Multi-source data fusion foam concrete construction process monitoring method and system

Abstract

The invention discloses a multi-source data fusion foam concrete construction process monitoring method and system, and relates to the technical field of concrete construction, the method comprises the following steps: obtaining a standard response data set of foam concrete, including raw material ratio parameters and performance index parameters which are stored in an associated manner; constructing and training a hybrid hierarchical performance prediction model, wherein the model comprises a shared prediction layer based on integrated learning and a plurality of independent output layers connected with the shared prediction layer; in combination with a multi-objective optimization algorithm, by taking compressive strength and cost optimization as an objective, constructing an evaluation function for global optimization, and obtaining a target mix proportion scheme; acquiring real-time process data associated with the target mix proportion scheme, and inputting the real-time process data into the mixing layering performance prediction model to obtain a prediction result; and comparing with a preset design target, and generating an adjustment instruction for adjusting the matching parameters of the subsequent stirring batches. The problem that in the prior art, the performance fluctuation of foam concrete of different batches is large due to the fact that the construction condition cannot be dynamically adjusted in real time is solved.

Description

Technical Field

[0001] This application relates to the field of concrete construction technology, specifically to a method and system for monitoring the construction process of foamed concrete using multi-source data fusion. Background Technology

[0002] With the development of the construction industry, foamed concrete is increasingly widely used in various projects. However, traditional construction monitoring methods mainly rely on manual experience and limited parameter monitoring, which makes it difficult to comprehensively and accurately reflect the actual performance and construction quality of foamed concrete. They also cannot be dynamically adjusted according to real-time construction conditions, resulting in large fluctuations in the performance of different batches of foamed concrete. Summary of the Invention

[0003] This application provides a method and system for monitoring the construction process of foamed concrete by integrating multi-source data, which solves the technical problem in the prior art that the construction situation cannot be dynamically adjusted in real time, resulting in large fluctuations in the performance of different batches of foamed concrete.

[0004] The technical solution to the above-mentioned technical problems in this application is as follows:

[0005] Firstly, this application provides a method for monitoring the construction process of foamed concrete using multi-source data fusion, the method comprising:

[0006] Obtain a standard response dataset for foamed concrete, wherein the standard response dataset includes associated stored raw material proportioning parameters and performance index parameters;

[0007] Based on the standard response dataset, a hybrid hierarchical performance prediction model is constructed and trained. The hybrid hierarchical performance prediction model includes a shared prediction layer based on ensemble learning and multiple independent output layers connected to the shared prediction layer.

[0008] By combining a multi-objective optimization algorithm with the goal of optimizing compressive strength and cost, and by constructing an evaluation function based on the hybrid stratified performance prediction model, global optimization is performed to obtain the target mix design.

[0009] Real-time process data associated with the target mix design is collected during real-time construction, and the real-time process data is input into the hybrid stratified performance prediction model to obtain performance prediction results;

[0010] The performance prediction results are compared with the preset design target, and adjustment instructions are generated based on the comparison results to adjust the mixing ratio parameters of subsequent batches.

[0011] Secondly, this application provides a multi-source data fusion-based monitoring system for the construction process of foamed concrete, including:

[0012] The data acquisition module is used to acquire the standard response dataset of foamed concrete, wherein the standard response dataset includes associated and stored raw material proportioning parameters and performance index parameters;

[0013] The model training module is used to construct and train a hybrid hierarchical performance prediction model based on the standard response dataset. The hybrid hierarchical performance prediction model includes a shared prediction layer based on ensemble learning and multiple independent output layers connected to the shared prediction layer.

[0014] The scheme acquisition module is used to combine a multi-objective optimization algorithm with the goal of optimizing compressive strength and cost, and to construct an evaluation function based on the hybrid hierarchical performance prediction model to perform global optimization and obtain the target mix ratio scheme.

[0015] The performance prediction module is used to collect real-time process data associated with the target mix design during real-time construction, and input the real-time process data into the hybrid layered performance prediction model to obtain the performance prediction results.

[0016] The instruction adjustment module is used to compare the performance prediction results with the preset design target, and generate adjustment instructions for adjusting the mixing ratio parameters of subsequent batches based on the comparison results.

[0017] This application provides one or more technical solutions, which have at least the following technical effects or advantages:

[0018] This application provides a method and system for monitoring the construction process of foamed concrete using multi-source data fusion. First, it utilizes multi-source data fusion to fully integrate information from various aspects, including raw material proportioning parameters and performance index parameters, to construct and train a mixed-layer performance prediction model. Based on real-time collected process data, it accurately predicts the performance of foamed concrete, providing a scientific basis for mix proportion adjustments during construction. By collecting and analyzing real-time process data associated with the target mix proportion scheme, potential problems during construction are identified. When the performance prediction results deviate from the preset design target, the system generates adjustment instructions to optimize the mix proportioning parameters for subsequent batches, thereby effectively reducing performance fluctuations between different batches of foamed concrete. Furthermore, by combining a multi-objective optimization algorithm to construct an evaluation function for global optimization, it not only ensures that the foamed concrete reaches the ideal compressive strength but also achieves good results in cost control, maximizing economic benefits while ensuring quality during construction.

[0019] Through the above technical solutions, the multi-source data fusion method and system for monitoring the construction process of foamed concrete provides a scientific, efficient and reliable solution for the construction of foamed concrete in the construction industry, which helps to improve construction quality and efficiency and promote the sustainable development of the construction industry. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the multi-source data fusion method for monitoring the construction process of foamed concrete provided in this application embodiment;

[0022] Figure 2 This is a schematic diagram of the structure of the foamed concrete construction process monitoring system with multi-source data fusion provided in the embodiments of this application.

[0023] The components represented by each number in the attached diagram are explained below:

[0024] Data acquisition module 11, model training module 12, scheme acquisition module 13, performance prediction module 14, instruction adjustment module 15. Detailed Implementation

[0025] This application provides a method and system for monitoring the construction process of foamed concrete by fusing multi-source data, which is used to address the technical problem in the prior art that the inability to dynamically adjust the construction situation in real time leads to large fluctuations in the performance of different batches of foamed concrete.

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0028] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use this application. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that this application can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid unnecessarily obscuring the description of this application. Therefore, this application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0029] Example 1, as Figure 1 As shown in the embodiments of this application, a method for monitoring the construction process of foamed concrete based on multi-source data fusion is provided, including:

[0030] S10: Obtain the standard response dataset of foamed concrete, wherein the standard response dataset includes associated stored raw material proportioning parameters and performance index parameters;

[0031] In this embodiment, a standard response dataset for foamed concrete is obtained by collecting historical construction project records. This dataset includes the raw material proportioning parameters and corresponding performance index parameters used in different projects. The raw material proportioning parameters include various key factors affecting the performance of foamed concrete, while the performance index parameters reflect the quality and characteristics of the foamed concrete.

[0032] The raw material proportioning parameters include at least the composition of the cementitious material, the dosage of the gas-generating agent, the water-to-material ratio, and the dosage of the gas-generating regulator. The performance indicators include at least compressive strength, dry density, fluidity, and expansion height.

[0033] In this embodiment, the acquired standard response dataset includes associated stored raw material proportioning parameters and performance index parameters. The raw material proportioning parameters include at least the composition of the cementitious materials, the dosage of the foaming agent, the water-to-material ratio, and the dosage of the foaming agent regulator. The composition of the cementitious materials determines the setting and hardening performance of the foamed concrete; different cementitious materials produce different strengths and durability during the hydration reaction. The dosage of the foaming agent affects the number and size of pores formed in the foamed concrete; a suitable dosage can give the foamed concrete good thermal insulation and sound insulation properties. The water-to-material ratio affects the fluidity and strength of the foamed concrete; an excessively high water-to-material ratio may reduce the concrete strength, while an excessively low ratio will affect its workability. The dosage of the foaming agent regulator can regulate the speed and stability of the foaming process, ensuring the uniformity of the foamed concrete.

[0034] Performance indicators include at least compressive strength, dry density, fluidity, and expansion height. Compressive strength is an indicator of the load-bearing capacity of foamed concrete structures, which is related to the safety and reliability of buildings; dry density reflects the porosity and quality of foamed concrete, which has a direct impact on thermal insulation and sound insulation effects; fluidity reflects the workability of foamed concrete during construction, and appropriate fluidity can ensure that the concrete fills the formwork uniformly; expansion height reflects the effect of the gas generation process and affects the volume stability of foamed concrete.

[0035] By storing and analyzing the correlation between raw material proportioning parameters and performance index parameters, a rich and accurate data foundation is provided for the subsequent construction and training of hybrid layered performance prediction models, thereby better enabling the monitoring and quality control of the foamed concrete construction process.

[0036] Specifically, step S10 in the method includes:

[0037] Obtain M raw material ratio parameters and N performance index parameters, and design a multi-factor orthogonal experiment accordingly to obtain an orthogonal experiment table, where M and N are both positive integers greater than or equal to 1;

[0038] A multi-factor orthogonal experiment is performed based on the orthogonal experimental table to obtain the multi-factor orthogonal experimental results, wherein the multi-factor orthogonal experimental results include a first raw material ratio parameter set and a first performance index parameter set;

[0039] Based on the results of the multi-factor orthogonal experiment, the performance sensitivity of the M raw material ratio parameters to the N performance index parameters was analyzed and determined.

[0040] Based on the performance sensitivity, the preliminary value range of the raw material ratio parameters of item M is obtained by extrapolation.

[0041] In this embodiment, firstly, M raw material proportioning parameters and N performance index parameters are obtained, and a multi-factor orthogonal experiment is designed to obtain an orthogonal experimental table, where M and N are both positive integers greater than or equal to 1. After obtaining the orthogonal experimental table, the multi-factor orthogonal experiment is executed to obtain experimental results containing the first set of raw material proportioning parameters and the first set of performance index parameters. The results of the multi-factor orthogonal experiment are analyzed to determine the performance sensitivity of each raw material proportioning parameter to different performance index parameters, that is, to clarify the degree of influence of each parameter on each performance index.

[0042] Then, the preliminary value range of the M raw material proportion parameters is obtained by extrapolation based on the performance sensitivity. For example, if the composition of the cementitious material is highly sensitive to the compressive strength and dry density of foamed concrete, the preliminary value range of the cementitious material composition is extrapolated based on the ideal range of compressive strength and dry density, combined with the test results. This provides a targeted parameter range for subsequent tests and optimizations, avoiding ineffective attempts in an excessively large parameter space and improving test efficiency and accuracy.

[0043] Furthermore, a standard response dataset for foamed concrete is obtained, wherein the standard response dataset includes associated stored raw material proportioning parameters and performance index parameters, and also includes:

[0044] Based on the preliminary value range, a second set of raw material ratio parameters for the M items is determined, and a single-factor experiment is conducted to obtain a second set of performance index parameters.

[0045] Based on the results of single-factor experiments, the slurry thickening curve and gas generation curve were extracted, and the thickening-gas generation matching degree was calculated accordingly.

[0046] Establish the correlation between the thickening-gas generation matching degree and the second set of performance index parameters;

[0047] Using the second raw material ratio parameter set from the single-factor test results as input factors, the thickening-gas generation matching degree as process parameters, and the second performance index parameter set as terminal performance, the standard response dataset is obtained by combining the correlation relationships and performing structured storage.

[0048] In this embodiment, firstly, after determining the preliminary value range of the M raw material proportioning parameters, a second set of raw material proportioning parameters is determined based on this value range to conduct single-factor experiments. In the single-factor experiments, one raw material proportioning parameter is changed each time while keeping other parameters constant, thereby testing the individual effect of each parameter on the performance of foamed concrete and obtaining the second set of performance index parameters.

[0049] Secondly, based on the results of single-factor experiments, the slurry thickening curve and gas generation curve were extracted. The slurry thickening curve reflects the consistency change of the foamed concrete slurry during mixing and setting, while the gas generation curve reflects the amount of gas generated by the gas-generating agent at different times. Analysis of these two curves reveals the physical changes of foamed concrete during the molding process. The thickening-gas generation matching degree was then calculated, which is an indicator of the coordination between the slurry thickening rate and the gas generation rate during foamed concrete molding. If the thickening rate is too fast and the gas generation rate is too slow, insufficient pore formation may occur, affecting the thermal insulation and sound insulation performance of the foamed concrete; conversely, if the gas generation rate is too fast and the thickening rate is too slow, pores may rupture, reducing the strength of the concrete.

[0050] Furthermore, a correlation is established between the thickening-gas generation matching degree and the second performance index parameter set. Using the second raw material proportion parameter set from the single-factor test results as input factors, the thickening-gas generation matching degree as process parameters, and the second performance index parameter set as terminal performance, the correlation is used for structured storage to obtain a standard response dataset. Based on this structured storage method, the data becomes more organized, providing a comprehensive and accurate data foundation for the subsequent construction and training of the hybrid hierarchical performance prediction model. This helps to more accurately predict the performance of foamed concrete and achieve effective monitoring and quality control of the construction process.

[0051] S20: Based on the standard response dataset, construct and train a hybrid hierarchical performance prediction model, which includes a shared prediction layer based on ensemble learning and multiple independent output layers connected to the shared prediction layer.

[0052] In this embodiment, firstly, since the standard response dataset contains associated stored raw material ratio parameters and performance index parameters, a hybrid hierarchical performance prediction model is built and trained based on the standard response dataset using artificial neural networks. The hybrid hierarchical performance prediction model includes a shared prediction layer based on ensemble learning and multiple independent output layers connecting the shared prediction layer.

[0053] Furthermore, the shared prediction layer of ensemble learning makes preliminary predictions on the performance of foamed concrete. By integrating different algorithms and models, it mines potential information in the data, improving the accuracy and stability of the predictions. Multiple independent output layers can then make individual predictions for different performance indicators. Each independent output layer focuses on a specific performance indicator, analyzing in detail the relationship between that indicator and the raw material proportioning parameters.

[0054] Specifically, step S20 in the method includes:

[0055] Initialize the configuration of the shared prediction layer based on the performance sensitivity.

[0056] Using the second raw material ratio parameter set as training input and the thickening-gas generation matching degree as supervision, multiple base models in the shared prediction layer are trained respectively, and integrated by combining a voting mechanism;

[0057] Using the thickening-gas generation matching degree as input and the second performance index parameter set as supervision, N independent output layers are constructed and trained for each of the N performance index parameters;

[0058] The hybrid hierarchical performance prediction model is generated by connecting the input terminals of the multiple independent output layers to the output terminal of the shared prediction layer.

[0059] In this embodiment, the shared prediction layer is first initialized based on performance sensitivity. Performance sensitivity reflects the degree of influence of various raw material ratio parameters on performance index parameters, thereby setting the initial parameters of the shared prediction layer.

[0060] Secondly, the second set of raw material proportioning parameters was determined after preliminary value range screening. This set, containing parameters that affect the performance of foamed concrete, was used as the training input. Thickening-gas evolution matching degree was used as supervision to train multiple base models in the shared prediction layer. Thickening-gas evolution matching degree is a key indicator for measuring the coordination between the thickening rate and gas evolution rate of the slurry during foamed concrete molding, providing accurate supervision information for the training of the base models. During training, each base model learns different features and patterns, which are then integrated using a voting mechanism. The voting mechanism combines the prediction results of multiple base models, improving the accuracy and stability of the shared prediction layer.

[0061] The voting mechanism makes decisions based on the prediction results of multiple base models. For example, a simple majority voting method selects the prediction result that appears most frequently as the final prediction result. By integrating multiple base models through the voting mechanism, the advantages of each base model are utilized, the limitations of a single model are reduced, and the reliability of the shared prediction layer for the preliminary prediction of foamed concrete performance is improved.

[0062] Subsequently, using the thickening-gas generation matching degree as input and the second performance index parameter set as supervision, N independent output layers are constructed and trained for each of the N performance index parameters. Each independent output layer focuses on a specific performance index and analyzes the relationship between that index and the raw material ratio parameters.

[0063] Finally, the inputs of multiple independent output layers are connected to the output of the shared prediction layer to generate a hybrid layered performance prediction model. The output of the shared prediction layer provides preliminary prediction information for the independent output layers, which then perform more detailed predictions based on this information. This enables the entire hybrid layered performance prediction model to accurately predict various performance indicators of foamed concrete based on the raw material proportioning parameters.

[0064] For example, a hybrid hierarchical performance prediction model is built and trained based on an artificial neural network.

[0065] First, data acquisition involves extracting the second raw material ratio parameter set and the second performance index parameter set, as well as the thickening-gas generation matching degree data, from the standard response dataset as training data.

[0066] Secondly, the model is built by designing a shared prediction layer and independent output layers. The shared prediction layer can employ multiple different types of base models, such as decision trees and support vector machines, which are integrated through ensemble learning. The independent output layers construct a corresponding number of sub-networks based on the number of performance metrics, and then initialize the parameters of the shared prediction layer according to performance sensitivity.

[0067] Finally, model training was performed using the second raw material ratio parameter set as input and the thickening-gas generation matching degree as the supervision signal to train multiple base models in the shared prediction layer. The training framework was constructed using the Adam optimizer and the mean squared error (MSE) loss function, with a batch size of 32 and a total of 50 training epochs. An early stopping mechanism (patience=5) was introduced, automatically terminating the training process when the validation set loss did not decrease for five consecutive epochs, resulting in the trained base models in the shared prediction layer. Each base model continuously adjusted its weights and biases to minimize the error between the prediction result and the supervision signal. A voting mechanism was used to integrate the prediction results of multiple base models, for example, simple majority voting, to obtain the final output of the shared prediction layer.

[0068] Using the thickening-gas-generating matching degree as input and the second performance index parameter set as supervision signal, N independent output layers are trained respectively. Each independent output layer, targeting a specific performance index, adjusts its parameters through backpropagation to improve the prediction accuracy of that performance index. The training framework is constructed using the Adam optimizer and the mean squared error (MSE) loss function, with a batch size of 32 and a total of 50 training epochs. An early stopping mechanism (patience=5) is introduced: if the validation set loss does not decrease for 5 consecutive epochs, the training process is automatically terminated. The inputs of the trained independent output layers are then connected to the output of a shared prediction layer to form a complete hybrid hierarchical performance prediction model. During the connection process, it is ensured that data can be correctly transferred and processed between different layers.

[0069] The initial configuration of the shared prediction layer based on the performance sensitivity includes:

[0070] Obtain the performance sensitivity of the raw material ratio parameters mentioned in item M, and form M performance sensitivity groups, each of which includes N performance sensitivity items;

[0071] Based on the results of the multi-factor orthogonal experiment, range and variance analysis were performed on all N performance index parameters, and the corresponding normalization was performed.

[0072] Using the normalized range as the first weighting coefficient and the normalized variance as the second weighting coefficient, the M performance sensitivity groups are traversed, and the N performance sensitivities in each performance sensitivity group are weighted and fused to obtain the ratio parameter intensity coefficient.

[0073] Based on the density coefficient of the ratio parameter, determine the base model scale ratio corresponding to the raw material ratio parameter M, and initialize multiple base models of the shared prediction layer according to the base model scale ratio.

[0074] In this embodiment, firstly, the performance sensitivity of M raw material proportioning parameters is obtained, forming M performance sensitivity groups, each containing N performance sensitivities, reflecting the degree of influence of each raw material proportioning parameter on different performance indicators. Based on the results of multi-factor orthogonal experiments, range and variance analysis are performed on the N performance indicator parameters.

[0075] Range analysis can identify the order of importance of each factor's influence on performance indicators, while analysis of variance can determine the significance of each factor's influence on performance indicators. Normalizing the analysis results makes the analysis results of different indicators comparable.

[0076] Secondly, using the normalized range as the first weighting coefficient and the normalized variance as the second weighting coefficient, the N performance sensitivities in the M performance sensitivity groups are weighted and fused to obtain the proportioning parameter intensity coefficient. The proportioning parameter intensity coefficient comprehensively considers the influence of range and variance on performance sensitivity, reflecting the importance of each raw material proportioning parameter.

[0077] Then, based on the intensity coefficient of the proportioning parameter, the base model size ratio corresponding to the M raw material proportioning parameters is determined. The base model size ratio reflects the importance and proportion of each raw material proportioning parameter in the shared prediction layer; parameters with higher importance correspond to relatively larger base model sizes. Multiple base models in the shared prediction layer are initialized according to the base model size ratio to improve the accuracy and stability of the prediction.

[0078] For example, assume that M is 3, representing the three raw material proportioning parameters: cementitious material dosage, foaming agent dosage, and water-cement ratio; and N is 2, representing the two performance parameters: compressive strength and dry density.

[0079] The performance sensitivity of three raw material proportioning parameters was obtained, forming three performance sensitivity groups, each containing two performance sensitivities. For example, the performance sensitivity of the amount of cementitious material to compressive strength and dry density is 0.6 and 0.4, respectively; the performance sensitivity of the amount of foaming agent to the two performance indicators is 0.3 and 0.7, respectively; and the performance sensitivity of the water-cement ratio to the two performance indicators is 0.5 and 0.5, respectively.

[0080] Assume the normalized ranges are 0.6 and 0.4, and the normalized variances are 0.7 and 0.3, respectively. Using the normalized range as the first weighting coefficient and the normalized variance as the second weighting coefficient, a weighted fusion of performance sensitivities is performed for each performance sensitivity group.

[0081] For the performance sensitivity group of cementitious material dosage, the unit weight coefficient of the mix proportion obtained after weighted fusion is (0.6×0.6+0.7×0.4)=0.64;

[0082] For the performance sensitivity group of foaming agent dosage, the obtained specific gravity coefficient of the ratio parameter is (0.3×0.6+0.7×0.4)=0.46;

[0083] For the performance sensitivity group of water-cement ratio, the obtained proportion parameter density coefficient is (0.5×0.6+0.5×0.7)=0.65.

[0084] Based on the density coefficient, the base model size ratios corresponding to the amounts of cementitious materials, foaming agent, and water-cement ratio are determined. For example, assuming the calculated base model size ratios are 0.3, 0.2, and 0.5, respectively, multiple base models in the shared prediction layer are then initialized according to these base model size ratios. For instance, if the shared prediction layer contains decision trees, support vector machines, and neural networks, resources are allocated and parameters are adjusted according to the base model size ratios. This allows the base model corresponding to the water-cement ratio to occupy a larger proportion in the shared prediction layer, thereby improving the accuracy and stability of foamed concrete performance prediction.

[0085] Further, using the second raw material ratio parameter set as training input and the thickening-gas generation matching degree as supervision, multiple base models in the shared prediction layer are trained respectively, including:

[0086] Based on the single-factor categories of the second raw material ratio parameter set, the second raw material ratio parameter set is divided into M single-parameter sets;

[0087] Match the corresponding thickening-gas generation matching degree for each of the M single parameter sets to form M single parameter sample sets;

[0088] Using the intensity coefficient of the ratio parameter as the exchange probability and the reciprocal of the intensity coefficient of the ratio parameter as the exchange ratio, perform iterative random exchanges among the M single-parameter sample sets, and train multiple base models in the shared prediction layer based on the results of the iterative random exchanges.

[0089] In this embodiment, firstly, based on the single-factor category of the second raw material ratio parameter set, the second raw material ratio parameter set is divided into M single-parameter sets. Each single-parameter set corresponds to one raw material ratio parameter. A corresponding thickening-gas generation matching degree is matched for each of the M single-parameter sets, forming M single-parameter sample sets. The thickening-gas generation matching degree serves as supervisory information, providing feedback for each single-parameter sample set, making the training process more targeted.

[0090] Secondly, iterative random exchanges are performed among M single-parameter sample sets, using the density coefficient of the matching parameter as the exchange probability and the reciprocal of the density coefficient as the exchange ratio. During the iterative random exchange process, each single-parameter sample set has a certain probability of exchanging data with other sample sets, and the exchange ratio is determined by the reciprocal of the density coefficient of the matching parameter.

[0091] Finally, based on the results of iterative random swapping, multiple base models in the shared prediction layer are trained and divided into M groups, corresponding to M raw material proportioning parameters. During training, each base model learns and adjusts based on the swapped sample set, continuously optimizing its weights and biases to improve its predictive ability for foamed concrete performance. Different base models can learn different features and patterns. By integrating the base models through ensemble learning, the advantages of each base model are fully utilized, reducing the limitations of a single model, thereby improving the reliability and accuracy of the shared prediction layer's preliminary prediction of foamed concrete performance.

[0092] For example, suppose M is 4, representing four different raw material ratio parameters, forming four single-parameter sample sets. During the iterative random exchange process, a single-parameter sample set may exchange data with other sample sets with a certain exchange probability. The exchange ratio is determined by the reciprocal of the density coefficient of the ratio parameter. After multiple iterative random exchanges, the exchanged sample sets are used to train multiple base models in the shared prediction layer. Each base model continuously adjusts its parameters to adapt to the new sample data, ultimately improving the prediction performance of the entire shared prediction layer.

[0093] S30: Combining a multi-objective optimization algorithm with the goal of optimizing compressive strength and cost, and using the hybrid stratified performance prediction model to construct an evaluation function for global optimization, the target mix design is obtained.

[0094] In this embodiment, a multi-objective optimization algorithm is used, with compressive strength and cost as optimization objectives. A hybrid hierarchical performance prediction model is used to construct an evaluation function for global optimization, and the optimal solution is found under various constraints.

[0095] The evaluation function is constructed based on compressive strength and cost. It compares the output of the hybrid stratified performance prediction model with the target value to calculate a comprehensive evaluation value. A global optimization approach is used to find the optimal solution among all possible combinations of raw material ratios that yields the best evaluation function value.

[0096] Specifically, step S30 in the method includes:

[0097] Initialize the population, where each individual in the population represents an alternative matching scheme;

[0098] A multi-objective evaluation function is constructed, wherein the first objective term of the multi-objective evaluation function is the performance prediction value output based on the hybrid hierarchical performance prediction model, and the second objective term is the material cost calculated based on the alternative mix proportion scheme.

[0099] Configure constraints based on the foamed concrete demand information of the target scenario;

[0100] The multi-objective optimization algorithm is run iteratively to find the best fit. The fitness of each individual in the population is calculated based on the multi-objective evaluation function and the constraints. The population is updated through selection, crossover, and mutation operations until the termination condition is met. The optimal multi-objective evaluation function value is then output as the target fit ratio scheme.

[0101] In this embodiment of the application, firstly, the population is initialized by randomly generating a certain number of individuals, each representing an alternative mix design that covers different combinations of raw material ratios.

[0102] Secondly, a multi-objective evaluation function is constructed. The first objective is based on the performance prediction value output by the hybrid stratified performance prediction model, such as performance indicators like compressive strength. The second objective is based on the material cost calculated from alternative mix proportion schemes.

[0103] Then, constraints are configured based on the foamed concrete requirements of the target scenario. These constraints include minimum compressive strength requirements, maximum cost limits, and specific workability requirements for foamed concrete. By setting these constraints, it is ensured that the optimized solution meets the needs of the actual application.

[0104] Then, a multi-objective optimization algorithm is run iteratively to find the best candidate. In each iteration, the fitness of each individual in the population is calculated based on the multi-objective evaluation function and constraints. The fitness value reflects the individual's performance in satisfying the objective function and constraints.

[0105] Among them, the selection operation retains individuals with higher fitness and eliminates individuals with lower fitness, so that the population evolves in a better direction; the crossover operation simulates gene exchange in biological heredity, exchanging some information between two excellent individuals to generate new individuals and increase the diversity of the population; the mutation operation randomly changes some genes of individuals to prevent the algorithm from getting stuck in local optima.

[0106] During the iteration process, the population is updated through selection, crossover, and mutation operations until a termination condition is met. The termination condition can be reaching a preset maximum number of iterations, or the population's fitness value no longer showing significant improvement within a certain number of iterations. When the termination condition is met, the individual with the optimal multi-objective evaluation function value is output as the target mix design. The target mix design achieves a good balance between the performance and cost of foamed concrete, meeting the performance requirements of the target scenario while ensuring reasonable cost. It provides scientific and effective mix design guidance for actual foamed concrete construction, improving construction quality and economic benefits.

[0107] S40: During real-time construction, collect real-time process data associated with the target mix proportion scheme, and input the real-time process data into the hybrid stratified performance prediction model to obtain performance prediction results;

[0108] In this embodiment of the application, during real-time construction, sensors and data acquisition devices are used to collect real-time process data associated with the target mix design, including the actual input of raw materials, mixing time, ambient temperature and humidity, etc. The real-time process data is input into the mixed layered performance prediction model. The model will predict the performance of foamed concrete based on the rules and characteristics learned in previous training, and obtain the performance prediction results.

[0109] Performance prediction results include multiple performance indicators of foamed concrete, such as compressive strength, dry density, and thickening-gas generation matching. By predicting these performance indicators, we can promptly understand the performance status of foamed concrete under current construction conditions and determine whether it meets the expected quality requirements. If the prediction results show that certain performance indicators are not up to standard, adjustment measures can be taken in a timely manner. For example, if the predicted compressive strength is insufficient, the amount of cementitious material can be appropriately increased; if the thickening-gas generation matching is not ideal, the amount of foaming agent or the mixing process can be adjusted.

[0110] S50: Compare the performance prediction results with the preset design target, and generate adjustment instructions for adjusting the mixing ratio parameters of subsequent batches based on the comparison results.

[0111] In this embodiment of the application, the performance prediction results are compared in detail with the preset design goals, which are usually determined based on specific engineering requirements and quality standards.

[0112] If there is a deviation between the performance prediction results and the preset design target, the system will generate corresponding adjustment instructions based on the type and degree of the deviation. The adjustment instructions are specifically targeted for different performance index deviations.

[0113] For example, when the predicted compressive strength is lower than the design target, adjustment instructions may require increasing the amount of cementitious material, and may also involve fine-tuning the water-cement ratio to ensure that the workability of the foamed concrete is maintained while increasing strength. If certain densities do not meet requirements, it may be necessary to adjust the amount of foaming agent or optimize the gradation of raw materials.

[0114] When addressing discrepancies in the thickening-gas generation matching, the adjustment instructions focus more on modifying the stirring process and time. If the matching is not ideal, the stirring time may be appropriately extended, or the stirring speed and method may be changed to promote better synergy between the thickening and gas generation processes.

[0115] In summary, compared to existing technologies, this application achieves multi-source data fusion for monitoring the construction process of foamed concrete through a mechanism data layer, an intelligent core layer, and a monitoring application layer. First, raw material proportioning parameters and performance index parameters are acquired, and corresponding multi-factor orthogonal experiments are designed. The results of these experiments are obtained, and the performance sensitivity of raw material proportioning parameters to performance index parameters is analyzed. Based on the obtained preliminary value range, a second set of raw material proportioning parameters is determined for single-factor experiments, and a refined database is established based on the experimental results. Then, an artificial neural network is used for accurate performance prediction, combined with a multi-objective optimization algorithm and a mixed-layer performance prediction model to construct an evaluation function for global optimization. Finally, the prediction results are compared with the preset design targets, and adjustment instructions are generated to adjust the proportioning parameters of subsequent batches. This effectively reduces uncertainties during construction and improves the quality stability of foamed concrete.

[0116] In summary, the embodiments of this application have at least the following technical effects:

[0117] This application provides a method for monitoring the construction process of foamed concrete using multi-source data fusion. First, it utilizes multi-source data fusion to fully integrate information from various aspects, including raw material proportioning parameters and performance indicators, to construct and train a mixed-layer performance prediction model. Based on real-time collected process data, it accurately predicts the performance of foamed concrete, providing a scientific basis for mix proportion adjustments during construction. By collecting and analyzing real-time process data associated with the target mix proportion scheme, potential problems during construction are identified. When the performance prediction results deviate from the preset design target, the system generates adjustment instructions to optimize the mix proportioning parameters for subsequent batches, effectively reducing performance fluctuations between different batches of foamed concrete. Furthermore, by combining a multi-objective optimization algorithm to construct an evaluation function for global optimization, it not only ensures that the foamed concrete reaches the ideal compressive strength but also achieves good cost control, maximizing economic benefits while ensuring quality during construction. Through the above technical solution, the multi-source data fusion method for monitoring the construction process of foamed concrete provides a scientific, efficient, and reliable solution for foamed concrete construction in the construction industry, helping to improve construction quality and efficiency and promoting the sustainable development of the construction industry.

[0118] Example 2, as Figure 2 As shown, based on the same inventive concept as the multi-source data fusion foamed concrete construction process monitoring method provided in Embodiment 1, this application also provides a multi-source data fusion foamed concrete construction process monitoring system, including:

[0119] Data acquisition module 11 is used to acquire a standard response dataset of foamed concrete, wherein the standard response dataset includes associated stored raw material proportioning parameters and performance index parameters;

[0120] The model training module 12 is used to construct and train a hybrid hierarchical performance prediction model based on the standard response dataset. The hybrid hierarchical performance prediction model includes a shared prediction layer based on ensemble learning and multiple independent output layers connected to the shared prediction layer.

[0121] The scheme acquisition module 13 is used to combine a multi-objective optimization algorithm with the goal of optimizing compressive strength and cost, and to construct an evaluation function based on the hybrid hierarchical performance prediction model to perform global optimization and obtain the target mix ratio scheme.

[0122] The performance prediction module 14 is used to collect real-time process data associated with the target mix ratio scheme during real-time construction, and input the real-time process data into the hybrid layered performance prediction model to obtain the performance prediction result;

[0123] The instruction adjustment module 15 is used to compare the performance prediction results with the preset design target, and generate adjustment instructions for adjusting the mixing ratio parameters of subsequent batches based on the comparison results.

[0124] In one embodiment, the data acquisition module 11 is specifically used for:

[0125] Obtain M raw material ratio parameters and N performance index parameters, and design a multi-factor orthogonal experiment accordingly to obtain an orthogonal experiment table, where M and N are both positive integers greater than or equal to 1;

[0126] A multi-factor orthogonal experiment is performed based on the orthogonal experimental table to obtain the multi-factor orthogonal experimental results, wherein the multi-factor orthogonal experimental results include a first raw material ratio parameter set and a first performance index parameter set;

[0127] Based on the results of the multi-factor orthogonal experiment, the performance sensitivity of the M raw material ratio parameters to the N performance index parameters was analyzed and determined.

[0128] Based on the performance sensitivity, the preliminary value range of the raw material ratio parameters of item M is obtained by extrapolation.

[0129] The raw material proportioning parameters include at least the composition of the cementitious material, the dosage of the gas-generating agent, the water-to-material ratio, and the dosage of the gas-generating regulator. The performance indicators include at least compressive strength, dry density, fluidity, and expansion height.

[0130] Furthermore, in one embodiment, a standard response dataset for foamed concrete is obtained, wherein the standard response dataset includes associated stored raw material proportioning parameters and performance index parameters, and further includes:

[0131] Based on the preliminary value range, a second set of raw material ratio parameters for the M items is determined, and a single-factor experiment is conducted to obtain a second set of performance index parameters.

[0132] Based on the results of single-factor experiments, the slurry thickening curve and gas generation curve were extracted, and the thickening-gas generation matching degree was calculated accordingly.

[0133] Establish the correlation between the thickening-gas generation matching degree and the second set of performance index parameters;

[0134] Using the second raw material ratio parameter set from the single-factor test results as input factors, the thickening-gas generation matching degree as process parameters, and the second performance index parameter set as terminal performance, the standard response dataset is obtained by combining the correlation relationships and performing structured storage.

[0135] In one embodiment, the model training module 12 is specifically used for:

[0136] Initialize the configuration of the shared prediction layer based on the performance sensitivity.

[0137] Using the second raw material ratio parameter set as training input and the thickening-gas generation matching degree as supervision, multiple base models in the shared prediction layer are trained respectively, and integrated by combining a voting mechanism;

[0138] Using the thickening-gas generation matching degree as input and the second performance index parameter set as supervision, N independent output layers are constructed and trained for each of the N performance index parameters;

[0139] The hybrid hierarchical performance prediction model is generated by connecting the input terminals of the multiple independent output layers to the output terminal of the shared prediction layer.

[0140] Furthermore, in one embodiment, initializing the shared prediction layer configuration based on the performance sensitivity includes:

[0141] Obtain the performance sensitivity of the raw material ratio parameters mentioned in item M, and form M performance sensitivity groups, each of which includes N performance sensitivity items;

[0142] Based on the results of the multi-factor orthogonal experiment, range and variance analysis were performed on all N performance index parameters, and the corresponding normalization was performed.

[0143] Using the normalized range as the first weighting coefficient and the normalized variance as the second weighting coefficient, the M performance sensitivity groups are traversed, and the N performance sensitivities in each performance sensitivity group are weighted and fused to obtain the ratio parameter intensity coefficient.

[0144] Based on the density coefficient of the ratio parameter, determine the base model scale ratio corresponding to the raw material ratio parameter M, and initialize multiple base models of the shared prediction layer according to the base model scale ratio.

[0145] Further, in one embodiment, using the second raw material ratio parameter set as training input and the thickening-gas generation matching degree as supervision, multiple base models in the shared prediction layer are trained respectively, including:

[0146] Based on the single-factor categories of the second raw material ratio parameter set, the second raw material ratio parameter set is divided into M single-parameter sets;

[0147] Match the corresponding thickening-gas generation matching degree for each of the M single parameter sets to form M single parameter sample sets;

[0148] Using the intensity coefficient of the ratio parameter as the exchange probability and the reciprocal of the intensity coefficient of the ratio parameter as the exchange ratio, perform iterative random exchanges among the M single-parameter sample sets, and train multiple base models in the shared prediction layer based on the results of the iterative random exchanges.

[0149] In one embodiment, the solution acquisition module 13 is specifically used for:

[0150] Initialize the population, where each individual in the population represents an alternative matching scheme;

[0151] A multi-objective evaluation function is constructed, wherein the first objective term of the multi-objective evaluation function is the performance prediction value output based on the hybrid hierarchical performance prediction model, and the second objective term is the material cost calculated based on the alternative mix proportion scheme.

[0152] Configure constraints based on the foamed concrete demand information of the target scenario;

[0153] The multi-objective optimization algorithm is run iteratively to find the best fit. The fitness of each individual in the population is calculated based on the multi-objective evaluation function and the constraints. The population is updated through selection, crossover, and mutation operations until the termination condition is met. The optimal multi-objective evaluation function value is then output as the target fit ratio scheme.

[0154] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0155] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0156] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for monitoring the construction process of foamed concrete using multi-source data fusion, characterized in that, include: Obtain a standard response dataset for foamed concrete, wherein the standard response dataset includes associated stored raw material proportioning parameters and performance index parameters; Based on the standard response dataset, a hybrid hierarchical performance prediction model is constructed and trained. The hybrid hierarchical performance prediction model includes a shared prediction layer based on ensemble learning and multiple independent output layers connected to the shared prediction layer. By combining a multi-objective optimization algorithm with the goal of optimizing compressive strength and cost, and by constructing an evaluation function based on the hybrid stratified performance prediction model, global optimization is performed to obtain the target mix design. Real-time process data associated with the target mix design is collected during real-time construction, and the real-time process data is input into the hybrid stratified performance prediction model to obtain performance prediction results; The performance prediction results are compared with the preset design target, and adjustment instructions are generated based on the comparison results to adjust the mixing ratio parameters of subsequent batches.

2. The method for monitoring the construction process of foamed concrete by multi-source data fusion as described in claim 1, characterized in that, Obtain a standard response dataset for foamed concrete, wherein the standard response dataset includes associated stored raw material proportioning parameters and performance index parameters, including: Obtain M raw material ratio parameters and N performance index parameters, and design a multi-factor orthogonal experiment accordingly to obtain an orthogonal experiment table, where M and N are both positive integers greater than or equal to 1; A multi-factor orthogonal experiment is performed based on the orthogonal experimental table to obtain the multi-factor orthogonal experimental results, wherein the multi-factor orthogonal experimental results include a first raw material ratio parameter set and a first performance index parameter set; Based on the results of the multi-factor orthogonal experiment, the performance sensitivity of the M raw material ratio parameters to the N performance index parameters was analyzed and determined. Based on the performance sensitivity, the preliminary value range of the raw material ratio parameters of item M is obtained by extrapolation.

3. The method for monitoring the construction process of foamed concrete by multi-source data fusion as described in claim 2, characterized in that, Obtain a standard response dataset for foamed concrete, wherein the standard response dataset includes associated stored raw material proportioning parameters and performance index parameters, and also includes: Based on the preliminary value range, a second set of raw material ratio parameters for the M items is determined, and a single-factor experiment is conducted to obtain a second set of performance index parameters. Based on the results of single-factor experiments, the slurry thickening curve and gas generation curve were extracted, and the thickening-gas generation matching degree was calculated accordingly. Establish the correlation between the thickening-gas generation matching degree and the second set of performance index parameters; Using the second raw material ratio parameter set from the single-factor test results as input factors, the thickening-gas generation matching degree as process parameters, and the second performance index parameter set as terminal performance, the standard response dataset is obtained by combining the correlation relationships and performing structured storage.

4. The method for monitoring the construction process of foamed concrete by multi-source data fusion as described in claim 3, characterized in that, Based on the standard response dataset, a hybrid hierarchical performance prediction model is constructed and trained. This model includes a shared prediction layer based on ensemble learning and multiple independent output layers connected to the shared prediction layer, comprising: Initialize the configuration of the shared prediction layer based on the performance sensitivity. Using the second raw material ratio parameter set as training input and the thickening-gas generation matching degree as supervision, multiple base models in the shared prediction layer are trained respectively, and integrated by combining a voting mechanism; Using the thickening-gas generation matching degree as input and the second performance index parameter set as supervision, N independent output layers are constructed and trained for each of the N performance index parameters; The hybrid hierarchical performance prediction model is generated by connecting the input terminals of the multiple independent output layers to the output terminal of the shared prediction layer.

5. The method for monitoring the construction process of foamed concrete by multi-source data fusion as described in claim 4, characterized in that, Based on the performance sensitivity, initialize the configuration of the shared prediction layer, including: Obtain the performance sensitivity of the raw material ratio parameters mentioned in item M, and form M performance sensitivity groups, each of the performance sensitivity groups including N performance sensitivity items; Based on the results of the multi-factor orthogonal experiment, range and variance analysis were performed on all N performance index parameters, and the corresponding normalization was performed. Using the normalized range as the first weighting coefficient and the normalized variance as the second weighting coefficient, the M performance sensitivity groups are traversed, and the N performance sensitivities in each performance sensitivity group are weighted and fused to obtain the ratio parameter intensity coefficient. Based on the density coefficient of the ratio parameter, determine the base model scale ratio corresponding to the raw material ratio parameter M, and initialize multiple base models of the shared prediction layer according to the base model scale ratio.

6. The method for monitoring the construction process of foamed concrete by multi-source data fusion as described in claim 4, characterized in that, Using the second raw material ratio parameter set as training input and the thickening-gas generation matching degree as supervision, multiple base models in the shared prediction layer are trained respectively, including: Based on the single-factor categories of the second raw material ratio parameter set, the second raw material ratio parameter set is divided into M single-parameter sets; Match the corresponding thickening-gas generation matching degree for each of the M single parameter sets to form M single parameter sample sets; Using the intensity coefficient of the ratio parameter as the exchange probability and the reciprocal of the intensity coefficient of the ratio parameter as the exchange ratio, perform iterative random exchanges among the M single-parameter sample sets, and train multiple base models in the shared prediction layer based on the results of the iterative random exchanges.

7. The method for monitoring the construction process of foamed concrete by multi-source data fusion as described in claim 6, characterized in that, Combining a multi-objective optimization algorithm, with the goal of optimizing compressive strength and cost, and using the aforementioned hybrid stratified performance prediction model to construct an evaluation function for global optimization, the target mix design is obtained, including: Initialize the population, where each individual in the population represents an alternative matching scheme; A multi-objective evaluation function is constructed, wherein the first objective term of the multi-objective evaluation function is the performance prediction value output based on the hybrid hierarchical performance prediction model, and the second objective term is the material cost calculated based on the alternative mix proportion scheme. Configure constraints based on the foamed concrete demand information of the target scenario; The multi-objective optimization algorithm is run iteratively to find the best fit. The fitness of each individual in the population is calculated based on the multi-objective evaluation function and the constraints. The population is updated through selection, crossover, and mutation operations until the termination condition is met. The optimal multi-objective evaluation function value is then output as the target fit ratio scheme.

8. The method for monitoring the construction process of foamed concrete by multi-source data fusion as described in claim 1, characterized in that, The raw material proportioning parameters include at least the composition of the cementitious material, the dosage of the gas-generating agent, the water-to-material ratio, and the dosage of the gas-generating regulator.

9. The method for monitoring the construction process of foamed concrete by multi-source data fusion as described in claim 1, characterized in that, The performance indicators include at least compressive strength, dry density, flowability, and expansion height.

10. A multi-source data fusion-based monitoring system for foamed concrete construction process, characterized in that, For performing the method according to any one of claims 1-9, comprising: The data acquisition module is used to acquire the standard response dataset of foamed concrete, wherein the standard response dataset includes associated and stored raw material proportioning parameters and performance index parameters; The model training module is used to construct and train a hybrid hierarchical performance prediction model based on the standard response dataset. The hybrid hierarchical performance prediction model includes a shared prediction layer based on ensemble learning and multiple independent output layers connected to the shared prediction layer. The scheme acquisition module is used to combine a multi-objective optimization algorithm with the goal of optimizing compressive strength and cost, and to construct an evaluation function based on the hybrid hierarchical performance prediction model to perform global optimization and obtain the target mix ratio scheme. The performance prediction module is used to collect real-time process data associated with the target mix design during real-time construction, and input the real-time process data into the hybrid layered performance prediction model to obtain the performance prediction results. The instruction adjustment module is used to compare the performance prediction results with the preset design target, and generate adjustment instructions for adjusting the mixing ratio parameters of subsequent batches based on the comparison results.

Images