Intelligent monitoring and early warning method for prefabricated data center

Abstract

The invention provides an intelligent monitoring and early warning method for a prefabricated data center, and relates to the technical field of intelligent monitoring and early warning, and the method comprises the steps: carrying out the modular design and manufacturing of a prefabricated center, and building a twin model which is mapped with a module; performing assembly fitting optimization according to the field scene data and the twin model, establishing an optimization assembly path, and establishing an assembly constraint mapped with the assembly path; controlling an assembly site to perform prefabricated central modular assembly based on the optimization assembly path, and performing point cloud data acquisition on the assembly site; carrying out twin model comparison by using the real-time updated point cloud data set, and carrying out deviation identification through assembly constraint; and performing joint early warning trigger analysis on the deviation identification result, and establishing a monitoring early warning result. By means of the method and device, the technical problem that in the prior art, adaptability and precision are low in the assembling process due to presetting of the assembling path and scheme can be solved, and the technical effects of improving the assembling precision and improving the production efficiency are achieved.

Description

Technical Field

[0001] The present application relates to the field of intelligent monitoring and early warning technology, and in particular to an intelligent monitoring and early warning method for a prefabricated data center. Background Art

[0002] With the continuous development of industrial automation and intelligent technology, intelligent manufacturing and modular production have become important development trends in modern manufacturing. Although existing technologies have made some progress in this regard, there are still some obvious defects.

[0003] At present, the assembly paths and schemes in the existing technology are usually preset and lack flexibility. Although there are certain automated equipment and sensors to assist, their functions are often limited to specific scenarios and cannot be dynamically adjusted according to actual conditions. Therefore, when the assembly environment or conditions change, the system cannot automatically make adaptive adjustments, which can easily lead to assembly deviations or substandard quality.

[0004] In summary, the prior art has technical problems such as lack of flexibility and dynamic adjustment capability due to the preset nature of assembly paths and schemes, which further affects the adaptability and precision of the assembly process, leading to assembly deviations and substandard quality. Summary of the invention

[0005] The purpose of this application is to provide an intelligent monitoring and early warning method for prefabricated data centers, in order to solve the technical problems in the prior art that the preset assembly paths and plans lead to a lack of flexibility and dynamic adjustment capabilities, further affecting the adaptability and accuracy of the assembly process, resulting in assembly deviations and substandard quality.

[0006] In view of the above problems, the present application provides an intelligent monitoring and early warning method for a prefabricated data center, including: modular design and manufacturing of the prefabricated center, and establishing a twin model mapped to the module; obtaining on-site scene data of the assembly site, performing assembly fitting optimization based on the on-site scene data and the twin model, establishing an optimal assembly path, and establishing assembly constraints mapped to the assembly path, wherein the assembly constraints include position constraints, assembly interface constraints, mechanical constraints, and motion constraints; based on the optimal assembly path, controlling the assembly site to perform modular assembly of the prefabricated center, and using a three-dimensional laser scanner to collect point cloud data at the assembly site to establish a real-time updated point cloud data set; using the real-time updated point cloud data set to compare the twin model, and identifying deviations through the assembly constraints to establish deviation identification results; performing a joint early warning trigger analysis on the deviation identification results to establish a monitoring and early warning result.

[0007] The technical solution provided in the present application has at least the following technical effects or advantages: by modularizing the design and manufacturing of the prefabrication center, and establishing a twin model mapped with the module; obtaining the on-site scene data of the assembly site, performing assembly fitting optimization according to the on-site scene data and the twin model, establishing an optimal assembly path, and establishing assembly constraints mapped with the assembly path, wherein the assembly constraints include position constraints, assembly interface constraints, mechanical constraints, and motion constraints; based on the optimal assembly path, controlling the assembly site to perform modular assembly of the prefabrication center, and using a three-dimensional laser scanner to collect point cloud data of the assembly site, and establishing a real-time updated point cloud data set; using the real-time updated point cloud data set to compare the twin model, and identifying deviations through the assembly constraints, and establishing deviation identification results; performing a joint early warning trigger analysis on the deviation identification results, and establishing a monitoring early warning result. That is to say, by achieving the technical goal of real-time optimization and dynamic adjustment of the assembly path, the technical effect of improving assembly accuracy, improving production efficiency, reducing the occurrence rate of deviations, and ensuring stable product quality is achieved.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 A flowchart of the intelligent monitoring and early warning method for the prefabricated data center in this application; Figure 2 A schematic diagram of a process for establishing an optimal assembly path in the intelligent monitoring and early warning method for a prefabricated data center in this application. DETAILED DESCRIPTION

[0011] This application solves the technical problem in the prior art that the pre-set nature of assembly paths and solutions leads to a lack of flexibility and dynamic adjustment capabilities, further affecting the adaptability and accuracy of the assembly process, resulting in assembly deviations and substandard quality by providing an intelligent monitoring and early warning method for prefabricated data centers. The technical goal of achieving real-time optimization and dynamic adjustment of assembly paths is achieved, achieving the technical effects of improving assembly accuracy, increasing production efficiency, reducing the occurrence of deviations, and ensuring stable product quality.

[0012] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0013] For example, please refer to the attached Figure 1 , this application provides an intelligent monitoring and early warning method for a prefabricated data center, which specifically includes the following steps: Step 1: Carry out modular design and manufacturing of the prefabrication center, and establish a twin model with module mapping.

[0014] Specifically, the modular design and manufacturing of the prefabrication center is a design method that splits the overall system into several independent modules. Each module is designed and produced separately according to specific functions or requirements, and can be easily combined together during later assembly. Through modular design, production efficiency can be improved, overall construction time can be reduced, and potential manufacturing complexity can be reduced.

[0015] Twin model refers to the real-time monitoring and optimization of the entire system by establishing a mapping relationship between the physical system and its digital twin model. The digital model can not only reflect the structure and performance of the physical module, but also simulate the impact of its working state and environmental changes on the operation of the system. For example, the twin model of the cooling system module can monitor the module's temperature, flow rate and other data in real time, predict possible failures or performance degradation of the system, and make adjustments in advance.

[0016] Modular design and twin models can accurately simulate and predict the operating status of modules through digital mapping, discover potential problems in advance, optimize the coordination between modules, and reduce the complexity and uncertainty of on-site assembly.

[0017] Step 2: Obtain on-site scene data of the assembly site, perform assembly fitting optimization based on the on-site scene data and the twin model, establish an optimal assembly path, and establish assembly constraints mapped to the assembly path, wherein the assembly constraints include position constraints, assembly interface constraints, mechanical constraints, and motion constraints.

[0018] Specifically, sensors or other devices are used to collect various types of information in the on-site environment, including temperature, humidity, spatial layout, component location, etc., to reflect the real-time conditions on site, thus providing a basis for the subsequent assembly process. On-site scene data helps designers understand the gap between the actual on-site situation and the expected design, thus providing support for optimizing the assembly process.

[0019] The assembly fitting is optimized based on the on-site scene data and the twin model. The twin model maps all the parameters and characteristics of the actual system. The on-site data is combined with the twin model through the optimization algorithm to automatically adjust and optimize the assembly plan, so that the actual assembly process can achieve the best effect in the shortest time. For example, module assembly requires a docking interface. The twin model can simulate and optimize the docking process of the interface to ensure the most accurate docking.

[0020] Establish an optimal assembly path. The optimal path refers to each step from the start of assembly to completion. By comparing the efficiency and risk of different paths, the most suitable path can be found to complete the assembly. For example, choose to assemble the bottom module first, then the upper module, or perform the assembly steps in the opposite process, which will affect the assembly efficiency and stability.

[0021] In addition, the assembly constraints mapped to the assembly path include position constraints, assembly interface constraints, mechanical constraints, and motion constraints. Position constraints mean that each component must be placed in a specific position during assembly to avoid errors or misalignment; assembly interface constraints involve interface matching between different modules, requiring interface position, shape, and size to meet standards to ensure seamless docking; mechanical constraints ensure that the module will not be deformed or damaged by external forces during the assembly process; motion constraints ensure that the relative movement between components during the assembly process is as expected to avoid interference or jamming. Constraints are used to ensure that the assembly process goes smoothly and improve overall efficiency and quality.

[0022] Step three: Based on the optimal assembly path, the assembly site is controlled to perform prefabricated center modular assembly, and point cloud data of the assembly site is collected using a three-dimensional laser scanner to establish a real-time updated point cloud data set.

[0023] Specifically, the assembly site is controlled based on the optimal assembly path to carry out prefabricated central modular assembly, so that each module is assembled in the best order and method. For example, the optimized path guides the installation of the power module first, and then the cooling module, thereby improving the overall installation efficiency and reducing interference and conflicts during the installation process. The optimized path can ensure that each module is installed in the best position and operation order on site, improving assembly quality and efficiency.

[0024] Use 3D laser scanning technology to scan the site with a laser beam and collect the reflected laser signals to obtain high-precision spatial data. Point cloud data refers to a set of 3D coordinate points composed of laser signals, which represents the geometric shape and spatial structure of the actual site. For example, a laser scanner can scan the precise position, size and relative position of each module on site with other modules. The geometric information of the site is obtained in real time through point cloud data to ensure consistency with the virtual model. The collected point cloud data is continuously updated and formed into a continuously updated data set to reflect the latest status of the assembly site. For example, as the modules are installed and assembled, the point cloud data will continue to change, reflecting the spatial position and relative relationship of the newly installed parts. The assembly path can be adjusted at any time through the real-time updated point cloud data set, and the assembly process can be optimized to ensure the smooth progress of the entire modular assembly process and avoid problems caused by the inconsistency between the actual situation and the virtual model.

[0025] Step 4: Use the real-time updated point cloud data set to perform the twin model comparison, and use the assembly constraints to identify deviations and establish deviation identification results.

[0026] Specifically, the point cloud data obtained by the 3D laser scanner is compared with the established virtual twin model, which can identify the differences between the site and the design, thereby ensuring that the on-site assembly process is consistent with the expected design. During the comparison process, the allowable and out-of-range deviations are judged according to the preset assembly constraints. Assembly constraints include parameter requirements such as position, size, and angle. For example, the angular error of the module is allowed to be two degrees in the design, but if the deviation exceeds two degrees during actual assembly, it is considered unqualified. Assembly constraints help to accurately screen out deviations that exceed the allowable error during the comparison process, thereby more accurately identifying problems in the assembly, thereby establishing deviation identification results, and generating reports or results on on-site assembly deviations. By establishing deviation identification results, all detected deviations and their properties, such as the type, size, location, and other information of the deviation, are obtained, and then problems in the assembly are obtained and adjusted in time.

[0027] Step 5: Conduct joint early warning trigger analysis on the deviation identification results to establish monitoring early warning results.

[0028] Specifically, the detected deviation results are combined with preset thresholds and rules to determine whether an early warning needs to be triggered. For example, if the deviation identification result shows that the module position deviation is seriously beyond the allowed range, an early warning is triggered, thereby reducing false alarms and ensuring that early warnings are only issued when they are really needed.

[0029] Based on the results of the early warning trigger analysis, monitoring early warning results are established, and clear early warning notifications or reports are generated to inform relevant personnel of problems in the module assembly process. Early warning results usually include the specific location, type, severity and recommended treatment measures of the deviation.

[0030] The intelligent monitoring and early warning method of the prefabricated data center can achieve the technical goal of real-time optimization and dynamic adjustment of the assembly path, and achieve the technical effects of improving assembly accuracy, increasing production efficiency, reducing the occurrence rate of deviations, and ensuring stable product quality.

[0031] Further, if Figure 2 As shown, the present application also includes: establishing a balance evaluation network, the balance evaluation network is a network for performing adaptation evaluation of the assembly path, the balance evaluation network includes a scene adaptation subnetwork, an assembly stability subnetwork, an assembly complexity subnetwork and a balance analysis subnetwork; when performing assembly path fitting optimization, the balance evaluation network is activated, and the path evaluation is performed through the scene adaptation subnetwork, the assembly stability subnetwork, and the assembly complexity subnetwork respectively, and the evaluation results are synchronized to the balance analysis subnetwork, and the balance fitness is output; optimization iteration is performed according to the balance fitness to establish an optimal assembly path.

[0032] Specifically, a balance evaluation network is established. The balance evaluation network is a framework for evaluating and optimizing the adaptability of assembly paths. The balance evaluation network includes a scene adaptation subnetwork, an assembly stability subnetwork, an assembly complexity subnetwork, and a balance analysis subnetwork, and each subnetwork has its specific function. The scene adaptation subnetwork is responsible for evaluating whether the assembly path is adapted to the actual environmental data of the current site, such as site size, equipment layout, and site conditions. The assembly stability subnetwork evaluates the stability of the assembly process to ensure that the components will not be offset or unstable during the assembly process. The assembly complexity subnetwork focuses on the complexity of the path and evaluates whether there are too many steps or difficult operations, which affects efficiency. The balance analysis subnetwork summarizes the evaluation results of each subnetwork and comprehensively analyzes the pros and cons of the path.

[0033] During the assembly path fitting and optimization process, the balance evaluation network is activated to evaluate different assembly paths. The scene adaptation subnetwork evaluates whether the path is adapted to the conditions on site, such as whether the site is sufficient to support the operation of the path and whether the space is appropriate. The assembly stability subnetwork analyzes the stability of the path to check whether the steps in the path are safe to prevent positional offset or other unstable factors during execution. The assembly complexity subnetwork checks whether the path is simple and effective to avoid wasting time and personnel burden due to too many operation steps. All evaluation results will be synchronously transmitted to the balance analysis subnetwork to summarize all information and output the balance fitness, which is used to represent the comprehensive performance of the path under all evaluation criteria.

[0034] The optimization iteration is performed based on the results of the equilibrium fitness. By continuously adjusting the assembly path, each step is optimized to achieve the best assembly efficiency and stability. Each iteration is based on the equilibrium fitness to ensure that the path is gradually improved and the most suitable assembly path is finally found.

[0035] The assembly path is optimized through four sub-networks: scenario adaptation, assembly stability, assembly complexity, and comprehensive balance analysis. Each sub-network is responsible for evaluating different path characteristics to help select the most suitable path in actual assembly. Through the iterative optimization process, an optimal assembly path can eventually be found to improve assembly efficiency, reduce risks, and reduce costs.

[0036] Furthermore, the present application also includes: executing individual coding, encoding the module's assembly path, assembly angle, and contact surface position into a vector; randomly generating N individuals according to the individual coding, and using the N individuals as the initialization population, where N is an integer greater than 1; configuring a mutation factor, performing individual mutation of the initialization population through the mutation factor, and performing cross-discrimination of the mutated individuals based on a preset random probability to establish updated individuals; and using balanced fitness to perform retention screening of updated individuals to complete optimization iterations.

[0037] Specifically, individual encoding is performed, that is, the key information in the assembly process is converted into digital form, and the module assembly path, assembly angle, contact surface position and other information are encoded into vectors. The assembly path refers to each step of the module assembly from the beginning to the completion, the assembly angle refers to the angular position of each component of the module during the assembly process, and the contact surface position refers to the contact point between different modules. By encoding, different information is integrated into a vector, which is convenient for subsequent optimization and processing.

[0038] After executing the individual encoding, multiple individuals are randomly generated. Each individual represents a possible assembly scheme. N represents the number of individuals generated. N is greater than 1, which means that more than one assembly scheme is considered. For example, if N is 5, 5 different assembly paths are generated, each of which may contain different assembly orders, assembly angles, etc. By generating multiple individuals, the chance of finding the optimal solution is increased.

[0039] The mutation factor refers to the parameter that controls the degree of individual variation. Through the mutation factor, individuals in the initial population can be mutated, some parameters of the individuals can be changed, and new individuals can be generated. For example, the assembly angle or contact surface position of a module can be randomly changed to create different assembly paths. Then, based on the preset random probability, the mutated individuals will undergo crossover discrimination, that is, through the crossover operation, some features of different individuals are combined into new individuals, which helps to discover more possible solutions and gradually approach the optimal solution.

[0040] After mutation and crossover operations, the quality of each individual is evaluated. Balanced fitness is an indicator used to evaluate the quality of each assembly path. The higher the fitness, the better the path, and vice versa. Individuals with higher fitness are screened out through balanced fitness, and the better individuals are retained for the next round of iteration. Iteration is carried out until the optimal assembly path is found.

[0041] Through individual encoding, population initialization, mutation operation, crossover discrimination and fitness screening, optimization iteration is performed. Each step is constantly adjusting and optimizing the assembly path, gradually improving the efficiency and quality of the path. Through multiple iterations and optimizations, the optimal assembly path can be found and the overall effect of the assembly process can be improved.

[0042] Furthermore, the present application also includes: establishing a basic variation factor; obtaining the number of iterations of the current iteration, and generating a first configuration constraint according to the number of iterations; establishing an iterative backtracking evaluation window, performing iterative backtracking analysis using the iterative backtracking evaluation window, and establishing a second configuration constraint according to the iterative backtracking analysis result; and compensating the basic variation factor using the first configuration constraint and the second configuration constraint to complete the variation factor configuration.

[0043] Specifically, the basic mutation factor is the initial parameter used to control the amplitude of the mutation operation. The basic mutation factor determines the range and degree of each individual mutation during the optimization process. The basic mutation factor affects the speed and quality of generating new individuals during the optimization process. For example, the basic mutation factor is set to a value that represents the proportion of change at each mutation. The larger the basic mutation factor, the wider the range of mutation; conversely, the smaller the mutation. By setting the basic mutation factor, it is ensured that there will be no excessive or insufficient mutation during the optimization process.

[0044] Each optimization iteration generates a new candidate solution, and the number of iterations represents the degree of progress of the optimization process. According to the current number of iterations, the first configuration constraint can be formulated. As the number of iterations increases, the conditions of the first configuration constraint will gradually tighten to promote the optimization process to converge to the optimal solution faster, and vice versa. For example, when the number of iterations reaches the preset value, higher requirements are required for the accuracy and stability of the assembly path to ensure that the final result meets the standards of actual operation.

[0045] The Iteration Backtracking Evaluation Window is a mechanism for reviewing and analyzing iteration results to ensure that the optimization process does not deviate from the optimal solution. Backtracking analysis helps analyze situations where the optimization effect is good and where excessive variation or instability may occur. Based on the results of the backtracking analysis, a second configuration constraint can be generated to guide the subsequent optimization process. For example, if the backtracking analysis finds that the path has a good optimization effect in the early stage, but fails in the later stage, the optimization strategy can be adjusted through the second configuration constraint to avoid the same situation from recurring in subsequent iterations.

[0046] The first configuration constraint and the second configuration constraint are obtained to compensate and adjust the basic mutation factor. The basic mutation factor will be adjusted accordingly according to the number of iterations and the results of the backtracking analysis to ensure that the optimization process is more accurate and effective. For example, if the first configuration constraint indicates that the mutation amplitude needs to be increased, and the second configuration constraint indicates that the mutation frequency needs to be reduced, the value of the mutation factor can be adjusted accordingly to make the mutation operation more in line with actual needs. After the configuration of the mutation factor is completed, the optimization process can approach the optimal solution more stably and efficiently.

[0047] Dynamic adjustment and compensation of mutation factors are achieved through steps such as establishing basic mutation factors, setting configuration constraints and backtracking analysis. The basic mutation factors determine the scope of mutation in the initial stage, and as the number of iterations increases and the backtracking analysis deepens, the configuration constraints gradually fine-tune the mutation factors to improve the optimization efficiency and ensure the stability and reliability of the final results, so that the optimization process gradually tends to the optimal solution, ensuring that the final result is the best.

[0048] Furthermore, the present application also includes: establishing error classification levels, the error classification levels including a first error classification level, a second error classification level, and a third error classification level; performing a joint early warning trigger analysis based on the error classification levels to obtain trigger results; and establishing a monitoring early warning result according to an early warning response plan corresponding to the trigger result.

[0049] Specifically, according to the size of the deviation and the impact on the assembly quality, the errors are divided into different levels, and an error grading level is established. The error grading level includes the first error grading level, the second error grading level, and the third error grading level. The error grading level helps to clarify the severity of each deviation and take different treatment measures according to different levels. For example, the first error grading level refers to a small deviation, which has little impact on the assembly quality and can be tolerated and does not need to be corrected immediately; the second error grading level represents a moderate deviation, which requires further observation or adjustment; the third error grading level represents a serious deviation, which may affect the overall assembly quality and requires immediate measures to repair it.

[0050] By evaluating the errors of different levels and combining other influencing factors, it is analyzed whether it is necessary to trigger an early warning. If the deviation exceeds the allowable range, an early warning will be triggered. Through joint analysis, the influence of multiple dimensions is comprehensively considered to determine whether emergency measures should be taken. For example, if the angle error and position error of the module are both at the third error classification level, and the error may affect the accuracy of the entire assembly process, an early warning needs to be triggered. According to the early warning response plan corresponding to the trigger result, a monitoring early warning result is established. When the early warning is triggered, a specific early warning report is generated according to the preset response plan. The early warning response plan is a specific measure taken for different trigger results, including adjusting the installation angle, recalibrating components, or pausing the assembly process for inspection, etc., to obtain the monitoring early warning results, including detailed information on the early warning trigger, error classification, response measures, and operations that need to be further performed, which are used to send them to relevant personnel in a timely manner to ensure that repair measures are taken in time.

[0051] By establishing an error classification level, the deviation is divided into different levels, and the warning trigger analysis is performed according to the severity of the error. Combined with the trigger results, the monitoring warning results are generated to guide the operator to take corresponding repair measures according to the warning response plan, thereby identifying and responding to the deviation, ensuring that the assembly quality is not affected by unqualified deviations, and improving the stability and accuracy of the assembly process.

[0052] Furthermore, the present application also includes: obtaining the accuracy requirements of the location point cloud; performing an acquisition effect evaluation when acquiring point cloud data at the assembly site through the three-dimensional laser scanner; performing adaptation analysis based on the acquisition effect evaluation results and the accuracy requirements, and establishing adaptation anomalies; calling the linked acquisition device for additional acquisition based on the adaptation anomaly, and establishing a real-time updated point cloud data set based on the additional acquisition.

[0053] Specifically, before point cloud data collection, the accuracy requirements of the point cloud data obtained, including the spatial resolution of the point cloud data, measurement error, point cloud density, etc., set clear standards for subsequent point cloud data collection to ensure that the data quality meets the requirements of actual use.

[0054] After using a laser scanner to collect field data, the collection effect needs to be evaluated. Collection effect evaluation is to check the quality of the collected point cloud data to determine whether it meets the preset accuracy requirements. For example, check whether the density of the point cloud meets the requirements, whether there are problems such as missing data or collection errors. If the collection effect evaluation shows that the data quality does not meet the requirements, adjustments and supplements are required.

[0055] According to the collection effect evaluation results and the accuracy requirements, the adaptation analysis is performed. Establishing the adaptation anomaly means comparing the collection effect evaluation results with the accuracy requirements to analyze whether there is a mismatch between the two. If the actual collected data does not meet the accuracy requirements, or there are problems in the collection process, it will be identified as an adaptation anomaly and further measures will be taken to solve it.

[0056] When an adaptation anomaly is detected, additional data collection is performed by starting other linkage devices to supplement or correct the existing point cloud data. Linkage collection devices include laser scanners, sensors or cameras, etc., to improve the accuracy and integrity of the data. For example, if there is a large deviation in the data collected initially, a second laser scanner is called to supplement it to ensure more accurate point cloud data. Through additional collection, the point cloud data set is updated in real time to ensure continuous optimization and adjustment of the data.

[0057] Start by obtaining the accuracy requirements of the location point cloud and clarify the quality standards for data collection. Identify the differences between the collected data and the expected target through evaluation of the collection effect and adaptation analysis of the accuracy requirements. When adaptation anomalies occur, call other devices for additional collection to ensure that the accuracy and integrity of the data are continuously updated, thereby ensuring that the point cloud data in the assembly process is accurate and supports precise assembly and monitoring.

[0058] Furthermore, the present application also includes: performing real-time video synchronization between the real-time updated point cloud dataset and the twin model to establish a real-time synchronization result; establishing a trigger color feature through the monitoring and early warning result, marking the real-time synchronization result based on the trigger color feature, and performing a visual display.

[0059] Specifically, the point cloud data collected by the laser scanner on site and the virtual twin model are displayed synchronously through video, so that the actual data on site and the ideal model can be displayed synchronously, helping operators to more intuitively see the gap between the site and the design. For example, if the position of a module on site deviates, the operator can see the comparison between the ideal position and the actual position of the module in the twin model synchronously through the video, so as to make adjustments more timely.

[0060] Color features are set for warning information based on monitoring and warning results to highlight potential problems. Triggering color features usually indicates the severity of deviations through color changes. For example, if the deviation of a module exceeds the allowable range, the display color of the module is changed to alert the operator. For example, normal green indicates no deviation, yellow indicates slight deviation, and red indicates severe deviation. Color changes can help personnel quickly identify and respond.

[0061] By triggering color features to mark the real-time synchronization results, deviations and anomalies on site are more visualized. For example, if an angle error occurs in a module on site and the error reaches the preset warning value, the module will be displayed in red in the real-time synchronization video screen, marking the anomaly at that location, ensuring that operators can quickly locate the problem area and take repair measures. Visual display can improve the speed and accuracy of personnel's perception of on-site problems and reduce human errors and response time.

[0062] Through real-time video synchronization, the on-site point cloud data and the virtual twin model are compared to help operators understand the on-site conditions in real time. Combined with the monitoring and early warning results, the color feature identification system allows anomalies and deviations to be intuitively displayed on the video screen through color changes, thereby improving the response speed and accuracy of on-site problems, providing a more intelligent and efficient tool for on-site assembly and monitoring, helping personnel to promptly discover and solve potential problems.

[0063] Furthermore, the present application also includes: recording warnings for all monitoring warning results and generating warning reports; using the warning reports to conduct abnormality focus analysis and establish adaptive abnormality feedback; and performing subsequent optimal assembly path construction optimization through the adaptive abnormality feedback.

[0064] Specifically, all warning information detected by the monitoring system is recorded in detail and an early warning report is generated. The monitoring early warning results are warnings of deviations or abnormal conditions obtained through real-time data analysis on site, while the early warning records are to organize these warnings into documents to ensure that all relevant information is retained for subsequent review and analysis. For example, if the module position deviation is detected to exceed the allowable range during the assembly process, this information will be recorded and an early warning report will be generated. The report will list the specific details of all warning events, including time, type, degree of deviation and other information.

[0065] After receiving the early warning report, we conduct an in-depth analysis of the abnormal events in the report, identify the abnormalities that need special attention, and establish an adaptive abnormal feedback mechanism based on the analysis results. Through abnormal attention analysis, we obtain deviations that have a significant impact on assembly quality, and classify and process them according to their characteristics. For example, when the installation angle of the module has a serious deviation, a feedback strategy is formulated based on its impact scope and severity. Adaptive abnormal feedback refers to automatically adjusting the feedback mechanism based on historical data and actual conditions to ensure timely response and appropriate measures when similar abnormalities occur.

[0066] According to the results of adaptive abnormal feedback, the assembly path is optimized and adjusted through the feedback mechanism to improve the accuracy and efficiency of the assembly process, thereby reducing or avoiding the occurrence of abnormalities. For example, if repeated adjustments and repairs are required due to excessive deviations, the path is automatically adjusted based on the feedback to select a more appropriate installation sequence or operation method to reduce the possibility of deviations and improve the stability of the entire assembly process.

[0067] The abnormal information monitored on site is collated through early warning records and report generation to provide a basis for subsequent analysis. Then, through abnormal attention analysis and adaptive abnormal feedback mechanism, key problems are identified and strategies are adjusted based on feedback. Finally, based on these feedback results, the assembly path is optimized to improve the accuracy and efficiency of the assembly process, thereby timely discovering and solving abnormal problems, and continuously optimizing and improving the assembly process to make it more efficient and stable.

[0068] In summary, the intelligent monitoring and early warning method for prefabricated data centers provided in the present application has the following technical effects: by modularly designing and manufacturing the prefabricated center, and establishing a twin model mapped with the module; obtaining the on-site scene data of the assembly site, performing assembly fitting optimization according to the on-site scene data and the twin model, establishing an optimal assembly path, and establishing assembly constraints mapped with the assembly path, wherein the assembly constraints include position constraints, assembly interface constraints, mechanical constraints, and motion constraints; based on the optimal assembly path, controlling the assembly site to perform modular assembly of the prefabricated center, and using a three-dimensional laser scanner to collect point cloud data at the assembly site, and establishing a real-time updated point cloud data set; using the real-time updated point cloud data set to compare the twin model, and identifying deviations through the assembly constraints, and establishing deviation identification results; performing a joint early warning trigger analysis on the deviation identification results, and establishing a monitoring and early warning result. That is to say, by achieving the technical goal of real-time optimization and dynamic adjustment of the assembly path, the technical effect of improving assembly accuracy, improving production efficiency, reducing the occurrence rate of deviations, and ensuring stable product quality is achieved.

[0069] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0070] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. An intelligent monitoring and early warning method for a prefabricated data center, characterized in that: Methods include: Carry out modular design and manufacturing of prefabrication centers, and establish twin models mapped with modules; Acquire on-site scene data of the assembly site, perform assembly fitting optimization according to the on-site scene data and the twin model, establish an optimal assembly path, and establish assembly constraints mapped to the assembly path, wherein the assembly constraints include position constraints, assembly interface constraints, mechanical constraints, and motion constraints; Based on the optimal assembly path, the assembly site is controlled to perform prefabricated center modular assembly, and a three-dimensional laser scanner is used to collect point cloud data at the assembly site to establish a real-time updated point cloud data set; Using the real-time updated point cloud data set to perform the twin model comparison, and using the assembly constraints to perform deviation identification, and establish a deviation identification result; Perform a joint early warning trigger analysis on the deviation identification results to establish a monitoring early warning result.

2. The intelligent monitoring and early warning method for a prefabricated data center according to claim 1, characterized in that: The step of performing assembly fitting optimization according to the on-site scene data and the twin model to establish an optimal assembly path includes: Establishing a balance evaluation network, wherein the balance evaluation network is a network for performing adaptation evaluation of assembly paths, and the balance evaluation network includes a scene adaptation subnetwork, an assembly stability subnetwork, an assembly complexity subnetwork, and a balance analysis subnetwork; When performing assembly path fitting optimization, the balance evaluation network is activated, and the path evaluation is performed through the scene adaptation subnetwork, the assembly stability subnetwork, and the assembly complexity subnetwork respectively, and the evaluation results are synchronized to the balance analysis subnetwork to output the balance fitness; An optimization iteration is performed according to the equilibrium fitness to establish an optimal assembly path.

3. The intelligent monitoring and early warning method for a prefabricated data center according to claim 2, characterized in that: The step of performing optimization iteration according to the equilibrium fitness and establishing an optimal assembly path comprises: Perform individual encoding to encode the module's assembly path, assembly angle, and contact surface position into a vector; Randomly generate N individuals according to the individual code, and use the N individuals as the initialization population, where N is an integer greater than 1; Configure a mutation factor, use the mutation factor to perform individual mutation of the initialized population, and perform crossover discrimination of the mutated individuals based on a preset random probability to establish an updated individual; The balanced fitness is used to update the retention and screening of individuals to complete the optimization iteration.

4. The intelligent monitoring and early warning method for a prefabricated data center according to claim 3, characterized in that: The configuration variation factors include: Establish basic variability factors; Obtaining the number of iterations of the current iteration, and generating a first configuration constraint according to the number of iterations; Establishing an iterative backtracking evaluation window, performing iterative backtracking analysis using the iterative backtracking evaluation window, and establishing a second configuration constraint according to the iterative backtracking analysis result; The first configuration constraint and the second configuration constraint are used to compensate for the basic variation factor to complete the variation factor configuration.

5. The intelligent monitoring and early warning method for a prefabricated data center according to claim 1, characterized in that: The performing joint early warning trigger analysis on the deviation identification result includes: Establishing error classification levels, wherein the error classification levels include a first error classification level, a second error classification level, and a third error classification level; Performing a joint early warning trigger analysis based on the error classification level to obtain a trigger result; Establish monitoring and early warning results based on the early warning response plan corresponding to the triggering results.

6. The intelligent monitoring and early warning method for a prefabricated data center according to claim 1, characterized in that: The method of collecting point cloud data at the assembly site using a three-dimensional laser scanner and establishing a real-time updated point cloud data set includes: The accuracy requirements for obtaining the location point cloud; When point cloud data of the assembly site is collected by the three-dimensional laser scanner, a collection effect evaluation is performed; Perform adaptation analysis based on the collection effect evaluation results and the accuracy requirements, and establish adaptation exceptions; According to the adaptation exception, the linkage acquisition device is called to perform additional acquisition, and a real-time updated point cloud data set is established based on the additional acquisition.

7. The intelligent monitoring and early warning method for a prefabricated data center according to claim 1, characterized in that: The method also includes: Performing real-time video synchronization on the real-time updated point cloud dataset and the twin model to establish a real-time synchronization result; A trigger color feature is established through the monitoring and early warning results, and the real-time synchronization result is marked based on the trigger color feature to perform a visual display.

8. The intelligent monitoring and early warning method for a prefabricated data center according to claim 1, characterized in that: The method also includes: Record all monitoring and early warning results and generate early warning reports; Using the early warning report to conduct abnormal attention analysis and establish adaptive abnormal feedback; The adaptive abnormal feedback is used to perform subsequent optimization of the optimal assembly path construction.

Images