Evaluation system for salt freezing resistance durability of composite green concrete in soil environment of frozen soil region

Abstract

The invention relates to the technical field of concrete durability evaluation, in particular to a compound green concrete salt freezing resistance durability evaluation system under a frozen soil environment, which comprises a data input unit, a decision information table storage unit, a rough set analysis processing unit and an evaluation output unit, the data input unit receives a sample data set containing freeze-thaw cycle times, dry-wet cycle times, freeze-thaw-dry-wet cycle times, long-term soaking time, composite salt concentration, gas content and fly ash mixing amount, and the decision information table storage unit stores a decision information table constructed based on a rough set theory. The influence factor combination and the relative dynamic elastic modulus percentage are associated, the rough set analysis processing unit establishes a mapping relation through attribute reduction and rule extraction and quantifies the weight of each factor, and the evaluation output unit outputs a relative dynamic elastic modulus predicted value, durability grade classification and key factor contribution sorting. And the salt freezing resistance durability of the concrete under the multi-factor coupling effect is accurately evaluated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of concrete durability evaluation, in particular to a composite green concrete salt and freeze resistance durability evaluation system in a frozen soil environment. BACKGROUND

[0002] Concrete durability evaluation is an important technology. In infrastructure construction in the northeast frozen soil area, concrete needs to withstand the multiple effects of freeze-thaw cycles, dry-wet alternation and combined salt erosion for a long time. The salt and freeze resistance durability of the concrete is directly related to the service life and safety performance of the engineering structure. Scientific evaluation of the salt and freeze resistance durability of the composite green concrete can provide key basis for material proportioning optimization, engineering design and maintenance, and is an important technical support for ensuring the long-term stability of the engineering in the frozen soil area.

[0003] However, the existing concrete durability evaluation method has the problem of insufficient evaluation accuracy under the coupling effect of multiple factors in the complex environment of the northeast frozen soil area. Traditional methods mostly establish evaluation models for single environmental factors and do not fully consider the synergistic effect of factors such as freeze-thaw-dry-wet cycle and combined salt concentration, resulting in a large deviation between the evaluation results and the actual service state. At the same time, the contribution weight of each factor is not quantified in the evaluation process, making it difficult to identify the key factors that play a leading role in durability, and the evaluation results lack pertinence. This problem makes it difficult to accurately predict the performance degradation trend of concrete in engineering practice, which may cause overuse or insufficient protection of materials, ultimately affecting the economy and safety of the engineering in the frozen soil area. In order to solve this technical problem, we provide a composite green concrete salt and freeze resistance durability evaluation system in a frozen soil environment. SUMMARY

[0004] The purpose of the present application is to provide a composite green concrete salt and freeze resistance durability evaluation system in a frozen soil environment to solve the problems raised in the background art.

[0005] 1. Since the traditional method does not consider the coupling effect of multiple factors, the evaluation results deviate greatly from the actual situation. Therefore, the present case constructs a decision information table containing factors such as freeze-thaw-dry-wet cycle, uses rough set algorithm to extract rules and establish mapping relationship, and can accurately evaluate the durability of concrete in complex environment.

[0006] 2. Since the existing method does not quantify the contribution weight of each factor, the evaluation lacks pertinence. Therefore, the present case determines the weight by calculating the attribute support degree change, generates the contribution ranking of key factors, and can clearly identify the dominant factors to provide basis for material optimization.

[0007] To achieve the above object, a composite green concrete salt frost durability evaluation system under permafrost soil environment is provided, a data input unit is used to receive a sample data set of the composite green concrete, the sample data set includes a plurality of preset influencing factors, and the system comprises:

[0008] A decision information table storage unit is used to store a predefined decision information table, the decision information table is constructed based on rough set theory, contains a plurality of records, each record corresponds to a specific combination of influencing factors and a percentage value of relative dynamic elastic modulus associated therewith, and the relative dynamic elastic modulus is used as a decision attribute to represent the dynamic elastic modulus loss rate of the concrete.

[0009] A rough set analysis processing unit is connected with the data input unit and the decision information table storage unit, used to query the decision information table according to the input sample data set, apply a rough set algorithm for attribute reduction and rule extraction, establish a mapping relationship between the influencing factors and the relative dynamic elastic modulus loss rate, and generate an anti-salt frost durability evaluation result.

[0010] An evaluation output unit is used to output the anti-salt frost durability evaluation result, including a predicted value of the relative dynamic elastic modulus and a corresponding durability grade classification.

[0011] Compared with the prior art, the system has the following beneficial effects:

[0012] 1. By integrating multiple working condition data such as salt immersion, salt corrosion-frost-thaw-dry-wet, coupling factors such as freeze-thaw-dry-wet cycles are included in the decision information table, the problem of disconnection between traditional single factor evaluation and actual environment is solved, and the mapping rules of factor combination and relative dynamic elastic modulus are extracted by means of rough set algorithm, which can accurately reflect the influence of multi-factor synergistic effect on concrete performance under the complex environment of northeast permafrost region, and the evaluation result is more in line with the engineering practice.

[0013] 2. The weight is determined by calculating the support degree change of each factor on the decision attribute, the dominant role of core factors such as freeze-thaw-dry-wet cycles is clear, the key factor contribution ranking is output to provide a targeted direction for material optimization, and the relative dynamic elastic modulus is divided into multiple durability grades to realize intuitive mapping from performance degradation degree to specific grade, which is convenient for engineers to quickly judge the concrete state.

[0014] 3. The double verification mechanism reversely checks the rule base by samples not participating in training, dynamically optimizes attribute reduction and mapping relationship, ensures that the evaluation result is stable and reliable, and the predicted value, grade classification and factor ranking output by the system fully meet the needs of permafrost engineering in material proportioning, structure maintenance and other aspects, provide scientific basis for engineering design and operation, and ensure long-term safe and stable operation of infrastructure. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a whole block diagram of the present application.

[0016] The meanings of the respective reference numerals in the figures are as follows.

[0017] 1, data input unit; 2, decision information table storage unit; 3, rough set analysis processing unit; 4, evaluation output unit. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0019] The present application provides a composite green concrete salt frost durability evaluation system under permafrost soil environment, please refer to Figure 1 As shown in the figure, it comprises a data input unit 1 for receiving a sample data set of the composite green concrete, and the sample data set comprises a plurality of preset influencing factors.

[0020] The sample data set comprises a plurality of preset influencing factors, specifically, the number of freeze-thaw cycles, the number of dry-wet cycles, the number of freeze-thaw-dry-wet cycles, the long-term soaking time, the composite salt concentration, the air content and the fly ash content.

[0021] In the complex soil environment of the permafrost region in the northeast, the salt frost durability of the composite green concrete is jointly affected by various environmental and material factors. The sample data set is processed by including key influencing factors and discretization, providing structured condition attribute values for subsequent analysis. These preset influencing factors specifically include the number of freeze-thaw cycles, the number of dry-wet cycles, the number of freeze-thaw-dry-wet cycles, the long-term soaking time, the composite salt concentration, the air content and the fly ash content, each reflecting the erosion effect and the self anti-deterioration ability of the concrete from different dimensions. The process of discretizing each influencing factor into a condition attribute value is as follows:

[0022] The number of freeze-thaw cycles is divided into 5 levels according to the actual number of freeze-thaw cycles of concrete, level 1 corresponds to 0-50 times (mild freeze-thaw), level 2 corresponds to 51-100 times (moderate freeze-thaw), level 3 corresponds to 101-200 times (severe freeze-thaw), level 4 corresponds to 201-300 times (extremely severe freeze-thaw), and level 5 corresponds to more than 301 times (extreme freeze-thaw). Each level represents an interval of freeze-thaw effect intensity, and the higher the value, the more severe the freeze-thaw erosion. The number of dry-wet cycles is divided into 4 levels according to the frequency of dry-wet alternation, level 1 is 0-30 times (low frequency alternation), level 2 is 31-60 times (medium frequency alternation), level 3 is 61-100 times (high frequency alternation), and level 4 is more than 101 times (extreme frequency alternation). This reflects the repeated effect of dry and wet environments on the surface and internal structure of concrete. The number of freeze-thaw-dry-wet cycles is used as an index of freeze-thaw and dry-wet coupling, and is divided into 5 levels according to the number of coupling cycles, level 1 is 0-40 times (mild coupling), level 2 is 41-80 times (moderate coupling), level 3 is 81-150 times (severe coupling), level 4 is 151-250 times (extremely severe coupling), and level 5 is more than 251 times (extreme coupling). This quantifies the cumulative effect of the superposition of the two effects, and this factor has an important weight because it directly reflects the typical combined erosion characteristics of permafrost regions. The long-term soaking time is divided into 4 levels according to the duration of concrete erosion in salt solution, level 1 is 0-30 days (short-term soaking), level 2 is 31-90 days (medium-term soaking), level 3 is 91-180 days (long-term soaking), and level 4 is more than 181 days (ultra-long-term soaking). The longer the time, the greater the depth of salt ion penetration, and the more significant the damage to the internal structure of concrete. The combined salt concentration is divided into 4 levels according to the total concentration percentage of multiple salts (such as NaCl, Na2SO4, etc.) in the erosion solution, level 1 is 0-2% (low concentration), level 2 is 2%-5% (medium concentration), level 3 is 5%-8% (high concentration), and level 4 is more than 8% (extremely high concentration). The higher the concentration, the more severe the salt erosion reaction, and the higher the risk of concrete surface spalling and internal cracking. The air content is divided into 3 levels according to the volume fraction of air bubbles in concrete, level 1 is 1%-3% (low air content), level 2 is 3%-5% (moderate air content), and level 3 is 5%-7% (high air content). Appropriate air bubbles (level 2) can alleviate the volume expansion pressure during freeze-thaw, and low or excessive air content may reduce the frost resistance. The fly ash content is divided into 4 levels according to the proportion of fly ash in the total amount of cementitious materials, level 1 is 0-10% (low dosage), level 2 is 10%-20% (medium dosage), level 3 is 20%-30% (high dosage), and level 4 is more than 30% (ultra-high dosage). Reasonable dosage (levels 2-3) can improve the density of concrete through the pozzolanic reaction, and excessive dosage may exacerbate damage due to insufficient early strength. This provides standardized input for the construction of decision information table based on rough set theory, enabling the combined effect of different factors to be systematically analyzed and accurately mapped to the performance degradation level assessment of concrete.

[0023] characterized in that it further comprises:

[0024] The decision information table storage unit 2 is used for storing a predefined decision information table, the decision information table is constructed based on rough set theory, contains multiple records, each record corresponds to a specific combination of influencing factors and the relative dynamic elastic modulus percentage value associated therewith, and the relative dynamic elastic modulus is used as a decision attribute to characterize the dynamic elastic modulus loss rate of concrete;

[0025] The specific way of constructing the decision information table based on rough set theory is as follows:

[0026] After the discrete processing of the influencing factors of the sample data set is completed, the decision information table needs to be constructed based on rough set theory. This process is the key to converting the original test data into a regularized evaluation basis, which not only connects the quantitative processing of each influencing factor in the previous section, but also provides a structured data basis for subsequent rough set analysis, so that the correlation between the influencing factors and the salt frost resistance of concrete can be systematically presented. The specific construction method of the decision information table is as follows:

[0027] Integrate the durability data of concrete under four typical test conditions: salt immersion condition (only immersed in composite salt solution), salt erosion-dry and wet condition (salt immersion coupled with dry and wet alternation), salt erosion-freezing and thawing condition (salt immersion coupled with freezing and thawing cycle), and salt erosion-freezing and thawing-dry and wet condition (salt immersion, freezing and thawing, and dry and wet triple coupling). These conditions cover the main erosion scenarios that concrete may encounter in the soil environment of the Northeast frozen soil region. When integrating the data, the discrete 7 influencing factors (freezing and thawing cycle times, etc.) are arranged as condition attribute columns, and the relative dynamic elastic modulus percentage is arranged as decision attribute column. This index directly reflects the degree of dynamic elastic modulus loss of concrete due to salt frost damage. Thus, an original data table containing condition attribute set and decision attribute is formed. The original data table is loaded through the ROSETTA software platform, and the attribute reduction process based on genetic algorithm is started:

[0028] The genetic algorithm aims to eliminate redundant attributes and retain core influencing factors. Each condition attribute is regarded as a gene, and attribute subsets are generated through selection, crossover, and mutation operations. The dependency of each subset on the decision attribute is calculated, i.e., whether the subset can fully explain the changes in the decision attribute. For example, if the dependency of the remaining attributes on the decision attribute remains above 90% after removing the fly ash content, the attribute is determined to be a redundant attribute. Conversely, if the dependency decreases by more than 20% after removing the freeze-thaw-dry-wet cycle times, the attribute is retained as a core factor. After multiple iterations (usually 50 generations of evolution), the most significant combination of core factors affecting the salt frost durability of concrete is selected. Based on the reduced core condition attributes and decision attributes, a decision information table is generated. Each record in the table corresponds to a discrete combination of core influencing factors ("freeze-thaw-dry-wet cycle times 3 levels + composite salt concentration 2 levels + air content 2 levels"), and the relative dynamic elastic modulus percentage value under this combination is associated to form a clear mapping relationship between condition attributes and decision attributes. For example, a record shows that when the freeze-thaw-dry-wet cycle times are 3 levels, the composite salt concentration is 3 levels, and the air content is 1 level, the relative dynamic elastic modulus percentage is 60%. This mapping directly reflects the performance degradation state of concrete under this factor combination, enabling the system to quickly query and generate reliable durability evaluation results based on input sample data.

[0029] The specific influencing factor combination in each record is composed of discrete numerical values:

[0030] In the construction of the decision information table, the specific influencing factor combination of each record needs to be presented in discrete numerical values. This processing not only caters to the quantitative classification of each influencing factor in the previous section, but also provides standardized input for attribute reduction and rule extraction of rough set algorithm, enabling the system to identify and associate factor strengths under different test conditions. The specific grading method is as follows:

[0031] The number of freeze-thaw cycles is divided into 5 levels according to the actual number of cycles, level 1 corresponds to 0-50 times (mild freeze-thaw effect), level 2 corresponds to 51-100 times (moderate freeze-thaw effect), level 3 corresponds to 101-200 times (severe freeze-thaw effect), level 4 corresponds to 201-300 times (extremely severe freeze-thaw effect), and level 5 corresponds to more than 301 times (extreme freeze-thaw effect). The value of each level intuitively reflects the intensity of freeze-thaw effect on concrete, for example, in the salt corrosion-freeze-thaw working condition, the freeze-thaw cycle of level 3 means that the concrete has suffered severe freeze-thaw erosion, and the internal pore structure may have been significantly deteriorated. The number of dry-wet cycles is divided into 4 levels according to the frequency of dry-wet alternation, level 1 is 0-30 times (low frequency alternation), level 2 is 31-60 times (medium frequency alternation), level 3 is 61-100 times (high frequency alternation), and level 4 is more than 101 times (extreme frequency alternation). In the salt corrosion-dry-wet working condition, dry-wet cycles of level 3 and above will accelerate the formation and spalling of salt frost on the surface of concrete, so this classification is directly related to the cumulative degree of surface damage. The number of freeze-thaw-dry-wet cycles, as the core index of freeze-thaw and dry-wet coupling, is divided into 5 levels according to the number of coupling, level 1 is 0-40 times (mild coupling), level 2 is 41-80 times (moderate coupling), level 3 is 81-150 times (severe coupling), level 4 is 151-250 times (extremely severe coupling), and level 5 is more than 251 times (extreme coupling). In the most complex working condition of salt corrosion-freeze-thaw-dry-wet, the higher the classification of this index, the stronger the superposition effect of the two actions, and the risk of internal micro-crack propagation of concrete also rises. The long-term soaking time is divided into 4 levels according to the number of days of salt solution erosion, level 1 is 0-30 days (short-term soaking), level 2 is 31-90 days (medium-term soaking), level 3 is 91-180 days (long-term soaking), and level 4 is more than 181 days (super-long-term soaking). In the pure salt immersion working condition, soaking time of level 3 and above will cause a large amount of salt ions to penetrate into the interior of the concrete, providing a material basis for salt freeze damage. The complex salt concentration is divided into 4 levels according to the total concentration percentage of multiple salts in the solution, level 1 is 0-2% (low concentration), level 2 is 2%-5% (medium concentration), level 3 is 5%-8% (high concentration), and level 4 is more than 8% (extremely high concentration). The higher the concentration, the more intense the reaction between salt ions and concrete hydration products, especially at high concentrations, the crystallization and expansion of sodium sulfate will significantly exacerbate the structural damage. The air content is divided into 3 levels according to the volume percentage of internal bubbles in concrete, level 1 is 1%-3% (low air content), level 2 is 3%-5% (moderate air content), and level 3 is 5%-7% (high air content).A moderate air content (Level 2) can buffer the expansion pressure of ice crystals during freeze-thaw cycles, while Level 1 or Level 3 may reduce the anti-freeze effect due to insufficient or uneven distribution of air bubbles. This characteristic makes it a key regulating factor in salt corrosion-freeze-thaw related conditions. Fly ash content is divided into four levels according to its proportion of the total cementitious material: Level 1 is 0-10% (low content), Level 2 is 10%-20% (medium content), Level 3 is 20%-30% (high content), and Level 4 is above 30% (ultra-high content). Medium and high content (Levels 2-3) can improve the salt corrosion resistance by improving the density of concrete, but ultra-high content may have the opposite effect due to insufficient early strength. Therefore, this classification needs to be considered in conjunction with other factors. This structured processing enables the subsequent attribute reduction of the genetic algorithm to accurately identify the core factors. The final decision information table also has a clear condition-decision mapping logic, providing a reliable data framework for the systematic evaluation of concrete's salt-freeze durability.

[0032] After the discretization of the combination of influencing factors is completed, the relative dynamic elastic modulus percentage associated with these combinations needs to be graded according to the degree of concrete performance degradation. This process is the quantitative definition of the decision attribute, which not only follows the structured processing of the condition attribute in the foregoing, but also provides clear result labels for the mapping relationship between conditions and decisions in the decision information table, enabling rough set analysis to directly associate factor combinations with durability status through rule extraction. The specific grading method is based on the test observation results of the dynamic elastic modulus loss rate of concrete under the action of multiple environmental factors. The relative dynamic elastic modulus percentage (i.e., the ratio of the dynamic elastic modulus after erosion to the initial value) is divided into 5 levels, each level corresponds to a discrete label, which is used to accurately quantify the durability status of concrete. Level 1: relative dynamic elastic modulus ≥ 90%): corresponding to the optimal state, the dynamic elastic modulus loss rate of concrete is ≤ 10%, the test observation shows that the internal structure is basically undamaged, the surface has no obvious salt frost or cracks, and the mechanical properties are still good. This level is commonly seen in mild erosion conditions such as short-term immersion (level 1) in salt water, low composite salt concentration (level 1), etc. Level 2: 80% ≤ relative dynamic elastic modulus < 90%): corresponding to the "good" state, the dynamic elastic modulus loss rate is between 10%-20%, a small amount of salt frost can be seen on the surface of the concrete in the test, but there is no obvious cracking, the internal pore structure has slight deterioration, but the overall mechanical properties can still meet the basic use requirements, and it often occurs in moderate erosion environments such as moderate frequency of dry-wet cycles (level 2), moderate freeze-thaw action (level 2), etc. Level 3: 70% ≤ relative dynamic elastic modulus < 80%): corresponding to the "medium" state, the dynamic elastic modulus loss rate reaches 20%-30%, the concrete surface shows obvious salt erosion and spalling, and local fine cracks can be seen, the internal structure damage has begun to accumulate, and it is easy to reach this level under harsh conditions such as severe freeze-thaw-dry-wet coupling (level 3), high composite salt concentration (level 3), etc. Level 4: 60% ≤ relative dynamic elastic modulus < 70%): corresponding to the "poor" state, the dynamic elastic modulus loss rate rises to 30%-40%, the test observation shows that the concrete surface has large-area spalling, the cracks extend to the deep layer, the internal density decreases significantly, and the mechanical properties decay significantly, which usually occurs in strong erosion environments such as extremely severe freeze-thaw action (level 4), super-long-term immersion (level 4), etc. Level 5: relative dynamic elastic modulus < 60%): corresponding to the "bad" state, the dynamic elastic modulus loss rate exceeds 40%, the concrete structure has suffered serious damage, with through cracks and even local fragmentation, and has basically lost its bearing capacity, which often occurs in extreme erosion conditions such as extremely severe freeze-thaw-dry-wet coupling (level 5), extremely high composite salt concentration (level 4), etc. For example, when the freeze-thaw-dry-wet cycle frequency is level 3, the composite salt concentration is level 3, and the air content is level 1, the relative dynamic elastic modulus percentage may fall within the range of 70%-80%, i.e., the decision attribute label is level 3, which directly reflects that the durability of concrete under this factor combination is in the "medium" state.This hierarchical approach is based on both objective observations of experimental data and the computability of decision attributes through discretization labels, providing standardized output targets for the subsequent rough set analysis processing unit 3 to establish the mapping relationship between influencing factors and durability states, making the logic of the entire evaluation system complete the closed loop from factor quantification to result classification.

[0033] Based on the division of the relative dynamic elastic modulus percentage into corresponding durability grades, it is used as a decision attribute to represent the dynamic elastic modulus loss rate, which is determined based on the internal mechanism of concrete material degradation and data correlation rules. This setting not only continues the quantitative processing of the decision attribute classification in the previous section, but also provides a scientific basis for the subsequent extraction of influencing factors and performance degradation correlation rules through rough set algorithms, making the core logic of the entire evaluation system coherent. The specific mechanism and implementation are as follows:

[0034] Firstly, the effectiveness of the index was verified by correlation analysis. Concrete samples with different air contents (1%-7%) were selected for comparative tests under the combined action of salt corrosion, freeze-thaw and wet-dry cycles. The relative dynamic elastic modulus and internal structure damage data of each sample were collected at different cycle times. The analysis found that the air content and the loss rate of relative dynamic elastic modulus were negatively correlated. When the air content was 3%-5%, the relative dynamic elastic modulus remained above 80% after 150 freeze-thaw and wet-dry cycles, and the internal crack density was less than 0.5 / mm². However, when the air content was less than 2%, the relative dynamic elastic modulus decreased to less than 65% after the same number of cycles, and the crack density reached 1.2 / mm². This result proves that the reduction in the percentage of relative dynamic elastic modulus is directly positively correlated with the internal structure damage of concrete (increased porosity and crack expansion). Therefore, this index can objectively reflect the degree of performance degradation caused by salt and freeze-thaw action, and it is scientific to use it as a decision attribute. After setting the relative dynamic elastic modulus as the decision attribute, the mapping relationship between the combination of condition attributes and macro performance degradation can be established. For example, when the combination of freeze-thaw and wet-dry cycle times 3 levels + composite salt concentration 3 levels + air content 1 level occurs, the corresponding relative dynamic elastic modulus percentage falls within the 70%-80% interval (3-level decision attribute), indicating that this factor combination will cause moderate degradation of concrete. However, the combination of freeze-thaw and wet-dry cycle times 5 levels + composite salt concentration 4 levels + air content 1 level corresponds to a relative dynamic elastic modulus of less than 60% (5-level decision attribute), indicating severe degradation. This mapping relationship links the dispersed influencing factors with the intuitive performance results, laying the foundation for rule extraction. When revealing the contribution of each factor to damage evolution through rough set rule extraction, based on the above mapping relationship, the system can automatically identify the key influencing path. For example, if the freeze-thaw and wet-dry cycle times are greater than or equal to 3 levels and the air content is less than or equal to 1 level, the support of the relative dynamic elastic modulus being less than or equal to 70% is 85%, indicating that the combination of these two factors plays a dominant role in damage evolution. However, the fly ash content of 4 levels only has a significant impact on the decision attribute when the composite salt concentration is less than or equal to 2 levels, indicating that its contribution is constrained by other factors. These rules not only quantify the strength of different factor combinations, but also reveal the internal logic of damage evolution, making the evaluation results not only able to predict the durability state of concrete, but also able to identify the key factors leading to degradation, providing targeted guidance for material optimization. In summary, using the relative dynamic elastic modulus as the decision attribute not only verifies its effectiveness in representing damage through correlation analysis, but also provides a carrier for rule extraction through the mapping relationship with condition attributes. This makes the entire evaluation system form a complete logical chain from index selection to rule extraction, ensuring that the analysis of the salt and freeze-thaw durability of composite green concrete is not only scientific and accurate, but also has practical application value.

[0035] The rough set analysis processing unit 3 is connected with the data input unit 1 and the decision information table storage unit 2, and is used for querying the decision information table according to the input sample data set, performing attribute reduction and rule extraction by applying a rough set algorithm, establishing a mapping relationship between influencing factors and the relative dynamic elastic modulus loss rate, and generating an anti-salt frost durability evaluation result.

[0036] After determining the relative dynamic elastic modulus as the decision attribute and establishing the mapping relationship between the conditional attributes and the decision attribute, the rough set analysis processing unit 3 needs to screen the core influencing factors through the attribute reduction process. This link not only connects the construction result of the decision information table, but also lays a foundation for the accuracy of subsequent rule extraction. By quantifying the contribution weight of each conditional attribute, the evaluation system can focus on the factors that play a key role in the anti-salt frost durability of concrete. The specific implementation is as follows:

[0037] The rough set analysis processing unit 3 calculates the dependency of each conditional attribute on the decision attribute. Dependency is an index that measures whether the set of conditional attributes can fully explain the changes in the decision attribute. The higher the value, the closer the association between the attribute set and the decision attribute. When calculating, all 7 conditional attributes in the decision information table are taken as the initial set. Through the dependency analysis function of the ROSETTA software, the initial dependency (denoted as γ0, usually set as the reference value 1.0) is obtained, and the iterative process of removing single conditional attributes is entered:

[0038] Each time, one attribute is removed from the condition attribute set (such as the first removal of "fly ash content"), a new attribute subset is formed, the dependence of the subset on the decision attribute (denoted as γᵢ) is recalculated through the ROSETTA software, and the support degree change amount Δγᵢ = γ0- γᵢ is calculated. For example, after removing the freeze-thaw-dry-wet cycle times, the dependence decreases from 1.0 to 0.65, Δγᵢ = 0.35, indicating that the removal of the attribute causes the explanatory power of the condition attribute to the decision attribute to decrease significantly, while after removing the fly ash content, the dependence remains at 0.96, Δγᵢ = 0.04, indicating that its influence on the overall association is small. Based on the support degree change amount, the independent contribution weight of each factor is determined: normalize Δγᵢ, that is, the contribution weight ωᵢ of a certain attribute = Δγᵢ / ΣΔγᵢ (ΣΔγᵢ is the sum of all attribute Δγᵢ). Assuming that the Δγᵢ of the 7 attributes are 0.35 (freeze-thaw-dry-wet cycle times), 0.20 (complex salt concentration), 0.15 (air content), 0.10 (freeze-thaw cycle times), 0.08 (dry-wet cycle times), 0.06 (long-term soaking time), and 0.04 (fly ash content), then the weight ω of freeze-thaw-dry-wet cycle times = 0.35 / (0.35+0.20+0.15+0.10+0.08+0.06+0.04) = 0.35, which becomes the core factor with the highest contribution weight, while the weight of fly ash content is only 0.04, which is a secondary factor, so the subsequent rule extraction focuses only on the factor combination that plays a leading role in the salt frost durability of concrete, which not only simplifies the analysis model but also ensures the reliability of the evaluation results. The logical progression formed by the decision information table constructed in the previous section provides key support for generating accurate salt frost durability evaluation results.

[0039] In the rule extraction stage, the weight coefficients are assigned based on the attribute importance ranking results, as follows:

[0040] After completing attribute reduction and determining the independent contribution weight of each condition attribute, the rule extraction stage needs to assign weight coefficients based on the attribute importance ranking results. This process further quantifies the strength of the core influencing factors, both carrying forward the attribute importance differences analyzed through the support degree change amount in the previous section and providing numerical basis for building a weighted evaluation rule set, so that the subsequent evaluation of the salt frost durability of concrete can more accurately reflect the actual influence of each factor. The specific implementation is as follows:

[0041] The support degree drop (Δγᵢ) calculated in the attribute reduction process is directly used as the basis for the importance score. The greater the support degree drop, the more critical the attribute is to the explanation of the decision attribute, and the higher its importance score. For example, the Δγᵢ of freeze-thaw-dry-wet cycle times calculated in the foregoing is 0.35, the composite salt concentration is 0.20, the air content is 0.15, the freeze-thaw cycle times is 0.10, the dry-wet cycle times is 0.08, the long-term soaking time is 0.06, and the fly ash content is 0.04. These values are directly used as the importance score of each attribute, clearly showing the core position of freeze-thaw-dry-wet cycle times among all factors and the relatively minor influence of fly ash content. The importance scores are normalized to obtain the weight coefficients: the importance score of each attribute is divided by the sum of the importance scores of all attributes, so that the sum of the final weight coefficients is 1. Taking the above values as an example, the sum is 0.35+0.20+0.15+0.10+0.08+0.06+0.04=1.0, so the weight coefficients of each attribute are the same as the importance score values, i.e., the weight of freeze-thaw-dry-wet cycle times is 0.35, the weight of composite salt concentration is 0.20, the weight of air content is 0.15, the weight of freeze-thaw cycle times is 0.10, the weight of dry-wet cycle times is 0.08, the weight of long-term soaking time is 0.06, and the weight of fly ash content is 0.04. This result is consistent with the actual erosion mechanism. The coupling effect of freeze-thaw and dry-wet is the main driving force for salt freeze damage of concrete in the frozen soil region of Northeast China, so it has the highest weight, while the fly ash content has a relatively limited effect under the superposition of multiple severe environmental factors, so it has the lowest weight. When the weight distribution results are used to build the weighted evaluation rule set, the system will incorporate the weight coefficients of each conditional attribute in each extracted rule. For example, a rule is: if the freeze-thaw-dry-wet cycle times ≥ 3 levels and the composite salt concentration ≥ 3 levels, then the relative dynamic elastic modulus ≤ 70%. The confidence of the rule will be strengthened according to the weights of the two factors (0.35+0.20=0.55), so that it has a higher priority in evaluation, while the rules containing fly ash content will have their confidence adjusted appropriately due to the low weight of this factor. This weighting method ensures that each rule in the rule set can play a corresponding role according to the actual importance of the factors, making the rule-based durability evaluation results not only consistent with the test data rules, but also accurately reflecting the contribution differences of different factors to the salt freeze resistance of concrete, and closely linking with the attribute reduction logic in the foregoing, providing a solid rule foundation for generating reliable evaluation results.

[0042] After completing the rule extraction and determining the weight coefficients of each condition attribute, a double verification mechanism is needed to establish the mapping relationship between the condition attribute combination and the decision attribute. This process is a strict check of the reliability of the rule base, which not only connects the initial rule base generated based on weight coefficients, but also ensures the accuracy of the mapping relationship through dynamic optimization, so that the final generated anti-salt freeze durability evaluation result can truly reflect the performance state of concrete in complex environments. The specific implementation is as follows:

[0043] Based on the allocated weight coefficients, an initial rule base is generated, taking the core condition attribute combination as input, combining its weight coefficient to calculate the influence strength on the decision attribute (relative dynamic elastic modulus level), forming a rule set of "if condition attribute A is x level and condition attribute B is y level, then the decision attribute is z level", for example, based on the highest weight "freeze-thaw-dry-wet cycle times" (0.35) and "complex salt concentration" (0.20), the rule "if freeze-thaw-dry-wet cycle times ≥ 4 levels and complex salt concentration ≥ 3 levels, then the relative dynamic elastic modulus level is 4 levels (poor)" is generated, and the confidence (such as the rule confidence is 0.55) accumulated by the weight coefficient is marked for each rule. The initial rule base is verified in reverse by MATLAB program:

[0044] From the test data set, 30% of the samples that did not participate in rule training (such as 100 groups of independent data under salt erosion-freeze-thaw-dry-wet conditions) are selected, the condition attribute values of these samples are input into the rule base, and the corresponding decision attribute prediction values are obtained. At the same time, the measured relative dynamic elastic modulus percentage and corresponding level in the sample are extracted, and the deviation between the predicted value and the measured value is calculated. The deviation calculation uses the weighted average of the absolute value of the level difference (such as the predicted level is 4, the measured is 5, the deviation is 1) and the relative error (such as the predicted relative dynamic elastic modulus is 65%, the measured is 58%, the relative error is 12%). The prediction accuracy of the rule base is evaluated comprehensively, and the deviation threshold (such as the absolute value of the level difference ≤1 and the relative error ≤10%) is set. If more than 30% of the samples in the verification result exceed the threshold, it is determined that the initial rule base is not reliable, and the genetic algorithm is triggered immediately to re-perform attribute reduction. When re-reducing, the genetic algorithm will adjust the attribute selection strategy (such as increasing the attention to "air content" which is directly related to the anti-freeze performance), optimize the core attribute combination, and then generate a new rule base. The reverse verification process is repeated until the sample deviation rate ≥90% in the continuous 3 times verification, and the prediction error is stable within the preset range (such as the average relative error ≤8%), so that the mapping relationship between the condition attribute combination and the decision attribute is established, making the mapping relationship adapt to the performance data law under different test conditions, and forming an organic link with the rule extraction process in the previous text, providing a solid guarantee for the final output of accurate anti-salt freeze durability evaluation results.

[0045] The evaluation output unit 4 is used to output the salt frost durability evaluation results, including the predicted value of relative dynamic elastic modulus and its corresponding durability grade classification.

[0046] After completing the verification of the combination of condition attributes and the mapping relationship of decision attributes, the evaluation output unit 4 will generate the salt frost durability evaluation results. This step is the presentation of the results of the entire analysis process, which not only establishes the mapping relationship through the double verification mechanism in the foregoing, but also meets the comprehensive needs of concrete durability evaluation in practical applications through multi-dimensional output, so that the evaluation results have both quantitative precision and can intuitively reflect the performance state and key influencing factors. The specific output content and implementation mode are as follows:

[0047] First, the generation of the predicted value of relative dynamic elastic modulus is based on the weighted calculation of weight coefficients and input influencing factor values. For the composite green concrete sample to be evaluated, the discrete values of the 7 influencing factors are first obtained, and then these values are multiplied by the corresponding weight coefficients and summed to obtain the weighted total score. Through the preset linear mapping formula, the weighted total score is converted into a specific predicted value of relative dynamic elastic modulus. For example, the weighted total score of a certain sample is 0.45, and the corresponding predicted value is 77.5%. This value directly quantifies the degree of retention of the dynamic elastic modulus of the concrete under the combined action of the current influencing factors;

[0048] Second, the durability grade classification is marked according to the interval to which the predicted value belongs. According to the established grading standard, the interval corresponding to the predicted value of relative dynamic elastic modulus is marked as the corresponding grade: ≥90% is grade 1 (excellent), 80%-90% is grade 2 (good), 70%-80% is grade 3 (medium), 60%-70% is grade 4 (poor), and <60% is grade 5 (poor). Taking the predicted value of 77.5% as an example, it belongs to the interval of 70%-80%, so the durability grade is marked as grade 3, which intuitively reflects the current performance degradation of the concrete;

[0049] Third, the key factor contribution ranking lists the influence intensity of each influencing factor on the current prediction result according to the weight coefficient from high to low. Based on the weight distribution determined in the rule extraction stage, the seven influencing factors are arranged in descending order of weight coefficient, and the actual input value (discretization level) of each factor is marked to clarify which factors are the main driving factors leading to the current durability state. For example, the ranking result of a certain sample is: freeze-thaw-dry-wet cycle times (0.35, level 3) > composite salt concentration (0.20, level 2) > air content (0.15, level 2) > freeze-thaw cycle times (0.10, level 2) > dry-wet cycle times (0.08, level 1) > long-term soaking time (0.06, level 2) > fly ash content (0.04, level 3), which clearly shows that freeze-thaw-dry-wet coupling effect and composite salt concentration are the core factors affecting the durability of the sample, providing a targeted direction for subsequent material optimization or maintenance measures, and enabling the entire evaluation system to provide scientific, comprehensive and practical reference for the salt frost durability evaluation of composite green concrete in the permafrost region of Northeast China.

[0050] In the present application, the sample data set containing freeze-thaw cycle times, dry-wet cycle times, freeze-thaw-dry-wet cycle times, long-term soaking time, composite salt concentration, air content and fly ash content is received by the data input unit 1, the decision information table storage unit 2 stores the decision information table constructed based on rough set theory, the association between influencing factor combination and relative dynamic elastic modulus percentage, the rough set analysis processing unit 3 establishes the mapping relationship through attribute reduction and rule extraction, quantifies the weight of each factor, and the evaluation output unit 4 outputs the relative dynamic elastic modulus prediction value, durability grade classification and key factor contribution ranking, accurately evaluating the salt frost durability of concrete under the coupling effect of multiple factors.

[0051] The basic principles, main features and advantages of the present application are shown and described above. It should be understood by those skilled in the art that the present application is not limited by the above examples, and the above examples and descriptions in the specification are only preferred examples of the present application and are not intended to limit the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A system for evaluating the salt and freeze resistance durability of composite green concrete in a permafrost soil environment, comprising a data input unit (1) for receiving a sample data set of the composite green concrete, the sample data set comprising a plurality of preset influencing factors; characterized in that, Also comprising: The decision information table storage unit (2) is used for storing the pre-defined decision information table, which is constructed based on rough set theory, contains multiple records, each record corresponds to a specific combination of influencing factors and the relative dynamic elastic modulus percentage value associated therewith, and the relative dynamic elastic modulus is used as a decision attribute to characterize the dynamic elastic modulus loss rate of concrete; The rough set analysis processing unit (3) is connected with the data input unit (1) and the decision information table storage unit (2), which is used to query the decision information table according to the input sample data set, apply rough set algorithm for attribute reduction and rule extraction, to establish the mapping relationship between influencing factors and relative dynamic elastic modulus loss rate, and generate the salt frost durability evaluation result; The evaluation output unit (4) is used for outputting the salt frost durability evaluation result, including the predicted value of relative dynamic elastic modulus and its corresponding durability grade classification.

2. The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 1, characterized in that: The sample data set includes multiple pre-set influencing factors, specifically the freeze-thaw cycle number, dry-wet cycle number, freeze-thaw-dry-wet cycle number, long-term immersion time, composite salt concentration, air content and fly ash content; Respectively converted into conditional attribute values through discretization processing, wherein the freeze-thaw cycle number corresponds to the freeze-thaw action intensity of the concrete, the dry-wet cycle number reflects the dry-wet alternating environment action frequency, the freeze-thaw-dry-wet cycle number characterizes the cumulative effect of freeze-thaw and dry-wet coupling, the long-term immersion time quantifies the continuous erosion time of salt solution, the composite salt concentration indicates the total concentration of multiple salts in the erosion solution, the air content describes the internal bubble content of the concrete, and the fly ash content reflects the mineral admixture proportion.

3. The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 2, characterized in that: The specific way of constructing the decision information table based on rough set theory is: Integrate the durability data of concrete under four test conditions of salt immersion, salt erosion-dry-wet, salt erosion-freeze-thaw and salt erosion-freeze-thaw-dry-wet, take the pre-set influencing factors as the conditional attribute column, and the relative dynamic elastic modulus percentage as the decision attribute column, to form a conditional attribute set; Load the original data table through the ROSETTA software platform, use genetic algorithm to reduce the conditional attribute set, eliminate redundant attributes and retain the core influencing factor combination, and generate a decision information table containing the mapping relationship between the simplified conditional attributes and the decision attributes.

4. The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 2, characterized in that: The specific influencing factor combination in each record is composed of discrete values: Wherein the freeze-thaw cycle number is classified according to the actual cycle number, the dry-wet cycle number is classified according to the dry-wet alternating frequency, the freeze-thaw-dry-wet cycle number is classified according to the coupling number, the long-term immersion time is classified according to the erosion days, the composite salt concentration is classified according to the solution concentration percentage, the air content is classified according to the volume percentage, and the fly ash content is classified according to the mixing ratio, and each level value represents the intensity interval of the factor under the specific test condition.

5. The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 1, characterized in that: The relative dynamic elastic modulus percentage associated with the influencing factor combination is classified according to the performance degradation degree of concrete, and the classification is based on the test observation results of the dynamic elastic modulus loss rate of concrete under multiple environmental factors, and each level value directly corresponds to the discrete label of the decision attribute, which is used to quantitatively represent the durability state of concrete.

6. The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 5, characterized in that: The mechanism of the relative dynamic elastic modulus as a decision attribute representing the dynamic elastic modulus loss rate lies in that the correlation analysis confirms that the air content is negatively correlated with the relative dynamic elastic modulus loss rate, and it is proved that the percentage reduction of the relative dynamic elastic modulus directly reflects the increase of the internal structure damage of concrete; Setting it as a decision attribute can establish the mapping relationship between the combination of condition attributes and macro performance degradation, and then reveal the contribution degree of each factor to damage evolution through rough set rule extraction.

7. The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 3, characterized in that: In the attribute reduction process, the rough set analysis processing unit (3) calculates the dependency of each condition attribute on the decision attribute, removes a single condition attribute in turn, and analyzes the change of the support degree of the remaining attribute set in the decision information table on the decision attribute through the ROSETTA software, thereby determining the independent contribution weight of each factor.

8. The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 3, characterized in that: In the rule extraction stage, the weight coefficients are assigned based on the attribute importance ranking results, as follows: The importance score of each condition attribute is calculated according to the support degree drop, and the importance score is normalized to a weight coefficient, so that the weight sum of the preset influencing factors is 1, wherein the freeze-thaw-dry-wet cycle weight is the highest, and the fly ash content weight is the lowest, and the weight distribution result is used to construct a weighted evaluation rule set. 9.The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 1, characterized in that: A double verification mechanism is adopted when establishing the mapping relationship: first, generate the initial rule base of condition attribute combination to decision attribute according to the weight coefficient, then verify the rule base in reverse through MATLAB program, input the untrained samples in the test data set, compare the deviation between the rule prediction value and the measured decision attribute value; when the deviation exceeds the threshold, trigger the genetic algorithm to reduce the attributes again, until the prediction error is stable within the preset range, and then determine that the mapping relationship is established. 10.The system for evaluating the salt and freeze resistance durability of the composite green concrete in the permafrost soil environment according to claim 1, characterized in that: The salt and freeze durability evaluation result contains three parts of output, as follows: First, the relative dynamic elastic modulus prediction value is generated by weighted calculation of the weight coefficient and the input influencing factor value; Second, the durability grade classification is marked according to the grade interval to which the prediction value belongs; Third, the key factor contribution ranking is listed according to the weight coefficient from high to low, and the influence strength of each influencing factor on the current prediction result.

Images