Method and system for preparing joint inorganic material of high-strength corrosion-resistant PC component

Abstract

The invention relates to the technical field of building materials, in particular to a method and system for preparing a joint inorganic material of a high-strength corrosion-resistant PC component, and the method comprises the steps: obtaining a basic data set, extracting material ratio data, calculating a component offset coefficient sequence and a thermosensitive gradient sequence, and generating a first complete optimization model through primary parameter optimization; identifying a target parameter pair according to the target performance index value, and performing secondary parameter optimization to generate a second complete optimization model; and finally, target parameters and corresponding matching temperature sequences are identified to complete material preparation. The invention further provides a corresponding system, electronic equipment and a storage medium. The material ratio and reaction conditions can be efficiently optimized, the mechanical property, corrosion resistance and long-term stability of the PC component joint material are remarkably improved, and meanwhile the test cost is reduced.

Description

Technical Field

[0001] This invention belongs to the field of building materials technology, specifically a method and system for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components. Background Technology

[0002] With the rapid development of industrialized construction and prefabricated structures, precast concrete (PC) components are increasingly widely used in modern building engineering. The performance of joint materials for high-strength and corrosion-resistant PC components directly affects the durability, safety, and service life of the overall structure. However, existing joint materials still have certain shortcomings in terms of strength, corrosion resistance, construction performance, and environmental friendliness, making it difficult to fully meet the needs of use under complex working conditions.

[0003] Currently, research on joint materials for PC components has made some progress, but many problems still need to be solved. For example, a heat-setting shaped joint material, disclosed in CN112805101B on September 20, 2022, proposes a shaped joint material with gibbsite, clay, and graphite as the main raw materials, and improves its heat sealing and flame retardant properties by adding organic additives. However, this technical solution mainly focuses on the fire resistance of the material, and has not deeply optimized the mechanical properties, corrosion resistance, and bonding strength with the PC component interface of the joint material. In addition, the use of organic additives in this material may affect its long-term durability, especially in humid or chemically corrosive environments, which may lead to a decline in material performance.

[0004] Meanwhile, a method for preparing an interface joint material based on industrial solid waste, with publication number CN114773011B, was published on January 3, 2023. This method proposes a way to prepare an interface joint material using industrial solid waste, which has advantages such as pumpability, rapid hardening, early strength, and micro-expansion. This technical solution demonstrates outstanding performance in terms of environmental protection and resource utilization, but its corrosion resistance and long-term stability still need improvement. Especially in high-salt, high-humidity, or acid-alkali corrosive environments, the material's corrosion resistance may be insufficient, affecting its application in harsh environments. Furthermore, the preparation process of this material has high requirements for the particle size distribution and blending ratio of the raw materials, which may increase the difficulty of quality control during the production process.

[0005] The aforementioned problems indicate that existing PC component joint materials still have certain shortcomings in terms of strength, corrosion resistance, long-term stability, and workability. Specifically, existing joint materials are insufficient to meet high-strength requirements in terms of mechanical properties, cannot adapt to complex environmental conditions in terms of corrosion resistance, and their workability and environmental friendliness have also fallen short of ideal levels. These problems not only limit the widespread application of PC components in high-performance buildings but also pose a potential threat to the safety and durability of projects.

[0006] In recent years, ultra-high performance concrete (UHPC) has received widespread attention and application in various fields due to its excellent mechanical properties, durability, and versatility. UHPC possesses extremely high compressive and tensile strength, making it suitable for bridge projects with longer spans and larger clearances; its high impermeability and corrosion resistance make it suitable for special projects such as underground utility tunnels and pipelines; furthermore, UHPC exhibits good toughness, meeting the requirements of projects with high impact resistance. For example, in 2010, Shao Xudong et al. from Hunan University proposed a "lightweight composite bridge deck structure of orthotropic steel plate-thin-layer UHPC," successfully overcoming the challenges of fatigue cracking of ordinary orthotropic steel plates and the easy damage of asphalt pavement on steel bridge decks, and further optimizing the special ultra-high toughness concrete (STC) for steel bridge decks. This innovation provides a more reliable solution for bridge engineering.

[0007] In the field of steel structure corrosion protection, UHPC also demonstrates significant advantages. The Haixin Bridge in Guangzhou, completed in 2021, uses an external UHPC protective layer for its arch feet and lower steel arch ribs, providing long-term and reliable corrosion protection for the steel structure. This application example provides a new approach for projects where steel structure corrosion coating updates are difficult, and also verifies the high durability of UHPC in complex environments. Furthermore, due to its lightweight, excellent corrosion resistance, and low energy consumption, UHPC aligns with energy conservation and environmental protection principles, and is therefore frequently used in decorative components such as exterior wall panels for buildings.

[0008] Despite the superior performance of UHPC in multiple fields, its high cost and technical barriers limit its widespread application in joint materials. Therefore, developing an inorganic joint material that combines high strength, corrosion resistance, and construction adaptability has become a pressing technical challenge. This invention aims to provide a method and system for preparing inorganic joint materials for high-strength, corrosion-resistant PC components. By optimizing the material formulation and preparation process, the mechanical properties, corrosion resistance, and construction adaptability of the joint material are significantly improved, while also considering environmental friendliness and economy, thereby meeting the demands of modern construction engineering for high-performance joint materials. Summary of the Invention

[0009] This invention provides a method and system for preparing inorganic materials for the joints of high-strength, corrosion-resistant PC components. Its main objective is to address the shortcomings of existing PC component joint materials in terms of mechanical properties, corrosion resistance, long-term stability, and construction adaptability. To achieve the above objective, this invention provides a method for preparing inorganic materials for the joints of high-strength, corrosion-resistant PC components, comprising: acquiring a basic dataset of PC component joint materials; sequentially extracting material proportioning data from the basic dataset; extracting component proportion sequences, reaction temperature sequences, and performance index values ​​from the material proportioning data; calculating component offset coefficient sequences and thermosensitive gradient sequences based on the component proportion sequences and reaction temperature sequences, respectively; performing a first-order parameter optimization on the material proportioning data based on the component offset coefficient sequences and thermosensitive gradient sequences to obtain an optimized parameter set; constructing a performance mapping matrix of the optimized parameter set based on the performance index values ​​to obtain an iterative optimization model; and determining whether the material proportioning data in the basic dataset has been optimized. After extraction is complete; if the material proportion data in the basic dataset is not completely extracted, return to the steps of sequentially extracting the material proportion data in the basic dataset; if the material proportion data in the basic dataset is completely extracted, use the iterative optimization model as the first fully optimized model; obtain the target performance index value, and identify the target parameter pair in the first fully optimized model based on the target performance index value; perform secondary parameter optimization on the first fully optimized model based on the target parameter pair to obtain the second fully optimized model; identify the target parameter on the second fully optimized model based on the target performance index value, identify the proportioning temperature sequence corresponding to the target parameter, and prepare the inorganic material for the joint of the high-strength corrosion-resistant PC component based on the proportioning temperature sequence.

[0010] Further, the calculation of the component offset coefficient sequence and the thermosensitive gradient sequence based on the component ratio sequence and the reaction temperature sequence respectively includes: sequentially extracting the component ratios from the component ratio sequence, and calculating the component offset coefficients based on the component ratios using the following formula to obtain the component offset coefficient sequence: α ij =k j ·p ij , where α ij k represents the offset coefficient of the j-th component in the i-th material proportion data. j p is the proportional weighting factor for the j-th component. ij Let β be the proportion of the j-th component in the i-th material ratio data; extract the reaction temperatures sequentially from the reaction temperature sequence, and calculate the thermosensitive gradient using the following formula based on the reaction temperatures to obtain the thermosensitive gradient sequence: β ij =c j ·(t ij -b0), where β ij c represents the thermodynamic gradient of the j-th reaction temperature in the i-th material ratio data. jLet t be the thermosensitive scaling factor for the j-th reaction temperature. ij Let be the j-th reaction temperature in the i-th material ratio data, and b0 be the reference thermosensitive intercept.

[0011] Further, before calculating the component offset coefficient sequence and the thermosensitive gradient sequence based on the component ratio sequence and the reaction temperature sequence respectively, the method further includes: obtaining the ratio threshold and ratio deviation threshold of the j-th component; and calculating the ratio weighting factor using the following formula based on the ratio threshold and ratio deviation threshold of the j-th component: where,

[0012]

[0013] Where, p j D represents the proportion threshold of the j-th component. j The proportional deviation threshold of the j-th component is represented; the thermosensitive threshold and temperature deviation threshold of the j-th reaction temperature are obtained; based on the thermosensitive threshold and temperature deviation threshold of the j-th reaction temperature, the thermosensitive proportional factor is calculated using the following formula:

[0014]

[0015] Among them, T j ΔT represents the thermistor threshold temperature of the j-th reaction. j This represents the temperature deviation threshold for the j-th reaction temperature.

[0016] Further, the step of performing parameter optimization on the material proportioning data based on the component offset coefficient sequence and the thermosensitive gradient sequence to obtain an optimized parameter set includes: obtaining an initial proportioning model; sequentially extracting component offset coefficients and thermosensitive gradients from the component offset coefficient sequence and the thermosensitive gradient sequence, respectively; determining optimization sites on the initial proportioning model based on the component offset coefficients; identifying the spatial distribution direction of the initial proportioning model and calculating the optimization direction angle based on the spatial distribution direction and the thermosensitive gradient; randomly selecting an optimization auxiliary angle within a pre-constructed optimization angle domain, wherein the optimization angle domain is (0, π); constructing an iterative optimization branch at the optimization site based on the optimization direction angle and the optimization auxiliary angle; determining whether the component offset coefficients and thermosensitive gradients have been completely extracted; if the component offset coefficients and thermosensitive gradients have not been completely extracted, updating the initial proportioning model using the iterative optimization branch and returning to the steps of sequentially extracting component offset coefficients and thermosensitive gradients from the component offset coefficient sequence and the thermosensitive gradient sequence, respectively; if the component offset coefficients and thermosensitive gradients have been completely extracted, the optimized parameter set is obtained.

[0017] Further, the step of identifying target parameter pairs in the first fully optimized model based on the target performance index value includes: sequentially identifying optimization end nodes in the first fully optimized model, wherein the optimization end node refers to the optimization node directly connected to the performance mapping matrix; identifying the performance mapping set on the optimization end node, and identifying the corresponding end interval of the performance mapping set; determining whether the target performance index value belongs to the corresponding end interval; if the target performance index value does not belong to the corresponding end interval, then returning to the step of sequentially identifying optimization end nodes in the first fully optimized model; if the target performance index value belongs to the corresponding end interval, then using the performance mapping set as the first mapping set to obtain multiple sets of first mapping sets; in the first... In a fully optimized model, adjacent mapping pairs with the same angle are identified sequentially. An adjacent mapping pair with the same angle refers to two performance mappings with the same optimization direction angle and adjacent optimization endpoints. The adjacent intervals of the adjacent mapping pairs with the same angle are identified. It is determined whether the target performance index value belongs to the adjacent interval with the same angle. If the target performance index value does not belong to the adjacent interval with the same angle, the process returns to the steps of sequentially identifying adjacent mapping pairs with the same angle in the first fully optimized model. If the target performance index value belongs to the adjacent interval with the same angle, the adjacent mapping pairs with the same angle are used as second mapping pairs, resulting in multiple sets of second mapping pairs. Based on the multiple sets of first mapping pairs and the multiple sets of second mapping pairs, the target parameter pairs are identified in the first fully optimized model using the target performance index value.

[0018] Further, the step of identifying target parameter pairs in the first fully optimized model using the target performance index value based on multiple sets of first mappings and multiple sets of second mappings includes: sequentially extracting first mapping sets from the multiple sets of first mappings, extracting first parameter pairs of the target performance index value from the first mapping sets, identifying first neighboring index pairs of the first parameter pairs, and obtaining multiple sets of first neighboring index pairs; sequentially extracting second mapping pairs from the multiple sets of second mappings, identifying second neighboring index pairs of the second mapping pairs, and obtaining multiple sets of second neighboring index pairs; and calculating the first neighboring distance set and the second neighboring distance set in the multiple sets of first neighboring index pairs and the multiple sets of second neighboring index pairs respectively using a pre-constructed distance formula, wherein the distance formula is as follows:

[0019] d p =|x p1 -x p2 |

[0020] d q =|y q1 -y q2 |

[0021] Where, d p Let x represent the first neighbor distance of the first neighbor index pair in the p-th group.p1 x represents the value of the first neighboring index in the first neighboring index pair of the p-th group. p2 d represents the value of the second neighboring index in the first neighboring index pair of group p; q y represents the second neighbor distance of the second neighbor index pair in the q-th group. q1 y represents the value of the first neighboring index in the second neighboring index pair of the q-th group. q2 The value of the second neighbor index in the second neighbor index pair of the qth group is represented by ||; the absolute value symbol is represented by ||; the minimum neighbor distance is identified in the first neighbor distance set and the second neighbor distance set; the target neighbor index pair corresponding to the minimum neighbor distance is identified, and the target parameter pair corresponding to the target neighbor index pair is identified.

[0022] Further, the step of performing secondary parameter optimization on the first fully optimized model based on the target parameters to obtain a second fully optimized model includes: obtaining the parameter relationship attributes and overlapping optimization paths of the target parameter pair; if the parameter relationship attributes are preset co-position end parameters, then conducting joint material preparation experiments based on the overlapping optimization paths and preset temperature gradients to obtain a set of test performance indicators; sequentially extracting test performance indicator values ​​from the set of test performance indicators, identifying the reaction temperature of the test performance indicator values, and calculating the test thermosensitive gradient of the reaction temperature; identifying the target end nodes of the target parameter pair, and constructing end optimization branches at the target end nodes based on the test thermosensitive gradients; constructing test mappings based on the test performance indicator values, and linking the test mappings. The second fully optimized model is obtained by reaching the terminal optimization branch. If the parameter relationship attribute is a preset angularly adjacent parameter, the joint material preparation test is carried out according to the overlapping optimization path and the preset component ratio gradient to obtain a set of test performance indicators. The test performance indicator values ​​are extracted sequentially from the set of test performance indicators, the component ratio of the test performance indicator values ​​is identified, and the test offset coefficient of the component ratio is calculated. The parent node and target optimization direction angle of the target parameter pair are identified, and the target terminal node is calculated according to the parent node and the test offset coefficient. The terminal optimization branch is constructed according to the target terminal node and the target optimization direction angle. The test mapping is constructed according to the test performance indicator values, and the test mapping is linked to the terminal optimization branch to obtain the second fully optimized model.

[0023] Further, the step of identifying the target parameter on the second fully optimized model based on the target performance index value includes: sequentially extracting test maps in the second fully optimized model to obtain the test performance index value corresponding to the test map; calculating the performance difference value between the test performance index value and the target performance index value to obtain a performance difference value set; extracting the minimum difference value from the performance difference value set and identifying the target parameter corresponding to the minimum difference value.

[0024] Further, the identification of the proportioning temperature sequence corresponding to the target parameter includes: extracting the target offset coefficient sequence and the target thermosensitive gradient sequence of the target parameter in the second fully optimized model; identifying the target component ratio sequence and the target reaction temperature sequence corresponding to the target offset coefficient sequence and the target thermosensitive gradient sequence; and constructing the proportioning temperature sequence based on the target component ratio sequence and the target reaction temperature sequence.

[0025] To achieve the above objectives, the present invention also provides a system for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components, comprising: a first fully optimized model generation module, used to acquire a basic dataset of PC component joint materials; sequentially extracting material proportion data from the basic dataset, extracting component ratio sequences, reaction temperature sequences, and performance index values ​​from the material proportion data, calculating component offset coefficient sequences and thermosensitive gradient sequences based on the component ratio sequences and reaction temperature sequences respectively; performing parameter optimization on the material proportion data based on the component offset coefficient sequences and thermosensitive gradient sequences to obtain an optimized parameter set, constructing a performance mapping matrix of the optimized parameter set based on the performance index values ​​to obtain an iterative optimization model; determining whether the material proportion data in the basic dataset has been completely extracted; if the material proportion data in the basic dataset has been completely extracted... If the data extraction is incomplete, return to the steps described above for sequentially extracting material proportion data from the basic dataset; if the material proportion data in the basic dataset has been extracted, then the iterative optimization model is used as the first fully optimized model; the target parameter pair identification module is used to obtain the target performance index value and identify the target parameter pair in the first fully optimized model based on the target performance index value; the second fully optimized model generation module is used to perform secondary parameter optimization on the first fully optimized model based on the target parameter pair to obtain the second fully optimized model; the high-strength corrosion-resistant PC component joint inorganic material preparation module is used to identify the target parameter on the second fully optimized model based on the target performance index value, identify the proportioning temperature sequence corresponding to the target parameter, and prepare the high-strength corrosion-resistant PC component joint inorganic material based on the proportioning temperature sequence.

[0026] To address the aforementioned problems, the present invention also provides an electronic device, comprising: a memory storing at least one instruction; and a processor executing the instruction stored in the memory to implement the above-described method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components.

[0027] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the above-described method for preparing inorganic materials for high-strength and corrosion-resistant PC component joints.

[0028] This invention constructs a first fully optimized model using existing basic datasets, significantly reducing the workload of experiments. However, since the performance index values ​​in the first fully optimized model may have significant errors compared to the target performance index values, further optimization of the proportioning and reaction condition parameters of the inorganic material for PC component joints is required. Firstly, target parameter pairs can be directly identified in the first fully optimized model based on the target performance index values. Then, secondary parameter optimization is performed on the first fully optimized model based on the target parameter pairs to obtain a second fully optimized model. This narrows the experimental scope to the process parameter range of the target parameter pairs. Finally, target parameters are identified on the second fully optimized model based on the target performance index values, and the corresponding proportioning temperature sequence is identified. This allows for the preparation of high-strength, corrosion-resistant inorganic materials for PC component joints based on the proportioning temperature sequence. Therefore, this invention can solve the shortcomings of current inorganic materials for PC component joints in terms of mechanical properties, corrosion resistance, long-term stability, and construction adaptability. Attached Figure Description

[0029] Figure 1 This is a schematic flowchart illustrating the method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components according to an embodiment of the present invention.

[0030] Figure 2 The structural framework diagram of the first fully optimized model generation process provided in the embodiments of the present invention is shown.

[0031] Figure 3 This is a schematic diagram of the module composition of the inorganic material preparation system for high-strength and corrosion-resistant PC component joints provided in an embodiment of the present invention. Detailed Implementation

[0032] This invention provides a method and system for preparing inorganic materials for the joints of high-strength, corrosion-resistant PC components. The core of this method lies in optimizing material ratios and reaction condition parameters through a data-driven approach, thereby improving the performance of the inorganic materials for PC component joints in terms of mechanical properties, corrosion resistance, long-term stability, and construction adaptability. The following is a detailed explanation... Figures 1 to 3 The specific embodiments of the present invention will be described in detail below.

[0033] Firstly, the first step of this invention is to obtain a basic dataset of PC component joint materials. This basic dataset includes historical experimental data, literature data, and material performance data from actual engineering applications (long-term stable data such as the UHPC protective layer parameters of Guangzhou Haixinqiao). This data is organized into material proportioning data, with each set containing a component ratio sequence, a reaction temperature sequence, and performance index values. The component ratio sequence describes the proportion of different components in the material, while the reaction temperature sequence records the reaction temperature of each component during the preparation process. Performance index values ​​are used to characterize the comprehensive performance of the material, such as compressive strength, flexural strength, corrosion resistance, and durability. Figure 2 As shown, the first fully optimized model generation module is responsible for processing this data and transforming it into the input form required for subsequent optimization processes.

[0034] To further analyze the material proportioning data, it is necessary to calculate the component offset coefficient sequence and the thermosensitive gradient sequence separately. The component offset coefficient sequence is used to quantify the impact of changes in component proportions on material properties, and its calculation formula is: α ij =k j ·p ij , where ɑ ij k represents the offset coefficient of the j-th component in the i-th material proportion data. j p is the proportional weighting factor for the j-th component. ij Let k be the proportion of the j-th component in the i-th material proportion data. Proportion weighting factor k j The calculation formula is:

[0035]

[0036] Where, p j D represents the proportion threshold of the j-th component. j This represents the proportion deviation threshold for the j-th component. The proportion threshold and proportion deviation threshold can be determined through experimental or empirical data. For example, in a specific embodiment, cement is one of the main components, with its proportion threshold set to 0.5 and proportion deviation threshold set to 0.05. Then, the proportion weighting factor k... j The calculated result is 10. Similarly, the thermosensitive gradient sequence is used to quantify the effect of reaction temperature changes on material properties, and its calculation formula is: β ij =c j ·(t ij -b0), where β ij c represents the thermodynamic gradient of the j-th reaction temperature in the i-th material ratio data. j Let t be the thermosensitive scaling factor for the j-th reaction temperature. ij Let b0 be the reference thermocouple intercept, where b is the j-th reaction temperature in the i-th material ratio data. Thermocouple scaling factor c is the reference thermocouple intercept. jThe calculation formula is:

[0037]

[0038] Among them, T j ΔT represents the thermistor threshold temperature of the j-th reaction. j This represents the temperature deviation threshold for the j-th reaction temperature. For example, in a specific embodiment, the thermosensitive threshold for the reaction temperature is set to 80°C, and the temperature deviation threshold is set to 5°C. Then, the thermosensitive scaling factor c... j The calculation result is 16.

[0039] Next, the material proportioning data is optimized based on the component offset coefficient sequence and the thermosensitive gradient sequence to obtain an optimized parameter set. The specific steps of this process are as follows: First, an initial proportioning model is obtained. In a specific example, the initial proportioning model is a multiple linear regression model, with the independent variables being the proportions of each component and the reaction temperature, and the dependent variable being the performance index value. The component offset coefficients and thermosensitive gradients are extracted sequentially from the component offset coefficient sequence and the thermosensitive gradient sequence, and these data are used to determine optimization points on the initial proportioning model. The selection of optimization points is based on the magnitude of the component offset coefficients and the thermosensitive gradients and their influence on material properties. Subsequently, the spatial distribution direction of the initial proportioning model is identified. Specifically, this involves calculating the partial derivatives of the performance index values ​​with respect to the proportion parameters of each component, constructing a gradient vector. The direction of this gradient vector is the spatial distribution direction, and the optimization direction angle is calculated based on the spatial distribution direction and the thermosensitive gradient. The formula for calculating the optimization direction angle is:

[0040]

[0041] Where θ represents the optimized direction angle, β ij and ɑ ij These represent the thermosensitive gradient and the component offset coefficient, respectively. An optimization auxiliary angle is randomly selected within the pre-constructed optimization angle domain (0, π), and an iterative optimization branch is constructed at the optimization point based on the optimization direction angle and the optimization auxiliary angle. Specifically, this involves synthesizing the optimization direction angle and the optimization auxiliary angle into a new update vector, and using this update vector to iteratively update the corresponding parameters in the initial proportioning model. The role of the iterative optimization branch is to gradually adjust the material proportions and reaction condition parameters to approximate the optimal solution. If the component offset coefficient and thermosensitive gradient are not fully extracted, the initial proportioning model is updated using the iterative optimization branch, and the above steps are repeated; if they are fully extracted, the optimization parameter set is obtained. The optimization parameter set contains the material proportion data and reaction condition parameters after one optimization.

[0042] After obtaining the optimized parameter set, a performance mapping matrix needs to be constructed based on the performance index values. Specifically, this matrix is ​​a hash table, where the key is a specific set of parameters in the optimized parameter set (such as a combination of component ratios and temperature), and the value is the corresponding performance index value. The performance mapping matrix is ​​then combined with the optimized parameter set to obtain the iterative optimization model. The core function of the performance mapping matrix is ​​to map the material performance index values ​​one-to-one with the parameters in the optimized parameter set, thereby achieving visualization and traceability of the performance index values. When all material proportioning data in the basic dataset has been extracted and optimized, the iterative optimization model is considered the first fully optimized model. The generation process of the first fully optimized model is as follows: Figure 2 As shown, the data flow and logical relationships between the various modules are clearly visible.

[0043] After constructing the first fully optimized model, the next step is to obtain the target performance index values ​​and identify target parameter pairs in the first fully optimized model based on these values. Target performance index values ​​are typically set by the user according to actual needs, such as requiring a material compressive strength of 50 MPa or a certain level of corrosion resistance. In the first fully optimized model, optimization endpoints are identified sequentially. Optimization endpoints are optimization nodes directly connected to the performance mapping matrix. Each optimization endpoint has a performance mapping set. The corresponding endpoint interval of the performance mapping set is used to determine whether the target performance index value belongs to that interval. This interval is defined as a closed interval formed by the minimum and maximum values ​​of all performance index values ​​contained in the performance mapping set. If the target performance index value does not belong to the corresponding endpoint interval, the process returns to identifying other optimization endpoints; if it does, the performance mapping set is used as the first mapping set, resulting in multiple sets of first mapping sets. Similarly, in the first fully optimized model, adjacent mapping pairs with the same angular direction are identified sequentially. Adjacent mapping pairs with the same angular direction are two performance mappings with the same optimization direction angle and adjacent optimization endpoints. Identify the adjacent intervals of the same-angle adjacent mapping pairs and determine whether the target performance index value belongs to the same-angle adjacent interval. If it does not, continue to identify other same-angle adjacent mapping pairs; if it does, use the same-angle adjacent mapping pair as the second mapping pair, obtaining multiple sets of second mapping pairs. Finally, based on the multiple sets of first mapping pairs and multiple sets of second mapping pairs, use the target performance index value to identify target parameter pairs in the first fully optimized model. The process of identifying target parameter pairs involves the use of a distance formula, which is: d p =|x p1 -x p2 | and d q =|y q1 -y q2 |, where d p and d p Let x represent the first neighbor distance and the second neighbor distance, respectively.p1 and x p2 This represents the two nearest neighbor index values ​​in the first nearest neighbor index pair, y q1 and y q2 This represents the two neighboring index values ​​in the second neighboring index pair. By calculating the minimum neighbor distance, the target neighboring index pair can be identified, and the target parameter pair can be further determined.

[0044] After obtaining the target parameter pair, secondary parameter optimization is required to generate a second fully optimized model. The specific steps of secondary parameter optimization depend on the parameter relationship attribute of the target parameter pair. If the parameter relationship attribute is a preset co-positional end parameter, the parameter relationship attribute is defined as follows: if two parameters in a target parameter pair correspond to the same optimization end node in the first fully optimized model, then its attribute is a co-positional end parameter. Then, based on the overlapping optimization path and the preset temperature gradient, a joint material preparation experiment is conducted to obtain a set of experimental performance indicators. For example, with the reaction temperature corresponding to the target parameter as the center and a step size of ±2℃, five temperature level points are set for the experiment. The set of experimental performance indicators contains multiple experimental performance indicator values, each corresponding to a specific set of reaction temperatures. By calculating the experimental thermosensitive gradient of the reaction temperature, the target end node of the target parameter pair can be further optimized. The calculation formula for the experimental thermosensitive gradient is consistent with the calculation formula for the aforementioned thermosensitive gradient sequence. If the parameter relationship attribute is a preset angularly adjacent parameter, and if two parameters in a target parameter pair correspond to two adjacent optimization end nodes with the same optimization direction angle, then its attribute is an angularly adjacent parameter. In this case, a joint material preparation experiment is conducted based on the overlapping optimization path and the preset component ratio gradient to obtain a set of experimental performance indicators. The set of experimental performance indicators contains multiple experimental performance indicator values, each corresponding to a specific component ratio. By calculating the experimental offset coefficient of the component ratio, the parent node of the target parameter pair and the target optimization direction angle can be further optimized. The calculation formula for the experimental offset coefficient is consistent with the calculation formula for the aforementioned component offset coefficient sequence. Finally, an experimental mapping is constructed based on the experimental performance indicator values, and the experimental mapping is linked to the end optimization branch to obtain the second fully optimized model.

[0045] Based on the second fully optimized model, target parameters are identified according to the target performance index values. The specific steps are as follows: Experimental mappings are extracted sequentially from the second fully optimized model, and the corresponding experimental performance index values ​​are obtained. The performance difference between the experimental performance index values ​​and the target performance index values ​​is calculated, resulting in a set of performance difference values. The formula for calculating the performance difference value is:

[0046] ΔP=|P 实验 -P 目标 |

[0047] Where ΔP represents the performance difference value, P 实验P represents the test performance index value. 目标 This represents the target performance index value. The minimum difference value is extracted from the set of performance difference values, and the target parameter corresponding to the minimum difference value is identified. The target parameter is the optimal material ratio and reaction condition parameter that satisfies the target performance index value.

[0048] Finally, the mixing temperature sequence corresponding to the target parameters is identified, and the inorganic material for the joints of high-strength and corrosion-resistant PC components is prepared based on the mixing temperature sequence. The generation process of the mixing temperature sequence is as follows: the target offset coefficient sequence and the target thermosensitive gradient sequence of the target parameters are extracted from the second fully optimized model, and the target component ratio sequence and the target reaction temperature sequence corresponding to the target offset coefficient sequence and the target thermosensitive gradient sequence are identified. The target component ratio sequence and the target reaction temperature sequence together constitute the mixing temperature sequence. For example, in a specific embodiment, the target component ratio sequence is 50% cement, 30% sand, and 20% additives, and the target reaction temperature sequence is 80℃, 90℃, and 100℃, then the mixing temperature sequence is (50%, 30%, 20%; 80℃, 90℃, 100℃). Based on the mixing temperature sequence, the material can be prepared according to the conventional inorganic material preparation process.

[0049] This invention also provides a system for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components, the structure of which is as follows: Figure 3 As shown, the system includes a first fully optimized model generation module, a target parameter pair identification module, a second fully optimized model generation module, and a high-strength, corrosion-resistant PC component joint inorganic material preparation module. The first fully optimized model generation module is responsible for acquiring the basic dataset and generating the first fully optimized model; the target parameter pair identification module is responsible for identifying target parameter pairs based on target performance index values; the second fully optimized model generation module is responsible for performing secondary parameter optimization on the target parameter pairs and generating the second fully optimized model; the high-strength, corrosion-resistant PC component joint inorganic material preparation module is responsible for identifying target parameters and preparing materials according to the mixing temperature sequence. Furthermore, this invention also provides an electronic device and a computer-readable storage medium for implementing the functions of the above-described method and system.

[0050] In summary, this invention achieves efficient preparation of inorganic materials for joints of high-strength and corrosion-resistant PC components through a data-driven approach, solving the shortcomings of existing technologies in terms of mechanical properties, corrosion resistance, long-term stability, and construction adaptability.

Claims

1. A method for preparing inorganic material for joints of high-strength and corrosion-resistant PC components, characterized in that, The method includes: Obtain the basic dataset for PC component joint materials; Material proportion data is extracted sequentially from the basic dataset. Component proportion sequence, reaction temperature sequence, and performance index value are extracted from the material proportion data. Component offset coefficient sequence and thermosensitive gradient sequence are calculated based on the component proportion sequence and reaction temperature sequence, respectively. Based on the component offset coefficient sequence and the thermosensitive gradient sequence, the material proportioning data is optimized once to obtain an optimized parameter set. Based on the performance index values, a performance mapping matrix of the optimized parameter set is constructed to obtain an iterative optimization model. Determine whether the material proportion data in the basic dataset has been completely extracted; If the material ratio data in the basic dataset has not been completely extracted, return to the steps described above for sequentially extracting the material ratio data from the basic dataset. If the material proportion data in the basic dataset has been extracted, then the iterative optimization model will be used as the first fully optimized model. Obtain the target performance index value, and identify the target parameter pair in the first fully optimized model based on the target performance index value; Based on the target parameters, a second fully optimized model is obtained by performing secondary parameter optimization on the first fully optimized model. Based on the target performance index value, the target parameters are identified on the second fully optimized model, the corresponding mixing temperature sequence is identified, and the inorganic material for the joint of high-strength corrosion-resistant PC components is prepared according to the mixing temperature sequence.

2. The method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components as described in claim 1, characterized in that, The calculation of the component offset coefficient sequence and the thermosensitive gradient sequence based on the component ratio sequence and the reaction temperature sequence respectively includes: The component proportions are extracted sequentially from the component proportion sequence. Based on the component proportions, the component offset coefficients are calculated using the following formula to obtain the component offset coefficient sequence: α ij =k j ·p ij , where α ij k represents the offset coefficient of the j-th component in the i-th material proportion data. j p is the proportional weighting factor for the j-th component. ij The proportion of the j-th component in the i-th material ratio data; The reaction temperatures are extracted sequentially from the reaction temperature sequence. Based on the reaction temperatures, the thermosensitive gradient is calculated using the following formula to obtain the thermosensitive gradient sequence: β ij =c j ·(t ij -b0), where β ij c represents the thermodynamic gradient of the j-th reaction temperature in the i-th material ratio data. j Let t be the thermosensitive scaling factor for the j-th reaction temperature. ij Let be the j-th reaction temperature in the i-th material ratio data, and b0 be the reference thermosensitive intercept.

3. The method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components as described in claim 2, characterized in that, Before calculating the component offset coefficient sequence and the thermosensitive gradient sequence based on the component ratio sequence and the reaction temperature sequence respectively, the method further includes: Obtain the proportion threshold and proportion deviation threshold of the j-th component; Based on the proportion threshold and proportion deviation threshold of the j-th component, the proportion weighting factor is calculated using the following formula: Where, p j D represents the proportion threshold of the j-th component. j This represents the proportional deviation threshold of the j-th component; Obtain the thermosensitive threshold and temperature deviation threshold for the j-th reaction temperature; Based on the thermosensitive threshold and temperature deviation threshold of the j-th reaction temperature, the thermosensitive scaling factor is calculated using the following formula: Among them, T j ΔT represents the thermistor threshold temperature of the j-th reaction. j This represents the temperature deviation threshold for the j-th reaction temperature.

4. The method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components as described in claim 3, characterized in that, The material proportioning data is optimized firstly based on the component offset coefficient sequence and the thermosensitive gradient sequence to obtain an optimized parameter set, including: Obtain the initial proportioning model; The component offset coefficients and thermal gradients are extracted sequentially from the component offset coefficient sequence and the thermal gradient sequence, respectively. The optimal sites are determined on the initial proportioning model based on the component offset coefficients. Identify the spatial distribution direction of the initial proportioning model, and calculate the optimized direction angle based on the spatial distribution direction and the thermal gradient; Randomly select an optimization auxiliary angle within the pre-constructed optimization angle domain, wherein the optimization angle domain is (0, π); Based on the optimized orientation angle and the optimized auxiliary angle, an iterative optimization branch is constructed at the optimized position; Determine whether the component offset coefficient and thermosensitive gradient have been completely extracted; If the component offset coefficients and thermosensitive gradients are not fully extracted, the initial proportioning model is updated using the iterative optimization branch, and the steps described above for extracting the component offset coefficients and thermosensitive gradients sequentially from the component offset coefficient sequence and the thermosensitive gradient sequence are returned. Once the component offset coefficients and thermosensitive gradients have been extracted, the optimized parameter set is obtained.

5. The method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components as described in claim 4, characterized in that, The step of identifying target parameter pairs in the first fully optimized model based on the target performance index value includes: In the first fully optimized model, the optimized end nodes are identified sequentially, wherein the optimized end nodes refer to the optimized nodes that are directly connected to the performance mapping matrix; Identify the performance mapping set on the optimized end node, and identify the corresponding end interval of the performance mapping set; Determine whether the target performance index value belongs to the same end interval; If the target performance index value does not belong to the same end interval, then return to the steps of sequentially identifying the optimization end nodes in the first fully optimized model. If the target performance index value belongs to the same end interval, then the performance mapping set is used as the first mapping set, and multiple sets of first mapping sets are obtained; In the first fully optimized model, adjacent mapping pairs with the same angle are identified sequentially, wherein the adjacent mapping pairs with the same angle refer to two performance mappings with the same optimization direction angle and adjacent optimization end nodes; Identify the adjacent intervals of the same-angle adjacent mapping pairs; Determine whether the target performance index value belongs to the adjacent interval at the same angle; If the target performance index value does not belong to the same-angle adjacent interval, then return to the steps of sequentially identifying same-angle adjacent mapping pairs in the first fully optimized model. If the target performance index value belongs to the same angle adjacent interval, then the same angle adjacent mapping pair is used as the second mapping pair, and multiple sets of second mapping pairs are obtained; Based on multiple sets of first mappings and multiple sets of second mappings, the target parameter pairs are identified in the first fully optimized model using the target performance index values.

6. The method for preparing inorganic materials for joints of high-strength, corrosion-resistant PC components as described in claim 5, characterized in that, The step of identifying target parameter pairs in the first fully optimized model using the target performance index value based on multiple sets of first mappings and multiple sets of second mappings includes: In the multiple sets of first mapping sets, first mapping sets are extracted sequentially. In the first mapping sets, first parameter pairs of the target performance index values ​​are extracted. First neighboring index pairs of the first parameter pairs are identified to obtain multiple sets of first neighboring index pairs. Extract the second mapping pairs sequentially from the multiple sets of second mapping pairs, identify the second neighbor index pairs of the second mapping pairs, and obtain multiple sets of second neighbor index pairs; The first neighbor distance set and the second neighbor distance set in the multiple sets of first neighbor index pairs and multiple sets of second neighbor index pairs are calculated using pre-constructed distance formulas, wherein the distance formulas are as follows: d p =|x p1 -x p2 | d q =|y q1 -y q2 | Where, d p Let x represent the first neighbor distance of the first neighbor index pair in the p-th group. p1 x represents the value of the first neighboring index in the first neighboring index pair of the p-th group. p2 d represents the value of the second neighboring index in the first neighboring index pair of group p; q y represents the second neighbor distance of the second neighbor index pair in the q-th group. q1 y represents the value of the first neighboring index in the second neighboring index pair of the q-th group. q2 This represents the value of the second neighboring index in the second neighboring index pair of the qth group; Identify the minimum neighbor distance in the first neighbor distance set and the second neighbor distance set; Identify the target proximity index pair corresponding to the minimum proximity distance, and identify the target parameter pair corresponding to the target proximity index pair.

7. The method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components as described in claim 6, characterized in that, The step of performing secondary parameter optimization on the first fully optimized model based on the target parameters to obtain a second fully optimized model includes: Obtain the parameter relationship attributes and overlapping optimization paths of the target parameter pairs; If the parameter relationship attribute is a preset co-position end parameter, then the joint material preparation test is carried out according to the overlapping optimization path and the preset temperature gradient to obtain a set of test performance indicators; The test performance index values ​​are extracted sequentially from the set of test performance indexes, the reaction temperature of the test performance index values ​​is identified, and the test thermosensitive gradient of the reaction temperature is calculated. Identify the target terminal nodes of the target parameter pairs, and construct terminal optimization branches at the target terminal nodes based on the experimental thermosensitive gradient; Based on the experimental performance index values, an experimental mapping is constructed, and the experimental mapping is linked to the terminal optimization branch to obtain the second fully optimized model; If the parameter relationship attribute is a preset adjacent parameter at the same angle, then the joint material preparation test is carried out according to the overlapping optimization path and the preset component ratio gradient to obtain a set of test performance indicators; The test performance index values ​​are extracted sequentially from the set of test performance indexes, the component proportions of the test performance index values ​​are identified, and the test offset coefficients of the component proportions are calculated. Identify the parent node and target optimization direction angle of the target parameter pair, and calculate the target end node based on the parent node and the experimental offset coefficient; Construct an end-optimization branch based on the target end node and the target optimization direction angle; Based on the experimental performance index values, an experimental mapping is constructed, and the experimental mapping is linked to the terminal optimization branch to obtain the second fully optimized model.

8. The method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components as described in claim 7, characterized in that, The step of identifying target parameters on the second fully optimized model based on the target performance index value includes: In the second fully optimized model, the test maps are extracted sequentially to obtain the test performance index values ​​corresponding to the test maps; Calculate the performance difference between the test performance index value and the target performance index value to obtain a set of performance difference values; Extract the minimum difference value from the set of performance difference values, and identify the target parameter corresponding to the minimum difference value.

9. The method for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components as described in claim 8, characterized in that, The temperature sequence corresponding to the identified target parameters includes: Extract the target offset coefficient sequence and target thermal gradient sequence of the target parameters from the second fully optimized model; Identify the target component ratio sequence and target reaction temperature sequence corresponding to the target offset coefficient sequence and the target thermosensitive gradient sequence; Construct a mixing temperature sequence based on the target component ratio sequence and the target reaction temperature sequence.

10. A system for preparing inorganic materials for joints of high-strength and corrosion-resistant PC components, characterized in that, The system includes: The first fully optimized model generation module is used to obtain a basic dataset of PC component joint materials; extract material proportion data sequentially from the basic dataset; extract component ratio sequences, reaction temperature sequences, and performance index values ​​from the material proportion data; calculate component offset coefficient sequences and thermosensitive gradient sequences based on the component ratio sequences and reaction temperature sequences, respectively; perform parameter optimization on the material proportion data based on the component offset coefficient sequences and thermosensitive gradient sequences to obtain an optimized parameter set; construct a performance mapping matrix of the optimized parameter set based on the performance index values ​​to obtain an iterative optimization model; determine whether the material proportion data in the basic dataset has been completely extracted; if the material proportion data in the basic dataset has not been completely extracted, return to the steps of sequentially extracting material proportion data in the basic dataset; if the material proportion data in the basic dataset has been completely extracted, use the iterative optimization model as the first fully optimized model. The target parameter pair identification module is used to obtain the target performance index value and identify the target parameter pair in the first fully optimized model based on the target performance index value. The second fully optimized model generation module is used to perform secondary parameter optimization on the first fully optimized model based on the target parameters to obtain the second fully optimized model. The high-strength corrosion-resistant PC component joint inorganic material preparation module is used to identify target parameters on the second fully optimized model according to the target performance index value, identify the mixing temperature sequence corresponding to the target parameters, and prepare high-strength corrosion-resistant PC component joint inorganic materials according to the mixing temperature sequence.

Images