Super-sulfur cement marine concrete improvement method based on prediction model

Abstract

The invention discloses a prediction model-based super-sulfur cement marine concrete improvement method, which comprises the following steps of: performing index screening to obtain an optimal mortar ratio set, and constructing a marine environment performance prediction model to obtain an initial marine environment performance set; according to the marine environment performance threshold value and the marine environment performance objective function, selecting an optimal super-sulfur cement maritime work mortar ratio, determining a steel fiber parameter range, establishing a mortar working performance simulation model, determining the mortar working performance objective function, performing parameter optimization to obtain optimal steel fiber parameters, performing a corrosion inhibition test to determine an optimal operation index, and performing a corrosion inhibition test to obtain the super-sulfur cement maritime work mortar. And determining an improvement scheme of the to-be-optimized maritime work concrete according to the optimal operation index, the optimal steel fiber parameter and the optimal ratio of the ultra-sulfur cement maritime work mortar. According to the method, the improvement speed and effect of the super-sulfur cement marine concrete can be improved, and an efficient and reliable material design tool can be provided for the field of ocean engineering.

Description

Technical Field

[0001] This invention relates to the field of concrete improvement, and more particularly to a method for improving ultrasulfur cement marine concrete based on a predictive model. Background Technology

[0002] With the increasing demand for high-performance concrete materials in the marine engineering field, marine concrete, as an important component of marine engineering structures, directly affects the safety and service life of engineering structures in terms of its durability and workability. Developing a scientific and systematic method for improving marine concrete is of great significance for enhancing the performance of marine concrete and extending the service life of marine engineering structures.

[0003] Current methods for preventing steel fiber corrosion primarily employ high-pollution and high-energy-consumption electroplating processes. Treating the steel fiber surface with tannic acid could improve its corrosion resistance. Furthermore, the complexation reaction between silica and tannic acid, and the hydrogen bonds between silica and mussel protein in shellfish aggregates, could enhance the interfacial bond strength between steel fibers and seawater sand concrete, thereby improving interfacial adhesion. Simultaneously, using supersulfur cement instead of ordinary silicate cement could improve its resistance to sulfate attack. However, traditional marine concrete mix design methods rely heavily on experience. There is still no definitive approach to determining steel fiber parameters, steel fiber treatment procedures, and the proportions of supersulfur cement based on the marine application environment of seawater sand concrete. Therefore, this invention proposes an improvement method for ultrasulfur cement marine concrete based on a predictive model. The method involves determining an optimal mortar mix design through geotechnical tests, then constructing a marine environmental performance prediction model to screen for optimal mix designs suitable for different marine environments. Furthermore, it optimizes steel fiber parameters by establishing a mortar workability simulation model, determining the optimal steel fiber treatment operation, and finally identifying the improvement method for the marine concrete to be optimized. This method not only significantly improves the performance of marine concrete but also provides an efficient and reliable material design tool for the marine engineering field, possessing significant practical value and broad market prospects. Summary of the Invention

[0004] The purpose of this invention is to provide an improvement method for ultrasulfur cement marine concrete based on a predictive model.

[0005] To achieve the above objectives, the present invention is implemented according to the following technical solution:

[0006] This invention includes the following steps:

[0007] Based on the material dosage range, multiple sets of ultrasulfur cement marine mortar mix proportions are determined to obtain an initial mortar mix proportion set. Geotechnical tests are conducted based on the initial mortar mix proportion set to obtain geotechnical indices. Based on the geotechnical indices threshold, the initial mortar mix proportion set is screened to obtain an optimal mortar mix proportion set.

[0008] Based on the preferred mortar mix ratio set, ultrasulfur cement marine concrete specimens were prepared, and the marine environmental performance under different marine environments was measured to construct a marine environmental performance prediction model. The marine environment of the marine concrete to be optimized and the preferred mortar mix ratio set were input into the marine environmental performance prediction model to obtain an initial marine environmental performance set.

[0009] Based on the marine environmental performance threshold, the initial marine environmental performance set is screened to obtain the preferred marine environmental performance set, and the marine environmental performance objective function is determined to select the optimal ultrasulfur cement marine mortar mix ratio.

[0010] Determine the range of steel fiber parameters, establish a multi-scale model of steel fiber supersulfur cement marine mortar based on the optimal mix ratio of supersulfur cement marine mortar and the range of steel fiber parameters, and conduct orthogonal experiments to establish a simulation model of mortar workability.

[0011] Determine the objective function for the workability of the mortar, and optimize the parameters of the steel fiber within the specified range based on the objective function and the workability simulation model to obtain the optimal steel fiber parameters.

[0012] The optimal operating parameters are determined by conducting rust-inhibition tests based on the optimal steel fiber parameters. Based on the optimal operating parameters, the optimal steel fiber parameters, and the optimal ultra-sulfur cement marine mortar mix ratio, the improvement scheme for the marine concrete to be optimized is determined.

[0013] Furthermore, the method for obtaining the preferred mortar mix proportion set includes:

[0014] Based on the material composition range, water-cement ratio range, and sand-cement ratio range of supersulfur cement, multiple sets of supersulfur cement marine mortar mix proportions were generated to obtain an initial mortar mix proportion set;

[0015] Supersulfur cement marine mortar was prepared according to the initial mortar mix design, and geotechnical tests were conducted to obtain the workability and strength indicators of the supersulfur cement marine mortar. The workability indicators include fluidity, fluidity over time, slump, setting time, plastic viscosity, and yield stress. The strength indicators include the compressive strength and flexural strength of the mortar specimens.

[0016] Based on the performance index threshold and the strength index threshold, the optimal mortar mix set is obtained by screening the ultrasulfur cement marine mortar mixes that meet the performance and strength requirements in the initial mortar mix set.

[0017] Furthermore, the method for obtaining the initial marine environmental performance set includes:

[0018] Based on the preferred mortar mix design, supersulfur cement marine concrete specimens were prepared, and their marine environmental performance under different marine environments was determined. The marine environmental performance included chloride ion diffusion coefficient, electrical flux, expansion rate, strength loss rate, Friedel salt formation, abnormal aggregation of gypsum / ettringite, chloride ion penetration depth, steel reinforcement corrosion current density, maximum pore size, and total porosity.

[0019] The marine environment, the ultra-sulfur cement marine mortar mix ratio of the optimized mortar mix ratio set, and the marine environmental performance of the corresponding concrete specimens are combined to form a comprehensive performance set. The comprehensive performance set is randomly divided into a training set and a test set at a ratio of 6:4. The training set is used to train the marine environmental performance prediction model, and the test set is used to evaluate the performance of the marine environmental performance prediction model.

[0020] The marine environmental performance prediction model includes a dual-channel input layer, a feature fusion layer, a physical constraint hidden layer, and a multi-task output layer.

[0021] The dual-channel input layer employs a formula channel to address the intrinsic characteristics of the mortar material and an environmental channel to address external erosion conditions.

[0022] The feature fusion layer cross-maps the mortar material ratio and environmental parameters into virtual reaction parameters through a reaction potential energy field generator, outputting a 5-dimensional reaction potential energy field, the expression of which is:

[0023]

[0024] Where Φ is the reaction potential field, and α i X represents the component reactivity weight. i Y represents the component concentration or content, β represents the environmental degradation coefficient, and Y represents the component concentration or content. env For environmental parameters;

[0025] The physical constraint hiding layer includes a pore evolution constraint layer, an ion diffusion constraint layer, and a product formation constraint layer, which are predicted by numerical simulation.

[0026] The multi-task output layer verifies whether the prediction results meet the material laws through a discriminator network and outputs the marine environmental performance prediction results.

[0027] The marine environment of the marine concrete to be optimized and the set of preferred mortar mix proportions are input into the marine environment performance prediction model to obtain the initial marine environment performance set.

[0028] Furthermore, the method for selecting the optimal mix proportion of supersulfur cement marine mortar includes:

[0029] An optimal marine environmental performance set is obtained by screening an initial marine environmental performance set based on marine environmental performance thresholds.

[0030] Based on the marine environmental performance objective function, the optimal ultrasulfur cement marine mortar mix ratio corresponding to the maximum value of the marine environmental performance objective function is selected as the optimal ultrasulfur cement marine mortar mix ratio. The expression is as follows:

[0031]

[0032] Among them Perform total For the objective function of marine environmental performance, w i S is the weight coefficient of the i-th sub-objective. i The score for the i-th sub-objective includes chloride ion resistance score S1, sulfate stability score S2, interface durability score S3, reinforcement protection score S4, and pore structure optimization score S5, where k1, k2, and k3 are chloride ion attenuation coefficients, and Dis(Cl) = 1 / 2. - ) is the chloride ion diffusion coefficient, Q is the electric flux, and d(Cl) - ) represents the chloride ion penetration depth, ε represents the expansion rate, and ε lim The expansion rate is the safety threshold, and Δf is the strength loss rate. lim As the safety threshold for strength loss, G binary F is an indicator of abnormal aggregation of gypsum / ettringite; 0 is used for no aggregation and 1 for aggregation. salt F represents the amount of Friedel salt produced. ref For the optimal Friedel salt formation, k4 is the steel reinforcement attenuation coefficient, and I... corr For the current density of steel reinforcement corrosion, d max For the maximum aperture, d crit For the critical aperture, η total η is the total porosity. crit This is the critical porosity.

[0033] Furthermore, the method for establishing a mortar workability simulation model includes:

[0034] The range of steel fiber content is determined based on the mortar content of the marine concrete to be optimized. The range of steel fiber parameters consists of the range of steel fiber content and the range of steel fiber size. The range of steel fiber size includes the range of steel fiber length and the range of steel fiber thickness.

[0035] Multiple groups of steel fiber supersulfur cement marine mortar were prepared based on the optimal mix ratio of supersulfur cement and the parameter range of steel fibers. Geotechnical tests were conducted to obtain the test performance indicators. The corresponding finite element model of steel fiber supersulfur cement marine mortar was constructed and finite element analysis was performed to obtain the simulated performance indicators. The performance deviation between the test performance indicators and the simulated performance indicators was calculated. The finite element model parameters were adjusted until the performance deviation was minimized, and the optimal finite element model of steel fiber supersulfur cement marine mortar was output as the mortar performance simulation model.

[0036] Furthermore, the method for obtaining the optimal steel fiber parameters includes:

[0037] The objective function for the workability of steel fiber reinforced supersulfur cement marine mortar is determined based on its workability indicators. The expression is as follows:

[0038]

[0039] Where P work Let M be the objective function for the workability of the mortar. i The flowability indicators include initial flowability M1, 30-minute flowability M2, 60-minute flowability M3, and slump M4. ref,i Liquidity indicator M i The corresponding benchmark flow value, β f V is the fiber resistance coefficient. f For steel fiber volume fraction, l f / d f For steel fiber length l f and diameter d f The ratio, where λ² is the stability weight, μ is the plastic viscosity, and μ opt For ideal plastic viscosity, μ max The maximum plastic viscosity is given by τ0, where τ is the yield stress. opt For the ideal yield stress, τ max For the maximum yield stress, T i To measure the initial setting time, T f The final setting time was measured. For the initial freezing time of the target, For the target final setting time, σ i σ is the initial setting tolerance time. f This refers to the final setting tolerance time;

[0040] A set of steel fiber parameters is randomly selected as the initial optimal location for the seagull population. Seagull optimization is then performed within the range of these parameters. A chaotic mapping is applied to the initial location of the seagull population, expressed as:

[0041]

[0042] Where X i,0 Let Lb be the initial position of particle i, Lb be the lower bound of the parameter, and Ub be the upper bound of the parameter. Let be the chaotic mapping value of particle i in the k-th iteration;

[0043] Update the seagull's position based on the spiral angle, using the following expression:

[0044]

[0045] in For the position update of particle i in the (k+1)th iteration, γ1 and γ2 are direction weight factors, r1 and r2 are random numbers in the range [0,1], and θ∈[0,2π] is a random spiral angle. D represents the optimal particle position in the population. X A is the relative distance threshold between particles. k ζ is the adaptive driving force coefficient, K is the helix angle attenuation coefficient, and K is the maximum number of iterations.

[0046] Arithmetic crossover was used for population crossover, Gaussian perturbation for population mutation, and a dynamic dimensionality learning strategy was employed to adjust particle positions. Based on the mortar performance simulation model and corresponding steel fiber parameters, the performance indicators of steel fiber supersulfur cement marine mortar were obtained. The fitness of the particle population was calculated, and the top 30% of individuals with the highest fitness were directly introduced into the next generation. The expression is as follows:

[0047]

[0048] in The position of particle i is adjusted using a dynamic dimension learning strategy after k iterations, where η = 0.5(1 + r3) is the learning rate, and r3 is a random number in the range [0,1]. Let I be the position of two randomly selected distinct individuals in dimension j at the current iteration. j Let j be a unit indicator vector. A set of randomly selected dimension indices. Let be the population fitness at the ikth iteration of particle . Let be the objective function for the workability of the mortar. As a coordinating indicator for condensation time, This is the penalty coefficient;

[0049] Repeat the iteration until the maximum number of iterations is reached or the increase rate of the objective function of mortar workability is less than 0.1% after 5 consecutive iterations, then stop the iteration and output the optimal steel fiber parameters.

[0050] Furthermore, the method for determining the improvement scheme of the marine concrete to be optimized includes:

[0051] The optimal steel fiber size is selected, and the steel fiber is processed according to the first operating index. The strength of the operating index is continuously improved, and the surface dense layer parameters are measured until the change rate of the surface dense layer parameters is less than 1%. The corresponding index strength is then taken as the second operating index. The first operating index includes tannic acid concentration, operating temperature, and operating time. The dense layer parameters include surface roughness, Si content, C content, and O content.

[0052] The concentration of nano-silica solution is determined based on the concentration of tannic acid, and the optimal operating index is composed of the second operating index and the concentration of nano-silica solution.

[0053] The optimal steel fiber parameters are processed using the optimal operating index. Based on the optimal steel fiber parameters and the optimal ultrasulfur cement marine mortar mix ratio, the optimal steel fiber ultrasulfur cement marine mortar mix ratio is determined. The optimal steel fiber ultrasulfur cement marine mortar mix ratio is then used to replace the mortar mix ratio of the marine concrete to be optimized, thus improving the marine concrete to be optimized.

[0054] Furthermore, the supersulfur cement marine concrete comprises: coarse aggregate, fine aggregate, and steel fiber supersulfur cement marine mortar; the coarse aggregate and the fine aggregate are specifically marine shellfish; the steel fiber supersulfur cement marine mortar comprises natural seawater, sea sand, supersulfur cement, and steel fibers; the supersulfur cement comprises phosphogypsum, steel slag, water-quenched blast furnace slag, sintering machine head ash, calcium carbonate powder, and calcium hydroxide; the calcium carbonate powder has a mesh size of 200; and the steel fibers have an iron content greater than 99%.

[0055] The beneficial effects of this invention are:

[0056] This invention is an improvement method for ultrasulfur cement marine concrete based on a predictive model. Compared with existing technologies, this invention has the following technical advantages:

[0057] This invention, through steps such as screening and optimizing mortar mix proportions, predicting marine environmental performance, determining the objective function of marine environmental performance, and optimizing steel fiber parameters and operational indicators, can improve the data preprocessing capabilities and enhance the model adaptability of supersulfur cement marine concrete improvement, thereby improving the efficiency and accuracy of supersulfur cement marine concrete improvement. Optimizing the supersulfur cement marine concrete improvement method can improve the scientific nature of the formulation and the efficiency of the improvement work, enabling the improvement of supersulfur cement marine concrete, improving the utilization rate of marine resources, and providing an efficient and reliable material design tool for the marine engineering field. It has significant practical value and broad market prospects. Attached Figure Description

[0058] Figure 1 This is a flowchart illustrating the steps of an improved method for ultrasulfur cement marine concrete based on a predictive model, according to the present invention. Detailed Implementation

[0059] The present invention will be further described below through specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0060] The present invention provides a method for improving ultrasulfur cement marine concrete based on a predictive model, comprising the following steps:

[0061] like Figure 1 As shown, this embodiment includes the following steps:

[0062] Based on the material dosage range, multiple sets of ultrasulfur cement marine mortar mix proportions are determined to obtain an initial mortar mix proportion set. Geotechnical tests are conducted based on the initial mortar mix proportion set to obtain geotechnical indices. Based on the geotechnical indices threshold, the initial mortar mix proportion set is screened to obtain an optimal mortar mix proportion set.

[0063] Based on the preferred mortar mix ratio set, ultrasulfur cement marine concrete specimens were prepared, and the marine environmental performance under different marine environments was measured to construct a marine environmental performance prediction model. The marine environment of the marine concrete to be optimized and the preferred mortar mix ratio set were input into the marine environmental performance prediction model to obtain an initial marine environmental performance set.

[0064] Based on the marine environmental performance threshold, the initial marine environmental performance set is screened to obtain the preferred marine environmental performance set, and the marine environmental performance objective function is determined to select the optimal ultrasulfur cement marine mortar mix ratio.

[0065] Determine the range of steel fiber parameters, establish a multi-scale model of steel fiber supersulfur cement marine mortar based on the optimal mix ratio of supersulfur cement marine mortar and the range of steel fiber parameters, and conduct orthogonal experiments to establish a simulation model of mortar workability.

[0066] Determine the objective function for the workability of the mortar, and optimize the parameters of the steel fiber within the specified range based on the objective function and the workability simulation model to obtain the optimal steel fiber parameters.

[0067] The optimal operating parameters are determined by conducting rust-inhibition tests based on the optimal steel fiber parameters. Based on the optimal operating parameters, the optimal steel fiber parameters, and the optimal ultra-sulfur cement marine mortar mix ratio, the improvement scheme for the marine concrete to be optimized is determined.

[0068] In this embodiment, the method for obtaining the preferred mortar mix proportion set includes:

[0069] Based on the material composition range, water-cement ratio range, and sand-cement ratio range of supersulfur cement, multiple sets of supersulfur cement marine mortar mix proportions were generated to obtain an initial mortar mix proportion set;

[0070] Supersulfur cement marine mortar was prepared according to the initial mortar mix design, and geotechnical tests were conducted to obtain the workability and strength indicators of the supersulfur cement marine mortar. The workability indicators include fluidity, fluidity over time, slump, setting time, plastic viscosity, and yield stress. The strength indicators include the compressive strength and flexural strength of the mortar specimens.

[0071] Based on the performance index threshold and the strength index threshold, the optimal mortar mix set is obtained by screening the ultra-sulfur cement marine mortar mixes that meet the performance and strength requirements in the initial mortar mix set.

[0072] In actual evaluation, the material composition range of supersulfur cement includes 10-40% phosphogypsum, 10-40% steel slag, 15-25% water-quenched blast furnace slag, 5-10% sintering machine head ash, 1-5% calcium carbonate powder, and 1-5% calcium hydroxide. Among them, sintering machine head ash and calcium hydroxide are activators, and calcium carbonate powder is used as a filler to improve the density of the cementitious system. At the same time, it acts as a crystal nucleus to accelerate the hydration of the supersulfur cement system. The water-cement ratio is between 0.35 and 0.45, and the sand-cement ratio is between 1.0 and 2.0.

[0073] Based on the threshold values ​​of workability indicators (flowability ≥170mm, flowability after time ≥150mm, slump, initial setting time ≥90min, final setting time ≤300min, plastic viscosity ≤50Pa*s, yield stress ≤35Pa) and strength indicators (3-day compressive strength ≥18MPa, 3-day flexural strength ≥4.5MPa), the optimal mortar mix proportions of ultrasulfur cement marine mortar that meet the workability and strength requirements are selected to obtain the optimal mortar mix proportion set.

[0074] In this embodiment, the method for obtaining the initial marine environmental performance set includes:

[0075] Based on the preferred mortar mix design, supersulfur cement marine concrete specimens were prepared, and their marine environmental performance under different marine environments was determined. The marine environmental performance included chloride ion diffusion coefficient, electrical flux, expansion rate, strength loss rate, Friedel salt formation, abnormal aggregation of gypsum / ettringite, chloride ion penetration depth, steel reinforcement corrosion current density, maximum pore size, and total porosity.

[0076] The marine environment, the ultra-sulfur cement marine mortar mix ratio of the optimized mortar mix ratio set, and the marine environmental performance of the corresponding concrete specimens are combined to form a comprehensive performance set. The comprehensive performance set is randomly divided into a training set and a test set at a ratio of 6:4. The training set is used to train the marine environmental performance prediction model, and the test set is used to evaluate the performance of the marine environmental performance prediction model.

[0077] The marine environmental performance prediction model includes a dual-channel input layer, a feature fusion layer, a physical constraint hidden layer, and a multi-task output layer.

[0078] The dual-channel input layer employs a formula channel to address the intrinsic characteristics of the mortar material and an environmental channel to address external erosion conditions.

[0079] The feature fusion layer cross-maps the mortar material ratio and environmental parameters into virtual reaction parameters through a reaction potential energy field generator, outputting a 5-dimensional reaction potential energy field, the expression of which is:

[0080]

[0081] Where Φ is the reaction potential field, and α i X represents the component reactivity weight.i Y represents the component concentration or content, β represents the environmental degradation coefficient, and Y represents the component concentration or content. env For environmental parameters;

[0082] The physical constraint hiding layer includes a pore evolution constraint layer, an ion diffusion constraint layer, and a product formation constraint layer, which are predicted by numerical simulation.

[0083] The multi-task output layer verifies whether the prediction results meet the material laws through a discriminator network and outputs the marine environmental performance prediction results.

[0084] The marine environment and the set of preferred mortar mix proportions for the marine concrete to be optimized are input into the marine environment performance prediction model to obtain the initial marine environment performance set;

[0085] In the actual evaluation, the chloride ion diffusion coefficient was determined according to GB / T 50082-2009 using the RCM method; the charge passing through the concrete over 6 hours was measured using the electric flux method; the specimens were immersed in 5% Na2SO4 solution and the expansion rate and strength loss were measured periodically; Friedel salt, ettringite (AFt), and gypsum were detected using X-ray diffraction; and the pore structure and chloride ion concentration were observed using SEM-EDS. - / S elemental distribution and reaction product morphology; electrochemical impedance spectroscopy was used to monitor the passivation film state of steel bars;

[0086] In the dual-channel input layer, the formula channel treats the 8-dimensional intrinsic characteristics of mortar materials, while the environment channel treats the 4-dimensional external erosion conditions (Cl). - Concentration, SO4 2- (Concentration, temperature, pH value);

[0087] In the physical constraint hidden layer, the basic hydration model is learned using a public cement database. The pore evolution constraint layer is used to output pore size distribution parameters, the ion diffusion constraint layer is used to output the modified Fick's second law, and the product formation constraint layer is used to perform the equilibrium of ettringite and Friedel salt, and to penalize predictions that violate the laws of materials science during backpropagation.

[0088] In the multi-task output layer, the material properties verified by the discriminator network include: capillary porosity ≥ chloride ion diffusion coefficient × 10. 12 , Amount of ettringite formed ∝ (SO4) 2- Concentration input - residual gypsum);

[0089] The composite loss function is used to adjust the model's prediction accuracy; the expression is:

[0090]

[0091] Where Loss is the composite loss function, n is the marine environmental performance dimension, and y j,predFor the predicted value of marine environmental performance j, y j,real For the actual value of marine environmental performance j, For the model parameter θ to violate the potential field Φ violation The gradient is given by λ1 = 0.15, which is the penalty weight.

[0092] In this embodiment, the method for selecting the optimal mix proportion of supersulfur cement marine mortar includes:

[0093] An optimal marine environmental performance set is obtained by screening an initial marine environmental performance set based on marine environmental performance thresholds.

[0094] Based on the marine environmental performance objective function, the optimal ultrasulfur cement marine mortar mix ratio corresponding to the maximum value of the marine environmental performance objective function is selected as the optimal ultrasulfur cement marine mortar mix ratio. The expression is as follows:

[0095]

[0096]

[0097] Among them Perform total For the objective function of marine environmental performance, w i S is the weight coefficient of the i-th sub-objective. i The score for the i-th sub-objective includes chloride ion resistance score S1, sulfate stability score S2, interface durability score S3, reinforcement protection score S4, and pore structure optimization score S5, where k1, k2, and k3 are chloride ion attenuation coefficients, and Dis(Cl) = 1 / 2. - ) is the chloride ion diffusion coefficient, Q is the electric flux, and d(Cl) - ) represents the chloride ion penetration depth, ε represents the expansion rate, and ε lim The expansion rate is the safety threshold, and Δf is the strength loss rate. lim As the safety threshold for strength loss, G binary F is an indicator of abnormal aggregation of gypsum / ettringite; 0 is used for no aggregation and 1 for aggregation. salt F represents the amount of Friedel salt produced. ref For the optimal Friedel salt formation, k4 is the steel reinforcement attenuation coefficient, and I... corr For the current density of steel reinforcement corrosion, d max For the maximum aperture, d crit For the critical aperture, η total η is the total porosity. crit The critical porosity;

[0098] In actual assessments, the specific criteria for selecting the initial marine environmental performance set based on marine environmental performance thresholds are: chloride ion diffusion coefficient < 5 × 10⁻⁶. -12 m 2 / s, current flux <1000 coulombs, expansion rate <0.5%, strength loss rate <30%, steel corrosion current density <0.1μA / cm 2 Maximum pore size < 50 nm, total porosity < 15%;

[0099] In the objective function for marine environmental performance, the weight coefficients of each sub-objective are set as 0.35 / 0.20 / 0.15 / 0.20 / 0.10, the chloride ion attenuation coefficients are set as 0.01 / 0.01 / 0.1, the steel reinforcement attenuation coefficient is set as 0.5, the expansion rate safety threshold is set as 0.1%, the strength loss safety threshold is set as 20%, the optimal Friedel salt formation is set as 5%, the critical pore size is set as 30 nm, and the critical porosity is set as 10%. The optimal mix ratio of supersulfur cement marine mortar is obtained. The mass ratio of supersulfur cement: seawater: sea sand is 500:225:1500. The supersulfur cement components are 20% phosphogypsum, 5% steel slag, 60% water-quenched blast furnace slag, 5% sintering machine head ash, 5% calcium carbonate powder, and 5% calcium hydroxide.

[0100] In this embodiment, the method for establishing a mortar workability simulation model includes:

[0101] The range of steel fiber content is determined based on the mortar content of the marine concrete to be optimized. The range of steel fiber parameters consists of the range of steel fiber content and the range of steel fiber size. The range of steel fiber size includes the range of steel fiber length and the range of steel fiber thickness.

[0102] Based on the optimal mix ratio of supersulfur cement marine mortar and the parameter range of steel fiber, multiple groups of steel fiber supersulfur cement marine mortar were prepared for geotechnical tests to obtain experimental performance indicators. The corresponding finite element model of steel fiber supersulfur cement marine mortar was constructed for finite element analysis to obtain simulated performance indicators. The performance deviation between the experimental performance indicators and the simulated performance indicators was calculated. The finite element model parameters were adjusted until the performance deviation was minimized, and the optimal finite element model of steel fiber supersulfur cement marine mortar was output as the mortar performance simulation model.

[0103] In actual evaluation, based on the mortar content V of the marine concrete to be optimized... s =0.5 determines the fiber scaling factor Where V S,0 =0.6 is the standard mortar content. The product of the reference steel fiber content range of 0.5%-2.0% and the fiber scaling factor of 0.8465 (0.423%-1.693%) is taken as the steel fiber content range. The steel fiber parameter range is composed of the steel fiber size range (length range 15mm-25mm, thickness range 0.4mm-0.7mm) and the steel fiber content range.

[0104] Performance deviation = |Experimental performance index - Simulated performance index| / corresponding standard value of performance index. Take the average of all performance deviations as the performance deviation. Adjust the finite element model parameters until the performance deviation is minimized. Output the optimal finite element model of steel fiber supersulfur cement marine mortar as the mortar performance simulation model.

[0105] In this embodiment, the method for obtaining optimal steel fiber parameters includes:

[0106] The objective function for the workability of steel fiber reinforced supersulfur cement marine mortar is determined based on its workability indicators. The expression is as follows:

[0107]

[0108] Where P work Let M be the objective function for the workability of the mortar. i The flowability indicators include initial flowability M1, 30-minute flowability M2, 60-minute flowability M3, and slump M4. ref,i Liquidity indicator M i The corresponding benchmark flow value, β f V is the fiber resistance coefficient. f For steel fiber volume fraction, l f / d f For steel fiber length l f and diameter d f The ratio, where λ² is the stability weight, μ is the plastic viscosity, and μ opt For ideal plastic viscosity, μ max The maximum plastic viscosity is given by τ0, where τ is the yield stress. opt For the ideal yield stress, τ max For the maximum yield stress, T i To measure the initial setting time, T f The final setting time was measured. For the initial freezing time of the target, For the target final setting time, σ i σ is the initial setting tolerance time. f This refers to the final setting tolerance time;

[0109] A set of steel fiber parameters is randomly selected as the initial optimal location for the seagull population. Seagull optimization is then performed within the range of these parameters. A chaotic mapping is applied to the initial location of the seagull population, expressed as:

[0110]

[0111] Where X i,0 Let Lb be the initial position of particle i, Lb be the lower bound of the parameter, and Ub be the upper bound of the parameter. Let be the chaotic mapping value of particle i in the k-th iteration;

[0112] Update the seagull's position based on the spiral angle, using the following expression:

[0113]

[0114] in For the position update of particle i in the (k+1)th iteration, γ1 and γ2 are direction weight factors, r1 and r2 are random numbers in the range [0,1], and θ∈[0,2π] is a random spiral angle. D represents the optimal particle position in the population. X A is the relative distance threshold between particles. k ζ is the adaptive driving force coefficient, K is the helix angle attenuation coefficient, and K is the maximum number of iterations.

[0115] Arithmetic crossover was used for population crossover, Gaussian perturbation for population mutation, and a dynamic dimensionality learning strategy was employed to adjust particle positions. Based on the mortar performance simulation model and corresponding steel fiber parameters, the performance indicators of steel fiber supersulfur cement marine mortar were obtained. The fitness of the particle population was calculated, and the top 30% of individuals with the highest fitness were directly introduced into the next generation. The expression is as follows:

[0116]

[0117] in The position of particle i is adjusted using a dynamic dimension learning strategy after k iterations, where η = 0.5(1 + r3) is the learning rate, and r3 is a random number in the range [0,1]. Let I be the position of two randomly selected distinct individuals in dimension j at the current iteration. j Let j be a unit indicator vector. A set of randomly selected dimension indices. Let be the population fitness at the ikth iteration of particle . Let be the objective function for the workability of the mortar. As a coordinating indicator for condensation time, This is the penalty coefficient;

[0118] Repeat the iteration until the maximum number of iterations is reached or the increase rate of the objective function of mortar workability is less than 0.1% after 5 consecutive iterations, and then stop the iteration and output the optimal steel fiber parameters;

[0119] In practical evaluation, the fiber resistance coefficient β is taken in the objective function of mortar workability. f =0.08, stability weight λ2=0.6, ideal plastic viscosity μ opt =30+8V f (unit: Pa·s), maximum plastic viscosity μ max =65 (unit: Pa·s), ideal yield stress τopt =15+5(l f / d f ) 0.7 (unit: Pa), maximum yield stress τ max =35 (unit:Pa), target initial setting time Target final freezing time Initial setting tolerance time σ i =15+5(l f / d f (unit:min), final setting tolerance time σ f =25+5(l f / d f (unit:min);

[0120] In the seagull optimization, the maximum number of iterations K = 100, the helix angle decay coefficient ζ = 0.8, and the penalty coefficient are set to... Direction weighting factor γ1 = 1.8 / γ2 = 1.87, particle relative distance threshold D X =10%;

[0121] Seagull optimization is performed, and the mortar work performance index corresponding to the steel fiber parameters in each iteration is simulated according to the work performance simulation model. The objective function of mortar work performance is calculated. The iteration is repeated until the maximum number of iterations is reached or the increase rate of the objective function of mortar work performance is less than 0.1% in 5 consecutive iterations. The optimal steel fiber parameters (fiber content 1.5%, steel fiber length 20mm, steel fiber thickness 0.5mm) are output.

[0122] In this embodiment, the method for determining the improvement scheme of the marine concrete to be optimized includes:

[0123] The optimal steel fiber size is selected, and the steel fiber is processed according to the first operating index. The strength of the operating index is continuously improved, and the surface dense layer parameters are measured until the change rate of the surface dense layer parameters is less than 1%. The corresponding index strength is then taken as the second operating index. The first operating index includes tannic acid concentration, operating temperature, and operating time. The dense layer parameters include surface roughness, Si content, C content, and O content.

[0124] The concentration of nano-silica solution is determined based on the concentration of tannic acid, and the optimal operating index is composed of the second operating index and the concentration of nano-silica solution.

[0125] The optimal steel fiber parameters are processed using the optimal operating index. The optimal steel fiber ultrasulfur cement marine mortar mix ratio is determined based on the optimal steel fiber parameters and the optimal ultrasulfur cement marine mortar mix ratio. The optimal steel fiber ultrasulfur cement marine mortar mix ratio is used to replace the mortar mix ratio of the marine concrete to be optimized to improve the marine concrete to be optimized.

[0126] In the actual evaluation, a rust inhibition test was conducted. The mass ratio of tannic acid solution to steel fiber was 4:1 to ensure that a uniform and dense coating was formed on the surface of the steel fiber, providing better corrosion protection. When the average roughness reduction rate and Si / C / O content increase rate were less than 1%, the second operating parameters were determined to be: operating temperature 25℃, tannic acid solution concentration 150g / L, and operating time 24 hours. The concentration of nano silica solution was determined to be 300g / L, which was twice the concentration of tannic acid solution.

[0127] The optimal operating index is used to process the steel fiber corresponding to the optimal steel fiber parameter, and the optimal steel fiber supersulfur cement marine mortar mix ratio is used to replace the mortar mix ratio of the marine concrete to be optimized to improve the marine concrete to be optimized.

[0128] In this embodiment, the ultrasulfur cement marine concrete comprises: coarse aggregate, fine aggregate, and steel fiber ultrasulfur cement marine mortar; the coarse aggregate and the fine aggregate are specifically marine shellfish; the steel fiber ultrasulfur cement marine mortar comprises natural seawater, sea sand, ultrasulfur cement, and steel fibers; the ultrasulfur cement comprises phosphogypsum, steel slag, water-quenched blast furnace slag, sintering machine head ash, calcium carbonate powder, and calcium hydroxide; the calcium carbonate powder has a mesh size of 200 mesh; and the steel fibers have an iron content greater than 99%.

[0129] The preparation steps for supersulfur cement marine concrete include:

[0130] Steel fiber cleaning: Use anhydrous ethanol to clean the oil and dirt on the surface of the steel fiber to ensure that the surface of the steel fiber is clean;

[0131] Steel fiber rust prevention treatment: Soak the cleaned steel fibers in a 150g / L concentration tannic acid solution at 25℃ for 24 hours and then remove them to form a coating on the surface of the steel substrate;

[0132] Nano-silica surface reinforcement: After tannic acid treatment, steel fibers are immersed in a 300g / L nano-silica solution, which causes the nano-silica and tannic acid to complex on the surface of the steel fibers, thereby improving the interfacial strength between the steel fibers and seawater sand concrete.

[0133] Concrete mixing: Add supersulfur cement according to the components and dry mix for 1 minute. Add sea sand and natural seawater according to the proportion and mix for 2 minutes. Add steel fibers that have been treated with rust inhibitors and mix again for 2 minutes to ensure that they are fully mixed to obtain steel fiber concrete mortar.

[0134] Pouring and Curing: Freshly mixed steel fiber reinforced supersulfur cement marine mortar is evenly mixed with coarse / fine aggregate (marine shellfish) and poured into the mold. It is then vibrated for an appropriate time to remove air bubbles and achieve high density. The mixture is then allowed to stand naturally to set. After demolding, standard curing is performed to ensure the concrete's strength and durability develop to their optimal state. Following standard curing, marine environment simulation and tidal simulation are conducted to enhance erosion resistance and improve long-term durability. Specifically, the marine environment simulation involves immersion in artificial seawater (Cl...) for 8-28 days. - =19g / L, SO4 2- =2.7g / L), the tidal simulation is as follows: during days 29-90, soaking for 12 hours and drying for 12 hours alternately.

[0135] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for improving ultrasulfur cement marine concrete based on a predictive model, characterized in that, Includes the following steps: S1. Determine multiple sets of ultrasulfur cement marine mortar mix proportions according to the material dosage range to obtain an initial mortar mix proportion set. Conduct geotechnical tests based on the initial mortar mix proportion set to obtain geotechnical indices. Filter the initial mortar mix proportion set according to the geotechnical indices threshold to obtain an optimal mortar mix proportion set. S2. Prepare ultrasulfur cement marine concrete specimens according to the preferred mortar mix ratio set, determine the marine environmental performance under different marine environments, construct a marine environmental performance prediction model, and input the marine environment of the marine concrete to be optimized and the preferred mortar mix ratio set into the marine environmental performance prediction model to obtain an initial marine environmental performance set. S3. Based on the marine environmental performance threshold, the initial marine environmental performance set is screened to obtain the preferred marine environmental performance set, and the marine environmental performance objective function is determined to select the optimal ultrasulfur cement marine mortar mix ratio. S4. Determine the range of steel fiber parameters, establish a multi-scale model of steel fiber supersulfur cement marine mortar based on the optimal ultrasulfur cement marine mortar mix ratio and the range of steel fiber parameters, and conduct orthogonal experiments to establish a simulation model of mortar workability. S5. Determine the objective function for mortar workability, and optimize the parameters of the steel fiber within the range of the steel fiber parameters according to the objective function for mortar workability and the workability simulation model to obtain the optimal steel fiber parameters. S6. Conduct rust-inhibition tests based on the optimal steel fiber parameters to determine the optimal operating parameters. Based on the optimal operating parameters, the optimal steel fiber parameters, and the optimal ultra-sulfur cement marine mortar mix ratio, determine the improvement scheme for the marine concrete to be optimized.

2. The method for improving ultrasulfur cement marine concrete based on a prediction model according to claim 1, characterized in that, The method for obtaining the preferred mortar mix proportion set includes: Based on the material composition range, water-cement ratio range, and sand-cement ratio range of supersulfur cement, multiple sets of supersulfur cement marine mortar mix proportions were generated to obtain an initial mortar mix proportion set; Supersulfur cement marine mortar was prepared according to the initial mortar mix design, and geotechnical tests were conducted to obtain the workability and strength indicators of the supersulfur cement marine mortar. The workability indicators include fluidity, fluidity over time, slump, setting time, plastic viscosity, and yield stress. The strength indicators include the compressive strength and flexural strength of the mortar specimens. Based on the performance index threshold and the strength index threshold, the optimal mortar mix set is obtained by screening the ultrasulfur cement marine mortar mixes that meet the performance and strength requirements in the initial mortar mix set.

3. The method for improving ultrasulfur cement marine concrete based on a predictive model according to claim 1, characterized in that, The method for obtaining the initial marine environmental performance set includes: Based on the preferred mortar mix design, supersulfur cement marine concrete specimens were prepared, and their marine environmental performance under different marine environments was determined. The marine environmental performance included chloride ion diffusion coefficient, electrical flux, expansion rate, strength loss rate, Friedel salt formation, abnormal aggregation of gypsum / ettringite, chloride ion penetration depth, steel reinforcement corrosion current density, maximum pore size, and total porosity. The marine environment, the ultra-sulfur cement marine mortar mix ratio of the optimized mortar mix ratio set, and the marine environmental performance of the corresponding concrete specimens are combined to form a comprehensive performance set. The comprehensive performance set is randomly divided into a training set and a test set at a ratio of 6:

4. The training set is used to train the marine environmental performance prediction model, and the test set is used to evaluate the performance of the marine environmental performance prediction model. The marine environmental performance prediction model includes a dual-channel input layer, a feature fusion layer, a physical constraint hidden layer, and a multi-task output layer. The dual-channel input layer employs a formula channel to address the intrinsic characteristics of the mortar material and an environmental channel to address external erosion conditions. The feature fusion layer cross-maps the mortar material ratio and environmental parameters into virtual reaction parameters through a reaction potential energy field generator, outputting a 5-dimensional reaction potential energy field, the expression of which is: Where Φ is the reaction potential field, and α i X represents the component reactivity weight. i Y represents the component concentration or content, β represents the environmental degradation coefficient, and Y represents the component concentration or content. env For environmental parameters; The physical constraint hiding layer includes a pore evolution constraint layer, an ion diffusion constraint layer, and a product formation constraint layer, which are predicted by numerical simulation. The multi-task output layer verifies whether the prediction results meet the material laws through a discriminator network and outputs the marine environmental performance prediction results. The marine environment of the marine concrete to be optimized and the set of preferred mortar mix proportions are input into the marine environment performance prediction model to obtain the initial marine environment performance set.

4. The method for improving ultrasulfur cement marine concrete based on a predictive model according to claim 1, characterized in that, The method for selecting the optimal mix proportion of supersulfur cement marine mortar includes: An optimal marine environmental performance set is obtained by screening an initial marine environmental performance set based on marine environmental performance thresholds. Based on the marine environmental performance objective function, the optimal ultrasulfur cement marine mortar mix ratio corresponding to the maximum value of the marine environmental performance objective function is selected as the optimal ultrasulfur cement marine mortar mix ratio. The expression is as follows: Among them Perform total For the objective function of marine environmental performance, w i S is the weight coefficient of the i-th sub-objective. i The score for the i-th sub-objective includes chloride ion resistance score S1, sulfate stability score S2, interface durability score S3, reinforcement protection score S4, and pore structure optimization score S5, where k1, k2, and k3 are chloride ion attenuation coefficients, and Dis(Cl) = 1 / 2. - ) is the chloride ion diffusion coefficient, Q is the electric flux, and d(Cl) - ) represents the chloride ion penetration depth, ε represents the expansion rate, and ε lim The expansion rate is the safety threshold, and Δf is the strength loss rate. lim As the safety threshold for strength loss, G binary F is an indicator of abnormal aggregation of gypsum / ettringite; 0 is used for no aggregation and 1 for aggregation. salt F represents the amount of Friedel salt produced. ref For the optimal Friedel salt formation, k4 is the steel reinforcement attenuation coefficient, and I... corr For the current density of steel reinforcement corrosion, d max For the maximum aperture, d crit For the critical aperture, η total η is the total porosity. crit This is the critical porosity.

5. The method for improving ultrasulfur cement marine concrete based on a predictive model according to claim 1, characterized in that, The method for establishing a simulation model of mortar workability includes: The range of steel fiber content is determined based on the mortar content of the marine concrete to be optimized. The range of steel fiber parameters consists of the range of steel fiber content and the range of steel fiber size. The range of steel fiber size includes the range of steel fiber length and the range of steel fiber thickness. Multiple groups of steel fiber supersulfur cement marine mortar were prepared based on the optimal mix ratio of supersulfur cement and the parameter range of steel fibers. Geotechnical tests were conducted to obtain the test performance indicators. The corresponding finite element model of steel fiber supersulfur cement marine mortar was constructed and finite element analysis was performed to obtain the simulated performance indicators. The performance deviation between the test performance indicators and the simulated performance indicators was calculated. The finite element model parameters were adjusted until the performance deviation was minimized, and the optimal finite element model of steel fiber supersulfur cement marine mortar was output as the mortar performance simulation model.

6. The method for improving ultrasulfur cement marine concrete based on a prediction model according to claim 1, characterized in that, The method for obtaining optimal steel fiber parameters includes: The objective function for the workability of steel fiber reinforced supersulfur cement marine mortar is determined based on its workability indicators. The expression is as follows: Where P work Let M be the objective function for the workability of the mortar. i The flowability indicators include initial flowability M1, 30-minute flowability M2, 60-minute flowability M3, and slump M4. ref,i Liquidity indicator M i The corresponding benchmark flow value, β f V is the fiber resistance coefficient. f For steel fiber volume fraction, l f / d f For steel fiber length l f and diameter d f The ratio, where λ² is the stability weight, μ is the plastic viscosity, and μ opt For ideal plastic viscosity, μ max The maximum plastic viscosity is given by τ0, where τ is the yield stress. opt For the ideal yield stress, τ max For the maximum yield stress, T i To measure the initial setting time, T f The final setting time was measured. For the initial freezing time of the target, For the target final setting time, σ i σ is the initial setting tolerance time. f This refers to the final setting tolerance time; A set of steel fiber parameters is randomly selected as the initial optimal location for the seagull population. Seagull optimization is then performed within the range of these parameters. A chaotic mapping is applied to the initial location of the seagull population, expressed as: Where X i,0 Let Lb be the initial position of particle i, Lb be the lower bound of the parameter, and Ub be the upper bound of the parameter. Let be the chaotic mapping value of particle i in the k-th iteration; Update the seagull's position based on the spiral angle, using the following expression: in For the position update of particle i in the (k+1)th iteration, γ1 and γ2 are direction weight factors, r1 and r2 are random numbers in the range [0,1], and θ∈[0,2π] is a random spiral angle. D represents the optimal particle position in the population. X A is the relative distance threshold between particles. k ζ is the adaptive driving force coefficient, K is the helix angle attenuation coefficient, and K is the maximum number of iterations. Arithmetic crossover was used for population crossover, Gaussian perturbation for population mutation, and a dynamic dimensionality learning strategy was employed to adjust particle positions. Based on the mortar performance simulation model and corresponding steel fiber parameters, the performance indicators of steel fiber supersulfur cement marine mortar were obtained. The fitness of the particle population was calculated, and the top 30% of individuals with the highest fitness were directly introduced into the next generation. The expression is as follows: in The position of particle i is adjusted using a dynamic dimension learning strategy after k iterations, where η = 0.5(1 + r3) is the learning rate, and r3 is a random number in the range [0,1]. Let I be the position of two randomly selected distinct individuals in dimension j at the current iteration. j Let j be a unit indicator vector. A set of randomly selected dimension indices. Let be the population fitness at the ikth iteration of particle . Let be the objective function for the workability of the mortar. As a coordinating indicator for condensation time, This is the penalty coefficient; Repeat the iteration until the maximum number of iterations is reached or the increase rate of the objective function of mortar workability is less than 0.1% after 5 consecutive iterations, then stop the iteration and output the optimal steel fiber parameters.

7. The method for improving ultrasulfur cement marine concrete based on a predictive model according to claim 1, characterized in that, The method for determining the improvement scheme of the marine concrete to be optimized includes: The optimal steel fiber size is selected based on the optimal steel fiber parameters. The steel fiber is processed according to the first operating index, and the strength of the operating index is continuously improved while the surface dense layer parameters are measured. When the change rate of the surface dense layer parameters is less than 1%, the corresponding index strength is taken as the second operating index. The first operating index includes tannic acid concentration, operating temperature, and operating time. The dense layer parameters include surface roughness, Si content, C content, and O content. The concentration of nano-silica solution is determined based on the concentration of tannic acid, and the optimal operating index is composed of the second operating index and the concentration of nano-silica solution. The optimal steel fiber parameters are processed using the optimal operating index. Based on the optimal steel fiber parameters and the optimal ultrasulfur cement marine mortar mix ratio, the optimal steel fiber ultrasulfur cement marine mortar mix ratio is determined. The optimal steel fiber ultrasulfur cement marine mortar mix ratio is used to replace the mortar mix of the marine concrete to be optimized, thus improving the marine concrete to be optimized.

8. The method for improving ultrasulfur cement marine concrete based on a predictive model according to claim 1, characterized in that, The supersulfur cement marine concrete comprises: coarse aggregate, fine aggregate, and steel fiber supersulfur cement marine mortar; the coarse aggregate and fine aggregate are specifically marine shellfish; the steel fiber supersulfur cement marine mortar comprises natural seawater, sea sand, supersulfur cement, and steel fibers; the supersulfur cement comprises phosphogypsum, steel slag, water-quenched blast furnace slag, sintering machine head ash, calcium carbonate powder, and calcium hydroxide; the calcium carbonate powder has a mesh size of 200; and the steel fibers have an iron content greater than 99%.

Images