Method for predicting fatigue life of solid waste fiber concrete

Abstract

The invention discloses a method for predicting the fatigue life of solid waste fiber concrete, and belongs to the technical field of concrete life detection. Comprising the following steps: 1, preparing m parts of concrete samples from solid wastes, carrying out alkali excitation on the concrete samples, and measuring and calculating the physical characteristics of the concrete samples; step 2, establishing a test environment, placing the concrete sample in the test environment, and measuring and calculating the properties of the concrete sample according to cycle time, including n indexes, to form a data set; 3, using the data set as calculation data, carrying out data standardization, then carrying out information amount calculation, and inputting the constructed concrete life prediction model; and 4, obtaining the service life of the concrete by means of the concrete service life prediction model. By adopting the method, the concrete service life prediction model is established, the strength of the concrete is predicted by simulating the index change in the environment, and finally the service life of the concrete is predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of concrete data calculation, and in particular to a method for predicting fatigue life of solid waste fiber concrete. Background Art

[0002] As the quality requirements of construction projects continue to increase, concrete, as a common building material, is widely used in projects that have been put into use for a long time. During use, concrete is affected by many factors, such as the environment and wind speed. These factors lead to a reduction in the service life of concrete, which in turn affects the structural integrity and service life of the building. Traditional methods for monitoring the crack resistance of concrete mainly rely on manual inspection and field tests, which have problems of low detection accuracy and low efficiency. Summary of the Invention

[0003] The purpose of this invention is to provide a method for predicting the fatigue life of solid waste fiber concrete, by collecting data of short-cycle concrete samples, establishing a concrete life prediction model to predict the strength change of concrete, and realizing the prediction of concrete life.

[0004] To achieve the above object, the present invention provides a method for predicting fatigue life of solid waste fiber concrete, comprising the following steps:

[0005] Step 1: Use solid waste to make m pieces of concrete samples, perform alkali excitation on the concrete samples, and measure the physical properties of the concrete samples;

[0006] Step 2: Establish a test environment, place the concrete sample in the test environment, and collect indicators of the concrete sample according to the cycle time, including n indicators, to form a data group;

[0007] Step 3: Use the data set as calculation data, first standardize the data, then calculate the amount of information, and input it into the constructed concrete life prediction model;

[0008] Step 4: Use the concrete life prediction model to obtain the service life of concrete.

[0009] Preferably, the specific process of step 1 is as follows:

[0010] Prepare m concrete samples with a specific size of 100mm*100mm*100mm, carry out standard curing, and perform alkali excitation on the concrete samples. Then, measure the physical properties of the concrete samples, including compressive strength, water absorption, and porosity.

[0011] Preferably, the specific process in step 2 is as follows:

[0012] The test environment includes marine environment, hot and humid environment and dry and cold environment. Concrete specimens are tested in the above environments. There are two cycle times, namely positive cycle and reverse cycle. The positive cycle is frequently tested in the early stage of the test, and the reverse cycle is frequently tested in the later stage of the test. The collected indicators include the physical properties in step one, including compressive strength, water absorption rate and porosity. The mass change rate and expansion rate of the concrete specimens are also tested. The above data are formed into data groups in chronological order.

[0013] Preferably, in step 3, the data set collected in step 2 is input into a concrete life prediction model. Here, the concrete life prediction model is specifically set to a LightGBM+CRITIC weighted model. The processing process is as follows:

[0014] First, the CRITIC weight calculation stage is carried out. The positive indicator is set to compressive strength and normalized. The negative indicator is set to water absorption, porosity and expansion rate, and reverse normalized. The cycle time is logarithmically transformed and then normalized. The above indicators are used to calculate the amount of information. The weight distribution result is obtained according to the calculated amount of information. The specific calculation process is as follows:

[0015] Normalize the compressive strength using the following formula:

[0016]

[0017] In the above formula, x' ij is the normalized value of the compressive strength of the i-th sample at the j-th day, x ij represents the compressive strength value of the i-th sample at the j-th day, x j represents the compressive strength value of all concrete specimens on day j;

[0018] The process of denormalizing water absorption, porosity, and swelling is as follows:

[0019]

[0020] In the above formula, a' ij Indicates the normalized value of water absorption, porosity or expansion rate, a ij Indicates the original value of water absorption, porosity or expansion rate corresponding to the i-th sample at the j-th day, a j It represents the water absorption, porosity or expansion rate of the corresponding concrete sample on the jth day;

[0021] The information volume is calculated using the contrast intensity formula, which is as follows:

[0022]

[0023] In the above formula, m is the number of concrete samples, Represents the standardized mean, including the preceding x' ij and a' ij , the formula for calculating the amount of information is as follows:

[0024] C j =σ j ×R j

[0025]

[0026] In the above formula, R j is the conflict of the current indicator, r jk represents the correlation coefficient between the j index and the k index, n represents the total number of indicators, C j represents the comprehensive information carrying capacity of the jth indicator;

[0027] The LightGBM model is constructed and parameter training is performed. The training process uses time series cross-validation and early stopping mechanism for auxiliary training. The weight distribution results obtained by CRITIC weight calculation are used to perform feature weighting processing.

[0028] Preferably, the parameters of the LightGBM model are set as follows, including tree structure parameters, regularization parameters, and training parameters; the tree structure control parameters are as follows: the number of leaf nodes is 30, the maximum depth of the tree is set to 5, and the minimum number of samples of leaf nodes is 20; the regularization parameters are as follows, the regularization coefficient L1 is set to 0.1, the regularization coefficient L2 is set to 0.2, and the classification minimum value-added threshold is set to 0.3; the training parameters are as follows: the learning rate is 0.3, the number of iterations is 300, and the number of early stopping rounds is set to 8.

[0029] 6. The fatigue life prediction method for solid waste fiber concrete according to claim 5 is characterized in that the weight distribution result is calculated as follows:

[0030] The formula for weight distribution is as follows:

[0031]

[0032] In the above formula, w j is the final weight, C k It represents the sum of the comprehensive information of all indicators.

[0033] Therefore, the present invention adopts the above-mentioned method for predicting fatigue life of solid waste fiber concrete, which has the following advantages:

[0034] In the present invention, by setting positive cycle and counter-cycle sampling methods, the degree of change of the physical properties of concrete samples at different time intervals can be collected, thereby realizing the prediction of the change of the concrete sample indicators. By establishing a mapping relationship between the cycle days, indicators and life span, and then through different simulation environments, the prediction of the concrete life span can be finally realized.

[0035] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 The present invention is a flowchart of a method for predicting fatigue life of solid waste fiber concrete. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. The specific model specifications need to be selected and determined based on the actual specifications of the device, etc. The specific selection calculation method adopts the existing technology in this field, so it will not be described in detail.

[0038] Example

[0039] like Figure 1 As shown, the present invention provides a method for predicting fatigue life of solid waste fiber concrete, comprising the following steps:

[0040] Step 1: Use solid waste to make m concrete samples, perform alkali activation on the concrete samples, and measure the physical properties of the concrete samples. Specifically, the concrete samples are made of solid waste materials, including steel slag and fly ash, and steel fiber is added accordingly. The concrete samples are rectangular and have a specific size of 100mm*100mm*100mm. They are cured under standard conditions and, after alkali activation, the physical properties of the concrete samples are measured, including compressive strength, water absorption, and porosity.

[0041] Step 2: Establish a test environment, place the concrete sample in the test environment, and collect indicators of the concrete sample according to the cycle time, including n indicators, to form a data group;

[0042] The test environment includes marine environment, hot and humid environment and dry and cold environment. Concrete specimens are tested in the above environments. There are two cycle times, namely positive cycle and reverse cycle. The positive cycle is frequently tested in the early stage of the test, which is suitable for calibrating concrete specimens with a large degree of change in the early stage of the test. The reverse cycle is frequently tested in the late stage of the test, which is suitable for calibrating concrete specimens with a large degree of change in the middle stage of the test, thereby determining the changes in concrete under different environments; the collection indicators in step two include the physical properties in step one, including compressive strength, water absorption rate and porosity, and will also detect the mass change rate and volume expansion rate of the concrete specimens, so as to better quantify the concrete specimens, and form data groups in chronological order.

[0043] Step 3: Use the data set as calculation data, perform data standardization, calculate the amount of information, and input it into the constructed concrete life prediction model;

[0044] The data set collected in step 2 is input into the concrete life prediction model. Here, the concrete life prediction model is specifically set to the LightGBM+CRITIC weighted model. The processing process is as follows:

[0045] First, the CRITIC weight calculation stage is carried out. The positive indicator is set to compressive strength and normalized using min-max. The reverse indicator is set to water absorption, porosity, and expansion rate, and reverse normalized. The cycle time is logarithmically transformed and then normalized. The above indicators are used to calculate the amount of information, and the weight distribution result is obtained according to the calculated amount of information. The specific calculation process is as follows:

[0046] The formula for the process of normalizing compressive strength is as follows:

[0047]

[0048] In the above formula, x' ij is the normalized value of compressive strength, x ij represents the compressive strength value of the i-th sample at the j-th day, x j represents the compressive strength value of all concrete specimens on day j;

[0049] The process of denormalizing water absorption, porosity, and swelling is as follows:

[0050]

[0051] In the above formula, a' ij represents the normalized value of water absorption, porosity and expansion rate, a ij Indicates the original value of water absorption, porosity or expansion rate corresponding to the i-th sample at the j-th day, a jIt represents the water absorption, porosity or expansion rate of the corresponding concrete sample on the jth day;

[0052] In the above formula, x' ij and a' ij All of them are indicators of concrete samples, using T' ij Indicates that the information volume is calculated using indicators and the contrast intensity formula is used for calculation. The formula is as follows:

[0053]

[0054] In the above formula, m is the number of concrete samples, Represents the standardized mean of the corresponding indicator; calculate the amount of information:

[0055] C j =σ j ×R j

[0056]

[0057] In the above formula, R j is the conflict of the current indicator, r jk represents the correlation coefficient between the j index and the k index, n represents the total number of indicators, C j represents the comprehensive information carrying capacity of the jth indicator;

[0058] The LightGBM model was constructed and parameter training was performed. The training process used time series cross-validation and early stopping mechanism for auxiliary training. The parameter settings of the LightGBM model were as follows, including tree structure parameters, regularization parameters, and training parameters; the tree structure control parameters were as follows: the number of leaf nodes was 30, the maximum depth of the tree was set to 5, and the minimum number of samples of leaf nodes was 20; the regularization parameters were as follows, the regularization coefficient L1 was set to 0.1, the regularization coefficient L2 was set to 0.2, and the classification minimum value-added threshold was set to 0.3; the training parameters were as follows: the learning rate was 0.3, the number of iterations was 300, and the number of early stopping rounds was set to 8.

[0059] According to the weight distribution result obtained by CRITIC weight calculation, feature weighting processing is performed. The weight distribution formula is as follows:

[0060]

[0061] In the above formula, w j Indicates the final weight of the corresponding indicator j, C k It represents the sum of the comprehensive information of all indicators.

[0062] Step 4: Refer to the weights obtained in step 3 and use the concrete life prediction model to obtain the service life of concrete. By inputting the above indicators into the concrete life prediction model

[0063] Therefore, the present invention adopts a method for predicting the fatigue life of solid waste fiber concrete. By setting positive cycle and counter-cycle sampling methods, the degree of change of physical properties of concrete samples at different time intervals can be collected, thereby realizing the prediction of changes in concrete sample indicators. By establishing a mapping relationship between cycle days, indicators and life span, and then through different simulation environments, the prediction of concrete life span is finally realized.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for predicting fatigue life of solid waste fiber concrete, characterized by: The following steps are involved: Step 1: Use solid waste to make m pieces of concrete samples, perform alkali excitation on the concrete samples, and measure the physical properties of the concrete samples; Step 2: Establish a test environment, place the concrete sample in the test environment, and collect indicators of the concrete sample according to the cycle time, including n indicators, to form a data group; Step 3: Use the data set as calculation data, first standardize the data, then calculate the amount of information, and input it into the constructed concrete life prediction model; Step 4: Use the concrete life prediction model to obtain the service life of concrete.

2. The fatigue life prediction method for solid waste fiber concrete according to claim 1, characterized in that: The specific process of step one is as follows: Prepare m concrete samples with a specific size of 100mm*100mm*100mm, carry out standard curing, and perform alkali excitation on the concrete samples. Then, measure the physical properties of the concrete samples, including compressive strength, water absorption, and porosity.

3. The fatigue life prediction method for solid waste fiber concrete according to claim 1, characterized in that: The specific process in step 2 is as follows: The test environment includes marine environment, hot and humid environment and dry and cold environment. Concrete specimens are tested in the above environments. There are two cycle times, namely positive cycle and reverse cycle. The positive cycle is frequently tested in the early stage of the test, and the reverse cycle is frequently tested in the later stage of the test. The collected indicators include the physical properties in step one, including compressive strength, water absorption rate and porosity. The mass change rate and expansion rate of the concrete specimens are also tested. The above data are formed into data groups in chronological order.

4. The fatigue life prediction method for solid waste fiber concrete according to claim 1, characterized in that: In step 3, the data set collected in step 2 is input into the concrete life prediction model. Here, the concrete life prediction model is specifically set to the LightGBM+CRITIC weighted model. The processing process is as follows: First, the CRITIC weight calculation stage is carried out. The positive indicator is set to compressive strength and normalized. The negative indicator is set to water absorption, porosity and expansion rate, and reverse normalized. The cycle time is logarithmically transformed and then normalized. The above indicators are used to calculate the amount of information. The weight distribution result is obtained according to the calculated amount of information. The specific calculation process is as follows: Normalize the compressive strength using the following formula: In the above formula, x' ij is the normalized value of the compressive strength of the i-th sample at the j-th day, x ij represents the compressive strength value of the i-th sample at the j-th day, x j represents the compressive strength value of all concrete specimens on day j; The process of denormalizing water absorption, porosity, and swelling is as follows: In the above formula, a' ij Indicates the normalized value of water absorption, porosity or expansion rate, a ij Indicates the original value of water absorption, porosity or expansion rate corresponding to the i-th sample at the j-th day, a j It represents the water absorption, porosity or expansion rate of the corresponding concrete sample on the jth day; The information volume is calculated using the contrast intensity formula, which is as follows: In the above formula, m is the number of concrete samples, Represents the standardized mean, including the preceding x' ij and a' ij , the formula for calculating the amount of information is as follows: C j =s j ×R j In the above formula, R j is the conflict of the current indicator, r jk represents the correlation coefficient between the j index and the k index, n represents the total number of indicators, C j represents the comprehensive information carrying capacity of the jth indicator; The LightGBM model is constructed and parameter training is performed. The training process uses time series cross-validation and early stopping mechanism for auxiliary training. The weight distribution results obtained by CRITIC weight calculation are used to perform feature weighting processing.

5. The method for predicting fatigue life of solid waste fiber concrete according to claim 4, characterized in that: The parameter settings of the LightGBM model are as follows, including tree structure parameters, regularization parameters, and training parameters; the tree structure control parameters are as follows: the number of leaf nodes is 30, the maximum depth of the tree is set to 5, and the minimum number of samples of leaf nodes is 20; the regularization parameters are as follows: the regularization coefficient L1 is set to 0.1, the regularization coefficient L2 is set to 0.2, and the classification minimum value-added threshold is set to 0.3; The training parameters are as follows: learning rate is 0.3, number of iterations is 300, and number of early stopping rounds is set to 8.

6. The method for predicting fatigue life of solid waste fiber concrete according to claim 5, characterized in that: The weight distribution results are calculated as follows: The formula for weight distribution is as follows: In the above formula, w j is the final weight, C k It represents the sum of the comprehensive information of all indicators.

Images