A method for mix proportion design of aluminum formwork concrete and a construction optimization system

Abstract

A method for mix proportion design of aluminum formwork concrete and a construction optimization system, the method comprising: collecting data of multiple aluminum forms to obtain a form data set; collecting multiple construction data during pouring of concrete to obtain a construction data set; identifying multiple finished concrete products based on image recognition technology to obtain a bubble defect data set; inputting the form data set, the construction data set and the bubble defect data set into a machine learning model to obtain a bubble probability prediction model; inputting multiple construction schemes into the bubble probability prediction model to obtain corresponding multiple bubble prediction results; selecting a construction scheme with the minimum bubble probability from the multiple bubble prediction results to obtain a corresponding target concrete mix proportion. By using the model to predict the relationship between the mix proportion and the surface bubble probability, the method can overcome the blindness of the empirical method, avoid a large number of trial and error, and quickly determine the optimization scheme of the mix proportion.

Description

Technical Field

[0001] This application relates to the field of data modeling, and in particular to a mix design method and construction optimization system for aluminum formwork concrete. Background Technology

[0002] Compared to other traditional formwork, aluminum formwork has advantages such as being lightweight and having good thermal conductivity, which can accelerate the construction progress and efficiency of concrete structures. However, aluminum formwork has poor chemical compatibility with concrete and high interfacial tension, which can easily create interfacial voids during demolding. This can lead to air trapped inside the concrete on the surface during the demolding process, forming air bubbles and voids. Many factors affect these air bubbles, including the concrete mix design parameters, vibration process, formwork material, and type of formwork lubricant.

[0003] In related technologies, the adjustment of mix proportion parameters mainly relies on the personal experience of engineers. Generally, a baseline mix proportion is initially determined based on experience from past engineering cases. During actual construction, the sand ratio is adjusted by observing the fluidity and workability of the concrete on-site. If air bubble defects occur, experienced engineers will slightly adjust the vibration parameters according to the location and distribution of the air bubbles.

[0004] However, the design of the empirical mixing ratio in this method relies heavily on personal experience, the mixing ratio scheme is difficult to reuse and promote, and the trial-and-error method to search for the optimal mixing ratio is time-consuming. Summary of the Invention

[0005] This application provides a mix design method and construction optimization system for aluminum formwork concrete. By using a model to predict the relationship between the mix proportion and the probability of surface air bubbles, it can overcome the blindness of the empirical method, avoid a lot of trial and error, and quickly determine the optimal mix proportion scheme.

[0006] Firstly, this application provides a mix design method for aluminum formwork concrete. The method includes: collecting material data and formwork interface agent data from multiple aluminum formworks to obtain a formwork data set; collecting multiple construction data during concrete pouring from the multiple aluminum formworks to obtain a construction data set; the construction data includes concrete mix proportion data and vibration data; the vibration data includes vibration frequency, vibration times, and vibration location; identifying multiple finished concrete samples using image recognition technology to obtain a bubble defect data set describing the number, size, and distribution characteristics of air bubbles; inputting the formwork data set, the construction data set, and the bubble defect data set into a machine learning model to obtain a bubble probability prediction model; inputting multiple construction schemes into the bubble probability prediction model to obtain corresponding multiple bubble prediction results; the construction scheme includes material data of the aluminum formwork, formwork interface agent data, concrete mix proportion data, and vibration data; the concrete mix proportion data of the multiple construction schemes are different; selecting the construction scheme with the lowest bubble probability from the multiple bubble prediction results to obtain the corresponding target concrete mix proportion.

[0007] In the above embodiments, template data, construction data, and bubble defect data obtained using image recognition technology are collected and input into a machine learning model to establish a bubble probability prediction model. This model can predict the bubble probability for different construction schemes and select the scheme with the lowest bubble probability to obtain the optimized concrete mix proportion. This achieves effective modeling of multiple factors affecting the generation of bubbles in concrete and intelligent prediction of bubble probability. Compared to previous methods that mainly relied on manual experience to determine the mix proportion, this embodiment can make the design of concrete mix proportions more scientific and reasonable, avoid a large number of trial and error, and quickly determine the optimal mix proportion scheme.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of inputting the template data set, the construction data set, and the bubble defect data set into the machine learning model to obtain the bubble probability prediction model, the method further includes: collecting multiple temperature data and multiple humidity data when the multiple aluminum templates are poured with concrete to obtain an environmental data set; inputting the environmental data set into the bubble probability prediction model to correct the bubble probability prediction model.

[0009] In the above embodiments, based on the original modeling, this method further collects environmental data on temperature and humidity during pouring, using this data as input to the corresponding environmental dataset and correcting the bubble probability prediction model. This allows the bubble probability prediction model to consider the influence of additional factors such as environmental temperature and humidity, thus improving the accuracy and reliability of the model's bubble probability prediction. Compared to the embodiments that only consider formwork, construction, and bubble data, the addition of environmental factors makes the concrete mix design more comprehensive and accurate, helping to further control the generation of concrete bubble defects.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, the method further includes: setting multiple objective functions for mix proportion optimization; the multiple objective functions include a function for minimizing bubble probability and a function for minimizing material cost; determining constraints on the mix proportion optimization variables; the constraints include water-cement ratio requirements and coarse aggregate dosage limits; using the non-dominated sorting genetic algorithm NSGA-II to optimize the multiple objective functions and obtain non-dominated solutions; setting weight coefficients to perform weighted evaluation on the non-dominated solutions and obtain evaluation values; and selecting the non-dominated solution with the highest evaluation value as the final mix proportion scheme.

[0011] In the above embodiments, by setting a multi-objective function that minimizes the bubble probability and material cost, the constraints for mix proportion optimization are determined. A non-dominated sorting genetic algorithm is then used to optimize the multi-objective function, obtaining a set of non-dominated solutions. A weighted evaluation method is then used to determine the optimal non-dominated solution, resulting in the final mix proportion scheme. Under the constraints, this approach achieves comprehensive optimization of the two key indicators of concrete bubble effect and material cost. Compared to traditional single-objective functions, the multi-objective method can obtain a more balanced and economical mix proportion scheme. Simultaneously, the non-dominated sorting genetic algorithm can effectively generate a global optimal solution, avoiding local optima, and ultimately achieving an economical and efficient optimization design of the concrete mix proportion.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, the setting of multiple objective functions for mix proportion optimization specifically includes: constructing a bubble probability prediction model based on deep learning, with the corresponding objective function for minimizing the bubble probability being minf1(x)=P(x); in the objective function for minimizing the bubble probability, x in f1(x) represents the mix proportion parameter, and P(x) represents the probability of bubble generation under the corresponding parameter; minimizing f1 can reduce the bubble probability; constructing a concrete material consumption calculation model, with the corresponding objective function for minimizing the material cost being minf2(x)=∑C i (x)*m i (x); In the objective function for minimizing material cost, x represents the mix proportion parameter, C i m is the unit price of the i-th type of concrete material. i (x) is the quantity of the i-th type of concrete material used in the mix proportion x. Minimizing f2 can reduce material costs.

[0013] In the above embodiments, by constructing a deep learning-based bubble probability prediction model and a concrete material consumption calculation model, two objective functions were determined: minimizing the bubble probability and minimizing the material cost. The deep learning-based bubble probability prediction model can achieve high-precision bubble probability prediction and is an important foundation for minimizing the bubble probability objective. The material consumption calculation model can accurately obtain the material cost of different mix proportions, achieving cost minimization. The establishment of these two models lays the model foundation for subsequent multi-objective optimization, enabling the optimization results to take into account both the concrete bubble effect and economic efficiency.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, in the concrete material consumption calculation module, the corresponding objective function for minimizing material cost is minf2(x)=∑C i (x)*m i Following step (x), the method further includes: constructing a prediction model for the early compressive strength of concrete, with the corresponding objective function for maximizing early strength being maxf3(x) = R. c(x,t0); R in the objective function that maximizes early intensity c (x,t0) represents the early compressive strength of concrete with mix proportion x, and t0 is the early time point. Maximizing f3 can improve the early strength. A prediction model for the later carbonation resistance of concrete is constructed, and the corresponding objective function for maximizing carbonation resistance is maxf4(x)=CR(x,t1). In this objective function for maximizing carbonation resistance, CR(x,t1) represents the later carbonation resistance of mix proportion x, and t1 is the expected later time of use. Maximizing f4 can improve carbonation resistance.

[0015] In the above embodiments, an early-age strength prediction model and a late-age carbonation resistance prediction model for concrete were further constructed, and two objective functions were determined: maximizing early-age strength and maximizing carbonation resistance. The early-age strength prediction model can predict the early-age strength levels of different mix proportions, and maximizing early-age strength is beneficial for construction. The carbonation resistance prediction model can evaluate long-term carbonation resistance, and maximizing carbonation resistance will improve the service life of concrete. The addition of these two new objective functions ensures that mix proportion optimization not only considers air bubbles and economy but also takes into account workability and service life, achieving overall optimization of multiple performance indicators.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of selecting the non-dominated solution with the highest evaluation value as the final mix proportion scheme, the method further includes: verifying whether the obtained final mix proportion scheme satisfies all constraints; if not, adjusting the weights of the plurality of objective functions and reselecting the optimal non-dominated solution.

[0017] In the above embodiments, after obtaining the optimized mix proportion scheme, it is verified whether all constraints are met. If not, the weight coefficients of the objective function are adjusted, and a new optimal non-dominated solution is selected. This verification and adjustment method ensures that the optimization result satisfies the constraints, avoiding the blind selection of infeasible mix proportion schemes. By adjusting the weight coefficients, a new optimal mix proportion can be obtained within the allowable range of constraints. Overall, the reliability and feasibility of the mix proportion optimization results are improved.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of selecting the construction scheme with the lowest bubble probability from the plurality of bubble prediction results to obtain the corresponding target concrete mix proportion, the method further includes: determining the printing scheme of the target concrete mix proportion; and printing out the target mix proportion.

[0019] In the above embodiments, a printing scheme for the target concrete mix proportion was determined, and the target mix proportion was printed out. This creates an intuitive physical record, facilitating on-site workers' query and use, and reducing the risk of misoperation. Furthermore, the printed output is not easily lost and can be stored and applied long-term. The standardized printed mix proportion template can improve the quality stability of different projects. The printed document is also stored as a project archive, meeting traceability requirements. Usage precautions can also be added to the printed version to avoid operational oversights.

[0020] Secondly, embodiments of this application provide a construction optimization system, which includes: a first data module for collecting material data and template interface agent data of multiple aluminum templates to obtain a template data set; a second data module for collecting multiple construction data during concrete pouring of the multiple aluminum templates to obtain a construction data set; the construction data includes concrete mix proportion data and vibration data; the vibration data includes vibration frequency, vibration times, and vibration location; a third data module for identifying multiple finished concrete products based on image recognition technology to obtain a bubble defect data set describing the number, size, and distribution characteristics of air bubbles; a model building module for inputting the template data set, the construction data set, and the bubble defect data set into a machine learning model to obtain a bubble probability prediction model; a model prediction module for inputting multiple construction schemes into the bubble probability prediction model to obtain corresponding multiple bubble prediction results; the construction scheme includes material data of aluminum templates, template interface agent data, concrete mix proportion data, and vibration data; the concrete mix proportion data of the multiple construction schemes are different; and a scheme determination module for selecting the construction scheme with the lowest bubble probability from the multiple bubble prediction results to obtain the corresponding target concrete mix proportion.

[0021] Thirdly, embodiments of this application provide a construction optimization system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, which includes computer instructions, and the one or more processors call the computer instructions to cause the construction optimization system to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a construction optimization system, cause the construction optimization system to perform the method described in the first aspect and any possible implementation thereof.

[0023] Fifthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a construction optimization system, cause the construction optimization system to perform the method described in the first aspect and any possible implementation thereof.

[0024] Understandably, the construction optimization systems provided in the second and third aspects, the computer program product provided in the fourth aspect, and the computer storage medium provided in the fifth aspect are all used to execute the methods provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0025] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0026] 1. By employing a method of collecting multi-source data and establishing a probability prediction model for air bubbles, it is possible to predict and model various factors affecting the generation of air bubbles in concrete. This effectively solves the inaccuracy problem of existing methods that mainly rely on manual experience to determine mix proportions. Consequently, it allows for more scientific and rational design of concrete mix proportions and control of air bubble defects in concrete. Simultaneously, it avoids extensive trial and error and quickly determines optimized mix proportion schemes.

[0027] 2. Because the model is corrected by taking into account temperature and humidity data, the prediction of bubble probability can be more accurate and reliable, effectively solving the problem of deviation in results caused by not considering environmental factors. Therefore, concrete mix design schemes can be more precise.

[0028] 3. Because a multi-objective optimization algorithm is used to solve the non-dominated solution set that satisfies the constraints and to evaluate and select the optimal solution, it is possible to achieve comprehensive optimization of multiple indicators such as early strength, bubble effect, material cost and carbonation resistance. This effectively solves the problem of incomplete and unbalanced optimization results caused by over-reliance on a single objective function, thereby obtaining a concrete mix design scheme that meets both technical requirements and is economical and efficient. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating the mix design method for aluminum formwork concrete in this embodiment of the application.

[0030] Figure 2 This is another flowchart illustrating the mix design method for aluminum formwork concrete in this application embodiment;

[0031] Figure 3 This is a schematic diagram of a functional module structure of the construction optimization system in an embodiment of this application;

[0032] Figure 4This is a schematic diagram of the physical device structure of the construction optimization system in the embodiments of this application. Detailed Implementation

[0033] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0034] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0035] To facilitate understanding, the application scenarios of the embodiments of this application are described below.

[0036] During the concrete pouring of a floor slab in a high-rise building, workers used aluminum formwork. The slab required C30 concrete, and a concrete mix design was created based on experience. Before pouring, the formwork was coated with oil. During pouring, concrete was sprayed from the pump truck pipes, filling the aluminum formwork. After the concrete had initially solidified, the formwork was removed, revealing numerous holes on the concrete surface, which were identified as air bubble defects. This batch of concrete failed to meet design strength requirements, resulting in significant economic losses for the construction company.

[0037] In related technologies, concrete mix proportions can be modified through human experience, such as by increasing the amount of fine aggregate, to prevent and improve concrete air bubbles.

[0038] In the implementation scenario of this technology, Engineer Wang was responsible for the quality management of the concrete construction of a floor slab. After conducting small-batch test pours based on the preliminary mix design, numerous air bubbles were found on the concrete surface. To prevent air bubble problems during large-scale construction, Engineer Wang decided to adjust the concrete mix proportions. First, drawing on years of pouring experience, he increased the fly ash content to improve the concrete's density, adjusting the fly ash to mineral powder ratio from 1:4 to 1:3. Simultaneously, he increased the fine sand content by 30 kg / m³. 3 .

[0039] Based on the revised mix design, Engineer Wang conducted another small-batch trial pouring to verify the results. The number of air bubbles on the concrete surface was significantly reduced after pouring. Therefore, he decided to use the revised concrete mix design in subsequent large-scale construction projects to prevent air bubble problems.

[0040] However, relying on manual experience to adjust the mix proportions is easily limited by the constraints of individual experience and may not necessarily find the optimal solution. Furthermore, adjusting the mix proportions manually is too subjective and arbitrary, failing to comprehensively consider all influencing factors, and the effectiveness and reliability of adjusting only individual parameters are difficult to guarantee.

[0041] The concrete mix design method in this application, which collects multi-source data to establish a bubble probability prediction model, enables the modeling and prediction of various factors affecting bubble generation. This not only overcomes the limitations of relying on human experience, but also enables automated and intelligent searching for the globally optimal mix design.

[0042] The following describes the scenario after adopting this embodiment.

[0043] Engineer Li's project plans to use a new type of aluminum formwork for pouring concrete floor slabs. To optimize the mix proportions and prevent air bubbles, he used a machine learning-based construction optimization system.

[0044] Engineer Li first collected template data such as technical parameters and coating agent brand for this template model and input it into the system's template dataset database. Then, he reviewed historical project construction records for this template, collecting data such as mix proportions and vibration parameters during pouring, and uploaded these to the system's construction dataset. Engineer Li sampled several batches of test-poured concrete, collected images, and labeled them with air bubble information, forming an air bubble defect dataset. The construction optimization system then invoked a machine learning model, using the three datasets as samples, to train an air bubble probability prediction model.

[0045] Engineer Li input several mix design options into the system, which predicted the probability of air bubbles for each option based on a trained model. The system returned mix design A, which had the lowest probability of air bubbles. Engineer Li adjusted the pouring mix design based on design A, and after actual pouring, the number of air bubble defects on the concrete surface was significantly reduced, meeting the quality requirements.

[0046] As can be seen, the machine learning-based construction optimization system in this application can not only intelligently and automatically optimize the concrete mix proportion, but also effectively solve the subjectivity and limitations of relying on human experience to design the mix proportion, thereby improving the quality and efficiency of mix proportion design and preventing the generation of concrete air bubble defects.

[0047] To facilitate understanding, the method provided in this implementation will be described in detail below, using the above scenario as an example. Please refer to [link / reference]. Figure 1This is a flowchart illustrating the mix design method for aluminum formwork concrete in this application embodiment.

[0048] S101. Collect material data and template interface agent data of multiple aluminum templates to obtain a template data set.

[0049] The construction optimization system will search for data on various specifications and models of aluminum formwork used in historical projects, collect the technical parameters of these aluminum formworks, including material grade, thickness, and locking type, and establish a unified formwork parameter table. At the same time, the construction optimization system will search for the construction records of these projects, collect the brand and model of the formwork coating materials used, coating thickness parameters, and other interface agent information, and organize it into an interface agent parameter table.

[0050] The construction optimization system associates the template models used in each project with the corresponding interface agent treatment methods, establishing a template-interface agent correspondence table. Finally, the template parameter table, interface agent parameter table, and template-interface agent correspondence table are integrated into a standardized template data set and stored in the database.

[0051] In some embodiments, the construction optimization system continues to collect template and interface agent data for new construction projects, updating the template dataset to enrich the dataset and improve model accuracy. Specific or outlier data in the dataset is labeled or removed to ensure the quality and validity of the dataset.

[0052] S102. Collect multiple construction data during concrete pouring using multiple aluminum formworks to obtain a construction data set.

[0053] Furthermore, the construction optimization system will search for construction records of historical concrete pouring projects using these aluminum formworks, and collect concrete mix design schemes, including parameters such as sand-cement ratio and admixture dosage. The system will also collect concrete vibration parameters for these projects, including information such as vibrator type, vibration frequency, vibration intensity, number of vibrations, and vibration location.

[0054] In some embodiments, for certain projects, the construction optimization system can also obtain specific operational data of construction personnel during the pouring process through querying, such as single-layer pouring height and vibration frequency.

[0055] Finally, the construction optimization system organizes and categorizes the collected mix proportion parameters, vibration parameters, and pouring operation data for each project, forming a standardized construction data set. Within this dataset, mix proportion parameters are correlated with vibration parameters and pouring operations on a project-by-project basis.

[0056] In some embodiments, the construction optimization system continuously supplements the data with new construction projects, constantly expanding the dataset's capacity and coverage. Simultaneously, it checks and cleans the dataset, deleting or correcting erroneous and abnormal data to ensure its accuracy.

[0057] S103. Based on image recognition technology, identify multiple finished concrete products to obtain a set of bubble defect data describing the number, size, and distribution characteristics of bubbles.

[0058] The construction optimization system collects images of the finished concrete surface after pouring using different aluminum formwork and mix proportions.

[0059] Specifically, the construction optimization system uses image processing methods to remove impurities while preserving concrete surface features. It then applies boundary detection algorithms from image recognition technology to identify bubble boundaries in the image, calculates the area of ​​each bubble, converts it to the equivalent sphere diameter, and obtains bubble size data. The system also counts the number of bubbles in each image, obtaining bubble count data. Finally, by analyzing the bubble distribution patterns in the image, the system determines whether the bubbles are concentrated or diffusely distributed.

[0060] Ultimately, the construction optimization system will be able to construct standardized data samples from each concrete sample, including the size, number, and distribution characteristics of air bubbles.

[0061] S104. Input the template data set, construction data set, and bubble defect data set into the machine learning model to obtain the bubble probability prediction model.

[0062] First, the construction optimization system will perform data cleaning and preprocessing on the collected template dataset, construction dataset, and bubble defect dataset. Data cleaning and preprocessing refers to removing duplicate and abnormal data.

[0063] The construction optimization system then determines the correspondence between datasets, i.e., which construction data corresponds to which template data. Next, it selects a suitable machine learning model for processing multi-source heterogeneous data, such as a deep learning neural network model.

[0064] According to the model's input requirements, the construction optimization system combines the three datasets into a feature matrix and a label vector. Further, the feature matrix and label vector are input into the selected model to configure its parameters. By training the model using the training set data, a preliminary bubble probability prediction model is obtained.

[0065] The construction optimization system uses validation set data to test the model's performance and further optimizes the model's parameters based on the results. Once the model's performance meets the requirements, the system retrains the model using the entire dataset to obtain the final bubble probability prediction model.

[0066] In some embodiments, the construction optimization system also evaluates the trained model to ensure its predictive performance is reliable. Simultaneously, it continuously improves the model's predictive capabilities by continuously adding new data to train it.

[0067] S105. Input multiple construction schemes into the bubble probability prediction model to obtain the corresponding multiple bubble prediction results.

[0068] Specifically, the construction optimization system searches the template dataset to determine the type of aluminum template to be used and its technical parameters. Simultaneously, it searches the interface agent dataset to select the corresponding interface treatment materials and methods for that template. Finally, the construction optimization system integrates the selected template data and interface treatment data with multiple construction schemes to form a complete sample of multiple construction schemes.

[0069] In some embodiments, users may design multiple candidate concrete mix proportions and vibration parameter combinations based on engineering requirements and experience data to form multiple initial construction schemes.

[0070] The construction optimization system sequentially inputs each construction plan sample into the pre-trained bubble probability prediction model. The model predicts a bubble probability value within the range of 0 to 1 for each construction plan sample. Simultaneously, the predicted bubble probability value for each construction plan is recorded, forming a bubble probability prediction result table.

[0071] The construction optimization system will also construct a bubble probability ranking table corresponding to multiple construction schemes based on the prediction results table.

[0072] S106. Select the construction scheme with the lowest bubble probability from multiple bubble prediction results to obtain the corresponding target concrete mix proportion.

[0073] Based on the bubble probability ranking table, the construction optimization system selects the construction scheme with the lowest bubble probability from multiple bubble prediction results. Further, the system searches for detailed concrete mix proportion parameters in this scheme, including the sand-cement ratio and coarse aggregate content. Finally, this mix proportion scheme is selected as the target mix proportion scheme, recorded in the optimization design table, and displayed to the user.

[0074] In some embodiments, after determining the printing scheme for the target concrete mix proportion, the construction optimization system will print out the target mix proportion for workers to prepare raw materials and mix concrete.

[0075] In some embodiments, the construction optimization system searches project logs from multiple pours using the same batch of aluminum formwork to collect temperature and humidity data for each pour. This data is collected in real-time from on-site data loggers during the pouring process, providing indoor temperature and humidity information. For projects with missing early data, the system supplements the environmental temperature and humidity data by searching for relevant meteorological records. Furthermore, the system organizes and categorizes the collected temperature and humidity data from each project's pouring day to create a standardized environmental dataset; within this dataset, temperature and humidity data are associated with the corresponding projects. The system continuously collects pouring environmental data for subsequent new projects, constantly expanding the capacity of the environmental dataset.

[0076] The construction optimization system utilizes the environmental dataset to supplement the feature variables of the bubble probability prediction model. Specifically, the system preprocesses the collected environmental dataset, converting it into an input format acceptable to the model. Then, by examining the structure of the bubble probability prediction model, it determines the interface location where temperature and humidity data are used as auxiliary input. The system maps the temperature and humidity data in the environmental dataset to the original template and construction data. It connects the auxiliary temperature and humidity data input to the corresponding interface location in the model. The original model parameters remain unchanged, freezing the main structure of the model. Only the newly added temperature and humidity data input portion is trained, adjusting the corresponding parameters. The model is then retrained on the training set to optimize the contribution of temperature and humidity data to the model.

[0077] In the above embodiments, machine learning methods are applied to achieve intelligent and automated mix design optimization, no longer relying solely on human experience. At the same time, multiple influencing factors are comprehensively considered to optimize the mix design scheme, resulting in more reliable effects. Furthermore, a standard dataset has been established to continuously accumulate data, thereby continuously improving the quality of mix design.

[0078] In practical applications, when implementing the mix design method for aluminum formwork concrete, the construction optimization system also needs to consider multiple indicators such as early strength, bubble effect, material cost and carbonation resistance to achieve a concrete mix design that meets both technical requirements and is economical and efficient.

[0079] The following provides supplementary information regarding the scenario in this embodiment.

[0080] Engineer Li's project involves pouring concrete for the basement roof slab. To address the air bubble problem arising from the mix design, he employed a construction optimization system. The system set two objective functions: minimizing the air bubble probability and cost, while also defining constraints such as the water-cement ratio. Furthermore, the system invoked a non-dominated sorting genetic algorithm to optimize and obtain a set of non-dominated solutions. By weighted scoring of these solutions, the system recommended the mix design with the highest score.

[0081] Engineer Li conducted a trial pour using the optimized mix proportions provided by the system, which significantly reduced air bubble defects. This optimized mix proportion scheme was then applied in subsequent large-scale construction, achieving high-quality pouring.

[0082] In light of the above scenarios, the method provided in this implementation will now be described in more detail. Please refer to [link / reference]. Figure 2 This is another flowchart illustrating the mix design method for aluminum formwork concrete in this application embodiment.

[0083] S201. Set multiple objective functions for mix ratio optimization.

[0084] The construction optimization system first determines the target variables for optimization—air bubble probability and material cost—based on project requirements data. Specifically, by setting a minimum air bubble probability, it aims to minimize the generation of air bubbles inside and on the surface of the concrete. Simultaneously, considering cost control requirements, a minimum material cost is set, meaning that while meeting technical specifications, the amount of material used is reduced, thereby lowering the cost of the mix design.

[0085] Specifically, the construction optimization system defines a bubble probability objective function, which involves establishing a bubble prediction model, inputting mix proportion variables, outputting a 0-1 probability value, and minimizing this probability. Simultaneously, the system defines a material cost objective function, inputting the unit price of each material and the mix proportion variables, calculating the total cost of the formula, and minimizing this cost. Finally, the system integrates these two objective functions into a set of objective functions for a multi-objective programming model.

[0086] In some embodiments, the construction optimization system constructs a bubble probability prediction model based on deep learning, with the corresponding objective function for minimizing the bubble probability being minf1(x) = P(x). In this objective function, x in f1(x) represents the mix proportion parameter, and P(x) represents the probability of bubble generation under the corresponding parameter. Minimizing f1 can reduce the bubble probability. Simultaneously, the construction optimization system constructs a concrete material consumption calculation model, with the corresponding objective function for minimizing material cost being minf2(x) = ∑C i (x)*m i (x); In the objective function for minimizing material cost, x represents the mix proportion parameter, C i m is the unit price of the i-th type of concrete material. i (x) is the quantity of the i-th type of concrete material used in the mix proportion x. Minimizing f2 can reduce material costs.

[0087] In some embodiments, the construction optimization system constructs a prediction model for the early compressive strength of concrete, with the corresponding objective function for maximizing early strength being maxf3(x) = R. c (x,t0); R in the objective function that maximizes early intensityc (x,t0) represents the early compressive strength of concrete with mix proportion x, and t0 is the early time point. Maximizing f3 can improve the early strength. At the same time, the construction optimization system will construct a prediction model for the later carbonation resistance of concrete. The corresponding objective function for maximizing carbonation resistance is maxf4(x)=CR(x,t1). In this objective function for maximizing carbonation resistance, CR(x,t1) represents the later carbonation resistance of mix proportion x, and t1 is the expected later use time. Maximizing f4 can improve carbonation resistance.

[0088] S202. Determine the constraints for the variables in the mix proportion optimization.

[0089] The construction optimization system determines the constraints of the mix proportion optimization variables based on historical data or user input.

[0090] Specifically, based on relevant specifications and experience data, the construction optimization system sets concrete mix proportions, i.e., the limiting ranges of each component, as constraints. For example, the water-cement ratio is constrained to 0.5 to control the fluidity and strength of the concrete; the sand-cement ratio is constrained to 30%–40% to ensure the compactness of the concrete; the water-cement ratio is constrained to 0.35–0.55 to affect the hydration reaction of the cementitious materials; and the coarse aggregate content is constrained to establish the maximum compaction packing rule.

[0091] In some embodiments, based on user selection, the construction optimization system may also set constraints such as the old material replacement rate limit for new concrete, the range constraints on concrete fluidity and slump, mixer speed, and mixing time.

[0092] S203. The non-dominated sorting genetic algorithm is used to optimize the multiple objective functions and obtain non-dominated solutions.

[0093] Specifically, the construction optimization system uses a non-dominated sorting genetic algorithm. The system defines chromosome encoding in the genetic algorithm, using binary encoding to represent the percentage of each component in the concrete mix proportion. The population is initialized by randomly generating multiple individuals that meet the constraints. For each generation of individuals, their fitness values ​​are calculated based on the objectives of minimizing bubble probability and minimizing cost. All individuals in the current generation are non-dominated sorted according to their fitness to determine different non-dominated levels. When selecting individuals to generate the next generation, individuals located in the first few non-dominated levels are prioritized to maintain diversity. Crossover and mutation operations are performed to generate a new population. The processes of fitness evaluation, non-dominated sorting, selection, crossover, and mutation are repeated. After a predetermined number of generations, the optimal non-dominated solution set is output.

[0094] The Non-dominated Sorting Genetic Algorithm (NSGA) is a genetic algorithm used to solve multi-objective optimization problems. Its main idea is to stratify the population based on non-dominance relationships. The best individuals in the non-dominated layers are retained to maintain population diversity. Genetic operators are used to generate a new population. The cycle of sorting, selection, and reproduction is repeated to obtain a Pareto-optimal solution set.

[0095] The specific steps include:

[0096] 1. Initialize the population.

[0097] 2. Perform non-dominated sorting on the population according to the objective function.

[0098] 3. Select the better solution based on the sorting results and crowding distance and keep it in the next generation.

[0099] 4. Perform crossover and mutation to generate new populations.

[0100] 5. Repeat steps 2-4 until the termination condition is met.

[0101] 6. Output the final non-dominated solution set.

[0102] Non-dominated solutions are a concept introduced for multi-objective optimization problems. In single-objective problems, each solution can be compared to others to distinguish their quality. However, in multi-objective problems, the concept of non-dominated solutions needs to be introduced for ranking. If in a solution, the value of each objective function is no worse than any other solution, then that solution is a non-dominated solution. That is, a non-dominated solution cannot be completely dominated by any other solution. The set of all non-dominated solutions in a multi-objective problem, i.e., the non-dominated solution set, constitutes the Pareto optimal front of the problem. These non-dominated solutions are all of good quality and cannot be compared in terms of quality. For a population, it can be divided into different non-dominated levels according to dominance relationships, and the solutions in each level are non-dominated solutions in that level. Multi-objective algorithms generally retain non-dominated solutions to maintain population diversity in order to obtain the Pareto optimal front.

[0103] In this scheme, the non-dominated solutions refer to the Pareto optimal concrete mix proportion schemes obtained after non-dominated sorting, including data such as the proportion / percentage of each component in the concrete, which have the optimal calculation results on the objective function.

[0104] Population is a crucial concept in genetic algorithms. In a genetic algorithm, the population consists of multiple individuals, each representing a possible solution. The population plays a vital role in the iterative process of the genetic algorithm, maintaining sample coverage of the entire solution space, ensuring diversity, and avoiding getting trapped in local optima. Individuals in the population communicate and update each other through genetic operators (heredity, crossover, mutation, etc.), driving the entire population towards better outcomes. As a large pool, the population can simultaneously accommodate and process multiple candidate solutions, increasing the global search capability. Evaluating the fitness of the population can distinguish between better and worse solutions, helping to preserve better genes. The population enables parallel computation, accelerating the algorithm's execution efficiency.

[0105] In this scheme, the population refers to the set of concrete mix proportion schemes that are continuously iterated and updated during the calculation process. This includes data such as the proportion / percentage of each component in the concrete. It is the initial concrete mix proportion scheme determined by the user or the construction optimization system, as well as the subsequent optimized scheme.

[0106] In genetic algorithms, genes are the basic units encoding individual characteristics, also called chromosomes. Individuals are composed of genes of a certain length, usually encoded in binary. Gene inheritance refers to the complete copying of the parent's gene sequence to the offspring, achieving the transmission of superior genes. Gene crossover refers to the recombination of the gene sequences of two parent individuals to generate new gene sequences, which serve as the genes of the offspring, creating new genotypes. Common types include single-point crossover, multi-point crossover, and uniform crossover. Gene mutation refers to the random alteration of certain gene loci in an individual, such as flipping locus values, introducing new genes, and increasing population diversity.

[0107] In this scheme, "gene" (chromosome) refers to the proportional data that encodes the mix proportions of each component in concrete. For example, the mixing ratio data such as sand and gravel ratio is encoded to form the "gene" of a concrete formula.

[0108] S204. Set weighting coefficients and perform a weighted evaluation on the non-dominated solution to obtain the evaluation value.

[0109] Specifically, the construction optimization system sets objective function weight coefficients w1 and w2 based on the relative importance of the optimization objectives. For each non-dominated solution i in the non-dominated solution set, its evaluation value f on each objective function is calculated. i1 and f i2 For each non-dominated solution i, a weighted comprehensive evaluation value F is calculated. i F i =w1f i1 +w2f i2 .

[0110] Each nondominated solution is determined by its comprehensive evaluation value F. i An evaluation rating is obtained. A higher F-value indicates a better overall effect after considering the weights. Thus, by setting weights, the non-dominated solution with the highest F-value is selected as the final optimization result.

[0111] S205. Select the non-dominated solution with the highest value among the evaluation values ​​as the final mix proportion scheme.

[0112] The construction optimization system applies the weighted evaluation values ​​F to all non-dominated solutions. i Sort the solutions from highest to lowest, and select the evaluation value F from the sorted non-dominated solution set. i The highest non-dominated solution, and the mix proportion scheme composed of the amounts of each component corresponding to that non-dominated solution, is the final determined optimal mix proportion scheme.

[0113] S206. Verify whether the obtained final mix proportion scheme meets all constraints.

[0114] The construction optimization system extracts the component dosage data from the final mix design and substitutes this data into the previously determined constraints for verification. For example, if the water-cement ratio constraint is 0.5, it verifies whether the water-cement ratio meets the 0.5 requirement; if the coarse aggregate dosage constraint is below a certain maximum value, it verifies whether the dosage is less than or equal to that maximum value.

[0115] The construction optimization system checks whether all constraints are met one by one.

[0116] S207. Obtain the final mix proportion scheme.

[0117] If all constraints pass verification, the mix design is feasible and can be used as the final optimization result, thus obtaining the final mix design.

[0118] S208. Adjust the weights of multiple objective functions and reselect the optimal nondominated solution.

[0119] If any constraint is not met, the mix design is infeasible. In this case, it is necessary to return to the previous round of optimization iterations and reselect the optimal non-dominated solution until a feasible optimal mix design that satisfies all constraints is obtained.

[0120] In this embodiment, a multi-objective optimization algorithm is used to solve the non-dominated solution set that satisfies the constraints and to evaluate and select the optimal solution. Therefore, it can achieve comprehensive optimization of multiple indicators such as early strength, bubble effect, material cost and carbonation resistance. It effectively solves the problem of incomplete and unbalanced optimization results caused by over-reliance on a single objective function, thereby obtaining a concrete mix design scheme that meets both technical requirements and is economical and efficient.

[0121] The construction optimization system in this application embodiment is described below from a module perspective. Please refer to [link / reference]. Figure 3 This is a schematic diagram of a functional module structure of the construction optimization system in this application embodiment.

[0122] The construction optimization system includes:

[0123] The first data module 301 is used to collect material data and interface agent data of multiple aluminum formworks to obtain a formwork data set; the second data module 302 is used to collect multiple construction data when the multiple aluminum formworks are used to pour concrete to obtain a construction data set; the construction data includes concrete mix proportion data and vibration data; the vibration data includes vibration frequency, vibration times and vibration location.

[0124] The third data module 303 is used to identify multiple finished concrete products based on image recognition technology to obtain a set of bubble defect data describing the number, size, and distribution characteristics of bubbles.

[0125] The model building module 304 is used to input the template data set, the construction data set, and the bubble defect data set into the machine learning model to obtain the bubble probability prediction model.

[0126] The model prediction module 305 is used to input multiple construction schemes into the bubble probability prediction model to obtain corresponding multiple bubble prediction results; the construction scheme includes material data of aluminum formwork, formwork interface agent data, concrete mix proportion data, and vibration data; the concrete mix proportion data of the multiple construction schemes are different;

[0127] The scheme determination module 306 is used to select the construction scheme with the lowest bubble probability from the multiple bubble prediction results to obtain the corresponding target concrete mix proportion.

[0128] In some embodiments, the construction optimization system further includes:

[0129] The fourth data module 307 is used to collect multiple temperature and humidity data during the concrete pouring of the multiple aluminum formworks to obtain an environmental data set;

[0130] The model correction module 308 is used to input the environmental data set into the bubble probability prediction model and correct the bubble probability prediction model.

[0131] In some embodiments, the model building module 304 in the construction optimization system specifically includes:

[0132] The function determination unit 3041 is used to set multiple objective functions for mix proportion optimization; these multiple objective functions include a function to minimize the bubble probability and a function to minimize the material cost.

[0133] The condition determination unit 3042 is used to determine the constraints of the mix proportion optimization variables; these constraints include water-cement ratio requirements and coarse aggregate dosage limits.

[0134] The model prediction module 305 in the construction optimization system specifically includes:

[0135] The function solving unit 3051 is used to optimize the multiple objective functions using the non-dominated sorting genetic algorithm NSGA-II to obtain non-dominated solutions;

[0136] Result determination unit 3052 is used to set weight coefficients, perform weighted evaluation on the non-dominated solution, and obtain the evaluation value;

[0137] The scheme determination module 306 in the construction optimization system is also used to select the non-dominated solution with the highest value among the evaluation values ​​as the final mix proportion scheme.

[0138] In some embodiments, the function determination unit 3041 specifically includes:

[0139] Bubble function subunit 30411 is used to construct a bubble probability prediction model based on deep learning. The corresponding objective function for minimizing the bubble probability is minf1(x)=P(x). In this objective function for minimizing the bubble probability, x in f1(x) represents the mix ratio parameter, and P(x) represents the bubble generation probability under the corresponding parameter. Minimizing f1 can reduce the bubble probability.

[0140] Material function subunit 30412 is used to construct the concrete material consumption calculation model, and the corresponding objective function for minimizing material cost is minf2(x)=∑C i (x)*m i (x); In the objective function for minimizing material cost, x represents the mix proportion parameter, C i m is the unit price of the i-th type of concrete material. i (x) is the quantity of the i-th type of concrete material used in the mix proportion x. Minimizing f2 can reduce material costs.

[0141] In some embodiments, the function determination unit 3041 further includes:

[0142] The compressive strength function subunit 30413 is used to construct a prediction model for the early compressive strength of concrete. The corresponding objective function for maximizing early strength is maxf3(x)=R. c (x,t0); R in the objective function that maximizes early intensity c (x,t0) represents the early compressive strength of concrete with mix proportion x, and t0 is the early time point. Maximizing f3 can improve the early strength.

[0143] The carbonation resistance function subunit 30414 is used to construct a prediction model for the later carbonation resistance performance of concrete. The corresponding objective function for maximizing carbonation resistance is maxf4(x)=CR(x,t1). In this objective function for maximizing carbonation resistance, CR(x,t1) represents the later carbonation resistance performance corresponding to the mix proportion x, and t1 is the expected later use time. Maximizing f4 can improve carbonation resistance.

[0144] In some embodiments, the construction optimization system further includes:

[0145] The scheme verification module 309 verifies whether the obtained final mix proportion scheme meets all constraints.

[0146] The scheme adjustment module 310 is used to adjust the weights of the multiple objective functions and reselect the optimal non-dominated solution when the final mix design does not meet all the constraints.

[0147] In some embodiments, the construction optimization system further includes:

[0148] Printing preparation module 311 is used to determine the printing scheme for the target concrete mix proportion;

[0149] The print output module 312 is used to print out the target mix proportion.

[0150] The construction optimization system in this application embodiment has been described above from the perspective of modular functional entities. The construction optimization system in this application embodiment is described below from the perspective of hardware processing. Please refer to [link / reference]. Figure 4 This is a schematic diagram of a physical device structure of the construction optimization system in this application embodiment.

[0151] It should be noted that, Figure 4 The structure of the construction optimization system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0152] like Figure 4As shown, the construction optimization system includes a Central Processing Unit (CPU) 401, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 402 or programs loaded from storage section 408 into Random Access Memory (RAM) 403, such as performing the methods described in the above embodiments. The RAM 403 also stores various programs and data required for system operation. The CPU 401, ROM 402, and RAM 403 are interconnected via a bus 404. An Input / Output (I / O) interface 405 is also connected to the bus 404.

[0153] The following components are connected to I / O interface 405: input section 406 including infrared detection devices, camera / video capture devices, mouse and keyboard, etc.; output section 407 including liquid crystal display (LCD) and warning lights, audio output devices, etc.; storage section 408 including hard disks, etc.; and communication section 409 including network interface cards such as LAN (Local Area Network) cards, modems, etc. Communication section 409 performs communication processing via a network such as the Internet. Drive 410 is also connected to I / O interface 405 as needed. Removable media 411, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on drive 410 as needed so that computer programs read from them can be installed into storage section 408 as needed.

[0154] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 409, and / or installed from removable medium 411. When the computer program is executed by central processing unit (CPU) 401, it performs the various functions defined in the present invention.

[0155] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0156] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0157] Specifically, the construction optimization system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the mix design method for aluminum formwork concrete provided in the above embodiment.

[0158] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the construction optimization system described in the above embodiments; or it may exist independently and not assembled into the construction optimization system. The storage medium carries one or more computer programs that, when executed by a processor of the construction optimization system, cause the construction optimization system to implement the methods provided in the above embodiments.

[0159] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0160] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0161] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website's site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0162] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A mix proportion design method of aluminum form concrete, applied to a construction optimization system, characterized in that, The method comprises: Collecting material data and template interface agent data of a plurality of aluminum templates to obtain a template data set; Collecting a plurality of construction data when the plurality of aluminum templates are used to pour concrete to obtain a construction data set; the construction data comprises concrete mix data and vibration data; the vibration data comprises vibration frequency, vibration times and vibration position; Identifying a plurality of finished concrete based on image recognition technology to obtain a bubble defect data set describing bubble quantity, size and distribution characteristics; Inputting the template data set, the construction data set and the bubble defect data set into a machine learning model to obtain a bubble probability prediction model; Inputting a plurality of construction schemes into the bubble probability prediction model to obtain a plurality of corresponding bubble prediction results; the construction schemes comprise material data of aluminum templates, template interface agent data, concrete mix data and vibration data; the concrete mix data of the plurality of construction schemes is different; Selecting a construction scheme with the smallest bubble probability from the plurality of bubble prediction results to obtain a corresponding target concrete mix; Setting a plurality of objective functions for mix optimization; the plurality of objective functions comprise a bubble probability minimization objective function and a material cost minimization objective function; Determining constraint conditions of mix optimization variables; the constraint conditions comprise water-cement ratio requirements and coarse aggregate usage limits; Optimizing the plurality of objective functions by using a non-dominated sorting genetic algorithm NSGA-II to obtain non-dominated solutions; Setting weight coefficients to perform weighted evaluation on the non-dominated solutions to obtain evaluation values; Selecting a non-dominated solution with the highest evaluation value from the evaluation values as a final mix scheme.

2. The method of claim 1, wherein, After the step of inputting the template data set, the construction data set and the bubble defect data set into the machine learning model to obtain the bubble probability prediction model, the method further comprises: Collecting a plurality of temperature data and a plurality of humidity data when the plurality of aluminum templates are used to pour concrete to obtain an environment data set; Inputting the environment data set into the bubble probability prediction model to correct the bubble probability prediction model.

3. The method of claim 1, wherein, The setting of the plurality of objective functions for mix optimization specifically comprises: A bubble probability prediction model based on deep learning is constructed, and a corresponding minimum bubble probability objective function is ; x in the minimum bubble probability objective function indicates a mixing ratio parameter, indicates a bubble generation probability under the corresponding parameter, and minimizing can reduce the bubble probability; A concrete material consumption calculation model is constructed, and a corresponding minimum material cost objective function is ; x in the minimum material cost objective function represents a mix proportion parameter, is a unit price of the i-th concrete material, is a use quantity of the i-th concrete material at the mix proportion x, and the minimum material cost can be reduced.​ 4. The method of claim 3, wherein, In the construction concrete material consumption calculation module, the corresponding minimum material cost objective function is After the step, the method further comprises: A concrete early compressive strength prediction model is constructed, and a corresponding maximum early strength objective function is ; wherein represents a mix proportion x in a concrete early compressive strength, is an early time point, and maximum can improve early strength; A prediction model for the later-stage carbonation resistance of concrete is constructed, and a maximized carbonation resistance objective function is ; the in the maximized carbonation resistance objective function represents the later-stage carbonation resistance of the mix proportion x, is a predicted later-stage time, and the maximized carbonation resistance can be improved.

5. The method of claim 1, wherein, After the step of selecting the non-dominated solution with the highest evaluation value as the final mix scheme, the method further comprises: Verifying whether the obtained final mix scheme satisfies all constraint conditions; If not, adjusting the weights of the plurality of objective functions and reselecting an optimal non-dominated solution.

6. The method of claim 1, wherein, After the step of selecting a construction scheme with the smallest bubble probability from the plurality of bubble prediction results to obtain a corresponding target concrete mix, the method further comprises: Determining a printing scheme of the target concrete mix; Printing and outputting the target mix.

7. A construction optimization system characterized by, Comprise: A first data module for collecting material data and template interface agent data of a plurality of aluminum templates to obtain a template data set; A second data module is configured to collect a plurality of construction data of the plurality of aluminum formworks pouring concrete, to obtain a construction data set; the construction data includes concrete mix data and vibrating data; the vibrating data includes vibrating frequency, vibrating times and vibrating position; A third data module is configured to identify a plurality of finished concrete based on image recognition technology, to obtain a bubble defect data set describing bubble quantity, size and distribution characteristics; A model construction module is configured to input the formwork data set, the construction data set and the bubble defect data set into a machine learning model, to obtain a bubble probability prediction model; A model prediction module is configured to input a plurality of construction schemes into the bubble probability prediction model, to obtain a plurality of corresponding bubble prediction results; the construction schemes include material data of the aluminum formwork, formwork interface agent data, concrete mix data and vibrating data; the concrete mix data of the plurality of construction schemes is different; A scheme determination module is configured to select a construction scheme with the smallest bubble probability from the plurality of bubble prediction results, to obtain a corresponding target concrete mix; The model construction module specifically includes: A function determination unit is configured to set a plurality of objective functions of mix optimization; the plurality of objective functions includes a bubble probability minimization objective function and a material cost minimization objective function; A condition determination unit is configured to determine constraint conditions of mix optimization variables; the constraint conditions include water-cement ratio requirements and coarse aggregate usage limits; The model prediction module specifically includes: A function solving unit is configured to use a non-dominated sorting genetic algorithm NSGA-II to optimize the plurality of objective functions, to obtain non-dominated solutions; A result determination unit is configured to set a weight coefficient, to perform weighted evaluation on the non-dominated solutions, to obtain evaluation values; The scheme determination module is further configured to select a non-dominated solution with the highest evaluation value as a final mix scheme.

8. A construction optimization system characterized by, One or more processors and memories; The memory is coupled with the one or more processors, and the memory is configured to store computer program codes including computer instructions, and the one or more processors are configured to invoke the computer instructions to enable the construction optimization system to perform the method in any one of claims 1-6. When the instructions are run on the construction optimization system, the construction optimization system is enabled to perform the method in any one of claims 1-6.

9. A computer-readable storage medium comprising instructions, characterized in that, ​

Images