JPH1151933A - Processing method for measured data on concrete property - Google Patents

Abstract

PROBLEM TO BE SOLVED: To specify the property or ultimate value of concrete with statistical and objective reliability by regressively analyzing the measured data on the property of concrete with a specific equation used as a model equation. SOLUTION: A regression curve is slotted from the measured data on the property of concrete with the equation Y=(A1-A2)/ 1+exp[(X-X0)/dx]}+A2 used as a model equation, where X0: center, dx: width, A1: Y initial value, A2: Y final value. The property or ultimate value of concrete is specified from the regression curve or the equation of the regression curve. When the measured data are processed, the property or ultimate value of the concrete can be specified with statistical and objective reliability even if the measured data are not the data to the ultimate value or from the somewhat irregular measured data containing an error. At least ten measured data are preferably used to give high reliability to the specified value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to measurement data relating to the properties of concrete, for example, data obtained by measuring rebound hardness or compressive strength with respect to the depth from the surface of a concrete structure exposed to harmful substances that corrode concrete, or Drying shrinkage strain for concrete age,
The present invention relates to a method for processing data obtained by measuring a neutralization depth or an adiabatic temperature rise amount.

[0002]

2. Description of the Related Art For example, in a concrete structure exposed to an aqueous solution containing a harmful substance that corrodes concrete, the harmful substance diffuses and penetrates from the surface, and the concrete on the surface layer is deteriorated and the strength is reduced. I do. The rate of the strength decrease is larger at the surface, and the rate of the strength decrease becomes smaller toward the inside, and the concrete strength becomes sound from a certain depth. The depth up to the sound concrete strength is called the "surface weakening depth", and the properties of concrete such as the surface weakening depth depend on the type of concrete used and the concrete structure exposed. Since it greatly depends on the type of harmful substances used and the age of the concrete structure, it is necessary to collect data on the concrete structure itself that was the subject of the survey.

[0003] Further, the drying shrinkage strain of concrete after production greatly differs from concrete to concrete depending on the type of cement used, the composition of the concrete, the properties of the aggregate used, and the water-cement ratio. Therefore, in order to accurately grasp the drying shrinkage strain of the produced concrete, it is necessary to collect data on the concrete itself. This is also true for concrete properties other than drying shrinkage strain, such as increasing the adiabatic temperature of concrete, neutralization, creep strain, strength development, heat of hydration or hydraulic conductivity, etc. This can be said in grasping the breathing of concrete, the penetration resistance of the proctor or the slump loss.

Here, in order to collect data on the properties of concrete as described above, it is often necessary to take a very long measurement period. For example, in order to grasp the final value of the drying shrinkage strain of concrete, the final value cannot be obtained unless the amount of strain of the concrete is measured at least for two to three years after the production of the concrete. The measurement data thus obtained also includes errors due to fluctuations in measurement conditions, irregularities in measuring equipment, variations in the capabilities of the measurer, and the like. Therefore, directly specifying the properties or final values of concrete from the measured data itself lacks reliability and objectivity.

Therefore, an object of the present invention is to provide a method for statistically and objectively determining the properties or the final value of concrete even if the measured data does not reach the final value, or from the measured data having some variation including an error. It is an object of the present invention to provide a method of processing measured data on concrete properties that can be reliably specified.

[0006]

In order to achieve the above-mentioned object, the present inventors have measured data on the properties of concrete, for example, the depth from the surface of a concrete structure whose surface is weakened. The data obtained by measuring the rebound hardness or compressive strength against, or the data obtained by measuring the dry shrinkage strain, the neutralization depth, or the adiabatic temperature rise against the concrete age, can be taken, that is,
Like the spark voltage and ignition rate of a car engine,
When the measurement data on the properties of concrete is plotted on a scatter plot, such as in a logistic model that often matches the measurement data such as the proportion of individuals where X responds to a certain stimulus, the tendency of the data to line up accurately If a model formula that can be expressed can be found, regression analysis of the measured data on the properties of concrete is performed using the model formula to derive a regression curve or the formula of the regression curve. Alternatively, if the final value is specified, tests and studies were repeated based on the idea that the specified property or final value of the concrete would have statistical objective reliability.

As a result, the measured data on the properties of concrete is as follows: Y = (A1−A2) / {1 + exp [(X−X0) / dx]} +
When a regression curve is drawn using an equation expressed as A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value), the regression curve improves the tendency of the measured data. The present invention was completed, and the present invention was completed.

That is, according to the present invention, the measured data on the properties of concrete is calculated as follows: Y = (A1−A2) / {1 + exp [(X−X0) / dx]} +
A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value) is subjected to regression analysis as a model expression. Processing method.

Further, the present invention is based on measurement data on the properties of concrete: Y = (A1−A2) / {1 + exp [(X−X0) / dx]} +
Draw a regression curve using the expression expressed as A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value) as a model expression,
A method for processing measurement data relating to the property of concrete, characterized by specifying the property or final value of concrete from the regression curve or the equation of the regression curve.

If the measurement data relating to the properties of the concrete according to the present invention described above is processed, the concrete data can be obtained not only from the measurement data up to the final value, but also from the measurement data having some variation including an error. The property or the end value can be specified with statistical objective reliability. In order to make the specific value highly reliable, it is preferable that the measured data have at least 10 or more.

Here, Y = (A1-A2) / {1 + exp [(X-X0) / dx]} +
When the formula of A2 is modified, (X−X0) / dx = ln [(Y−A1) / (A2−Y)]. Here, X−X0 = S, (Y−A1) / (A2−Y)
= W, dx is moved to the right side, and dx = k,
The above equation becomes S = kInW or S = klogW. Since this equation is a basic equation based on Boltzmann's principle, the following equation is used: Y = (A1−A2) / {1 + exp [(X−X0) / dx]} +
A regression curve or a regression equation obtained by using the equation expressed as A2 as a model equation may be referred to as a regression curve or a regression equation by Boltzmann regression.

As the measurement data on the properties of concrete suitable for being treated according to the present invention, the rebound hardness or the compressive strength with respect to the depth from the surface of the concrete structure whose surface is weakened was measured. Data or data obtained by measuring the drying shrinkage strain, the neutralization depth, or the adiabatic temperature rise with respect to the concrete age. Further, to further describe, data obtained by measuring creep strain, strength development, heat of hydration, or hydraulic conductivity with respect to the age of concrete, and concrete breathing, proctor penetration resistance, or slump loss over a relatively short period of time. Is measured. This is because the properties of concrete as described above reflect phenomena that proceed according to the second law of thermodynamics, that is, reflection of phenomena such as dissipation of matter, diffusion of matter, change in order, change in viscosity, or dissipation of heat. Because it is possible.

[0013]

Test Examples Hereinafter, test examples in which the above-described data processing method according to the present invention has been found will be described.

Test Example 1 A core was sampled from a concrete structure having a weakened surface by boring, and the rebound hardness of the upper surface of the core specimen was measured with a Schmid hammer. Next, the core specimen was scraped off from the upper surface thereof using a polishing machine of several mm, and the rebound hardness of the newly polished surface was measured again with a Schmidt hammer. Thereafter, the above-mentioned polishing and the measurement of the rebound hardness were repeated until the depth at which the polishing was removed was considered to have sufficiently exceeded the weakening depth, and data showing the change in the rebound hardness with respect to the depth from the surface were obtained. Table 1 shows the measured data.

[0015]

[Table 1]

Next, the measured data is plotted on a scatter diagram, and Y = (A1−A2) / {1 + exp [(X−X0) / dx]} +
A regression curve was drawn using an expression expressed as A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value) as a model expression. The plot of this regression curve was made by Microcal,
Using ORIGIN's regression analysis software, X
To the depth data from the surface, and to Y the measured data of the rebound hardness corresponding to the depth from the surface, respectively, and the constants in the above formula determined by the data, ie, X0: center,
When dx: width, A1: Y initial value, A2: Y final value were calculated, X0 = −0.111, dx = 1.89, A1 = −0.977, A2 = 39.2.
Met. Therefore, the equation of the regression curve drawn on the scatter diagram is: Y = 38.2 / {1 + exp [(X + 0.111) /1.89]} +
39.2. FIG. 1 shows a scatter plot in which a regression curve is plotted.

According to the scatter plot shown in FIG. 1, the regression curve by Boltzmann regression has a tendency to take a certain degree of variation data including a measurement error and the like, such as measurement data indicating a change in rebound hardness with respect to the depth from the surface. It turned out that it became the curve shown well.

Test Example 2 Concrete using ordinary Portland cement (hereinafter, "NC")
Data obtained by measuring the drying shrinkage strain with respect to the material age performed in the past were plotted on a scatter diagram, and a regression curve by Boltzmann regression was drawn on the scatter diagram using the same regression analysis software as in Test Example-1. . FIG. 2 shows a scatter diagram in which this regression curve is plotted. Further, an accelerated neutralization test was similarly performed on concrete using NC, and data obtained by measuring the neutralization depth with respect to the material age were obtained. The data was plotted on a scatter diagram, and a regression curve by Boltzmann regression was drawn on the scatter diagram using the same regression analysis software as in Test Example-1. FIG. 3 shows a scatter diagram in which this regression curve is plotted.

From the scatter plots shown in FIGS. 2 and 3, the regression curve by Boltzmann regression is a curve that clearly shows the tendency indicated by the data obtained by measuring the drying shrinkage strain or the neutralization depth with respect to the age. I understood. In each figure, the final value of each concrete calculated from the equation of the obtained regression curve is described.

Test Example 3 For concrete using NC and Class B blast furnace cement (hereinafter, referred to as "BB"), the data obtained by measuring the adiabatic temperature rise with respect to the material age were plotted on a scatter diagram. Test example on scatter diagram
A regression curve by Boltzmann regression was drawn using the same regression analysis software as in 1. FIG. 4 shows a scatter diagram in which this regression curve is plotted. Also, the adiabatic temperature rise amount is calculated from a regression curve obtained by performing a regression analysis on the measurement data using the following three formulas (conventional formula 1, conventional formula 2 and conventional formula 3) proposed conventionally. Table 2 shows the result of calculating the final value of.

T = K [1-exp (-at)] {conventional formula 1 T = K [1- (1 + at) exp (-at)] {conventional formula 2 T = K [1-exp ( −at ^b )] {Conventional formula 3 (where, T: adiabatic temperature rise, K: T final value, t: material age, a, b: constant)

[0022]

[Table 2]

It has been known that the regression curves drawn by the above three formulas (conventional formula 1, conventional formula 2 and conventional formula 3), which have been proposed in the past, are slightly shifted from the plotted points of the data. From the scatter plot shown in FIG. 4, it was found that the regression curve by Boltzmann regression almost completely coincided with the plotted points of the data. Also, from Table 2, a regression equation by Boltzmann regression (in Table 2, "Boltzmann equation"
It was described. ), The value of the ultimate adiabatic temperature rise was found to be very close to the average of the ultimate adiabatic temperature rise obtained from each of the three conventional formulas.

[Test Example 4] As a representative of the low heat-generating concrete, for a concrete using belite cement (hereinafter, referred to as "LC"), data obtained by measuring the adiabatic temperature rise with respect to the material age are plotted on a scatter diagram. On this scatter diagram, a regression curve by Boltzmann regression was drawn using the same regression analysis software as in Test Example 1 using all the measured data. FIG. 5 shows a scatter plot in which this regression curve is plotted.
Shown in

From the scatter plot shown in FIG. 5, the regression curve by Boltzmann regression is a curve that is considered to have a stagnant temperature rise on the age of 15 days even though the temperature is still rising. It turned out not to match the point.
Such a tendency was also observed when the above-mentioned three conventional formulas were used.

Therefore, in concrete using LC, the reaction of alite which generates heat at the beginning of hydration and the reaction of belite which generates heat after hydration alternate in 3 to 4 days. Considering that the reaction is exchanged on the way, the data obtained by measuring the adiabatic temperature rise for the above material age is divided into 4 days of material age when the reaction is considered to be exchanged, and a test is performed using each of the divided measurement data. Regression curves by Boltzmann regression were drawn using the same regression analysis software as in Example-1. FIG. 6 shows a scatter plot in which this regression curve is plotted.

From the scatter plot shown in FIG. 6, it was found that the regression curve by Boltzmann regression created by dividing the data almost completely coincides with the plotted points of the data.
It has been known that a regression curve using a conventional formula cannot be matched with a plot point even by using such a dividing means.

[Test Example-5] The data of the adiabatic temperature rise with respect to the material age of the concrete using BB measured in the above Test Example-3 is shown in Table 3, and the data up to the final use material age in the calculation shown in Table 3. Test example using each
Regression formulas by Boltzmann regression were respectively derived using the same regression analysis software as in 1, and the results of calculating the ultimate values of the adiabatic temperature rises by the obtained regression formulas are also shown in Table 3.

[0029]

[Table 3]

From Table 3, the final value of each adiabatic temperature rise calculated from the regression equation by Boltzmann regression after 4.5 days for the final material age used for calculation is the data up to the last measurement date, ie, 10 days of material age. And the final value of the adiabatic temperature rise calculated from the regression equation by Boltzmann regression obtained using
Since there is only a difference within 1 ° C., it was found that the properties of concrete, such as the ultimate value of the adiabatic temperature rise, can be accurately specified even from data of a relatively short measurement period.

From the test examples described above, Y = (A1−A2) / {1 + exp [(X−X0) / dx]} + on the scatter plot in which measured data on the properties of concrete is plotted.
When a regression curve is drawn using an equation expressed as A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value), the regression curve improves the tendency of the measured data. If the concrete property or final value is specified from this regression curve or the equation of this regression curve, the specified concrete property or final value has statistical objective reliability. I understood that.

[0032]

【Example】

[Example-A] Examples in which the surface weakening depths of four concrete water storage tanks that have stored an extremely thin aqueous hydrochloric acid solution for a long period of time were specified by the above-described data processing method according to the present invention are described below. Shown in

Example-A1 Cores (core numbers 1 to 4) were drilled from four concrete water tanks.
And the rebound hardness of the upper surface of this core specimen was measured with a Schmid hammer, and then a few mm of the surface layer was scraped off from the upper surface using a grinder, and the rebound hardness of the newly polished surface was measured using a Schmid hammer. Was measured again. Thereafter, the above polishing and the measurement of the rebound hardness were repeated until the depth at which the polished and removed depth was considered to have sufficiently exceeded the weakening depth, and data showing the change in the rebound hardness with respect to the depth from the surface of each core specimen were obtained. Obtained.

Next, the obtained measured data are plotted on a scatter diagram, and Y = (A1−A2) / {1 + exp [(X−X0) / dx]} +
The regression curves were drawn using the formulas A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value) as model expressions, and the surface weakening depth was specified on the figure. . The plotting of this regression curve was performed in the same manner as in Test Example 1 above, except that the Microca
ORIGIN regression analysis software manufactured by I company was used. The surface weakening depth is specified on the diagram by dividing the regression curve by a straight line connecting the horizontal part of the regression curve (the part that is not theoretically horizontal but becomes approximately horizontal as shown) and the ultimate rebound hardness. The vertical line was lowered from the point where the vertical line crossed, and the depth from the surface at the point where the vertical line intersected the horizontal axis was defined as the surface weakening depth. The scatter diagrams used to determine the surface weakening depth are shown in FIGS. Table 4 shows the surface weakening depth of each of the specified core specimens.

[0035]

[Table 4]

[Example-A2] Data showing the change in the rebound hardness with respect to the depth from the surface of each core specimen measured in the above-mentioned Example-A1 were obtained as follows: Y = (A1-A2) / ｛1 + exp [ (X-X0) / dx]｝ +
A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value) is subjected to regression analysis as a model expression, and each core data is calculated from the obtained regression expression. The surface weakening depth of the specimen was calculated. The calculation of the surface weakening depth from the regression equation is
Since it is difficult to obtain the surface weakening depth directly from the ultimate rebound hardness, an approximate ultimate rebound hardness is introduced as a rebound hardness as close as possible to the ultimate rebound hardness, and this approximate ultimate rebound hardness is set to 99.9% of the ultimate rebound hardness. Then, the surface weakening depth corresponding to the approximate ultimate rebound hardness was calculated. Table 4 also shows the surface weakening depth of each core specimen calculated from the regression equation.

In the above embodiment, as the data for the regression analysis, data obtained by measuring the rebound hardness with respect to the depth from the surface were used. However, the rebound hardness measured by a Schmidt hammer can be converted into the compressive strength of concrete. Therefore, the index for determining the surface weakening depth may be rebound hardness or compressive strength. Further, in the above embodiment, the surface weakening depth was specified by using a core specimen collected from the structure by boring, but regardless of the core specimen, by a method of directly polishing the surface of the target structure. The data of the polishing depth and the rebound hardness may be obtained, and the surface weakening depth may be specified by the above-described method.

[Example B] The ultimate adiabatic temperature rise of concrete using LC and the material age at which the ultimate adiabatic temperature rise was measured were determined by the above-described data processing method according to the present invention for a relatively short measurement period. Examples specified from the above data are shown below.

For concrete using LC, the change in the adiabatic temperature rise was measured from 4 days to 21 days of age. In addition, from the material age of 4 days, the test examples described above show that rehydration curves by Boltzmann regression do not match when the data of this period are included due to the hydration heat generated by the reaction of alite during the first few days. 4 because it was known.

The data of the adiabatic temperature rise amount with respect to the measured material age is expressed as follows: Y = (A1−A2) / {1 + exp [(X−X0) / dx]} +
A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value) is subjected to regression analysis as a model equation, and the ultimate adiabatic temperature rise is calculated from the obtained regression equation. After calculating,
33.6 ° C.

Further, a curve based on the regression equation was drawn by extending the material age on the scatter diagram, and a material age reaching the ultimate adiabatic temperature rise amount was identified from this imaginary line. there were. The specification of the material age reaching the ultimate adiabatic temperature rise on the diagram is obtained by connecting the horizontal part of the imaginary line (the part that is not theoretically horizontal but becomes approximately horizontal in the figure) and the ultimate adiabatic temperature rise. The perpendicular was lowered from the point where the straight line diverged from the imaginary line, and the material age at the point where this perpendicular intersected the horizontal axis was defined as the material age that eventually reached the adiabatic temperature rise. FIG. 11 shows a scatter diagram used for specifying a material age reaching the amount of ultimate adiabatic temperature rise.

[0042]

As described above, according to the method for processing measured data relating to the properties of concrete according to the present invention described above,
Even if it is not the measurement data up to the final value, or from the measurement data with some errors including errors,
There is an effect that the property or the final value of the concrete can be specified with statistical objective reliability.

[Brief description of the drawings]

FIG. 1 is a scatter diagram showing data obtained by measuring the rebound hardness of a concrete structure having a weakened surface with respect to a depth from the surface, and a regression curve by Boltzmann regression on the measured data.

FIG. 2 is a scatter diagram showing data obtained by measuring the drying shrinkage strain of concrete using NC with respect to a material age performed in the past, and a regression curve by Boltzmann regression on the measured data.

FIG. 3 is a scatter diagram showing data obtained by performing an accelerated neutralization test on concrete using NC and measuring the neutralization depth with respect to the age, and drawing a regression curve by Boltzmann regression on the measured data.

FIG. 4 is a scatter diagram showing data obtained by measuring an adiabatic temperature rise amount with respect to material age for concrete using NC and BB, and a regression curve by Boltzmann regression on the measured data.

FIG. 5 is a scatter diagram showing data obtained by measuring the adiabatic temperature rise with respect to the age of concrete using LC, and a regression curve by Boltzmann regression using all the measured data collectively on the measured data. is there.

FIG. 6 shows data obtained by measuring the adiabatic temperature rise with respect to the material age of concrete using LC, and, on the measured data, the material age divided into four days when the reaction is considered to be exchanged, and each of the divided measurements was performed. FIG. 4 is a scatter diagram depicting a regression curve by Boltzmann regression using data.

FIG. 7 is a scatter diagram used to specify the surface weakening depth of core number 1 collected by boring from a concrete water tank that has stored an extremely thin aqueous hydrochloric acid solution for a long period of time.

FIG. 8 is a scatter diagram used for specifying the surface weakening depth of core number 2 collected by boring from a concrete water storage tank that has stored an extremely thin aqueous hydrochloric acid solution for a long period of time.

FIG. 9 is a scatter diagram used to specify the surface weakening depth of core number 3 collected by boring from a concrete water storage tank that has stored an extremely thin aqueous hydrochloric acid solution for a long period of time.

FIG. 10 is a scatter diagram used to specify the surface weakening depth of core number 4 collected by boring from a concrete water tank that has stored an extremely thin aqueous hydrochloric acid solution for a long period of time.

FIG. 11 is a scatter diagram used to specify a material age reaching the ultimate adiabatic temperature rise of concrete using LC.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Satoshi Kajio 1-2-23 Kiyosumi, Koto-ku, Tokyo Inside Japan Research Institute, Inc. (72) Inventor Kazuya Tamagi 1-2-23, Kiyosumi Koto-ku, Tokyo Inside the Japan Consultant Co., Ltd. (72) Inventor, Yoshinori Yamazaki 1-2-23 Kiyosumi, Koto-ku, Tokyo Inside Japan Central Co., Ltd. (72) Inventor, Hisao Abe 1-2-3, Shibaura, Minato-ku, Tokyo Shimizu (72) Inventor Mitsuaki Watanabe 1-2-3 Shibaura, Minato-ku, Tokyo Shimizu Corporation

Claims (4)

[Claims] 1. The measured data on the properties of concrete is expressed as follows: Y = (A1-A2) / {1 + exp [(X-X0) / dx]} +
A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value) is subjected to regression analysis as a model expression. Processing method. 2. From the measurement data on the properties of concrete, Y = (A1-A2) / {1 + exp [(X-X0) / dx]} +
Draw a regression curve using the expression expressed as A2 (where X0: center, dx: width, A1: Y initial value, A2: Y final value) as a model expression,
A method for processing measured data relating to the property of concrete, wherein the property or the final value of the concrete is specified from the regression curve or the equation of the regression curve. 3. The measurement data relating to the property of the concrete is data obtained by measuring a rebound hardness or a compressive strength with respect to a depth from a surface of a concrete structure whose surface is weakened. Or a method for processing measured data relating to the properties of concrete according to 2. 4. The method according to claim 1, wherein the measurement data relating to the property of the concrete is data obtained by measuring a drying shrinkage strain, a neutralization depth, or an adiabatic temperature rise with respect to the age of the concrete. A method for processing measured data relating to the properties of concrete described.