Vibration and temperature control cooperative control method and system for concrete forming

Abstract

The invention discloses a vibration temperature control cooperative control method and system for concrete forming, and the method comprises the steps: building an initial prediction model of a vibration temperature control bidirectional coupling effect through concrete mix proportion parameters, historical vibration control parameters and initial temperature field data; the model is dynamically corrected through vibration time sequence data and temperature field space data which are collected in real time to obtain a real-time coupling effect prediction model, and then the real-time model is used for predicting the compactness deviation and the temperature gradient of the next control period; then vibration parameter adjustment amount and temperature control measure intervention intensity are calculated based on the predicted value through a multi-target collaborative optimization algorithm to generate a collaborative control instruction, and finally collaborative optimization of vibration and temperature control in the concrete pouring process is achieved according to the instruction. According to the invention, precise collaborative optimization of vibration and temperature control in the concrete pouring process can be realized, and the stability of concrete forming quality is effectively guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, and in particular to a vibration and temperature control collaborative control method and system for concrete forming. BACKGROUND

[0002] In the field of concrete forming, the existing technology usually adopts a relatively independent control method for vibration control and temperature control during the concrete pouring process. Specifically, vibration control usually sets fixed vibration parameters (including vibration frequency, amplitude, vibration duration, etc.) in advance according to the type of concrete (such as strength grade, slump, etc.) and the structural characteristics of the pouring site, and starts the vibration equipment according to the preset parameters during the pouring process to make the concrete compact through mechanical vibration; while temperature control usually formulates a preliminary temperature control scheme according to the ambient temperature and the hydration heat characteristics of the concrete before pouring, such as embedding cooling water pipes or arranging insulation layers in the concrete, and manually or semi-automatically adjusting the flow of the cooling water pipes or adjusting the coverage range of the insulation measures to control the temperature rise and temperature gradient in the concrete when the temperature exceeds the preset threshold value during the pouring process. In addition, some existing technologies attempt to simply associate vibration parameters with temperature data, for example, appropriately reducing the vibration frequency when the temperature is too high, but such association is mostly linear adjustment based on experience, lacking systematic modeling and dynamic response analysis of the complex coupling relationship between vibration and temperature. Therefore, the existing technology cannot achieve precise collaborative optimization of vibration and temperature control during the concrete pouring process, resulting in difficulty in effectively guaranteeing the stability of the concrete forming quality. SUMMARY

[0003] The present application provides a vibration and temperature control collaborative control method and system for concrete forming, which can achieve precise collaborative optimization of vibration and temperature control during the concrete pouring process, effectively guaranteeing the stability of the concrete forming quality.

[0004] An embodiment of the present application provides a vibration and temperature control collaborative control method for concrete forming, comprising: constructing an initial prediction model of bidirectional coupling effect of vibration and temperature control based on concrete mix proportion parameters, historical vibration control parameters and initial temperature field data; real-time collecting vibration time series data and temperature field spatial data during the concrete pouring process, and inputting the vibration time series data and temperature field spatial data into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model; using the real-time coupling effect prediction model to predict the compaction evolution trend and temperature gradient of the concrete pouring in the next control period, and generating a compaction deviation prediction value and a temperature gradient prediction value of the concrete pouring in the next control period; According to the compaction degree deviation prediction value and the temperature gradient prediction value, a vibration parameter adjustment amount and a temperature control measure intervention intensity are calculated by a multi-objective collaborative optimization algorithm, and a collaborative control instruction of the next control period is obtained based on the calculation result, the collaborative control instruction including a vibration control instruction and a temperature control adjustment instruction; Based on the collaborative control instruction, the concrete pouring system is controlled to perform collaborative optimization of vibration and temperature control during concrete pouring in the next control period.

[0005] As an improvement of the above scheme, the initial prediction model of the vibration-temperature control two-way coupling effect is constructed based on the concrete mix proportion parameters, historical vibration control parameters and initial temperature field data, including the following sub-steps: The concrete mix proportion parameters are processed by feature coding to obtain a material property feature vector; Based on the material property feature vector and the historical vibration control parameters, a vibration response prediction sub-model is constructed; Based on the material property feature vector and the initial temperature field data, a temperature evolution prediction sub-model is constructed; The vibration response prediction sub-model and the temperature evolution prediction sub-model are two-way coupled and associated to obtain the initial prediction model of the vibration-temperature control two-way coupling effect.

[0006] As an improvement of the above scheme, the vibration time series data and the temperature field space data in the concrete pouring process are collected in real time, and the vibration time series data and the temperature field space data are input into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model, including the following sub-steps: The vibration time series data in the concrete pouring process are collected in real time by a vibration sensor array to obtain a vibration signal time series; The temperature field space data in the concrete pouring process are collected in real time by a temperature sensor array to obtain a temperature distribution space matrix; The vibration signal time series and the temperature distribution space matrix are input into the initial prediction model to calculate a model prediction error; Based on the model prediction error, the coupling parameters of the initial prediction model are updated; The initial prediction model after parameter updating is verified for convergence to obtain a real-time coupling effect prediction model.

[0007] As an improvement of the above scheme, the real-time coupling effect prediction model is used to predict the compaction degree evolution trend and the temperature gradient of the next control period of concrete pouring, to generate the compaction degree deviation prediction value and the temperature gradient prediction value of the next control period of concrete pouring, including the following sub-steps: According to the pouring progress, a control period time window is divided to determine the prediction time interval of the next control period; The real-time coupling effect prediction model is calculated forwardly in the prediction time interval to obtain a density evolution trend curve and a temperature gradient distribution surface; The density evolution trend curve is subjected to target density deviation quantification processing to generate a density deviation prediction value of concrete pouring in the next control period; The temperature gradient distribution surface is subjected to maximum gradient value extraction processing to generate a temperature gradient prediction value of concrete pouring in the next control period.

[0008] As an improvement of the above scheme, the vibration parameter adjustment amount and the temperature control measure intervention intensity are calculated by a multi-objective collaborative optimization algorithm according to the density deviation prediction value and the temperature gradient prediction value, and the collaborative control instruction of the next control period is obtained based on the calculation result, the collaborative control instruction including a vibration control instruction and a temperature control adjustment instruction, including the following sub-steps: Based on the density deviation prediction value and the temperature gradient prediction value, a density optimization weight and a temperature control optimization weight are configured to obtain a multi-objective collaborative optimization weight matrix of density optimization and temperature control optimization; The vibration parameter adjustment range constraint and the temperature control measure intervention intensity constraint are set to obtain optimization constraint conditions of multi-objective collaborative optimization; The multi-objective collaborative optimization weight matrix and the optimization constraint conditions are taken as inputs, and the non-dominated sorting genetic algorithm is used to solve the optimal solution set of the vibration parameter adjustment amount and the temperature control measure intervention intensity to obtain the optimal vibration control instruction and the optimal temperature control adjustment instruction of the next control period.

[0009] Another embodiment of the present application correspondingly provides a vibration and temperature control collaborative control system for concrete forming, comprising: A construction module is configured to construct an initial prediction model of vibration and temperature control two-way coupling effect based on concrete mix proportion parameters, historical vibration control parameters and initial temperature field data; A correction module is configured to collect vibration time series data and temperature field space data in a concrete pouring process in real time, and input the vibration time series data and the temperature field space data into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model; A prediction module is configured to use the real-time coupling effect prediction model to predict the density evolution trend and the temperature gradient of concrete pouring in the next control period to generate a density deviation prediction value and a temperature gradient prediction value of concrete pouring in the next control period; a calculation module, configured to calculate a vibration parameter adjustment amount and an intervention intensity of a temperature control measure by a multi-objective collaborative optimization algorithm according to the compactness deviation prediction value and the temperature gradient prediction value, and obtain a collaborative control instruction of the next control period based on a calculation result, the collaborative control instruction comprising a vibration control instruction and a temperature control adjustment instruction; a collaborative control module, configured to control, based on the collaborative control instruction, the concrete pouring system to perform collaborative optimization of vibration and temperature control in the next control period of concrete pouring.

[0010] As an improvement of the above-mentioned scheme, the construction module is specifically configured to: perform feature coding processing on the concrete mix proportion parameters to obtain a material property feature vector; construct a vibration response prediction sub-model based on the material property feature vector and the historical vibration control parameters; construct a temperature evolution prediction sub-model based on the material property feature vector and the initial temperature field data; couple and correlate the vibration response prediction sub-model and the temperature evolution prediction sub-model in a bidirectional manner to obtain an initial prediction model of vibration-temperature control bidirectional coupling effect.

[0011] As an improvement of the above-mentioned scheme, the correction module is specifically configured to: acquire vibration signal time sequence data by real-time acquisition of vibration time sequence data in the concrete pouring process through a vibration sensor array; acquire temperature distribution spatial matrix data by real-time acquisition of temperature field spatial data in the concrete pouring process through a temperature sensor array; input the vibration signal time sequence data and the temperature distribution spatial matrix data into the initial prediction model to calculate a model prediction error; update coupling parameters of the initial prediction model based on the model prediction error; verify the convergence of the initial prediction model after parameter updating to obtain a real-time coupling effect prediction model.

[0012] As an improvement of the above-mentioned scheme, the prediction module is specifically configured to: determine a prediction time interval of the next control period according to a control period time window divided according to a pouring progress; perform forward iteration calculation of the real-time coupling effect prediction model in the prediction time interval to obtain a compactness evolution trend curve and a temperature gradient distribution surface; perform target compactness deviation quantification processing on the compactness evolution trend curve to generate a compactness deviation prediction value of the next control period of concrete pouring; The temperature gradient distribution surface is subjected to maximum gradient value extraction processing to generate a temperature gradient prediction value of the concrete pouring in the next control period.

[0013] As an improvement of the above scheme, the calculation module is specifically configured to: Based on the density deviation prediction value and the temperature gradient prediction value, a density optimization weight and a temperature control optimization weight are configured to obtain a multi-objective collaborative optimization weight matrix of density optimization and temperature control optimization. The vibration parameter adjustment range constraint and the temperature control measure intervention intensity constraint are set to obtain optimization constraint conditions of multi-objective collaborative optimization. The multi-objective collaborative optimization weight matrix and the optimization constraint conditions are input, and the non-dominated sorting genetic algorithm is used to solve an optimal solution set of the vibration parameter adjustment amount and the temperature control measure intervention intensity to obtain the optimal vibration control instruction and the optimal temperature control adjustment instruction in the next control period.

[0014] Compared with the prior art, the embodiments of the present application have the following beneficial effects: First, an initial prediction model of vibration and temperature control bidirectional coupling effect is constructed based on concrete mixing ratio parameters, historical vibration control parameters and initial temperature field data; then, real-time coupling effect prediction model is obtained by dynamically correcting the initial prediction model through real-time acquisition of vibration time sequence data and temperature field space data in the pouring process, so that the model can reflect the actual state of the current pouring; the real-time model is used to predict the density deviation and temperature gradient in the next control period; then, the vibration parameter adjustment amount and the temperature control measure intervention intensity are calculated through a multi-objective collaborative optimization algorithm based on the prediction values to generate a collaborative control instruction; finally, the pouring system is controlled to perform collaborative optimization of vibration and temperature control according to the instruction. The construction and dynamic correction of the vibration and temperature control bidirectional coupling effect model can accurately reflect the mutual influence of the two, and the multi-objective collaborative optimization algorithm ensures the coordination of vibration and temperature control adjustment, thereby realizing precise collaborative control of vibration and temperature control in the concrete forming process. Compared with the prior art which fails to construct a vibration and temperature control bidirectional coupling effect prediction model and realize collaborative optimization, the embodiments of the present application realize precise collaboration of vibration and temperature control by constructing and dynamically correcting the coupling model and generating a collaborative instruction through multi-objective optimization, thereby realizing precise collaborative optimization of vibration and temperature control in the concrete pouring process and effectively ensuring the stability of the concrete forming quality. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a flowchart of a vibration and temperature control collaborative control method for concrete forming provided by an embodiment of the present application; Figure 2 is a structural schematic diagram of a vibration and temperature control collaborative control system for concrete forming provided by an embodiment of the present application. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0017] Referring to Figure 1 is a flowchart of a concrete forming vibration and temperature control collaborative control method provided by an embodiment of the present application. The concrete forming vibration and temperature control collaborative control method comprises the following steps. S10, constructing an initial prediction model of vibration and temperature control two-way coupling effect based on concrete mixing ratio parameters, historical vibration control parameters and initial temperature field data; S11, collecting vibration time series data and temperature field space data in a concrete pouring process in real time, and inputting the vibration time series data and the temperature field space data into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model; S12, using the real-time coupling effect prediction model to predict the density evolution trend and temperature gradient of concrete pouring in a next control period, to generate a density deviation prediction value and a temperature gradient prediction value of concrete pouring in the next control period; S13, calculating vibration parameter adjustment amount and temperature control measure intervention intensity through a multi-objective collaborative optimization algorithm according to the density deviation prediction value and the temperature gradient prediction value, and obtaining a collaborative control instruction of the next control period based on the calculation result, wherein the collaborative control instruction comprises a vibration control instruction and a temperature control adjustment instruction; S14, controlling a concrete pouring system to perform collaborative optimization of vibration and temperature control of concrete pouring in the next control period based on the collaborative control instruction.

[0018] Compared with the prior art, the embodiments of the present application have the following beneficial effects: Firstly, an initial prediction model of vibration and temperature control two-way coupling effect is constructed based on concrete mix proportion parameters, historical vibration control parameters and initial temperature field data; then, through real-time collection of vibration time series data and temperature field space data in the pouring process, the initial prediction model is dynamically corrected to obtain a real-time coupling effect prediction model, so that the model can reflect the actual state of the current pouring; the real-time model is used to predict the density deviation and temperature gradient of the next control period; then, based on the predicted values, a multi-objective collaborative optimization algorithm is used to calculate the vibration parameter adjustment amount and the intervention intensity of temperature control measures to generate collaborative control instructions; finally, the pouring system is controlled according to the instructions to realize collaborative optimization of vibration and temperature control. The construction and dynamic correction of the vibration and temperature control two-way coupling effect model can accurately reflect the mutual influence of the two, and the multi-objective collaborative optimization algorithm ensures the coordination of vibration and temperature control adjustment, so as to realize precise collaborative control of vibration and temperature control in the concrete forming process. Compared with the prior art which fails to construct a vibration and temperature control two-way coupling effect prediction model and realize collaborative optimization, the embodiment of the present application realizes precise collaboration of vibration and temperature control by constructing and dynamically correcting the coupling model and generating collaborative instructions by combining multi-objective optimization, so as to realize precise collaborative optimization of vibration and temperature control in the concrete pouring process and effectively ensure the stability of the concrete forming quality.

[0019] As one of the examples, the initial prediction model of vibration and temperature control two-way coupling effect is constructed based on concrete mix proportion parameters, historical vibration control parameters and initial temperature field data, including the following sub-steps: The concrete mix proportion parameters are processed by feature coding to obtain a material property feature vector; A vibration response prediction sub-model is constructed based on the material property feature vector and the historical vibration control parameters; A temperature evolution prediction sub-model is constructed based on the material property feature vector and the initial temperature field data; The vibration response prediction sub-model and the temperature evolution prediction sub-model are two-way coupled and associated to obtain the initial prediction model of vibration and temperature control two-way coupling effect.

[0020] In the embodiment, firstly, the concrete mix proportion parameters are processed by feature coding to convert them into feature vectors capable of representing the material properties of the concrete; then, based on the material property feature vectors and historical vibration control parameters, a vibration response prediction sub-model capable of predicting the response of the concrete under the action of vibration is constructed; at the same time, the material property feature vectors and initial temperature field data are combined to construct a temperature evolution prediction sub-model capable of reflecting the temperature change process of the concrete; finally, the vibration response prediction sub-model and the temperature evolution prediction sub-model are bidirectionally coupled and associated, so that the influence of vibration on temperature and the reaction of temperature on vibration response can be reflected between the two sub-models, thereby forming an initial prediction model of the bidirectional coupling effect of vibration and temperature control. Therefore, the embodiment extracts key information of the material by feature coding, constructs sub-models and then couples them, which not only considers the change laws of vibration and temperature respectively, but also reflects the interaction between the two, which is beneficial to overall collaborative control. In summary, the embodiment accurately extracts the material properties affecting vibration response and temperature evolution through feature coding of the concrete mix proportion parameters; the vibration response prediction sub-model and the temperature evolution prediction sub-model constructed respectively can reflect the change laws of vibration and temperature; and the bidirectional coupling and association of the two make the initial prediction model accurately reflect the mutual influence between vibration and temperature control, providing a reliable initial model framework for subsequent real-time correction of the model and realization of collaborative optimization of vibration and temperature control, and improving the accuracy and applicability of the initial prediction model.

[0021] Specifically, the working process of the embodiment is exemplarily as follows: When the concrete mix proportion parameters are processed by feature coding, 10 core parameters such as cement strength grade, water-binder ratio, sand ratio, fly ash content, slag content, water reducing agent content, coarse aggregate particle size distribution and fine aggregate fineness modulus are extracted, min-max standardization processing is adopted for continuous parameters (such as water-binder ratio), one-hot encoding conversion is adopted for discrete parameters (such as cement strength grade), and the particle size distribution uniformity index of the distribution type parameter (such as coarse aggregate particle size distribution) is calculated as a feature value. Seven parameters strongly related to vibration response and temperature evolution are selected by Pearson correlation coefficient, and a material property feature vector with a dimension of 7x1 is constructed , wherein the continuous parameter standardization formula is: , in the formula, is the original parameter value, are the historical extreme values of the parameter, respectively.

[0022] When the vibration response prediction sub-model is constructed based on the material property feature vector and historical vibration control parameters, historical vibration frequency , vibration amplitude , vibration duration , vibration acceleration , and vibration point spacing are selected.The input matrix consists of 5 parameters. ( (For the number of historical samples), an improved gated recurrent unit (GRU) model is adopted, and an adaptive weighting module based on material properties is added to the input layer. This is achieved through calculation... and The mutual information values ​​of each parameter determine the weighting coefficients. The influence weights of vibration parameters under different material properties are dynamically adjusted. The model output is the concrete density growth rate. The expression is: In the formula Here is the weight matrix of the GRU network. For bias terms, This is a weighted diagonal matrix.

[0023] When constructing a temperature evolution prediction sub-model based on material property feature vectors and initial temperature field data, the initial temperature field data is discretized into a 10×10 grid temperature matrix. , combined thermal conductivity Specific heat capacity To obtain the same thermal parameters, a spatiotemporal convolutional network (STCN) architecture is adopted. First, the spatial distribution features of the temperature field are extracted through 2D convolutional layers, and then the temporal evolution is captured through 1D convolutional layers. At the same time, a material property correction factor is introduced. ( For the Sigmoid function, To correct the weight matrix, the temperature conduction rate is dynamically adjusted. The model output is... Temperature field matrix at time ,satisfy: In the formula For STCN network parameters, As a bias term, this model can effectively reflect the differences in temperature field evolution of concrete with different mix proportions.

[0024] When bidirectionally coupling the vibration response prediction sub-model and the temperature evolution prediction sub-model, design the coupling coefficient matrix. , The coefficient representing the influence of temperature change on vibration response (obtained through linear regression of temperature gradient and density change rate in historical data). The coefficient representing the influence of vibration on temperature evolution (determined based on theoretical calculations of vibration energy consumption converted into heat energy). A two-way feedback mechanism is established: the output of the vibration response sub-model. pass The initial temperature matrix of the modified temperature evolution sub-model ( Output of the temperature evolution sub-model pass adjusting the weight coefficient of the vibration response sub-model , for the temperature gradient). The final initial prediction model expression is: , realizing dynamic coupling simulation of vibration and temperature field.

[0025] As one of the examples, the vibration time series data and the temperature field spatial data in the concrete pouring process are collected in real time, and the vibration time series data and the temperature field spatial data are input into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model, including the following sub-steps: The vibration time series data in the concrete pouring process are collected in real time by the vibration sensor array to obtain a vibration signal time series; The temperature field spatial data in the concrete pouring process are collected in real time by the temperature sensor array to obtain a temperature distribution spatial matrix; The vibration signal time series and the temperature distribution spatial matrix are input into the initial prediction model to calculate a model prediction error; Based on the model prediction error, the coupling parameters of the initial prediction model are updated; The initial prediction model after parameter updating is verified for convergence to obtain a real-time coupling effect prediction model.

[0026] In this embodiment, first, the vibration time series data and the temperature field spatial data in the concrete pouring process are collected in real time by the vibration sensor array and the temperature sensor array respectively to form a vibration signal time series and a temperature distribution spatial matrix that can reflect the real-time state; then the two types of real-time data are input into the initial prediction model, and a model prediction error is calculated by comparing the model output and the actual monitoring data to quantify the deviation of the model from the actual pouring state; then the coupling parameters in the initial prediction model are updated based on the prediction error to make the model adapt to the dynamic changes of the current pouring process; finally, the model after parameter updating is verified for convergence to ensure the stability and reliability of the model, so as to obtain a real-time coupling effect prediction model that can accurately reflect the real-time pouring state. Therefore, through the application of the vibration sensor array and the temperature sensor array, the vibration and temperature real-time data in the pouring process can be comprehensively and accurately collected; by calculating the model prediction error and updating the coupling parameters, the initial prediction model can dynamically adapt to the real-time changes of the concrete pouring process; and the convergence verification ensures the stability and reliability of the model after parameter updating, and the real-time coupling effect prediction model obtained finally can accurately reflect the coupling relationship between vibration and temperature under the current pouring state, providing a model basis for prediction and collaborative control in the next control cycle, and improving the dynamic adaptation ability of the model to the actual pouring process.

[0027] In this embodiment, the working process is exemplarily as follows: When collecting vibration time series data in the concrete pouring process in real time through the vibration sensor array, 8 groups of piezoelectric vibration sensors are arranged outside the pouring formwork and near the internal vibrating rod. The sensor sampling frequency is set to 2 kHz, the collection cycle is 10 s / time, and the collection parameters include vibration acceleration, vibration frequency and vibration displacement. During the collection process, the anti-interference filter circuit is used to remove the mechanical noise of the pouring equipment (filter frequency range 0.5-500 Hz). The 100-cycle data collected by each group of sensors are spliced in time sequence to form a vibration signal time series sequence with the dimension of 8x2000 ( The sampling time is 1-2000), wherein represents the vibration signal fusion value of the group sensor at the time.

[0028] When collecting temperature field space data in the concrete pouring process in real time through the temperature sensor array, the distributed optical fiber temperature sensor is used. The sensor array is arranged in a 5cmx5cm grid in the pouring area, covering the top, middle and bottom of the pouring section (a total of 3 layers, 36 monitoring points in each layer). The sensor temperature is collected every 5 minutes. The temperature data of the 3 layers of monitoring points at the same time are mapped to a 3x6x6 three-dimensional matrix according to the spatial position, and the edge area temperature value is completed through the spatial interpolation algorithm (using inverse distance weighted interpolation). Finally, it is converted into a 10x10 temperature distribution space matrix , wherein each element in the matrix represents the actual temperature value of the row and column grid point.

[0029] When the vibration signal time series sequence and the temperature distribution space matrix are input into the initial prediction model to calculate the model prediction error, the vibration signal time series sequence is first extracted (extracting time domain peak value, root mean square value and frequency domain main frequency, etc.), and is converted into a vibration feature vector matching the input dimension of the initial prediction model . The and the temperature distribution space matrix are input into the initial prediction model, and the predicted temperature field matrix and the predicted vibration response feature are obtained. The improved weighted error formula is used to calculate the model prediction error , and the formula is: , wherein is the temperature error weight (value 0.6, set according to the priority of temperature control and vibration control), , are the first and second characteristic values of the actual and predicted temperature field matrices, respectively are the first and second characteristic values of the actual and predicted temperature field matrices, respectively are the first and second characteristic values of the actual and predicted temperature field matrices, respectively , are the first and second characteristic values of the actual and predicted temperature field matrices, respectively are the first and second characteristic values of the actual and predicted temperature field matrices, respectively

[0030] When updating the coupling parameters of the initial prediction model based on the model prediction error, the gradient descent method is used to update the coupling coefficient matrix in the initial prediction model to minimize the prediction error is the objective function, and the error is calculated by taking the partial derivative of the coupling coefficient , with respect to the coupling coefficient , , where is the learning rate, is the current number of update iterations, and the weight parameters , of the GRU and STCN networks in the initial prediction model are fine-tuned, with a fine-tuning amplitude of 1 / 5 of the initial training step size, to ensure the stability of the core architecture of the model.

[0031] When verifying the convergence of the initial prediction model after updating the parameters, the prediction errors , , of three consecutive update cycles are continuously collected, and the convergence threshold is set (according to the engineering precision requirement), and if and are satisfied, it is determined that the model parameter update converges; if not, the number of iterations is increased and the parameter update step is re-executed until the convergence condition is met. After convergence, the model after parameter update is determined as the real-time coupling effect prediction model, which can real-time match the vibration and temperature field coupling state of the current pouring process.

[0032] As one of the examples, the real-time coupling effect prediction model is used to predict the density evolution trend and temperature gradient of the next control cycle of concrete pouring, generate the density deviation prediction value and temperature gradient prediction value of the next control cycle of concrete pouring, including the following sub-steps: According to the pouring progress, the control cycle time window is divided to determine the prediction time interval of the next control cycle; The real-time coupling effect prediction model is iteratively calculated in the prediction time interval to obtain the density evolution trend curve and temperature gradient distribution surface; The compaction degree evolution trend curve is subjected to target compaction degree deviation quantitative processing to generate a compaction degree deviation prediction value of the concrete pouring in the next control period. The temperature gradient distribution surface is subjected to maximum gradient value extraction processing to generate a temperature gradient prediction value of the concrete pouring in the next control period.

[0033] In the embodiment, first, the control period time window is divided according to the concrete pouring progress, the prediction time interval of the next control period is determined, and a clear time boundary is set for the prediction process; then, the real-time coupling effect prediction model is subjected to forward iteration calculation within the prediction time interval, the compaction degree evolution trend curve and the temperature gradient distribution surface reflecting the internal state change of the concrete are obtained through dynamic deduction of the model; then, for the compaction degree evolution trend curve, the deviation degree from the preset target compaction degree is calculated by using the target compaction degree deviation quantitative processing mode, and a compaction degree deviation prediction value of the next control period is generated; at the same time, for the temperature gradient distribution surface, the key temperature gradient index is obtained through maximum gradient value extraction processing, and a temperature gradient prediction value of the next control period is generated. This process is constrained by the time interval, relies on the iteration calculation of the real-time coupling effect prediction model, and obtains specific prediction values through targeted data processing, thereby providing clear quantitative basis for subsequent collaborative optimization and regulation.

[0034] In this embodiment, specifically, when determining the predicted time interval for the next control cycle by dividing the control cycle time window according to the pouring progress, 1 / 3 of the initial setting time of the concrete is first used as the base cycle length. The pouring volume per unit time is calculated in combination with the real-time pouring rate (obtained through the pouring pump flow sensor). Then, the cycle length is dynamically adjusted according to the difference between the designed volume of the area to be poured and the volume already poured: if the remaining volume is less than the base cycle pouring volume, the cycle is shortened by the ratio of the remaining volume to the pouring rate; if the remaining volume is greater than the base cycle pouring volume, the base length is maintained, and the predicted time interval is set from the current time to the end of the base length. At the same time, a certain buffer time is reserved at the end of the interval to cope with the rate fluctuations during the pouring process, ensuring that the predicted interval is accurately matched with the actual pouring progress. When performing forward iterative calculations on the real-time coupling effect prediction model within the prediction time interval to obtain the density evolution trend curve and temperature gradient distribution surface, an improved variable step-size iterative method is adopted: a small step size (e.g., 10s / step) is used in the early stage of prediction to capture initial change details; a medium step size (e.g., 30s / step) is used in the middle stage to balance accuracy and efficiency; and a large step size (e.g., 60s / step) is used in the later stage to focus on the final trend. Each iteration uses the vibration parameters and temperature field data of the current step as model input, and outputs the density value and temperature field matrix of that step. The density values ​​of all steps are sorted by time to form the evolution curve; the gradient value of each step's temperature field matrix is ​​calculated using a three-dimensional gradient algorithm (combining spatial coordinates and time dimension), and superimposed to form the temperature gradient distribution surface. This variable step-size design improves computational efficiency while maintaining prediction accuracy. When quantifying the target density deviation of the density evolution trend curve and generating the density deviation prediction value, an innovative dynamic weight deviation algorithm is adopted: first, the target density threshold range is determined. (As specified in the design specifications), extract the predicted density value of the curve at the end of the prediction interval. Calculate the deviation value : In the formula To reduce the penalty weight for insufficient density, To mitigate the impact of excessive compaction, differentiated weights are used to prioritize undercompacted areas, making deviation quantification more aligned with actual engineering needs. The maximum gradient value is extracted from the temperature gradient distribution surface. When generating predicted temperature gradient values, a spatiotemporal weighted extraction method is employed: first, the gradient values ​​at all spatiotemporal points on the surface are traversed. ( For spatial coordinates, (For time), calculate the weight coefficients for each point. ,in Spatial weights (1.0 for core regions, 0.6 for edge regions). Time weights are assigned (1.0 for later stages and 0.7 for earlier stages), and then calculated using the formula. Determine the final predicted value.

[0035] As one example, the step of calculating the vibration parameter adjustment amount and the temperature control intervention intensity based on the predicted density deviation value and the predicted temperature gradient value using a multi-objective collaborative optimization algorithm, and obtaining the collaborative control command for the next control cycle based on the calculation results, wherein the collaborative control command includes a vibration control command and a temperature control adjustment command, includes the following sub-steps: Based on the predicted values ​​of density deviation and temperature gradient, density optimization weights and temperature control optimization weights are configured to obtain a multi-objective collaborative optimization weight matrix for density optimization and temperature control optimization. By setting constraints on the adjustment range of vibration parameters and the intervention intensity of temperature control measures, the optimization constraints for multi-objective collaborative optimization are obtained. Using the multi-objective collaborative optimization weight matrix and the optimization constraints as input, the optimal solution set for the vibration parameter adjustment amount and the temperature control intervention intensity is obtained by solving the non-dominated sorting genetic algorithm, thus obtaining the optimal vibration control command and the optimal temperature control adjustment command for the next control cycle.

[0036] In this embodiment, firstly, based on the predicted values ​​of density deviation and temperature gradient, a multi-objective collaborative optimization weight matrix is ​​formed by configuring density optimization weights and temperature control optimization weights, clarifying the priority of the two types of optimization objectives in collaborative control. Next, constraints on the vibration parameter adjustment range and the intervention intensity of temperature control measures are set to construct boundary conditions for multi-objective collaborative optimization, ensuring the engineering feasibility of the optimization results. Finally, the multi-objective collaborative optimization weight matrix and optimization constraints are input into a non-dominated sorting genetic algorithm. The algorithm solves for the optimal solution set of vibration parameter adjustment and temperature control intervention intensity, which is then transformed into the optimal vibration control command and temperature control adjustment command for the next control cycle. This embodiment achieves scientific quantification and collaborative optimization of vibration and temperature control adjustment quantities by balancing optimization objectives through weight configuration, ensuring engineering applicability through constraint setting, and solving for the optimal solution through a specific algorithm.

[0037] Specifically, when configuring density optimization weights and temperature control optimization weights based on the predicted density deviation and temperature gradient, a dynamic adaptive weight adjustment mechanism is adopted: Let the predicted density deviation be... The predicted temperature gradient value is First, calculate the normalized deviation coefficients of the two. ( (to allow maximum density deviation) and ( (for temperature gradient limits), then using the formula , Determine the weights, where To optimize the weights for density, Weights are optimized for temperature control. Finally, a multi-objective collaborative optimization weight matrix is ​​constructed. This dynamic weighting mechanism can automatically adjust the optimization priority based on the actual degree of deviation. When the standard is exceeded Automatically increase to ensure that key indicators are optimized first.

[0038] When setting constraints on the vibration parameter adjustment range and the intervention intensity of temperature control measures, the vibration parameter adjustment range constraint includes: vibration frequency adjustment amount. Amplitude adjustment amount Vibration duration adjustment amount Temperature control intervention intensity constraints include: cooling water pipe flow rate adjustment coefficient. Insulation layer thickness adjustment amount Simultaneously, coupling constraints are set: the sum of the vibration power increment and the temperature control energy consumption increment does not exceed a predetermined proportion of the rated total energy consumption, forming a complete set of optimization constraints. .

[0039] Using the multi-objective collaborative optimization weight matrix and optimization constraints as input, the improved non-dominated sorting genetic algorithm is used to solve the problem. The following improvements are made to the traditional NSGA-II: First, a real-number encoding method is adopted to encode the vibration parameter adjustment amount. and the intensity of temperature control intervention measures The first is to encode a chromosome of length 5; the second is to design a weighted composite fitness function. ,in Optimize the objective function for density ( , (Adjusted prediction bias) Optimize the objective function for temperature control ( , (For adjusted gradient prediction). The algorithm process includes: initializing the population (e.g., size 100), calculating fitness and performing non-dominated sorting, selecting parents using improved crowding distance (introducing weight matrix correction), generating offspring through adaptive crossover (crossover probability dynamically adjusted with population diversity) and polynomial mutation, and performing constraint checks (eliminating violations). After the individual populations are eliminated, a new population is formed. After 50 iterations, the solution closest to the ideal point in the Pareto optimal front is taken as the optimal solution, and the vibration control command for the next control cycle (including...) is output. ) and temperature control adjustment commands (including ).

[0040] See Figure 2 This is a schematic diagram of a vibration and temperature control system for concrete molding according to an embodiment of the present invention. The vibration and temperature control system for concrete molding includes: The construction module 10 is configured to construct an initial prediction model of the vibration-temperature control bidirectional coupling effect based on concrete mix proportion parameters, historical vibration control parameters and initial temperature field data; The correction module 11 is configured to collect vibration time series data and temperature field space data in a real-time manner during the concrete pouring process, and input the vibration time series data and the temperature field space data into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model; The prediction module 12 is configured to predict the density evolution trend and the temperature gradient of the concrete pouring in a next control period by using the real-time coupling effect prediction model, and generate a density deviation prediction value and a temperature gradient prediction value of the concrete pouring in the next control period. The calculation module 13 is configured to calculate a vibration parameter adjustment amount and a temperature control measure intervention intensity by using a multi-objective collaborative optimization algorithm according to the density deviation prediction value and the temperature gradient prediction value, and obtain a collaborative control instruction of the next control period based on a calculation result, wherein the collaborative control instruction includes a vibration control instruction and a temperature control adjustment instruction. The collaborative control module 14 is configured to control the concrete pouring system to perform collaborative optimization of vibration and temperature control during the concrete pouring in the next control period based on the collaborative control instruction.

[0041] Compared with the prior art, the embodiment of the present application has the following beneficial effects: First, an initial prediction model of the vibration-temperature control bidirectional coupling effect is constructed based on concrete mix proportion parameters, historical vibration control parameters and initial temperature field data. Then, the initial prediction model is dynamically corrected to obtain a real-time coupling effect prediction model by collecting vibration time series data and temperature field space data in a real-time manner during the pouring process, so that the model can reflect the actual state of the current pouring. The real-time model is used to predict the density deviation and the temperature gradient in the next control period. Then, a vibration parameter adjustment amount and a temperature control measure intervention intensity are calculated by using a multi-objective collaborative optimization algorithm based on the prediction values, and a collaborative control instruction is generated. Finally, the pouring system is controlled to perform collaborative optimization of vibration and temperature control according to the instruction. The construction and dynamic correction of the vibration-temperature control bidirectional coupling effect model can accurately reflect the mutual influence of the two, and the multi-objective collaborative optimization algorithm ensures the coordination of vibration and temperature control adjustment, thereby realizing precise collaborative control of vibration and temperature control in the concrete forming process. Compared with the prior art which fails to construct a vibration-temperature control bidirectional coupling effect prediction model and realize collaborative optimization, the embodiment of the present application realizes precise collaboration of vibration and temperature control by constructing and dynamically correcting the coupling model and generating a collaborative instruction by using multi-objective optimization, thereby realizing precise collaborative optimization of vibration and temperature control in the concrete pouring process and effectively ensuring the stability of the concrete forming quality.

[0042] As one of the examples, the construction module is specifically configured to: characteristic coding processing is performed on the concrete mix proportion parameters to obtain a material characteristic feature vector; Based on the material characteristic feature vector and the historical vibration control parameters, a vibration response prediction sub-model is constructed; Based on the material characteristic feature vector and the initial temperature field data, a temperature evolution prediction sub-model is constructed; The vibration response prediction sub-model and the temperature evolution prediction sub-model are bidirectionally coupled and associated to obtain an initial prediction model of the vibration-temperature control bidirectional coupling effect.

[0043] As an improvement of the above scheme, the correction module is specifically configured to: The vibration time series data in the concrete pouring process is collected in real time by the vibration sensor array to obtain a vibration signal time series sequence; The temperature field spatial data in the concrete pouring process is collected in real time by the temperature sensor array to obtain a temperature distribution spatial matrix; The vibration signal time series sequence and the temperature distribution spatial matrix are input into the initial prediction model to calculate the model prediction error; Based on the model prediction error, the coupling parameters of the initial prediction model are updated; The initial prediction model after parameter updating is verified for convergence to obtain a real-time coupling effect prediction model.

[0044] As one of the examples, the prediction module is specifically configured to: According to the pouring progress, a control period time window is divided to determine a prediction time interval of the next control period; The real-time coupling effect prediction model is calculated forward in the prediction time interval to obtain a density evolution trend curve and a temperature gradient distribution surface; The density evolution trend curve is subjected to target density deviation quantization processing to generate a density deviation prediction value of the concrete pouring in the next control period; The temperature gradient distribution surface is subjected to maximum gradient value extraction processing to generate a temperature gradient prediction value of the concrete pouring in the next control period.

[0045] As one of the examples, the calculation module is specifically configured to: Based on the density deviation prediction value and the temperature gradient prediction value, a density optimization weight and a temperature control optimization weight are configured to obtain a multi-objective collaborative optimization weight matrix of density optimization and temperature control optimization; The vibration parameter adjustment range constraint and the temperature control measure intervention intensity constraint are set to obtain optimization constraint conditions for multi-objective collaborative optimization; The non-dominated sorting genetic algorithm is used to solve an optimal solution set of the vibration parameter adjustment amount and the intervention intensity of the temperature control measure by taking the multi-target collaborative optimization weight matrix and the optimization constraint condition as inputs, so as to obtain optimal vibration control instructions and optimal temperature control adjustment instructions in the next control cycle.

[0046] It should be noted that the system embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. In addition, the connection relationship between the modules in the system embodiment provided by the present application indicates that there is a communication connection between them, which can be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0047] The above describes the preferred embodiments of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which are also considered within the scope of protection of the present application.

Claims

1. A method for coordinated vibration and temperature control in concrete molding, characterized in that, include: Based on concrete mix proportion parameters, historical vibration control parameters, and initial temperature field data, an initial prediction model for the two-way coupling effect of vibration and temperature control is constructed. Vibration time-series data and temperature field spatial data during the concrete pouring process are collected in real time, and the vibration time-series data and temperature field spatial data are input into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model. Using the real-time coupling effect prediction model, the density evolution trend and temperature gradient of concrete pouring in the next control cycle are predicted, and the density deviation and temperature gradient prediction values ​​of concrete pouring in the next control cycle are generated. Based on the predicted density deviation and the predicted temperature gradient, the vibration parameter adjustment amount and the intervention intensity of temperature control measures are calculated through a multi-objective collaborative optimization algorithm. Based on the calculation results, the collaborative control command for the next control cycle is obtained. The collaborative control command includes a vibration control command and a temperature control adjustment command. Based on the aforementioned coordinated control command, the concrete pouring system is controlled to perform coordinated optimization of vibration and temperature control during concrete pouring in the next control cycle.

2. The vibration and temperature control method for concrete forming as described in claim 1, characterized in that, The initial prediction model for the two-way coupling effect of vibration and temperature control, based on concrete mix proportion parameters, historical vibration control parameters, and initial temperature field data, includes the following sub-steps: The concrete mix proportion parameters are processed by feature encoding to obtain material property feature vectors. Based on the material property feature vector and the historical vibration control parameters, a vibration response prediction sub-model is constructed. Based on the material property feature vector and the initial temperature field data, a temperature evolution prediction sub-model is constructed; The vibration response prediction sub-model and the temperature evolution prediction sub-model are bidirectionally coupled to obtain the initial prediction model of the vibration-temperature control bidirectional coupling effect.

3. The vibration and temperature control method for concrete forming as described in claim 2, characterized in that, The process of acquiring real-time vibration time-series data and temperature field spatial data during concrete pouring, and then inputting this data into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model, includes the following sub-steps: Vibration time sequence data during concrete pouring is collected in real time by a vibration sensor array to obtain the vibration signal time sequence. The temperature field spatial data during the concrete pouring process is collected in real time by a temperature sensor array to obtain the temperature distribution spatial matrix. The vibration signal time sequence and the temperature distribution spatial matrix are input into the initial prediction model to calculate the model prediction error. Based on the model prediction error, update the coupling parameters of the initial prediction model; The convergence of the initial prediction model after parameter updates is verified to obtain the real-time coupling effect prediction model.

4. The vibration and temperature control method for concrete forming as described in claim 3, characterized in that, The method of using the real-time coupling effect prediction model to predict the density evolution trend and temperature gradient of concrete pouring in the next control cycle, and generating predicted values ​​for density deviation and temperature gradient of concrete pouring in the next control cycle, includes the following sub-steps: The control cycle time window is divided according to the pouring progress, and the prediction time interval of the next control cycle is determined. The real-time coupling effect prediction model is performed in forward iterative calculation within the prediction time interval to obtain the density evolution trend curve and the temperature gradient distribution surface. The density evolution trend curve is subjected to target density deviation quantification processing to generate the predicted density deviation value of concrete pouring in the next control cycle. The maximum gradient value is extracted from the temperature gradient distribution surface to generate the predicted temperature gradient value for concrete pouring in the next control cycle.

5. The vibration and temperature control method for concrete forming as described in claim 4, characterized in that, The step involves calculating the vibration parameter adjustment amount and the temperature control intervention intensity based on the predicted density deviation and the predicted temperature gradient using a multi-objective collaborative optimization algorithm, and obtaining the collaborative control command for the next control cycle based on the calculation results. The collaborative control command includes a vibration control command and a temperature control adjustment command, and includes the following sub-steps: Based on the predicted values ​​of density deviation and temperature gradient, density optimization weights and temperature control optimization weights are configured to obtain a multi-objective collaborative optimization weight matrix for density optimization and temperature control optimization. By setting constraints on the adjustment range of vibration parameters and the intervention intensity of temperature control measures, the optimization constraints for multi-objective collaborative optimization are obtained. Using the multi-objective collaborative optimization weight matrix and the optimization constraints as input, the optimal solution set for the vibration parameter adjustment amount and the temperature control intervention intensity is obtained by solving the non-dominated sorting genetic algorithm, thus obtaining the optimal vibration control command and the optimal temperature control adjustment command for the next control cycle.

6. A vibration and temperature control system for concrete molding, characterized in that, include: The module is used to build an initial prediction model of the two-way coupling effect of vibration and temperature control based on concrete mix proportion parameters, historical vibration control parameters and initial temperature field data. The correction module is used to collect vibration time-series data and temperature field spatial data in real time during the concrete pouring process, and input the vibration time-series data and temperature field spatial data into the initial prediction model for dynamic correction to obtain a real-time coupling effect prediction model. The prediction module is used to predict the density evolution trend and temperature gradient of concrete pouring in the next control cycle using the real-time coupling effect prediction model, and generate the predicted values ​​of density deviation and temperature gradient of concrete pouring in the next control cycle. The calculation module is used to calculate the vibration parameter adjustment amount and the temperature control intervention intensity based on the predicted density deviation value and the predicted temperature gradient value through a multi-objective collaborative optimization algorithm, and to obtain the collaborative control command for the next control cycle based on the calculation results. The collaborative control command includes a vibration control command and a temperature control adjustment command. The collaborative control module is used to control the concrete pouring system to perform collaborative optimization of vibration and temperature control during concrete pouring in the next control cycle based on the collaborative control instructions.

7. The vibration and temperature control system for concrete forming as described in claim 6, characterized in that, The building module is specifically used for: The concrete mix proportion parameters are processed by feature encoding to obtain material property feature vectors. Based on the material property feature vector and the historical vibration control parameters, a vibration response prediction sub-model is constructed. Based on the material property feature vector and the initial temperature field data, a temperature evolution prediction sub-model is constructed; The vibration response prediction sub-model and the temperature evolution prediction sub-model are bidirectionally coupled to obtain the initial prediction model of the vibration-temperature control bidirectional coupling effect.

8. The vibration and temperature control system for concrete forming as described in claim 7, characterized in that, The correction module is specifically used for: Vibration time sequence data during concrete pouring is collected in real time by a vibration sensor array to obtain the vibration signal time sequence. The temperature field spatial data during the concrete pouring process is collected in real time by a temperature sensor array to obtain the temperature distribution spatial matrix. The vibration signal time sequence and the temperature distribution spatial matrix are input into the initial prediction model to calculate the model prediction error. Based on the model prediction error, update the coupling parameters of the initial prediction model; The convergence of the initial prediction model after parameter updates is verified to obtain the real-time coupling effect prediction model.

9. The vibration and temperature control system for concrete forming as described in claim 8, characterized in that, The prediction module is specifically used for: The control cycle time window is divided according to the pouring progress, and the prediction time interval of the next control cycle is determined. The real-time coupling effect prediction model is performed in forward iterative calculation within the prediction time interval to obtain the density evolution trend curve and the temperature gradient distribution surface. The density evolution trend curve is subjected to target density deviation quantification processing to generate the predicted density deviation value of concrete pouring in the next control cycle. The maximum gradient value is extracted from the temperature gradient distribution surface to generate the predicted temperature gradient value for concrete pouring in the next control cycle.

10. The vibration and temperature control system for concrete forming as described in claim 9, characterized in that, The calculation module is specifically used for: Based on the predicted values ​​of density deviation and temperature gradient, density optimization weights and temperature control optimization weights are configured to obtain a multi-objective collaborative optimization weight matrix for density optimization and temperature control optimization. By setting constraints on the adjustment range of vibration parameters and the intervention intensity of temperature control measures, the optimization constraints for multi-objective collaborative optimization are obtained. Using the multi-objective collaborative optimization weight matrix and the optimization constraints as input, the optimal solution set for the vibration parameter adjustment amount and the temperature control intervention intensity is obtained by solving the non-dominated sorting genetic algorithm, thus obtaining the optimal vibration control command and the optimal temperature control adjustment command for the next control cycle.

Images