Intelligent maintenance method and equipment for concrete construction

Abstract

The invention provides an intelligent maintenance method and equipment for concrete construction, and belongs to the technical field of concrete construction. According to the method, a concrete construction data set corresponding to each monitoring equipment group can be determined; determining a quality risk score according to the concrete construction data set and a preset quality risk assessment algorithm; based on the concrete construction data set, each quality risk score and a GBDT model, determining a maintenance resource distribution level of each preset monitoring grid region, so as to control the maintenance equipment to execute a maintenance strategy corresponding to the maintenance resource distribution level; determining maintenance resource distribution deviation information corresponding to each preset monitoring grid region according to each maintenance resource distribution level and a corresponding preset resource distribution level curve; and performing clustering analysis on the maintenance resource distribution deviation information based on a preset deviation difference degree analysis model, and determining a to-be-intervened maintenance grid region in the concrete construction monitoring region, so as to generate corresponding intervened maintenance prompt information and send the information to a maintenance management terminal.

Description

Technical Field

[0001] The present application relates to the technical field of concrete construction, and in particular to an intelligent maintenance method and equipment for concrete construction. Background Art

[0002] Concrete construction occupies a core position in construction projects, and its quality is directly related to the safety and durability of building structures. Maintenance, as a key step in concrete construction, has a decisive influence on the development of concrete performance. The formulation of traditional maintenance strategies mainly relies on the experience of staff and lacks accurate data support. This leads to unstable maintenance effects, especially when dealing with complex situations, such as hydration heat problems in large-volume concrete construction. It is difficult to accurately judge maintenance measures based on experience alone, which may cause excessive temperature stress, resulting in cracks and seriously affecting concrete performance.

[0003] Moreover, the unreasonable allocation of maintenance resources is also a major drawback of traditional maintenance methods. In large-scale construction projects, concrete in different areas has differences in structure, size and environmental conditions, but traditional methods often adopt a unified maintenance resource allocation method, resulting in uneven resource allocation, excess resources in some areas causing waste, and insufficient resources in some areas affecting concrete quality. Traditional maintenance methods are difficult to apply to increasingly complex building structures and high requirements for concrete construction quality.

[0004] Based on this, there is an urgent need for a technical solution for efficient, accurate and intelligent maintenance of concrete during concrete construction. Summary of the invention

[0005] To solve the above problems, the embodiments of the present application provide an intelligent maintenance method and equipment for concrete construction.

[0006] On the one hand, an embodiment of the present application provides an intelligent maintenance method for concrete construction, the method comprising:

[0007] Determine the concrete construction data set corresponding to each monitoring equipment group; wherein one of the monitoring equipment groups is used to collect maintenance monitoring parameters within a preset monitoring grid area of ​​the concrete construction monitoring area;

[0008] Determine the quality risk score of the corresponding preset monitoring grid area according to the concrete construction data set and the preset quality risk assessment algorithm;

[0009] Based on the concrete construction data set, each of the quality risk scores and the pre-trained gradient boosting decision tree (GBDT) model, determine the maintenance resource allocation level of each of the preset monitoring grid areas to control the maintenance equipment to execute the maintenance strategy corresponding to the maintenance resource allocation level;

[0010] Determine the maintenance resource allocation deviation information corresponding to each of the preset monitoring grid areas according to each of the maintenance resource allocation levels and the corresponding preset resource allocation level curve;

[0011] Based on the preset deviation difference analysis model, cluster analysis is performed on the maintenance resource allocation deviation information to determine the grid area to be intervened for maintenance in the concrete construction monitoring area, so as to generate corresponding intervention maintenance prompt information and send it to the maintenance management terminal.

[0012] In one implementation of the present application, determining the concrete construction data set corresponding to each monitoring equipment group specifically includes:

[0013] Acquire the initial monitoring data group collected by each monitoring device group for the corresponding preset monitoring network area; wherein the preset monitoring grid area is obtained by dividing the grid units with a shape similar to that of the concrete construction monitoring area; the initial monitoring data group includes at least the following maintenance monitoring parameters: stress value, strain value, hydration heat value, ambient temperature value and ambient humidity value;

[0014] Input each of the initial monitoring data groups into a pre-trained Gaussian Markov random field GMRF model to determine the maintenance difference feature matrix and fused monitoring data between each of the preset monitoring grid areas according to the model output result; wherein the maintenance difference feature matrix includes the spatial correlation covariance matrix between the initial monitoring data groups of each of the preset monitoring grid areas; the fused monitoring data includes the monitoring data group adjusted according to the spatial correlation between each of the initial monitoring data groups;

[0015] The maintenance difference feature matrix, the fused monitoring data and the initial monitoring data group are merged to construct the concrete construction data set.

[0016] In one implementation of the present application, the quality risk score of the preset monitoring grid area is determined according to the concrete construction data set and the preset quality risk assessment algorithm, specifically including:

[0017] According to the concrete construction data set corresponding to the preset maintenance monitoring period, the mean value and standard deviation of each parameter corresponding to the first monitoring parameter group are calculated; wherein the first monitoring parameter group is composed of stress value, strain value and hydration heat value;

[0018] Input the corresponding parameters in the initial monitoring data group of the concrete construction data set at the current monitoring time, the mean value of each parameter, the standard deviation of each parameter, and the preset environmental factor impact index into the preset quality risk assessment algorithm to calculate the quality risk score of the corresponding preset monitoring grid area;

[0019] Among them, the formula of the preset quality risk assessment algorithm is as follows:

[0020]

[0021] Among them, R i represents the quality risk score of the i-th preset monitoring grid area; σ i represents the stress value of the i-th preset monitoring grid area at the current monitoring moment; μ σ represents the mean stress value of the i-th preset monitoring grid area during the preset maintenance monitoring period; σ σ represents the standard deviation of the stress value of the i-th preset monitoring grid area during the preset maintenance monitoring period; ε i represents the strain value of the i-th preset monitoring grid area at the current monitoring moment; μ ε represents the mean strain value of the u-th preset monitoring grid area in the preset maintenance monitoring period; σ ε represents the standard deviation of the strain value of the i-th preset monitoring grid area during the preset maintenance monitoring period; Q i represents the hydration heat value of the i-th preset monitoring grid area at the current monitoring moment; μ Q represents the average hydration heat value of the i-th preset monitoring grid area during the preset maintenance monitoring period; σ Q represents the standard deviation of the calorific value of hydration of the preset monitoring grid area in the preset maintenance monitoring period; e is a natural constant; λ is the influence coefficient of the preset environmental factor; E i represents the preset environmental factor impact index of the i-th preset monitoring grid area at the current monitoring moment.

[0022] In one implementation of the present application, the preset environmental factor impact index is obtained based on the ambient temperature value, the ambient humidity value, and the preset environmental factor impact index calculation formula; wherein the preset environmental factor impact index calculation formula is specifically as follows:

[0023]

[0024] Among them, T i represents the ambient temperature value of the i-th preset monitoring grid area at the current monitoring moment; T opt represents the preset optimal ambient temperature value at the current monitoring time; T ad represents the preset adjustable ambient temperature range value at the current monitoring moment; H i represents the ambient humidity value of the i-th preset monitoring grid area at the current monitoring time; H optrepresents the preset optimal ambient humidity value at the current monitoring time; H ad represents the preset adjustable ambient humidity range value at the current monitoring moment; α is the preset ambient temperature influence weight; β is the preset ambient humidity influence weight.

[0025] In one implementation of the present application, based on the concrete construction data set, each of the quality risk scores and the pre-trained gradient boosting decision tree GBDT model, the maintenance resource allocation level of each of the preset monitoring grid areas is determined, specifically including:

[0026] Constructing a model input feature vector according to the concrete construction data set and each of the quality risk scores;

[0027] The model is input into the GBDT model of the feature vector to determine the maintenance resource demand prediction value according to the model output result; wherein the GBDT model is trained based on a number of historical maintenance data samples and their corresponding quality risk scores;

[0028] After determining that the maintenance resource demand forecast value is greater than the first risk threshold, determining that the maintenance resource allocation level is the first allocation level, so that the maintenance equipment performs a first maintenance strategy with a first heating power and a curing agent spraying cycle being a first spraying cycle;

[0029] After determining that the maintenance resource demand forecast value is less than or equal to the first risk threshold and greater than or equal to the second risk threshold, determining that the maintenance resource allocation level is a second allocation level, so that the maintenance equipment executes a second maintenance strategy of a second heating power and a curing agent spraying cycle of a second spraying cycle; wherein the second heating power is less than the first heating power, and the second spraying cycle is greater than the first spraying cycle;

[0030] After determining that the predicted value of the maintenance resource demand is less than the second risk threshold, the maintenance resource allocation level is determined to be the third allocation level, so that the maintenance equipment performs heating maintenance with a third heating duration and the maintenance agent spraying cycle is a third maintenance strategy with a third spraying cycle; wherein the third heating duration is less than the preset normal heating duration, the third spraying cycle is greater than the preset normal spraying cycle, and the preset normal spraying cycle is greater than the second spraying cycle.

[0031] In one implementation of the present application, the maintenance resource allocation deviation information corresponding to each of the preset monitoring grid areas is determined according to each of the maintenance resource allocation levels and the corresponding preset resource allocation level curve, specifically including:

[0032] According to the time correspondence, each maintenance resource allocation level is matched with the corresponding preset resource allocation level curve to determine whether there is an allocation level offset in each preset monitoring grid area; wherein the preset resource allocation level curve includes the correspondence between the maintenance resource allocation level and the maintenance monitoring time within a preset maintenance monitoring period;

[0033] If so, determine the allocation level offset between each of the preset monitoring grid areas, generate an allocation level offset vector corresponding to the preset monitoring grid area, and use the allocation level offset vector as the maintenance resource allocation deviation information.

[0034] In one implementation of the present application, before determining the maintenance resource allocation deviation information corresponding to each of the preset monitoring grid areas according to each of the maintenance resource allocation levels and the corresponding preset resource allocation level curves, the method further includes:

[0035] Obtaining historical maintenance resource allocation data corresponding to the concrete construction monitoring area through a preset concrete construction sample database;

[0036] Through a nonlinear regression model, the historical maintenance resource allocation data is curve-fitted to generate an initial resource allocation grade curve corresponding to each of the preset monitoring grid areas;

[0037] Determine a continuous maintenance resource allocation level sequence corresponding to each of the preset monitoring grid areas within a preset sampling time to generate a sampling resource allocation level curve;

[0038] Calculate the similarity of grade changes between the sampled resource allocation grade curve and the initial resource allocation grade curve, and determine that the initial resource allocation grade curve is the corresponding preset resource allocation grade curve when the grade change similarity is greater than a preset threshold; otherwise, add the historical maintenance resource allocation data to the continuous maintenance resource allocation grade sequence and refit the initial resource allocation grade curve.

[0039] In one implementation of the present application, cluster analysis is performed on each of the maintenance resource allocation deviation information based on a preset deviation difference analysis model to determine the maintenance grid area to be intervened in the concrete construction monitoring area, specifically including:

[0040] Input each maintenance resource allocation deviation information into the pre-trained preset deviation difference analysis model, so that the preset deviation difference analysis model traverses each maintenance resource allocation deviation information according to the preset field radius and the minimum number of samples to generate a corresponding plurality of clusters; wherein the preset deviation difference analysis model is a density-based clustering algorithm DBSCAN model; and one cluster includes at least one preset monitoring grid area;

[0041] According to the deviation mean and preset deviation threshold corresponding to each of the cluster clusters, the cluster cluster that meets the preset intervention maintenance condition is determined, and the preset monitoring grid area corresponding to the cluster cluster is used as the grid area to be intervened for maintenance; the preset intervention maintenance condition is that the deviation mean is greater than the preset deviation threshold; the preset deviation threshold is obtained based on the sum of the average value of each of the deviation means and the standard deviation of each of the deviation means of a preset multiple.

[0042] In one implementation of the present application, the method further includes:

[0043] According to the preset maintenance monitoring period, determine the maintenance period sub-interval corresponding to the current monitoring moment; wherein different maintenance period sub-intervals correspond to different data collection time intervals;

[0044] Determine the corresponding data collection time interval according to the maintenance period subinterval;

[0045] The next data collection time is determined according to the data collection time adjacent to the current monitoring time and the data collection time interval, so as to control the monitoring equipment group to obtain the maintenance monitoring parameters; the monitoring equipment group at least includes: strain gauges, temperature sensors, humidity sensors and hydration heat sensors arranged in a three-dimensional grid.

[0046] On the other hand, the embodiment of the present application further provides an intelligent maintenance device for concrete construction, the device comprising:

[0047] At least one processor; and a memory in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute an intelligent maintenance method for concrete construction as described above.

[0048] Compared with the prior art, the present invention has the following significant effects:

[0049] Through the above scheme, this application integrates multi-parameter monitoring technology to capture key data of concrete interior and environment in real time, providing a scientific basis for maintenance decision-making, significantly improving the accuracy and intelligence level of maintenance operations, reducing the reliance on manual experience, and making the maintenance process more scientific and efficient. This application quantitatively evaluates the quality risk of concrete, and intelligently adjusts the allocation of maintenance resources according to the evaluation results, realizes the optimal allocation of resources, avoids excessive or insufficient investment in resources, effectively reduces maintenance costs, and improves resource utilization efficiency. In addition, this application can also timely discover deviations in the maintenance process, use the preset deviation difference analysis model for precise positioning, generate targeted intervention maintenance prompts, and ensure the timeliness and effectiveness of maintenance measures. Furthermore, this application improves the intelligence and accuracy of concrete maintenance, reduces the reliance on manual experience, and realizes the reasonable allocation and efficient utilization of maintenance resources, providing a strong guarantee for the improvement of concrete construction quality and the safety and stability of construction projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0051] Figure 1 A schematic diagram of a process of an intelligent maintenance method for concrete construction in an embodiment of the present application;

[0052] Figure 2 This is a schematic diagram of the structure of an intelligent maintenance equipment for concrete construction in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. 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 making creative work are within the scope of protection of the present application.

[0054] The embodiments of the present application provide an intelligent maintenance method and equipment for concrete construction, which are used to solve technical problems existing in traditional concrete maintenance, such as extensiveness, over-reliance on manual experience, difficulty in efficient, accurate and intelligent maintenance, and unreasonable allocation of maintenance resources.

[0055] The following describes in detail various embodiments of the present application in conjunction with the accompanying drawings.

[0056] The present application embodiment provides an intelligent maintenance method for concrete construction, such as Figure 1 As shown, the method may include steps S101-S105:

[0057] S101, the server determines the concrete construction data set corresponding to each monitoring equipment group.

[0058] Among them, a monitoring equipment group is used to collect maintenance monitoring parameters within a preset monitoring grid area of ​​the concrete construction monitoring area.

[0059] It should be noted that the server, as the executor of the intelligent maintenance method for concrete construction, is only an example, and the executor is not limited to the server, and this application does not make specific restrictions on this. The executor can also be an electronic device connected to the maintenance equipment by wire or wireless; the maintenance equipment includes but is not limited to heating equipment, humidification equipment, and film-coating maintenance equipment.

[0060] In the embodiment of the present application, the above-mentioned determination of the concrete construction data set corresponding to each monitoring equipment group specifically includes:

[0061] Obtain the initial monitoring data group collected by each monitoring equipment group for the corresponding preset monitoring network area. The preset monitoring grid area is divided by grid units with a shape similar to that of the concrete construction monitoring area. The initial monitoring data group includes at least the following maintenance monitoring parameters: stress value, strain value, hydration heat value, ambient temperature value and ambient humidity value. Input each initial monitoring data group into a pre-trained Gaussian Markov random field GMRF model to determine the maintenance difference feature matrix and fused monitoring data between each preset monitoring grid area according to the model output result. The maintenance difference feature matrix includes the spatial correlation covariance matrix between the initial monitoring data groups of each preset monitoring grid area. The fused monitoring data includes the monitoring data group adjusted according to the spatial correlation between each initial monitoring data group. The maintenance difference feature matrix, the fused monitoring data and the initial monitoring data group are merged to construct a concrete construction data set.

[0062] That is to say, the present application arranges monitoring equipment groups for different preset monitoring grid areas. The monitoring equipment groups can be arranged in a three-dimensional grid, that is, each device in the same monitoring equipment group is placed on a three-dimensional grid node in the same preset monitoring grid area. The three-dimensional grid node can be a key area in the preset monitoring grid area, such as a beam-column node, a large-span panel, a wall corner, etc., which can be arranged by the user according to the actual use scenario, and the present application does not make specific restrictions on this. The server can obtain the maintenance monitoring parameters collected by the monitoring equipment group within the preset maintenance monitoring period to form an initial monitoring data group. The preset monitoring grid area is obtained by dividing the grid units with a shape similar to that of the concrete construction monitoring area. For example, if the concrete construction monitoring area is a rectangular parallelepiped, the grid unit is a small rectangular parallelepiped grid area with a similar shape. The concrete construction monitoring area is divided by an integer multiple of grid units to obtain an integer multiple of preset monitoring grid areas. For example, if the concrete construction monitoring area is divided by 10 grid units, the preset monitoring grid area is 10.

[0063] After the server obtains the initial monitoring data group of each preset monitoring grid area, the initial monitoring data group can be input into the pre-trained Gaussian Markov Random Field (GMRF) model for processing. The GMRF model predefines the neighborhood structure, constructs the covariance matrix, and then trains with several historical monitoring data group samples after data preprocessing (such as cleaning outliers and filling missing values). The trained GMRF model can output fused monitoring data (including maintenance monitoring parameters of different preset monitoring grid areas) after fusion of stress values, strain values, hydration heat values, ambient temperature values, and ambient humidity values. These data also need to be organized into a format acceptable to the GBDT model, such as a two-dimensional array or data frame, where each row represents a sample and each column represents a feature. And generate a spatial correlation covariance matrix (i.e., maintenance difference feature matrix), which is used to characterize the spatial correlation of data between each initial monitoring data group. Before being input into the Gradient Boosting Decision Tree (GBDT) model, the matrix needs to be processed by feature engineering, such as calculating the stress value difference and average value of adjacent preset monitoring grid areas. The server adds the maintenance difference feature matrix and fused monitoring data obtained above, and each of the initial monitoring data groups to the concrete construction data set, which can be stored in a database connected to the server.

[0064] Through the above scheme, the GMRF model can be used to fuse data, capture the spatial dependencies between data, and construct a data set containing spatial correlation relationships for subsequent reference.

[0065] In one embodiment of the present application, since the preset maintenance monitoring period may be relatively long, such as 7 days, 14 days, etc., in order to effectively acquire monitoring data and reduce invalid monitoring, the method further includes:

[0066] According to the preset maintenance monitoring period, determine the maintenance period sub-interval corresponding to the current monitoring moment. Different maintenance period sub-intervals correspond to different data collection time intervals. Determine the corresponding data collection time interval according to the maintenance period sub-interval. According to the data collection moment and data collection time interval adjacent to the current monitoring moment, determine the next data collection moment so as to control the monitoring equipment group to obtain the maintenance monitoring parameters. The monitoring equipment group at least includes: strain gauges, temperature sensors, humidity sensors and hydration heat sensors arranged in a three-dimensional grid.

[0067] That is to say, the present application can pre-divide the maintenance period sub-intervals according to the preset maintenance monitoring period, for example, the first day is the first maintenance period sub-interval, the second to third days are the second maintenance period sub-interval... At the same time, the data collection time monitoring is pre-specified for different first maintenance period sub-intervals, and the monitoring equipment group performs data collection according to the data collection time interval or the server performs steps S101-S105 of the present application according to the data collection time interval. The data collection time interval can be set by the user according to the actual usage scenario, and the present application does not make specific restrictions on this. The present application can determine the data collection time closest to the current monitoring time through the current monitoring time, and calculate the time difference between the current monitoring period and the most recent data collection period. Through the time difference and the data collection time interval, the next data collection time is calculated. For example, if the time difference is 30 minutes and the data collection time interval is 40 minutes, then the next data collection time is ten minutes later. The maintenance monitoring parameters collected by each sensor of the monitoring equipment group are obtained at the data collection time determined by the server.

[0068] It should be noted that the present application can execute S101-S105 in real time, or it can be executed according to the data collection time interval through the above embodiment. The specific setting can be made by the user according to the actual usage scenario and needs, and is not specifically limited here.

[0069] S102, the server determines the quality risk score of the corresponding preset monitoring grid area according to the concrete construction data set and the preset quality risk assessment algorithm.

[0070] In the embodiment of the present application, the quality risk score of the corresponding preset monitoring grid area is determined according to the concrete construction data set and the preset quality risk assessment algorithm, specifically including:

[0071] According to the concrete construction data set corresponding to the preset maintenance monitoring period, the mean value and standard deviation of each parameter corresponding to the first monitoring parameter group are calculated. Among them, the first monitoring parameter group consists of stress value, strain value and hydration heat value. The corresponding parameters and the mean value, standard deviation and influence index of each parameter in the initial monitoring data group of the concrete construction data set at the current monitoring time are input into the preset quality risk assessment algorithm to calculate the quality risk score of the corresponding preset monitoring grid area.

[0072] In other words, the present application can calculate the parameter mean and parameter standard deviation corresponding to the stress value, strain value and hydration heat value through a preset quality risk assessment algorithm and a concrete construction data set, and then calculate the quality risk score of the preset monitoring grid area.

[0073] Among them, the formula of the preset quality risk assessment algorithm is as follows:

[0074]

[0075] Among them, R i represents the quality risk score of the i-th preset monitoring grid area; σ i represents the stress value of the i-th preset monitoring grid area at the current monitoring time; μ σ represents the mean stress value of the i-th preset monitoring grid area during the preset maintenance monitoring period; σ σ represents the standard deviation of the stress value of the i-th preset monitoring grid area during the preset maintenance monitoring period; ε i represents the strain value of the i-th preset monitoring grid area at the current monitoring time; μ ε represents the mean strain value of the i-th preset monitoring grid area during the preset maintenance monitoring period; σ ε represents the standard deviation of the strain value of the i-th preset monitoring grid area during the preset maintenance monitoring period; Q i represents the hydration heat value of the i-th preset monitoring grid area at the current monitoring moment; μ Q represents the mean hydration heat value of the i-th preset monitoring grid area during the preset maintenance monitoring period; σ Q represents the standard deviation of the calorific value of hydration in the preset monitoring grid area during the preset maintenance monitoring period; e is a natural constant; λ is the influence coefficient of the preset environmental factor; E i represents the influence index of the preset environmental factor of the ith preset monitoring grid area at the current monitoring moment. The above stress value can be derived from the strain value obtained by the strain gauge.

[0076] The above preset environmental factor impact index is obtained based on the ambient temperature value, ambient humidity value, and the preset environmental factor impact index calculation formula; the preset environmental factor impact index calculation formula is as follows:

[0077]

[0078] Among them, T i represents the ambient temperature value of the i-th preset monitoring grid area at the current monitoring time; T opt Indicates the preset optimal ambient temperature value at the current monitoring time; T ad Indicates the preset adjustable ambient temperature range value at the current monitoring moment; H i represents the ambient humidity value of the i-th preset monitoring grid area at the current monitoring time; H opt Indicates the preset optimal ambient humidity value at the current monitoring time; H ad It represents the preset adjustable ambient humidity range value at the current monitoring moment; α is the preset ambient temperature influence weight; β is the preset ambient humidity influence weight.

[0079] In addition, the present application can preset a risk score threshold. When the quality risk score is greater than the preset risk score threshold, the early warning mechanism can be triggered to send an alarm message to the maintenance management terminal so that relevant personnel can take timely measures to intervene and handle the corresponding preset monitoring grid area to avoid the occurrence of quality accidents.

[0080] The quality risk of concrete can be quantitatively assessed through the above formula that takes into account structural performance parameters such as stress, strain, hydration heat value, and the impact of environmental factors on structural quality risk. At the same time, according to the calculated quality risk score, potential risk points in the concrete structure can be discovered in a timely manner. Quality risk assessment through the above formula can realize automatic data processing and real-time monitoring. This can not only reduce manual intervention and errors, but also improve management efficiency, allowing relevant personnel to understand the quality risk status of the structure more timely and accurately, and take corresponding management measures.

[0081] S103, the server determines the maintenance resource allocation level of each preset monitoring grid area based on the concrete construction data set, each quality risk score and the pre-trained gradient boosting decision tree (GBDT) model, so as to control the maintenance equipment to execute the maintenance strategy corresponding to the maintenance resource allocation level.

[0082] In the embodiment of the present application, based on the concrete construction data set, each quality risk score and the pre-trained gradient boosting decision tree GBDT model, the maintenance resource allocation level of each preset monitoring grid area is determined, specifically including:

[0083] According to the concrete construction data set and each quality risk score, the model input feature vector is constructed. The model is input into the feature vector GBDT model to determine the maintenance resource demand prediction value according to the model output results. Among them, the GBDT model is trained based on several historical maintenance data samples and their corresponding quality risk scores.

[0084] That is to say, the present application can train the initialized GBDT model (including setting model parameters, such as learning rate, number of iterations, number and depth of trees, etc.) in advance through several historical maintenance data samples and quality risk scores, and the corresponding pre-marked maintenance resource demand prediction values. The historical maintenance data samples and quality risk scores can be data after data cleaning, missing value processing, and feature selection. Before the GBDT model is trained, the historical maintenance data samples and quality risk scores are encoded into vectors, and the vectors contain sample elements encoded by the maintenance difference feature matrix, fusion monitoring data, and initial monitoring data groups. Through the training data, the GBDT model iteratively trains a series of decisions, and in each round of iteration, adjusts the new decision tree according to the prediction error of the current model until the output accuracy of the GBDT model is greater than the preset accuracy threshold. Subsequently, the server encodes the concrete construction data set and each quality risk score respectively, and constructs a model input feature vector corresponding to the preset monitoring grid area. The model input feature vector is predicted using the trained GBDT model to obtain the maintenance resource demand prediction value of the preset monitoring grid area.

[0085] Further, after determining that the predicted value of the maintenance resource demand is greater than the first risk threshold, the maintenance resource allocation level is determined to be the first allocation level, so that the maintenance equipment executes the first heating power and the curing agent spraying cycle is the first maintenance strategy of the first spraying cycle. For example, the power of the heating equipment is increased to the maximum, and the curing agent spraying cycle is shortened to 50% of the preset normal spraying cycle.

[0086] After determining that the predicted value of the maintenance resource demand is less than or equal to the first risk threshold and greater than or equal to the second risk threshold, the maintenance resource allocation level is determined to be the second allocation level, so that the maintenance equipment executes the second heating power and the curing agent spraying cycle is the second maintenance strategy of the second spraying cycle. The second heating power is less than the first heating power, and the second spraying cycle is greater than the first spraying cycle. For example, the power of the heating equipment is kept at a medium level, and the curing agent spraying cycle is shortened to 75% of the preset normal spraying cycle.

[0087] After determining that the predicted value of the maintenance resource demand is less than the second risk threshold, the maintenance resource allocation level is determined to be the third allocation level, so that the maintenance equipment performs heating maintenance for the third heating duration and the maintenance agent spraying cycle is the third maintenance strategy of the third spraying cycle. Among them, the third heating duration is less than the preset normal heating duration, the third spraying cycle is greater than the preset normal spraying cycle, and the preset normal spraying cycle is greater than the second spraying cycle. For example, the use time of the heating equipment is reduced to 70% of the preset normal heating duration, and the maintenance agent spraying cycle is extended to 120% of the preset normal spraying cycle.

[0088] Through the above scheme, the strategy of maintenance resource allocation can be reasonably set. Combined with the spatial correlation between the preset monitoring grid areas analyzed by GMRF, the mutual influence of maintenance between different areas can be obtained. The GBDT model can be used to further reasonably obtain the predicted value of maintenance resource demand, so as to obtain differentiated maintenance strategies more comprehensively and reasonably for concrete maintenance.

[0089] S104, the server determines the maintenance resource allocation deviation information corresponding to each preset monitoring grid area according to each maintenance resource allocation level and the corresponding preset resource allocation level curve.

[0090] In the embodiment of the present application, the maintenance resource allocation deviation information corresponding to each preset monitoring grid area is determined according to each maintenance resource allocation level and the corresponding preset resource allocation level curve, specifically including:

[0091] According to the time correspondence, each maintenance resource allocation level is matched with the corresponding preset resource allocation level curve to determine whether there is an allocation level offset in each preset monitoring grid area. The preset resource allocation level curve includes the correspondence between the maintenance resource allocation level and the maintenance monitoring time within the preset maintenance monitoring period. In the case of determining that there is an allocation level offset in each preset monitoring grid area, the allocation level offset between each preset monitoring grid area is determined, and an allocation level offset vector corresponding to the preset monitoring grid area is generated, and the allocation level offset vector is used as the maintenance resource allocation deviation information.

[0092] In other words, the server can match the maintenance resource allocation level preset by the curve from the preset resource allocation level curve according to the collection time (monitoring time) of the maintenance detection parameters corresponding to the maintenance resource allocation level, which is referred to as the preset level below. The server calculates whether the difference between the maintenance resource allocation level and the preset level is greater than the preset deviation, such as 1, which indicates that there is an allocation level offset. At this time, the server calculates the allocation level offset between each preset monitoring grid area. Among them, the calculation formula of the allocation level offset is as follows:

[0093]

[0094] Among them, O ij represents the allocation level offset between the ith preset monitoring grid area and the jth preset monitoring grid area; a is the first weight corresponding to the ith preset monitoring grid area, and is the allocation level offset change weight relative to its own preset level; f i The difference between the maintenance resource allocation level and the predetermined level for the i-th preset monitoring grid area; is the predetermined level of the i-th preset monitoring grid area; x1 is the first normalized weight for normalizing the difference between the maintenance resource allocation level and the predetermined level; b is the second weight corresponding to the j-th preset monitoring grid area, s+b=1; f j is the difference between the maintenance resource allocation level of the jth preset monitoring grid area and its corresponding predetermined level (obtained according to the preset resource allocation level curve corresponding to the jth preset monitoring grid area); x2 is the second normalized weight for normalizing the difference between the maintenance resource allocation level of the jth preset monitoring grid area and its corresponding predetermined level. The specific values ​​of a, b, x1, x2 and the preset deviation can be adjusted by the user according to the actual scenario and are not specifically limited here.

[0095] After obtaining the allocation level offset of a preset monitoring grid area relative to other preset monitoring grid areas, the allocation level offset vector corresponding to the preset monitoring grid area can be constructed according to the regional distance relationship, thereby obtaining the maintenance resource allocation deviation information.

[0096] In addition, in another embodiment of the present application, before determining the maintenance resource allocation deviation information corresponding to each preset monitoring grid area according to each maintenance resource allocation level and the corresponding preset resource allocation level curve, the method further includes:

[0097] The historical maintenance resource allocation data corresponding to the concrete construction monitoring area is obtained through the preset concrete construction sample database. The historical maintenance resource allocation data is curve-fitted through the nonlinear regression model to generate the initial resource allocation grade curve corresponding to each preset monitoring grid area. The continuous maintenance resource allocation grade sequence corresponding to each preset monitoring grid area within the preset sampling time is determined to generate the sampling resource allocation grade curve. The grade change similarity of the sampling resource allocation grade curve and the initial resource allocation grade curve is calculated to determine the initial resource allocation grade curve as the corresponding preset resource allocation grade curve when the grade change similarity is greater than the preset threshold. When the grade change similarity is not greater than the preset threshold, the historical maintenance resource allocation data is added to the continuous maintenance resource allocation grade sequence to refit the initial resource allocation grade curve.

[0098] That is to say, the present application pre-sets the historical maintenance resource allocation data stored in the concrete construction sample database, performs nonlinear regression processing, and fits to obtain the initial resource allocation grade curve. The server records the continuous maintenance resource allocation grade within the sampling time in the previous preset sampling time within the preset maintenance detection period, and generates a sampled resource allocation grade curve. Then, the curve similarity between the sampled resource allocation grade curve and the initial resource allocation grade curve is calculated, and the curve similarity is used as the grade change similarity. The greater the curve similarity, the more similar the characterization curve. The server can calculate by algorithms such as the inverse of the Euclidean distance and the Pearson correlation coefficient, and the present application does not make specific restrictions on this. Once the grade change similarity is greater than the preset threshold, the initial resource allocation grade curve is used as the preset resource allocation grade curve for the corresponding preset monitoring grid area. Otherwise, the initial resource allocation grade curve is refitted, and the refitted initial resource allocation grade curve is used as the preset resource allocation grade curve. The above preset threshold is set by the user according to actual use, and the present application does not make specific restrictions on this.

[0099] Through the above scheme, reliable resource allocation level reference data can be provided for the intelligent maintenance of the concrete construction monitoring area in the future, and a preset resource allocation level curve that best matches the actual construction scenario can be provided.

[0100] S105, the server performs cluster analysis on the deviation information of each maintenance resource allocation based on a preset deviation difference analysis model, determines the grid area to be intervened for maintenance in the concrete construction monitoring area, generates corresponding intervention maintenance prompt information and sends it to the maintenance management terminal.

[0101] The maintenance management terminal can be a mobile phone, computer or other device of maintenance-related personnel, and this application does not make any specific restrictions on this.

[0102] In the embodiment of the present application, cluster analysis is performed on the maintenance resource allocation deviation information based on the preset deviation difference analysis model to determine the maintenance grid area to be intervened in the concrete construction monitoring area, specifically including:

[0103] Input each maintenance resource allocation deviation information into a pre-trained preset deviation difference analysis model, so that the preset deviation difference analysis model traverses each maintenance resource allocation deviation information according to the preset field radius and the minimum number of samples to generate corresponding multiple clusters. Among them, the preset deviation difference analysis model is a density-based clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN) model. A cluster includes at least one preset monitoring grid area. According to the deviation mean and preset deviation threshold corresponding to each cluster, determine the cluster that meets the preset intervention maintenance condition, and use the preset monitoring grid area corresponding to the cluster as the maintenance grid area to be intervened. The preset intervention maintenance condition is that the deviation mean is greater than the preset deviation threshold. The preset deviation threshold is obtained based on the sum of the average value of each deviation mean and the standard deviation of each deviation mean of the preset multiple.

[0104] That is to say, this application is equipped with a DBSCAN model with pre-trained model parameters (including neighborhood radius and minimum number of samples). The server inputs the pre-processed maintenance resource allocation deviation information into the DBSCAN model, and then the algorithm traverses each point in the data set corresponding to the input data, and determines the category of the point based on the neighborhood radius and the minimum number of samples. If the number of samples of a point in its neighborhood is greater than or equal to the minimum number of samples, the point is a core point, and the density-connected points are divided into a cluster with the core point as the center; if a point is not a core point and is in the neighborhood of a core point, the point is a boundary point and belongs to the cluster where the corresponding core point is located; if a point is neither a core point nor in the neighborhood of any core point, the point is a noise point.

[0105] Subsequently, the server will also calculate the deviations between each cluster and obtain the mean deviation between the cluster and other clusters. If the mean deviation is greater than the preset deviation threshold preset by the user, it means that the cluster needs to be paid special attention, and the preset monitoring grid area corresponding to the cluster will be used as the maintenance grid area to be intervened. The preset deviation threshold can be the sum of the average value of the deviation means corresponding to all clusters and the standard deviation of the deviation means corresponding to all clusters with a preset multiple (such as twice). The preset deviation threshold can also be other setting methods, which are specifically specified by the user, and this application does not make specific limitations on this.

[0106] Through the above scheme, this application integrates multi-parameter monitoring technology to capture key data of concrete interior and environment in real time, providing a scientific basis for maintenance decision-making, significantly improving the accuracy and intelligence level of maintenance operations, reducing the reliance on manual experience, and making the maintenance process more scientific and efficient. This application quantitatively evaluates the quality risk of concrete, and intelligently adjusts the allocation of maintenance resources according to the evaluation results, realizes the optimal allocation of resources, avoids excessive or insufficient investment in resources, effectively reduces maintenance costs, and improves resource utilization efficiency. In addition, this application can also timely discover deviations in the maintenance process, use the preset deviation difference analysis model for precise positioning, generate targeted intervention maintenance prompts, and ensure the timeliness and effectiveness of maintenance measures. Furthermore, this application improves the intelligence and accuracy of concrete maintenance, reduces the reliance on manual experience, and realizes the reasonable allocation and efficient utilization of maintenance resources, providing a strong guarantee for the improvement of concrete construction quality and the safety and stability of construction projects.

[0107] Figure 2 A schematic diagram of the structure of an intelligent maintenance device for concrete construction provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the device includes:

[0108] At least one processor; and a memory in communication with the at least one processor. The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0109] Determine the concrete construction data set corresponding to each monitoring equipment group. Among them, one monitoring equipment group is used to collect maintenance monitoring parameters in a preset monitoring grid area of ​​the concrete construction monitoring area. According to the concrete construction data set and the preset quality risk assessment algorithm, determine the quality risk score of the corresponding preset monitoring grid area. Based on the concrete construction data set, each quality risk score and the pre-trained gradient boosting decision tree GBDT model, determine the maintenance resource allocation level of each preset monitoring grid area to control the maintenance equipment to execute the maintenance strategy corresponding to the maintenance resource allocation level. According to each maintenance resource allocation level and the corresponding preset resource allocation level curve, determine the maintenance resource allocation deviation information corresponding to each preset monitoring grid area. Based on the preset deviation difference analysis model, cluster analysis is performed on each maintenance resource allocation deviation information to determine the maintenance grid area to be intervened in the concrete construction monitoring area, so as to generate the corresponding intervention maintenance prompt information and send it to the maintenance management terminal.

[0110] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0111] The device and method provided in the embodiments of the present application correspond one to one, and therefore, the device also has similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the device will not be repeated here.

[0112] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0113] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An intelligent maintenance method for concrete construction, characterized in that: The method comprises: Determine the concrete construction data set corresponding to each monitoring equipment group; wherein one of the monitoring equipment groups is used to collect maintenance monitoring parameters within a preset monitoring grid area of ​​the concrete construction monitoring area; Determine the quality risk score of the corresponding preset monitoring grid area according to the concrete construction data set and the preset quality risk assessment algorithm; Based on the concrete construction data set, each of the quality risk scores and the pre-trained gradient boosting decision tree (GBDT) model, the maintenance resource allocation level of each of the preset monitoring grid areas is determined to control the maintenance equipment to execute the maintenance strategy corresponding to the maintenance resource allocation level; Determine the maintenance resource allocation deviation information corresponding to each of the preset monitoring grid areas according to each of the maintenance resource allocation levels and the corresponding preset resource allocation level curve; Based on the preset deviation difference analysis model, cluster analysis is performed on the maintenance resource allocation deviation information to determine the grid area to be intervened for maintenance in the concrete construction monitoring area, so as to generate corresponding intervention maintenance prompt information and send it to the maintenance management terminal.

2. The intelligent maintenance method for concrete construction according to claim 1, characterized in that: Determine the concrete construction data set corresponding to each monitoring equipment group, including: Acquire the initial monitoring data group collected by each monitoring device group for the corresponding preset monitoring network area; wherein the preset monitoring grid area is obtained by dividing the grid units with a shape similar to that of the concrete construction monitoring area; the initial monitoring data group includes at least the following maintenance monitoring parameters: stress value, strain value, hydration heat value, ambient temperature value and ambient humidity value; Input each of the initial monitoring data groups into a pre-trained Gaussian Markov random field GMRF model to determine the maintenance difference feature matrix and fused monitoring data between each of the preset monitoring grid areas according to the model output result; wherein the maintenance difference feature matrix includes the spatial correlation covariance matrix between the initial monitoring data groups of each of the preset monitoring grid areas; the fused monitoring data includes the monitoring data group adjusted according to the spatial correlation between each of the initial monitoring data groups; The maintenance difference feature matrix, the fused monitoring data and the initial monitoring data group are merged to construct the concrete construction data set.

3. The intelligent maintenance method for concrete construction according to claim 1, characterized in that: Determine the quality risk score of the corresponding preset monitoring grid area according to the concrete construction data set and the preset quality risk assessment algorithm, specifically including: According to the concrete construction data set corresponding to the preset maintenance monitoring period, the mean value and standard deviation of each parameter corresponding to the first monitoring parameter group are calculated; wherein the first monitoring parameter group is composed of stress value, strain value and hydration heat value; Input the corresponding parameters in the initial monitoring data group of the concrete construction data set at the current monitoring time, the mean value of each parameter, the standard deviation of each parameter, and the preset environmental factor impact index into the preset quality risk assessment algorithm to calculate the quality risk score of the corresponding preset monitoring grid area; Among them, the formula of the preset quality risk assessment algorithm is as follows: Among them, R i represents the quality risk score of the i-th preset monitoring grid area; σ i represents the stress value of the i-th preset monitoring grid area at the current monitoring moment; μ σ represents the mean stress value of the i-th preset monitoring grid area during the preset maintenance monitoring period; σ σ represents the standard deviation of the stress value of the i-th preset monitoring grid area during the preset maintenance monitoring period; ε i represents the strain value of the i-th preset monitoring grid area at the current monitoring moment; μ ε represents the mean strain value of the i-th preset monitoring grid area during the preset maintenance monitoring period; σ ε represents the standard deviation of the strain value of the i-th preset monitoring grid area during the preset maintenance monitoring period; Q i represents the hydration heat value of the i-th preset monitoring grid area at the current monitoring moment; μ Q represents the average calorific value of hydration of the i-th preset monitoring grid area during the preset maintenance monitoring period; σ Q represents the standard deviation of the calorific value of hydration of the i-th preset monitoring grid area during the preset maintenance monitoring period; e is a natural constant; λ is the influence coefficient of the preset environmental factor; E i represents the preset environmental factor impact index of the i-th preset monitoring grid area at the current monitoring moment.

4. The intelligent maintenance method for concrete construction according to claim 3, characterized in that: The preset environmental factor impact index is obtained based on the ambient temperature value, the ambient humidity value, and the preset environmental factor impact index calculation formula; wherein the preset environmental factor impact index calculation formula is specifically as follows: Among them, T i represents the ambient temperature value of the i-th preset monitoring grid area at the current monitoring moment; T opt represents the preset optimal ambient temperature value at the current monitoring time; T ad represents the preset adjustable ambient temperature range value at the current monitoring moment; H i represents the ambient humidity value of the i-th preset monitoring grid area at the current monitoring time; H opt represents the preset optimal ambient humidity value at the current monitoring time; H ad represents the preset adjustable ambient humidity range value at the current monitoring moment; α is the preset ambient temperature influence weight; β is the preset ambient humidity influence weight.

5. The intelligent maintenance method for concrete construction according to claim 1, characterized in that: Based on the concrete construction data set, each of the quality risk scores and the pre-trained gradient boosting decision tree (GBDT) model, the maintenance resource allocation level of each of the preset monitoring grid areas is determined, specifically including: Constructing a model input feature vector according to the concrete construction data set and each of the quality risk scores; The model is input into the GBDT model of the feature vector to determine the maintenance resource demand prediction value according to the model output result; wherein the GBDT model is trained based on a number of historical maintenance data samples and their corresponding quality risk scores; After determining that the maintenance resource demand forecast value is greater than the first risk threshold, determining that the maintenance resource allocation level is the first allocation level, so that the maintenance equipment performs a first maintenance strategy with a first heating power and a curing agent spraying cycle being a first spraying cycle; After determining that the maintenance resource demand forecast value is less than or equal to the first risk threshold and greater than or equal to the second risk threshold, determining that the maintenance resource allocation level is a second allocation level, so that the maintenance equipment executes a second maintenance strategy of a second heating power and a curing agent spraying cycle of a second spraying cycle; wherein the second heating power is less than the first heating power, and the second spraying cycle is greater than the first spraying cycle; After determining that the predicted value of the maintenance resource demand is less than the second risk threshold, the maintenance resource allocation level is determined to be the third allocation level, so that the maintenance equipment performs heating maintenance with a third heating duration and the maintenance agent spraying cycle is a third maintenance strategy with a third spraying cycle; wherein the third heating duration is less than the preset normal heating duration, the third spraying cycle is greater than the preset normal spraying cycle, and the preset normal spraying cycle is greater than the second spraying cycle.

6. The intelligent maintenance method for concrete construction according to claim 1, characterized in that: According to each of the maintenance resource allocation levels and the corresponding preset resource allocation level curves, the maintenance resource allocation deviation information corresponding to each of the preset monitoring grid areas is determined, specifically including: According to the time correspondence, each maintenance resource allocation level is matched with the corresponding preset resource allocation level curve to determine whether there is an allocation level offset in each preset monitoring grid area; wherein the preset resource allocation level curve includes the correspondence between the maintenance resource allocation level and the maintenance monitoring time within a preset maintenance monitoring period; If so, determine the allocation level offset between each of the preset monitoring grid areas, generate an allocation level offset vector corresponding to the preset monitoring grid area, and use the allocation level offset vector as the maintenance resource allocation deviation information.

7. The intelligent maintenance method for concrete construction according to claim 6, characterized in that: Before determining the maintenance resource allocation deviation information corresponding to each of the preset monitoring grid areas according to each of the maintenance resource allocation levels and the corresponding preset resource allocation level curves, the method further includes: Obtaining historical maintenance resource allocation data corresponding to the concrete construction monitoring area through a preset concrete construction sample database; Through a nonlinear regression model, the historical maintenance resource allocation data is curve-fitted to generate an initial resource allocation grade curve corresponding to each of the preset monitoring grid areas; Determine a continuous maintenance resource allocation level sequence corresponding to each of the preset monitoring grid areas within a preset sampling time to generate a sampling resource allocation level curve; Calculate the similarity of grade changes between the sampled resource allocation grade curve and the initial resource allocation grade curve, and determine that the initial resource allocation grade curve is the corresponding preset resource allocation grade curve when the grade change similarity is greater than a preset threshold; otherwise, add the historical maintenance resource allocation data to the continuous maintenance resource allocation grade sequence and refit the initial resource allocation grade curve.

8. The intelligent maintenance method for concrete construction according to claim 1, characterized in that: Based on the preset deviation difference analysis model, cluster analysis is performed on the maintenance resource allocation deviation information to determine the maintenance grid area to be intervened in the concrete construction monitoring area, specifically including: Input each maintenance resource allocation deviation information into the pre-trained preset deviation difference analysis model, so that the preset deviation difference analysis model traverses each maintenance resource allocation deviation information according to the preset field radius and the minimum number of samples to generate a corresponding plurality of clusters; wherein the preset deviation difference analysis model is a density-based clustering algorithm DBSCAN model; and one cluster includes at least one preset monitoring grid area; According to the deviation mean and preset deviation threshold corresponding to each of the cluster clusters, the cluster cluster that meets the preset intervention maintenance condition is determined, and the preset monitoring grid area corresponding to the cluster cluster is used as the grid area to be intervened for maintenance; the preset intervention maintenance condition is that the deviation mean is greater than the preset deviation threshold; the preset deviation threshold is obtained based on the sum of the average value of each of the deviation means and the standard deviation of each of the deviation means of a preset multiple.

9. The intelligent maintenance method for concrete construction according to claim 1, characterized in that: The method further comprises: According to the preset maintenance monitoring period, determine the maintenance period sub-interval corresponding to the current monitoring moment; wherein different maintenance period sub-intervals correspond to different data collection time intervals; Determine the corresponding data collection time interval according to the maintenance period subinterval; The next data collection time is determined according to the data collection time adjacent to the current monitoring time and the data collection time interval, so as to control the monitoring equipment group to obtain the maintenance monitoring parameters; the monitoring equipment group at least includes: strain gauges, temperature sensors, humidity sensors and hydration heat sensors arranged in a three-dimensional grid.

10. An intelligent maintenance device for concrete construction, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute an intelligent maintenance method for concrete construction as described in any one of claims 1 to 9 above.

Images