Concrete proportion optimization method and system

Abstract

The invention discloses a concrete proportion optimization method and system, and relates to the technical field of concrete proportion optimizing.The method comprises the steps that multiple concrete samples are obtained, attribute detection is conducted on the various concrete samples, a corresponding training concrete proportion set is obtained, a preset initial attribute prediction model is trained through the training concrete proportion set, and a prediction result is obtained; a target attribute prediction model is generated, an initial particle swarm is constructed with the concrete proportion as particles, the initial particle swarm is optimized according to the target attribute prediction model and a preset concrete evaluation function, and the optimized concrete proportion is obtained. The technical problems that traditional concrete proportion optimization mainly depends on artificial experience and multiple tests, a large amount of time and resources need to be consumed, the rapid changing construction requirement is difficult to adapt, and the concrete generation reliability is reduced are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of concrete mix ratio optimization, and in particular to a concrete mix ratio optimization method and system. Background Art

[0002] During the concrete production process, its mix design directly determines the mechanical properties, durability, and construction feasibility of the structure. Therefore, how to optimize the concrete mix is ​​crucial.

[0003] Traditional concrete mix optimization mainly relies on manual experience and multiple experiments, which requires a lot of time and resources. Moreover, the stability of concrete cannot be guaranteed as environmental factors change. It is difficult to adapt to rapidly changing construction needs, which reduces the reliability of concrete production. Summary of the Invention

[0004] The present invention provides a concrete mix ratio optimization method and system, which solves the technical problems that traditional concrete mix ratio optimization mainly relies on manual experience and multiple experiments, requires a lot of time and resources, is difficult to adapt to rapidly changing construction needs, and reduces the reliability of concrete production.

[0005] A first aspect of the present invention provides a method for optimizing concrete mix ratio, comprising:

[0006] Acquire multiple concrete samples, perform property detection on each of the concrete samples, and obtain corresponding training concrete mix sets;

[0007] Using the training concrete mix set to train a preset initial property prediction model to generate a target property prediction model;

[0008] An initial particle group is constructed using the concrete mix ratio as particles, and the initial particle group is optimized according to the target attribute prediction model and a preset concrete evaluation function to obtain an optimized concrete mix ratio.

[0009] Optionally, the step of performing property detection on various concrete samples to obtain corresponding training concrete mix sets includes:

[0010] Using the slump cone method, the slump of each concrete sample is tested to obtain a plurality of slumps;

[0011] Performing hydration treatment on the various concrete samples respectively to obtain a plurality of target concrete samples and a plurality of concrete test data;

[0012] Performing compressive strength tests on each of the target concrete samples to obtain multiple critical loads;

[0013] Ratio processing is performed on each of the critical loads and the preset pressure-bearing area to obtain multiple compressive strengths;

[0014] Data preprocessing is performed on each of the compressive strengths, each of the concrete test data, and each of the slumps to obtain a corresponding training concrete mix set.

[0015] Optionally, the training concrete mix set includes a first training set and a second training set, the initial attribute prediction model includes a first prediction model and a second prediction model, and the step of using the training concrete mix set to train the preset initial attribute prediction model to generate a target attribute prediction model includes:

[0016] Using the first training set to train the first prediction model to generate a first target prediction model;

[0017] Using the second training set to train the second prediction model to generate a second target prediction model;

[0018] The first target prediction model and the second target prediction model are sequentially connected to generate a target attribute prediction model.

[0019] Optionally, the step of optimizing the initial particle swarm according to the target attribute prediction model and a preset concrete evaluation function to obtain an optimized concrete mix ratio includes:

[0020] Inputting each concrete mix ratio in the initial particle group into the target property prediction model to obtain multiple compressive strengths and multiple slumps;

[0021] Based on a preset concrete evaluation function, respectively calculating the compressive strength evaluation index and the slump evaluation index corresponding to each concrete mix ratio according to each compressive strength and each slump;

[0022] The initial particle group is optimized using the respective compressive strength evaluation indexes and the respective slump evaluation indexes to obtain an optimized concrete mix ratio.

[0023] Optionally, the concrete evaluation function includes a compressive strength evaluation function and a slump evaluation function, and the step of calculating the compressive strength evaluation index and the slump evaluation index corresponding to each concrete mix ratio according to each compressive strength and each slump based on the preset concrete evaluation function comprises:

[0024] Inputting each of the concrete mix ratios into the compressive strength evaluation function to obtain a plurality of compressive strength reference values ​​and a plurality of compressive strength dynamic correction values;

[0025] Based on the preset compressive strength weight, a weighted calculation is performed on the compressive strength, the compressive strength reference value, and the compressive strength dynamic correction value associated with each concrete mix ratio to obtain a compressive strength evaluation index corresponding to each concrete mix ratio;

[0026] Inputting each of the concrete mix ratios into the slump evaluation function to obtain a plurality of slump reference values ​​and a plurality of slump dynamic correction values;

[0027] Based on the preset slump weight, weighted calculation is performed on the slump, slump reference value and slump dynamic correction value associated with each concrete mix ratio to obtain the slump evaluation index corresponding to each concrete mix ratio.

[0028] Optionally, the step of optimizing the initial particle group using the respective compressive strength evaluation indices and the respective slump evaluation indices to obtain an optimized concrete mix ratio comprises:

[0029] Performing ratio processing on the preset reference compressive coefficient and each of the compressive strength evaluation indexes to obtain a plurality of first ratios;

[0030] Performing difference processing on a preset benchmark slump evaluation index and each of the slump evaluation indices to obtain a plurality of first differences;

[0031] performing sum processing on the absolute value of each of the first differences and the corresponding first ratio respectively to obtain multiple fitnesses;

[0032] Updating the initial particle swarm according to each fitness to obtain a corresponding updated particle swarm;

[0033] Determining whether the update times of the initial particle swarm is greater than or equal to a preset iteration threshold;

[0034] If the update number is less than the iteration threshold, the updated particle group is used as a new initial particle group, and the step of inputting each concrete mix ratio in the initial particle group into the target attribute prediction model to obtain multiple compressive strengths and multiple slumps is skipped and executed;

[0035] If the update number is greater than or equal to the iteration threshold, the concrete mix ratio corresponding to the minimum fitness value in the updated particle swarm is selected as the optimized concrete mix ratio.

[0036] A second aspect of the present invention provides a concrete mix ratio optimization system, comprising:

[0037] A detection module is used to obtain multiple concrete samples, perform property detection on each of the concrete samples, and obtain a corresponding training concrete mix set;

[0038] A training module, configured to train a preset initial property prediction model using the training concrete mix set to generate a target property prediction model;

[0039] The optimization module is used to construct an initial particle group using the concrete mix ratio as particles, and optimize the initial particle group according to the target attribute prediction model and a preset concrete evaluation function to obtain an optimized concrete mix ratio.

[0040] A third aspect of the present invention provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of any one of the above-described methods for optimizing concrete mix proportions.

[0041] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements any of the above-mentioned concrete mix ratio optimization methods.

[0042] A fifth aspect of the present invention provides a computer program product, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program comprises program instructions, wherein when the program instructions are executed by a computer, the computer is caused to execute any one of the above-described methods for optimizing concrete mix proportions.

[0043] It can be seen from the above technical solutions that the present invention has the following advantages:

[0044] The present invention obtains a variety of concrete samples, performs attribute detection on various concrete samples, obtains corresponding training concrete mix sets, uses the training concrete mix sets to train a preset initial attribute prediction model, generates a target attribute prediction model, constructs an initial particle group using the concrete mix as a particle, optimizes the initial particle group according to the target attribute prediction model and a preset concrete evaluation function, and obtains an optimized concrete mix. This overcomes the technical problems that traditional concrete mix optimization mainly relies on manual experience and multiple experiments, requires a large amount of time and resources, is difficult to adapt to rapidly changing construction needs, and reduces the reliability of concrete generation. Compared with traditional concrete mix optimization methods, the present invention uses the training concrete mix set to train a preset initial attribute prediction model to generate a target attribute prediction model, quantifies the effects of material ratio and environmental factors on compressive strength and slump, and simultaneously optimizes the initial particle group according to the target attribute prediction model and a preset concrete evaluation function to obtain an optimized concrete mix, thereby improving the optimization efficiency of the concrete mix and the reliability of concrete generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 A flowchart of the steps of a concrete mix ratio optimization method provided in Example 1 of the present invention;

[0047] Figure 2 A flowchart of the steps of a concrete mix ratio optimization method provided in the second embodiment of the present invention;

[0048] Figure 3 A line graph showing changes in the compressive strength evaluation index of concrete provided in Example 2 of the present invention;

[0049] Figure 4 A line graph showing changes in the slump evaluation index of concrete provided in Example 2 of the present invention;

[0050] Figure 5 This is a structural block diagram of a concrete mix ratio optimization system provided in Example 3 of the present invention;

[0051] Figure 6 This is a structural block diagram of a computer device provided in Example 4 of the present invention. DETAILED DESCRIPTION

[0052] The embodiments of the present invention provide a concrete mix ratio optimization method and system for solving the technical problems that traditional concrete mix ratio optimization mainly relies on manual experience and multiple experiments, requires a lot of time and resources, is difficult to adapt to rapidly changing construction needs, and reduces the reliability of concrete generation.

[0053] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0054] See also Figure 1 , Figure 1 This is a flowchart of the steps of a concrete mix ratio optimization method provided in Example 1 of the present invention.

[0055] The present invention provides a method for optimizing concrete mix ratio, comprising:

[0056] Step 101: Obtain multiple concrete samples, perform property detection on the various concrete samples, and obtain corresponding training concrete mix sets.

[0057] Concrete specimen refers to a small concrete sample prepared according to a specific mix ratio (water-cement ratio, sand ratio, etc.) for testing performance.

[0058] The training concrete mix set refers to the training data set of the deep learning network, including but not limited to the compressive strength, slump, water-cement ratio, sand ratio, ambient temperature, ambient humidity, cement activity and aggregate moisture content of various concrete samples.

[0059] In an embodiment of the present invention, multiple concrete samples are constructed based on different water-cement ratios and sand ratios. The slump of each concrete sample is tested using the slump cone method to obtain multiple slumps. Each concrete sample is then hydrated to obtain multiple target concrete samples and multiple concrete test data. Compressive strength testing is performed on each target concrete sample to obtain multiple critical loads. Each critical load and a preset bearing area are input into a preset compressive strength function to obtain multiple compressive strengths. Data preprocessing is performed on each compressive strength, each concrete test data, and each slump to obtain a corresponding training concrete mix set.

[0060] It should be noted that the compressive strength function is specifically:

[0061]

[0062] in, is the compressive strength, is the critical load, is the pressure bearing area.

[0063] Step 102: Use the training concrete mix set to train the preset initial property prediction model to generate a target property prediction model.

[0064] In an embodiment of the present invention, a training concrete mix set is used to train a preset initial attribute prediction model to generate a target attribute prediction model. The initial attribute prediction model includes a first prediction model and a second prediction model. For example, the first prediction model is trained using the water-cement ratio, sand ratio, ambient temperature, and ambient humidity from the training concrete mix set as inputs, and cement activity and aggregate moisture content as labels. The second prediction model is trained using the cement activity and aggregate moisture content from the training concrete mix set as inputs, and compressive strength and slump as labels.

[0065] Step 103: construct an initial particle swarm using the concrete mix ratio as particles, and optimize the initial particle swarm according to the target attribute prediction model and the preset concrete evaluation function to obtain an optimized concrete mix ratio.

[0066] In an embodiment of the present invention, an initial particle swarm is constructed, wherein each particle in the initial particle swarm corresponds to a concrete mix ratio. Each concrete mix ratio in the initial particle swarm is input into a target attribute prediction model to obtain multiple compressive strengths and multiple slumps. Based on a preset concrete evaluation function, a compressive strength evaluation index and a slump evaluation index corresponding to each concrete mix ratio are calculated based on each compressive strength and each slump. The initial particle swarm is optimized using each compressive strength evaluation index and each slump evaluation index to obtain an optimized concrete mix ratio.

[0067] In an embodiment of the present invention, a plurality of concrete samples are obtained and property tests are performed on the various concrete samples to obtain a corresponding training concrete mix set. The training concrete mix set is used to train a preset initial property prediction model to generate a target property prediction model. An initial particle group is constructed using the concrete mix as a particle. The initial particle group is optimized according to the target property prediction model and a preset concrete evaluation function to obtain an optimized concrete mix. This overcomes the technical problem that traditional concrete mix optimization relies mainly on manual experience and multiple experiments, requires a large amount of time and resources, is difficult to adapt to rapidly changing construction needs, and reduces the reliability of concrete generation. Compared with traditional concrete mix optimization methods, the present invention uses a training concrete mix set to train a preset initial property prediction model to generate a target property prediction model, quantifies the effects of material ratio and environmental factors on compressive strength and slump, and simultaneously optimizes the initial particle group according to the target property prediction model and a preset concrete evaluation function to obtain an optimized concrete mix, thereby improving the optimization efficiency of the concrete mix and the reliability of concrete generation.

[0068] See also Figure 2 , Figure 2 This is a flowchart of the steps of a concrete mix ratio optimization method provided in Example 2 of the present invention.

[0069] The present invention provides a method for optimizing concrete mix ratio, comprising:

[0070] Step 201: Acquire multiple concrete samples, perform property detection on the various concrete samples, and obtain corresponding training concrete mix sets.

[0071] Furthermore, step 201 includes the following sub-steps:

[0072] S11. Use the slump cone method to test the slump of various concrete samples to obtain multiple slumps.

[0073] In the embodiments of the present invention, the slump cone method is used to test the slump of various concrete samples, thereby obtaining multiple slumps. For example, for one concrete sample, the unset concrete sample is placed in three layers into a standard slump cone. Each layer is rammed 25 times. The cone is then lifted vertically, and the height difference after collapse is measured as the slump.

[0074] It should be noted that the slump of concrete reflects its workability, that is, the fluidity, plasticity and anti-segregation ability of the concrete mixture. The larger the slump, the better the fluidity of the concrete, and the more conducive it is to pumping and pouring. Medium slump is suitable for ordinary cast-in-place structures, such as beams and slabs, which can balance fluidity and stability. Low slump is mainly used for dry and hard concrete, such as prefabricated components and pavements. Too large a slump may lead to segregation and water seepage, affecting the strength and durability of the concrete after hardening. Too small a slump will make it difficult to vibrate and compact it, and it will easily form holes after the concrete solidifies.

[0075] S12. Perform hydration treatment on various concrete samples respectively to obtain multiple target concrete samples and multiple concrete test data.

[0076] The target concrete sample refers to the concrete sample after solidification.

[0077] Concrete test data refers to the ambient temperature, ambient humidity, cement activity and aggregate moisture content during the hydration reaction of concrete samples.

[0078] In an embodiment of the present invention, various unsolidified concrete samples are subjected to hydration reaction respectively, and the ambient temperature, ambient humidity, cement activity and aggregate moisture content during the hydration reaction are obtained as concrete test data, thereby obtaining multiple target concrete samples and multiple concrete test data.

[0079] S13. Perform compressive strength tests on each target concrete sample to obtain multiple critical loads.

[0080] In the embodiment of the present invention, pressure is applied to each target concrete sample at a constant rate to obtain a critical load when each target concrete sample is destroyed.

[0081] S14. Ratio processing is performed on each critical load and a preset pressure-bearing area to obtain multiple compressive strengths.

[0082] The bearing area refers to the bearing area of ​​the target concrete specimen.

[0083] In the embodiment of the present invention, the ratios of the critical loads to the preset pressure-bearing areas are calculated respectively to obtain a plurality of compressive strengths.

[0084] It should be noted that compressive strength reflects the ability of concrete to resist pressure damage and directly reflects its bearing capacity and durability. The higher the compressive strength, the better the performance of the concrete.

[0085] S15. Perform data preprocessing on each compressive strength, each concrete test data, and each slump to obtain a corresponding training concrete mix set.

[0086] Data preprocessing refers to cleaning and smoothing operations on data.

[0087] In the embodiment of the present invention, cleaning and smoothing operations are performed on each compressive strength, each concrete test data and each slump to generate a training concrete mix set consisting of multiple compressive strengths, multiple concrete test data and multiple slumps.

[0088] Step 202: Use the training concrete mix set to train the preset initial property prediction model to generate a target property prediction model.

[0089] Furthermore, the training concrete mix set includes a first training set and a second training set, and the initial attribute prediction model includes a first prediction model and a second prediction model. Step 202 includes the following sub-steps:

[0090] S21. Use the first training set to train the first prediction model to generate a first target prediction model.

[0091] The first training set refers to a data set consisting of multiple water-cement ratios, multiple sand ratios, multiple ambient temperatures, multiple ambient humidity, multiple cement activities, and multiple aggregate moisture contents.

[0092] In an embodiment of the present invention, a first prediction model is trained using a first training set to generate a first target prediction model. For example, A1: The first training set is input into the first prediction model, and first training prediction data (i.e., cement activity and aggregate moisture content predicted by the first prediction model) is output. A2: Based on a preset loss function (the loss function may be a mean square error function or a mean absolute error function), a first training loss function value for the first training set is calculated based on the first training prediction data. A3: When the first training loss function value is greater than or equal to a preset standard loss function value, the model parameters of the first prediction model are adjusted, and the execution jumps to A1-A3 until the first training loss function value is less than the preset standard loss function value. A4: When the first training loss function value is less than the preset standard loss function value, the first target prediction model is generated.

[0093] It should be noted that the first prediction model is a long short-term memory network. The structure of the first prediction model is as follows: the input layer contains 4 neurons for inputting water-cement ratio, sand ratio, ambient temperature, and ambient humidity; the first hidden layer contains 64 neurons, activated by the ReLU activation function; the second hidden layer contains 32 neurons, activated by the ReLU activation function; the output layer contains 2 neurons, which output cement activity and aggregate moisture content.

[0094] S22: Use the second training set to train the second prediction model to generate a second target prediction model.

[0095] The second training set refers to a data set consisting of multiple cement activities, multiple aggregate moisture contents, multiple compressive strengths, and multiple slumps.

[0096] In an embodiment of the present invention, a second prediction model is trained using a second training set to generate a second target prediction model. For example, step B1: Input the second training set into the second prediction model, and output second training prediction data (i.e., the compressive strength and slump predicted by the second prediction model). Step B2: Based on a preset loss function (the loss function may be a mean square error function or a mean absolute error function), calculate a second training loss function value for the second training set based on the second training prediction data. Step B3: When the second training loss function value is greater than or equal to a preset standard loss function value, adjust the model parameters of the second prediction model and jump to step B1-B3 until the second training loss function value is less than the preset standard loss function value. Step B4: When the second training loss function value is less than the preset standard loss function value, generate a second target prediction model.

[0097] It should be noted that the second prediction model is a long short-term memory network. The structure of the second prediction model is as follows: input layer: contains 2 neurons for inputting cement activity and aggregate moisture content; first hidden layer: contains 128 neurons, activated by ReLU activation function; second hidden layer: contains 64 neurons, activated by ReLU activation function; output layer: contains 2 neurons, used to output compressive strength and slump.

[0098] S23. Connect the first target prediction model and the second target prediction model in sequence to generate a target attribute prediction model.

[0099] In an embodiment of the present invention, a first target prediction model and a second target prediction model are used to construct a target attribute prediction model, wherein the target attribute prediction model includes the first target prediction model and the second target prediction model connected in sequence.

[0100] Step 203: construct an initial particle group using concrete mix ratios as particles, and input each concrete mix ratio in the initial particle group into the target attribute prediction model to obtain multiple compressive strengths and multiple slumps.

[0101] In an embodiment of the present invention, an initial particle group is randomly generated, wherein each particle in the initial particle group corresponds to a concrete mix ratio (the concrete mix ratio includes water-cement ratio, sand ratio, ambient temperature and ambient humidity), and a target attribute prediction model is used to generate the compressive strength and slump corresponding to each particle.

[0102] Step 204 : Based on a preset concrete evaluation function, the compressive strength evaluation index and the slump evaluation index corresponding to each concrete mix ratio are calculated according to each compressive strength and each slump.

[0103] Furthermore, the concrete evaluation function includes a compressive strength evaluation function and a slump evaluation function. Step 204 includes the following sub-steps:

[0104] S31. Input each concrete mix ratio into a compressive strength evaluation function to obtain a plurality of compressive strength reference values ​​and a plurality of compressive strength dynamic correction values.

[0105] In the embodiment of the present invention, the compressive strength reference value and the compressive strength dynamic correction value corresponding to each concrete mix ratio are calculated using a compressive strength evaluation function.

[0106] It should be noted that the compressive strength evaluation function is specifically:

[0107] ;

[0108] in, is the compressive strength benchmark value, is the dynamic correction value of compressive strength, For the quality of water, is the quality of cement, is the first fitting coefficient, is the second fitting coefficient, is the third fitting coefficient, is the sand ratio, is the sand ratio of the concrete sample with the highest compressive strength, is the cement activity weight coefficient, is the cement activity in time period t, is the mean cement activity, is the weight coefficient of aggregate moisture content, is the aggregate moisture content in time period t, is the average moisture content of aggregate, is the ambient temperature weight coefficient, is the deviation between the ambient temperature of the time period t and the ambient temperature of the concrete specimen with the highest compressive strength, is the ambient humidity weight coefficient, is the deviation of the ambient humidity at time period t from the ambient humidity of the concrete specimen with the highest compressive strength, t is the index of the time period, t n is the nth time period, n is the number of time periods, is the water-cement ratio.

[0109] It is worth mentioning that see Figure 3 As shown in the figure, the sand ratio in the compressive strength evaluation function is kept at the sand ratio of the concrete sample with the highest compressive strength. Ten concrete samples with different water-cement ratios are prepared, and the compressive strength of the samples is measured respectively. The relationship curve between compressive strength and water-cement ratio is fitted ( ), the first fitting coefficient and the second fitting coefficient are determined based on the least squares method. Keeping the water-cement ratio as the average water-cement ratio of the concrete sample, ten additional concrete samples with different sand ratios are prepared, and the compressive strength is tested respectively. The relationship curve between compressive strength and sand ratio is fitted ( ), and the third fitting coefficient is fitted based on the least squares method.

[0110] It should be noted that the compressive strength benchmark value reflects the dominant influence of water-cement ratio and sand ratio on compressive strength under fixed ambient temperature and humidity, and embodies the nonlinear relationship between compressive strength and water-cement ratio and sand ratio. The larger the water-cement ratio, the lower the compressive strength. This exponential law indicates that increasing the water-cement ratio leads to a decrease in compressive strength, and the water-cement ratio should be sufficient for hydration. At the same time, a high sand ratio may increase porosity in the concrete, reducing compressive strength, while a low sand ratio may lead to poor fluidity and reduce compressive strength. Therefore, both high and low sand ratios will lead to reduced compressive strength. It represents the penalty term on compressive strength when the sand ratio deviates from the sand ratio of the concrete sample with the highest compressive strength.

[0111] It should be noted that cement activity changes with time and environmental factors. High temperature accelerates hydration, but may lead to insufficient strength in the later stage. Low temperature inhibits hydration and delays strength development. The mean cement activity of concrete samples is used as the reference value of cement activity. When the cement activity in time period t is less than the mean cement activity, the cement activity is insufficient. <0, cement activity weakens the compressive strength. When the cement activity in time period t is greater than the mean cement activity, the cement activity exceeds the reference benchmark value and strengthens the compressive strength. The greater the cement activity, the more obvious the strengthening effect. Aggregate moisture content affects the water-cement ratio. The mean aggregate moisture content of the concrete sample is used as the reference benchmark value of aggregate moisture content. When the aggregate moisture content in time period t is less than the mean aggregate moisture content, the aggregate moisture content has not reached the benchmark value. <0, at which point the aggregate moisture content weakens the compressive strength. This weakening effect becomes weaker as the aggregate moisture content increases. When the aggregate moisture content in time period t is greater than the average aggregate moisture content, the aggregate moisture content exceeds the benchmark value, resulting in an increase in actual mixing water and slurry porosity, which in turn leads to a decrease in compressive strength. The greater the aggregate moisture content in time period t, the more the aggregate moisture content in time period t deviates from the benchmark value, and the more obvious the weakening effect on compressive strength. Temperature and humidity also have an optimal temperature and humidity that maximize the compressive strength. When the ambient temperature and humidity in time period t deviate from the optimal temperature and humidity, the compressive strength will be weakened, and the greater the deviation, the more obvious the weakening effect. The continuous influence of the entire stage from the start time to the end time is captured by integration. The dynamic correction term of compressive strength is proportional to the cement activity of the time period, inversely proportional to the difference between the aggregate moisture content and the reference benchmark value of the aggregate moisture content, inversely proportional to the difference between the ambient temperature and the optimal temperature, and inversely proportional to the difference between the ambient humidity and the optimal humidity. Cement activity directly determines the production and microstructure of hydration products and is the decisive factor for compressive strength. It has the highest weight coefficient. Temperature has a more significant effect on early strength and can directly affect the rate of hydration reaction. The weight coefficient is second. Excessive aggregate moisture content will increase porosity, thereby affecting compressive strength. It is not a direct effect and the effect is milder, with a lower weight coefficient. Humidity mainly affects surface maintenance and has a more indirect and slow effect on overall strength. Its weight coefficient is the lowest, and it is taken =0.4, =0.2, =0.3, =0.1, the relationship between compressive strength, water-cement ratio and sand ratio is shown in Table 1.

[0112] Table 1

[0113]

[0114] S32. Perform weighted calculations on the compressive strength, compressive strength reference value, and compressive strength dynamic correction value associated with each concrete mix ratio based on a preset compressive strength weight to obtain a compressive strength evaluation index corresponding to each concrete mix ratio.

[0115] In the embodiment of the present invention, the compressive strength, compressive strength reference value and compressive strength dynamic correction value associated with each concrete mix ratio are respectively input into a preset first weighting function to obtain the compressive strength evaluation index corresponding to each concrete mix ratio.

[0116] It should be noted that the first weighting function is specifically:

[0117] ;

[0118] in, is the compressive strength evaluation index, is the compressive strength, is the first compressive strength weight coefficient, is the second compressive strength weight coefficient, is the third compressive strength weight coefficient.

[0119] S33. Input each concrete mix ratio into a slump evaluation function to obtain a plurality of slump reference values ​​and a plurality of slump dynamic correction values.

[0120] In the embodiment of the present invention, the slump reference value and the slump dynamic correction value corresponding to each concrete mix ratio are calculated using a slump evaluation function.

[0121] It should be noted that the slump evaluation function is specifically:

[0122] ;

[0123] in, is the slump reference value, is the dynamic correction value of slump, is the temperature sensitivity coefficient, is the humidity sensitivity coefficient, is the deviation between the ambient temperature of the time period t and the ambient temperature of the concrete sample with the highest slump, is the maximum ambient temperature of the concrete specimen at all time periods, is the lowest ambient temperature of the concrete specimen at all time periods, is the ambient humidity of the concrete sample with the highest slump, is the mean water-cement ratio, is the maximum sand ratio of the concrete sample, is the minimum sand ratio of the concrete sample.

[0124] It is worth mentioning that see Figure 4 As shown in the figure, the larger the water-cement ratio, the higher the slurry fluidity, the larger the slump, and the larger the slump reference value. Used for normalization to ensure consistency with the water-cement ratio. It is used to measure the penalty when the sand ratio deviates from the optimal sand ratio. The maximum sand ratio of the concrete sample is used as the benchmark value. When the sand ratio is too high or too low, aggregate accumulation or insufficient slurry wrapping will occur, resulting in decreased fluidity and thus reduced slump. This penalty is amplified by the quadratic method.

[0125] It is worth mentioning that the dynamic correction value of slump quantifies the dynamic effects of water-cement ratio, cement activity, aggregate moisture content, ambient temperature and ambient humidity on slump. When <0, the cement activity is higher than the benchmark cement activity, the hydration reaction is accelerated, and the free water is consumed quickly, resulting in a decrease in slump. The stronger the cement activity, the faster the hydration reaction, the more it promotes the decrease in slump. The higher the cement activity, the smaller the slump. When the actual temperature is greater than 0, the aggregate moisture content is higher than the baseline value, the actual mixing water increases, the paste fluidity improves, and the slump increases, which has a positive correction effect. The higher the aggregate moisture content, the more obvious the positive correction effect. The optimal temperature represents the ideal ambient temperature for the concrete hydration reaction rate. When the actual temperature is higher, the hydration reaction rate is accelerated, resulting in faster slump loss. The greater the deviation, the more obvious the acceleration of the hydration reaction rate and the faster the slump loss. When the actual ambient temperature is lower, the hydration reaction rate slows down, but the low temperature increases the viscosity of the concrete, thereby reducing the slump. The lower the actual ambient temperature, the higher the viscosity of the concrete. The value represents the deviation of the ambient humidity from that of the concrete sample with the highest slump, and applies only to cases where the actual ambient humidity is lower than that of the concrete sample with the highest slump. When the actual ambient humidity is low, water evaporation accelerates, the concrete paste thickens, and the slump decreases. Furthermore, the lower the ambient humidity, the faster the water evaporation and the lower the slump. When the actual ambient humidity is high, water evaporation almost stops, but the free water content of the concrete paste does not increase, and the slump remains essentially unchanged. The dynamic slump correction value is inversely proportional to the cement activity, directly proportional to the aggregate moisture content, inversely proportional to the deviation of the actual ambient temperature from the optimal temperature, and inversely proportional to the actual ambient humidity. The relationship between compressive strength, water-cement ratio, and sand ratio is shown in Table 2.

[0126] Table 2

[0127]

[0128] It should be noted that the relationship between the hydration rate and the ambient temperature is described based on the Arrhenius equation (i.e., the Arrhenius equation). The specific formula is:

[0129] ;

[0130] Where, is the hydration rate at temperature T, is the prefactor, is the standard gas constant, is the hydration activation energy.

[0131] When the ambient temperature is equal to the optimal temperature, the linearization is performed and the expression of the temperature sensitivity coefficient can be obtained as follows:

[0132] ;

[0133] in, for The hydration rate, is the ambient temperature of the concrete sample with the highest slump, is the hydration rate.

[0134] S34. Perform weighted calculations on the slump, slump reference value, and slump dynamic correction value associated with each concrete mix ratio based on a preset slump weight to obtain a slump evaluation index corresponding to each concrete mix ratio.

[0135] In the embodiment of the present invention, the slump, slump reference value and slump dynamic correction value associated with each concrete mix ratio are respectively input into a preset second weighting function to obtain the slump evaluation index corresponding to each concrete mix ratio.

[0136] It should be noted that the second weighting function is specifically:

[0137] ;

[0138] in, is the slump evaluation index, is the first slump weight coefficient, is the second slump weight coefficient, is the third slump weight coefficient, is the slump.

[0139] Step 205: Optimize the initial particle group using various compressive strength evaluation indices and various slump evaluation indices to obtain an optimized concrete mix ratio.

[0140] Furthermore, step 205 includes the following sub-steps:

[0141] S41 , performing ratio processing on a preset reference compressive coefficient and each compressive strength evaluation index to obtain a plurality of first ratios.

[0142] S42: performing difference processing on the preset reference slump evaluation index and each slump evaluation index to obtain a plurality of first differences.

[0143] S43 , performing sum processing on the absolute value of each first difference and the corresponding first ratio respectively to obtain multiple fitness levels.

[0144] In the embodiment of the present invention, each compressive strength evaluation index and the corresponding slump evaluation index are respectively input into a preset fitness function to obtain multiple fitness levels.

[0145] It should be noted that the adaptation function is specifically:

[0146] ;

[0147] in, For fitness, It is the best slump evaluation index.

[0148] S44. Update the initial particle swarm according to each fitness to obtain a corresponding updated particle swarm.

[0149] In an embodiment of the present invention, the dynamic weight of each particle in the initial particle swarm is calculated based on the fitness and a preset dynamic weight function. The particle velocity of each particle is adjusted based on the dynamic weight and the particle velocity function to obtain a corresponding updated particle swarm.

[0150] It should be noted that the dynamic weight function is specifically:

[0151] ;

[0152] in, is the dynamic weight of the i-th particle, is the initial weight, and =0.9, is the current iteration number, is the maximum number of iterations, is the normalized deviation of the ith particle, is the fitness of the i-th particle, is the minimum fitness in the current iteration, is the maximum fitness in the current iteration, and i is the index of the particle.

[0153] It should be noted that the particle velocity function is specifically:

[0154] ;

[0155] in, is the velocity of the i-th particle at the d+1th iteration, is the velocity of the i-th particle at the d-th iteration, is the individual learning factor, To avoid the algorithm falling into the local optimal first random number, and, is the number of particles in the history iteration The smallest position, is the position of the i-th particle at the d-th iteration, is the position of all particles that minimizes fitness in historical iterations, is the social learning factor, To avoid the algorithm falling into the local optimum, a second random number is used.

[0156] S45. Determine whether the update times of the initial particle swarm is greater than or equal to a preset iteration threshold.

[0157] S46. If the number of updates is less than the iteration threshold, the updated particle group is used as a new initial particle group, and the process jumps to the step of inputting each concrete mix ratio in the initial particle group into the target property prediction model to obtain multiple compressive strengths and multiple slumps.

[0158] In the embodiment of the present invention, it is determined whether the update number of the initial particle swarm is greater than or equal to a preset iteration threshold. If the update number is less than the iteration threshold, the updated particle swarm is used as the new initial particle swarm, and the process jumps to step 203 to step 205.

[0159] S47. If the number of updates is greater than or equal to the iteration threshold, the concrete mix ratio corresponding to the minimum fitness value in the updated particle swarm is selected as the optimized concrete mix ratio.

[0160] In the embodiment of the present invention, when the number of updates is greater than or equal to the iteration threshold, the concrete mix ratio corresponding to the minimum fitness value in the updated particle swarm is selected as the optimized concrete mix ratio.

[0161] It should be noted that the current solution The closer to the optimal solution , indicating that the better the current optimization process is, the smaller the normalized deviation is. The dynamic weight combines the normalized deviation and the number of iterations. In the early stage of iteration, it tends to search globally to avoid falling into the local optimum. The number of iterations d is inversely proportional to the dynamic weight, and the normalized deviation is proportional to the dynamic weight. In the early stage of iteration, d<< , the dynamic weight is larger, the particle speed is more affected by the historical speed, and tends to be a global search. In the later stage of iteration, Approaching , Approaching 0, dynamic weight Mainly affected Influence, The larger it is, the worse the current solution is, which in turn affects the speed change to move away from the poor solution area. The smaller it is, the better the current solution is, and the speed change is slowed down to further search in the better solution area; the iteration is repeated until the upper limit of the iteration number is reached, and the water-cement ratio, sand ratio, ambient temperature and ambient humidity corresponding to the optimal position searched among all particles when the upper limit of the iteration number is reached are output as the global optimal solution, and the concrete mix ratio is optimized according to the global optimal solution.

[0162] In an embodiment of the present invention, a plurality of concrete samples are obtained and property tests are performed on the various concrete samples to obtain a corresponding training concrete mix set. The training concrete mix set is used to train a preset initial property prediction model to generate a target property prediction model. An initial particle group is constructed using the concrete mix as a particle. The initial particle group is optimized according to the target property prediction model and a preset concrete evaluation function to obtain an optimized concrete mix. This overcomes the technical problem that traditional concrete mix optimization relies mainly on manual experience and multiple experiments, requires a large amount of time and resources, is difficult to adapt to rapidly changing construction needs, and reduces the reliability of concrete generation. Compared with traditional concrete mix optimization methods, the present invention uses a training concrete mix set to train a preset initial property prediction model to generate a target property prediction model, quantifies the effects of material ratio and environmental factors on compressive strength and slump, and simultaneously optimizes the initial particle group according to the target property prediction model and a preset concrete evaluation function to obtain an optimized concrete mix, thereby improving the optimization efficiency of the concrete mix and the reliability of concrete generation.

[0163] See also Figure 5 , Figure 5 This is a structural block diagram of a concrete mix ratio optimization system provided in Example 3 of the present invention.

[0164] The present invention provides a concrete mix ratio optimization system, comprising:

[0165] The detection module 301 is used to obtain multiple concrete samples, perform property detection on each concrete sample, and obtain a corresponding training concrete mix set;

[0166] The training module 302 is used to train the preset initial property prediction model using the training concrete mix set to generate a target property prediction model;

[0167] The optimization module 303 is used to construct an initial particle swarm using the concrete mix ratio as particles, and optimize the initial particle swarm according to the target attribute prediction model and the preset concrete evaluation function to obtain an optimized concrete mix ratio.

[0168] Furthermore, the detection module 301 includes:

[0169] The first detection submodule is used to perform slump detection on various concrete samples using the slump cone method to obtain multiple slumps;

[0170] The hydration submodule is used to perform hydration treatment on various concrete samples respectively to obtain multiple target concrete samples and multiple concrete test data;

[0171] The second detection submodule is used to perform compressive strength testing on each target concrete sample to obtain multiple critical loads;

[0172] Ratio processing is performed on each critical load and the preset pressure-bearing area to obtain multiple compressive strengths;

[0173] The preprocessing submodule is used to preprocess the data of each compressive strength, each concrete test data and each slump to obtain the corresponding training concrete mix set.

[0174] Furthermore, the training concrete mix set includes a first training set and a second training set, the initial attribute prediction model includes a first prediction model and a second prediction model, and the training module 302 includes:

[0175] A first training submodule is used to train the first prediction model using the first training set to generate a first target prediction model;

[0176] A second training submodule is used to train the second prediction model using the second training set to generate a second target prediction model;

[0177] The connection submodule is used to connect the first target prediction model and the second target prediction model in sequence to generate a target attribute prediction model.

[0178] Furthermore, the optimization module 303 includes:

[0179] The first analysis submodule is used to input each concrete mix ratio in the initial particle group into the target property prediction model to obtain multiple compressive strengths and multiple slumps;

[0180] The second analysis submodule is used to calculate the compressive strength evaluation index and slump evaluation index corresponding to each concrete mix ratio according to each compressive strength and each slump based on a preset concrete evaluation function;

[0181] The optimization submodule is used to optimize the initial particle group using various compressive strength evaluation indices and various slump evaluation indices to obtain the optimized concrete mix ratio.

[0182] Furthermore, the concrete evaluation function includes a compressive strength evaluation function and a slump evaluation function. The second analysis submodule includes:

[0183] The first analysis unit is used to input each concrete mix ratio into the compressive strength evaluation function to obtain multiple compressive strength reference values ​​and multiple compressive strength dynamic correction values;

[0184] The second analysis unit is configured to perform weighted calculations on the compressive strength, compressive strength reference value, and compressive strength dynamic correction value associated with each concrete mix ratio based on a preset compressive strength weight, to obtain a compressive strength evaluation index corresponding to each concrete mix ratio;

[0185] The third analysis unit is used to input each concrete mix ratio into the slump evaluation function to obtain multiple slump reference values ​​and multiple slump dynamic correction values;

[0186] The fourth analysis unit is used to perform weighted calculations on the slump, slump reference value and slump dynamic correction value associated with each concrete mix ratio based on a preset slump weight, so as to obtain a slump evaluation index corresponding to each concrete mix ratio.

[0187] Furthermore, the submodules are optimized, including:

[0188] a ratio unit, configured to perform ratio processing on the preset reference compressive coefficient and each compressive strength evaluation index to obtain a plurality of first ratios;

[0189] A difference unit is used to perform difference processing on a preset reference slump evaluation index and each slump evaluation index to obtain a plurality of first differences;

[0190] a fitness unit, configured to perform sum processing on the absolute value of each first difference and the corresponding first ratio respectively to obtain a plurality of fitnesses;

[0191] An updating unit is used to update the initial particle swarm according to each fitness to obtain a corresponding updated particle swarm;

[0192] a fifth analyzing unit, configured to determine whether the number of updates of the initial particle swarm is greater than or equal to a preset iteration threshold;

[0193] If the number of updates is less than the iteration threshold, the updated particle group is used as the new initial particle group, and the execution jumps to the step of inputting each concrete mix ratio in the initial particle group into the target attribute prediction model to obtain multiple compressive strengths and multiple slumps;

[0194] If the number of updates is greater than or equal to the iteration threshold, the concrete mix ratio corresponding to the minimum fitness value in the updated particle swarm is selected as the optimized concrete mix ratio.

[0195] See also Figure 6 , Figure 6 This is a structural block diagram of a computer device provided in Example 4 of the present invention.

[0196] An electronic device according to an embodiment of the present invention includes: a memory 401 and a processor 402, wherein the memory 401 stores a computer program; when the computer program is executed by the processor 402, the processor 402 executes the concrete mix ratio optimization method according to any of the above embodiments.

[0197] Memory 401 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Memory 401 has storage space 403 for program code 413 for executing any of the method steps described above. For example, storage space 403 for program code may include individual program codes 413 for implementing various steps in the method described above. These program codes may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards, or floppy disks. The program codes may be compressed, for example, in a suitable format. When executed by a processing device, these codes cause the processing device to execute the various steps in the method described above. These program codes may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards, or floppy disks. The program codes may be compressed, for example, in a suitable format. When these codes are executed by a computing and processing device, the computing and processing device is caused to execute the various steps in the concrete mix ratio optimization method described above.

[0198] The fifth embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the concrete mix ratio optimization method according to any of the above embodiments is implemented.

[0199] Embodiment 6 of the present invention further provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the concrete mix ratio optimization method as described in any of the above embodiments.

[0200] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0201] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0202] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0203] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0204] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0205] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A concrete mix optimization method, characterized in that: include: Acquire multiple concrete samples, perform property detection on each of the concrete samples, and obtain corresponding training concrete mix sets; Using the training concrete mix set to train a preset initial property prediction model to generate a target property prediction model; An initial particle group is constructed using the concrete mix ratio as particles, and the initial particle group is optimized according to the target attribute prediction model and a preset concrete evaluation function to obtain an optimized concrete mix ratio.

2. The concrete mix optimization method according to claim 1, characterized in that: The step of performing attribute detection on various concrete samples to obtain corresponding training concrete mix sets includes: Using the slump cone method, the slump of each concrete sample is tested to obtain a plurality of slumps; Performing hydration treatment on the various concrete samples respectively to obtain a plurality of target concrete samples and a plurality of concrete test data; Performing compressive strength tests on each of the target concrete samples to obtain multiple critical loads; Ratio processing is performed on each of the critical loads and the preset pressure-bearing area to obtain multiple compressive strengths; Data preprocessing is performed on each of the compressive strengths, each of the concrete test data, and each of the slumps to obtain a corresponding training concrete mix set.

3. The concrete mix optimization method according to claim 1, characterized in that: The training concrete mix set includes a first training set and a second training set, the initial attribute prediction model includes a first prediction model and a second prediction model, and the step of using the training concrete mix set to train the preset initial attribute prediction model to generate a target attribute prediction model includes: Using the first training set to train the first prediction model to generate a first target prediction model; Using the second training set to train the second prediction model to generate a second target prediction model; The first target prediction model and the second target prediction model are sequentially connected to generate a target attribute prediction model.

4. The concrete mix optimization method according to claim 1, characterized in that: The step of optimizing the initial particle swarm according to the target attribute prediction model and a preset concrete evaluation function to obtain an optimized concrete mix ratio includes: Inputting each concrete mix ratio in the initial particle group into the target property prediction model to obtain multiple compressive strengths and multiple slumps; Based on a preset concrete evaluation function, respectively calculating the compressive strength evaluation index and the slump evaluation index corresponding to each concrete mix ratio according to each compressive strength and each slump; The initial particle group is optimized using the respective compressive strength evaluation indexes and the respective slump evaluation indexes to obtain an optimized concrete mix ratio.

5. The concrete mix optimization method according to claim 4, characterized in that: The concrete evaluation function includes a compressive strength evaluation function and a slump evaluation function. The step of calculating the compressive strength evaluation index and the slump evaluation index corresponding to each concrete mix ratio according to each compressive strength and each slump based on the preset concrete evaluation function includes: Inputting each of the concrete mix ratios into the compressive strength evaluation function to obtain a plurality of compressive strength reference values ​​and a plurality of compressive strength dynamic correction values; Based on the preset compressive strength weight, a weighted calculation is performed on the compressive strength, the compressive strength reference value, and the compressive strength dynamic correction value associated with each concrete mix ratio to obtain a compressive strength evaluation index corresponding to each concrete mix ratio; Inputting each of the concrete mix ratios into the slump evaluation function to obtain a plurality of slump reference values ​​and a plurality of slump dynamic correction values; Based on the preset slump weight, weighted calculation is performed on the slump, slump reference value and slump dynamic correction value associated with each concrete mix ratio to obtain the slump evaluation index corresponding to each concrete mix ratio.

6. The method for optimizing concrete mix ratio according to claim 4, characterized in that: The step of optimizing the initial particle group by using the respective compressive strength evaluation indexes and the respective slump evaluation indexes to obtain an optimized concrete mix ratio comprises: Performing ratio processing on the preset reference compressive coefficient and each of the compressive strength evaluation indexes to obtain a plurality of first ratios; Performing difference processing on a preset benchmark slump evaluation index and each of the slump evaluation indices to obtain a plurality of first differences; performing sum processing on the absolute value of each of the first differences and the corresponding first ratio respectively to obtain multiple fitnesses; Updating the initial particle swarm according to each fitness to obtain a corresponding updated particle swarm; Determining whether the update times of the initial particle swarm is greater than or equal to a preset iteration threshold; If the update number is less than the iteration threshold, the updated particle group is used as a new initial particle group, and the step of inputting each concrete mix ratio in the initial particle group into the target attribute prediction model to obtain multiple compressive strengths and multiple slumps is skipped and executed; If the update number is greater than or equal to the iteration threshold, the concrete mix ratio corresponding to the minimum fitness value in the updated particle swarm is selected as the optimized concrete mix ratio.

7. A concrete mix optimization system, characterized in that: include: A detection module is used to obtain multiple concrete samples, perform property detection on each of the concrete samples, and obtain a corresponding training concrete mix set; A training module, configured to train a preset initial property prediction model using the training concrete mix set to generate a target property prediction model; The optimization module is used to construct an initial particle group using the concrete mix ratio as particles, and optimize the initial particle group according to the target attribute prediction model and a preset concrete evaluation function to obtain an optimized concrete mix ratio.

8. An electronic device, characterized in that: The method comprises a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the concrete mix ratio optimization method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the concrete mix ratio optimization method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer is caused to execute the concrete mix ratio optimization method according to any one of claims 1 to 6.

Images