Concrete micro-parameter design method and system under complex environmental conditions
By acquiring the image data of concrete samples and combining KL transformation and genetic algorithm optimization, the problem of insufficient reliability and accuracy of concrete mesoscopic parameter design in the prior art is solved, and high reliability and high accuracy in complex environments are achieved Designed to adapt to complex environments and extend the life of the structure.
Patent Information
- Application Number
- CN202410976793.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-07-22
AI Technical Summary
The existing detailed concrete parameter design scheme relies on empirical design, resulting in insufficient reliability and accuracy, which cannot meet the refinement and reliability requirements in complex environments.
By obtaining the image data of concrete samples, the projection probability density information and boundary information are used to extract mesoscopic parameters in combination with KL transformation, and parameter optimization is performed based on genetic algorithm and associated chromosome storage genetic algorithm to realize the design of mesoscopic parameters of concrete.
It improves the reliability and accuracy of the detailed parameter design of concrete, can better adapt to complex environmental conditions, reduce design and maintenance costs, and extend the service life of the structure.
Smart Images

Figure CN118522386B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of intelligent manufacturing, and specifically relates to a concrete micro-parameter design method and system under complex environmental conditions. Background Art
[0002] With the development of economy and technology and the improvement of people's living standards, concrete materials have been widely used in people's production and life. The performance of concrete materials will directly affect the safety and reliability of people's production and life.
[0003] Good concrete meso-parameter design results can not only improve the adaptability and stability of concrete structures in complex environments, but also effectively reduce design and maintenance costs while extending the service life of the structure. Therefore, the design of concrete meso-parameters has always been a research hotspot in the industry.
[0004] However, the current concrete meso-parameter design scheme often adopts an empirical design scheme, that is, it relies on empirical parameters to design concrete meso-parameters. However, the reliability and accuracy of such empirical schemes are poor, and they can no longer meet the current industry's refinement and reliability requirements. Summary of the invention
[0005] One of the purposes of the present invention is to provide a concrete mesoscopic parameter design method under complex environmental conditions with high reliability and good accuracy.
[0006] The second purpose of the present invention is to provide a system for realizing the concrete micro-parameter design method under the complex environmental conditions.
[0007] The method for designing concrete microscopic parameters under complex environmental conditions provided by the present invention comprises the following steps:
[0008] S1. Obtaining a concrete sample under a set environment and obtaining image data corresponding to the concrete sample;
[0009] S2. Extracting mesoscopic parameters of the concrete sample according to the projection probability density information and boundary information of the image data obtained in step S1, combined with KL transformation; the mesoscopic parameters include spatial position parameters, shape parameters and attribute parameters;
[0010] S3. Based on the mesoscopic parameters obtained in step S2, the process information of the genetic algorithm is stored using the companion chromosome and the parameters of the genetic algorithm are corrected, thereby optimizing the mesoscopic parameters of the concrete and completing the design of the mesoscopic parameters of the concrete under complex environmental conditions.
[0011] The step S1 specifically includes the following steps:
[0012] Obtain concrete samples under set conditions to ensure that the size of the samples meets the requirements of the set concrete experiment;
[0013] For the obtained concrete sample, an image of a cross section of the concrete sample is obtained, and filtering is performed on the obtained image to obtain image data corresponding to the concrete sample.
[0014] The step S2 specifically includes the following steps:
[0015] Acquisition of spatial position parameters: based on the projection probability density of the image data of the concrete sample, the spatial position parameters of the concrete sample are acquired;
[0016] Acquisition of shape parameters: Perform Fourier transform on the boundary of the image data of the concrete sample to obtain the shape parameters of the concrete sample;
[0017] Acquisition of property parameters: Based on the image data of the concrete sample and KL transformation, the property parameters of the concrete sample are obtained.
[0018] The acquisition of the spatial position parameters specifically includes the following steps:
[0019] Based on the projection probability density of the image data of the concrete sample, the spatial position parameter is calculated using the following formula : Where n is the total number of pixels; is the probability density of the initial state; The concrete microstructure in the spatial direction The position coordinates on ; For state t at position The projected probability density on .
[0020] The acquisition of the shape parameters specifically includes the following steps:
[0021] Based on the boundary of the image data of the concrete sample, the boundary is represented as a series of coordinate points , where j is the index parameter of the boundary point;
[0022] The boundary coordinates Converted to plural form, expressed as , i is the imaginary unit;
[0023] Boundary coordinates The complex form of discrete Fourier transform is obtained to obtain a series of Fourier coefficients for , where N is the total number of boundary points and k is the frequency index;
[0024] The Fourier coefficients are calculated using the following formula Simplify and get the simplified Fourier coefficients : Where nn is the set threshold;
[0025] According to the simplified Fourier coefficients , and reconstruct the simplified boundary by inverse discrete Fourier transform for ;
[0026] Finally, the shape parameters are calculated for .
[0027] The acquisition of the attribute parameters specifically includes the following steps:
[0028] In order to study the bulk density of concrete and damping Based on the KL transformation, the concrete volume density and damping The random field of is discretized into a set of independent standard normal random variables, expressed as
[0029]
[0030] In the formula for The corresponding random field is The value is or ; is a Gaussian random field The mean of ; is a Gaussian random field The standard deviation of ; is the eigenvalue of KL transformation; is the characteristic function of KL transformation; is the random variable used to generate the random field; is the spatial position of the parameter field;
[0031] Take the first n terms of the KL expansion method to calculate and simplify it to In the formula For the simplified The corresponding random field;
[0032] Finally, As a property parameter.
[0033] The step S3 specifically includes the following steps:
[0034] A. Initialize the parameters of the genetic algorithm and generate potential solutions;
[0035] B. Calculate the fitness value of the individual based on the spatial position parameters, shape parameters and attribute parameters, and evaluate the individuals in the current population;
[0036] C. Use companion chromosomes to complete the recording of initial population information;
[0037] D. Select individuals according to their fitness values;
[0038] E. Genetic manipulation;
[0039] F. According to the data information of the companion chromosome, the probability of genetic operation is corrected in real time;
[0040] G. Update the data information of the companion chromosome;
[0041] H. Generate new population;
[0042] I. Determine whether the set termination condition is met:
[0043] If the set termination condition is reached, the genetic algorithm ends and the current optimal solution is output as the mesoscopic parameter design result of concrete;
[0044] If the set termination condition is not met, return to step B for the next round of iteration.
[0045] The step A specifically comprises the following steps:
[0046] During initialization, different individuals including spatial position parameters, shape parameters and attribute parameters are randomly generated;
[0047] When generating individuals, a feasible initial solution A is found in the constraint domain according to the control equation; specifically, the following steps are included:
[0048] Set a unit vector with a random direction for and a sufficiently large number M to ensure that The composed vector domain can geometrically cover the entire feasible domain composed of the boundary conditions of the governing equations;
[0049] For a unit vector in a random direction :
[0050] like If the digital point of is in the feasible domain, then the current digital point is an individual in the initial population;
[0051] like If the digital point of is not in the feasible region, then The M in the replace, ;like If the digital point is still not in the feasible region, then In use replace, ; Repeat this i times until The digital points are in the feasible domain; finally, it is repeated n times to generate n initial individuals; the above process is expressed by the following formula: In the formula is the position parameter of the initial state; is the shape parameter of the initial state; It is the attribute parameter of the initial state.
[0052] The step B specifically comprises the following steps:
[0053] The fitness value of an individual is calculated using the following formula: : In the formula is the integral upper bound parameter, and ; is the stress at the current point; is the strain at the current point; is the first intermediate parameter, and , E is Young's elastic modulus, is the first undetermined parameter, is the second undetermined parameter, is the third undetermined parameter, is the fourth undetermined parameter; is the second intermediate parameter, and , is the initial volume density, is the fifth undetermined parameter, is the sixth undetermined parameter, is the seventh undetermined parameter; express , and ;
[0054] A companion chromosome is set for each individual to store the historical fitness of the corresponding individual, the weight of the position parameter, the weight of the shape parameter and the weight of the attribute parameter;
[0055] The weight value is calculated using the following formula: In the formula is the weight value of the ii-th parameter, and the value of ii ranges from 1 to 3, corresponding to the position parameter, shape parameter, and attribute parameter in sequence; is the number of storage cells allocated to the iith parameter in the calculation.
[0056] The step C specifically comprises the following steps:
[0057] The fitness value of the i-th individual , the historical fitness Fh of the current population, the weights of the location parameters, shape parameters and attribute parameters in the population The corresponding data information is stored in the companion chromosome;
[0058] The historical fitness Fh of the current population is expressed as .
[0059] The step D specifically comprises the following steps:
[0060] According to the fitness value of the individual, several individuals are randomly selected as a group, and the fitness value is selected from each group The top several individuals are cross-operated to select the fitness value The lowest individual undergoes mutation operation.
[0061] The step E specifically comprises the following steps:
[0062] Genetic operations include crossover and mutation;
[0063] Set the initial genetic operation probability ;
[0064] The following formula is used for crossover: In the formula is the degree of crossover between two individuals, and its value is ; is the i-th individual after crossover; is the jth individual after crossover; is the i-th individual before crossover; is the jth individual before crossover;
[0065] The following formula is used for mutation: In the formula is the i-th individual before mutation; is the jth element of the i-th individual before mutation; is the jth element of the ith individual after mutation; is the degree of individual variation, and its value is ; For individuals The maximum value of the elements in; For individuals The minimum value of the elements in ; is the i-th individual after mutation.
[0066] The step F specifically comprises the following steps:
[0067] The following formula is used to update the genetic operation probability of the i-th individual: In the formula is the probability of genetic operation, wherein the genetic operation includes crossover and mutation; is the maximum probability of genetic operation; is the minimum probability of genetic operation; is the maximum value of the current historical fitness value.
[0068] The step I specifically comprises the following steps:
[0069] Determine whether the set termination condition is met: the termination condition includes that the number of iterations reaches the set maximum number of iterations, or the current historical fitness value exceeds the set threshold;
[0070] If the set termination condition is reached, the genetic algorithm ends and the current optimal solution is output as the mesoscopic parameter design result of concrete;
[0071] If the set termination condition is not met, return to step B for the next round of iteration.
[0072] The present invention also provides a system for implementing the concrete meso-parameter design method under the complex environmental conditions, comprising a data acquisition module, a parameter extraction module and a parameter design module; the data acquisition module, the parameter extraction module and the parameter design module are connected in series in sequence; the data acquisition module is used to acquire concrete samples under a set environment, and acquire image data corresponding to the concrete samples, and upload the data information to the parameter extraction module; the parameter extraction module is used to extract the meso-parameters of the concrete samples according to the received data information, the projection probability density information and the boundary information of the acquired image data, combined with the KL transformation, and upload the data information to the parameter design module; the parameter design module is used to store the process information of the genetic algorithm based on the received data information, and perform parameter correction of the genetic algorithm using a companion chromosome, thereby optimizing the meso-parameters of the concrete and completing the design of the meso-parameters of the concrete under the complex environmental conditions.
[0073] The method and system for designing concrete meso-parameters under complex environmental conditions provided by the present invention aim at the design requirements of concrete meso-parameters under complex environmental conditions, obtain the parameters of concrete samples through experiments, and realize the design of concrete meso-parameters under complex environmental conditions based on genetic algorithms. Moreover, the method and system have higher reliability and better accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0074] Figure 1 The figure is a schematic diagram of the method flow of the present invention.
[0075] Figure 2It is a schematic diagram of an iterative convergence curve of the companion genetic algorithm of an embodiment of the method of the present invention.
[0076] Figure 3 Schematic diagram of the functional modules of the system of the present invention. DETAILED DESCRIPTION
[0077] like Figure 1 The method flow diagram of the method of the present invention is shown as follows: The concrete micro-parameter design method under complex environmental conditions disclosed by the present invention comprises the following steps:
[0078] S1. Obtaining a concrete sample under a set environment and obtaining image data corresponding to the concrete sample; specifically comprising the following steps:
[0079] Obtain concrete samples under the set environment. In order to ensure the accuracy and reliability of the experiment, it is necessary to ensure that the size of the sample meets the general concrete experimental requirements and the surface quality is good to avoid uncertainties that may be introduced during the preparation process;
[0080] For the obtained concrete sample, an image of the cross section of the concrete sample is obtained (which can be obtained through an electron microscope), and the obtained image is filtered (including preprocessing to eliminate possible noise, and edge detection and image segmentation operations, so as to accurately extract the information required for the microscopic parameters of the concrete), to obtain the image data corresponding to the concrete sample;
[0081] S2. Extracting the microscopic parameters of the concrete sample according to the projection probability density information and boundary information of the image data obtained in step S1 in combination with the KL transform; specifically comprising the following steps:
[0082] Extracting microscopic parameters of the concrete sample according to the image data obtained in step S1; the microscopic parameters include spatial position parameters, shape parameters and property parameters; these microscopic parameters have an important influence on the mechanical properties of concrete, so it is very important to accurately extract and record them;
[0083] Acquisition of spatial position parameters: based on the projection probability density of the image data of the concrete sample, the spatial position parameters of the concrete sample are acquired; specifically, the steps include:
[0084] Based on the projection probability density of the image data of the concrete sample, the spatial position parameter is calculated using the following formula : Where n is the total number of pixels; is the probability density of the initial state; The concrete microstructure in the spatial direction The position coordinates on ; For state t at position The projected probability density on ;
[0085] Acquisition of shape parameters: Perform Fourier transform on the boundary of the image data of the concrete sample to obtain the shape parameters of the concrete sample; specifically, the steps include:
[0086] Based on the boundary of the image data of the concrete sample, the boundary is represented as a series of coordinate points , where j is the index parameter of the boundary point;
[0087] The boundary coordinates Converted to plural form, expressed as , i is the imaginary unit;
[0088] Boundary coordinates The complex form of discrete Fourier transform is obtained to obtain a series of Fourier coefficients for , where N is the total number of boundary points and k is the frequency index;
[0089] The Fourier descriptor is derived from A set of coefficients selected from the Fourier transform, usually low-frequency coefficients are selected because they contain the main features of the shape; therefore, the first n low-frequency coefficients are retained and the remaining coefficients are set to zero; the following formula is used to convert the Fourier coefficients Simplify and get the simplified Fourier coefficients : Where nn is the set threshold;
[0090] According to the simplified Fourier coefficients , and reconstruct the simplified boundary by inverse discrete Fourier transform for ; The original shape is approximated, but the details are simplified;
[0091] Finally, the shape parameters are calculated for ;
[0092] Acquisition of attribute parameters: Based on the image data of the concrete sample and KL transformation, the attribute parameters of the concrete sample are obtained; specifically, the steps include:
[0093] Based on the theoretical support of KL transformation and related literature, the volume density, damping and other property parameters of concrete can be effectively simulated by random fields; this method not only improves the accuracy and efficiency of calculation, but also provides more scientific and reasonable data support for the design and analysis of concrete structures;
[0094] In order to study the bulk density of concrete and damping Based on the KL transformation, the concrete volume density and damping The random field of is discretized into a set of independent standard normal random variables, expressed as In the formula for The corresponding random field is The value is or ; is a Gaussian random field The mean of ; is a Gaussian random field The standard deviation of ; is the eigenvalue of KL transformation; is the characteristic function of KL transformation; is the random variable used to generate the random field; is the spatial position of the parameter field;
[0095] In order to ensure the calculation accuracy and improve the calculation efficiency, the first n terms of the KL expansion method are calculated and simplified to In the formula For the simplified The corresponding random field;
[0096] Finally, As attribute parameters;
[0097] S3. According to the mesoscopic parameters obtained in step S2, based on the genetic algorithm, the process information of the genetic algorithm is stored by using the companion chromosome and the parameters of the genetic algorithm are corrected, so as to optimize the mesoscopic parameters of the concrete and complete the design of the mesoscopic parameters of the concrete under complex environmental conditions; specifically, the steps are as follows:
[0098] A. Initialize the parameters of the genetic algorithm and generate potential solutions; specifically, the following steps are included:
[0099] During initialization, different individuals including spatial position parameters, shape parameters and attribute parameters are randomly generated;
[0100] In order to make better use of computing resources, when generating individuals, a feasible initial solution A is found in the constraint domain according to the control equation, that is, an initial individual that satisfies the boundary conditions of the control equation is generated; specifically, the following steps are included:
[0101] Set a unit vector with a random direction for and a sufficiently large number M to ensure that The composed vector domain can geometrically cover the entire feasible domain composed of the boundary conditions of the governing equations;
[0102] For a unit vector in a random direction :
[0103] like If the digital point of is in the feasible domain, then the current digital point is an individual in the initial population;
[0104] like If the digital point of is not in the feasible region, then The M in the replace, ;like If the digital point is still not in the feasible region, then In use replace, ; Repeat this i times until The digital points are in the feasible domain; finally, it is repeated n times to generate n initial individuals; the above process is expressed by the following formula: In the formula is the position parameter of the initial state; is the shape parameter of the initial state; is the attribute parameter of the initial state;
[0105] B. Based on the spatial position parameters, shape parameters and attribute parameters, calculate the fitness value of the individual and evaluate the individuals in the current population; specifically, the following steps are included:
[0106] The fitness value of an individual is calculated using the following formula: : In the formula is the integral upper bound parameter, and ; is the stress at the current point; is the strain at the current point; is the first intermediate parameter, and , E is Young's elastic modulus, is the first undetermined parameter, is the second undetermined parameter, is the third undetermined parameter, is the fourth undetermined parameter; is the second intermediate parameter, and , is the initial volume density, is the fifth undetermined parameter, is the sixth undetermined parameter, is the seventh undetermined parameter; express , and ;
[0107] A companion chromosome is set for each individual to store the historical fitness of the corresponding individual, the weight of the position parameter, the weight of the shape parameter, and the weight of the attribute parameter; this information can be used to match and guide subsequent genetic operations;
[0108] The weight value is calculated using the following formula: In the formula is the weight value of the ii-th parameter, and the value of ii ranges from 1 to 3, corresponding to the position parameter, shape parameter, and attribute parameter in sequence; is the number of storage units allocated to the ii-th parameter in the calculation, and is also the data length ratio of the ii-th parameter;
[0109] C. Use the companion chromosome to complete the recording of the initial population information; specifically, the following steps are included:
[0110] The fitness value of the i-th individual , the historical fitness of the current population Fh, the weights of the position parameters, shape parameters and attribute parameters in the population The corresponding data information is stored in the companion chromosome;
[0111] The historical fitness Fh of the current population is expressed as ;
[0112] D. Select individuals according to their fitness values; specifically, the following steps are included:
[0113] According to the fitness value of the individual, randomly select several (preferably 4 to 6) individuals as a group, and select the fitness value from each group The top several (preferably 2) individuals undergo the crossover operation in step E, and the fitness value The lowest individual undergoes mutation operation.
[0114] E. Genetic operation; specifically includes the following steps:
[0115] Genetic operations include crossover and mutation;
[0116] First, define the initial genetic operation probability ;
[0117] The following formula is used for crossover: In the formula is the degree of crossover between two individuals, and its value is ; is the i-th individual after crossover; is the jth individual after crossover; is the i-th individual before crossover; is the jth individual before crossover;
[0118] The following formula is used for mutation: In the formula is the i-th individual before mutation; is the jth element of the i-th individual before mutation; is the jth element of the ith individual after mutation; is the degree of individual variation, and its value is ; For individuals The maximum value of the elements in ; For individuals The minimum value of the elements in ; is the i-th individual after mutation;
[0119] F. According to the data information of the companion chromosome, the probability of genetic operation is corrected in real time; specifically, the following steps are included:
[0120] The probability of crossover and mutation in traditional genetic algorithms is fixed. This fixed scheme has low reliability and accuracy. Therefore, the weight information in the companion chromosome is used to adjust the probability. The following formula is used to update the probability of genetic operation of the i-th individual: In the formula is the probability of genetic operation, wherein the genetic operation includes crossover and mutation; is the maximum probability of genetic operation; is the minimum probability of genetic operation; is the maximum value of the current historical fitness value;
[0121] G. Update the data information of the companion chromosome; when generating a new population, not only the genetic information of the individual is transmitted, but also the new companion chromosome information needs to be updated or generated;
[0122] H. Generate a new population; Generate a new population using the above information and genetic operations;
[0123] I. Determine whether the set termination condition is met: the termination condition includes that the number of iterations reaches the set maximum number of iterations, or the current historical fitness value exceeds the set threshold;
[0124] If the set termination condition is reached, the genetic algorithm ends and the current optimal solution is output as the mesoscopic parameter design result of concrete;
[0125] If the set termination condition is not met, return to step B for the next round of iteration.
[0126] The method of the present invention is further described below in conjunction with embodiments:
[0127] like Figure 2 FIG. 1 is a schematic diagram showing the convergence of population fitness with the number of iterations when the genetic algorithm of the present invention is used to optimize concrete microscopic parameters in an embodiment of the method of the present invention; FIG. Figure 2 It can be seen from the main curve and inset curve in that the algorithm has good convergence performance under different temperature conditions. After 2000 iterations, the fitness drops to The following shows that the results are fully converged and meet the expected calculation requirements;
[0128] Table 1 is a schematic diagram of the prediction results of the method of the present invention under three temperature conditions:
[0129] Table 1 Schematic diagram of prediction results under three temperature conditions
[0130]
[0131] From Table 1, we can see that changes in temperature directly affect the physical and chemical processes inside concrete, such as changes in the state of water (freezing, evaporation), the breaking and reorganization of chemical bonds, and the expansion or contraction of materials. These changes are reflected in the microscopic parameters of concrete, including but not limited to mineralogical characteristics, pore structure and particle state. From Table 1, we can see that with the increase in temperature, the values of most parameters also gradually increase. At lower temperatures, these parameter values are lower, which may be related to the low activity of the material and the solidification state of water. As the temperature rises, the material becomes active and the internal structure changes, which is reflected in the increase in parameter values.
[0132] Such values can provide important information about the performance of materials under different environmental conditions for engineering applications. For example, when designing cold-resistant concrete, the values of microscopic parameters at low temperatures are particularly important; and when designing concrete for use in high-temperature environments, attention should be paid to parameter changes at high temperatures. In addition, these values can also help researchers predict the durability and life of materials, as well as their performance under extreme temperatures. With this data, concrete can be more accurately designed and formulated to suit specific conditions of use. At lower temperatures (such as arrive The low values of the microscopic parameters may be related to the freezing of water inside the concrete and the low reactivity of the material itself. Under these conditions, the toughness and strength of the concrete may be high, but the risk of brittle fracture also increases. arrive The increase in the value reflects the changes in the internal structure and properties of concrete, such as water evaporation, breaking of chemical bonds, and thermal expansion of materials.
[0133] The changes in the values of these microscopic parameters reveal the sensitivity of concrete materials to temperature changes and provide performance indicators that the material may exhibit under different temperature conditions. This temperature dependence of concrete microscopic parameters is crucial for designing concrete structures under specific conditions, helps predict material performance changes, and guides the optimization and improvement of concrete. These data are valuable information for materials science research and engineering applications, and can design concrete materials that are more suitable for specific environmental conditions.
[0134] like Figure 3 The figure shows a schematic diagram of the functional modules of the system of the present invention: the system disclosed in the present invention for realizing the concrete meso-parameter design method under the complex environmental conditions comprises a data acquisition module, a parameter extraction module and a parameter design module; the data acquisition module, the parameter extraction module and the parameter design module are connected in series in sequence; the data acquisition module is used to acquire concrete samples under a set environment, and acquire image data corresponding to the concrete samples, and upload the data information to the parameter extraction module; the parameter extraction module is used to extract the meso-parameters of the concrete samples according to the received data information, the projection probability density information and the boundary information of the acquired image data, combined with the KL transformation, and upload the data information to the parameter design module; the parameter design module is used to store the process information of the genetic algorithm based on the received data information, and perform parameter correction of the genetic algorithm using the companion chromosome, so as to optimize the meso-parameters of the concrete and complete the design of the meso-parameters of the concrete under the complex environmental conditions.
Claims
1. A method for designing concrete mesoscopic parameters under complex environmental conditions. Features The steps include: S1. Obtaining a concrete sample under a set environment and obtaining image data corresponding to the concrete sample; S2. Extracting the mesoscopic parameters of the concrete sample according to the projection probability density information and boundary information of the image data obtained in step S1 in combination with KL transformation; the mesoscopic parameters include spatial position parameters, shape parameters and attribute parameters; specifically comprising the following steps: Acquisition of spatial position parameters: based on the projection probability density of the image data of the concrete sample, the spatial position parameters of the concrete sample are acquired; Acquisition of shape parameters: Perform Fourier transform on the boundary of the image data of the concrete sample to obtain the shape parameters of the concrete sample; Acquisition of attribute parameters: Based on the image data of the concrete sample and KL transformation, the attribute parameters of the concrete sample are obtained; The acquisition of the spatial position parameters specifically includes the following steps: Based on the projection probability density of the image data of the concrete sample, the spatial position parameter is calculated using the following formula : Where n is the total number of pixels; is the probability density of the initial state; The concrete microstructure in the spatial direction The position coordinates on ; For state t at position The projected probability density on ; The acquisition of the shape parameters specifically includes the following steps: Based on the boundary of the image data of the concrete sample, the boundary is represented as a series of coordinate points , where j is the index parameter of the boundary point; The boundary coordinates Converted to plural form, expressed as , i is the imaginary unit; Boundary coordinates The complex form of discrete Fourier transform is obtained to obtain a series of Fourier coefficients for , where N is the total number of boundary points and k is the frequency index; The Fourier coefficients are calculated using the following formula Simplify and get the simplified Fourier coefficients : Where nn is the set threshold; According to the simplified Fourier coefficients , and reconstruct the simplified boundary by inverse discrete Fourier transform for ; Finally, the shape parameters are calculated for ; The acquisition of the attribute parameters specifically includes the following steps: In order to study the bulk density of concrete and damping Based on the KL transformation, the concrete volume density and damping The random field of is discretized into a set of independent standard normal random variables, expressed as In the formula for The corresponding random field is The value is or ; is a Gaussian random field The mean of ; is a Gaussian random field The standard deviation of ; is the eigenvalue of KL transformation; is the characteristic function of KL transformation; is the random variable used to generate the random field; is the spatial position of the parameter field; Take the first n terms of the KL expansion method to calculate and simplify it to In the formula For the simplified The corresponding random field; Finally, As attribute parameters; S3. According to the mesoscopic parameters obtained in step S2, based on the genetic algorithm, the process information of the genetic algorithm is stored by using the companion chromosome and the parameters of the genetic algorithm are corrected, so as to optimize the mesoscopic parameters of the concrete and complete the design of the mesoscopic parameters of the concrete under complex environmental conditions; specifically, the steps are as follows: A. Initialize the parameters of the genetic algorithm and generate potential solutions; specifically, the following steps are included: During initialization, different individuals including spatial position parameters, shape parameters and attribute parameters are randomly generated; When generating individuals, a feasible initial solution A is found in the constraint domain according to the control equation; specifically, the following steps are included: Set a unit vector with a random direction for and a sufficiently large number M to ensure that The composed vector domain can geometrically cover the entire feasible domain composed of the boundary conditions of the governing equations; For a unit vector in a random direction : like If the digital point of is in the feasible domain, then the current digital point is an individual in the initial population; like If the digital point of is not in the feasible region, then The M in the replace, ;like If the digital point is still not in the feasible region, then In use replace, ; Repeat this i times until The number of points is in the feasible domain; finally, it is repeated n times to generate n initial individuals; The above process is expressed by the following formula: In the formula is the position parameter of the initial state; is the shape parameter of the initial state; is the attribute parameter of the initial state; B. Based on the spatial position parameters, shape parameters and attribute parameters, calculate the fitness value of the individual and evaluate the individuals in the current population; specifically, the following steps are included: The fitness value of an individual is calculated using the following formula: : In the formula is the integral upper bound parameter, and ; is the stress at the current point; is the strain at the current point; is the first intermediate parameter, and , E is Young's elastic modulus, is the first undetermined parameter, is the second undetermined parameter, is the third undetermined parameter, is the fourth undetermined parameter; is the second intermediate parameter, and , is the initial volume density, is the fifth undetermined parameter, is the sixth undetermined parameter, is the seventh undetermined parameter; express , and ; A companion chromosome is set for each individual to store the historical fitness, position parameter weight, shape parameter weight and attribute parameter weight of the corresponding individual; The weight value is calculated using the following formula: In the formula is the weight value of the ii-th parameter, and the value of ii ranges from 1 to 3, corresponding to the position parameter, shape parameter, and attribute parameter in sequence; is the number of storage cells allocated to the iith parameter in the calculation; C. Use the companion chromosome to complete the recording of the initial population information; specifically, the following steps are included: The fitness value of the i-th individual , the historical fitness Fh of the current population, the weights of the location parameters, shape parameters and attribute parameters in the population The corresponding data information is stored in the companion chromosome; The historical fitness Fh of the current population is expressed as ; D. Select individuals according to their fitness values; specifically, the following steps are included: According to the fitness value of the individual, several individuals are randomly selected as a group, and the fitness value is selected from each group The top several individuals are cross-operated to select the fitness value The lowest individual performs mutation operation; E. Genetic operation; specifically includes the following steps: Genetic operations include crossover and mutation; First, define the initial genetic operation probability ; The following formula is used for crossover: In the formula is the degree of crossover between two individuals, and its value is ; is the i-th individual after crossover; is the jth individual after crossover; is the i-th individual before crossover; is the jth individual before crossover; The following formula is used for mutation: In the formula is the i-th individual before mutation; is the jth element of the i-th individual before mutation; is the jth element of the ith individual after mutation; is the degree of individual variation, and its value is ; For individuals The maximum value of the elements in; For individuals The minimum value of the elements in ; is the i-th individual after mutation; F. According to the data information of the companion chromosome, the probability of genetic operation is corrected in real time; specifically, the following steps are included: The genetic operation frequency of the i-th individual is updated using the following formula: In the formula is the probability of genetic operation, wherein the genetic operation includes crossover and mutation; is the maximum probability of genetic operation; is the minimum probability of genetic operation; is the maximum value of the current historical fitness value; G. Update the data information of the companion chromosome; H. Generate new population; I. Determine whether the set termination condition is met: If the set termination condition is reached, the genetic algorithm ends and the current optimal solution is output as the mesoscopic parameter design result of concrete; If the set termination condition is not met, return to step B for the next round of iteration; The specific steps include: Determine whether the set termination condition is met: the termination condition includes that the number of iterations reaches the set maximum number of iterations, or the current historical fitness value exceeds the set threshold; If the set termination condition is reached, the genetic algorithm ends and the current optimal solution is output as the mesoscopic parameter design result of concrete; If the set termination condition is not met, return to step B for the next round of iteration.
2. A system for implementing the concrete mesoscopic parameter design method under complex environmental conditions as described in claim 1, Features It includes a data acquisition module, a parameter extraction module and a parameter design module; the data acquisition module, the parameter extraction module and the parameter design module are connected in series in sequence; the data acquisition module is used to acquire concrete samples under a set environment, acquire image data corresponding to the concrete samples, and upload the data information to the parameter extraction module; the parameter extraction module is used to extract the microscopic parameters of the concrete samples according to the received data information, the projection probability density information and the boundary information of the acquired image data, combined with the KL transformation, and upload the data information to the parameter design module; The parameter design module is used to optimize the mesoscopic parameters of concrete based on the received data information and the genetic algorithm, using the companion chromosome to store the process information of the genetic algorithm and to modify the parameters of the genetic algorithm, so as to complete the design of the mesoscopic parameters of concrete under complex environmental conditions.
Citation Information
Patent Citations
Durable concrete mix proportion optimization method based on RF-NSGA-II
CN111986737A
Method and system for detecting compressive strength of cement concrete and electronic equipment
CN117153297A
Concrete pavement system and method
US20090129862A1
Cement KILN modeling for improved operation
US20230104043A1
Computer-implemented method for automated planning deployment of radio communication devices in an environment
US20230139903A1
