Iot-based prefabricated building prefabricated component early warning method and system

Abstract

The application relates to a prefabricated component early warning method and system for fabricated buildings based on the Internet of Things. The method classifies Internet of Things data obtained by monitoring a target prefabricated component based on preset Internet of Things data classification information and inputs a corresponding combined learning model. Based on the combined learning model, a corresponding model prediction relationship and a correlation prediction relationship are obtained. Based on the model prediction relationship and the correlation prediction relationship, current prediction data of the target prefabricated component is obtained. The health state warning information of the target prefabricated component is triggered based on the current prediction data. The application has the effect of predicting the health state of the component to find potential problems in advance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of prefabricated component management, and in particular to an assembly type building prefabricated component early warning method and system based on the Internet of Things. BACKGROUND

[0002] At present, with the development of assembly type building prefabricated components, the effective management method of assembly type building prefabricated components has also developed rapidly. Among them, the comprehensive management method using the Internet of Things technology combined with a digital management platform is favored by most managers. Using the Internet of Things technology can realize the monitoring of the whole life cycle of the prefabricated component. Combined with the monitoring data and the digital management platform, the production of the assembly type building prefabricated component is optimized, and the economic benefit is improved.

[0003] In related technologies, the prefabricated component digital management platform usually accesses the Internet of Things data and monitors the prefabricated component through digital means. However, this kind of monitoring method can only be found after the state of the prefabricated component appears problems, and lacks the prediction of the health status of the prefabricated component, so it is difficult to find potential problems in advance. SUMMARY

[0004] In order to realize the prediction of the health status of the component to find potential problems in advance, the present application provides an assembly type building prefabricated component early warning method and system based on the Internet of Things.

[0005] In the first aspect, the above application aims to achieve the following technical solutions:

[0006] Based on the preset Internet of Things data classification information, the Internet of Things data obtained by monitoring the target prefabricated component is classified and input into the corresponding combined learning model;

[0007] Based on the combined learning model, the corresponding model prediction relationship and the associated prediction relationship are obtained;

[0008] Based on the model prediction relationship and the associated prediction relationship, the current prediction data of the target prefabricated component is obtained;

[0009] Based on the current prediction data, the health state warning information of the target prefabricated component is triggered.

[0010] By adopting the technical scheme, in the process of running the digital management platform, the data detected by the Internet of Things equipment for monitoring the target prefabricated component is transmitted to the digital management platform in a wireless or wired manner. The Internet of Things equipment can be a positioning tag installed on the target prefabricated component, a video monitoring device for monitoring the production process of the target prefabricated component, and an environmental sensor for monitoring the environmental parameters of the target prefabricated component. Different Internet of Things data formats and forms are different. Therefore, inputting the data of the Internet of Things equipment into the corresponding combined learning model facilitates more accurate learning of the characteristics of the target prefabricated component at different stages, and further realizes more accurate prediction. The model prediction relationship refers to a relationship for predicting the health status of the target prefabricated component by using the Internet of Things data of the current stage, and the correlation prediction relationship refers to a relationship for predicting the health status of the target prefabricated component by using the Internet of Things data of multiple stages. Therefore, by using the model prediction relationship and the correlation prediction relationship, real-time monitoring and prediction of the target prefabricated component are realized, and the accuracy of the prediction is improved, so that potential problems can be found in advance.

[0011] In a preferred example, the application can be further configured to obtain the Internet of Things data classification information in the following manner:

[0012] The Internet of Things data is classified based on a preset component whole-process cycle to obtain Internet of Things classification data. The component whole-process cycle refers to the whole-process cycle of the prefabricated component from design to installation and use.

[0013] By adopting the technical scheme, the preset component whole-process cycle refers to the whole-process cycle of the prefabricated component from design to installation and use. The component whole-process cycle includes the conceptual design stage, the structural design stage, the production preparation stage, the production manufacturing stage, the quality inspection stage, the component packaging stage, the component transportation stage, the on-site installation stage, the on-site inspection stage, and the operation and maintenance stage. Therefore, the Internet of Things data of the Internet of Things equipment is classified according to different stages to avoid confusion of the data, and analysis of the data of different stages divided by time facilitates prediction.

[0014] In a preferred example, the application can be further configured to obtain the combined learning model in the following manner:

[0015] Obtain historical Internet of Things data obtained by monitoring prefabricated components in the past, identify the historical Internet of Things data, and distribute the input to multiple independent learning models;

[0016] Generate a time-space model based on the historical Internet of Things data;

[0017] Combine the time-space model and the multiple independent learning models to obtain a combined learning model.

[0018] By adopting the technical solution, the IoT data monitored by the historical IoT device is identified before the production of a new batch of new-specification target prefabricated components each time, it is judged whether the health status of the actual target prefabricated component each time can be correctly predicted when a problem occurs, and whether the IoT data collected for prediction is related to the problem of the health status of the target prefabricated component, that is, whether correct data is input into the learning model. In this way, the learning model with high prediction accuracy at different stages is selected, and the IoT data collected when the prediction accuracy at different stages is high is set. The learning model suitable for different stages is allocated, and the time-space model is generated based on different stages of the whole process cycle of the component. The time-space model is based on time sequence, and the IoT data input at different stages is learned and analyzed correspondingly, so as to realize the prediction of the health status of the component.

[0019] In a preferred example, the application can be further configured to: based on the combined learning model, obtain a corresponding model prediction relationship and a correlation prediction relationship, specifically including:

[0020] Based on the multiple independent learning models in the combined learning model, multiple learning results are obtained;

[0021] Based on the time-space model and the multiple learning results in the combined learning model, a model prediction relationship and a correlation prediction relationship are obtained.

[0022] By adopting the above technical solution, the multiple learning results refer to the results of multiple learning models learning the IoT data at different stages. For example, the multiple learning results at the production preparation stage refer to the learning results of the multiple independent learning models corresponding to the production preparation stage. Therefore, at each stage, the corresponding multiple learning results are used to generate a model prediction relationship for predicting the health status of the target prefabricated component based on the IoT data at the current stage, and a correlation prediction relationship for predicting the health status of the target prefabricated component in combination with the IoT data at the current stage and the subsequent stage. In this way, the accuracy of prediction is improved through two predictions of the model prediction relationship and the correlation prediction relationship, so as to discover potential problems in advance.

[0023] In a preferred example, the application can be further configured to: the model prediction relationship includes a current stage critical threshold, the correlation prediction relationship includes a correlation stage critical threshold, the current stage critical threshold is greater than the correlation stage critical threshold, and the current prediction data of the target prefabricated component is obtained based on the model prediction relationship and the correlation prediction relationship, respectively, specifically including:

[0024] The Internet of Things data is compared with the current stage critical threshold of the model prediction relationship and the association stage critical threshold of the association prediction relationship respectively, to obtain the current prediction data of the target prefabricated component.

[0025] By adopting the above technical scheme, in the model prediction relationship and the association prediction relationship, the current stage critical threshold and the association stage critical threshold are included respectively, the current stage critical threshold is used to judge the health state of the target prefabricated component in the current stage, and the association stage critical threshold is used to judge the health state of the target prefabricated component in the subsequent stage, the current stage critical threshold is not less than the association stage critical threshold because the current stage critical threshold only relies on the Internet of Things data in the current stage, and the association stage critical threshold needs to combine the Internet of Things data in the current stage and the subsequent stage, therefore, the association prediction relationship corresponding to the association stage critical threshold needs to use a larger amount of data, and the prediction result of the general association prediction relationship will be more accurate, but in order to guarantee the prediction in advance and the prediction accuracy in the current stage, the current stage critical threshold with a smaller amount of data is not less than the association stage critical threshold, so as to improve the accuracy of the prediction data in the current stage.

[0026] In a second aspect, the above-mentioned application purpose is achieved by the following technical scheme:

[0027] A prefabricated component early warning system for fabricated buildings based on the Internet of Things, comprising:

[0028] A combined learning model updating module is configured to classify the Internet of Things data obtained by monitoring the target prefabricated component based on preset Internet of Things data classification information, and input the Internet of Things data into a corresponding combined learning model.

[0029] A relationship formula acquisition module is configured to obtain a model prediction relationship and an association prediction relationship based on the combined learning model.

[0030] A current prediction data acquisition module is configured to obtain current prediction data of the target prefabricated component based on the model prediction relationship and the association prediction relationship respectively.

[0031] A health state warning instruction module is configured to trigger health state warning information of the target prefabricated component based on the current prediction data.

[0032] Optionally, the prefabricated component early warning system for fabricated buildings based on the Internet of Things further comprises:

[0033] The Internet of Things classification data acquisition module is configured to classify Internet of Things data based on a preset component whole-process cycle to obtain Internet of Things classification data, wherein the component whole-process cycle refers to a whole-process cycle of a prefabricated component from design to installation and use.

[0034] Optionally, the prefabricated component early warning system based on the Internet of Things further comprises:

[0035] The multi-path learning model allocation module is configured to acquire historical Internet of Things data obtained by monitoring prefabricated components in the past, identify the historical Internet of Things data, and allocate the historical Internet of Things data to multiple independent learning models.

[0036] The time-space model generation module is configured to generate a time-space model based on the historical Internet of Things data.

[0037] The combined learning model combination module is configured to combine the time-space model and the multiple independent learning models to obtain a combined learning model.

[0038] In a third aspect, the above-mentioned application objectives are achieved by the following technical solutions:

[0039] A computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above-mentioned prefabricated component early warning method based on the Internet of Things when executing the computer program.

[0040] In a fourth aspect, the above-mentioned application objectives are achieved by the following technical solutions:

[0041] A computer-readable storage medium stores a computer program, and the computer program is executable on a processor to implement the steps of the above-mentioned prefabricated component early warning method based on the Internet of Things.

[0042] In summary, the present application has at least one of the following beneficial technical effects:

[0043] 1. In the process of running the digital management platform, the data detected by the Internet of Things equipment for monitoring the target prefabricated component is transmitted to the digital management platform through wireless or wired means. The Internet of Things equipment can be a positioning tag installed on the target prefabricated component, a video monitoring device for monitoring the production process of the target prefabricated component, and an environmental sensor for monitoring the environmental parameters of the target prefabricated component. Different Internet of Things data formats and forms are different, so inputting the data of the Internet of Things equipment into the corresponding combined learning model facilitates more accurate learning of the characteristics of the target prefabricated component at different stages, and further realizes more accurate prediction; the model prediction relationship refers to a relationship for predicting the health status of the target prefabricated component using the current stage of Internet of Things data, and the associated prediction relationship refers to a relationship for predicting the health status of the target prefabricated component using multiple stages of Internet of Things data, so that through the model prediction relationship and the associated prediction relationship, real-time monitoring and prediction of the target prefabricated component are realized, and the accuracy of the prediction is improved to discover potential problems in advance;

[0044] 2. The preset component whole process cycle refers to the whole process cycle of the prefabricated component from design to installation and use, wherein the component whole process cycle includes the conceptual design stage, the structural design stage, the production preparation stage, the production manufacturing stage, the quality inspection stage, the component packaging stage, the component transportation stage, the on-site installation stage, the on-site inspection stage and the operation and maintenance stage, so that the Internet of Things data of the Internet of Things equipment is classified according to different stages to avoid confusion of the data, and analysis of the data of different stages divided by time facilitates prediction;

[0045] 3. Before the production of each new batch of target prefabricated components of new specifications begins, the historical Internet of Things data monitored by the Internet of Things equipment is identified to determine whether the health status of each actual target prefabricated component can be correctly predicted when a problem occurs, and whether the predicted Internet of Things data is related to the problem of the health status of the target prefabricated component, that is, whether the correct data is input into the learning model. In this way, a learning model with a high prediction accuracy for different stages is selected, and the Internet of Things data collected when the prediction accuracy for different stages is high is set, different learning models are allocated to different stages, and a time-space model is generated based on different stages of the component whole process cycle. The time-space model is based on time sequence, and the Internet of Things data input at different stages is learned and analyzed accordingly, so as to realize prediction of the health status of the component. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 is an implementation flowchart of the prefabricated component early warning method based on the Internet of Things in the embodiment of the application;

[0047] Figure 2is an implementation flowchart of the method for obtaining the IoT data classification information in the embodiment of the present application;

[0048] Figure 3 is an implementation flowchart of the method for obtaining the combined learning model in the embodiment of the present application;

[0049] Figure 4 is an implementation flowchart of S20 of the prefabricated component early warning method for fabricated buildings based on the Internet of Things in the embodiment of the present application;

[0050] Figure 5 is an implementation flowchart of S30 of the prefabricated component early warning method for fabricated buildings based on the Internet of Things in the embodiment of the present application;

[0051] Figure 6 is a principle block diagram of the prefabricated component early warning system for fabricated buildings based on the Internet of Things in the embodiment of the present application;

[0052] Figure 7 is an internal structure diagram of the prefabricated component early warning computer device for fabricated buildings based on the Internet of Things in the embodiment of the present application. DETAILED DESCRIPTION

[0053] The following will be described in detail with reference to the accompanying drawings. Figures 1-7 The present application will be further described in detail.

[0054] In an embodiment, as shown in the accompanying drawings, Figure 1 The present application discloses a prefabricated component early warning method for fabricated buildings based on the Internet of Things, which specifically comprises the following steps:

[0055] S10: Based on the preset IoT data classification information, the IoT data obtained by monitoring the target prefabricated component is classified and input into the corresponding combined learning model.

[0056] In the embodiment, the digital management platform is used to collect the data detected by the IoT device for monitoring the target prefabricated component, the target prefabricated component is a prefabricated concrete component, the IoT device includes a positioning tag installed on the target prefabricated component, a video monitoring device for monitoring the production process of the target prefabricated component, and an environmental sensor for monitoring the environmental parameters of the target prefabricated component, and the IoT data refers to the data detected and obtained by the IoT device. The digital management platform is also used to obtain other data and information of the non-IoT device detection data of the target prefabricated component, and the obtaining method can be associated with other systems, for example, a target prefabricated component design system and a target prefabricated component three-dimensional building information model. The IoT data classification information refers to the IoT data obtained in different stages of the whole life cycle of the target prefabricated component. The combined learning model refers to a learning model for predicting the health status of the target prefabricated component.

[0057] Specifically, in the process of running the digital management platform, the data detected by the Internet of Things device for monitoring the target prefabricated component is transmitted to the digital management platform in a wireless or wired manner. For example, the Internet of Things data of the video monitoring device in the production stage of the target prefabricated component can be transmitted to the digital management platform in a wired manner, the Internet of Things data of the positioning tag in the transportation stage can be transmitted to the digital management platform in a wireless manner, and the preset Internet of Things data classification information is classification information obtained by dividing the data acquired by different Internet of Things devices and different systems at different time points according to different stages of the whole life cycle of the target prefabricated component. Therefore, in the process of running the digital management platform, based on the Internet of Things data classification information, the Internet of Things data of the target prefabricated component monitored by different Internet of Things devices at different time points and the data acquired by different systems are classified and input into a corresponding combined learning model. The combined learning model refers to a learning model for predicting the health state of the target prefabricated component. The combined learning model refers to a big data model capable of autonomous learning to achieve a prediction task. Therefore, continuously inputting the Internet of Things data to train the combined learning model can improve the prediction accuracy of the combined learning model.

[0058] S20: obtaining a corresponding model prediction relationship and a correlation prediction relationship based on the combined learning model.

[0059] In this embodiment, the model prediction relationship refers to a relationship for predicting the health state of the target prefabricated component based on the Internet of Things data of the current stage. The correlation prediction relationship refers to a relationship for predicting the health state of the target prefabricated component in combination with the Internet of Things data of the current stage and the subsequent stage.

[0060] Specifically, the combined learning model refers to a big data model capable of autonomous learning to achieve a prediction task. The combined learning model has been trained and learned with a large amount of data before being put into use, and has a prediction capability. Therefore, when the Internet of Things data is input into the combined learning model, that is, when the Internet of Things data acquired at a corresponding stage is input into the combined learning model, according to a specific time point and data content in the Internet of Things data, the combined learning model is used to generate a relational expression corresponding to the specific time point for predicting the health status of the target prefabricated component based on the data content, that is, a model prediction relational expression. For example, in the production stage of the target prefabricated component, the specific time point is included in the Internet of Things data, and the data content is video data of a video monitoring device. Therefore, the generated model prediction relational expression is a relational expression for predicting the health status of the target prefabricated component by judging the state of the target prefabricated component in each frame of image of the video data, such as the information in each frame of image of the video data based on the surface coating integrity of the target prefabricated component, whether there is a crack on the surface of the target prefabricated component, or whether the size of the target prefabricated component is normal. When the surface coating integrity of the target prefabricated component is less than a corresponding threshold value, there is a crack on the surface of the target prefabricated component, or the size of the target prefabricated component is not normal, the health status of the target prefabricated component is predicted to be poor by the model prediction relational expression. In addition, by using the combined learning model, a relational expression for predicting the health status of the target prefabricated component in combination with the current Internet of Things data and subsequent Internet of Things data is generated, that is, a correlation prediction relational expression. For example, in the production stage of the target prefabricated component, the specific time point is included in the Internet of Things data, and the data content is video data of a video monitoring device. Therefore, the generated correlation prediction relational expression is a relational expression for predicting the health status of the target prefabricated component by the state of the target prefabricated component in the current video data of each frame of image and the subsequent Internet of Things data, such as the information in each frame of image of the video data based on the surface coating integrity of the target prefabricated component, whether there is a crack on the surface of the target prefabricated component, or whether the size of the target prefabricated component is normal. When the surface coating integrity of the target prefabricated component is less than a corresponding threshold value, there is a crack on the surface of the target prefabricated component, or the size of the target prefabricated component is not normal, and in the subsequent quality inspection stage, it is detected that the surface coating integrity of the target prefabricated component is further reduced, there is still a crack on the surface of the target prefabricated component or the crack is further reduced, and the size of the target prefabricated component is still not normal, the health status of the target prefabricated component is predicted to be poor by the correlation prediction relational expression. It should be noted that the prediction condition of the model prediction relational expression is generally greater than that of the correlation prediction relational expression.

[0061] S30: Obtain the current prediction data of the target prefabricated component based on the model prediction relational expression and the correlation prediction relational expression, respectively.

[0062] In the present embodiment, the current prediction data refers to the prediction results of the model prediction relational expression and the correlation prediction relational expression.

[0063] Specifically, at different stages of the target precast component, the Internet of Things data at the corresponding time point is input into the corresponding model prediction relationship and the associated prediction relationship, and the output result of the model prediction relationship and the associated prediction relationship is the current prediction data of the predicted health state of the target precast component.

[0064] S40: Triggering the health state warning information of the target precast component based on the current prediction data.

[0065] In the embodiment, the component health state warning information refers to information indicating that the health state of the target precast component is poor.

[0066] Specifically, based on the current prediction data, if the current prediction data indicates that the corresponding Internet of Things data meets the condition of the model prediction relationship and / or the associated prediction relationship, at this time, it is predicted that the health state of the target precast component is poor, therefore, the information indicating that the health state of the target precast component is poor is sent to the monitoring management end monitoring the state of the target precast component, that is, the component health state warning information is triggered.

[0067] In an embodiment, as shown in Figure 2 The Internet of Things data classification information is obtained in the following manner:

[0068] S01: Based on the preset component whole process cycle, the Internet of Things data is classified to obtain Internet of Things classification data, and the component whole process cycle refers to the whole process cycle of the precast component from design to installation and use.

[0069] Specifically, the preset component whole process cycle refers to the whole process cycle of the precast component from design to installation and use, in the embodiment, the component whole process cycle includes the conceptual design stage, the structural design stage, the production preparation stage, the production manufacturing stage, the quality inspection stage, the component packaging stage, the component transportation stage, the on-site installation stage, the on-site inspection stage and the operation and maintenance stage, therefore, the Internet of Things classification data obtained by classifying the Internet of Things data includes: conceptual design stage data, structural design stage data, production preparation stage Internet of Things data, production manufacturing stage Internet of Things data, quality inspection stage Internet of Things data, component packaging stage Internet of Things data, component transportation stage Internet of Things data, on-site installation stage Internet of Things data, on-site inspection stage and operation and maintenance stage Internet of Things data.

[0070] It should be noted that the conceptual design stage data and the structural design stage data are data of the prefabricated component in the design stage, and thus are not obtained through the Internet of Things device, but are obtained through the prefabricated component design system associated with the digital management platform. Since the conceptual design stage data and the structural design stage data are not obtained through the Internet of Things device, the conceptual design stage data and the structural design stage data are mainly used to provide the machine learning model for learning, that is, the machine learning model learns the conceptual design stage data and the structural design stage data when the health state of the prefabricated component is in trouble, so that when the digital management platform is used after the prefabricated component production stage starts, the conceptual design stage data and the structural design stage data are input into the digital management platform, so as to predict whether the health state of the current prefabricated component is in trouble according to the conceptual design stage data and the structural design stage data.

[0071] In an embodiment, as shown in FIG. 2, the combined learning model is obtained in the following manner: Figure 3

[0072] S02: Obtain historical Internet of Things data obtained by monitoring the prefabricated component, identify the historical Internet of Things data, and distribute the input to the multiple independent learning models.

[0073] In this embodiment, the historical Internet of Things data refers to the historical Internet of Things data obtained. The multiple independent learning models refer to a plurality of different independent machine learning models.

[0074] ​Specifically, a plurality of different independent machine learning models are built in the digital management platform. Before the production of a new batch of prefabricated components of a new specification starts each time, historical Internet of Things data obtained by the Internet of Things devices monitoring the prefabricated components is acquired, i.e., historical Internet of Things data. The historical Internet of Things data is identified and analyzed to determine the prediction accuracy of different machine learning models when the health status of the actual prefabricated components appears to be problematic, and whether the predicted collected Internet of Things data is associated with the problem of the health status of the prefabricated components, i.e., whether the correct Internet of Things data is input into the machine learning model. In this way, the machine learning models with a prediction accuracy higher than the average in different stages are divided, and among the machine learning models with a prediction accuracy higher than the average in different stages, the machine learning models associated with the problem of the health status of the prefabricated components are further divided, and the divided machine learning models are used as multiple independent learning models. It should be noted that the purpose of further dividing the machine learning models associated with the problem of the health status of the prefabricated components is to improve the accuracy of prediction and reduce the situation that the collected Internet of Things data is not the required data in the actual operation of the digital management platform, because the collected Internet of Things data by the machine learning model and the data corresponding to the problem causing the health status of the prefabricated components are not the same data.

[0075] S03: generating a time-space model based on the historical Internet of Things data.

[0076] In this embodiment, the time-space model refers to a model formed by arranging the multiple independent learning models in time sequence in different stages.

[0077] Specifically, according to the historical Internet of Things data, the route of the prefabricated component is determined, and the route is divided into a plurality of coordinate points (i.e., coordinate points in the same coordinate system as the data of the positioning tag installed on the prefabricated component). Based on the different stages of the whole life cycle of the prefabricated component and the coordinate points, a space model based on time and location is formed, thereby forming a time-space model.

[0078] S04: combining the time-space model and the multiple independent learning models to obtain a combined learning model.

[0079] Specifically, the data input and output interfaces of the multiple independent learning models corresponding to different stages are associated with the positions corresponding to the time-space model to form a combined learning model, for example, taking different stages of the whole life cycle of the prefabricated component as the vertical axis and the coordinate points formed by the route of the prefabricated component movement as the horizontal axis to form a multiple independent learning model output result matrix. When the multiple independent learning model outputs the result, the output result is displayed at the corresponding position.

[0080] In an embodiment, as shown in FIG. 2, in step S20, based on the combined learning model, a corresponding model prediction relationship and an associated prediction relationship are obtained, specifically including: Figure 4

[0081] S21: Based on the multiple independent learning models in the combined learning model, a multiple learning result is obtained.

[0082] In this embodiment, the multiple learning result refers to the learning result of the multiple independent learning models.

[0083] Specifically, when the Internet of Things data is input into the multiple independent learning models, the Internet of Things data is identified, the current time point is determined, the first Internet of Things data factor that will cause the health status of the target prefabricated component to have a problem at the current time point is determined by the learning function of the multiple independent learning models, and the corresponding threshold value for predicting that the health status of the target prefabricated component will have a problem is determined. The content corresponding to the first Internet of Things data factor in the data content of the Internet of Things data is extracted. If the content corresponding to the first Internet of Things data factor is not in the data content of the Internet of Things data, the multiple learning result has no content. In addition, the second Internet of Things data factor that will cause the health status of the target prefabricated component to have a problem in combination with the current time point and subsequent Internet of Things data is determined, and the threshold value for predicting that the health status of the target prefabricated component will have a problem corresponding to the current Internet of Things data and the subsequent Internet of Things data, respectively, is determined. The content corresponding to the second Internet of Things data factor in the data content of the Internet of Things data is extracted. Similarly, if the content corresponding to the second Internet of Things data factor is not in the data content of the Internet of Things data, the multiple learning result has no content. In this way, the first Internet of Things data factor that will cause the health status of the target prefabricated component to have a problem at the current time point, the threshold value for predicting that the health status of the target prefabricated component will have a problem corresponding to the first Internet of Things data factor, the content corresponding to the first Internet of Things data factor in the data content of the Internet of Things data, the second Internet of Things data factor that will cause the health status of the target prefabricated component to have a problem in combination with the current time point and subsequent Internet of Things data, the threshold value for predicting that the health status of the target prefabricated component will have a problem corresponding to the second Internet of Things data factor, and the content corresponding to the second Internet of Things data factor in the data content of the Internet of Things data are associated to obtain the multiple learning result corresponding to the current time point.​

[0084] S22: Obtain a model prediction relationship and a correlation prediction relationship based on the time-domain space model in the combined learning model and the multi-path learning result.

[0085] Specifically, based on the multi-path learning result in the time-domain space model, i.e., the learning result of the multi-path independent learning model, the learning result of each machine learning model is used to form a model prediction relationship. The first Internet of Things data factor in the learning result of each machine learning model that will cause the health state of the target prefabricated component to have a problem at the current time point, the threshold value for predicting that the health state of the target prefabricated component will have a problem corresponding to the first Internet of Things data factor, and the content corresponding to the first Internet of Things data factor in the data content of the Internet of Things data are extracted to form the model prediction relationship. The second Internet of Things data factor in the learning result of each machine learning model that will cause the health state of the target prefabricated component to have a problem in combination with the current time point and subsequent Internet of Things data, the threshold value for predicting that the health state of the target prefabricated component will have a problem corresponding to the second Internet of Things data factor, and the content corresponding to the second Internet of Things data factor in the data content of the Internet of Things data are extracted to form the correlation prediction relationship.

[0086] Further, since the conceptual design stage data and the structural design stage data are not obtained through the Internet of Things device, the conceptual design stage data and the structural design stage data are mainly used to provide the machine learning model for learning, i.e., through the machine learning model to learn the conceptual design stage data and the structural design stage data when predicting that the health state of the target prefabricated component will have a problem, and the subsequent stage Internet of Things data, to generate the correlation prediction relationship between the conceptual design stage data and the structural design stage data and the subsequent stage Internet of Things data.

[0087] In an embodiment, as shown in FIG. 30, in step S30, the model prediction relationship includes a current stage critical threshold value, and the correlation prediction relationship includes a correlation stage critical threshold value. The current stage critical threshold value is greater than the correlation stage critical threshold value. Based on the model prediction relationship and the correlation prediction relationship respectively, the current prediction data of the target prefabricated component is obtained, which specifically includes: Figure 5

[0088] S31: Compare the Internet of Things data with the current stage critical threshold value of the model prediction relationship and the correlation stage critical threshold value of the correlation prediction relationship respectively to obtain the current prediction data of the target prefabricated component.

[0089] ​In the embodiment, the current stage critical threshold refers to a threshold in the model prediction relationship for predicting that the health state of the target prefabricated component is problematic. The association stage critical threshold refers to a threshold in the association prediction relationship for predicting that the health state of the target prefabricated component is problematic. In the model prediction relationship, a threshold for predicting that the health state of the target prefabricated component is problematic, i.e. the current stage critical threshold, is included, and by analogy, the association prediction relationship also includes a threshold for predicting that the health state of the target prefabricated component is problematic, i.e. the association stage critical threshold. Since the current stage critical threshold only relies on the Internet of Things data of the current stage, and the association stage critical threshold needs to combine the Internet of Things data of the current stage and the subsequent stage, the association prediction relationship corresponding to the association stage critical threshold needs to use a larger amount of data, so the prediction result of the association prediction relationship is more accurate, but in order to ensure the advance of the prediction and the prediction accuracy in the current stage, the current stage critical threshold is not less than the association stage critical threshold.

[0090] Specifically, the data corresponding to the current stage critical threshold and the association stage critical threshold in the Internet of Things data are compared with the current stage critical threshold and the association stage critical threshold, and the comparison result is taken as the current prediction data, i.e. when the data corresponding to the current stage critical threshold is greater than the current stage critical threshold, it is predicted that the health state of the target prefabricated component is problematic, and when the data corresponding to the association stage critical threshold is greater than the association stage critical threshold, it is predicted that the health state of the target prefabricated component is problematic.

[0091] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0092] In an embodiment, a prefabricated component early warning system based on the Internet of Things for fabricated buildings is provided, which corresponds one-to-one to the prefabricated component early warning method based on the Internet of Things for fabricated buildings in the above embodiment. As shown in the figure, the prefabricated component early warning system based on the Internet of Things for fabricated buildings includes a combined learning model updating module, a relationship formula acquisition module, a current prediction data acquisition module, and a health state warning instruction module. The functions of each functional module are described in detail as follows: Figure 6

[0093] The combined learning model updating module is configured to classify the Internet of Things data obtained by monitoring the target prefabricated component based on the preset Internet of Things data classification information, and input the corresponding combined learning model;

[0094] The relationship formula acquisition module is configured to obtain the corresponding model prediction relationship and association prediction relationship based on the combined learning model.​

[0095] The current prediction data acquisition module is configured to acquire current prediction data of the target prefabricated component based on the model prediction relationship and the correlation prediction relationship respectively.

[0096] The health state warning instruction module is configured to trigger health state warning information of the target prefabricated component based on the current prediction data.

[0097] Optionally, the prefabricated component early warning system based on the Internet of Things further comprises:

[0098] The Internet of Things classification data acquisition module is configured to classify the Internet of Things data based on a preset component whole-process cycle to obtain Internet of Things classification data, wherein the component whole-process cycle refers to a whole-process cycle of the prefabricated component from design to installation and use.

[0099] Optionally, the prefabricated component early warning system based on the Internet of Things further comprises:

[0100] The multi-path learning model allocation module is configured to acquire historical Internet of Things data obtained by monitoring the prefabricated component, identify the historical Internet of Things data, and allocate the historical Internet of Things data to the multi-path independent learning models.

[0101] The time-space model generation module is configured to generate a time-space model based on the historical Internet of Things data.

[0102] The combined learning model combination module is configured to combine the time-space model and the multi-path independent learning models to obtain a combined learning model.

[0103] Optionally, the relationship acquisition module comprises:

[0104] The multi-path learning result acquisition submodule is configured to acquire multi-path learning results based on the multi-path learning models in the combined learning model.

[0105] The relationship acquisition submodule is configured to acquire the model prediction relationship and the correlation prediction relationship based on the time-space model and the multi-path learning results in the combined learning model.

[0106] Optionally, the model prediction relationship comprises a current stage critical threshold, and the correlation prediction relationship comprises a correlation stage critical threshold, wherein the current stage critical threshold is greater than the correlation stage critical threshold, and the current prediction data acquisition module comprises:

[0107] The current prediction data acquisition submodule is configured to compare the Internet of Things data with the current stage critical threshold of the model prediction relationship and the correlation stage critical threshold of the correlation prediction relationship respectively to obtain the current prediction data of the target prefabricated component.

[0108] The specific limitations of the prefabricated component early warning system based on the Internet of Things can refer to the limitations of the prefabricated component early warning method based on the Internet of Things described above, and will not be repeated here. Each module in the prefabricated component early warning system based on the Internet of Things can be realized by software, hardware, and combinations thereof, in whole or in part. The above-mentioned each module can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so that the processor calls and executes the operations corresponding to each of the above modules.

[0109] In one embodiment, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in Figure 7 The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store Internet of Things data classification information, a combined learning model, a model prediction relationship, a correlation prediction relationship, current prediction data, and health state warning instructions. The network interface of the computer device is used to communicate with external terminals through network connections. The computer program is executed by the processor to implement a prefabricated component early warning method based on the Internet of Things.

[0110] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the following steps when executing the computer program:

[0111] Based on the preset Internet of Things data classification information, the Internet of Things data obtained by monitoring the target prefabricated component is classified and input into the corresponding combined learning model;

[0112] Based on the combined learning model, the corresponding model prediction relationship and the correlation prediction relationship are obtained;

[0113] Based on the model prediction relationship and the correlation prediction relationship respectively, the current prediction data of the target prefabricated component is obtained;

[0114] Based on the current prediction data, the health state warning information of the target prefabricated component is triggered.

[0115] In one embodiment, a computer readable storage medium is provided, and a computer program is stored on the computer readable storage medium, and the computer program is executed by the processor to implement the following steps:

[0116] Based on the preset Internet of Things data classification information, the Internet of Things data obtained by monitoring the target prefabricated component is classified, and the corresponding combined learning model is input;

[0117] Based on the combined learning model, the corresponding model prediction relationship and the associated prediction relationship are obtained.

[0118] Based on the model prediction relationship and the associated prediction relationship respectively, the current prediction data of the target prefabricated component is obtained.

[0119] Based on the current prediction data, the health state warning information of the target prefabricated component is triggered.

[0120] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0121] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is exemplified. In actual application, the above-mentioned functions can be completed by different functional units and modules according to needs, that is, the internal structure of the system is divided into different functional units or modules to complete all or part of the above-mentioned functions.

[0122] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for early warning of prefabricated components for prefabricated buildings based on the Internet of Things, characterized in that, The IoT-based early warning method for prefabricated building components includes: Based on the preset IoT data classification information, the IoT data obtained by monitoring the target prefabricated components will be classified and input into the corresponding combined learning model. The combined learning model is obtained through the following method: Historical IoT data obtained from monitoring prefabricated components is acquired, the historical IoT data is identified, and the data is assigned to multiple independent learning models. Based on the historical IoT data, a temporal-space model is generated; By combining the temporal-space model and the multi-path independent learning model, a combined learning model is obtained; Based on the combined learning model, the corresponding model prediction relation and association prediction relation are obtained; The process of obtaining the corresponding model prediction relation and association prediction relation based on the combined learning model specifically includes: Based on the multi-path independent learning model in the combined learning model, multi-path learning results are obtained; Based on the temporal-spatial model and the multi-path learning results in the combined learning model, model prediction relation and correlation prediction relation are obtained. The model prediction relation is a relation used to predict the health status of the target prefabricated component based on the IoT data of the current stage. The correlation prediction relation is a relation used to predict the health status of the target prefabricated component by combining the IoT data of the current stage and the subsequent stages. Based on the model prediction relationship and the correlation prediction relationship, the current prediction data of the target prefabricated component is obtained respectively; Based on the current prediction data, a health status warning message for the target precast component is triggered. The model prediction relation includes a current stage critical threshold, and the correlation prediction relation includes a correlation stage critical threshold. The current stage critical threshold is greater than the correlation stage critical threshold. The step of obtaining the current prediction data of the target precast component based on the model prediction relation and the correlation prediction relation specifically includes: The IoT data is compared with the current stage critical threshold of the model prediction relation and the association stage critical threshold of the association prediction relation to obtain the current prediction data of the target prefabricated component.

2. The early warning method for prefabricated components of assembled buildings based on the Internet of Things according to claim 1, characterized in that, The IoT data classification information is obtained through the following methods: Based on a preset component lifecycle, IoT data is classified to obtain IoT classification data. The component lifecycle refers to the entire lifecycle of prefabricated components from design to installation and use.

3. An early warning system for prefabricated building components based on the Internet of Things, characterized in that, The IoT-based early warning system for prefabricated building components includes: The combined learning model update module is used to classify the IoT data obtained from monitoring the target prefabricated components based on the preset IoT data classification information, and input the classification data into the corresponding combined learning model. The IoT-based early warning system for prefabricated building components also includes: The multi-path learning model allocation module is used to acquire historical IoT data obtained from monitoring prefabricated components, identify the historical IoT data, and allocate the input to multiple independent learning models. The temporal-spatial model generation module is used to generate a temporal-spatial model based on the historical IoT data. A combined learning model combination module is used to combine the temporal-space model and the multiple independent learning models to obtain a combined learning model. The relation acquisition module is used to obtain the corresponding model prediction relation and association prediction relation based on the combined learning model; The relational expression acquisition module includes: The multi-path learning result acquisition submodule is used to obtain multi-path learning results based on the multiple independent learning models in the combined learning model; The relation acquisition submodule is used to obtain the model prediction relation and the correlation prediction relation based on the temporal-spatial model and the multi-path learning results in the combined learning model. The model prediction relation is a relation used to predict the health status of the target prefabricated component based on the IoT data of the current stage, and the correlation prediction relation is a relation used to predict the health status of the target prefabricated component by combining the IoT data of the current stage and the subsequent stages. The current prediction data acquisition module is used to acquire the current prediction data of the target precast component based on the model prediction relation and the correlation prediction relation, respectively. The health status warning instruction module is used to trigger a health status warning message for the target precast component based on the current prediction data.

4. The prefabricated building component early warning system based on the Internet of Things according to claim 3, characterized in that, The IoT-based early warning system for prefabricated building components also includes: The IoT classification data acquisition module is used to classify IoT data based on a preset component lifecycle to obtain IoT classification data. The component lifecycle refers to the entire lifecycle of prefabricated components from design to installation and use.

5. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the Internet of Things-based early warning method for prefabricated building components as described in any one of claims 1 to 2.

6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the Internet of Things-based early warning method for prefabricated components of prefabricated buildings as described in any one of claims 1 to 2.

Images