A method for predicting the service life of ALC green wall panels

Abstract

The present invention discloses a method for predicting the service life of an ALC green environmentally friendly wallboard, and relates to the technical field of data processing. The method comprises: obtaining application environment status information of a target ALC environmentally friendly wallboard, and constructing a first use constraint; determining multiple surface defect recognition results; determining K test points; simulating periodic temperature fluctuations in wallboard applications; determining K hysteresis factors and K attenuation factors; performing factor density analysis on the K hysteresis factors and the K attenuation factors, respectively, to determine the hysteresis factor in the test set and the attenuation factor in the test set; calling a service life predictor to analyze the hysteresis factor in the test set and the attenuation factor in the test set, and determining the service life prediction result of the target ALC environmentally friendly wallboard. The present invention solves the technical problems of low reliability and inaccurate prediction of the service life prediction of ALC environmentally friendly wallboards in the prior art, and achieves the technical effect of improving the accuracy of the service life prediction of ALC environmentally friendly wallboards.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for predicting the service life of an ALC green environmentally friendly wallboard. Background Art

[0002] At present, the service life of wallboards often relies on empirical data or simple material durability tests. This method does not take into account the environmental variables of wallboards in actual applications, such as the combined effects of external factors such as temperature, humidity, wind speed, rainfall, etc. For example, in a coastal area, wallboards may be exposed to high humidity and salt spray environments and suffer from long-term corrosion, but if predictions are made based only on general laboratory test data, the service life may be overestimated. In addition, traditional methods often lack real-time monitoring of surface defects of wallboards and cannot effectively identify early damage such as tiny cracks or peeling, resulting in rapid deterioration of wallboard performance under external forces such as temperature fluctuations, but this is not fully considered in the prediction, ultimately affecting the accuracy and reliability of the prediction.

[0003] The existing technology has the technical problems of low reliability and inaccurate prediction of the service life of ALC environmentally friendly wall panels. Summary of the invention

[0004] The present application provides a method for predicting the service life of an ALC green environmentally friendly wallboard, which is used to solve the technical problems of low reliability and inaccurate prediction of the service life of an ALC environmentally friendly wallboard in the prior art.

[0005] In view of the above problems, the present application provides a method for predicting the service life of an ALC green environmentally friendly wallboard, the method comprising:

[0006] Acquire application environment status information of the target ALC environmentally friendly wallboard and construct the first usage constraint;

[0007] Collecting surface images of the target ALC environmentally friendly wallboard, and using a wall surface state identifier to perform defect identification on the wallboard surface image set obtained by the collection, and determining a plurality of surface defect identification results, wherein the plurality of surface defect identification results include a plurality of defect locations;

[0008] Performing hierarchical screening of test points based on the multiple defect positions and the multiple surface defect recognition results to determine K test points, where K is an integer greater than or equal to 1;

[0009] K inner temperature sensors and K outer temperature sensors are respectively arranged on the inner side and the outer side of the wallboard at the K test points, and heating lamps and cooling devices are arranged on the outer side of the wallboard at the K test points, and the heating lamps and cooling devices are regulated according to the application environment status information to simulate the periodic temperature fluctuation in the wallboard application;

[0010] By extracting the temperatures collected by the K inner temperature sensors and the K outer temperature sensors, K periodic inner temperature sets and K periodic outer temperature sets are obtained;

[0011] Performing curve fitting on the K periodic inner temperature sets and the K periodic outer temperature sets respectively to obtain K periodic inner temperature fluctuation curves and K periodic outer temperature fluctuation curves, and performing mapping analysis on the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves to determine K hysteresis factors and K attenuation factors;

[0012] Performing factor density analysis on the K time lag factors and the K attenuation factors respectively to determine a time lag factor in a test set and an attenuation factor in a test set;

[0013] The service life predictor is called to analyze the time lag factor in the test set and the attenuation factor in the test set to determine the service life prediction result of the target ALC environmentally friendly wallboard.

[0014] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0015] The present application obtains the application environment status information of the target ALC environmentally friendly wallboard, constructs the first usage constraint, then collects the surface image of the target ALC environmentally friendly wallboard, and uses the wall surface status identifier to perform defect identification on the collected wallboard surface image set, and determines multiple surface defect identification results, wherein the multiple surface defect identification results include multiple defect positions, and then based on the multiple defect positions and the multiple surface defect identification results, hierarchical screening of test points is performed to determine K test points, wherein K is an integer greater than or equal to 1, and K inner temperature sensors and K outer temperature sensors are respectively arranged on the inner and outer sides of the wallboard at the K test points, and heating lamps and cooling equipment are arranged on the outer sides of the wallboard at the K test points, and the heating lamps and cooling equipment are regulated according to the application environment status information to simulate the periodic temperature in the wallboard application. Fluctuation, then extract the temperatures collected by K inner temperature sensors and K outer temperature sensors, obtain K periodic inner temperature sets and K periodic outer temperature sets, and then perform curve fitting on the K periodic inner temperature sets and K periodic outer temperature sets respectively, obtain K periodic inner temperature fluctuation curves and K periodic outer temperature fluctuation curves, and perform mapping analysis on the K periodic inner temperature fluctuation curves and K periodic outer temperature fluctuation curves, determine K lag factors and K attenuation factors, perform factor density analysis on the K lag factors and K attenuation factors respectively, determine the lag factor in the test set and the attenuation factor in the test set, and then call the service life predictor to analyze the lag factor in the test set and the attenuation factor in the test set to determine the service life prediction result of the target ALC environmentally friendly wallboard. The technical effect of fitting the actual use scenario of the environmentally friendly wallboard and improving the prediction accuracy is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A schematic diagram of a process for predicting the service life of an ALC green environmentally friendly wallboard provided in an embodiment of the present application;

[0018] Figure 2 A schematic diagram of a process for determining a test set hysteresis factor and a test set attenuation factor in a service life prediction method for an ALC green environmentally friendly wallboard provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] The present application provides a method for predicting the service life of ALC green environmentally friendly wall panels, which is used to solve the technical problems of low reliability and inaccurate prediction of the service life of ALC environmentally friendly wall panels in the prior art.

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0021] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0022] Embodiment 1, as Figure 1 As shown, the present application provides a method for predicting the service life of an ALC green environmentally friendly wallboard, wherein the method comprises:

[0023] S100: Acquire application environment status information of the target ALC environmentally friendly wallboard and construct a first usage constraint;

[0024] In one embodiment, the target ALC environmentally friendly wallboard is any ALC wallboard for which service life prediction is required. The application environment status information is used to describe data related to the application physical environment, chemical environment, and usage scenario of the target ALC environmentally friendly wallboard, thereby reflecting the specific environmental conditions of the target ALC environmentally friendly wallboard when it is used. The application environment status information is used as the first usage constraint to pave the way for subsequent temperature testing that fits the actual application scenario of the target ALC environmentally friendly wallboard, thereby achieving the technical effect of improving the accuracy and reliability of service life prediction.

[0025] Preferably, by deploying sensors or using existing environmental monitoring equipment, key physical parameters, chemical parameters and scene parameters of the environment where the ALC environmentally friendly wall panels are located or are about to be used are obtained, and the application environment status information is obtained after the data is summarized. Key physical parameters include: temperature (temperature fluctuation range between day and night, and temperature change trend throughout the year), humidity (air humidity in the wall panel application site, especially changes in humid, wet or dry environments), wind speed (whether it is affected by strong winds, especially the impact of wind speed on the wall panel surface), rainfall and snowfall (the average annual rainfall, snowfall and precipitation frequency in the area, especially extreme climate conditions of heavy rain and snow accumulation), etc.

[0026] Preferably, chemical parameters include atmospheric corrosivity: for example, the salt content in the air (especially in coastal areas) or the content of industrial pollutants (such as sulfur dioxide, nitrogen oxides, etc.), and the pH value. Scene parameters include the installation location of the wall panel, the load conditions (such as whether the wall panel bears additional physical loads), and the vibration impact (whether there is obvious mechanical vibration in the application environment).

[0027] S200: collecting surface images of the target ALC environmentally friendly wallboard, and using a wall surface state identifier to perform defect identification on a collection of collected wallboard surface images, and determining a plurality of surface defect identification results, wherein the plurality of surface defect identification results include a plurality of defect locations;

[0028] In one possible embodiment, a high-precision image acquisition device is selected, such as a high-definition camera, a 3D scanner, or an industrial vision system, so as to capture the details of the surface of the target ALC environmentally friendly wallboard and obtain a wall surface image set. Then, a wall surface state identifier is used to perform defect recognition on the wall surface image set to determine the defect type (cracks, large-area cracks, local bulging, hollow shells, and falling off, etc.), defect degree, and defect location of the target ALC environmentally friendly wallboard surface, and obtain the multiple surface defect recognition results. The multiple surface defect recognition results include multiple defect locations.

[0029] Preferably, a plurality of sample wallboard surface image sets and a plurality of sample surface defect recognition results are obtained as training data, and supervised learning is performed on a framework constructed based on a convolutional neural network to learn the mapping relationship between the wallboard surface image and the surface defect recognition result, until the training converges, and the trained wall surface identifier is obtained. The wall surface identifier is used to efficiently, accurately and intelligently identify the wallboard surface defects reflected in the wallboard surface image.

[0030] Step S200 ensures accurate detection and positioning of surface defects of the target ALC environmentally friendly wallboard, lays the foundation for the selection of test points in subsequent life prediction, and analyzes the spatial distribution of defects.

[0031] S300: Performing hierarchical screening of test points based on the multiple defect positions and the multiple surface defect recognition results to determine K test points, where K is an integer greater than or equal to 1;

[0032] Further, based on the multiple defect positions and the multiple surface defect recognition results, the test point hierarchical screening is performed to determine K test points. The step S300 of the embodiment of the present application further includes:

[0033] Aggregating the plurality of surface defect recognition results of the same type to obtain a plurality of aggregated surface defect recognition result sets;

[0034] Counting the quantities included in the plurality of aggregated surface defect recognition result sets respectively to obtain a plurality of statistics;

[0035] Calculate the ratios of the multiple statistics to the sum of the multiple statistics respectively, and multiply the calculation results by the preset number of test points to obtain multiple test numbers, wherein the preset number of test points is K;

[0036] The multiple aggregate surface defect recognition result sets are screened based on the multiple test quantities, and the K test points are determined in combination with the multiple defect positions according to a mapping relationship between the screening results and the defect positions.

[0037] Furthermore, step S300 of the embodiment of the present application further includes:

[0038] Extracting a first aggregated surface defect identification result set from the plurality of aggregated surface defect identification result sets and matching a first test quantity from the plurality of test quantities;

[0039] Obtaining the maximum defect value and the minimum defect value in the first polymer surface defect recognition result set, and constructing a preset test interval in combination with the first test quantity;

[0040] Taking the first polymerized surface defect recognition result corresponding to the maximum defect value in the first polymerized surface defect recognition result set as the center, performing step-by-step diffusion iteration in the first polymerized surface defect recognition result according to the preset test interval, and determining to extract a plurality of first polymerized surface defect recognition results;

[0041] taking the center and the plurality of first aggregated surface defect recognition results as a first screening surface defect recognition result set;

[0042] The plurality of aggregated surface defect identification result sets are screened according to the plurality of test quantities to determine a plurality of screened surface defect identification result sets.

[0043] In one embodiment, in order to perform temperature testing on the target ALC environmentally friendly wall panels, thereby analyzing the time lag and attenuation of the wall panels at different defect locations, and providing reliable data support for the subsequent service life prediction of the wall panels, it is necessary to perform hierarchical screening of test points based on the multiple surface defect identification results and the multiple defect locations, wherein the hierarchical screening is to determine the different numbers of test points according to the proportion of different surface defect types, and then obtain the thermal insulation function that can reflect the target ALC environmentally friendly wall panels when surface defects exist, thereby achieving the technical effect of paving the way for subsequent service life prediction.

[0044] In a possible embodiment, surface defects are classified according to type, such as cracks, peeling, holes, etc., and surface defect results belonging to the same type are aggregated together to form multiple aggregated surface defect identification result sets. Then, the number of defects contained in the multiple aggregated surface defect identification result sets is counted respectively to obtain multiple statistics. The larger the statistic, the higher the frequency of the corresponding defect type in the target ALC environmentally friendly wallboard, and the more attention it needs to pay when predicting the service life.

[0045] Calculate the ratio of the multiple statistics to the sum of the multiple statistics respectively, and multiply the calculation result by the preset number of test points (the total number of tests for temperature testing pre-set by technicians in this field) to obtain multiple test numbers, where the preset number of test points is K.

[0046] The plurality of aggregated surface defect recognition result sets are screened based on the plurality of test numbers. The greater the number of tests, the greater the number of defect positions corresponding to the surface defect recognition results in the aggregated surface defect recognition result set that are selected as test points. The screening result includes a plurality of screened surface defect recognition result sets. Further, based on a one-to-one mapping relationship between the screening result and the defect position, the K test points are determined in combination with the plurality of defect positions.

[0047] Preferably, a first aggregate surface defect recognition result set is extracted from the plurality of aggregate surface defect recognition result sets, and the first test quantity in the plurality of test quantities is matched. The maximum defect value and the minimum defect value in the first aggregate surface defect recognition result set are obtained, and a preset test interval is constructed in combination with the first test quantity. Optionally, the difference between the maximum defect value and the minimum defect value is calculated, and then the difference is placed in the first test quantity to determine the preset test interval. Thus, the surface defect situation of the target ALC environmentally friendly wallboard is comprehensively grasped.

[0048] Exemplarily, the first test quantity is 10, the minimum crack width is 0.2 mm, the maximum crack width is 3 mm, (3-0.2) / 10=0.28 (mm), and the preset test spacing is 0.28 mm.

[0049] The first aggregated surface defect identification result corresponding to the maximum defect value in the first aggregated surface defect identification result set is taken as the center, and gradually diffused outward from the center position according to the size of the preset test supervision to determine multiple first diffused defect values ​​of multiple first test points. The first aggregated surface defect identification results closest to the multiple first diffused defect values ​​in the first aggregated surface defect identification result set are summarized, and the first screened surface defect identification result set is obtained by combining the center. Based on the same principle, the multiple aggregated surface defect identification result sets are screened according to the multiple test quantities to determine multiple screened surface defect identification result sets.

[0050] For example, the first test quantity is 10, the minimum crack width is 0.2 mm, the maximum crack width is 3 mm, and the preset test spacing is 0.28 mm, then the 10 first diffusion defect values ​​are 2.72, 2.44, 2.16, 1.88, 1.6, 1.32, 1.04, 0.76, 0.48, and 0.2 respectively. The aggregate surface defect identification results closest to the 10 first diffusion defect values ​​and the maximum crack width in the first aggregate surface defect identification result set are counted as the first screening surface defect identification result.

[0051] S400: K inner temperature sensors and K outer temperature sensors are respectively arranged on the inner side and the outer side of the wallboard at the K test points, and heating lamps and cooling devices are arranged on the outer side of the wallboard at the K test points, and the heating lamps and cooling devices are regulated according to the application environment status information to simulate periodic temperature fluctuations in the wallboard application;

[0052] S500: extracting the temperatures collected by the K inner temperature sensors and the K outer temperature sensors to obtain K periodic inner temperature sets and K periodic outer temperature sets;

[0053] In one embodiment, a high-precision, fast-response temperature sensor is selected to capture the temperature fluctuation on the wallboard surface. Sensors are fixed on the inner surface of the wallboard at the K test points to ensure close contact with the wall so as to accurately measure the inner temperature. External temperature sensors are fixed on the outer surface of the wallboard at the K test points to monitor the direct impact of external environmental temperature changes on the wallboard.

[0054] Then, heating lamps and cooling devices are arranged on the outside of the wallboard at the K test points. The heating lamps and cooling devices are used to change the temperature of the outside of the wallboard at the test points. Preferably, the cooling device should be installed at a position that can effectively reduce the temperature of the outside of the wallboard. The distance and angle of the heating lamp should ensure that the test area can be evenly heated without causing local overheating or partial heating.

[0055] Preferably, the heating lamp and the cooling device are regulated according to the temperature fluctuation range in the application environment status information to simulate the periodic temperature fluctuation in the wallboard application. Exemplarily, to simulate the temperature fluctuation of 48 hours, the heating lamp is set to work for 8 hours, then the cooling device is used to work for 16 hours, then the heating lamp is controlled to work for 8 hours, and then the cooling device is used to work for 16 hours to cool the wallboard, thereby simulating the temperature changes during the day and night in two cycles.

[0056] In one embodiment, the inner and outer temperature data of K test points are recorded in real time by connecting the data acquisition system of the sensor. The sensor should operate at a sufficiently high sampling frequency to capture subtle fluctuations in temperature, especially during the conversion process of heating and cooling. The collected temperature data can be stored in a local data storage device or transmitted wirelessly to a server for further processing. Then, the local data storage device or server is connected to retrieve the collected temperature data to obtain the K periodic inner temperature sets and K periodic outer temperature sets. Among them, the K periodic inner temperature sets and the K periodic outer temperature sets reflect the changes in the inner and outer temperatures of the target ALC environmentally friendly wallboard at the K test points. The technical effect of providing data support for subsequent time lag analysis and attenuation analysis is achieved.

[0057] S600: performing curve fitting on the K periodic inner temperature sets and the K periodic outer temperature sets respectively to obtain K periodic inner temperature fluctuation curves and K periodic outer temperature fluctuation curves, and performing mapping analysis on the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves to determine K time lag factors and K attenuation factors;

[0058] Further, curve fitting is performed on the K periodic inner temperature sets and the K periodic outer temperature sets respectively to obtain K periodic inner temperature fluctuation curves and K periodic outer temperature fluctuation curves, and mapping analysis is performed on the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves to determine K time lag factors and K attenuation factors. Step S600 of the embodiment of the present application also includes:

[0059] Acquire a preset fitting model, and use the preset fitting model to fit the K periodic inner temperature sets and the K periodic outer temperature sets to obtain the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves;

[0060] Extracting the extreme values ​​of the K periodic outer temperature fluctuation curves to obtain K periodic outer extreme value sequences, wherein the K periodic outer extreme value sequences include K periodic outer extreme value time point sequences;

[0061] Extracting the extreme values ​​of the K periodic inner temperature fluctuation curves to obtain K periodic inner extreme value sequences, wherein the K periodic inner extreme value sequences include K periodic inner extreme value time point sequences;

[0062] Performing a mapping time difference analysis based on the K periodic outer extreme value time point sequences and the K periodic inner extreme value time point sequences, determining K initial time lag factor sets, and performing mean processing on the K initial time lag factor sets to obtain K time lag factors;

[0063] Based on the K periodic outer extreme value sequences and the K periodic inner extreme value sequences, a mapping temperature value attenuation analysis is performed to determine K initial attenuation factor sets, and the K initial attenuation factor sets are averaged to obtain K attenuation factors.

[0064] Furthermore, step S600 of the embodiment of the present application further includes:

[0065] Construct the preset fitting model, wherein the preset fitting model includes a fitting formula, and the fitting formula is:

[0066] ;

[0067] in, is the periodic temperature fluctuation curve, A is the difference between the maximum and minimum values ​​of the periodic temperature set, is the fluctuation period, t is the temperature, and B is the average value of the periodic temperature set;

[0068] The K periodic inner temperature sets and the K periodic outer temperature sets are respectively input into the preset fitting model to obtain the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves.

[0069] In a possible embodiment, the temperature change on the outside is transmitted to the inside through the target ALC environmentally friendly wallboard, so the time that the temperature change on the inside lags behind the temperature change on the outside is used as a lag factor. The larger the lag factor, the better the thermal insulation performance of the target ALC environmentally friendly wallboard and the longer the service life. At the same time, when the temperature change on the outside is transmitted to the inside through the target ALC environmentally friendly wallboard, the degree of temperature fluctuation will be weakened, so the degree of weakening of the temperature fluctuation is used as an attenuation factor. The smaller the attenuation factor, the better the thermal insulation performance of the target ALC environmentally friendly wallboard.

[0070] By performing curve fitting on the K periodic inner temperature sets and the K periodic outer temperature sets respectively, the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves are obtained, thereby achieving the goal of facilitating subsequent time lag analysis and attenuation analysis. Among them, the K periodic inner temperature fluctuation curves reflect the fluctuation of the periodic changes of the inner temperature of the K test points. The K periodic outer temperature sets reflect the fluctuation of the periodic changes of the outer temperature of the K test points.

[0071] Preferably, the preset fitting model is constructed, wherein the preset fitting model includes a fitting formula, and the fitting formula is:

[0072] ;

[0073] in, is the periodic temperature fluctuation curve, A is the difference between the maximum and minimum values ​​of the periodic temperature set, is the fluctuation period, t is the temperature, and B is the average value of the periodic temperature set. Then, the K periodic inner temperature sets and the K periodic outer temperature sets are respectively input into the preset fitting model, and after model fitting, the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves are obtained.

[0074] According to the time sequence of the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves, since the temperature transfer has time continuity, the temperature points in the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves have a one-to-one mapping relationship. By mapping and analyzing the curves, the K time lag factors and K attenuation factors are determined. The technical effect of providing data support for subsequent service life prediction is achieved.

[0075] In one embodiment, the extreme values ​​(maxima or minima) of the K periodic outer temperature fluctuation curves are extracted to obtain K periodic outer extreme value sequences, wherein the K periodic outer extreme value sequences include K periodic outer extreme value time point sequences. A periodic outer extreme value time point is a time node at which a periodic outer extreme value appears. The extreme values ​​of the K periodic inner temperature fluctuation curves are extracted to obtain K periodic inner extreme value sequences, wherein the K periodic inner extreme value sequences include K periodic inner extreme value time point sequences. The K periodic inner extreme value sequences include K periodic inner extreme value time point sequences.

[0076] Mapping time difference analysis is performed based on the K periodic outer extreme value time point sequences and the K periodic inner extreme value time point sequences. Preferably, the time difference between the first periodic outer extreme value time point and the first periodic inner extreme value time point is calculated respectively, and the calculation results are used as K first initial time lag factors. Furthermore, the time difference between the second periodic outer extreme value time point and the second periodic inner extreme value time point is calculated respectively, and the calculation results are used as K second initial time lag factors. By analogy, K initial time lag factor sets are determined. After obtaining the K initial time lag factor sets, the K initial time lag factor sets are averaged to obtain K time lag factors. Among them, the K time lag factors reflect the average time lag of the K test points.

[0077] Furthermore, based on the magnitude of each extreme value in the K periodic outer extreme value sequences and the K periodic inner extreme value sequences, combined with the one-to-one mapping relationship between the extreme values ​​in the K periodic outer extreme value sequences and the K periodic inner extreme value sequences, a mapping temperature value attenuation analysis is performed to determine K initial attenuation factor sets. Then, the K initial attenuation factor sets are averaged to determine the average attenuation level of the temperature fluctuation amplitude of each test point, and K attenuation factors are obtained.

[0078] Further, based on the K periodic outer extreme value sequences and the K periodic inner extreme value sequences, a mapping temperature value attenuation analysis is performed to determine K initial attenuation factor sets. Step S600 of the embodiment of the present application further includes:

[0079] Calculate the difference between two adjacent maximum and minimum values ​​in the K periodic outer extreme value sequences and the K periodic inner extreme value sequences in order of precedence, and obtain K outer temperature fluctuation amplitude sequences and K inner temperature fluctuation amplitude sequences;

[0080] The inner temperature fluctuation amplitudes in the K inner temperature fluctuation amplitude sequences are sequentially compared with the outer temperature fluctuation amplitudes in the K outer temperature fluctuation amplitude sequences to obtain K initial attenuation factor sets.

[0081] In a possible embodiment, the difference between two adjacent maximum and minimum values ​​in the K periodic outer extreme value sequences and the K periodic inner extreme value sequences is calculated in order to obtain K outer temperature fluctuation amplitude sequences and K inner temperature fluctuation amplitude sequences. The outer temperature fluctuation amplitude is the temperature fluctuation difference between two adjacent periodic outer extreme values.

[0082] Furthermore, the inner temperature fluctuation amplitudes in the K inner temperature fluctuation amplitude sequences are sequentially compared with the outer temperature fluctuation amplitudes in the K outer temperature fluctuation amplitude sequences to obtain K initial attenuation factor sets, thereby achieving a technical effect of reliably analyzing the attenuation of the inner and outer temperature fluctuation amplitudes.

[0083] S700: performing factor density analysis on the K time lag factors and the K attenuation factors respectively to determine a test set time lag factor and a test set attenuation factor;

[0084] Further, such as Figure 2 As shown, factor density analysis is performed on the K time lag factors and the K attenuation factors respectively to determine the time lag factor in the test set and the attenuation factor in the test set. Step S700 of the embodiment of the present application further includes:

[0085] Calculating the mean values ​​of the K hysteresis factors and the K attenuation factors respectively to obtain a hysteresis factor mean value and an attenuation factor mean value;

[0086] Taking the mean of the lag factor and the mean of the attenuation factor as the analysis starting points respectively, iterating among the K lag factors and the K attenuation factors according to a preset lag search bandwidth and a preset attenuation search bandwidth, and determining a dense set of lag factors and a dense set of attenuation factors, wherein the preset lag search bandwidth is the difference between the lag factors of two adjacent iterations, and the preset attenuation search bandwidth is the difference between the attenuation factors of two adjacent iterations;

[0087] The dense set of time lag factors and the dense set of attenuation factors are traversed to determine the central value, and the time lag factor in the test set and the attenuation factor in the test set are obtained.

[0088] Further, taking the mean of the lag factor and the mean of the attenuation factor as the analysis starting point, respectively, iterating among the K lag factors and the K attenuation factors according to the preset lag search bandwidth and the preset attenuation search bandwidth, and determining a dense set of lag factors and a dense set of attenuation factors, step S700 of the embodiment of the present application further includes:

[0089] Taking the mean of the time lag factors as the analysis starting point, iterating among the K time lag factors according to the preset time lag search bandwidth to determine the search time lag factor;

[0090] Calculating the mean of the lag factor and the mean concentration density and the search concentration density of the search lag factor respectively based on the preset lag search bandwidth;

[0091] When the mean concentration density is less than or equal to the search concentration density, the search lag factor is used as the starting point of the stage, and the K lag factors are iterated continuously according to the lag search bandwidth until the concentration density of the previous iteration is greater than the concentration density of this iteration, and the search lag factor obtained in this iteration is used as the target lag factor;

[0092] Adding the lag factors whose factor differences from the K lag factors to the target lag factor are less than or equal to the preset lag search bandwidth into the lag factor dense set;

[0093] Taking the mean value of the attenuation factor as the analysis starting point, iterating through the K attenuation factors according to the preset attenuation search bandwidth to determine a target attenuation factor;

[0094] The attenuation factors whose factor differences from the target attenuation factor among the K attenuation factors are less than or equal to the preset attenuation search bandwidth are added into the dense set of attenuation factors.

[0095] In one embodiment, after obtaining the K time lag factors and the K attenuation factors, in order to analyze the overall time lag and overall attenuation of the target ALC environmentally friendly wallboard, it is necessary to screen out representative test set time lag factors and test set attenuation factors from the K time lag factors and K attenuation factors.

[0096] Preferably, the means of the K lag factors and the K attenuation factors are calculated respectively to obtain the lag factor means and the attenuation factor means. Further, the lag factor means and the attenuation factor means are respectively used as the analysis starting points, and the K lag factors and the K attenuation factors are iterated according to the preset lag search bandwidth and the preset attenuation search bandwidth to determine the lag factor dense set and the attenuation factor dense set, wherein the preset lag search bandwidth is the difference between the lag factors of two adjacent iterations, and the preset attenuation search bandwidth is the difference between the attenuation factors of two adjacent iterations.

[0097] The medians of the dense set of lag factors and the dense set of attenuation factors are extracted respectively, and the extracted results are used as the lag factors in the test set and the attenuation factors in the test set, thereby achieving the technical effect of obtaining representative data from the K lag factors and the K attenuation factors.

[0098] Specifically, the mean of the hysteresis factor is taken as the starting point of the analysis, and the K hysteresis factors are iterated according to the preset hysteresis search bandwidth to determine the search hysteresis factor. The difference between the search hysteresis factor and the mean of the hysteresis factor is the preset hysteresis search bandwidth. Furthermore, an area is constructed with the mean of the hysteresis factor as the center and the preset hysteresis search bandwidth as the radius, and the number of hysteresis factors contained in the area is counted, and the statistical result is divided by the area of ​​the area, and the calculated result is used as the mean concentration density. Based on the same principle, the search concentration density of the search hysteresis factor is calculated according to the preset hysteresis search bandwidth.

[0099] The mean concentration density reflects the density of the hysteresis factors clustered around the mean of the hysteresis factors, and the search concentration density reflects the density of the hysteresis factors clustered around the search hysteresis factor.

[0100] When the mean concentration density is less than or equal to the search concentration density, it indicates that the search lag factor is more representative. At this time, the search lag factor is used as the starting point of the stage, and the K lag factors are iterated according to the lag search bandwidth until the concentration density of the previous iteration is greater than the concentration density of this iteration, indicating that the edge of the area with denser distribution has been reached, and the search lag factor obtained in this iteration is used as the target lag factor.

[0101] Furthermore, the lag factors whose factor difference to the target lag factor among the K lag factors is less than or equal to the preset lag search bandwidth are added to the dense set of lag factors. Based on the same principle as that of obtaining the dense set of lag factors, the mean of the attenuation factors is used as the starting point of the analysis, and the K attenuation factors are iterated according to the preset attenuation search bandwidth to determine the target attenuation factor. Then, the attenuation factors whose factor difference to the target attenuation factor among the K attenuation factors is less than or equal to the preset attenuation search bandwidth are added to the dense set of attenuation factors. The technical effect of intensively screening the K lag factors and the K attenuation factors is achieved.

[0102] S800: calling a service life predictor to analyze the time lag factor in the test set and the attenuation factor in the test set to determine a service life prediction result of the target ALC environmentally friendly wallboard.

[0103] In one embodiment, the service life predictor is used to predict and analyze the service life of the target ALC environmentally friendly wallboard from two dimensions: the time lag factor and the concentrated attenuation factor. The test concentrated time lag factor and the concentrated attenuation factor are input into the trained service life predictor for analysis to obtain the service life prediction result. The service life prediction result reflects the service life of the target ALC environmentally friendly wallboard.

[0104] Preferably, multiple sample test set time lag factors, multiple sample test set attenuation factors and multiple sample service life prediction results are obtained as training data, wherein the multiple sample service life prediction results have labels, and supervised training is performed on the framework constructed based on the feedforward neural network, and multiple sample service life prediction results with labels are used for verification during the training process, and the many-to-one mapping relationship between the test set time lag factor and the test set attenuation factor and the service life prediction result is learned until the training converges, and the service life predictor that has been trained is obtained. The technical effect of improving the accuracy of service life prediction is achieved.

[0105] In summary, the embodiments of the present application have at least the following technical effects:

[0106] This application can accurately identify surface defects, determine test points that fit the actual situation, and conduct in-depth analysis of temperature transfer characteristics and accurate life prediction through reliable simulation of the actual environment. It has achieved the technical effect of improving the efficiency and accuracy of prediction and is applicable to the life management of ALC environmentally friendly wallboards in different environments.

[0107] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0109] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for predicting the service life of an ALC green environmentally friendly wallboard, characterized in that: The method comprises: Acquire application environment status information of the target ALC environmentally friendly wallboard and construct the first usage constraint; Collecting surface images of the target ALC environmentally friendly wallboard, and using a wall surface state identifier to perform defect identification on the wallboard surface image set obtained by the collection, and determining a plurality of surface defect identification results, wherein the plurality of surface defect identification results include a plurality of defect locations; Performing hierarchical screening of test points based on the multiple defect positions and the multiple surface defect recognition results to determine K test points, where K is an integer greater than or equal to 1; K inner temperature sensors and K outer temperature sensors are respectively arranged on the inner side and the outer side of the wallboard at the K test points, and heating lamps and cooling devices are arranged on the outer side of the wallboard at the K test points, and the heating lamps and cooling devices are regulated according to the application environment status information to simulate the periodic temperature fluctuation in the wallboard application; By extracting the temperatures collected by the K inner temperature sensors and the K outer temperature sensors, K periodic inner temperature sets and K periodic outer temperature sets are obtained; Performing curve fitting on the K periodic inner temperature sets and the K periodic outer temperature sets respectively to obtain K periodic inner temperature fluctuation curves and K periodic outer temperature fluctuation curves, and performing mapping analysis on the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves to determine K time lag factors and K attenuation factors; Performing factor density analysis on the K time lag factors and the K attenuation factors respectively to determine a test set time lag factor and a test set attenuation factor; The service life predictor is called to analyze the time lag factor in the test set and the attenuation factor in the test set to determine the service life prediction result of the target ALC environmentally friendly wallboard.

2. A method for predicting the service life of an ALC green environmentally friendly wallboard as claimed in claim 1, characterized in that: Performing factor density analysis on the K time lag factors and the K attenuation factors respectively to determine the time lag factor in the test set and the attenuation factor in the test set, including: Calculating the mean values ​​of the K time lag factors and the K attenuation factors respectively to obtain a time lag factor mean value and an attenuation factor mean value; Taking the mean of the lag factor and the mean of the attenuation factor as the analysis starting points respectively, iterating among the K lag factors and the K attenuation factors according to a preset lag search bandwidth and a preset attenuation search bandwidth, and determining a dense set of lag factors and a dense set of attenuation factors, wherein the preset lag search bandwidth is the difference between the lag factors of two adjacent iterations, and the preset attenuation search bandwidth is the difference between the attenuation factors of two adjacent iterations; The dense set of time lag factors and the dense set of attenuation factors are traversed to determine the central value, and the time lag factor in the test set and the attenuation factor in the test set are obtained.

3. A method for predicting the service life of an ALC green environmentally friendly wallboard as claimed in claim 2, characterized in that: Taking the mean of the lag factor and the mean of the attenuation factor as the analysis starting point respectively, iterating among the K lag factors and the K attenuation factors according to the preset lag search bandwidth and the preset attenuation search bandwidth, and determining a dense set of lag factors and a dense set of attenuation factors, including: Taking the mean of the time lag factors as the analysis starting point, iterating among the K time lag factors according to the preset time lag search bandwidth to determine the search time lag factor; Calculating the mean of the lag factor and the mean concentration density and the search concentration density of the search lag factor respectively based on the preset lag search bandwidth; When the mean concentration density is less than or equal to the search concentration density, the search lag factor is used as the starting point of the stage, and the K lag factors are iterated continuously according to the lag search bandwidth until the concentration density of the previous iteration is greater than the concentration density of this iteration, and the search lag factor obtained in this iteration is used as the target lag factor; Adding the lag factors whose factor differences from the K lag factors to the target lag factor are less than or equal to the preset lag search bandwidth into the lag factor dense set; Taking the mean value of the attenuation factor as the analysis starting point, iterating through the K attenuation factors according to the preset attenuation search bandwidth to determine a target attenuation factor; The attenuation factors whose factor differences from the target attenuation factor among the K attenuation factors are less than or equal to the preset attenuation search bandwidth are added into the dense set of attenuation factors.

4. A method for predicting the service life of an ALC green environmentally friendly wallboard as claimed in claim 1, characterized in that: Curve fitting is performed on the K periodic inner temperature sets and the K periodic outer temperature sets respectively to obtain K periodic inner temperature fluctuation curves and K periodic outer temperature fluctuation curves, and mapping analysis is performed on the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves to determine K hysteresis factors and K attenuation factors, including: Acquire a preset fitting model, and use the preset fitting model to fit the K periodic inner temperature sets and the K periodic outer temperature sets to obtain the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves; Extracting the extreme values ​​of the K periodic outer temperature fluctuation curves to obtain K periodic outer extreme value sequences, wherein the K periodic outer extreme value sequences include K periodic outer extreme value time point sequences; Extracting the extreme values ​​of the K periodic inner temperature fluctuation curves to obtain K periodic inner extreme value sequences, wherein the K periodic inner extreme value sequences include K periodic inner extreme value time point sequences; Performing a mapping time difference analysis based on the K periodic outer extreme value time point sequences and the K periodic inner extreme value time point sequences, determining K initial time lag factor sets, and performing mean processing on the K initial time lag factor sets to obtain K time lag factors; Based on the K periodic outer extreme value sequences and the K periodic inner extreme value sequences, a mapping temperature value attenuation analysis is performed to determine K initial attenuation factor sets, and the K initial attenuation factor sets are averaged to obtain K attenuation factors.

5. A method for predicting the service life of an ALC green environmentally friendly wallboard as claimed in claim 4, characterized in that: include: Construct the preset fitting model, wherein the preset fitting model includes a fitting formula, and the fitting formula is: ; in, is the periodic temperature fluctuation curve, A is the difference between the maximum and minimum values ​​of the periodic temperature set, is the fluctuation period, t is the temperature, and B is the average value of the periodic temperature set; The K periodic inner temperature sets and the K periodic outer temperature sets are respectively input into the preset fitting model to obtain the K periodic inner temperature fluctuation curves and the K periodic outer temperature fluctuation curves.

6. A method for predicting the service life of an ALC green environmentally friendly wallboard as claimed in claim 5, characterized in that: Based on the K periodic outer extreme value sequences and the K periodic inner extreme value sequences, a mapping temperature value attenuation analysis is performed to determine K initial attenuation factor sets, including: Calculate the difference between two adjacent maximum and minimum values ​​in the K periodic outer extreme value sequences and the K periodic inner extreme value sequences in order of precedence, and obtain K outer temperature fluctuation amplitude sequences and K inner temperature fluctuation amplitude sequences; The inner temperature fluctuation amplitudes in the K inner temperature fluctuation amplitude sequences are sequentially compared with the outer temperature fluctuation amplitudes in the K outer temperature fluctuation amplitude sequences to obtain K initial attenuation factor sets.

7. The method for predicting the service life of an ALC green environmentally friendly wallboard according to claim 1, characterized in that: Performing hierarchical screening of test points based on the multiple defect positions and the multiple surface defect recognition results to determine K test points includes: Aggregating the plurality of surface defect recognition results of the same type to obtain a plurality of aggregated surface defect recognition result sets; Counting the quantities included in the plurality of aggregated surface defect recognition result sets respectively to obtain a plurality of statistics; Calculate the ratios of the multiple statistics to the sum of the multiple statistics respectively, and multiply the calculation results by the preset number of test points to obtain multiple test numbers, wherein the preset number of test points is K; The multiple aggregate surface defect recognition result sets are screened based on the multiple test quantities, and the K test points are determined in combination with the multiple defect positions according to a mapping relationship between the screening results and the defect positions.

8. A method for predicting the service life of an ALC green environmentally friendly wallboard as claimed in claim 7, characterized in that: include: Extracting a first aggregated surface defect identification result set from the plurality of aggregated surface defect identification result sets and matching a first test quantity from the plurality of test quantities; Obtaining the maximum defect value and the minimum defect value in the first polymer surface defect recognition result set, and constructing a preset test interval in combination with the first test quantity; Taking the first polymerized surface defect recognition result corresponding to the maximum defect value in the first polymerized surface defect recognition result set as the center, performing step-by-step diffusion iteration in the first polymerized surface defect recognition result according to the preset test interval, and determining to extract a plurality of first polymerized surface defect recognition results; taking the center and the plurality of first aggregated surface defect recognition results as a first screening surface defect recognition result set; The plurality of aggregated surface defect identification result sets are screened according to the plurality of test quantities to determine a plurality of screened surface defect identification result sets.

Images