Soil cement mix design method and soil cement construction method

Abstract

The present invention provides a soil cement mix design method that uses artificial intelligence to set mix design parameters that affect the physical properties of soil cement, thereby enabling the soil cement to exhibit strength in the elastic region and suppressing mud formation, and a soil cement construction method including this method. [Solution] The present invention is a soil cement mix design method comprising at least the steps of (A) implementing a trained artificial intelligence, (B) outputting lead mix examples, (C) measuring the physical properties of a test specimen, and (D) setting the indicated mix. Furthermore, the present invention is a soil cement construction method that includes at least (i) a step of collecting soil and sand, (ii) a step of material testing using the soil and sand, (iii) a step of setting a specified mix design using the above mix design method, and (iv) a step of test construction using the specified mix, and further optionally includes (v) a step of resetting the specified mix design using the above mix design method if the target strength is not met in the above test construction.

Description

[Technical Field]

[0001] The present invention relates to a soil cement mix design method that uses artificial intelligence (AI) to set the specified mix design for soil cement, or to readjust (correct) it if necessary, and to a soil cement construction method using this method. [Background technology]

[0002] Incidentally, the above-mentioned mix design is typically determined by laboratory mix design tests, and the mix design elements included in this design are mainly the type of cement, the unit cement content, and the test moisture content. Furthermore, as shown in Figure 1, the test moisture content consists of the moisture content on the left side of the quadratic curve that intersects the line representing the target compressive strength (lower limit moisture content), the test moisture content that indicates the peak strength of the curve (peak strength moisture content), and the moisture content on the right side of the curve (upper limit moisture content). As shown in Figure 2, this curve is determined by changing the unit cement content and test moisture content, etc., based on the type and characteristics of the on-site soil generated at the construction site (hereinafter simply referred to as "soil") obtained from tests on the physical and chemical properties of the soil (hereinafter referred to as "soil properties") (hereinafter referred to as "material tests").

[0003] The above-mentioned soil cement is a mixture of soil and sand with cement, water, etc., or a solidified product thereof. Construction methods using soil cement mainly include the soil cement method for erosion control facilities (which includes compaction type and flow type) and the ground improvement method which involves excavating soft ground and replacing it with soil cement. The two methods differ in their target soil cement strength, workability, etc. (hereinafter sometimes referred to as "quality" or "physical properties") and construction methods. Of these, the soil cement method for erosion control is a method of constructing erosion control facilities by solidifying them with soil cement, whereas conventional erosion control facilities were constructed with concrete. Both construction methods utilize soil generated through excavation, eliminating the need for soil removal and disposal, thus reducing costs, noise, and vibration associated with soil transportation. Furthermore, erosion control soil cement generally has a viscosity of 3.0 N / mm². 2It possesses the strength within the elastic range (the region where strain and stress are proportional), and offers excellent workability for tasks such as spreading and compacting.

[0004] The mix design for the soil cement method is carried out through (i) soil sampling, (ii) material testing, and (iii) setting the specified mix through laboratory mix testing (see Non-Patent Document 1, page 52, etc.).

[0005] However, the soil used in the above-mentioned indoor mix design tests (for example, surface soil sampled manually at the construction site) and the soil sampled during the construction phase (for example, soil mixed with surface and deep soil after extensive and deep excavation using construction machinery) may have different soil properties due to the different sampling areas. This difference in soil properties affects the quality of the soil cement. To confirm this effect, a test construction (hereinafter referred to as "test construction") is carried out at the construction site based on the specified mix design obtained in the mix design. If the soil cement of the specified mix does not meet the required quality, such as strength, as a result of the test construction, material tests and indoor mix design tests will be repeated using the soil sample of the construction site, and the specified mix design will be readjusted (corrected).

[0006] However, since the strength of soil cement is usually measured using the compressive strength of a 28-day-old specimen, construction is suspended until this strength is achieved. Furthermore, as mentioned above, the specified mix design includes a wide range of items such as the type of cement, the unit cement content, and the test water content. Thus, in conventional soil-cement construction methods, the design of soil-cement mixes is complicated and time-consuming and costly because it requires repeated material tests and laboratory mix tests on a wide range of mix components.

[0007] Therefore, in order to streamline the design of soil cement mixes, Patent Document 1 proposes a method for determining whether soil can be used, and Patent Document 2 proposes a method for designing soil cement mixes. Of these, the method described in Patent Document 1 includes the steps of determining the amount of microorganisms, the amount of humic acid, and the water content ratio contained in the soil and comparing them with a criterion (threshold) for determining whether the soil is usable, and determining that the soil is usable if each of the above amounts of the soil is below the threshold.

[0008] Furthermore, the method described in Patent Document 2 includes the step of collecting soil and sand from the construction site, The steps to determine the target strength of the construction target using soil cement, A step to calculate the fine-grained content of the on-site soil used in the soil cement mix, This includes a step of determining whether the target strength, which is determined from the fine-grained content calculated from the relationship between the fine-grained content, the amount of cement added, and the compressive strength, is met. If it is determined that the target strength is met, the next step is to determine the type of soil cement that meets the target strength. The steps include determining the amount of cement to add based on the relationship between the fine particle content, the type of cement, and the amount of cement added, The first step is to determine the amount of water to add based on the relationship between the fine particle content and the amount of cement and water added. The next step is to conduct an indoor mix design test based on the determined amounts of cement and water to be added. This method includes [something].

[0009] However, the method described in Patent Document 1 is limited to determining whether or not soil can be used, and the method described in Patent Document 2 is complicated and involves many steps. [Prior art documents] [Non-patent literature]

[0010] [Non-Patent Document 1] Japan Landslide and Erosion Control Technology Center, "Soil Cement Construction Manual for Erosion Control," Revised 2nd Edition, published September 30, 2016. [Patent Documents]

[0011] [Patent Document 1] Japanese Patent Publication No. 2019-74485 [Patent Document 2] Japanese Patent Publication No. 2016-218031 [Overview of the project] [Problems that the invention aims to solve]

[0012] As described above, sabo soil cement requires long-term durability and strength in the elastic region, and unlike the soil cement used in the aforementioned ground improvement method, solidification of soil and sand by cement hydration is essential. Further, since cement hydration is affected by various factors, the mix design of sabo soil cement is particularly complicated and requires much time and cost. Accordingly, an object of the present invention is to provide a soil cement mix design method that can develop strength in the elastic region and suppress mudification of soil cement by setting the aforementioned mix items affecting the physical properties of soil cement such as the hydration reaction of cement using artificial intelligence, and a soil cement construction method including the method. [Means for Solving the Problems]

[0013] As a result of intensive studies to achieve the above object, the present inventors have completed the inventions of the following "1" to [7].

[0014] [1] A soil cement mix design method comprising at least the following steps (A) to (D). (A) A trained artificial intelligence implementation step of causing artificial intelligence to perform machine learning using training data including at least the soil properties used for the soil cement, cement type, unit cement amount, test water content ratio, unit volume mass of the soil cement, compressive strength, and mixing ratio of the improving material to soil and sand, and implementing the trained artificial intelligence. (B) A lead mix example output step of inputting at least the soil properties of the soil and sand at the construction target site, the type of cement used for the soil cement, the unit cement amount, and the test water content ratio into the trained artificial intelligence, and causing the trained artificial intelligence to output a lead mix example (a mix example serving as a starting point for deriving (leading to) the specified mix) including at least the unit cement amount and the test water content ratio. (C) A specimen physical property value measurement step of preparing a soil cement specimen based on the output lead mix example, and measuring at least the short-term strength of the specimen and the mass of the specimen. (D) Setting of the indicated mix: The physical properties of the specimen measured above are input into the trained artificial intelligence, which predicts at least the unit cement content, test water content ratio, and long-term strength of the soil cement and outputs mix design examples, and a mix design example that satisfies the target long-term strength is selected from among these mix design examples and set as the indicated mix design. [2](E) If the long-term strength of the specimen prepared based on the specified mix design set in step (D) above does not meet the target long-term strength (hereinafter referred to as "target strength"), the method for designing the soil cement mix according to [1] above, further comprising as an optional step of resetting the specified mix design by repeating steps (B) to (D) above. [3] The soil cement mix design method described in [1] above, wherein the (D) step above includes at least the following (D1) to (D5) steps. (D1) For each unit cement quantity, a regression analysis is performed with the above test water content as the explanatory variable and the above predicted long-term strength as the dependent variable, and a quadratic regression equation is obtained for each unit cement quantity. (D2) A prediction process for peak intensity, etc., which uses the above quadratic regression equation to calculate and predict peak intensity, peak intensity water content ratio, upper water content ratio and lower water content ratio, and upper and lower water content ratio intensities based on the upper and lower water content ratios. (D3) The first linear regression analysis process involves performing a regression analysis with the above unit cement quantity as the explanatory variable and the above peak intensity as the dependent variable to obtain the first linear regression equation. Here, the first linear regression equation is the regression equation for the lower limit of predicted strength with respect to the amount of cement per unit, and the optimal amount of cement per unit is determined using this regression equation. (D4) A regression analysis is performed with the above unit cement amount as the explanatory variable and the above upper and lower limit water content strength as the dependent variable to obtain a second linear regression equation. Then, the value of the target strength is substituted into the second linear regression equation to calculate the unit cement amount to be described in the specified mix design. The reason for substituting the target compressive strength value into the second linear regression equation is to determine the optimal moisture content when the optimal amount of cement is added, based on the relationship between the unit cement amount and the peak moisture content. (D5) After performing a regression analysis with the above unit cement quantity as the explanatory variable and the above peak intensity water content as the dependent variable to obtain a third linear regression equation, the value of the unit cement quantity described in the above-mentioned mix design is substituted into the third linear regression equation to calculate the value of the peak intensity water content, and this value of the peak intensity water content is set as the design water content described in the mix design. Note that the term "moisture content" is used for the test moisture content until the specified mix design is finalized, and then it is called the "design moisture content" after the specified mix design is adopted. [4] The soil cement mix design method described in [3] above, wherein the above quadratic and primary regression equations are equations (1) below. y=ax 2 +bx+c ···(1) However, in equation (1), x is the explanatory variable, y is the dependent variable, and a, b, and c are the regression coefficients. In particular, when x is the test moisture content and y is the long-term strength, a < 0; when x is the unit cement content and y is the peak strength or upper / lower limit moisture content strength; and when x is the unit cement content and y is the peak strength moisture content ratio, a = 0. [5] The soil cement mix design method described in [1] above, wherein the soil type is one or more selected from the following: particle size obtained from a soil particle size test, density obtained from a soil particle density test, oven-dry specific gravity and water absorption rate obtained from a coarse aggregate density and water absorption rate test, oven-dry specific gravity and water absorption rate obtained from a fine aggregate density and water absorption rate test, maximum dry density and optimal water content obtained from a soil compaction test by compaction, water content obtained from a soil natural water content test, and soil color obtained from an organic impurity test. [6] The soil cement mix design method described in [1] above, wherein the short-term strength is the compressive strength before 7 days of age, and the long-term strength is the compressive strength after 28 days of age. [7] (i) a soil and sand collection step, (ii) a material testing step using the soil and sand, (iii) a setting step of a specified mix design using the mix design method described in any of [1] to [6] above, and (iv) a test construction step using the specified mix, Furthermore, as an optional step, (v) if the target strength is not met in the above test construction, a soil cement construction method including a step of resetting the specified mix design using the above mix design method. [Effects of the Invention]

[0015] The present invention's mix design method allows for the easy and accurate setting or resetting of the specified mix using artificial intelligence. Furthermore, since the soil-cement construction method of the present invention includes a step of setting or resetting the specified mix using the above mix design method, the soil-cement construction method can be implemented quickly. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing the relationship between the test water content of soil and the compressive strength of soil cement. [Figure 2] This graph plots the measured water content of soils with different soil types and the measured compressive strength of soil cement at 28 days of age, categorized by unit cement content (C-80 to C-200). The soil types, from left to right in the quadratic curve group, are those with high suitability for use as soil cement (e.g., crushed stone), those with moderate suitability (e.g., sandy soil), and those with low suitability (e.g., cohesive soil). The higher the suitability of the soil and the higher the unit cement content, the larger the regression coefficient (absolute value) of the quadratic term, resulting in a more pointed shape of the quadratic curve. The horizontal line represents the target strength (mix strength, 4.5 N / mm2) in the laboratory mix design test. [Figure 3] This graph plots the measured water content of the test material and the measured compressive strength of soil cement at 28 days of age, categorized by three different unit cement amounts. The unit cement amounts for the soil cement are 50 kg / m³, 80 kg / m³, and 110 kg / m³, respectively, from bottom to top on the quadratic curve. The horizontal line represents the target strength (mix strength, 3.0 N / mm²) at the site. [Figure 4]Figure 3 is a graph plotting the unit cement content, peak strength (○), and upper and lower water content strength (●). The unit cement content indicated by the intersection of the target strength at the site (horizontal line, 3.0 N / mm2) and the dashed line is the optimal unit cement content (131.9 kg / m3) for the specified mix design. [Figure 5] Figure 3 is a graph plotting the unit cement content against the peak strength water content. The peak strength water content shown for the optimal cement content (131.9 kg / m3) corresponds to the optimal test water content (15.9%) for the specified mix design. [Modes for carrying out the invention]

[0017] The soil cement mix design method of the present invention includes at least the steps of (A) implementing a trained artificial intelligence, (B) outputting lead mix examples, (C) measuring the physical properties of a test specimen, and (D) setting the specified mix. Furthermore, the soil cement construction method of the present invention includes at least the steps of (i) collecting soil and sand, (ii) conducting material tests using the soil and sand, (iii) setting a specified mix design using the above mix design method, and (v) conducting a test construction using the specified mix design. The soil cement mix design method and the soil cement construction method of the present invention will be described in detail below.

[0018] 1. Mix Design Method for Soil Cement (A) Implementation process of trained artificial intelligence Step (A) described above is a process of training artificial intelligence using training data that includes at least the soil type, cement type, unit cement amount, test water content, unit volume mass of soil cement, compressive strength, and mixing ratio of soil improvement material into the soil, and then implementing the trained artificial intelligence.

[0019] The above soil type is one or more selected from the following, as shown in Table 1, etc.: particle size obtained from soil particle size distribution tests, density obtained from soil particle density tests, density and absolute dry specific gravity and water absorption rate obtained from coarse aggregate density and water absorption rate tests, density and absolute dry specific gravity and water absorption rate obtained from fine aggregate density and water absorption rate tests, maximum dry density and optimal water content obtained from soil compaction tests by compaction, water content obtained from natural soil water content tests, and soil color obtained from organic impurity tests.

[0020] Furthermore, the type of cement mentioned above is not particularly limited and may be one or more selected from blast furnace cement type A, blast furnace cement type B, blast furnace cement type C, ordinary Portland cement, silica cement, fly ash cement, eco-cement, and cement-based solidifying agents. Here, the cement-based solidifying agent is a composite material with cement as the base material, and the other solidifying components and their proportions are determined according to the difficulty of solidification, such as for general soft soil, special soil (general-purpose type), and high-organic soil, as well as the conditions of the construction site. For example, commercially available cement-based solidifying agents include GeoSet (registered trademark, manufactured by Taiheiyo Cement Corporation) 200 for general-purpose use, and GeoSet 225 (manufactured by Taiheiyo Cement Corporation) and Youth Stabilizer (manufactured by UBE Mitsubishi Corporation) for high-organic soil use. Considering cost and solidification performance, the above solidifying agent is preferably blast furnace cement type B and a cement-based solidifying agent.

[0021] (B) Output process for lead formulation example Step (B) described above is a step in which the trained artificial intelligence is given at least the soil type of the soil at the construction site, the type of cement to be used in the soil cement, the unit cement amount, and the test moisture content, and the trained artificial intelligence outputs a lead mix example that includes at least the unit cement amount and the test moisture content.

[0022] (C) Measurement process for the physical properties of the test specimen Step (C) above is a step in which a soil cement test specimen is prepared based on the lead mix example output above, and the short-term strength and mass of the test specimen are measured at least. The strength measurement should be carried out in accordance with JIS A 1108-2006 "Test method for compressive strength of concrete". Here, the short-term strength mentioned above is the compressive strength of soil cement before 7 days of age. In the field of soil cement, the compressive strength at 7 days of age is generally used as the short-term strength. However, if the long-term strength can be predicted based on the strength at a shorter age, the short-term strength of the specimen can be obtained in a shorter period of time. Therefore, as long as the prediction accuracy can be maintained at a high level, the shorter the age of the short-term material, the better.

[0023] (D) Setting process for the indicated formulation Step (D) described above involves inputting the measured physical properties of the test specimen into the trained artificial intelligence, which then predicts at least the unit cement content, test water content ratio, and long-term strength of the soil cement and outputs mix design examples. From these examples, a mix design example that satisfies the target strength is selected and set as the recommended mix design. Here, the long-term strength mentioned above refers to the compressive strength of soil cement after 28 days of age. However, in the field of soil cement, the compressive strength at 28 days of age is generally used as the long-term strength. Furthermore, if the long-term strength of the specimen prepared based on the specified mix design set in step (D) above does not meet the target strength, the process may optionally include a step of repeating steps (B) to (D) above to reset the specified mix design. By performing this optional step, a predicted strength exceeding the target strength can be easily obtained simply by changing the unit cement amount and the test water content for input and output. Therefore, the present invention, which utilizes artificial intelligence, eliminates the need for the conventional 28-day compressive strength test, which was time-consuming and laborious, and consequently makes mix design quicker and simpler.

[0024] The above (D) process is, more specifically, a process that includes at least the following (D1) to (D5) processes. That is, (D1) Quadratic regression analysis process The above (D1) process involves performing a regression analysis for each unit cement quantity, using the above-mentioned test water content as the explanatory variable and the above-mentioned predicted long-term strength as the dependent variable, and obtaining a quadratic regression equation for each unit cement quantity. Figures 2 and 3 show a graph of the (D1) process.

[0025] (D2) Prediction process for peak intensity, etc. The above (D2) process is a process of calculating and predicting peak intensity, peak intensity moisture content, upper and lower moisture content limits, and upper and lower moisture content limits based on the upper and lower moisture content limits, using the above quadratic regression equation. Incidentally, since the above quadratic regression equation is symmetrical with respect to the peak intensity moisture content, the compressive strength on the quadratic curve shown by the upper moisture content limit and the compressive strength on the quadratic curve shown by the lower moisture content limit are the same value, and there is only one value for the upper and lower moisture content limits based on the upper and lower moisture content limits. Note that the above upper and lower moisture content limits can be substituted with the peak intensity moisture content ±2~3%. The reason for substituting the upper and lower moisture content limits with the peak intensity moisture content ±2~3% is that it is difficult to manage construction using a single value for the peak intensity moisture content, so a certain range is provided. Figure 1 shows a conceptual diagram of the (D2) process. Incidentally, the peak intensity (maximum value) can be obtained by differentiating the above quadratic regression equation. Here, the upper and lower moisture content limits are substituted with the peak intensity moisture content ±2-3% because it is difficult to manage construction using a single value for the peak intensity moisture content, thus allowing for a certain range.

[0026] (D3) First linear regression analysis process The above (D3) process involves performing a regression analysis with the above unit cement quantity as the explanatory variable and the above peak intensity as the dependent variable to obtain the first linear regression equation.

[0027] (D4) Calculation process of unit cement quantity The above (D4) process involves performing a regression analysis with the above unit cement quantity as the explanatory variable and the above upper and lower limit water content strength as the dependent variable to obtain a second linear regression equation, and then substituting the target strength value into the second linear regression equation to calculate the unit cement quantity value to be described in the specified mix design.

[0028] (D5) Calculation process of the design water content The above (D5) process involves performing a regression analysis with the above unit cement quantity as the explanatory variable and the above peak strength water content as the dependent variable to obtain a third linear regression equation. Then, the value of the unit cement quantity described in the specified mix design is substituted into the third linear regression equation to calculate the value of the peak strength water content, and this value of the peak strength water content is set as the design water content described in the specified mix design.

[0029] The above quadratic and linear regression equations are given by equation (1) below. y=ax 2 +bx+c ···(1) However, in equation (1), x is the explanatory variable, y is the dependent variable, and a, b, and c are the regression coefficients. In particular, when x is the test moisture content and y is the long-term strength, a < 0; when x is the unit cement content and y is the peak strength or upper / lower limit moisture content strength; and when x is the unit cement content and y is the peak strength moisture content ratio, a = 0.

[0030] 2. Soil-cement construction method Next, the soil cement construction method of the present invention will be described. The soil cement construction method of the present invention comprises at least the steps of (i) collecting soil and sand, (ii) conducting material testing using the soil and sand, (iii) setting a target mix using the mix design method described in any of [1] to [6] above, and (iv) conducting a test construction using the target mix, and further comprising as an optional step, (v) if the target strength is not achieved in the test construction, a step of resetting (correcting) the target mix using the mix design method above. The following provides a detailed explanation of steps (i) through (v) above.

[0031] (i) Soil extraction process The depth to which soil can be excavated and the amount of soil that can be extracted are investigated to understand the variability of the soil. If there is a large variation in the soil, a decision is made on whether to separate and extract the soil, or to mix separately extracted soil, before extracting the soil.

[0032] (ii) Material testing process using the soil and sand The material testing items are those related to soil properties that affect the physical properties of soil cement, such as particle size, density, water absorption rate, compaction properties, water content, and organic impurities. These test items should be tested in accordance with the test methods of Japanese Industrial Standards (JIS) etc., as specified in Table 1, etc., which define each item.

[0033] (iii) Setting of the specified formulation The specified formulation is set using the formulation design method of the present invention.

[0034] (iv) Test construction process Test specimens will be prepared based on the specified formulation described above, and their physical properties, such as long-term strength, will be measured. As mentioned above, the soil used in the indoor mix design test and the soil sampled during the construction phase may differ in soil type. Therefore, to confirm the effect of these soil type differences on the quality of the soil cement, a trial construction will be carried out at the construction site based on the specified mix design obtained in the indoor mix design test.

[0035] (v) Process of resetting the indicated formulation If the soil cement specimens do not meet the required quality as a result of the above test construction, the specified mix design will be reset using the mix design method of the present invention. [Examples]

[0036] The present invention will be described below with reference to examples, but the present invention is not limited to these examples.

[0037] (A) Implementation of pre-trained artificial intelligence The soil properties used in soil cement (each item of the material test shown in Tables 1 and 2), cement type, unit cement content, test water content, and unit volume mass (based on the mass of the test specimen, the volume of the test specimen is measured in 1 m³) have been accumulated to date. 3The AI ​​was trained using training data that included the mass (converted to a specific value), compressive strength at 7 and 28 days of age, and the mixing ratio of soil improvement material to the soil. The material test methods in Table 2 are the same as those described in Table 1.

[0038] [Table 1]

[0039] [Table 2]

[0040] (B) Output of lead formulation example Next, the soil properties, cement type, unit cement content, and test water content of the soil (hereinafter referred to as "sample soil") from the construction site (alluvial fan) measured using the test methods shown in Tables 1 and 2 were input into the trained artificial intelligence. As a result, the trained artificial intelligence output lead mix design examples (2a, 2b, and 2c) consisting of the unit cement content, test water content, and unit volume mass shown in Table 3. The lead mix design examples are shown in Table 3.

[0041] [Table 3]

[0042] (C) Measurement of physical properties of the test specimen Based on the lead mix examples (2a, 2b, and 2c) shown in Table 3, the above soil samples, which had been air-dried in a room at 20°C for 18 hours, a cement-based solidifying agent (product name: Blast Furnace Cement Type B, manufactured by Taiheiyo Cement Co., Ltd.), and water were mixed in a pan-type mixer to obtain the above three types of soil cement mixtures. Next, the mixture is poured into a mold and molded, then sealed and demolded after 7 days. Three specimens with a diameter of 125 mm and a height of 250 mm are prepared for each formulation example. After measuring the mass of each specimen, the volume of the specimen is calculated based on its mass. 3The mass was calculated by converting it to a specific value, and then the unit volume mass was determined. Furthermore, the compressive strength of the above specimens at 7 days of age was measured in accordance with JIS A 1108-2006 "Test Method for Compressive Strength of Concrete". The above unit volume mass and compressive strength are shown in Table 3. However, the above unit volume mass and compressive strength are the average values ​​of 3 specimens for each mix design example.

[0043] (D) Setting process for the indicated formulation The unit cement content, test moisture content, unit volume mass, and compressive strength at 7 days of age for each of the mix design examples (2a, 2b, and 2c) listed in Table 3 were input into the trained artificial intelligence. As a result, the trained artificial intelligence output the unit cement content, test moisture content, unit volume mass, and compressive strength at 28 days of age for mix design examples 1a-1c and 3a-3c in Table 4, and the test moisture content, unit volume mass, and compressive strength at 28 days of age for mix design examples 2a-2c. These output results are shown in Table 4.

[0044] [Table 4]

[0045] (D1) Quadratic regression analysis process For each of the unit cement quantities listed in Table 4, regression analysis was performed with the above-mentioned test moisture content as the explanatory variable and the predicted compressive strength at 28 days of age as the dependent variable, and a quadratic regression equation was obtained for each unit cement quantity. The results are graphed and shown in Figure 3. Incidentally, the unit cement quantities for the white triangles (△), white circles (〇), and white squares (□) are 50 kg / m³, respectively. 3 , 80 kg / m 3 , and 110 kg / m 3 That is the case.

[0046] (D2) Prediction process for peak intensity, etc. Next, using the above quadratic regression equation, we predicted the peak intensity, peak intensity water content ratio, upper and lower water content ratio limits, and the upper and lower water content ratio intensities based on the upper and lower water content ratio limits.

[0047] (D3) First primary regression analysis process Regression analysis was performed with the above unit cement content as the explanatory variable and the above peak strength as the objective variable, to obtain a first primary regression equation. As for the upper and lower limit water content ratio strength, the strength at a water content ratio of ±3% of the peak strength water content ratio was adopted.

[0048] (D4) Calculation process of unit cement content Regression analysis was performed with the above unit cement content as the explanatory variable and the above upper and lower limit water content ratio strength as the objective variable to obtain a second primary regression equation, then the value of target strength (3.0N / mm 2 ) was substituted into this second primary regression equation, to calculate the value of the unit cement content (103.3kg / m 3 ) to be described in the specified mix proportion.

[0049] (D5) Calculation process of design water content ratio Regression analysis was performed with the above unit cement content as the explanatory variable and the above peak strength water content ratio as the objective variable to obtain a third primary regression equation, then the value of the unit cement content (103.3kg / m 3 ) described in the above specified mix proportion was substituted into this third primary regression equation, to calculate the value of the peak strength water content ratio (10.8%), and the value of this peak strength water content ratio was taken as the design water content ratio to be described in the specified mix proportion, to set the specified mix proportion. The specified mix proportion obtained as described above is shown in Table 5. Note that the unit cement content is a value rounded at 5.0kg / m 3 (for example, 103.3kg / m 3 is 105kg / m 3 ), and the design water content ratio is also a value rounded to 0.1% (for example, 10.81% is 10.8%, and 10.85% is 10.9%). Further, in accordance with the specified mix proportion in Table 5, soil cement specimens were prepared, and their compressive strength at the material age of 28 days was measured in accordance with JIS A 1108-2006 "Method of test for compressive strength of concrete". As a result, the compressive strength at the material age of 28 days exceeded the above target site strength of 1.5N / mm 2 , with the minimum strength being 2.7N / mm 2 , and the maximum strength being 3.81N / mm 2 .

[0050] Table 5

Claims

1. A method for designing the mix of soil cement, comprising at least the following steps (A) to (D). (A) Implementation process of a trained artificial intelligence, which involves using training data that includes at least the soil type, cement type, unit cement amount, test water content, unit volume mass of the soil cement, compressive strength, and mixing ratio of soil improvement material to soil, to train the artificial intelligence. (B) Output process for lead mix design, in which the trained artificial intelligence is given at least the soil type of the soil at the construction site, the type of cement to be used in the soil cement, the unit cement amount, and the test moisture content, and the trained artificial intelligence outputs a lead mix design example (a mix design example that serves as a starting point for deriving (leading) the specified mix design) that includes at least the unit cement amount and the test moisture content. (C) A step to measure the physical properties of a specimen, in which a soil cement specimen is prepared based on the lead mix example output above, and the short-term strength and mass of the specimen are measured. (D) Setting of a specified mix: The physical properties of the specimen measured above are input into the trained artificial intelligence, which then predicts at least the unit cement content, test water content ratio, and long-term strength of the soil cement and outputs mix design examples. From these mix design examples, a mix design that satisfies the target long-term strength is set as the specified mix design.

2. (E) If the long-term strength of the specimen prepared based on the specified mix design set in step (D) above does not meet the target long-term strength, the method for designing the soil cement mix according to claim 1 further includes as an optional step a step of resetting the specified mix design, which involves repeating steps (B) to (D) above to reset the specified mix design.

3. The soil cement mix design method according to claim 1, wherein the above step (D) includes at least the following steps (D1) to (D5). (D1) For each unit amount of cement, a regression analysis is performed with the above test water content as the explanatory variable and the above predicted long-term strength as the dependent variable, and a quadratic regression equation is obtained for each unit amount of cement. (D2) A prediction process for peak intensity, etc., which uses the above quadratic regression equation to calculate and predict peak intensity, peak intensity water content ratio, upper water content ratio and lower water content ratio, and upper and lower water content ratio intensities based on the upper and lower water content ratios. (D3) The first linear regression analysis process, in which regression analysis is performed with the above unit cement quantity as the explanatory variable and the above peak intensity as the dependent variable to obtain the first linear regression equation. (D4) A regression analysis is performed with the above unit cement amount as the explanatory variable and the above upper and lower limit water content strength as the dependent variable to obtain a second linear regression equation. Then, the value of the target strength is substituted into the second linear regression equation to calculate the unit cement amount to be described in the specified mix design. (D5) After performing a regression analysis with the above unit cement quantity as the explanatory variable and the above peak intensity water content as the dependent variable to obtain a third linear regression equation, the value of the peak intensity water content is calculated by substituting the value of the unit cement quantity described in the above specified mix design into the third linear regression equation, and this value of the peak intensity water content is set as the design water content described in the specified mix design.

4. The soil cement mix design method according to claim 3, wherein the above-mentioned quadratic and primary regression equations are the following equations (1). y=ax ２ +bx+c ・・・(1) However, in equation (1), x is the explanatory variable, y is the dependent variable, and a, b, and c are the regression coefficients. In particular, when x is the test moisture content and y is the long-term strength, a < 0; when x is the unit cement amount and y is the peak strength or upper / lower limit moisture content strength; and when x is the unit cement amount and y is the peak strength moisture content ratio, a = 0.

5. The soil type design method for soil cement according to claim 1, wherein the soil type is one or more selected from the following: particle size obtained from a soil particle size test, density obtained from a soil particle density test, absolute dry specific gravity and water absorption rate obtained from a coarse aggregate density and water absorption rate test, absolute dry specific gravity and water absorption rate obtained from a fine aggregate density and water absorption rate test, maximum dry density and optimal water content obtained from a soil compaction test by compaction, water content obtained from a soil natural water content test, and soil color tone obtained from an organic impurity test.

6. The soil cement mix design method according to claim 1, wherein the short-term strength is the compressive strength before 7 days of age, and the long-term strength is the compressive strength after 28 days of age.

7. (i) a soil and sand collection step, (ii) a material testing step using the soil and sand, (iii) a setting step of a specified mix design using the mix design method described in any one of claims 1 to 6, and (iv) a test construction step using the specified mix, Furthermore, as an optional step, (v) if the target strength is not met in the above test construction, a soil cement construction method including a step of resetting the specified mix design using the above mix design method.

Images