Deep learning-based steel template service life evaluation method

Abstract

The invention discloses a steel template service life evaluation method based on deep learning, and relates to the technical field of steel template management, and the method comprises the following steps: collecting the use data and image data of a steel template, analyzing and obtaining the physical index data, structural data and material quality data of the steel template, carrying out the preprocessing of the collected data, and obtaining the service life of the steel template; the physical index data comprises the strength, the abrasion and the corrosion of the steel template, and the structural data comprises the three-dimensional geometrical shape and the size data. According to the method, the physical strength, the surface condition and the material attribute of the steel template are comprehensively analyzed based on deep learning, so that the performance difference of different parts caused by factors such as stress distribution, corrosive wear and the like is reflected more meticulously and accurately; and the deep learning algorithm can automatically learn and extract characteristics having key influence on the service life of the steel template from a large amount of multi-dimensional data, so that the accuracy of evaluating the service life of the steel template is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of steel formwork management, and in particular to a method for evaluating the service life of a steel formwork based on deep learning. Background Art

[0002] As an important infrastructure material, steel formwork is widely used in construction engineering, bridge engineering, water conservancy engineering and other fields. It is mainly used for the forming and support of concrete structures, which significantly reduces the pore water pressure and air bubble removal in the concrete pressure by traditional formwork, forming an ideal rough interface, thereby improving the project quality and construction efficiency. The wide application of steel formwork makes the assessment of its service life particularly important. Deep learning, as an important branch of the field of artificial intelligence, has powerful data processing and pattern recognition capabilities. Therefore, this technical advantage can be used to conduct refined assessment and prediction of steel formwork, thereby improving the accuracy and scientificity of the assessment.

[0003] In the existing technology, the evaluation of the service life of steel formwork focuses more on physical indicators such as strength, wear, and rust of the steel formwork, while ignoring the influence of the steel formwork structure and material quality. Therefore, how to improve the comprehensiveness of the steel formwork life evaluation indicators and accurately evaluate its service life is the problem we need to solve. To this end, a steel formwork service life evaluation method based on deep learning is proposed. Summary of the invention

[0004] The purpose of the present invention is to provide a steel formwork service life assessment method based on deep learning to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A method for evaluating the service life of a steel formwork based on deep learning comprises the following steps:

[0007] Step 1: Collect usage data and image data of the steel template, analyze and obtain physical index data, structural data and material quality data of the steel template, and pre-process the collected data to obtain a labeled data set, wherein the physical index data includes the strength, wear and corrosion of the steel template, the structural data includes three-dimensional geometric shape and size data, and the material quality data includes chemical composition and grain structure;

[0008] Step 2: Use deep learning algorithms to analyze physical index data, structural data, and material quality data, perform feature extraction, and obtain template strength features, performance features, and material features to form a comprehensive feature set;

[0009] Step 3, based on the extracted template strength characteristics, performance characteristics and material characteristics, combined with the associated data of the comprehensive feature set, analyze the influence of the template strength characteristics, performance characteristics and material characteristics on the service life of the steel template;

[0010] Step 4, comprehensively considering the strength characteristics, performance characteristics and material characteristics of the formwork, constructing a life prediction model, obtaining a correction prediction coefficient, and evaluating the remaining service life range of the steel formwork;

[0011] Step 5: Based on the assessment results of the remaining service life of the steel formwork, formulate a corresponding maintenance and replacement plan.

[0012] A further improvement of the technical solution of the present invention is that in step 1, the process of obtaining the labeled data set is:

[0013] Step 101, using a universal testing machine to perform tensile and compressive mechanical property tests on the steel template to obtain tensile strength and yield strength data, and visually inspecting the steel template to record the wear marks and rust degree on the surface of the steel template, and using a laser scanner to measure the wear depth and rust area;

[0014] Step 102, using a 3D laser scanner to perform an all-round scan of the steel template to obtain its 3D geometric shape and size data, and post-processing the scan results, such as denoising, splicing, alignment, etc., to generate a complete 3D model;

[0015] Step 103, using a spectrometer to perform chemical composition analysis on the steel template to determine the content of its main elements and trace elements, and cutting a cross-section sample of the steel template for metallographic preparation, using a metallographic microscope to observe the grain morphology, size, distribution and orientation of the material, and analyzing its organizational structure;

[0016] Step 104, preprocessing the collected data, wherein the image data is cropped, scaled, and color corrected to meet the analysis requirements, and the physical indicators and material quality data are cleaned to remove outliers and noise;

[0017] Step 105 , integrating the preprocessed physical index data, structural data and material quality data into a data set for storage, and annotating detailed attribute information for each sample in the data set.

[0018] A further improvement of the technical solution of the present invention is that in step 2, the process of obtaining the comprehensive feature set is:

[0019] Step 201, standardizing the preprocessed physical index data, structural data and material quality data, and scaling the data to the same scale;

[0020] Step 202, using a deep learning model to extract features from the data, for physical indicator data, extracting the template strength features of tensile strength and yield strength, for structural data, extracting the performance features of three-dimensional geometric shape and size, and for material quality data, analyzing the chemical composition and grain structure data to extract material features;

[0021] In step 203, the features extracted from different data types are fused to form a comprehensive feature set that fully reflects the physical properties, structural integrity and material attributes of the steel formwork.

[0022] A further improvement of the technical solution of the present invention is that in step 3, the process of analyzing the influence of the template strength characteristics, performance characteristics and material characteristics on the service life of the steel template is as follows:

[0023] Step 301, based on the relevant data of the comprehensive feature set, extract the tensile strength and yield strength data of the template strength feature, extract the wear degree, fatigue life, and corrosion resistance data of the performance feature, and extract the chemical composition, hardness, internal inclusions, and crack structure data of the material feature;

[0024] Step 302, evaluating the correlation between the template strength characteristics, performance characteristics and material characteristics and service life;

[0025] Step 303, based on the correlation between tensile strength, yield strength and service life, analyze the influence of strength characteristics on the durability of the template; based on the correlation between wear degree, fatigue life, corrosion resistance and service life, analyze the influence of performance degradation of the template during use on the service life; based on the influence of chemical composition, hardness, internal inclusions, and crack structure on the durability of the template, analyze the influence of cracks and inclusion defects on the early failure of the template;

[0026] Step 304, based on the results of the correlation degree analysis and in combination with the relevant data of the comprehensive feature set, respectively calculate the strength evaluation index, the performance evaluation index and the material evaluation index, and analyze the influence of the template strength characteristics, performance characteristics and material characteristics on the service life.

[0027] A further improvement of the technical solution of the present invention is that the calculation formula of the strength evaluation index is:

[0028]

[0029] Among them, I S is the strength assessment index, S i is the tensile strength of the i-th sample, S min and S max is the minimum and maximum value of tensile strength, σ i is the yield strength of the i-th sample, σ y and σu is the yield strength and ultimate tensile strength, n is the number of samples, α and β are weight parameters used to adjust the contribution of tensile strength and yield strength to the evaluation index. As the strength increases, the evaluation index I S It tends to increase, but the growth rate is regulated by α and β;

[0030] The calculation formula of the performance evaluation index is:

[0031]

[0032] Among them, I P is the performance evaluation index, WR(s) is the wear rate at time point s, FL(s) is the predicted fatigue life at time point s, and FL avg is the average fatigue life, k is the weight, γ is the attenuation parameter, t is the upper limit of the integral, with the decrease of wear rate and the increase of fatigue life, the evaluation index I P tends to 1;

[0033] The calculation formula of the material evaluation index is:

[0034]

[0035] Among them, I M is the material evaluation index, H j is the hardness contribution of the jth chemical component, H min is the minimum value of hardness, H max is the maximum value of hardness, DS j is the j-th defect score, m is the number of defect types, λ is the penalty weight of the defect score, and with the increase of hardness and the decrease of defect score, the evaluation index I M Tends to increase.

[0036] A further improvement of the technical solution of the present invention is that in step 4, the process of obtaining the correction prediction coefficient is:

[0037] Step 401, extracting associated feature data of template strength features, performance features, and material features based on the comprehensive feature set, analyzing the degradation trends of the strength, performance, and material features of the steel template, establishing a performance degradation model using a deep learning algorithm, and analyzing the relationship between the associated features and the remaining service life;

[0038] Step 402, using the associated feature data to train the performance degradation model, and using a cross-validation method to verify the prediction ability of the model, capturing the dynamic changes in the strength, performance and material characteristics of the steel formwork during use, and predicting its future degradation trend, to ensure the accuracy and generalization ability of the model;

[0039] Step 403, according to the prediction result of the performance degradation model and the actual use of the steel formwork, combined with the strength evaluation index, the performance evaluation index and the material evaluation index, calculate the correction prediction coefficient, and analyze the influence of the result of the correction prediction coefficient on the prediction value of the remaining life of the steel formwork;

[0040] Step 404, based on the correction prediction coefficient classification, to analyze the remaining service life range of the steel formwork, and to classify corresponding evaluation thresholds for different steel formwork degradation levels, the degradation levels of the steel formwork are low degradation level, medium degradation level and high degradation level, wherein the remaining service life of the steel formwork gradually decreases from low to high degradation level;

[0041] Step 405, for a new steel formwork or an existing formwork whose remaining service life needs to be evaluated, input its strength characteristics, performance characteristics and material characteristics data, and use the performance degradation model combined with the correction prediction coefficient to predict the remaining service life of the steel formwork.

[0042] A further improvement of the technical solution of the present invention is that the calculation formula of the correction prediction coefficient is:

[0043]

[0044] Where C is the correction prediction coefficient, n is the number of samples, is the strength evaluation index of the i-th sample, is the performance evaluation index of the i-th sample, is the material evaluation index of the i-th sample, q, w, e are used to adjust the weight parameters of strength, performance and material evaluation index to the comprehensive evaluation, δ is the scaling factor for adjusting the overall comprehensive evaluation result, m is the number of defect types, DS j Score the j-th defect, DS max is the maximum value of the defect score, which is used for normalization, and λ is the penalty weight of the defect score, which controls the negative impact of the defect on the correction prediction coefficient.

[0045] A further improvement of the technical solution of the present invention is that: the multiple degradation levels correspond to multiple evaluation thresholds, wherein the evaluation thresholds include an upper threshold and a lower threshold;

[0046] The multiple degradation levels and the multiple evaluation thresholds satisfy the following relationship:

[0047] Low degradation level C≥C h , the steel formwork is in good condition and is expected to have a long remaining service life, without the need for immediate overhaul or replacement;

[0048] Medium degradation level C l ≤C <C h, the remaining service life of the steel formwork is at a medium level, and it may be necessary to pay attention to its performance changes and consider maintenance or inspection in the future;

[0049] High Degradation Level 0 <C<C l , the remaining service life of the steel formwork is short, and it is necessary to pay close attention to its performance changes and consider overhauling, replacing or taking other measures as soon as possible to ensure safe use;

[0050] Among them, C is the correction prediction coefficient, C h is the lower threshold corresponding to the low degradation level and the upper threshold corresponding to the medium degradation level, C l is the lower threshold corresponding to the medium degradation level and the upper threshold corresponding to the high degradation level.

[0051] A further improvement of the technical solution of the present invention is that in step 5, the process of formulating the corresponding maintenance and replacement plan is:

[0052] Step 501, analyzing the remaining service life assessment result of the steel formwork to determine its current degradation level, so as to clarify the current state and potential risks of the steel formwork;

[0053] Step 502, based on the current state of the steel formwork and the determined degradation level, a maintenance plan corresponding to the degradation level is formulated, and according to the maintenance plan, the steel formwork is repaired, such as surface treatment, rust removal, repainting, reinforcement or replacement of damaged parts;

[0054] Step 503, setting replacement standards, and formulating a replacement plan for steel formwork that meets or exceeds the replacement standards;

[0055] Step 504, during the implementation of the maintenance and replacement plan, the usage and performance of the steel formwork are continuously monitored. If it is found that there is a large deviation between the actual usage and the predicted results, the maintenance and replacement plan is adjusted in a timely manner.

[0056] Due to the adoption of the above technical solution, the present invention has the following technical advances compared with the prior art:

[0057] 1. The present invention provides a method for assessing the service life of steel formwork based on deep learning. Based on deep learning, a comprehensive analysis is performed on the physical strength, surface condition and material properties of the steel formwork, which can more carefully and accurately reflect the performance differences in different parts caused by factors such as stress distribution, corrosion and wear. In addition, the deep learning algorithm can automatically learn from large amounts of multi-dimensional data and extract features that have a key impact on the service life of the steel formwork, thereby significantly improving the accuracy of the assessment of the service life of the steel formwork.

[0058] 2. The present invention provides a method for assessing the service life of steel formwork based on deep learning. It provides early warning information by identifying signs of steel formwork performance degradation and potential risks in advance, and collects data such as the strength, performance, and material characteristics of the steel formwork and inputs them into a trained model to analyze and obtain the prediction results of the remaining service life. This assessment capability enables relevant personnel to understand the degradation of the steel formwork in a timely manner, and formulate timely maintenance and replacement plans accordingly, thereby avoiding potential safety risks and production delays. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0060] Figure 1 is a flow chart of the method of the present invention;

[0061] Figure 2 A flow chart for analyzing the influence of the template strength characteristics, performance characteristics and material characteristics on the service life of the steel template in the present invention;

[0062] Figure 3 This is a flow chart for obtaining the correction prediction coefficients of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] Embodiment 1, as Figure 1 , Figure 2 As shown, the present invention provides a method for evaluating the service life of a steel formwork based on deep learning, comprising the following steps:

[0065] Step 1: Collect the usage data and image data of the steel template, analyze and obtain the physical index data, structural data and material quality data of the steel template, and pre-process the collected data to obtain a labeled data set, wherein the physical index data includes the strength, wear and corrosion of the steel template, the structural data includes the three-dimensional geometric shape and size data, and the material quality data includes the chemical composition and grain structure. Use a universal testing machine to perform tensile and compressive mechanical property tests on the steel template to obtain tensile strength and yield strength data, and perform a visual inspection on the steel template to record the wear marks and rust degree on the surface of the steel template, and use a laser scanner to measure the wear depth and rust area. Use a three-dimensional laser scanner to perform a full-scale scan of the steel template to obtain its three-dimensional geometric shape and Dimensional data, and post-process the scanning results, such as denoising, splicing, alignment, etc., to generate a complete 3D model, use a spectrometer to analyze the chemical composition of the steel template, determine its main elements and trace element content, and cut the cross-section sample of the steel template for metallographic preparation, use a metallographic microscope to observe the grain morphology, size, distribution and orientation of the material, analyze its organizational structure, and pre-process the collected data, including cropping, scaling and color correction of the image data to meet the analysis requirements, cleaning the physical indicators and material quality data, removing outliers and noise, integrating the pre-processed physical indicator data, structural data and material quality data into the data set for storage, and annotating detailed attribute information for each sample in the data set;

[0066] Step 2: Use deep learning algorithms to analyze physical indicator data, structural data, and material quality data, perform feature extraction, obtain template strength features, performance features, and material features, respectively, and form a comprehensive feature set. Standardize the preprocessed physical indicator data, structural data, and material quality data, scale the data to the same scale, and use a deep learning model to extract features from the data. For physical indicator data, extract template strength features of tensile strength and yield strength, extract performance features of three-dimensional geometric shape and size for structural data, and extract material features from chemical composition and grain structure data for material quality data. Fusion the features extracted from different data types to form a comprehensive feature set that comprehensively reflects the physical properties, structural integrity, and material properties of the steel template;

[0067] Step 3, according to the extracted template strength characteristics, performance characteristics and material characteristics, combined with the associated data of the comprehensive feature set, analyze the influence of the template strength characteristics, performance characteristics and material characteristics on the service life of the steel template, based on the relevant data of the comprehensive feature set, extract the tensile strength and yield strength data of the template strength characteristics, extract the wear degree, fatigue life and corrosion resistance data of the performance characteristics, extract the chemical composition, hardness, internal inclusions and crack structure data of the material characteristics, evaluate the correlation between the template strength characteristics, performance characteristics and material characteristics and the service life, based on the correlation between the tensile strength and yield strength and the service life, analyze the influence of the strength characteristics on the durability of the template, based on the correlation between the wear degree, fatigue life, corrosion resistance and the service life, analyze the influence of the performance degradation of the template during use on the service life, based on the influence of chemical composition, hardness, internal inclusions and crack structure on the durability of the template, analyze the influence of cracks and inclusion defects on the early failure of the template, based on the results of the correlation degree analysis, combined with the relevant data of the comprehensive feature set, respectively calculate the strength evaluation index, performance evaluation index and material evaluation index, and analyze the influence of the template strength characteristics, performance characteristics and material characteristics on the service life;

[0068] Furthermore, the calculation formula of the strength assessment index is:

[0069]

[0070] Among them, I S is the strength assessment index, S i is the tensile strength of the i-th sample, S min and S max is the minimum and maximum value of tensile strength, σ i is the yield strength of the i-th sample, σ y and σ u is the yield strength and ultimate tensile strength, n is the number of samples, α and β are weight parameters used to adjust the contribution of tensile strength and yield strength to the evaluation index. As the strength increases, the evaluation index I S It tends to increase, but the growth rate is regulated by α and β;

[0071] The calculation formula of the performance evaluation index is:

[0072]

[0073] Among them, I P is the performance evaluation index, WR(s) is the wear rate at time point s, FL(s) is the predicted fatigue life at time point s, and FL avg is the average fatigue life, k is the weight, γ is the attenuation parameter, t is the upper limit of the integral, with the decrease of wear rate and the increase of fatigue life, the evaluation index IP tends to 1;

[0074] The calculation formula of the material evaluation index is:

[0075]

[0076] Among them, I M is the material evaluation index, H j is the hardness contribution of the jth chemical component, H min is the minimum value of hardness, H max is the maximum value of hardness, DS j is the j-th defect score, m is the number of defect types, λ is the penalty weight of the defect score, and with the increase of hardness and the decrease of defect score, the evaluation index I M tend to increase;

[0077] Step 4, comprehensively considering the strength characteristics, performance characteristics and material characteristics of the formwork, constructing a life prediction model, obtaining a correction prediction coefficient, and evaluating the remaining service life range of the steel formwork;

[0078] Step 5: Based on the assessment results of the remaining service life of the steel formwork, formulate a corresponding maintenance and replacement plan.

[0079] Embodiment 2, as Figure 3 As shown, based on Example 1, the present invention provides a technical solution: Preferably, in step 4, the process of obtaining the correction prediction coefficient is:

[0080] Based on the comprehensive feature set, the associated feature data of the template strength feature, performance feature and material feature are extracted, the degradation trend of the strength, performance and material features of the steel template is analyzed, a performance degradation model is established using a deep learning algorithm, the relationship between the associated features and the remaining service life is analyzed, the performance degradation model is trained using the associated feature data, and the prediction ability of the model is verified using the cross-validation method. The dynamic changes in the strength, performance and material characteristics of the steel template during use are captured, and its future degradation trend is predicted to ensure the accuracy and generalization ability of the model. According to the prediction results of the performance degradation model and the actual use of the steel template, combined with the strength evaluation index, performance evaluation index and other indicators, the performance degradation model is used to predict the future degradation trend of the steel template. The estimation index and material evaluation index are calculated, the correction prediction coefficient is calculated, and the influence of the result of the correction prediction coefficient on the prediction value of the remaining life of the steel formwork is analyzed. Based on the classification of the correction prediction coefficient, the remaining service life range of the steel formwork is analyzed, and the corresponding evaluation thresholds are divided for different steel formwork degradation levels. The degradation levels of the steel formwork are low degradation level, medium degradation level and high degradation level, wherein the degradation level of the steel formwork is from low to high, and the remaining service life gradually decreases. For new steel formwork or existing formworks whose remaining service life needs to be evaluated, input their strength characteristics, performance characteristics and material characteristics data, and use the performance degradation model combined with the correction prediction coefficient to predict the remaining service life of the steel formwork;

[0081] Furthermore, the calculation formula of the correction prediction coefficient is:

[0082]

[0083] Where C is the correction prediction coefficient, n is the number of samples, is the strength evaluation index of the i-th sample, is the performance evaluation index of the i-th sample, is the material evaluation index of the i-th sample, q, w, e are used to adjust the weight parameters of strength, performance and material evaluation index to the comprehensive evaluation, δ is the scaling factor for adjusting the overall comprehensive evaluation result, m is the number of defect types, DS j Score the j-th defect, DS max is the maximum value of the defect score, which is used for normalization. λ is the penalty weight of the defect score, which controls the negative impact of the defect on the correction prediction coefficient. The value range of C is (0,1]. When q, w, e are large, the corresponding evaluation index has a greater impact on C. The value of δ will affect the magnitude of C, but will not change its relative size relationship. The larger the value of λ, the more significant the negative impact of the defect on C, resulting in a smaller value of C. As n increases (that is, the number of samples increases), if the difference in evaluation index between samples is not large, the value of C tends to be stable; if the difference is significant, the value of C fluctuates greatly, and m and DS j The change of directly affects the sum of defect scores, which in turn affects the value of C;

[0084] Furthermore, the plurality of degradation levels correspond to a plurality of evaluation thresholds, wherein the evaluation threshold comprises an upper threshold and a lower threshold;

[0085] The multiple degradation levels and multiple evaluation thresholds satisfy the following relationship:

[0086] Low degradation level C≥C h , the steel formwork is in good condition and is expected to have a long remaining service life, without the need for immediate overhaul or replacement;

[0087] Medium degradation level C l ≤C <C h , the remaining service life of the steel formwork is at a medium level, and it may be necessary to pay attention to its performance changes and consider maintenance or inspection in the future;

[0088] High Degradation Level 0 <C<C l , the remaining service life of the steel formwork is short, and it is necessary to pay close attention to its performance changes and consider overhauling, replacing or taking other measures as soon as possible to ensure safe use;

[0089] Among them, C is the correction prediction coefficient, C h is the lower threshold corresponding to the low degradation level and the upper threshold corresponding to the medium degradation level, C l is the lower threshold corresponding to the medium degradation level and the upper threshold corresponding to the high degradation level;

[0090] In step 5, the process of formulating the corresponding maintenance and replacement plan is as follows:

[0091] Analyze the remaining service life assessment results of the steel formwork and determine its current degradation level to clarify the current status and potential risks of the steel formwork. Based on the current status of the steel formwork and the determined degradation level, formulate a maintenance plan corresponding to the degradation level, and repair the steel formwork according to the maintenance plan, such as surface treatment, rust removal, repainting, reinforcement or replacement of damaged parts, etc. Set replacement standards and formulate replacement plans for steel formworks that meet or exceed the replacement standards. In the process of implementing the maintenance and replacement plans, continuously monitor the usage and performance of the steel formwork. If it is found that there is a large deviation between the actual usage and the predicted results, adjust the maintenance and replacement plans in a timely manner.

[0092] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for evaluating the service life of steel formwork based on deep learning, characterized in that: The following steps are involved: Step 1: Collect usage data and image data of the steel template, analyze and obtain physical index data, structural data and material quality data of the steel template, and pre-process the collected data to obtain a labeled data set, wherein the physical index data includes the strength, wear and corrosion of the steel template, the structural data includes three-dimensional geometric shape and size data, and the material quality data includes chemical composition and grain structure; Step 2: Use deep learning algorithms to analyze physical index data, structural data, and material quality data, perform feature extraction, and obtain template strength features, performance features, and material features to form a comprehensive feature set; Step 3, based on the extracted template strength characteristics, performance characteristics and material characteristics, combined with the associated data of the comprehensive feature set, analyze the influence of the template strength characteristics, performance characteristics and material characteristics on the service life of the steel template; Step 4, comprehensively considering the strength characteristics, performance characteristics and material characteristics of the formwork, constructing a life prediction model, obtaining a correction prediction coefficient, and evaluating the remaining service life range of the steel formwork; Step 5: Based on the assessment results of the remaining service life of the steel formwork, formulate a corresponding maintenance and replacement plan.

2. According to a method for evaluating the service life of a steel formwork based on deep learning according to claim 1, it is characterized in that: In step 1, the process of obtaining the labeled data set is as follows: Step 101, using a universal testing machine to perform tensile and compressive mechanical property tests on the steel template to obtain tensile strength and yield strength data, and visually inspecting the steel template to record the wear marks and rust degree on the surface of the steel template, and using a laser scanner to measure the wear depth and rust area; Step 102, using a 3D laser scanner to perform an all-round scan on the steel template to obtain its 3D geometric shape and size data, and post-processing the scan results to generate a complete 3D model; Step 103, using a spectrometer to perform chemical composition analysis on the steel template to determine the content of its main elements and trace elements, and cutting a cross-section sample of the steel template for metallographic preparation, using a metallographic microscope to observe the grain morphology, size, distribution and orientation of the material, and analyzing its organizational structure; Step 104, preprocessing the collected data, wherein the image data is cropped, scaled, and color corrected to meet the analysis requirements, and the physical indicators and material quality data are cleaned to remove outliers and noise; Step 105 , integrating the preprocessed physical index data, structural data and material quality data into a data set for storage, and annotating detailed attribute information for each sample in the data set.

3. According to a method for evaluating the service life of a steel formwork based on deep learning according to claim 2, it is characterized in that: In step 2, the process of obtaining the comprehensive feature set is as follows: Step 201, standardizing the pre-processed physical index data, structural data and material quality data; Step 202, using a deep learning model to extract features from the data, for physical indicator data, extracting the template strength features of tensile strength and yield strength, for structural data, extracting the performance features of three-dimensional geometric shape and size, and for material quality data, analyzing the chemical composition and grain structure data to extract material features; In step 203, the features extracted from different data types are fused to form a comprehensive feature set that fully reflects the physical properties, structural integrity and material attributes of the steel formwork.

4. According to the method for evaluating the service life of steel formwork based on deep learning in claim 3, it is characterized in that: In step 3, the process of analyzing the influence of the template strength characteristics, performance characteristics and material characteristics on the service life of the steel template is as follows: Step 301, based on the relevant data of the comprehensive feature set, extract the tensile strength and yield strength data of the template strength feature, extract the wear degree, fatigue life, and corrosion resistance data of the performance feature, and extract the chemical composition, hardness, internal inclusions, and crack structure data of the material feature; Step 302, evaluating the correlation between the template strength characteristics, performance characteristics and material characteristics and service life; Step 303, based on the correlation between tensile strength, yield strength and service life, analyze the influence of strength characteristics on the durability of the template; based on the correlation between wear degree, fatigue life, corrosion resistance and service life, analyze the influence of performance degradation of the template during use on the service life; based on the influence of chemical composition, hardness, internal inclusions, and crack structure on the durability of the template, analyze the influence of cracks and inclusion defects on the early failure of the template; Step 304, based on the results of the correlation degree analysis and in combination with the relevant data of the comprehensive feature set, respectively calculate the strength evaluation index, the performance evaluation index and the material evaluation index, and analyze the influence of the template strength characteristics, performance characteristics and material characteristics on the service life.

5. The method for evaluating the service life of a steel formwork based on deep learning according to claim 4 is characterized in that: The calculation formula of the strength evaluation index is: Among them, I S is the strength assessment index, S i is the tensile strength of the i-th sample, S min and S max is the minimum and maximum value of tensile strength, σ i is the yield strength of the i-th sample, σ y and σ u are the yield strength and ultimate tensile strength, n is the number of samples, α and β are weight parameters used to adjust the contribution of tensile strength and yield strength to the evaluation index; The calculation formula of the performance evaluation index is: Among them, I P is the performance evaluation index, WR(s) is the wear rate at time point S, FL(s) is the predicted fatigue life at time point s, and FL avg is the average fatigue life, k is the weight, γ is the attenuation parameter, and t is the upper limit of integration; The calculation formula of the material evaluation index is: Among them, I M is the material evaluation index, H j is the hardness contribution of the jth chemical component, H min is the minimum value of hardness, H max is the maximum value of hardness, DS j is the j-th defect score, m is the number of defect types, and λ is the penalty weight of the defect score.

6. The method for evaluating the service life of a steel formwork based on deep learning according to claim 5 is characterized in that: In step 4, the process of obtaining the correction prediction coefficient is as follows: Step 401, extracting associated feature data of template strength features, performance features, and material features based on the comprehensive feature set, analyzing the degradation trends of the strength, performance, and material features of the steel template, establishing a performance degradation model using a deep learning algorithm, and analyzing the relationship between the associated features and the remaining service life; Step 402, using the associated feature data to train the performance degradation model, and using a cross-validation method to verify the prediction ability of the model, capturing the dynamic changes in the strength, performance and material characteristics of the steel formwork during use, and predicting its future degradation trend; Step 403, according to the prediction result of the performance degradation model and the actual use of the steel formwork, combined with the strength evaluation index, the performance evaluation index and the material evaluation index, calculate the correction prediction coefficient, and analyze the influence of the result of the correction prediction coefficient on the prediction value of the remaining life of the steel formwork; Step 404, based on the correction prediction coefficient classification, to analyze the remaining service life range of the steel formwork, and to classify corresponding evaluation thresholds for different steel formwork degradation levels, the degradation levels of the steel formwork are low degradation level, medium degradation level and high degradation level, wherein the remaining service life of the steel formwork gradually decreases from low to high degradation level; Step 405, for a new steel formwork or an existing formwork whose remaining service life needs to be evaluated, input its strength characteristics, performance characteristics and material characteristics data, and use the performance degradation model combined with the correction prediction coefficient to predict the remaining service life of the steel formwork.

7. The method for evaluating the service life of a steel formwork based on deep learning according to claim 6 is characterized in that: The calculation formula of the correction prediction coefficient is: Where C is the correction prediction coefficient, n is the number of samples, is the strength evaluation index of the i-th sample, is the performance evaluation index of the i-th sample, is the material evaluation index of the i-th sample, q, w, e are used to adjust the weight parameters of strength, performance and material evaluation index to the comprehensive evaluation, δ is the scaling factor for adjusting the overall comprehensive evaluation result, m is the number of defect types, DS j Score the j-th defect, DS max is the maximum value of the defect score, which is used for normalization, and λ is the penalty weight of the defect score.

8. The method for evaluating the service life of a steel formwork based on deep learning according to claim 7, characterized in that: The plurality of degradation levels correspond to the plurality of evaluation thresholds, wherein the evaluation thresholds include an upper threshold and a lower threshold; The multiple degradation levels and the multiple evaluation thresholds satisfy the following relationship: Low degradation level C≥C h ; Medium degradation level C l ≤C <C h ; High Degradation Level 0 <C<C l ; Among them, C is the correction prediction coefficient, C h is the lower threshold corresponding to the low degradation level and the upper threshold corresponding to the medium degradation level, C l is the lower threshold corresponding to the medium degradation level and the upper threshold corresponding to the high degradation level.

9. The method for evaluating the service life of a steel formwork based on deep learning according to claim 8, characterized in that: In step 5, the process of formulating the corresponding maintenance and replacement plan is as follows: Step 501, analyzing the remaining service life assessment result of the steel formwork to determine its current degradation level, so as to clarify the current state and potential risks of the steel formwork; Step 502, based on the current state of the steel formwork and the determined degradation level, a maintenance plan corresponding to the degradation level is formulated, and the steel formwork is repaired according to the maintenance plan; Step 503, setting replacement standards, and formulating a replacement plan for steel formwork that meets or exceeds the replacement standards; Step 504, during the implementation of the maintenance and replacement plan, the usage and performance of the steel formwork are continuously monitored. If it is found that there is a large deviation between the actual usage and the predicted results, the maintenance and replacement plan is adjusted in a timely manner.

Images