Hydration Heat Effect and Temperature Control Measures of Long-Span U-Shaped Aqueducts

Abstract

This study presents a comprehensive analysis of hydration heat-induced temperature and stress fields in a U-shaped aqueduct during the casting phase, integrating field measurements and numerical simulations. The key findings are as follows: (1) Thermal Evolution Characteristics: Both experimental and numerical results demonstrated consistent thermal behavior, characterized by a rapid temperature rise, subsequent rapid cooling, and eventual stabilization near ambient conditions. The peak temperature is observed at the centroid of the bearing section’s base slab, reaching 83.8 °C in field tests and 87.0 °C in simulations. (2) Stress Field Analysis: Numerical modeling reveals critical stress conditions in the outer concrete layers within high-temperature zones. The maximum tensile stress reaches 6.37 MPa, exceeding the allowable value of the tensile strength of the current concrete (1.85 MPa) by 244%, indicating a significant risk of thermal cracking. (3) Temperature Gradient and Cooling Rate Anomalies: Both methodologies identify non-compliance with critical control criteria. Internal-to-surface temperature differentials exceed the 25 °C threshold. Daily cooling rates at monitored locations surpass 2.0 °C/d during the initial 5–6 days of the cooling phase, elevating cracking risks associated with excessive thermal gradients. (4) Mitigation Strategy Proposal: Implementation of a hydration heat control system is recommended; compared to single-layer systems, the proposed mid-depth double-layer steel pipe cooling system (1.2 m/s flow) reduced peak temperature by 23.8 °C and improved cooling efficiency by 28.7%. The optimized water circulation maintained thermal balance between concrete and cooling water, achieving water savings and cost reduction while ensuring structural quality. (5) The cooling system proposed in this paper has certain limitations in terms of applicable environment and construction difficulty. Future research can combine with a BIM system to dynamically control the tube cooling system in real time.

1. Introduction

Prestressed concrete U-shaped aqueducts are extensively utilized in hydraulic engineering due to their advantages, including lightweight, favorable structural stress performance, efficient hydraulic conditions, and cost-effective cross-sections. The existing research focuses on the hydration heat temperature control measures of conventional cross-section aqueducts [1,2,3,4]. Previous studies on these U-shaped aqueducts have primarily concentrated on structural optimization design [5,6,7], seismic performance [8,9,10], and dynamic fluid–structure interaction [11,12], and less attention has been paid to the temperature field and stress field caused by hydration reaction during concrete pouring.

These U-shaped aqueducts are thin-walled concrete structures characterized by length dimensions significantly exceeding their thickness. Their cross-sectional shapes, structural configurations, and construction technologies differ from those of mass concrete structures, such as dams and bridge abutments, making them more sensitive to temperature control measures, including pouring temperature, pipe cooling systems, and insulation techniques. In addition, due to the thin webs and the high thermal conductivity of the steel formwork covering the surface, heat dissipates easily, while the heat produced by the hydration reaction of concrete, which has low thermal conductivity, accumulates inside. This phenomenon results in a substantial temperature difference between the interior and surface of the aqueduct’s concrete. The resulting temperature gradient induces significant thermal stresses [13]. These stresses can lead to temperature-induced cracks in the concrete, diminishing the aqueduct’s load-bearing capacity, accelerating steel corrosion, and severely compromising its service life [14,15].

The thermodynamic behavior of the U-shaped aqueduct is more complicated than that of a rectangular/box-shaped aqueduct due to its special geometric shape (the bottom is circular and thick to bear the bending moment, and the side wall becomes thinner upwards): the large amount of cement in the thick bottom area, the concentration of hydration heat, and the difficulty of heat dissipation lead to a high core temperature; the thin side wall has fast heat dissipation and low temperature, and the huge temperature difference between the inside and outside and the cross section causes temperature stress, which significantly increases the risk of cracking. At the same time, the U-shaped grooved bottom plate and the side wall form a strong integral shell structure through a smooth transition of continuous arc surface, and each part is strongly constrained by each other during temperature deformation. The rectangular/box aqueduct bottom plate and the side wall are mostly right-angled connections, the constraint complexity and continuity are low, and the temperature stress control is relatively easier. Due to the excessive internal and external temperature difference and the too fast cooling rate during the construction period, as well as the strong constraint at the intermittent surface, the concrete of the side wall of the Shahe aqueduct generates excessive horizontal temperature tensile stress on its inner surface, causing the inner side of the side wall to crack, which poses a certain threat to the safe operation of the aqueduct [16]. This project case highlights the engineering urgency necessary for the analysis of the hydration heat effect of the U-shaped thin-walled aqueduct and the study of temperature control measures.

Based on a 42 m span U-shaped aqueduct as the engineering background, this study monitors and analyzes the hydration heat temperature field and stress field during the pouring stage of the U-shaped aqueduct through a combination of field experiments and numerical simulations to provide references for related engineering projects.

2. Project Overview

A U-shaped C50 prestressed concrete aqueduct body with a simply supported beam-type structural arrangement is employed for this aqueduct, featuring a trough width of 6.4 m, a trough depth of 5.57 m, and a trough wall thickness of 0.35 m. The bottom of the trough body includes a thickened zone measuring 1 m in thickness. The end ribs have a length of 1.48 m, with 1 m post-cast strips set at both ends. The transition section has a width of 3 m. Transverse square tie rods with a side length of 0.5 m (width) × 0.4 m (height) are arranged along the longitudinal direction at the top, spaced at intervals of 2.5 m. Based on the structural bearing and durability requirements, C50 concrete is selected for the aqueduct body, and its 28 d cube compressive strength can meet the bearing requirements under the combined action of the prestressed system, self-weight, and water load. The durability design of impermeability grade P8 is adapted to the long-term water environment service characteristics of the aqueduct. The size design is verified by multi-objective optimization. The groove width and groove depth parameters ensure the uniformity of flow velocity under the design flow rate through hydraulic calculation. The stability of the 0.35 m thin-walled side wall under the action of water pressure is verified by finite element analysis. The thickening area of the 1 m bottom plate is oriented to strengthen the bending moment concentration characteristics of the bearing. All structural dimensions are demonstrated by the special demonstration of structural bearing capacity to achieve the balance between hydraulic performance and structural safety. A cross-sectional schematic diagram is shown in Figure 1.

The aqueduct trough span structure is constructed using a cast-in-place bracket method. The bracket is designed as a beam–column structure composed of steel pipe pier supports and Bailey beams. All outer formworks are made of steel formworks. First, the steel formworks are erected, followed by the binding of steel bars, and, finally, the concrete is poured. After the initial setting of the concrete, water curing is applied. Then, the prestressing tensioning operation is conducted. The concrete pouring technique of, “placing concrete from one end to the other end, with a staggered-symmetrical-layered-segmented distribution of concrete for the left and right webs and the bottom slab”, is adopted.

It is necessary to monitor the temperature of the concrete surface layer, middle layer, and bottom layer, and take appropriate corrective measures when deviations in the measured data are observed to ensure dynamic management of temperature control during the concrete construction of the aqueduct.

The layout principle of the temperature test elements in this field test are as follows:

• The half axis of the plane symmetry axis of the concrete pouring body can be selected in the test area, and the monitoring points in the test area should be arranged according to the plane stratification.
• In the test area, the location and number of monitoring points can be determined according to the distribution of temperature field in the concrete pouring body and the provisions of temperature control.
• On each test axis, the monitoring points should not be less than four, and should be arranged according to the plane size of the structure.
• Along the thickness direction of the concrete pouring body, at least surface, bottom, and center temperature measuring points should be arranged, and the spacing of measuring points should not be greater than 500 mm.
• The effect of thermal insulation maintenance and the number of environmental temperature monitoring points should be determined according to specific needs.
• The surface temperature of the concrete pouring body should be the temperature of 50 mm within the surface of the concrete pouring body.
• The bottom temperature of the concrete pouring body should be the temperature at 50 mm above the bottom of the concrete pouring body.

The on-site test-temperature measuring elements are arranged as follows:

• Five and seven measuring points are arranged along the thickness direction at the upper and lower parts of the left and right webs of the bearing section, respectively, and are numbered incrementally from left to right.
• Three and four measuring points are arranged along the thickness direction at the upper and lower parts of the left and right webs of the mid-span section, respectively, and are numbered incrementally from left to right.
• Three measuring points are arranged along the thickness direction at the bisector point of the mid-span reinforced concrete tie rod and are numbered incrementally from bottom to top. In addition, nine measuring points are arranged along the thickness direction at the bisector point in the transverse groove direction of the bearing bottom slab, numbered incrementally from bottom to top.
• Six measuring points are arranged along the thickness direction at the bisector point in the transverse groove direction of the mid-span bottom slab and are numbered incrementally from bottom to top.

Since the U-shaped aqueduct is a centrally symmetric structure, only the test data and simulation data from the left half of the aqueduct are considered in the analysis of the hydration heat results. The measurement positions are specified as follows: the location of T1-1–3 (inside the tie rod) is designated as measuring position 1; T1-4–6 as measuring position 2; T1-7–10 as measuring position 3; T1-11–16 as measuring position 4; T2-1–5 as measuring position 5; T2-6–12 as measuring position 6; and T2-13–21 as measuring position 7. It is necessary to ensure that the maximum thermal stress remains below the allowable tensile strength during the analysis to prevent structural cracking. Therefore, it is essential to monitor the thermal stress generated during the hydration heat process of concrete and to provide real-time feedback and early warning. The temperature elements are arranged as shown in Figure 2. The strain monitoring elements for the field test are arranged as shown in Figure 3. The construction photos, temperature elements, and strain monitoring elements of the U-shaped aqueduct are shown in Figure 4.

The project is located in Nanchong City, Sichuan Province. The concrete pouring time is April 2024. According to the measured data, the measured temperature of the concrete during the pouring period is 20 °C, and the average wind speed is 3 m/s.

3. Finite Element Simulation Analysis

3.1. Model Establishment

The hydration heat analysis module of Midas/FEA NX 2024 software was utilized for modeling to study the hydration heat temperature field and stress field of the U-shaped aqueduct. Given that the number of solid elements in the meshed U-shaped aqueduct structure is substantial, a 1/4 finite element model was developed based on the principle of symmetry to enhance computational efficiency, as illustrated in Figure 5. The model comprises 146,265 nodes and 141,704 solid elements. The X-axis is oriented along the bridge direction, with a length of 20.98 m; the Y-axis is aligned with the transverse groove direction, with a width of 8.0 m; and the Z-axis represents the vertical groove direction.

The modeling assumptions are as follows: (1) the thermal parameters of the concrete material, such as specific heat and thermal conductivity, remain constant; (2) the concrete is considered a homogeneous and isotropic material; (3) the influence of the steel mesh is disregarded; and (4) based on the temperature monitoring data recorded at the beginning of pouring, the concrete casting temperature is assumed to be 20 °C.

3.2. Boundary Conditions

The cast-in-place construction method on supports is adopted; therefore, translational constraints in the Z-direction are applied at the base of the bottom slab in the model. In addition, since the structure is simplified as a 1/4 symmetric model, translational degrees of freedom in the X-direction are constrained on the symmetric surface perpendicular to the X-axis, while translational degrees of freedom in the Y-direction are constrained on the symmetric surface perpendicular to the Y-axis.

In addition to the aforementioned constraints, convection boundaries must be defined to represent the heat exchange between the concrete surface and the surrounding fluid caused by temperature differences. During the pouring stage of the aqueduct, the structure is enclosed by steel formwork. Except for the upper surface and the symmetric boundaries, all surfaces involved in convection with the external environment are in contact with the steel formwork. In the symmetric model, the symmetric surface is treated as an adiabatic boundary without the assignment of convection or fixed temperature conditions.

3.3. Thermal Parameter Calculation

When heat loss during the construction process of the concrete structure is not considered, meaning all heat generated from cement hydration is assumed to contribute to structural heating, the estimation formulas for the absolute temperature rise, maximum temperature, and internal and surface maximum temperatures due to concrete hydration heat are applied as follows:

$$T_{\max }={\displaystyle \frac{\left(m_{\mathrm{c}}+K·F\right)Q}{c·\rho }}$$

(1)

where T_max is the maximum hydration heat temperature rise of concrete, °C; m_c is the cement content per cubic meter of concrete, kg/m³; K is the adjustment coefficient for hydration heat of different admixture contents; F is the amount of active admixtures in concrete, kg/m³; Q is the hydration heat per kilogram of cement, kJ/kg; c is the specific heat capacity of concrete, kJ/kg▪°C; and ρ is the mass density of concrete, kg/m³.

C50 concrete is selected for pouring in this aqueduct, and the admixture in the concrete is fly ash. The adiabatic temperature rise is calculated as follows:

$$T_{\max }={\displaystyle \frac{\left(400+0.25\times 71\right)\times 377}{0.96\times 2400}}=67.99 °\mathrm{C}$$

The convection coefficient of the concrete surface in the air can be calculated using the following formula:

$${\beta }_{\mathrm{q}}=21.8+13.53{\upsilon }_{\mathrm{a}}$$

(2)

where υ_a is the wind speed, m/s.

When formwork or an insulation layer is present on the concrete surface, its influence must be considered, and the convection coefficient must be adjusted accordingly. The equivalent convection coefficient β between the concrete surface and the air, accounting for the formwork or insulation layer, is computed using the following formula:

$$\beta ={\displaystyle \frac{1}{{\displaystyle \sum \frac{{\delta }_{\mathrm{i}}}{{\lambda }_{\mathrm{i}}}+\frac{1}{{\beta }_{\mathrm{q}}}}}}$$

(3)

where β_q is the convection coefficient of the concrete surface in the air, kJ/(m²▪h▪°C); δ_i is the thickness of each insulation layer material, m; and λ_i is the thermal conductivity of each insulation layer material, kJ/(m²▪h▪°C).

Based on the relevant data and observational data analysis, the wind speed is considered as 3 m/s in the calculation of this aqueduct project, and steel formwork is utilized for the construction of the U-shaped aqueduct surface. The thermal conductivity of the steel formwork is 163.29 kJ/(m▪h▪°C), and the calculated convection coefficient is 61.68 kJ/(m▪h▪°C).

3.4. Division of Pouring Stages

The U-shaped aqueduct adopts the concrete pouring technology of, “staggered from one end to the other end, left and right webs and bottom plates-symmetry-stratified-segmented distribution”. In the Midas/FEA NX 2024 software simulation construction stage division, the U-shaped aqueduct is divided into three construction stages to simulate the pouring of aqueduct concrete. The first stage is to pour a 1/4 circle, the second stage is to pour a complete semicircle, and the third stage is to pour the vertical section, that is, to complete the pouring (this does not simulate the pull rod). According to the actual situation of the site, the simulation pouring takes 16 h, 6 h, and 2 h.

4. Analysis of Measured Results

4.1. Measured Temperature-Time History

The range of the temperature element of the resistance type temperature element selected in this test is −30~70 °C, and the accuracy is ±0.5 °C. Under the environment of 25 °C, the temperature measurement error is not more than 0.3 °C, and the temperature element is installed with protective measures, which can meet the requirements of data acquisition and storage for more than 20 days of continuous testing.

The measured temperature variation curves at each measuring point are shown in Figure 6a–f. The overall temperature of the concrete at the T2-17 measuring point, located at the thickness center of measuring position 7, is the highest, with a maximum temperature of 82.7 °C occurring 23 h after pouring. Although the U-shaped aqueduct is a thin-walled structure, the large local concrete pouring volume necessitates corresponding temperature control measures during the construction process.

In general, the measured temperature variation curves at each measuring point initially rises rapidly, then decreases rapidly, and eventually stabilizes to the atmospheric temperature. The temperature of the inner measuring points is higher than that of the outer measuring points. Due to the influence of ambient temperature changes, the temperature fluctuation at the surface measuring points is more pronounced. For measuring points located in symmetrical positions, since steel formwork serves as the convection boundary on both sides, the temperature changes tend to be consistent.

4.2. Measured Stress-Time History

In this test, the accuracy of steel bar gauge and concrete strain gauge are 0.5% F.S and 0.1% F.S, respectively, and the errors are within the allowable range of instrument calibration.

The stress monitoring data from the longitudinal measuring points on the web plate and the circumferential measuring points on the bottom slab of each cross-section were extracted for analysis, and the comparison between the measured stress and the allowable tensile strength is presented in Figure 7a–f.

The following can be seen:

• During the concrete pouring process of the U-shaped aqueduct, the stress at the surface of the concrete aqueduct is tensile. As the hydration reaction progresses, the tensile stress gradually increases and then decreases. In the later stages of concrete pouring, thermal stress fluctuates, which is mainly caused by temperature differences.
• The variation trend of the stress values at each internal measuring point of the U-shaped aqueduct is consistent with the temperature variation trend at the corresponding locations. In the first two days, the compressive stress increases rapidly with the rise in hydration heat temperature. From the third day onwards, the compressive stress gradually decreases, and internal stress changes become stable in the following days.
• The internal measuring points of the aqueduct are subjected to certain compressive stresses during the initial temperature rise stage, with the mid-span section experiencing higher compression and the beam end experiencing lower compression.

5. Analysis of Simulation Results

5.1. Thermal Field Simulation Analysis

5.1.1. Overall Temperature Distribution

After applying the specified material parameters and boundary conditions to the finite element model of the U-shaped aqueduct, the transient temperature field after pouring was obtained using the thermal analysis solution function of Midas/FEA NX finite element software. The analysis time for the hydration heat temperature field of the integrally poured aqueduct was 192 h. The structural temperature distribution nephograms at 32 h and 168 h after the start of pouring were extracted, as shown in Figure 8a,b.

The following can be seen:

• The temperature nephograms at each time generally exhibit a distribution of high internal temperature and low external temperature. The overall trend is characterized by rapid heating, slow cooling, and the eventual stabilization of the ambient temperature.
• During the temperature rise stage, following the pouring of the aqueduct concrete, the cement hydration releases a large amount of hydration heat, causing the internal temperature of the concrete to rise rapidly and reach a peak. The highest temperature is 87.2 °C, occurring 32 h after pouring.
• The highest temperature appears inside the bottom slab of the aqueduct beam section with a uniform cross-section of 2.48 m at the beam end. This section has a horizontal trough width of 6.5 m and a vertical trough thickness of 1.95 m, representing the thickest part during the overall casting stage. The large pouring thickness results in a high degree of hydration heat and limited internal heat dissipation.
• Along the vertical direction of the aqueduct, the temperature is highest at the middle position and gradually decreases vertically toward the inner and outer sides. The temperature contour lines of the cross-section are distributed in an annular pattern.

5.1.2. Comparison Between Thermal Field Simulation and Measured Results

The arrangement of temperature elements is described in Section 1. The temperature calculation results for nodes located at the same position in the model within 192 h after the start of pouring were extracted. Two measuring points (the central measuring point and one outermost measuring point) were selected at each measuring position, and their comparison to the measured results is presented in Figure 9a–f.

Figure 8 indicates the following:

• The measured curve is generally consistent with the model temperature curve. The temperature variation curves of the concrete at each measuring point exhibit a pattern of, “rapid rise → rapid decrease → alignment with external air temperature and tendency toward atmospheric temperature”.
• Both the field test and numerical simulation results demonstrate that the central T2-17 measuring point at measuring position 7 (i.e., the bottom slab of the bearing section) reaches the highest temperature, with maximum values of 83.8 °C and 87.2 °C, respectively. This measuring position does not satisfy the requirement of Section 3.0.4 of the “Technical Specification for Temperature Monitoring and Control of Mass Concrete” (GB/T 51028—2015) [17], which stipulates that, “the temperature rise of the concrete pouring body based on the mold entry temperature should not exceed 50 °C”.
• The temperature of the inner measuring points is higher than that of the outer measuring points. This occurs because concrete is a poor thermal conductor, and the heat generated by the cement hydration reaction tends to accumulate internally and is not easily dissipated. The surface measuring points of the concrete are affected by both internal hydration heat and heat exchange with the air, making them more susceptible to atmospheric temperature variations.

A comparison of the maximum temperature reached by the concrete at the central measuring points of each measuring position under simulation and measured conditions, as well as the time at which the maximum temperature is reached, is presented in

Table 1. Comparison of simulated and measured highest temperature at the central measuring point.

Position	Point Number	Maximum Temperature (°C)			Time of Max Temperature (h)
		Emulation	Measurement	Temp Difference	Emulation	Measurement	Time Gap
2	T1-5	55.0	63.5	8.5	30	30	0
3	T1-9	69.1	62.3	6.8	20	25	5
4	T1-13	74.3	75.4	1.1	20	20	0
5	T2-3	70.0	77.6	7.6	36	31	5
6	T2-9	76.0	81.5	5.5	36	32	4
7	T2-17	87.0	82.7	4.3	28	23	5

.

Table 1 reveals the following:

• The average temperature difference between the simulated and measured values at each measuring point ranges from 1.0 °C to 8.5 °C, with the peak temperature time deviation remaining within 5 h.
• The simulated maximum temperature is lower than the measured value, primarily due to parameter settings. For instance, the concrete pouring temperature in the simulation is set at 20 °C, whereas, in actual construction, the pouring temperature frequently exceeds 20 °C. Excluding these factors, the simulation demonstrates satisfactory performance, and the results can be utilized to predict the risk of concrete cracking induced by hydration heat.

5.2. Thermal Stress Field Simulation Analysis

5.2.1. Temperature Stress Distribution

Finite element simulation was utilized to analyze the principal stress fields at several representative time points. The principal stress fields induced by hydration heat at 32 h and 168 h after concrete placement are shown in Figure 10a,b, where tensile stress is indicated as positive and compressive stress as negative. The following observations were made:

• During the temperature rise phase of the U-shaped aqueduct concrete, a considerable amount of is heat generated by cement hydration, leading to an increase in internal temperature. As a result, thermal expansion occurs, producing compressive stress in the interior and tensile stress on the surface of the aqueduct.
• During the temperature rise phase, the local maximum tensile stress of 6.37 MPa appears at 32 h after placement. The region of maximum stress, shown in Figure 9a, is characterized by significant steel formwork constraints and abrupt structural dimensional changes.
• During the temperature drop phase of the U-shaped aqueduct concrete, as the internal and external temperatures gradually decrease, the concrete volume contracts. This shrinkage generates tensile stress in the interior, gradually offsetting the compressive stress formed during the temperature rise. Therefore, the compressive stress inside the aqueduct decreases to zero and rapidly transitions to tensile stress.

5.2.2. Comparison Between Thermal Field Simulation and Measured Results

Nodes at the same locations in the model as the longitudinal measurement points on the web and the circumferential measurement points on the base slab were extracted to examine the temperature stress variation pattern in aqueduct concrete. The results calculated 192 h after the start of placement are displayed in Figure 11a–f.

The following conclusions can be drawn from Figure 10:

• Analysis of stress variation at the mid-span and 1/4-span stress monitoring nodes shows that the stress values at both locations follow similar trends within 8 days of placement.
• The stress values at the mid-span and end-section stress monitoring nodes remain below the allowable tensile strength of the concrete at that stage, indicating that the U-shaped aqueduct satisfies the crack resistance requirements.
• A deviation exists between the simulated and measured values. Possible reasons for this discrepancy include instrumental errors and environmental interference, as well as the effects of concrete shrinkage, creep, and ambient temperature on the collected strain data. These factors can result in the measured stress being higher than the actual stress within the aqueduct. Nonetheless, the overall trend and magnitude of both results are consistent, further confirming the accuracy of the simulation model.

6. Optimization Analysis of Hydration Heat Pipe Cooling System

6.1. Cooling Pipe Layout and Key Parameters

The pipe cooling system demonstrates strong controllability, allowing effective temperature regulation by adjusting parameters such as pipe diameter, water temperature, and water flow rate, which leads to significant cooling effects. Due to the densely packed reinforcement in the upper sections of the web and base slab, the cooling pipes are strategically positioned in the middle of the Jinjiang aqueduct’s base slab.

A model incorporating the pipe cooling system was developed using the finite element software Midas/FEA NX. Low-temperature cooling water was circulated through the system to reduce the temperature rise caused by concrete hydration heat via thermal exchange between the concrete and the low-temperature fluid. The Midas/FEA NX analysis model is shown in Figure 12.

During the calculations, the effect of reinforcement bars in the concrete, the damping coefficients of the pipes, and head losses were neglected. The cooling water was considered a one-dimensional, steady, and uniform flow. The convective heat transfer between the cooling pipes and the concrete comprises two processes: thermal conduction from the concrete surface to the water flow and the movement of thermal fluid elements. Accordingly, mesh generation and node coupling for the cooling pipe thermal flow elements were completed to enable heat exchange exclusively between the primary concrete nodes and the additional nodes of the thermal flow elements.

Assuming turbulent flow within the condenser pipes and neglecting the thermal resistance of the metallic condenser pipes, the cooling water indirectly transferred heat to the concrete through the cold-water pipes. Therefore, the convective heat transfer coefficient of the water was adjusted, adopting a value of 2000 kJ/(m²▪h▪°C). The cooling water temperature was set to 20 °C. The cooling pipes were made of galvanized stainless steel with a wall thickness of 2.5 mm and a diameter of 40 mm, and the flow velocity was maintained at 1.2 m/s.

6.2. Optimization Simulation Results

After arranging the cold-water pipes in the finite element model of the U-shaped aqueduct, the structural temperature distribution nephograms at 24 h and 192 h after pouring were extracted and are shown in Figure 13.

As observed from the overall temperature nephogram, this construction scheme reduces the maximum temperature by 23.8 °C and enhances the cooling effect by 28.7%. The longitudinal placement of the cold-water pipes at the center of the aqueduct base slab effectively lowers the temperature at the core of the concrete, minimizing the internal and external temperature differential. The temperature peak is reached at 30 h. The aqueduct beam-end base slab has a thickness of 1.95 m, and without the inclusion of cooling pipes, the internal temperature can reach up to 82.7 °C. This demonstrates the feasibility of the construction scheme. In addition, the cold-water circulation equipment can be reused, which significantly shortens the construction period, reduces project costs, and effectively diminishes the temperature gradient, mitigating the occurrence of temperature-induced cracks.

The principal stress field nephograms at several representative time points of the finite element simulation are selected, and the analysis results of the hydration heat stress field at 24 h and 192 h after pouring are extracted and are shown, respectively, with tensile stress represented as positive and compressive stress as negative, as illustrated in Figure 14.

The stress states of the concrete in the mid-span section and the end section of the aqueduct meet the structural safety requirements, indicating that the use of cold-water pipes can effectively prevent the occurrence of temperature cracks.

7. Conclusions and Discussion

The hydration heat monitoring and numerical simulation analysis of the aqueduct body are conducted based on a 42 m span U-shaped aqueduct as the engineering background. The finite element software Midas/FEA NX is utilized to calculate the temperature field and thermal stress of the U-shaped aqueduct. The calculated results are then compared to the measured data to validate the pattern of hydration heat temperature variation and the distribution of the hydration heat stress field within the aqueduct. The main conclusions are as follows:

• During the layered pouring stage, the maximum concrete temperature of the U-shaped aqueduct reached 87.2 °C, occurring at 32 h after pouring. This peak was located at the center of the bottom slab (with a thickness of 1.95 m) in the constant-section beam segment 2.48 m from the beam end of the aqueduct body. For measuring point 7 (bottom slab of the support section), the central temperature was also as high as 83.8 °C (measured) and 87.0 °C (simulated). Due to the difficulty in dissipating hydration heat in the central area of such large-thickness sections, these regions have become key focus areas for construction temperature control.
• The concrete temperature exhibited a variation pattern of, “rapid temperature rise—rapid temperature drop—stabilization to ambient temperature”. Temperatures at inner-layer measuring points were generally higher; temperatures at symmetric measuring points were significantly affected by boundary conditions; and temperatures at outer-layer measuring points fluctuated more obviously with ambient temperature.
• During the temperature rise stage, relatively large tensile stresses appeared locally in the beam end section. Although these stresses did not exceed the allowable tensile strength of the concrete at that stage, both on-site monitoring and simulations showed that during the first 5–6 days of the cooling stage the daily temperature drop rate exceeded 2.0 °C/d, and the temperature difference between inner and surface layers exceeded 25 °C. There is a risk of cracking caused by excessively rapid cooling or excessive temperature differences.
• The reference value of these research results for other aqueduct projects needs to be adjusted according to specific scenarios. Regarding differences in aqueduct types, for aqueducts with other cross-sections such as rectangular and trapezoidal ones, due to differences in heat dissipation area and thickness distribution (e.g., rectangular aqueducts have more uniform thickness distribution between side walls and bottom slabs), the location of the temperature peak may shift from the “center of thickness” to “sectional abrupt change areas”. Thus, targeted optimization of monitoring points is required. In terms of climate conditions, in high-temperature and high-humidity areas, the concrete heat dissipation rate decreases, leading to potentially higher and longer-lasting temperature peaks. Enhanced measures such as water cooling or surface insulation should be implemented. In cold regions, it is necessary to balance the needs of “freezing prevention” and “temperature control” to avoid excessive insulation causing the temperature difference between inner and surface layers to exceed the limit.
• Temperature control optimization for different engineering conditions can be carried out from three aspects: Adaptation of concrete grades: High-grade concrete (e.g., C50 and above) has a higher peak hydration heat. It is recommended to use mineral admixtures (fly ash, slag) to replace cement to reduce hydration heat and adjust the mix proportion to reduce the unit cement dosage. For low-grade concrete (e.g., C30), the focus can be on optimizing the pouring interval to control stress using early strength growth. Optimization of cross-section design: For large-thickness sections (>1.5 m), it is recommended to set heat dissipation holes or embedded cooling pipes to reduce the central temperature through active temperature control. For small-thickness sections (<0.8 m), surface moisture retention should be strengthened to avoid surface cracking caused by abrupt changes in ambient temperature. Adaptation to geographical locations: In high-altitude areas with large day-night temperature differences, temperature-responsive insulation materials (e.g., phase change insulation layers) should be used. In rainy areas, the timing of formwork removal should be optimized to prevent direct rainwater scouring on the surface of high-temperature concrete, which may induce temperature difference stress.
• Based on the research results, a temperature control decision-making framework for aqueduct projects is proposed as follows: Identification of key sections: Prioritize locating large-section areas with a thickness > 1.5 m (e.g., support bottom slabs, constant-section segments of beam ends). Predict the location of temperature peaks using finite element simulation and arrange multi-layer temperature sensors. Coupling of climate parameters: Collect meteorological data (daily maximum temperature, wind speed, humidity) of the project area for more than 5 years before construction and establish a “climate-heat dissipation rate” correlation model to predict risk periods of temperature drop. Dynamic control indicators: Take “temperature difference between inner and surface layers ≤ 25 °C and daily temperature drop rate ≤ 2.0 °C/d” as core thresholds, and formulate hierarchical measures for different stages: strengthen insulation (e.g., covering with flame-retardant quilts) during the temperature rise stage; adopt gradient removal of insulation layers combined with spray moisturizing during the cooling stage; and immediately activate water cooling when thresholds are exceeded. Coordination of materials and processes: Select the type and dosage of admixtures according to the concrete grade and optimize the layered pouring thickness and intermittent time based on cross-section size to reduce interlayer temperature difference stress.