prediction models or structural damage models.

[0035] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 A flowchart illustrating the method for constructing the multidimensional correlation features between bridge deflection and temperature according to an embodiment of the present invention is shown. Figure 2 A comparison chart of the original temperature and the original deflection is shown; Figure 3 The results of temperature-induced deflection extraction are shown in the figure. Figure 4 A comparison graph of temperature and temperature-induced deflection after alignment is shown; Figure 5 A graph showing the correlation between temperature increment and temperature-induced deflection increment at different time intervals is presented. Figure 6 A schematic diagram showing the heating and cooling characteristics is provided. Figure 7 A schematic diagram illustrating the characteristics of historical lag is shown; Figure 8 A schematic diagram illustrating the nonlinear characteristics of the reference temperature is shown. Figure 9 A schematic diagram illustrating the nonlinear characteristics of temperature difference intensity is shown. Figure 10 A schematic diagram illustrating the diurnal characteristics of the environment is shown; Figure 11 The results of the effective sample screening are shown in the figure; Figure 12 A block diagram of a bridge deflection and temperature multidimensional correlation feature construction device according to an embodiment of the present invention is shown; Figure 13 A schematic diagram of the structure of the electronic device provided in an embodiment of the present invention is shown. Detailed Implementation

[0038] In the description of this invention, 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," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention 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. Therefore, they should not be construed as limitations on this invention.

[0039] Furthermore, the terms "first" and "second" 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 "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0040] In this invention, unless otherwise explicitly 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 connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0041] This invention provides a method for constructing a multidimensional correlation feature between bridge deflection and temperature. (See also...) Figure 1 As shown, it includes: Step 101: Obtain ambient temperature monitoring data and bridge deflection monitoring data, and preprocess the bridge deflection monitoring data to obtain temperature-induced deflection data; By installing real-time monitoring equipment on the bridge structure, ambient temperature monitoring data and bridge deflection monitoring data are collected.

[0042] Both the ambient temperature monitoring data and the bridge deflection monitoring data are detection sequences. For example, the original ambient temperature monitoring data T can be defined as:

[0043]

[0044] The original bridge deflection monitoring data D is:

[0045]

[0046] in, and These represent the sampling times for temperature and bridge deflection, respectively. and These represent the corresponding temperature value and bridge deflection value, respectively. and These represent the number of temperature and bridge deflection data collected, respectively.

[0047] For the bridge deflection monitoring data D, preprocessing is performed to extract the temperature-induced deflection, thereby forming a corresponding temperature-induced deflection sequence, i.e., temperature-induced deflection data. For example, wavelet decomposition algorithm can be used to extract temperature-induced deflection data from the bridge deflection monitoring data D. .

[0048] Step 102: Synchronize the ambient temperature monitoring data and the temperature-induced deflection data in time to obtain a synchronized temperature sequence and a synchronized temperature-induced deflection sequence; In this embodiment, for ambient temperature monitoring data and temperature-induced deflection data, the common time interval between the two sequences can be determined. Let the set of temperature sampling times be... The set of temperature-induced deflection sampling times is Take the start and end times corresponding to the intersection:

[0049]

[0050] in, This is the start time of the common time interval. This is the end time of the common time interval.

[0051] Within a public time period at fixed intervals Generate multiple time points at equal intervals :

[0052]

[0053] N is the number of sampling times.

[0054] In this embodiment, the ambient temperature and bridge deflection monitoring sequences can be synchronized using nearest neighbor matching. For each synchronization moment... Based on ambient temperature monitoring data T and temperature-induced deflection data, respectively In the middle, the sampling time and The closest data point is used as the value at that synchronization moment to obtain the synchronization temperature sequence. and synchronous temperature-induced deflection sequence If the synchronization time If the time interval between the two original sampling points is equal, then the earlier value can be uniformly specified to ensure the causality of the data and avoid introducing future information.

[0055] Step 103: Based on the synchronous temperature sequence and the synchronous temperature-induced deflection sequence, determine the multidimensional relationship characteristics between ambient temperature and bridge deflection. The multidimensional relationship characteristics include: heating and cooling characteristics, historical lag characteristics, nonlinear characteristics of base temperature, nonlinear characteristics of temperature difference intensity, and environmental diurnal cycle characteristics. The multidimensional relationship characteristics are constructed as inputs to the bridge temperature effect prediction model or structural damage identification model.

[0056] To more accurately characterize the complex relationship between ambient temperature and main beam deflection, which exhibits significant nonlinearity, asymmetry, and historical time delay (thermal inertia), a multidimensional feature of ambient temperature and main beam deflection is constructed.

[0057] Among them, the heating and cooling characteristics can be obtained by separating the temperature increment through a rectification function, and are used to characterize the thermodynamic asymmetric response; the historical lag characteristics can include the temperature increment and the temperature-induced deflection increment of the previous moment, and are used to characterize the thermal inertia lag effect; the base temperature nonlinear characteristics can be calculated from the synchronous temperature sequence, and are used to characterize the difference in thermal expansion coefficients between different absolute temperature ranges; the temperature difference intensity nonlinear characteristics can be calculated from the temperature increment, and are used to characterize the nonlinear influence of the temperature difference amplitude; the environmental diurnal cycle characteristics can be calculated from the synchronous moment, and are used to characterize the influence of diurnal cycle changes on the bridge temperature response.

[0058] After determining the multidimensional relationship features, they can be input into the bridge temperature effect prediction model or the structural damage identification model to obtain the model processing results, such as the prediction results of the bridge temperature effect or the identification results of the bridge structural damage, for downstream use.

[0059] The method for constructing multidimensional correlation features between bridge deflection and temperature provided in this embodiment comprehensively integrates key characteristics such as temperature time-series evolution, cross-time-period time delay effects, and thermodynamic asymmetric responses through the constructed multidimensional relationship features. It can deeply reconstruct the complex physical processes of bridge temperature response, including nonlinearity, asymmetry, and historical time delay, effectively compensating for the shortcomings of traditional features being singular and lacking physical information. This significantly improves the completeness and accuracy of feature representation, effectively weakens the influence of noise interference and abnormal fluctuations, enhances the stability and reliability of model input, and provides high-quality, robust data support for downstream temperature prediction models or structural damage models.

[0060] In some optional implementations, the method further includes: setting an incremental time window, and determining the temperature increment and temperature-induced deflection increment corresponding to the incremental time window based on the synchronous temperature sequence and the synchronous temperature-induced deflection sequence; the temperature increment and temperature-induced deflection increment are used to determine the multidimensional relationship features.

[0061] In this embodiment, the incremental time window can be determined based on the sampling interval. Specifically, as shown above, the sampling interval is... Length of the incremental time window It can be set as: Then the bridge is The temperature increment and temperature-induced deflection increment at time t are defined as follows:

[0062]

[0063]

[0064] in, for Temperature increment over time For the synchronous temperature sequence The synchronized temperature value at time k, where k is the index of the current time and also corresponds to the index of the end time of the incremental time window. This represents the number of sampling steps corresponding to the incremental time window. This is the index of the start time of the incremental time window. for Temperature-induced deflection increment at time t, In the synchronous temperature-induced deflection sequence The synchronous temperature-induced deflection value at any given time.

[0065] In this embodiment, n is the number of sampling steps corresponding to the incremental time window, which can be a preset value, such as n=1, 3, 5, etc., depending on the actual situation. For example, If n=12, then .

[0066] In this embodiment, an incremental time window is introduced to define the temperature increment and the temperature-induced deflection increment. Bridge temperature-induced deflection is driven by the amount of temperature change rather than absolute temperature; the incremental form of the temperature-induced deflection increment can directly reflect the core driving force of thermodynamic changes. Structural temperature changes lag behind ambient temperature, and the correlation between instantaneous temperature difference and deflection is extremely low. By using an incremental time window of appropriate length, the bridge's thermal inertia time constant can be effectively covered, improving the correlation between temperature and deflection. Furthermore, incremental calculation can automatically eliminate long-term trend terms and baseline drift in the temperature and deflection sequences, filter low-frequency noise from the monitoring system, and provide clean basic data for subsequent temperature rise / fall separation and the construction of nonlinear characteristics.

[0067] In some optional implementations, the heating and cooling characteristics are as follows:

[0068]

[0069]

[0070] in, As a characteristic of temperature rise, As a cooling characteristic, This is the temperature increment.

[0071] In this implementation, the heating and cooling characteristics can be calculated separately using the rectification function. Only extract the change in temperature during the rise. Only the change in temperature is extracted. Because the thermal response (such as deflection change) of bridge structures during heating and cooling is often asymmetric, it can be used to distinguish the different effects of heating and cooling on deflection in bridge temperature-induced deflection modeling.

[0072] In some optional implementations, the historical lag characteristics include the temperature increment characteristics of the previous time step and the temperature-induced deflection increment characteristics of the previous time step, specifically defined as:

[0073]

[0074]

[0075] This represents the temperature increment at the previous time step (i.e., time step k-1). The temperature-induced deflection increment at the previous moment; this is understandable. That is, the temperature increment at time k-1, and , That is, the temperature-induced deflection increment at time k-1, and .

[0076] In this embodiment, the historical lag characteristic can reflect the memory effect of the bridge's thermal response.

[0077] In some optional implementations, the base temperature nonlinearity characteristic includes a reference temperature characteristic and a reference temperature quadratic characteristic, specifically defined as:

[0078]

[0079]

[0080] in, As a reference temperature characteristic, The characteristic is the square root of the reference temperature. The number of sampling steps corresponding to the incremental time window; For the synchronous temperature sequence The synchronous temperature value at a given time is used as a reference temperature characteristic. .

[0081] The thermal response of a bridge is related not only to the temperature difference but also to the base temperature at which the temperature change occurs. In order to characterize the difference in the thermal expansion coefficient of a bridge in different absolute temperature ranges, the reference temperature and its square were calculated as features.

[0082] Furthermore, the nonlinear characteristics of the temperature difference intensity include the absolute value characteristic of the temperature difference and the quadratic characteristic of the temperature difference, specifically defined as:

[0083]

[0084]

[0085] in, Characterized by the absolute value of the temperature difference. It exhibits a quadratic characteristic of temperature difference. This is the temperature increment.

[0086] In this embodiment, to reflect the magnitude of the temperature difference and its nonlinear effects, the absolute value and the square of the temperature difference are calculated as features.

[0087] In some optional implementations, the environmental diurnal cycle characteristic is:

[0088]

[0089]

[0090] in, for The daily periodic sine component corresponding to the time. for The daily cosine component corresponding to the time. express The clock hour corresponding to the time (the value ranges from 0 to 23).

[0091] in, It is a dimensionless real number representing the precise location of the current synchronization moment within a 24-hour period, expressed in hours, and ranging from [0, 24]. Specifically, From synchronization time The formula is derived from this and is as follows: =From The number of hours extracted. For example... For example, 9:45, the converted hour is 9.75, that is... .

[0092] Bridge temperature response typically exhibits a distinct diurnal cycle; periodic coding is used to calculate time. The corresponding time characteristics. Furthermore, sine and cosine coding is used instead of directly using... As a time feature, it can eliminate the time wrap-around problem caused by abrupt changes in values ​​around midnight (such as a sudden change from 23.9 to 0), and can ensure the continuity and periodicity of the time feature.

[0093] In some optional implementations, the feature vector corresponding to the multidimensional relation features... Specifically, it can be expressed as:

[0094]

[0095]

[0096] The features of each dimension can be found in the relevant descriptions above, and will not be repeated here.

[0097] Optionally, to ensure the validity of the input feature samples, the following valid sample discrimination rule can be defined: if a feature has missing values ​​at a certain time, or if the absolute value of the temperature difference at that time is less than a preset threshold, then the sample is not included in the prediction model. The valid sample discrimination formula is as follows:

[0098]

[0099] Among them, when Time indicates the first One sample was valid and was retained. Indicates the first Multidimensional relational features of a sample, when This indicates that the sample is invalid. For a preset temperature difference screening threshold (e.g., ).

[0100] In this embodiment, by constructing 11 multidimensional relationship features across five categories—heating and cooling characteristics, historical lag characteristics, base temperature nonlinearity characteristics, temperature difference intensity nonlinearity characteristics, and environmental diurnal cycle characteristics—these features can be directly used as standard inputs for subsequent bridge temperature effect prediction models and structural damage identification models. This allows for accurate reproduction of complex responses such as asymmetric heating and cooling and thermal inertia time lags, reducing prediction errors and achieving high-precision temperature-induced deflection prediction. Subsequently, damage identification using the prediction residuals of the health state benchmark model can significantly reduce false alarm and false negative rates. This scheme can provide data support for bridge stiffness degradation analysis, remaining life prediction, and maintenance decisions.

[0101] For ease of understanding, the present invention will be further described in detail below with reference to specific embodiments.

[0102] Taking the bridge inspection and health project of a prefabricated T-beam bridge as an example. The bridge is 167 meters long, with a span combination of... Meters, one span of the main beam was selected as the monitoring object. Monitoring system configuration: Temperature sensor acquisition frequency is [missing information]. The deflection sensor, with a sampling frequency of 1 Hz, is installed at the mid-span of the main beam to monitor the deflection response. Measured data from one week of operation on the bridge are used to describe the specific implementation steps of this invention in detail.

[0103] Obtain the raw temperature monitoring sequence of the bridge monitoring system and deflection monitoring sequence During this period, a total of 973,245 monitoring points were used for the bridge's original deflection, and 2,016 monitoring points were used for temperature. A comparison of the time histories of the original temperature and original deflection data is provided below. Figure 2 As shown.

[0104] The wavelet decomposition algorithm is used to extract the temperature-induced deflection sequence from the original deflection monitoring sequence. , specifically Figure 3 As shown.

[0105] In this case, the temperature sensor sampling frequency is: Hz; the deflection sensor collects data at fixed intervals of 1Hz. Generate synchronization time sequence For each synchronization moment The corresponding data points with the closest sampling times are selected from the temperature sequence and the temperature-induced deflection sequence, respectively, as the synchronization values ​​for that time, thus obtaining the synchronization temperature sequence. and synchronous temperature-induced deflection sequence .

[0106] After synchronization and alignment, the result is obtained One sample. Alignment comparison of temperature and temperature-induced deflection, for example... Figure 4 As shown.

[0107] Eight candidate incremental time windows were preset for comparative testing, with window lengths set to... For each candidate incremental time window The Pearson correlation coefficients for the temperature increment and the temperature-induced deflection increment were calculated separately, and the results are as follows: Figure 5 As shown. The results indicate that when the incremental window... As the time interval was gradually increased from 10 min to 120 min, the correlation coefficient between the temperature increment and the temperature-induced deflection increment showed a significant upward trend, and at... The correlation coefficient reached its peak at [location], with the Pearson correlation coefficient exceeding 0.85.

[0108] Based on the above comparison, this embodiment ultimately selects... As the optimal incremental time window. Corresponding to a fixed... The synchronous sampling interval, this incremental time window is equivalent to 24 sampling steps. The temperature increment and temperature-induced deflection increment for the entire time period are calculated using this incremental time window, which serve as the basic input for subsequent multidimensional relationship features.

[0109] Based on the measured data of the bridge in the case study over one week, the specific steps of this invention are used to construct multidimensional characteristics of ambient temperature and main beam deflection, as follows: Figure 6 (including temperature rise characteristics) With cooling characteristics ), Figure 7 (Including the temperature increment characteristics of the previous moment) Characteristics of temperature-induced deflection increment at the previous time step ), Figure 8 (including reference temperature characteristics) and the quadratic characteristic of the reference temperature ), Figure 9 (including absolute temperature difference characteristics) and the quadratic characteristics of temperature difference ), Figure 10 (including environmental daily cycle characteristics) , As shown in the figure.

[0110] After completing the construction of multidimensional features, select The temperature difference threshold was used for screening. Invalid time periods with extremely small temperature differences were removed. The original 1992 synchronous samples were ultimately transformed into 1833 sets of high-quality multidimensional continuous incremental feature vectors. These feature matrices were directly used as the standard input for subsequent nonlinear prediction models. The results of the effective sample screening are as follows: Figure 11 As shown.

[0111] The method for constructing multidimensional correlation features between bridge deflection and temperature provided in this embodiment, by building a multidimensional extended feature system, can comprehensively integrate key characteristics such as temperature time-series evolution, cross-time-period time delay effects, and thermodynamic asymmetric responses. It can deeply reconstruct the complex physical processes of bridge temperature response, such as nonlinearity and thermal inertia, effectively overcoming the shortcomings of traditional single features and missing physical information. By deeply reconstructing complex physical processes such as thermal inertial hysteresis effects and spatial distribution differences, it can accurately reproduce the intrinsic correlation between structural temperature and temperature-induced deflection, fully revealing the thermodynamic mechanism of bridge temperature effects. Furthermore, it can significantly improve the completeness and accuracy of feature representation, effectively weaken the influence of noise interference and abnormal fluctuations, enhance the stability and reliability of model input, and provide high-quality, robust data support for downstream temperature effect prediction models.

[0112] The above details the method and process for constructing the multidimensional correlation characteristics between bridge deflection and temperature. This method can also be implemented using a corresponding device, the structure and function of which will be described in detail below.

[0113] Based on the same inventive concept, embodiments of the present invention also provide a device for constructing multidimensional correlation features between bridge deflection and temperature, see [link to related document]. Figure 12 As shown, the device includes: Data acquisition and preprocessing module 1201: used to acquire ambient temperature monitoring data and bridge deflection monitoring data, and preprocess the bridge deflection monitoring data to obtain temperature-induced deflection data; Time synchronization module 1202: used to synchronize the ambient temperature monitoring data and the temperature-induced deflection data in time to obtain a synchronized temperature sequence and a synchronized temperature-induced deflection sequence; Feature construction module 1203: used to determine the multidimensional relationship features between ambient temperature and bridge deflection based on the synchronous temperature sequence and the synchronous temperature-induced deflection sequence. The multidimensional relationship features include: heating feature and cooling term feature, historical lag feature, base temperature nonlinear feature, temperature difference intensity nonlinear feature, and environmental diurnal cycle feature. The multidimensional relationship features are used as input to the bridge temperature effect prediction model or structural damage identification model.

[0114] In some optional implementations, the feature construction module 1203 is further configured to: A preset incremental time window is defined, and the temperature increment and temperature-induced deflection increment corresponding to the incremental time window are determined based on the synchronized temperature sequence and the synchronized temperature-induced deflection sequence; the temperature increment and temperature-induced deflection increment are used to determine the multidimensional relationship features.

[0115] In some optional implementations, the temperature increment and the temperature-induced deflection increment are defined as follows:

[0116]

[0117]

[0118] in, for Temperature increment over time For the synchronous temperature sequence The synchronized temperature value at any given time. The number of sampling steps corresponding to the incremental time window. for Temperature-induced deflection increment at time t, In the synchronous temperature-induced deflection sequence The synchronous temperature-induced deflection value at any given time.

[0119] In some optional implementations, the heating characteristic and the cooling characteristic are:

[0120]

[0121]

[0122] in, As a characteristic of temperature rise, As a cooling characteristic, This is the temperature increment.

[0123] In some optional implementations, the historical lag characteristics include the temperature increment characteristics of the previous time step and the temperature-induced deflection increment characteristics of the previous time step, specifically defined as:

[0124]

[0125]

[0126] This represents the temperature increment from the previous moment. The temperature-induced deflection increment at the previous moment.

[0127] In some optional implementations, the base temperature nonlinearity characteristic includes a reference temperature characteristic and a reference temperature quadratic characteristic, specifically defined as:

[0128]

[0129]

[0130] in, As a reference temperature characteristic, The characteristic is the square root of the reference temperature. The number of sampling steps corresponding to the incremental time window. For the synchronous temperature sequence The synchronized temperature value at any given time; The nonlinear characteristics of the temperature difference intensity include the absolute value characteristic of the temperature difference and the quadratic characteristic of the temperature difference, specifically defined as:

[0131]

[0132]

[0133] in, Characterized by the absolute value of the temperature difference. It exhibits a quadratic characteristic of temperature difference. This is the temperature increment.

[0134] In some optional implementations, the environmental diurnal cycle characteristic is:

[0135]

[0136]

[0137] in, for The daily periodic sine component corresponding to the time. for The daily cosine component corresponding to the time. express The hour corresponding to the given time.

[0138] This invention also provides a computer storage medium storing computer-executable instructions, including a program for performing the above-described screening method for Alzheimer's disease and related cognitive impairments. The computer-executable instructions can execute the methods in any of the above-described method embodiments.

[0139] The computer storage medium can be any available medium or data storage device that a computer can access, including but not limited to magnetic storage (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical storage (e.g., CD, DVD, BD, HVD), and semiconductor storage (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0140] Figure 13 A structural block diagram of an electronic device according to another embodiment of the present invention is shown. The electronic device 1100 may be a host server with computing capabilities, a personal computer (PC), or a portable computer or terminal, etc. The specific embodiments of the present invention do not limit the specific implementation of the electronic device.

[0141] The electronic device 1100 includes at least one processor 1110, a communications interface 1120, a memory array 1130, and a bus 1140. The processor 1110, the communications interface 1120, and the memory 1130 communicate with each other via the bus 1140.

[0142] The communication interface 1120 is used to communicate with network elements, including, for example, virtual machine management centers and shared storage.

[0143] Processor 1110 is used to execute programs. Processor 1110 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0144] Memory 1130 is used for executable instructions. Memory 1130 may include high-speed RAM and may also include non-volatile memory, such as at least one disk storage device. Memory 1130 may also be a memory array. Memory 1130 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules. The instructions stored in memory 1130 can be executed by processor 1110 to enable processor 1110 to perform the screening method for Alzheimer's disease and related cognitive impairment in any of the above method embodiments.

[0145] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for constructing a multidimensional correlation feature between bridge deflection and temperature, characterized in that, include: Ambient temperature monitoring data and bridge deflection monitoring data are acquired, and the bridge deflection monitoring data is preprocessed to obtain temperature-induced deflection data. The ambient temperature monitoring data and the temperature-induced deflection data are synchronized in time to obtain a synchronized temperature sequence and a synchronized temperature-induced deflection sequence. Based on the synchronous temperature sequence and synchronous temperature-induced deflection sequence, the multidimensional relationship characteristics between ambient temperature and bridge deflection are determined. These multidimensional relationship characteristics include: heating and cooling characteristics, historical lag characteristics, nonlinear characteristics of base temperature, nonlinear characteristics of temperature difference intensity, and environmental diurnal cycle characteristics. The multidimensional relationship features are used as inputs to bridge temperature effect prediction models or structural damage identification models.

2. The method according to claim 1, characterized in that, The method further includes: A preset incremental time window is defined, and the temperature increment and temperature-induced deflection increment corresponding to the incremental time window are determined based on the synchronized temperature sequence and the synchronized temperature-induced deflection sequence; the temperature increment and temperature-induced deflection increment are used to determine the multidimensional relationship features.

3. The method according to claim 2, characterized in that, The temperature increment and the temperature-induced deflection increment are defined as follows: in, for Temperature increment over time For the synchronous temperature sequence The synchronized temperature value at any given time. The number of sampling steps corresponding to the incremental time window. for Temperature-induced deflection increment at time t, In the synchronous temperature-induced deflection sequence The synchronous temperature-induced deflection value at any given time.

4. The method according to claim 2 or 3, characterized in that, The heating characteristic and the cooling characteristic are as follows: in, As a characteristic of temperature rise, As a cooling characteristic, This is the temperature increment.

5. The method according to claim 2 or 3, characterized in that, The historical lag characteristics include the temperature increment characteristics of the previous time step and the temperature-induced deflection increment characteristics of the previous time step, specifically defined as: This represents the temperature increment from the previous moment. The temperature-induced deflection increment at the previous moment.

6. The method according to claim 2 or 3, characterized in that, The nonlinear characteristics of the base temperature include the reference temperature characteristics and the quadratic characteristics of the reference temperature, specifically defined as follows: in, As a reference temperature characteristic, The characteristic is the square root of the reference temperature. The number of sampling steps corresponding to the incremental time window. For the synchronous temperature sequence The synchronized temperature value at any given time; The nonlinear characteristics of the temperature difference intensity include the absolute value characteristic of the temperature difference and the quadratic characteristic of the temperature difference, specifically defined as: in, Characterized by the absolute value of the temperature difference. It exhibits a quadratic characteristic of temperature difference. This is the temperature increment.

7. The method according to claim 2 or 3, characterized in that, The diurnal characteristic of the environment is as follows: in, for The daily periodic sine component corresponding to the time. for The daily cosine component corresponding to the time. express The hour corresponding to the given time.

8. A device for constructing a multidimensional correlation between bridge deflection and temperature, characterized in that, include: Data acquisition and preprocessing module: used to acquire ambient temperature monitoring data and bridge deflection monitoring data, and preprocess the bridge deflection monitoring data to obtain temperature-induced deflection data; Time synchronization module: used to synchronize the ambient temperature monitoring data and the temperature-induced deflection data in time to obtain a synchronized temperature sequence and a synchronized temperature-induced deflection sequence; Feature construction module: used to determine the multidimensional relationship features between ambient temperature and bridge deflection based on the synchronous temperature sequence and synchronous temperature-induced deflection sequence. The multidimensional relationship features include: heating feature and cooling term feature, historical lag feature, base temperature nonlinear feature, temperature difference intensity nonlinear feature, and environmental diurnal cycle feature. The multidimensional relationship features are used as inputs to bridge temperature effect prediction models or structural damage identification models.

9. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions, which are used to execute the method for constructing the multidimensional correlation feature of bridge deflection and temperature as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, include: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the bridge deflection and temperature multidimensional correlation feature construction method according to any one of claims 1 to 7.

Images