Intelligent temperature control method and system for concrete pouring of large-size bearing platform

Abstract

The invention is suitable for the technical field of building construction, and provides an intelligent temperature control method and system for large-size bearing platform concrete pouring, and the method comprises the following steps: collecting concrete internal temperature data and environment temperature data in real time; monitoring inlet and outlet water temperature and water pipe flow of the cooling water pipe; performing statistical analysis on the collected temperature data based on a machine learning algorithm, and identifying a temperature change trend and an abnormal point; according to the statistical analysis result, an optimal temperature control scheme is automatically generated, and the optimal temperature control scheme comprises cooling water flow adjustment, a mold entering temperature control target and an inside and outside temperature difference control target; determining a raw material temperature control scheme and a concrete mix proportion according to the environment temperature data; and the concrete pouring progress is determined, and the vibrating time and position of the insertion type vibrator are controlled so as to ensure the vibrating quality. According to the method, the accuracy and reliability of temperature change trend prediction are remarkably improved, and a more scientific and accurate basis is provided for subsequent temperature control decision making.

Description

Technical Field

[0001] This invention relates to the field of building construction technology, specifically to an intelligent temperature control method and system for pouring large-volume foundation concrete. Background Technology

[0002] In large-scale infrastructure construction, the pouring of large-volume concrete foundations is a common and critical construction step. Due to their large structural dimensions, large-volume concrete releases a significant amount of heat during cement hydration. Concrete is a poor conductor of heat, making heat difficult to dissipate, leading to a rapid increase in the internal temperature of the concrete. Simultaneously, the concrete surface dissipates heat quickly, creating a large temperature difference between the interior and surface. This temperature stress can cause cracks in the concrete, severely impacting the structural safety and durability of the foundation, reducing the project's service life, and potentially even causing safety accidents.

[0003] Traditional methods of concrete temperature control primarily rely on manual experience for temperature monitoring and regulation. This involves embedding temperature sensors within the concrete and periodically reading temperature data, then judging whether adjustments are needed based on experience. However, this method has several limitations. Manual monitoring data is not only infrequent but also fails to comprehensively and accurately reflect the dynamic changes in the concrete's internal temperature. Furthermore, manual data analysis and temperature control plan development are inefficient, unable to respond promptly to rapid temperature changes, and thus hinder precise temperature control management.

[0004] Therefore, there is a need to provide an intelligent temperature control method and system for large-volume foundation concrete pouring, which aims to solve the above problems. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of this invention is to provide an intelligent temperature control method and system for large-volume foundation concrete pouring, so as to solve the problems existing in the above-mentioned background technology.

[0006] This invention is implemented as follows: a smart temperature control method for pouring large-volume foundation concrete, the method comprising the following steps: Real-time acquisition of concrete internal temperature data and ambient temperature data; monitoring of inlet and outlet water temperatures and flow rates in cooling water pipes; The collected temperature data is statistically analyzed using machine learning algorithms to identify temperature change trends and anomalies. Based on the statistical analysis results, the optimal temperature control scheme is automatically generated. The optimal temperature control scheme includes cooling water flow adjustment, mold entry temperature control target, and internal and external temperature difference control target. Determine the raw material temperature control scheme and concrete mix proportion based on ambient temperature data; Determine the concrete pouring schedule and control the vibration time and position of the immersion vibrator to ensure vibration quality.

[0007] Another objective of this invention is to provide an intelligent temperature control system for large-volume foundation concrete pouring, the system comprising: The data monitoring and acquisition module is used to collect real-time data on the internal temperature of concrete and ambient temperature; and to monitor the inlet and outlet water temperatures and flow rates of cooling water pipes. The temperature change recognition module is used to perform statistical analysis on the collected temperature data based on machine learning algorithms to identify temperature change trends and anomalies. The optimal temperature control scheme module is used to automatically generate the optimal temperature control scheme based on statistical analysis results. The optimal temperature control scheme includes cooling water flow adjustment, mold entry temperature control target, and internal and external temperature difference control target. The raw material control module is used to determine the raw material temperature control scheme and concrete mix proportion based on ambient temperature data. The vibration control module is used to determine the concrete pouring progress and control the vibration time and position of the immersion vibrator to ensure the vibration quality.

[0008] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention constructs an enhanced sample set by coupling temperature characteristics with cooling flow data and uses a sliding time window to divide time-series samples, providing a training foundation for the model that reflects the "regulation-response" relationship. By training a long short-term memory network and a random forest model in parallel and establishing online evaluation and switching logic, it ensures that the system can adaptively select the model with higher prediction accuracy under the current data characteristics as the dominant model. Furthermore, by introducing a model rolling update mechanism, the dominant model parameters are continuously fine-tuned using real-time data, enabling the temperature prediction model to dynamically adapt to the complex and changing conditions at the construction site. Ultimately, this significantly improves the accuracy and reliability of temperature change trend prediction, providing a more scientific and accurate basis for subsequent temperature control decisions.

[0009] 2. This invention utilizes the isolated forest algorithm and the self-organizing map neural network model to detect and fuse global and local anomaly patterns in parallel, achieving multi-dimensional and highly sensitive identification of temperature anomalies. By correlating anomalies with construction logs and environmental parameters in a multi-dimensional manner, it achieves a leap from "discovering anomalies" to "diagnosing the cause." Crucially, by cross-validating the preliminary diagnostic conclusions with the prediction results of the current dominant prediction model, a closed-loop mechanism of mutual verification between "diagnosis and prediction" is constructed. This not only significantly improves the confidence of the anomaly diagnosis conclusions but also feeds back the verified diagnostic conclusions to optimize the prediction model itself. This enables the entire system to have the continuous evolutionary ability to learn from anomalies and self-correct, greatly enhancing the robustness and intelligence of the intelligent temperature control system.

[0010] 3. This invention combines vibration parameters with casting geometry for simulation analysis, generating a vibration parameter-density effect mapping table. Its reliability is verified by comparing with real-time sensor data, establishing a vibration strategy library based on measured data correction. Furthermore, after selecting preliminary solutions through multi-objective optimization within the strategy library, the invention creatively integrates vibration strategies with predicted temperature change trends. Based on the different thermal characteristics of high-temperature accumulation zones and temperature difference-sensitive zones, regionally differentiated adjustments are made to vibration parameters, achieving deep synergy between vibration technology and temperature control. By optimizing concrete density and pore distribution at the microstructural level, it indirectly promotes heat dissipation and a uniform temperature field, jointly ensuring concrete quality from both structural and thermophysical performance dimensions, producing an unexpected synergistic enhancement effect.

[0011] 4. This invention extracts multi-dimensional features from real-time vibration data streams to form a quality evaluation vector, and compares it with preset standards to generate quantified quality evaluation results with spatial location labels, achieving precise and location-based perception of vibration effects. By correlating real-time evaluation results with expected targets in the validated vibration strategy library, defect areas are accurately identified. Furthermore, by combining predicted temperature change trends to perform secondary diagnosis of defects, adjustment instructions matching the regional thermal state are generated, realizing intelligent linkage between vibration quality control and temperature crack prevention. Finally, by feeding the adjustment effect back to the knowledge base and mapping table, a complete "perception-evaluation-diagnosis-adjustment-learning" closed loop is formed, enabling the system's vibration parameter recommendation and adjustment decision-making capabilities to continuously self-optimize as the project progresses, ultimately ensuring vibration quality and indirectly serving the core objective of temperature control. Attached Figure Description

[0012] Figure 1 A flowchart for an intelligent temperature control method for pouring large-volume foundation concrete.

[0013] Figure 2 A flowchart for identifying temperature change trends and anomalies in an intelligent temperature control method for large-volume foundation concrete pouring.

[0014] Figure 3 A flowchart for generating the optimal temperature control scheme in the intelligent temperature control method for large-volume foundation concrete pouring.

[0015] Figure 4 A flowchart for determining the raw material temperature control scheme in the intelligent temperature control method for large-volume foundation concrete pouring. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0018] like Figure 1 As shown in the figure, this invention provides an intelligent temperature control method for pouring large-volume foundation concrete, the method comprising the following steps: S100 collects real-time data on the internal temperature of concrete and ambient temperature; monitors the inlet and outlet water temperatures and flow rates of cooling water pipes. S200 uses machine learning algorithms to perform statistical analysis on the collected temperature data and identify temperature change trends and anomalies. S300 automatically generates the optimal temperature control scheme based on statistical analysis results. The optimal temperature control scheme includes cooling water flow adjustment, mold entry temperature control target, and internal and external temperature difference control target. S400, determine the raw material temperature control scheme and concrete mix proportion based on ambient temperature data; S500 determines the concrete pouring schedule and controls the vibration time and position of the immersion vibrator to ensure vibration quality.

[0019] It should be noted that traditional manual monitoring methods have a low data collection frequency, making it impossible to obtain real-time and comprehensive information on the dynamic changes in the internal temperature of concrete and the ambient temperature. This makes it difficult to detect temperature anomalies in a timely manner, resulting in delayed temperature control measures and an inability to effectively prevent cracking. Relying on manual experience to formulate temperature control schemes makes it difficult to conduct in-depth analysis and mining of large amounts of temperature data, and to accurately identify temperature change trends and anomalies. This results in a lack of scientific basis for the formulation of temperature control schemes, preventing precise adjustments based on actual conditions and leading to unsatisfactory temperature control effects. Traditional methods often consider the internal temperature of concrete in isolation when formulating temperature control schemes, without fully considering the interplay between multiple factors such as ambient temperature, raw material temperature, concrete mix proportions, and pouring progress, thus failing to achieve comprehensive and systematic temperature control management. The embodiments of this invention aim to solve the above problems.

[0020] In this embodiment of the invention, based on the symmetry of the foundation structure, a quarter of the structure is selected as the main testing area, where key temperature measurement points are arranged. Two high-precision digital temperature sensors are embedded at each key temperature measurement point to ensure data integrity. Φ42×3mm cooling water pipes are used, laid out according to the design drawings, ensuring reasonable pipe spacing to form an effective cooling network. A pressure test is performed on the water pipes before use to prevent leakage and water blockage. This embodiment of the invention collects real-time data on the internal temperature of the concrete and ambient temperature, automatically transmitting the data to a cloud server. Simultaneously, it monitors the inlet and outlet temperatures and flow rate of the cooling water pipes to ensure the normal operation of the cooling system. By collecting real-time data on the internal temperature of the concrete and ambient temperature, the dynamic changes in concrete temperature can be comprehensively and timely monitored, providing accurate and detailed data support for subsequent temperature control decisions. This helps to detect temperature anomalies in a timely manner and take preventative measures to prevent cracking.

[0021] Then, machine learning algorithms are used to statistically analyze the collected temperature data, identify temperature change trends and anomalies, and predict temperature changes over a future period, providing a basis for temperature control measures. Based on the statistical analysis results, an optimal temperature control scheme is automatically generated. This optimal scheme includes adjustments to cooling water flow, target temperature control for the formwork, and target temperature difference control between the inside and outside of the concrete. This achieves full-process monitoring and optimization of multiple targets, such as raw material temperature during the formwork stage, cooling water flow during the construction stage, and the internal and external temperature difference. The statistical analysis of the collected temperature data using machine learning algorithms can automatically identify temperature change trends and anomalies, and automatically generate the optimal temperature control scheme based on the analysis results. This method fully utilizes the advantages of data mining and intelligent algorithms, making the temperature control scheme more scientific and accurate. It can adjust the cooling water flow, target temperature control for the formwork, and target temperature difference control in real time according to actual conditions, effectively improving the temperature control effect. In addition, the raw material temperature control scheme and concrete mix proportions are determined based on ambient temperature data, such as adjusting cement dosage, water-cement ratio, and admixture proportions, to optimize concrete performance. Finally, the concrete pouring schedule is determined, and the vibration time and position of the immersion vibrator are controlled to ensure the vibration quality. By comprehensively considering the interaction between multiple factors, a comprehensive and systematic temperature control management is achieved, ensuring the quality of concrete and structural safety from multiple aspects.

[0022] like Figure 2 As shown, in a preferred embodiment of the present invention, the step of statistically analyzing the collected temperature data based on machine learning algorithms to identify temperature change trends and anomalies specifically includes: S201, clean the collected data, remove outliers, and perform smoothing filtering; S202, extract key features of temperature data, including heating rate, cooling rate, peak internal and external temperature difference, and temperature gradient distribution, to form a feature vector; S203: Based on historical engineering data, a training set is constructed to train a long short-term memory network or a random forest model to predict temperature change trends. S204 uses the isolated forest algorithm or self-organizing map neural network to identify temperature anomalies, and correlates them with construction logs and environmental parameters to diagnose the causes and determine the anomaly type.

[0023] In this embodiment of the invention, the collected data on the internal temperature of concrete, ambient temperature, inlet and outlet temperatures of cooling water, and flow rate are cleaned to remove outliers (such as jumps caused by sensor malfunctions) and smoothed. Then, key features of the temperature data are extracted, including heating rate, cooling rate, peak internal and external temperature difference, and temperature gradient distribution, forming a feature vector. A training set is then constructed based on historical engineering data to train a Long Short-Term Memory (LSTM) network or a random forest model to predict temperature change trends (such as the temperature evolution curve for the next 24 hours). Combined with real-time monitoring data, the prediction parameters are dynamically adjusted through a rolling model update mechanism to ensure that the prediction accuracy meets engineering requirements. Finally, an isolated forest algorithm or a self-organizing map neural network is used to identify temperature anomalies (such as localized overheating / overcooling areas), and these are correlated with construction logs and environmental parameters (such as wind speed and humidity) for causal diagnosis to determine the anomaly type, which includes cooling water pipe blockage, sensor drift, and sudden changes in pouring thickness.

[0024] As a preferred embodiment of the present invention, the step of constructing a training set based on historical engineering data, training a long short-term memory network or a random forest model, and predicting the temperature change trend specifically includes: S2031, The feature vector is time-series aligned and correlated with the cooling water pipe flow data of the corresponding historical time period to generate an enhanced training sample set; wherein, the enhanced training sample set is used to characterize the correlation between concrete temperature changes and active cooling measures; S2032, The enhanced training sample set is divided using a sliding time window method to generate a divided time-series training sample set; wherein, the divided time-series training sample set contains multiple training instances arranged in chronological order. S2033, The Long Short-Term Memory Network model is trained using the partitioned time-series training sample set to generate a first temperature prediction model; wherein, the first temperature prediction model is used to capture and predict the time-series dependency features in the temperature data. S2034, The random forest model is trained using the partitioned time-series training sample set to generate a second temperature prediction model; wherein, the second temperature prediction model is used to capture and predict the nonlinear relationship between temperature features; S2035, Establish online evaluation and switching logic for model performance, compare the prediction errors of the first temperature prediction model and the second temperature prediction model on recent validation data in real time, obtain the first comparison result, and select one of the first temperature prediction model and the second temperature prediction model as the current dominant prediction model based on the first comparison result. S2036, Start the model rolling update mechanism, use the latest collected and processed real-time temperature feature data as incremental data, fine-tune the parameters of the current dominant prediction model, and generate an optimized temperature prediction model. S2037, input the real-time collected concrete internal temperature data and ambient temperature data into the optimized temperature prediction model to obtain the output of the optimized temperature prediction model, and use the output of the optimized temperature prediction model as the predicted temperature change trend.

[0025] It should be noted that in the step of constructing a training set based on historical engineering data, training a long short-term memory network or random forest model, and predicting the temperature change trend, in this embodiment of the invention, for training the first temperature prediction model (long short-term memory network model), the partitioned time-series training sample set is input into the network, a time step is set to capture the short-term fluctuations and long-term evolution trends of temperature data, and the key information in the historical temperature sequence is selectively memorized and redundant information is forgotten using the gating mechanism of its internal forget gate, input gate, and output gate. The focus is on learning the time-series dependencies of features such as the heating rate and the peak internal and external temperature difference in the feature vector. The mean square error between the predicted temperature and the actual temperature is minimized by the backpropagation algorithm over time, and the network weights and bias parameters are iteratively optimized until the model loss function converges, generating a first temperature prediction model specifically for capturing the time-series dependency features of concrete temperature.

[0026] For training the second temperature prediction model (random forest model), multiple differentiated decision trees are constructed by randomly sampling with replacement from the partitioned time-series training sample set. Each tree uses only a subset of features for node splitting. During training of each decision tree, the optimal split point is found using criteria such as the Gini coefficient or information gain, ensuring that each generated tree learns the nonlinear relationships between temperature features from different perspectives. Finally, the prediction results of all decision trees (e.g., taking the average) are integrated as the model output to generate a second temperature prediction model specifically designed to capture the complex nonlinear mapping relationships between features such as temperature gradient distribution and cooling rate. The two models trained through this process complement each other in principle. The Long Short-Term Memory (LSTM) network model excels at capturing temporal patterns, while the random forest model excels at handling nonlinear features. This provides a foundation of candidate models with different strengths for subsequent online performance evaluation and switching, addressing the performance limitations of a single model in temperature prediction under complex construction environments.

[0027] In this embodiment of the invention, in the step of constructing a training set based on historical engineering data, training a long short-term memory network or random forest model, and predicting the temperature change trend, this invention constructs an enhanced training sample set that can characterize the dynamic relationship between "active cooling and temperature response" by temporally aligning and associating temperature feature vectors with historical cooling water pipe flow data. This overcomes the limitation of traditional prediction methods that isolate temperature data and ignore the influence of active regulation. By using a sliding time window method to divide the time-series samples, it ensures that the model can learn the continuous temporal dependence of the concrete hydration heat process.

[0028] By training a Long Short-Term Memory (LSTM) network model and a Random Forest model in parallel and establishing online evaluation and switching logic, the model with smaller prediction error is selected as the dominant model in real time, thus solving the problem of insufficient generalization ability of a single model under complex working conditions. Furthermore, by introducing a rolling model update mechanism, the parameters of the dominant prediction model are fine-tuned using the latest collected real-time temperature feature data, enabling the prediction model to dynamically adapt to changes in material properties and environmental conditions. Ultimately, this significantly improves the accuracy of temperature change trend prediction and its adaptability to dynamic changes at the construction site.

[0029] As a preferred embodiment of the present invention, the step of identifying temperature anomalies using the isolated forest algorithm or self-organizing map neural network, and correlating them with construction logs and environmental parameters for causal diagnosis to determine the anomaly type specifically includes: S2041, Synchronize the real-time collected concrete internal temperature data with the feature vector to construct a real-time temperature monitoring dataset; S2042, the real-time temperature monitoring dataset is simultaneously input into a pre-trained isolated forest algorithm model and a self-organizing map neural network model to perform anomaly detection in parallel, thereby obtaining global anomaly points output by the isolated forest algorithm model and local anomaly patterns output by the self-organizing map neural network model; wherein, the isolated forest algorithm model is used to identify global anomaly points that deviate from the overall temperature pattern, and the self-organizing map neural network model is used to identify local anomaly patterns that aggregate in the temperature gradient distribution feature space. S2043, The global outliers output by the isolated forest algorithm model and the local outlier patterns output by the self-organizing map neural network model are fused to generate a comprehensive list of outliers; wherein, the outliers in the comprehensive list of outliers are used to mark risk areas where there are temperature anomalies in the spatial and temporal dimensions. S2044, perform multi-dimensional correlation matching between the comprehensive list of abnormal points and the construction logs and historical environmental parameter data of the same time period to obtain the correlation matching results; wherein, the construction log records include pouring progress, vibration operation records and cooling water pipe valve operation records; the historical environmental parameter data includes ambient temperature and wind speed; S2045, Based on the results of the association matching, execute the anomaly cause diagnosis logic, map the anomaly point to the construction event or environmental condition that is most likely to cause the anomaly point, and generate a preliminary anomaly cause diagnosis report; the anomaly types include: cooling water pipe blockage anomaly, sensor drift anomaly, sudden change in pouring thickness anomaly, and insulation measure failure anomaly. S2046, Cross-validate the preliminary abnormality cause diagnosis report with the predicted temperature change trend to generate the diagnosed abnormality type and cause diagnosis conclusion; wherein, the cross-validation includes: comparing the time and spatial location of each abnormal point in the comprehensive abnormal point list with the predicted potential risk areas in the predicted temperature change trend; when the compared abnormal point is in the predicted temperature change trend, the area corresponding to the compared abnormal point is identified as a risk trend of continuous temperature rise or excessive internal and external temperature difference, so as to generate the diagnosed abnormality type and cause diagnosis conclusion; S2047, The abnormality type and cause diagnosis conclusion are fed back to the first temperature prediction model and the second temperature prediction model as incremental data as part of the model rolling update mechanism, which is used to correct the current dominant prediction model.

[0030] In this embodiment of the invention, in the step of identifying temperature anomalies using the isolated forest algorithm or self-organizing map neural network, and performing cause diagnosis by associating construction logs and environmental parameters to determine the type of anomaly, this invention utilizes the isolated forest algorithm model to identify global anomalies and the self-organizing map neural network model to capture local anomaly patterns, and fuses them to generate a comprehensive list of anomaly locations. This achieves multi-dimensional and highly robust perception of temperature anomalies, overcoming the potential for missed or false detections that may occur with a single algorithm.

[0031] By cross-referencing and matching the anomaly list with construction logs and environmental parameters across multiple dimensions, and mapping anomaly locations to specific construction events (such as cooling water pipe valve operation or vibration work) or environmental conditions (such as sudden changes in wind speed), a leap from "anomaly detection" to "cause diagnosis" is achieved. Crucially, by cross-validating the preliminary diagnostic conclusions with the risk trends output by the current dominant prediction model, a closed loop of mutual verification between "diagnosis" and "prediction" is constructed. The verified diagnostic conclusions are then confirmed, which not only improves the reliability of the diagnosis but also feeds the confirmed conclusions back as incremental data to the prediction model for self-correction. This enables the system to continuously learn from anomalies and evolve, greatly enhancing the reliability and intelligence of intelligent diagnosis.

[0032] like Figure 3 As shown, in a preferred embodiment of the present invention, the step of automatically generating the optimal temperature control scheme specifically includes: S301, determine the mold entry temperature control target and the internal and external temperature difference control target based on the temperature change trend and ambient temperature data; S302. Determine whether the temperature change trend is in a heating or cooling period, and adjust the cooling water flow rate. During the heating period, use high flow rate turbulent flow to enhance heat dissipation, and during the cooling period, use low flow rate laminar flow to avoid thermal shock.

[0033] In this embodiment of the invention, a pre-defined correspondence is established between the mold entry temperature control target and the internal and external temperature difference control target, and the ambient temperature and its changing trend. The changing trend includes a heating period and a cooling period. During the heating period, high-flow turbulent flow is used to enhance heat dissipation, while during the cooling period, low-flow laminar flow is used to avoid thermal shock and to slow down the flow rate to prevent the concrete from cooling too quickly.

[0034] like Figure 4 As shown, in a preferred embodiment of the present invention, the step of determining the raw material temperature control scheme and concrete mix proportion based on ambient temperature data specifically includes: S401, Determine the temperature control scheme for cementitious materials: When the ambient temperature is higher than the high temperature threshold, use ice water mixing or liquid nitrogen cooling; when the ambient temperature is lower than the low temperature threshold, use warm water mixing, with a water temperature ≥15℃. S402, Determine the aggregate temperature control scheme: Monitor the aggregate temperature in real time using an infrared thermometer and adjust the spray cooling system or underfloor heating device. S403, based on historical engineering data, trains an LSTM neural network to establish a nonlinear mapping relationship between temperature data and mix proportion parameters; S404 is designed with real-time adjustment logic for mix proportions, including dynamic adjustment of cement dosage, intelligent optimization of water-cement ratio, and coordinated adjustment of admixture proportions.

[0035] In this embodiment of the invention, the raw material temperature control scheme is designed by first considering the cementitious materials. When the ambient temperature is higher than the high-temperature threshold, such as in summer, ice water is used for mixing or liquid nitrogen cooling. When the ambient temperature is lower than the low-temperature threshold, such as in winter, warm water is used for mixing, with a water temperature ≥15℃, to ensure that the temperature of cement, mineral powder, and other cementitious materials entering the tank is controlled within the target range (e.g., ≤30℃ in summer, ≥5℃ in winter). Then, the aggregate temperature is considered. The aggregate temperature is monitored in real time using an infrared thermometer. When the aggregate temperature is too high, the spray cooling system is activated; when the aggregate temperature is too low, the underfloor heating device is activated. In addition, to optimize the concrete mix proportion, an LSTM neural network is trained based on historical engineering data to establish a nonlinear mapping relationship between temperature data and mix proportion parameters (cement dosage, water-cement ratio, admixture ratio). The model input includes real-time temperature, ambient humidity, wind speed, etc., and the output is the optimal mix proportion adjustment suggestion.

[0036] In addition, a real-time mix design logic is implemented: when the internal temperature of the concrete exceeds a preset threshold, the system automatically reduces the cement dosage by 3%-8% and simultaneously increases the retarder dosage to slow down the hydration reaction. The water-cement ratio is dynamically adjusted by combining temperature data with concrete workability monitoring (such as real-time feedback from a slump meter). For example, in high-temperature environments, the water-cement ratio is reduced by 0.02-0.05 to reduce water consumption and prevent concrete segregation. Based on a temperature prediction model, the proportion of admixtures with low heat of hydration (such as silica fume and limestone powder) is increased, while the dosage of high heat of hydration cement is reduced. For example, during summer construction, the fly ash dosage is increased from 20% to 30%, and the cement dosage is reduced by 10%.

[0037] As a preferred embodiment of the present invention, the step of determining the concrete pouring progress and controlling the vibration time and position of the immersion vibrator specifically includes: S501 can acquire the planar position, pouring height and time nodes of concrete pouring in real time and form a three-dimensional pouring progress heat map. S502, based on concrete workability and environmental conditions, generates initial recommended values ​​for vibration parameters through knowledge base queries; S503 uses the discrete element method to analyze the concrete density distribution under different vibration parameters and optimizes the vibration strategy. S504, real-time monitoring of vibration force, vibration depth and vibration time; S505 automatically judges the vibration effect based on preset vibration quality evaluation standards and generates adjustment instructions.

[0038] In this embodiment of the invention, the planar position, pouring height, and time nodes of concrete pouring can be obtained in real time through RFID tags or GPS positioning modules installed on the pouring equipment, forming a three-dimensional pouring progress heatmap. Combined with a BIM model or digital twin system, the real-time progress is compared with the preset construction plan to identify progress deviations. Furthermore, based on concrete workability (such as slump and spread) and environmental conditions, initial recommended values ​​for vibration parameters (such as vibration frequency, amplitude, insertion depth, and duration) can be generated through a knowledge base query. Then, the discrete element method is used to analyze the concrete density distribution under different vibration parameters, optimizing the vibration strategy: high-frequency vibration is used for the surface layer, and low-frequency vibration is used for deeper layers. Additionally, by integrating acceleration and displacement sensors onto the vibrator, the vibration force, vibration depth, and vibration time are monitored in real time. Then, based on preset vibration quality evaluation standards, the vibration effect is automatically judged, and adjustment instructions are generated, such as extending the vibration time or adjusting the vibration position.

[0039] As a preferred embodiment of the present invention, the step of analyzing the concrete density distribution under different vibration parameters using the discrete element method and optimizing the vibration strategy specifically includes: S5031, Based on the three-dimensional pouring progress heat map, obtain the geometric information of the current pouring layer, and combine the vibration frequency, amplitude and insertion depth in the initial recommended values ​​of the vibration parameters with the geometric information of the current pouring layer to generate the vibration parameters to be analyzed; S5032, Perform vibration process simulation analysis on each set of generated vibration parameters to be analyzed, and output concrete density distribution data corresponding to each set of generated vibration parameters to be analyzed; wherein, the concrete density distribution data is used to characterize the degree of compaction of concrete at different locations. S5033, extract the density values ​​of key areas from the concrete density distribution data, and establish a correspondence between the density values ​​of key areas and the vibration parameters to be analyzed to form a vibration parameter-density effect mapping table; S5034, the density distribution trend in the vibration parameter-density effect mapping table is compared with the actual density effect inverted by the vibration data stream monitored in real time by the acceleration sensor and displacement sensor to obtain a second comparison result. Based on the second comparison result, the data in the mapping table is marked with confidence level to generate a vibration strategy library for data verification. S5035, in the vibration strategy library for data verification, based on the preset compaction threshold and energy consumption evaluation algorithm, the vibration parameters that meet the compaction requirements and have the lowest overall energy consumption are selected to generate a preliminary optimized vibration scheme; S5036, the preliminary optimized vibration scheme is analyzed in conjunction with the predicted temperature change trend. Based on the high-temperature accumulation area and the temperature difference sensitive area identified in the temperature change trend, the vibration parameters are adjusted regionally to generate the final vibration strategy, which is then used as the optimized vibration strategy. Specifically, in the high-temperature accumulation area, enhanced vibration parameters are used to improve the internal pore structure and promote heat dissipation; in the temperature difference sensitive area, moderate vibration parameters are used to avoid excessive disturbance.

[0040] In this embodiment of the invention, in the step of analyzing the density distribution of concrete under different vibration parameters using the discrete element method and optimizing the vibration strategy, this invention combines the vibration parameters with the geometric information shown in the three-dimensional pouring progress heatmap to simulate and analyze the vibration process, generate a vibration parameter-density effect mapping table, and establish a quantitative relationship of the internal density distribution of concrete under different parameters.

[0041] By comparing and marking the consistency between the simulated compaction trend in the mapping table and the actual compaction effect inverted from real-time monitoring data such as accelerometers, and generating a vibration strategy library validated by field data, the guiding value of simulation analysis results for actual construction is ensured. Furthermore, after selecting preliminary solutions through multi-objective optimization in the validated strategy library, the vibration strategy is creatively synergistically analyzed with the predicted temperature change trend. For high-temperature accumulation areas, enhanced vibration parameters are used to optimize the internal pore structure and promote heat dissipation; for areas sensitive to internal and external temperature differences, moderate vibration parameters are used to avoid excessive disturbance and exacerbate surface moisture evaporation. This achieves deep synergy between vibration technology and temperature control objectives, indirectly assisting in achieving temperature control objectives from the perspective of improving the microstructure of concrete.

[0042] In a preferred embodiment of the present invention, the step of automatically determining the vibration effect and generating adjustment instructions based on a preset vibration quality evaluation standard specifically includes: S5051 extracts the force-depth curve features characterizing vibration stability, the duration features characterizing the effective vibration time, and the spatial coverage features characterizing the vibration range from the real-time monitored vibration force, vibration depth, and vibration time data stream, forming a vibration quality evaluation feature vector for the current vibration operation. S5052, compare the vibration quality evaluation feature vector of the current vibration operation with the preset vibration quality evaluation standard. The preset vibration quality evaluation standard includes the vibration force stability threshold, the effective vibration depth range, and the minimum continuous vibration time, to obtain the quality compliance score of the current vibration operation. S5053, The quality compliance score of the current vibration operation is bound to the current vibration position information obtained from the three-dimensional pouring progress heat map to generate a vibration quality evaluation result with spatial position label. S5054, perform correlation analysis between the vibration quality evaluation results with spatial location tags and the expected compaction data of the corresponding location in the vibration strategy library of the data verification, identify the defect areas where the actual vibration effect deviates significantly from the expected target, and generate a list of vibration quality defects. S5055, compare the list of vibration quality defects with the optimized vibration strategy, and perform a secondary diagnosis based on the predicted temperature change trend to obtain the execution result of the adjustment command; wherein, the execution result of the adjustment command includes: for defect areas located in high-temperature accumulation zones and with insufficient vibration, generate an adjustment command with the core of strengthening vibration; for defect areas located in zones sensitive to internal and external temperature differences and with excessive vibration, generate an adjustment command with the core of weakening vibration and assisting surface insulation. S5056, based on the execution result of the adjustment instruction, update the vibration parameter-compaction effect mapping table, and store the current vibration quality evaluation, diagnosis process and effective adjustment instructions as new knowledge entries in the knowledge base to optimize the generation logic of the initial recommended values ​​of subsequent vibration parameters, and at the same time use the execution result of the adjustment instruction as the final adjustment instruction.

[0043] In this embodiment of the invention, in the step of automatically judging the vibration effect and generating adjustment instructions based on a preset vibration quality evaluation standard, the invention extracts the stability of the force-depth curve, effective action time, and spatial coverage features from the real-time vibration data stream to form a vibration quality evaluation feature vector. This vector is then compared with a preset standard to generate a quantified quality evaluation result with spatial location labels, achieving accurate and traceable online evaluation of the vibration effect. By correlating the real-time evaluation results with the expected compaction at the corresponding location in the data-verified vibration strategy library, defective areas where the actual vibration effect deviates significantly from the expected target are accurately identified.

[0044] Furthermore, by combining predicted temperature change trends, secondary diagnosis of defect areas is performed. For insufficient vibration defects located in high-temperature clusters, enhanced vibration commands are generated to simultaneously improve compaction and heat dissipation. For excessive vibration defects located in areas sensitive to internal and external temperature differences, commands to reduce vibration and supplement insulation are generated to avoid cracking risks. This achieves intelligent linkage between vibration quality control and temperature crack prevention. Finally, by feeding back effective adjustment experience to the knowledge base and updating the vibration parameter-compactness effect mapping table, a complete "evaluation-diagnosis-adjustment-learning" closed loop is formed, enabling the system's decision-making capabilities to continuously self-optimize as construction progresses.

[0045] This invention also provides an intelligent temperature control system for large-volume foundation concrete pouring, the system comprising: The data monitoring and acquisition module 100 is used to collect real-time internal temperature data of concrete and ambient temperature data; and to monitor the inlet and outlet water temperatures and flow rates of cooling water pipes. The temperature change recognition module 200 is used to perform statistical analysis on the collected temperature data based on machine learning algorithms to identify temperature change trends and anomalies. The optimal temperature control scheme module 300 is used to automatically generate the optimal temperature control scheme based on statistical analysis results. The optimal temperature control scheme includes cooling water flow adjustment, mold entry temperature control target, and internal and external temperature difference control target. The raw material control module 400 is used to determine the raw material temperature control scheme and concrete mix proportion based on ambient temperature data. The vibration control module 500 is used to determine the concrete pouring progress and control the vibration time and position of the immersion vibrator to ensure the vibration quality.

[0046] In a preferred embodiment of the present invention, the temperature change recognition module 200 includes: The data preprocessing unit is used to clean the collected data, remove outliers, and perform smoothing filtering. The key feature extraction unit is used to extract key features of temperature data, including heating rate, cooling rate, peak internal and external temperature difference, and temperature gradient distribution, forming a feature vector. The temperature change trend unit is used to build a training set based on historical engineering data, train a long short-term memory network or a random forest model, and predict the temperature change trend. The anomaly identification unit is used to identify temperature anomalies using the isolated forest algorithm or self-organizing map neural network, and to perform cause diagnosis by associating them with construction logs and environmental parameters to determine the anomaly type.

[0047] In a preferred embodiment of the present invention, the optimal temperature control scheme module 300 includes: The control target determination unit is used to determine the mold inlet temperature control target and the internal and external temperature difference control target based on the temperature change trend and ambient temperature data. The cooling water flow rate adjustment unit is used to determine whether the temperature change trend is in a heating or cooling period, and adjusts the cooling water flow rate accordingly. During the heating period, high flow rate turbulent flow is used to enhance heat dissipation, while during the cooling period, low flow rate laminar flow is used to avoid thermal shock.

[0048] In a preferred embodiment of the present invention, the raw material control module 400 includes: The cementitious material temperature control unit is used to determine the cementitious material temperature control scheme: when the ambient temperature is higher than the high temperature threshold, ice water mixing or liquid nitrogen cooling is used; when the ambient temperature is lower than the low temperature threshold, warm water mixing is used, with a water temperature ≥15℃. The aggregate temperature control unit is used to determine the aggregate temperature control scheme: it monitors the aggregate temperature in real time through an infrared thermometer and adjusts the spray cooling system or the underfloor heating device. The mix proportion determination unit is used to train an LSTM neural network based on historical engineering data and establish a nonlinear mapping relationship between temperature data and mix proportion parameters. The dynamic adjustment unit is used to design the real-time adjustment logic of the mix proportion, including dynamic adjustment of cement dosage, intelligent optimization of water-cement ratio, and coordinated adjustment of admixture ratio.

[0049] The above description only details the preferred embodiments of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0050] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0051] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0052] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the disclosure in the specification and embodiments. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

Claims

1. A method for intelligent temperature control during the pouring of large-volume foundation concrete, characterized in that, The method includes the following steps: Real-time acquisition of concrete internal temperature data and ambient temperature data; monitoring of inlet and outlet water temperatures and flow rates in cooling water pipes; The collected temperature data is statistically analyzed using machine learning algorithms to identify temperature change trends and anomalies. Based on the statistical analysis results, the optimal temperature control scheme is automatically generated. The optimal temperature control scheme includes cooling water flow adjustment, mold entry temperature control target, and internal and external temperature difference control target. Determine the raw material temperature control scheme and concrete mix proportion based on ambient temperature data; Determine the concrete pouring schedule and control the vibration time and position of the immersion vibrator to ensure vibration quality.

2. The intelligent temperature control method for large-volume foundation concrete pouring according to claim 1, characterized in that, The steps involved in statistically analyzing the collected temperature data using machine learning algorithms to identify temperature change trends and outliers include: The collected data is cleaned, outliers are removed, and smoothing filtering is performed. Extract key features from temperature data, including heating rate, cooling rate, peak internal and external temperature difference, and temperature gradient distribution, to form a feature vector. A training set is built based on historical engineering data, and a long short-term memory network or random forest model is trained to predict temperature change trends. The isolated forest algorithm or self-organizing map neural network is used to identify temperature anomalies, and the causes are diagnosed by correlating them with construction logs and environmental parameters to determine the type of anomaly.

3. The intelligent temperature control method for large-volume foundation concrete pouring according to claim 2, characterized in that, The steps of constructing a training set based on historical engineering data, training a long short-term memory network or a random forest model, and predicting temperature change trends specifically include: The feature vector is time-series aligned and correlated with the cooling water pipe flow data of the corresponding historical time period to generate an enhanced training sample set; wherein, the enhanced training sample set is used to characterize the correlation between concrete temperature changes and active cooling measures; The enhanced training sample set is divided using a sliding time window method to generate a divided time-series training sample set; wherein, the divided time-series training sample set contains multiple training instances arranged in chronological order. The Long Short-Term Memory Network model is trained using the partitioned time-series training sample set to generate a first temperature prediction model; wherein, the first temperature prediction model is used to capture and predict time-series dependent features in temperature data. The random forest model is trained using the partitioned time-series training sample set to generate a second temperature prediction model; wherein, the second temperature prediction model is used to capture and predict the nonlinear relationship between temperature features; Establish an online evaluation and switching logic for model performance, compare the prediction errors of the first temperature prediction model and the second temperature prediction model on recent validation data in real time, obtain the first comparison result, and select one of the first temperature prediction model and the second temperature prediction model as the current dominant prediction model based on the first comparison result. The model rolling update mechanism is activated, and the latest collected and processed real-time temperature feature data is used as incremental data to fine-tune the parameters of the current dominant prediction model, generating an optimized temperature prediction model. The real-time collected internal temperature data of concrete and ambient temperature data are input into the optimized temperature prediction model to obtain the output of the optimized temperature prediction model, and the output of the optimized temperature prediction model is used as the predicted temperature change trend.

4. The intelligent temperature control method for large-volume foundation concrete pouring according to claim 3, characterized in that, The steps of identifying temperature anomalies using the isolated forest algorithm or self-organizing map neural network, and correlating them with construction logs and environmental parameters for causal diagnosis to determine the anomaly type, specifically include: The real-time collected concrete internal temperature data is synchronized with the feature vector to construct a real-time temperature monitoring dataset; The real-time temperature monitoring dataset is simultaneously input into a pre-trained isolated forest algorithm model and a self-organizing map neural network model to perform anomaly detection in parallel, thereby obtaining global anomaly points output by the isolated forest algorithm model and local anomaly patterns output by the self-organizing map neural network model. The global outliers output by the isolated forest algorithm model and the local outlier patterns output by the self-organizing map neural network model are fused to generate a comprehensive list of outlier locations. The comprehensive list of abnormal locations is matched with construction logs and historical environmental parameter data for the same time period in multiple dimensions to obtain the matching results. Based on the results of the association matching, the anomaly cause diagnosis logic is executed to map the anomaly points to the construction events or environmental conditions that are most likely to cause the anomalies, and generate a preliminary anomaly cause diagnosis report. The preliminary abnormality cause diagnosis report is cross-validated with the predicted temperature change trend to generate the diagnosed abnormality type and cause diagnosis conclusion. The diagnosed anomaly type and cause diagnosis conclusions are fed back to the first temperature prediction model and the second temperature prediction model as incremental data as part of the model rolling update mechanism. This data is used to correct the prediction deviation of the current dominant prediction model under the corresponding abnormal conditions, and the diagnosed anomaly type is used as the final determined anomaly point type.

5. The intelligent temperature control method for large-volume foundation concrete pouring according to claim 4, characterized in that, The steps for automatically generating the optimal temperature control scheme specifically include: Determine the mold inlet temperature control target and the internal and external temperature difference control target based on the temperature change trend and ambient temperature data; Determine whether the temperature change trend is in a warming or cooling phase, and adjust the cooling water flow rate accordingly. During the warming phase, use high flow rate turbulent flow to enhance heat dissipation, and during the cooling phase, use low flow rate laminar flow to avoid thermal shock.

6. The intelligent temperature control method for large-volume foundation concrete pouring according to claim 5, characterized in that, The steps of determining the raw material temperature control scheme and concrete mix proportion based on ambient temperature data specifically include: Determine the temperature control scheme for the cementitious material: when the ambient temperature is higher than the high temperature threshold, use ice water mixing or liquid nitrogen cooling; when the ambient temperature is lower than the low temperature threshold, use warm water mixing. Determine the aggregate temperature control scheme: monitor the aggregate temperature in real time using an infrared thermometer and adjust the spray cooling system or underfloor heating device accordingly; An LSTM neural network was trained based on historical engineering data to establish a nonlinear mapping relationship between temperature data and mix proportion parameters. The design incorporates a real-time adjustment logic for the mix proportion, including dynamic adjustment of cement dosage, intelligent optimization of water-cement ratio, and coordinated adjustment of admixture proportion.

7. The intelligent temperature control method for large-volume foundation concrete pouring according to claim 6, characterized in that, The steps of determining the concrete pouring progress and controlling the vibration time and position of the immersion vibrator specifically include: By installing RFID tags or GPS positioning modules on the pouring equipment, the planar position, pouring height and time nodes of concrete pouring can be obtained in real time to form a three-dimensional pouring progress heat map. Based on concrete workability and environmental conditions, initial recommended values ​​for vibration parameters are generated by querying a knowledge base. The discrete element method was used to analyze the density distribution of concrete under different vibration parameters and to optimize the vibration strategy. By integrating an acceleration sensor and a displacement sensor onto the vibratory rod, the vibration force, vibration depth, and vibration time can be monitored in real time. Based on preset vibration quality evaluation standards, the vibration effect is automatically determined and adjustment instructions are generated.

8. The intelligent temperature control method for large-volume foundation concrete pouring according to claim 7, characterized in that, The steps of analyzing the concrete density distribution under different vibration parameters using the discrete element method and optimizing the vibration strategy specifically include: Based on the three-dimensional pouring progress heat map, the geometric information of the current pouring layer is obtained. The vibration frequency, amplitude and insertion depth in the initial recommended values ​​of the vibration parameters are combined with the geometric information of the current pouring layer to generate the vibration parameters to be analyzed. For each set of generated vibration parameters to be analyzed, a vibration process simulation analysis is performed, and concrete density distribution data corresponding to each set of generated vibration parameters to be analyzed is output; wherein, the concrete density distribution data is used to characterize the degree of compaction of concrete at different locations. Extract the density values ​​of key areas from the concrete density distribution data, and establish a correspondence between the density values ​​of key areas and the vibration parameters to be analyzed to form a vibration parameter-density effect mapping table; The density distribution trend in the vibration parameter-density effect mapping table is compared with the actual density effect inverted by the vibration data stream monitored in real time by the acceleration sensor and displacement sensor to obtain a second comparison result. Based on the second comparison result, the data in the mapping table is marked with confidence level to generate a vibration strategy library for data verification. In the vibration strategy library verified by the data, the vibration parameters that meet the compaction requirements and have the lowest overall energy consumption are selected based on the preset compaction threshold and energy consumption evaluation algorithm to generate a preliminary optimized vibration scheme. The preliminary optimized vibration scheme is analyzed in conjunction with the predicted temperature change trend. Based on the high-temperature accumulation area and the temperature difference sensitive area between inside and outside the temperature change trend, the vibration parameters are adjusted regionally to generate the final vibration strategy, which is then used as the optimized vibration strategy.

9. The intelligent temperature control method for large-volume foundation concrete pouring according to claim 8, characterized in that, The step of automatically determining the vibration effect and generating adjustment instructions based on a preset vibration quality evaluation standard specifically includes: From the real-time monitored data streams of vibration force, vibration depth, and vibration time, we extract the force-depth curve features that characterize vibration stability, the duration features that characterize the effective vibration time, and the spatial coverage features that characterize the vibration range, to form a vibration quality evaluation feature vector for the current vibration operation. The vibration quality evaluation feature vector of the current vibration operation is compared with the preset vibration quality evaluation standard, which includes the vibration force stability threshold, the effective vibration depth range, and the minimum continuous vibration time, to obtain the quality compliance score of the current vibration operation. The quality compliance score of the current vibration operation is bound to the current vibration location information obtained from the three-dimensional pouring progress heat map to generate a vibration quality evaluation result with spatial location label. The vibration quality evaluation results with spatial location tags are correlated with the expected compaction data of the corresponding location in the vibration strategy library verified by the data. Defect areas where the actual vibration effect deviates significantly from the expected target are identified, and a list of vibration quality defects is generated. The list of vibration quality defects is compared with the optimized vibration strategy, and a secondary diagnosis is performed in conjunction with the predicted temperature change trend to obtain the execution result of the adjustment command. The execution result of the adjustment command includes: for defective areas located in high-temperature accumulation zones and with insufficient vibration, an adjustment command with the core of strengthening vibration is generated; for defective areas located in zones sensitive to internal and external temperature differences and with excessive vibration, an adjustment command with the core of weakening vibration and assisting surface insulation is generated. Based on the execution result of the adjustment command, the vibration parameter-compaction effect mapping table is updated, and the current vibration quality evaluation, diagnosis process and effective adjustment command are stored as new knowledge entries in the knowledge base to optimize the generation logic of the initial recommended values ​​of subsequent vibration parameters. At the same time, the execution result of the adjustment command is used as the final adjustment command.

10. A smart temperature control system for large-volume foundation concrete pouring, characterized in that, The system applies the intelligent temperature control method for large-volume foundation concrete pouring as described in any one of claims 1 to 9, and the system includes: The data monitoring and acquisition module is used to collect real-time data on the internal temperature of concrete and ambient temperature; and to monitor the inlet and outlet water temperatures and flow rates of cooling water pipes. The temperature change recognition module is used to perform statistical analysis on the collected temperature data based on machine learning algorithms to identify temperature change trends and anomalies. The optimal temperature control scheme module is used to automatically generate the optimal temperature control scheme based on statistical analysis results. The optimal temperature control scheme includes cooling water flow adjustment, mold entry temperature control target, and internal and external temperature difference control target. The raw material control module is used to determine the raw material temperature control scheme and concrete mix proportion based on ambient temperature data. The vibration control module is used to determine the concrete pouring progress and control the vibration time and position of the immersion vibrator to ensure the vibration quality.

Images