A method for maintaining and optimizing prefabricated parts for road and bridge construction in response to temperature changes

Abstract

The application relates to the field of road and bridge construction and maintenance, and discloses a prefabricated part maintenance optimization method for road and bridge construction suitable for temperature changes, which comprises the following steps: arranging a temperature sensing array after the prefabricated part is cast and formed to collect multi-point temperature time series data; performing space-time coupling interpolation on the multi-point temperature time series data based on a heat conduction partial differential equation to generate a prefabricated part three-dimensional transient temperature field; performing gradient operator operation on the three-dimensional transient temperature field to obtain a temperature gradient field and a temperature change rate field; identifying a temperature abnormal aggregation area according to the local vorticity characteristics of the temperature gradient field and determining a temperature violent fluctuation area according to the extreme value point position of the temperature change rate field; and adjusting the flow direction distribution coefficient and the flow size of the maintenance medium in each area for the temperature abnormal aggregation area and the temperature violent fluctuation area. The application realizes accurate identification and differential regulation of abnormal temperature areas in the prefabricated part maintenance process.

Description

Technical Field

[0001] This invention relates to the field of road and bridge construction and maintenance technology, specifically to an optimized method for the maintenance of prefabricated components used in road and bridge construction that adapts to temperature changes. Background Technology

[0002] In road and bridge construction, temperature control during the curing process of precast components directly affects concrete strength development, structural dimensional accuracy, and crack resistance. For large precast components, factors such as ambient temperature fluctuations and uneven distribution of hydration heat often lead to complex internal temperature gradients, forming localized abnormal areas. If a uniform curing medium supply method is used, it is difficult to eliminate these temperature anomalies. Existing curing methods often involve setting temperature sensors at predetermined locations to collect discrete point temperatures, and then adjusting the curing environment based on single-point temperature values ​​or simple thresholds. This method, based on discrete monitoring and global control, cannot reflect the spatial continuity and evolution of the internal temperature field of the precast component, resulting in insufficient accuracy in identifying temperature anomaly areas and difficulty in capturing potential quality risks caused by heat flow convergence or rapid temperature changes. Furthermore, existing technologies typically only uniformly adjust the temperature or overall flow rate of the curing medium, failing to implement differentiated flow direction and flow distribution for different areas. This leads to low utilization efficiency of curing resources, the potential persistence of temperature anomaly areas, and even the development of structural stress concentration and surface microcracks. To address the aforementioned issues, the technical challenges include: accurately extracting the spatial locations of temperature anomalies from transient temperature fields—anomalies characterized by the continuous accumulation of heat in localized areas that are difficult to detect using conventional temperature difference detection; and identifying areas of drastic temperature changes, which pose a risk of thermal shock damage due to rapid temperature fluctuations within a short period. Furthermore, after identifying these two types of anomaly areas, it is necessary to determine how to adjust the flow direction and volume of the maintenance medium according to the characteristics of each area to achieve refined maintenance optimization. Summary of the Invention

[0003] This invention provides an optimized method for the maintenance of precast components used in road and bridge construction that adapts to temperature changes. Based on multi-point temperature sensing data and partial differential equations of heat conduction, the three-dimensional transient temperature field of the precast component is reconstructed. By calculating the local curl characteristics of the temperature gradient field, regions with abnormal temperature accumulation are identified. Regions with drastic temperature fluctuations are extracted by identifying the extreme points of the temperature change rate field. The flow direction distribution coefficient and flow rate of the maintenance medium are adjusted for the above two types of regions, thereby achieving differentiated control of abnormal temperature regions during the maintenance of precast components and improving the uniformity and quality stability of maintenance.

[0004] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides an optimized method for the maintenance of precast components for road and bridge construction that adapts to temperature changes, comprising the following steps: After the precast component is cast, a temperature sensor array is deployed according to preset monitoring points to collect multi-point temperature time-series data of the surface and interior of the precast component; based on the partial differential equation of heat conduction, the multi-point temperature time-series data is subjected to spatial and temporal coupled interpolation processing to generate a three-dimensional transient temperature field of the precast component; gradient operator operation is performed on the three-dimensional transient temperature field to obtain the spatial distribution of the temperature gradient field, and the temperature change rate at each spatial location at adjacent sampling times is calculated to generate a temperature change rate field; the temperature anomaly accumulation area is identified according to the local curl characteristics of the temperature gradient field, and the temperature drastic fluctuation area is determined according to the extreme point position of the temperature change rate field; for the temperature anomaly accumulation area and the temperature drastic fluctuation area, the flow direction distribution coefficient and flow rate of the maintenance medium in each area are adjusted respectively, thereby realizing the fine control of the spatial distribution and temporal change of the precast component maintenance temperature.

[0005] As a preferred embodiment of the present invention, the coupled interpolation process specifically involves: using the partial differential equation of heat conduction as the governing equation, using the multi-point temperature time series data as discrete boundary conditions, discretizing the time and spatial coordinates using a seven-point difference scheme, solving for the discrete node temperature values ​​at each time step using the Gauss-Seidel iterative method to obtain the temperature time series of the overall three-dimensional mesh nodes of the preform, and then mapping the mesh node temperatures to arbitrary spatial locations using cubic spline interpolation to form a continuous three-dimensional transient temperature field. This process can utilize limited discrete measurement point information to accurately reconstruct the true temperature distribution inside and on the surface of the preform, providing a reliable data foundation for subsequent analysis.

[0006] Furthermore, the temperature gradient field is generated as follows: at each grid node of the three-dimensional transient temperature field, the partial temperature derivatives along the length, width, and height of the preform are calculated using the second-order central difference formula. The vector formed by the partial derivatives in the three directions is used as the temperature gradient vector of that node. After parallel calculation for all grid nodes, the node gradient vector is extended to the entire spatial domain through bilinear interpolation to form a continuous temperature gradient field. The gradient field obtained in this way can intuitively reflect the direction and intensity of heat transfer, making it easier to locate the risk of thermal stress concentration.

[0007] As another technical solution of the present invention, the identification process of the temperature anomaly accumulation region is as follows: The curl vector of each spatial location in the temperature gradient field is calculated, i.e., the vorticity is obtained by performing a cross product operation on the temperature gradient vector. A percentile threshold for the vorticity amplitude is set, and continuous spatial voxels with vorticity amplitudes exceeding the percentile threshold are marked as candidate anomaly accumulation regions. Then, isolated voxels are eliminated through connected component analysis, and the multiple connected voxels with the largest areas are retained as temperature anomaly accumulation regions. This method can automatically shield detection noise and accurately capture the core area of ​​heat accumulation inside the precast component, helping to prevent material damage induced by local high temperatures.

[0008] For determining regions of drastic temperature fluctuations, the preferred approach of this invention is to perform a morphological top-hat transformation on the temperature change rate field to extract raised regions with a change rate amplitude higher than the local background. The geometric center of each raised region is designated as an extremum point. A spherical space centered on the extremum point and with a preset radius as its diameter is defined as the region of drastic temperature fluctuations. If the spherical regions corresponding to different extremum points overlap, they are merged into a single fluctuation region. This accurately delineates the spatial range where temperature rises and falls most significantly over time, providing a clear target area for dynamic stabilization control.

[0009] For the identified abnormal accumulation and fluctuation areas, when adjusting the flow direction and flow rate of the curing medium, multiple sets of independently adjustable medium guide plates are pre-embedded in the curing cover layer of the precast component. Based on the relative azimuth angle between the center coordinates of each area and the outlet of the guide plate, the flow direction is changed by rotating the guide plate angle, ensuring that the number of guide plate outlets corresponding to each abnormal accumulation and fluctuation area is proportional to the volume of that area. Simultaneously, the arithmetic mean of the temperature gradient modulus of all grid nodes within each area is calculated. This mean is subtracted from the preset optimal gradient modulus to obtain the difference. A positive difference decreases the opening of the guide plate outlet, while a negative difference increases the opening. The adjustment amount is proportional to the absolute value of the difference, and the proportionality coefficient is segmented according to the distance of the area from the curing medium source. This spatially differentiated flow direction and flow rate distribution strategy can enhance heat dissipation in areas of concentrated heat and provide buffered insulation in areas of rapid temperature change, causing the temperature gradient at various points on the precast component to quickly converge to the ideal range. This effectively reduces internal stress caused by excessive temperature differences and improves the macroscopic quality of the precast component.

[0010] To stabilize the temperature level during the curing process, the method of this invention further includes: continuously adjusting the heating power of the curing medium using a proportional-integral-derivative (PID) controller based on the overall statistical distribution characteristics of the temperature change rate field. More preferably, the method involves statistically analyzing the rate of change values ​​at all spatial locations within the entire temperature change rate field, calculating its probability density distribution function, extracting the skewness coefficient and kurtosis coefficient of the distribution function as overall statistical distribution characteristics, linearly combining the skewness coefficient and kurtosis coefficient into a comprehensive index, using this comprehensive index as the controlled variable, setting the target value of the comprehensive index to zero, and outputting a correction amount for the heating power through the PID controller to adjust the power of the electric heating element in real time. Specifically, the linear combination involves multiplying the skewness coefficient by a first weighting coefficient and then adding it to the kurtosis coefficient multiplied by a second weighting coefficient. The sum of the first and second weighting coefficients is 1. The first weighting coefficient takes a larger value in the early stage of curing the precast component and a smaller value in the later stage, while the second weighting coefficient takes a smaller value in the early stage and a larger value in the later stage. The variation of the weighting coefficients is smoothly transitioned within the curing cycle using a cosine function. This method dynamically adjusts the heating power based on the symmetry and concentration of temperature change distribution, which can suppress temperature field shifts and abnormal peaks, ensuring that the precast components remain in a stable temperature evolution state throughout the curing process, thus guaranteeing the uniformity and strength consistency of the final structure.

[0011] The technical effects and advantages provided by the present invention in the above technical solution are as follows: Temperature anomaly clustering regions are identified by calculating the local curl characteristics of the temperature gradient field. The vorticity amplitude obtained by the cross product of the temperature gradient vector represents the degree of heat rotation and accumulation. Continuous spatial voxels are marked and screened using percentile thresholds, and isolated voxels are then removed through connected component analysis to obtain the multiple connected regions with the largest areas as temperature anomaly clustering regions. This method captures heat flow convergence phenomena based on vector field characteristics, and can detect closed flow regions or thermal accumulations that are difficult to find by relying solely on temperature amplitude or gradient magnitude. This makes the identification results closer to the behavior of local energy stagnation in actual physical processes, avoids missing slowly forming hotspots, and provides clear spatial target areas for subsequent differentiated control. Regions with drastic temperature fluctuations are determined based on the location of extreme points in the temperature rate of change field. Morphological top-hat transformation is performed on the temperature rate of change field to extract raised regions with rate of change amplitudes higher than the local background. The geometric center of the raised region is taken as the extreme point, and a spherical space with a preset radius is generated with the extreme point as the center. Overlapping regions are then merged. This processing suppresses the slowly varying background in the rate of change field, highlights local blocks with significant time dynamics, and can adaptively extract areas with temperature rise and fall rates significantly different from the surrounding areas without relying on preset thresholds. This accurately locates the spatial range affected by thermal shock, allowing subsequent flow regulation to directly act on the location of the fluctuation source and suppress the further development of temperature change amplitude.

[0012] The distribution coefficient and flow rate of the curing medium are adjusted separately for areas with abnormal temperature accumulation and areas with severe temperature fluctuations. Multiple independently adjustable medium guide vanes are pre-embedded, and the flow direction is changed by rotating the vanes according to the relative azimuth angle between the center coordinates of each area and the vane outlet. This ensures that the number of vane outlets matching each area is proportional to the volume of that area. Furthermore, the opening of the vane outlets is linearly adjusted based on the difference between the average temperature gradient modulus and the preset optimal modulus within the area to change the local flow rate. This distribution mechanism, based on regional spatial positioning and temperature gradient deviation, achieves coordinated control of the direction and intensity of medium supply to each abnormal area. Heat accumulation areas receive enhanced heat dissipation flow direction and flow rate, while areas with severe temperature fluctuations receive buffered flow rate to mitigate thermal shock. This creates a non-uniform, target-oriented medium distribution within the curing cover layer, improving the uniformity of the temperature field and the stability of its changes over time, and reducing internal stress caused by temperature differences. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0014] Figure 1 This is a flowchart of an optimized method for the maintenance of precast components used in road and bridge construction to adapt to temperature changes. Figure 2 This is a flowchart of the three-dimensional transient temperature field generation process for prefabricated components; Figure 3 It is a time-series data curve of temperature from sensors on the surface and inside the precast component; Figure 4 This is a distribution diagram of the temperature gradient modulus and temperature change rate along the length of the preform; Figure 5 This is a probability density distribution diagram of the vorticity amplitude in the temperature gradient field; Figure 6 It is a graph showing the probability density distribution and statistical characteristics of the temperature change rate of the precast component. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0016] See Figure 1This invention provides an optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, comprising the following steps: after the precast component is cast, a temperature sensor array is deployed according to preset monitoring points to collect multi-point temperature time-series data of the surface and interior of the precast component; based on the partial differential equation of heat conduction, the multi-point temperature time-series data is subjected to coupled spatial and temporal interpolation processing to generate a three-dimensional transient temperature field of the precast component; gradient operator operation is performed on the three-dimensional transient temperature field to obtain the spatial distribution of the temperature gradient field, and the temperature change rate at each spatial location at adjacent sampling times is calculated to generate a temperature change rate field; based on the local curl characteristics of the temperature gradient field, an abnormal temperature accumulation area is identified, and based on the extreme point location of the temperature change rate field, a region of severe temperature fluctuation is determined; for the abnormal temperature accumulation area and the region of severe temperature fluctuation, the flow direction distribution coefficient and flow rate of the maintenance medium in each region are adjusted respectively.

[0017] Example 1: In specific implementation, please refer to Figure 2 After the precast component is cast, a temperature sensor array is installed at preset monitoring points on its surface and inside. The temperature sensor array contains multiple contact thermocouples or fiber Bragg grating temperature sensors, arranged at preset intervals along the length, width, and height of the precast component. Surface sensors are attached to the inside of the template, while internal sensors are embedded in the concrete mixture. The temperature sensor array collects temperature data from each monitoring point at a fixed sampling period, forming multi-point temperature time-series data.

[0018] A three-dimensional transient temperature field of the prefabricated component is generated by coupled spatial and temporal interpolation processing of multi-point temperature time-series data. Specifically, the process involves using the partial differential equation of heat conduction as the governing equation and the multi-point temperature time-series data as discrete boundary conditions. The partial differential equation of heat conduction is: in, This represents the temperature at any given time at any spatial location inside the precast component. Indicates time, This represents the thermal diffusivity of the precast concrete. For the Laplace operator. Thermal diffusivity. The value of is obtained through parameter identification based on the concrete mix proportion and the measured temperature rise curve during the initial curing stage. During identification, an instantaneous heat source is applied locally to the precast component and the temperature response is recorded. The thermal diffusivity is then obtained by inversion and solution. The specific value.

[0019] In the spatial domain, the prefabricated component is treated as a cuboid computational domain. A three-dimensional orthogonal mesh is created along the length, width, and height directions of the prefabricated component. The spacing between mesh nodes in the length, width, and height directions is determined based on the density of the sensor array in those directions, ensuring that the spatial position of each sensor coincides with or corresponds to the nearest mesh node through interpolation. In the time domain, constant-step discretization is used. The time step is determined based on the sampling period of the temperature sensor array and is not greater than the maximum allowable time step that satisfies the numerical stability condition.

[0020] The partial differential equation for heat conduction is discretized using a seven-point difference scheme. For internal grid nodes, forward difference is used to approximate the time derivative term at the time level, and a second-order central difference scheme is used for the Laplace operator at the spatial level, involving the temperature values ​​of the current node and its two adjacent nodes in the length direction, width direction, and height direction (a total of seven nodes). The partial differential equation for heat conduction is transformed into a linear algebraic equation for each internal grid node at each time step. The left side of the equation represents the unknown temperature of the node at the current time step, and the right side represents the combination of the known temperature of the previous time step and the unknown temperatures of adjacent nodes.

[0021] The boundary condition processing method is as follows: For grid nodes located on the surface of the precast component or corresponding to the sensor installation location, the temperature value at the corresponding time in the multi-point temperature time series data is directly used as the known temperature value of the grid node at the corresponding time step, and substituted into the discrete equation system as the first type of boundary condition. For boundary grid nodes not covered by measurement points, the temperature value is assigned from the temperature time series data of the nearest sensor measurement point through nearest neighbor interpolation, forming a closed set of boundary conditions.

[0022] At each time step, all internal grid nodes form a sparse linear equation system. The temperature field of the previous time step is used as the initial value for iteratively solving the temperature field of the current time step, and the Gauss-Seidel iterative method is employed. The Gauss-Seidel iterative method is executed as follows: all internal nodes are traversed in the natural order of the grid nodes along the length, width, and height directions. Each time the temperature value of a node is updated, the latest updated temperature value of that node and the latest temperature values ​​obtainable from adjacent nodes in the current iteration are immediately used to calculate the new temperature value of that node. After each iteration, the maximum absolute deviation between the new temperature values ​​of all nodes and the temperature values ​​of the previous iteration step is calculated, and the maximum absolute deviation is compared with a preset convergence tolerance. The preset convergence tolerance is set to one-tenth of the sensor's measurement accuracy. If the maximum absolute deviation is less than the preset convergence tolerance, the iteration stops, and the discrete node temperature values ​​for the current time step are obtained; otherwise, the next iteration continues until the convergence condition is met.

[0023] The above discretization and iterative solution process is executed sequentially for all time steps to obtain the temperature time series of the three-dimensional mesh nodes of the precast component at all sampling times, that is, each mesh node corresponds to a time series of temperature change over time.

[0024] After acquiring the temperature time series of all grid nodes, cubic spline interpolation is used to map the grid node temperatures to arbitrary spatial locations, forming a continuous three-dimensional transient temperature field. The cubic spline interpolation process is as follows: a separate cubic spline interpolation function is constructed for the time series of each grid node, using natural boundary conditions, i.e., the second derivative at the endpoints of the interpolation interval is zero, the interpolation nodes are each sampling time, and the function value is the node temperature value at each time. Piecewise cubic polynomials ensure the continuity of the function values ​​and the first derivative at the nodes, thus obtaining the temperature value of each grid node at any time. In the spatial dimension, cubic spline interpolation is first performed along the length of the precast component, obtaining a series of continuous curves in the length direction; then, cubic spline interpolation is performed on the length direction interpolation results in the width direction; finally, cubic spline interpolation is performed on the height direction interpolation results of the first two steps. Using a spline interpolation function in the form of a three-dimensional tensor product, the discrete grid node temperatures are extended to arbitrary coordinate positions throughout the entire spatial domain. Thus, a continuous three-dimensional transient temperature field covering the entire time and spatial domains of the precast component is generated.

[0025] See Figure 3 In the figure, the horizontal axis represents curing time in hours (h), and the vertical axis represents temperature in degrees Celsius (°C). The curves correspond to the time-series temperature data measured by multiple temperature sensors installed on the surface and inside the precast component. Surface sensors 1 and 2 are represented by solid blue lines and dashed orange lines, while internal sensors 1, 2, and 3 are represented by dashed green lines, dotted red lines, and squared purple lines, respectively.

[0026] Overall, the temperature measured by all sensors went through a process of first rising and then falling. During the rising phase, the temperature gradually climbed, reaching its peak at around 20 hours, and then entered a slow cooling phase, stabilizing at around 20°C, close to the ambient temperature, at around 120 hours.

[0027] Specifically, the internal sensors exhibited the highest temperatures. Internal sensor 1 reached its peak temperature of approximately 65°C, while internal sensor 3 peaked at approximately 58°C, and internal sensor 2 peaked at approximately 57°C. This indicates that the internal temperature of the preform is significantly higher than the surface temperature, and there are certain temperature differences at different locations within the preform. Surface sensors 1 and 2 reached peak temperatures of approximately 44°C and 42°C, respectively, which are significantly lower than those of the internal sensors. Furthermore, the shape and values ​​of their temperature curves are similar, indicating a relatively uniform surface temperature distribution.

[0028] Around the 30th hour, the temperatures of surface sensors 1 and 2 began to drop rapidly, and the rate of drop was significantly faster than that of the internal sensors, indicating that the surface temperature was greatly affected by the external environment. At the same time, the temperature of the internal sensors dropped more slowly, which helped to form an internal temperature gradient.

[0029] Example 2: In the specific implementation, after obtaining the three-dimensional transient temperature field, gradient operator operations are performed on the three-dimensional transient temperature field. The three-dimensional transient temperature field is generated by Example 1, and each grid node in the cuboid spatial domain corresponding to the prefabricated part has a definite temperature value. The gradient operator operation is applied to the three-dimensional transient temperature field to obtain the spatial distribution of the temperature gradient field.

[0030] For each grid node in the three-dimensional transient temperature field, the partial temperature derivatives along the length, width, and height of the preform are calculated using the second-order central difference formula. Taking the calculation of the partial derivatives along the length of the preform as an example, the second-order central difference formula is as follows: in, This represents the partial derivative of the temperature at the mesh node along the length of the preform. Indicates that the index in the length direction of the precast component is Width direction index is The height direction index is The temperature values ​​of the grid nodes, Indicates that the index in the length direction of the precast component is Width direction index is The height direction index is The temperature values ​​of the grid nodes, , , All are integer indices. This indicates the spacing between two adjacent grid nodes along the length of the precast component. The value is determined when dividing the three-dimensional transient temperature field grid, and is taken as half of the spacing of the sensor array along the length direction.

[0031] The partial temperature derivative in the width direction of the preform is calculated using the same second-order central difference formula, with the width direction index used in the calculation. and The grid node temperature values, with the grid spacing in the width direction as the denominator. The partial derivative of temperature in the height direction uses the height direction index. and The grid node temperature values, with the grid spacing in the height direction as the denominator. . and The value of is determined by the grid spacing in the width and height directions, respectively.

[0032] The three directional partial derivatives calculated for a single grid node are combined into a vector, which serves as the temperature gradient vector for that grid node, denoted as... The above temperature partial derivative calculation and vector construction operations are performed on all mesh nodes. The temperature gradient vector calculations of all mesh nodes are independent of each other and are completed synchronously in parallel to obtain a set of discrete temperature gradient vectors on the three-dimensional mesh nodes.

[0033] After obtaining the temperature gradient vectors of all grid nodes, bilinear interpolation is used to extend the node gradient vectors to the entire spatial domain, forming a continuous temperature gradient field. For any non-grid node location within the prefabricated spatial domain, its temperature gradient vector is synthesized by interpolation of the temperature gradient vectors of the eight corner nodes of the cubic grid cell surrounding that location. The interpolation synthesis process is as follows: First, on the length-width plane of the cubic grid cell, bilinear interpolation is performed using the temperature gradient vectors of the four corner nodes for the relative coordinates of the spatial location in the length and width directions, resulting in four interpolation results on that plane; then, along the height direction, linear interpolation is performed on the four interpolation results based on the relative coordinates of the spatial location in the height direction, finally obtaining the temperature gradient vector at that location. By traversing all non-grid node locations to complete the interpolation, a continuous spatial temperature gradient field is generated.

[0034] The temperature change rate field is generated as follows: For each grid node in the three-dimensional transient temperature field, the temperature value of that grid node at two adjacent sampling times is extracted. The temperature value at the later sampling time is subtracted from the temperature value at the earlier sampling time, and the temperature difference is divided by the time interval between the two sampling times to obtain the temperature change rate of that grid node within that time interval. This is performed synchronously on all grid nodes to obtain the temperature change rate values ​​for each grid node. Subsequently, using the same bilinear interpolation method as the temperature gradient field, the temperature change rate values ​​of the grid nodes are extended to the entire spatial domain. Specifically, at non-node locations, a continuous temperature change rate field is generated by performing length-width plane bilinear interpolation and height-direction linear interpolation on the temperature change rate values ​​of the eight corner nodes surrounding the cubic element.

[0035] See Figure 4 In the figure, the horizontal axis represents the position of the preform along its length (unit: meters), the left side of the vertical axis represents the temperature gradient modulus (unit: °C / m), and the right side of the vertical axis represents the temperature change rate (unit: °C / h). In the legend, the blue solid line is the temperature gradient modulus curve, and the orange dashed line is the temperature change rate curve.

[0036] As shown in the figure, the temperature gradient modulus exhibits a fluctuating trend of first increasing and then decreasing along the length of the preform. Specifically, at the initial position (approximately 0-2m), the temperature gradient modulus gradually increases from approximately 2.5℃ / m to around 3.5℃ / m; in the middle section (approximately 2-6m), the temperature gradient modulus further increases to a peak of approximately 4.6℃ / m, then slowly decreases to 3.7℃ / m; at the end section (approximately 6-10m), the fluctuation of the temperature gradient modulus weakens, generally tending towards the range of 2.5-3.5℃ / m. This trend reflects the non-uniform spatial distribution of the temperature gradient within the preform, with the peak region likely corresponding to key areas where internal temperature changes are more drastic.

[0037] The temperature change rate curve exhibits a spatial distribution characteristic different from that of the temperature gradient modulus. In the 0–2 m length range, the temperature change rate is relatively high with the largest fluctuation, peaking at approximately 2.2 °C / h, then gradually decreasing to approximately 0.5 °C / h (2–8 m range), with a trough in the middle section. In the 8–10 m range, the temperature change rate rises again, with increased fluctuation and stabilizing between 1.0 and 2.0 °C / h. This trend in temperature change rate indicates relatively drastic temperature fluctuations at both ends of the precast component's length, while the fluctuations are relatively gentle in the middle.

[0038] The spatial distribution of the temperature gradient modulus and the rate of temperature change exhibits certain complementary characteristics. The peak value of the temperature gradient modulus is concentrated in the middle section of the preform, while the peak value of the rate of temperature change is mainly distributed in the two ends. This phenomenon indicates that the spatial variation patterns of the temperature gradient and temperature change are different. Therefore, it is necessary to adopt control strategies for optimizing the flow direction and flow rate of the curing medium in different regions to achieve the uniformity and stability of the temperature field inside the preform.

[0039] Example 3: In practice, local curl feature analysis is performed on the generated temperature gradient field to identify regions of temperature anomaly accumulation. The temperature gradient field was obtained in Example 2, and a temperature gradient vector exists at the center of each voxel within the preform's spatial domain.

[0040] Calculate the curl vector at each spatial location in the temperature gradient field. The curl vector is obtained through the curl operation of the temperature gradient vector, which is the cross product of the gradient operator and the temperature gradient vector. In Cartesian coordinates, let the temperature gradient vector be... ,in, This represents the component of the temperature gradient vector along the length of the preform. This represents the component of the temperature gradient vector along the width of the preform. This represents the component of the temperature gradient vector along the height of the preform. Curl vector. The formula for calculation is: in, This represents the partial derivative of the height component of the temperature gradient vector with respect to the width component. This represents the partial derivative of the width component of the temperature gradient vector with respect to the height component. This represents the partial derivative of the length component of the temperature gradient vector with respect to the height coordinate. This represents the partial derivative of the height component of the temperature gradient vector with respect to the length component. This represents the partial derivative of the width component of the temperature gradient vector with respect to the length component. This represents the partial derivative of the length component of the temperature gradient vector with respect to the width component. All of these partial derivatives are calculated on a discrete grid using the second-order central difference formula, with the grid spacing consistent with the spacing used when dividing the temperature gradient field grid.

[0041] curl vector The vorticity amplitude is obtained by taking the square root of the sum of the squares of the components of the curl vector, i.e. ,in, , , These represent the components of the curl vector in the length, width, and height directions, respectively. After calculating all spatial positions within the precast component's spatial domain, the spatial distribution of the vorticity amplitude is obtained.

[0042] A percentile threshold for vorticity amplitude is set. The percentile threshold is set to the 95th percentile, based on the following criteria: all vorticity amplitude values ​​are sorted from smallest to largest, and the vorticity amplitude at the 95th percentile position is used as the threshold. This ensures that the top 5% of spatial voxels with the largest vorticity amplitudes are included in the candidate range. This proportion is based on the statistical regularity that thermal anomalies inside the precast component usually account for a relatively small proportion of the total volume. Spatial voxels with vorticity amplitudes exceeding the percentile threshold are marked as foreground voxels in the binary image, and the remaining voxels are marked as background voxels, forming a candidate set of binary voxels.

[0043] Connectivity analysis is performed on the candidate binary voxel set. The 26-neighborhood connectivity criterion is used: for each foreground voxel, it is checked whether its 26 adjacent voxels in the three coordinate directions and all diagonal directions are also foreground voxels; interconnected foreground voxels are grouped into the same connected component. After traversing all foreground voxels to complete the connected component labeling, the number of voxels contained in each connected component is counted. A lower limit for the number of voxels is set as a preset isolation rejection threshold, with a value of 8. This threshold is based on the number of voxels corresponding to the smallest meaningful cluster structure in 3D space consisting of a single voxel and all 26 neighboring voxels. Connected components with fewer than 8 voxels are identified as isolated voxels and rejected. Connected components with 8 or more voxels are retained, and the set of connected voxels with the largest number of voxels in the retained connected components is identified as a temperature anomaly cluster region. The upper limit for the number of retained connected voxels is determined based on the actual situation and is not fixed.

[0044] The region of drastic temperature fluctuation was determined by performing a morphological top-hat transformation on the temperature rate of change field. The temperature rate of change field, obtained in Example 2, represents the rate of temperature change over time at each spatial location. The top-hat transformation used a disk-shaped structuring element to perform a grayscale opening operation on the temperature rate of change field. The radius parameter of the disk-shaped structuring element was selected based on the relationship between the feature size of the prefabricated component and the expected size of the fluctuation region, taking a value of one-twentieth of the shortest side length of the prefabricated component. This ensured that the structuring element was sufficient to cover local background changes while retaining raised areas above the background. Subtracting the opening operation result from the original temperature rate of change field yielded the top-hat transformation response map. Higher values ​​in the response map indicate a more prominent temperature rate of change relative to the local background at the corresponding spatial location.

[0045] An adaptive threshold segmentation is performed on the top-hat transform response map. The segmentation threshold is twice the mean of all values ​​in the top-hat transform response map. This multiple is automatically determined and rounded by Otsu's method on the overall gray-level histogram of the response map. Continuous spatial regions with response values ​​higher than the segmentation threshold are extracted as raised regions. The geometric center coordinates of each raised region are calculated. The geometric center coordinates are determined by the arithmetic mean of the coordinates of all voxels within the raised region in three directions. The calculation result is used as the extreme point corresponding to that raised region.

[0046] A spherical spatial region with a preset radius is established centered on each extreme point, and this spherical spatial region is defined as a region of severe temperature fluctuation. The preset radius is taken as the average value of the spacing between the prefabricated sensors, ensuring that the spherical region at least covers the typical neighborhood range surrounding the central sensor. After generating the corresponding spherical regions for all extreme points, it is checked whether there is spatial overlap between the spherical regions corresponding to different extreme points. The criterion for spatial overlap is that the distance between the centers of two spherical regions is less than twice the preset radius. The overlapping spherical regions are merged into a single connected fluctuation region, ultimately forming the entire region of severe temperature fluctuation.

[0047] See Figure 5 In the figure, the horizontal axis represents the vorticity amplitude of the curl vector of the temperature gradient field, ranging from near zero to approximately 15. The vertical axis represents the probability density distribution of the vorticity amplitude, reflecting the statistical frequency of the vorticity amplitude in the precast component's spatial domain. The blue bars are histograms of the probability density distribution of the vorticity amplitude, showing that most vorticity amplitudes are concentrated in the range of 0 to approximately 8.5, with a higher probability density in the range of 0 to 2, indicating that most spatial voxels have low curl amplitudes and a relatively smooth temperature gradient field. The yellow bars represent voxels with vorticity amplitudes exceeding the 95th percentile threshold of 8.51, representing anomalous voxel regions with high vorticity amplitudes. The red dashed line in the figure marks the 95th percentile threshold of 8.51, serving as the threshold for identifying temperature anomaly clustering regions.

[0048] Example 4: In practice, multiple independently adjustable media guide plates are pre-embedded within the curing cover layer of the precast components. These media guide plates are arranged in an array with alternating rows and columns within the curing cover layer. Each group of media guide plates is driven by an independent servo motor and can rotate around a vertical axis to change the spray direction of the curing media. An electrically operated regulating valve is installed at the outlet of each media guide plate. By changing the valve opening, the flow cross-sectional area at the outlet of the media guide plate is adjusted, thereby regulating the flow rate of the curing media passing through that outlet. The curing media is water mist or steam, which is transported from the curing media source to each media guide plate via pipelines.

[0049] Regions of temperature anomaly accumulation and regions of severe temperature fluctuation were identified using the method in Example 3. Each region of temperature anomaly accumulation and each region of severe temperature fluctuation was represented as a set of continuous voxels in space. For each region of temperature anomaly accumulation and each region of severe temperature fluctuation, its geometric center coordinates were extracted. The geometric center coordinates were calculated by taking the arithmetic mean of the coordinate values ​​of all voxels within the region along the length, width, and height directions of the preform, thus obtaining the coordinate values ​​of the geometric center on the three coordinate axes.

[0050] For the geometric center of each region, calculate the relative azimuth angle of that geometric center relative to the outlet positions of each medium guide vane within the curing layer. The outlet positions of the medium guide vanes are determined by the pre-embedded design drawings, and the three-dimensional coordinates of each outlet are known. The relative azimuth angle is calculated as follows: taking the projection point of the medium guide vane outlet position onto the horizontal plane formed by the length and width directions of the precast component as the origin, and the projection point of the geometric center onto the horizontal plane as the target point, calculate the direction vector from the origin to the target point, and solve for the angle between this direction vector and the reference axis of the length direction to obtain the relative azimuth angle. For each region, select several medium guide vanes that are closest to its geometric center in a straight line, and drive the servo motors of these medium guide vanes to rotate, adjusting the spray direction angle of the medium guide vanes to the calculated relative azimuth angle, thereby pointing the curing medium jet direction towards the geometric center of the corresponding region.

[0051] The allocation method for the number of medium guide vane outlets in each region is as follows: The total volume of all areas with abnormal temperature accumulation and areas with drastic temperature fluctuations is calculated, and the total number of medium guide vanes pre-embedded in the curing cover layer is determined simultaneously. The number of medium guide vane outlets per unit volume is calculated by dividing the total number of medium guide vanes by the total volume of all regions. For each individual region, its volume is multiplied by the number of medium guide vane outlets per unit volume, and the resulting product is rounded down to the nearest integer as the number of medium guide vane outlets allocated to that region. If the rounded result is zero, at least one medium guide vane outlet is allocated. The allocation process follows the principle of proximity between regions and medium guide vanes, mapping them one-to-one until each region is allocated the corresponding number of medium guide vane outlets.

[0052] Simultaneously, for each region with concentrated temperature anomalies and region with drastic temperature fluctuations, the arithmetic mean of the temperature gradient magnitudes of all grid nodes within that region is calculated. The temperature gradient magnitude is obtained by calculating the magnitude from the temperature gradient vector obtained in Example 2. The temperature gradient magnitude of each grid node is denoted by the symbol... The modulus of the regional average temperature gradient is indicated by the symbol [symbol missing]. Indicated by the symbol for the preset optimal temperature gradient modulus. express, The method for determining the optimal temperature gradient modulus is as follows: Based on the mix proportions, cross-sectional dimensions, and temperature stress control requirements of the precast concrete in the curing process specifications, the temperature gradient modulus generated inside the precast component under ideal uniform temperature drop conditions is obtained through finite element simulation calculations. This temperature gradient modulus is then used as the preset optimal temperature gradient modulus. For each region, the calculation... and The difference.

[0053] The outlet opening of the medium guide plate in the corresponding region is linearly adjusted based on the difference, and the adjustment formula is as follows: in, This indicates the adjusted outlet opening of the medium guide plate. This represents the reference opening degree, which is 50% of the cross-sectional area of ​​the medium guide plate outlet in its fully open state. This is the opening adjustment ratio coefficient. The average temperature gradient modulus within the region. To preset the optimal temperature gradient modulus. When Greater than hour, For positive, The outlet opening of the medium guide plate is reduced to decrease the local maintenance medium flow rate and slow down the rate of temperature change in that area; when Less than hour, Negative, The opening of the medium guide plate outlet is increased to increase the flow rate of the local curing medium and accelerate the rate of temperature change in the area.

[0054] Opening adjustment ratio coefficient The distance is segmented based on the distance from the area to the curing medium source. The distance is calculated as follows: the path length along the pipeline from the geometric center of the area to the outlet of the curing medium source is taken as the distance from the area to the curing medium source, denoted by the symbol... Indicate. Will The range of values ​​is divided into three intervals: Less than the first distance threshold This is the near-source region. Greater than or equal to And less than the second distance threshold For the middle source region, Greater than or equal to For the distant source region. First distance threshold. Take one-third of the total length of the maintenance medium pipeline, and use the second distance threshold. Take two-thirds of the total length of the maintenance medium pipeline. The opening adjustment ratio coefficient corresponding to the near-source section is taken as [value missing]. The opening adjustment ratio coefficient corresponding to the middle source interval is taken as: The opening adjustment ratio coefficient corresponding to the far-source interval is taken as: All three conditions are met. . , , The specific values ​​were obtained by conducting hydraulic tests on the maintenance medium pipeline: flow meters and pressure sensors were installed at different distances to measure the change in flow rate corresponding to a unit change in valve opening. Then, the proportional coefficient was calculated based on the correspondence between the flow rate change and the temperature gradient modulus adjustment requirements, so that the opening adjustment range in the far-source area is larger under the same gradient difference, in order to compensate for the flow rate attenuation caused by the pressure loss along the pipeline during the transportation of the maintenance medium.

[0055] Example 5: In specific implementation, the temperature change rate values ​​corresponding to all spatial locations are extracted from the generated temperature change rate field. The temperature change rate field is obtained by the method in Example 2, covering the entire spatial domain of the prefabricated component, with each spatial voxel having a temperature change rate value. The temperature change rate values ​​of all spatial voxels are combined into a sample set, denoted as . , where subscript Voxel number, total sample size It equals the total number of space voxels.

[0056] For sample set The probability density distribution function is calculated. Using an equally spaced histogram statistical method, the range of temperature change rate is divided into several equally wide intervals. The width of each interval is determined based on the total number of samples and the range. The number of intervals is [number to be filled in]. The square root of the value is rounded up. The frequency of the temperature change rate falling within each interval is counted, and then the frequency is divided by the product of the total number of samples and the interval width to obtain the probability density value. The probability density distribution function curve of the temperature change rate field is formed by plotting the midpoint of the interval as the x-axis and the probability density value as the y-axis.

[0057] Skewness and kurtosis coefficients are extracted from the probability density function as overall statistical distribution characteristics. The skewness coefficient is represented by the symbol... The kurtosis coefficient is indicated by the symbol. express. The calculation method is as follows: first calculate the sample mean of the rate of temperature change. and sample standard deviation , For all The arithmetic mean, For all and The arithmetic square root of the mean of the sum of the squares of the differences; then calculate each Standardized value Sum the cubed standardized values ​​and divide by the total number of samples. The skewness coefficient is obtained. . The calculation method is as follows: sum the fourth power of each standardized value and divide by the total number of samples. Then subtract 3. Subtracting 3 makes the kurtosis coefficient of the normal distribution zero. It reflects the direction and degree of asymmetry in the distribution of the rate of temperature change. It reflects how steep or flat the distribution of the rate of temperature change is relative to the normal distribution.

[0058] The skewness coefficient and kurtosis coefficient are linearly combined into a comprehensive index, denoted by the symbol... The formula for calculating the comprehensive index is as follows: Among them, the first weighting coefficient Second weighting coefficient Let be a coefficient that varies with time, satisfying First weighting coefficient A larger value is used in the early stage of precast component curing, and a smaller value is used in the later stage of curing; second weighting coefficient A smaller value is used in the early stage of precast component curing, and a larger value is used in the later stage of curing. The variation of the weighting coefficient is smoothly transitioned within the curing cycle using a cosine function. Specifically, the total curing time is set as follows: Indicates the current moment using This indicates that the timing begins from the start of maintenance, phase Then the first weight coefficient Second weighting coefficient Initial stages of maintenance Approaching 0 Approaching 0 Approaching 1, Approaching 1 Approaching 0; late stage of maintenance near , near , Close to -1, Approaching 0 Approaching 1. This change means that the comprehensive index in the early stage of maintenance is mainly dominated by the skewness coefficient, focusing on controlling the overall directional deviation of temperature changes; while the comprehensive index in the later stage of maintenance is mainly dominated by the kurtosis coefficient, focusing on controlling the dispersion and extreme fluctuations of temperature changes.

[0059] Based on comprehensive indicators Let be the controlled variable, and set its target value to zero. When the temperature change rate distribution is perfectly symmetrical and its kurtosis matches a normal distribution... A value of 0 indicates that the temperature change is uniform overall. The heating power of the curing medium is continuously adjusted using a proportional-integral-derivative controller to eliminate... Deviation from the target value.

[0060] The formula for calculating the discrete incremental output heating power correction of the proportional-integral-derivative controller is as follows: in, This represents the heating power correction amount calculated in the current control cycle; This represents the comprehensive index value calculated during the current control cycle; This represents the comprehensive index value calculated in the previous control cycle. This represents the comprehensive index value calculated in the control cycle two weeks prior. This represents the control period of the proportional-integral-derivative (PID) controller, and its value is consistent with the sampling period of the temperature sensor array. Indicates proportional gain. Indicates integral gain. This represents the differential gain. The proportional gain... Integral gain and differential gain The step response of the curing medium heating system was determined through testing: During the commissioning phase of the curing system, a power step signal of known amplitude was applied to the electric heating element, and the time response curve of the comprehensive index was recorded. Based on the overshoot, rise time, and steady-state error of the response curve, the critical proportional gain method or the attenuation curve method was used to calculate the results. , , The specific values ​​are determined, and these three gain values ​​are stored in the controller's parameter register.

[0061] Each control cycle performs a proportional-integral-derivative (PID) controller operation to obtain the heating power correction amount. .Will The heating power command is added to the current reference value and becomes the new heating power command. The reference value is the preset power curve value set according to the curing stage in the curing process specification. The added heating power command is limited, with the upper limit being the rated power of the electric heating element and the lower limit being zero power. The limited heating power command is sent to the electric heating element for execution, adjusting the heating power of the curing medium in real time, thereby changing the temperature of the curing medium and thus affecting the overall statistical distribution characteristics of the temperature change rate field inside the precast component, thus affecting the comprehensive index. Approaching zero.

[0062] See Figure 6 The horizontal axis of the graph represents the rate of temperature change, measured in degrees Celsius per hour (°C / h), which measures the rate at which the temperature of the precast space voxel changes over time. The vertical axis represents the probability density, reflecting the frequency of the temperature change rate distribution in the sample set. The graph uses blue bars to represent the histogram probability density distribution of the actual temperature change rate, solid blue lines to represent the kernel density estimation curve, and dashed orange lines to represent the normal distribution curves corresponding to the mean and variance. The legend clearly indicates the meaning of each curve for easy comparison and analysis.

[0063] As shown in the figure, the actual temperature change rate distribution exhibits a slight right skew, with the curve peaking in the range of approximately 2.0–3.0 °C / h, where the probability density reaches a high of nearly 0.4, indicating that the temperature change rate is relatively concentrated within this range. The overall distribution is relatively smooth, and the kernel density estimation curve largely matches the normal distribution curve, but there are differences in the peak height and tail shape, showing that the actual temperature change rate has certain non-normal characteristics.

[0064] The skewness coefficient in the figure is 0.378, indicating that the temperature change rate distribution is somewhat positively skewed, with a longer tail on the right side of the data, reflecting that the temperature change rate in some areas is slightly higher than the overall average during the maintenance process; the kurtosis coefficient is -0.145, slightly below zero, indicating that the distribution is slightly flatter than the normal distribution, with a lighter tail, and relatively fewer extreme temperature fluctuations.

[0065] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An optimized method for the curing of precast components used in road and bridge construction that adapts to temperature changes, characterized in that, Includes the following steps: After the precast component is cast, a temperature sensor array is set up according to the preset monitoring points to collect multi-point temperature time-series data of the surface and interior of the precast component. Based on the partial differential equation of heat conduction, the multi-point temperature time series data is subjected to coupled interpolation processing in the spatial and temporal domains to generate a three-dimensional transient temperature field of the preform. Gradient operator operations are performed on the three-dimensional transient temperature field to obtain the spatial distribution of the temperature gradient field. At the same time, the rate of temperature change at each spatial location at adjacent sampling times is calculated to generate a temperature rate of change field. The temperature anomaly accumulation region is identified based on the local curl characteristics of the temperature gradient field, and the region of drastic temperature fluctuation is determined based on the location of the extreme point of the temperature change rate field. For the areas where abnormal temperatures accumulate and the areas where temperatures fluctuate drastically, the flow direction distribution coefficient and flow rate of the curing medium in each area are adjusted respectively. The gradient operator operation is performed on the three-dimensional transient temperature field to obtain the spatial distribution of the temperature gradient field, specifically as follows: At each grid node of the three-dimensional transient temperature field, the temperature partial derivatives along the length, width and height of the preform are calculated using the second-order central difference formula. The vector formed by the partial derivatives in the three directions is used as the temperature gradient vector of the node. After parallel calculation for all grid nodes, the node gradient vector is extended to the entire spatial domain by bilinear interpolation to form a continuous temperature gradient field.

2. The optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, as described in claim 1, is characterized in that... The process of performing coupled spatial and temporal interpolation on the multi-point temperature time-series data based on the partial differential equation of heat conduction to generate a three-dimensional transient temperature field of the preform is as follows: Using the heat conduction partial differential equation as the governing equation, the multi-point temperature time series data is used as discrete boundary conditions. The time and spatial coordinates are discretized using a seven-point difference scheme. The discrete node temperature values ​​at each time step are solved by the Gauss-Seidel iterative method to obtain the temperature time series of the overall three-dimensional mesh nodes of the preform. Then, the mesh node temperature is mapped to any spatial location by cubic spline interpolation to form a continuous three-dimensional transient temperature field.

3. The optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, as described in claim 1, is characterized in that... The step of identifying temperature anomaly clustering regions based on the local curl characteristics of the temperature gradient field specifically involves: The curl vector at each spatial location in the temperature gradient field is calculated, i.e., the vorticity is obtained by performing a cross product operation on the temperature gradient vector. A percentile threshold for the vorticity amplitude is set, and continuous spatial voxels with vorticity amplitudes exceeding the percentile threshold are marked as candidate anomalous clustering regions. Then, isolated voxels are removed through connected component analysis, and the multiple connected voxels with the largest areas are retained as temperature anomalous clustering regions.

4. The optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, as described in claim 1, is characterized in that... The determination of the region of drastic temperature fluctuation based on the location of the extreme points of the temperature change rate field specifically involves: A morphological top-hat transformation is performed on the temperature change rate field to extract raised regions with a change rate amplitude higher than the local background. The geometric center of each raised region is taken as an extreme point. A spherical space with the extreme point as the center and a preset radius as the diameter is defined as a region of violent temperature fluctuation. If the spherical regions corresponding to different extreme points overlap, they are merged into a single fluctuation region.

5. The optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, as described in claim 1, is characterized in that... Specifically, for the areas of abnormal temperature accumulation and the areas of drastic temperature fluctuation, the flow direction distribution coefficient and flow rate of the curing medium in each area are adjusted, respectively: Multiple independently adjustable medium guide vanes are pre-embedded in the curing cover layer of the precast component. Based on the center coordinates of each region and the relative azimuth angle of the guide vane outlet, the medium flow direction is changed by rotating the guide vane angle. This makes the number of guide vane outlets corresponding to each abnormal accumulation region and fluctuation region proportional to the volume of that region. At the same time, based on the difference between the average temperature gradient modulus of each region and the preset optimal gradient modulus, the opening of the corresponding guide vane outlet is linearly adjusted to change the local flow rate.

6. The optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, as described in claim 5, is characterized in that... The method of linearly adjusting the opening of the corresponding guide vane outlet based on the difference between the average temperature gradient modulus of each region and the preset optimal gradient modulus is as follows: Calculate the arithmetic mean of the temperature gradient modulus of all grid nodes in each abnormal clustering area and fluctuation area. Subtract this mean from the preset optimal gradient modulus to obtain the difference. When the difference is positive, the opening of the guide plate outlet decreases; when the difference is negative, the opening increases. The opening adjustment is proportional to the absolute value of the difference, and the proportional coefficient is segmented according to the distance of the area from the maintenance medium source.

7. The optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, as described in claim 1, is characterized in that... The method further includes: continuously adjusting the heating power of the curing medium using a proportional-integral-derivative controller based on the overall statistical distribution characteristics of the temperature change rate field.

8. The optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, as described in claim 7, is characterized in that... Based on the overall statistical distribution characteristics of the temperature change rate field, the heating power of the curing medium is continuously adjusted using a proportional-integral-derivative controller, specifically as follows: The rate of change values ​​of all spatial locations in the entire temperature change rate field are statistically analyzed, and their probability density distribution function is calculated. The skewness coefficient and kurtosis coefficient of the distribution function are extracted as the overall statistical distribution characteristics. The skewness coefficient and kurtosis coefficient are linearly combined into a comprehensive index. This comprehensive index is used as the controlled variable, and the target value of the comprehensive index is set to zero. The heating power correction is output through the proportional-integral-derivative controller to adjust the power of the electric heating element in real time.

9. The optimized maintenance method for precast components used in road and bridge construction that adapts to temperature changes, as described in claim 8, is characterized in that... The method of linearly combining the skewness coefficient and kurtosis coefficient into a comprehensive index is as follows: The skewness coefficient multiplied by the first weighting coefficient is added to the kurtosis coefficient multiplied by the second weighting coefficient. The sum of the first weighting coefficient and the second weighting coefficient is 1. The first weighting coefficient takes a larger value in the early stage of curing of the precast component and a smaller value in the later stage of curing. The second weighting coefficient takes a smaller value in the early stage of curing and a larger value in the later stage of curing. The change pattern of the weighting coefficients adopts a cosine function to smoothly transition within the curing cycle.

Images