Photovoltaic construction quality real-time monitoring method based on Internet of Things

Abstract

The invention is suitable for the technical field of construction quality, and provides a photovoltaic construction quality real-time monitoring method based on the Internet of Things, and the method comprises the steps: determining a current target construction stage according to an obtained construction plan, and determining at least one quantitative quality index according to the construction quality specification of a photovoltaic component; constructing a quality verification task model in combination with Internet of Things sensor resources and image acquisition equipment resources deployed for the target construction stage on the construction site; issuing the quality verification task model to an edge intelligent gateway corresponding to the construction site, driving equipment by the edge intelligent gateway to execute the quality verification task model, acquiring multi-dimensional sensing data and image data, and performing fusion analysis; when a result corresponding to the fusion analysis triggers an abnormal condition, generating and sending early warning information associated with the target construction stage and the quantitative quality index, and recording a structured quality event; and the timeliness and the accuracy of quality monitoring are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of construction quality technology, and in particular to a method for real-time monitoring of photovoltaic construction quality based on the Internet of Things. Background Technology

[0002] The construction quality of photovoltaic power plants directly affects their long-term operational safety and power generation efficiency. In existing technologies, quality control of key processes such as photovoltaic module installation and support frame positioning primarily relies on manual measurements and records taken by construction workers based on blueprints, combined with on-site spot checks by supervisors. Some solutions are beginning to deploy sensors at specific locations for fixed-point monitoring, or utilize drones for periodic aerial photography to assist in inspections. These methods provide data recording during the construction process to some extent.

[0003] However, manual measurement and spot checks are discrete and lagging, making it impossible to continuously track quality throughout the entire construction process in real time, and difficult to promptly detect and correct construction deviations. Secondly, even with the introduction of sensors, monitoring is usually isolated and static, collecting single parameters only from preset fixed points, lacking the ability to dynamically adjust monitoring targets and strategies according to the construction progress. Furthermore, different quality indicators are collected separately by different tools or personnel, making effective spatiotemporal correlation and fusion analysis impossible, and hindering a comprehensive assessment of the quality compliance of complex installation actions. This results in low monitoring efficiency, insufficient accuracy in judgment, and an inability to achieve pre-emptive warnings and in-process control. Summary of the Invention

[0004] This invention provides a method for real-time monitoring of photovoltaic construction quality based on the Internet of Things, which solves the problems of low monitoring efficiency, insufficient judgment accuracy, and inability to achieve early warning and in-process control.

[0005] This invention provides a method for real-time monitoring of photovoltaic construction quality based on the Internet of Things, applied to a construction quality monitoring platform. The method includes: Obtain the construction plan for the target photovoltaic construction project; Based on the construction plan, determine the current target construction stage, and based on the construction quality specifications for photovoltaic components, determine at least one quantitative quality indicator corresponding to the target construction stage. Based on the quantitative quality indicators, and combined with the IoT sensor resources and image acquisition equipment resources deployed at the construction site for the target construction phase, a quality verification task model is constructed. The quality verification task model is sent to the edge smart gateway corresponding to the construction site. The edge smart gateway drives the IoT sensors and image acquisition devices to execute the quality verification task model in order to collaboratively collect multi-dimensional sensing data and image data of the target photovoltaic components, and perform fusion analysis based on the collected data. When the results of the fusion analysis trigger abnormal conditions, an early warning message associated with the target construction stage and quantitative quality indicators is generated and sent, and a structured quality event containing abnormal component identifiers, timestamps and associated data snapshots is recorded.

[0006] Furthermore, the quantitative quality indicators include at least one of the following: Position accuracy indicators are used to characterize the deviation of the installation position, tilt angle, or azimuth angle of photovoltaic modules or brackets from the design value; Mechanical condition indicators are used to characterize the stress, strain, or installation torque of supports, clamps, or connecting components; Electrical condition indicators are used to characterize the operating temperature or open-circuit voltage of photovoltaic modules during installation.

[0007] Furthermore, determining at least one quantitative quality indicator corresponding to the target construction stage based on the construction quality specifications for photovoltaic components includes: From the photovoltaic engineering construction quality acceptance specifications, photovoltaic module product technical specifications, and photovoltaic support structure design specifications, extract the allowable deviation threshold, safe load limit, or normal working parameter range associated with the target construction stage, and define the allowable deviation threshold, safe load limit, or normal working parameter range as the corresponding quantitative quality indicators.

[0008] Furthermore, the content encapsulated in the quality verification task model includes: Data acquisition rules are used to specify the sensor types, sampling frequencies, and image acquisition angles and time points involved in acquiring multidimensional sensor data and image data. Fusion analysis rules are used to specify the logic and methods for correlating, comparing, or calculating multidimensional sensor data and image data to verify the quantitative quality indicators. Anomaly response rules are used to specify the warning level, notification method, and event information to be recorded when the verification result does not meet the quantitative quality indicators.

[0009] Furthermore, based on the quantitative quality indicators and combined with the IoT sensor resources and image acquisition equipment resources deployed at the construction site, a quality verification task model is constructed, including: Based on the quantitative quality indicators, determine the data types and analysis objectives required to verify the indicators; Based on the data type, sensors and devices that can provide corresponding data are matched from the IoT sensor resources and image acquisition device resources to form a resource set; Based on the analysis objectives and resource set, a quality verification task model is generated that includes the data acquisition rules, fusion analysis rules, and anomaly response rules.

[0010] Furthermore, the generation of fusion analysis rules based on the analysis objective and resource set includes: When the quantitative quality index is the position accuracy index, the fusion analysis rule is configured as follows: perform Kalman filtering noise reduction processing on the position data collected by the lidar, and compare the processed data with the design coordinates from the engineering blueprint; When the quantitative quality index is a mechanical condition index, the fusion analysis rule is configured to: compare the real-time stress data collected by the stress sensor with the preset safe load threshold, and combine the analysis results of image recognition to make a comprehensive judgment on the compliance of the installation action; When the quantitative quality index is an electrical condition index, the fusion analysis rule is configured to: perform trend analysis on the temperature data sequence collected by the temperature sensor and compare it with the nominal temperature range of the current model photovoltaic module under the same environmental conditions.

[0011] Furthermore, the quality verification task model, driven by the edge intelligent gateway to utilize IoT sensors and image acquisition devices, collaboratively collects multi-dimensional sensing data and image data of the target photovoltaic component, including: According to the data acquisition rules in the quality verification task model, acquisition instructions are generated and issued for the IoT sensors and image acquisition devices, so that the sensors and devices can acquire data from the same target photovoltaic component at specified time points or event triggering conditions.

[0012] Furthermore, the fusion analysis based on the collected data includes: The multi-dimensional sensing data stream containing timestamps and component identifiers collected by the IoT sensor is spatiotemporally aligned and correlated with the image data stream containing corresponding timestamps and viewpoint information collected by the image acquisition device. The fusion analysis rules defined in the quality verification task model are executed to perform cross-validation and comprehensive analysis on the spatiotemporally aligned multidimensional sensing data and image data, so as to generate verification conclusions for the quantitative quality indicators.

[0013] Furthermore, the method further includes: obtaining early warning strategy configuration information bound to the target construction stage; generating and sending early warning information associated with the target construction stage and quantitative quality indicators, including: Based on the fusion analysis results and the early warning strategy configuration information, early warning information with specific levels, content formats, and target recipients is generated and sent.

[0014] Furthermore, the early warning strategy configuration information includes a tiered response strategy, which defines the early warning level, notification method, and recipient list corresponding to different anomaly severity levels; the step of generating and sending early warning information based on the fusion analysis results and the early warning strategy configuration information includes: Determine the severity of the anomalies indicated by the fusion analysis results; The corresponding early warning level, notification method, and recipient list are matched according to the tiered response strategy. According to the matching notification method, a warning message is sent to the targets in the recipient list.

[0015] As can be seen from the above technical solutions, the present invention has the following advantages: This invention determines the current target construction stage and corresponding quantitative quality indicators by acquiring construction plans, and dynamically constructs a quality verification task model that integrates data acquisition, analysis, and response rules based on this. This model is then distributed to an edge intelligent gateway, driving multiple types of sensors to collaboratively collect and analyze data, and generating early warnings when anomalies are detected. This invention solves the problem of monitoring strategies being unable to adapt to different process requirements by dynamically binding the task model to the construction stage; it improves the accuracy of anomaly identification by executing the model at the edge and collaborating with multi-source data fusion analysis; and it realizes a transformation in photovoltaic construction quality management from post-construction inspection to in-process proactive verification and intelligent control, effectively improving the timeliness and accuracy of quality monitoring. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for real-time monitoring of photovoltaic construction quality based on the Internet of Things in this invention. Figure 2 This is a schematic diagram of the process for constructing the quality verification task model in this invention; Figure 3 This is a schematic diagram of the process for performing fusion analysis based on the collected data in this invention; Figure 4 This is a schematic diagram illustrating the process of generating and sending early warning information based on the fusion analysis results and early warning strategy configuration information in this invention. Detailed Implementation

[0017] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “corresponding to,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] Example 1 Please see Figure 1 The method provided in this application is applied to a construction quality monitoring platform, which is typically deployed in the cloud and is responsible for coordinating the overall quality monitoring strategy for photovoltaic construction projects. It includes the following steps: S1. Obtain the construction plan for the target photovoltaic construction project; The construction plan is a guiding document for project construction, existing in electronic document or database form. Its content includes at least the overall project construction schedule, key milestones, and the task breakdown and sequence of each stage. The construction quality monitoring platform can automatically obtain structured construction plan data from the project management system via an application programming interface (API), or managers can upload planning documents conforming to a predetermined format. The platform parses this plan to establish a timeline model of the construction phases, serving as the basis for dynamically determining the current monitoring stage.

[0019] S2. Determine the current target construction stage based on the construction plan, and determine at least one quantitative quality indicator corresponding to the target construction stage based on the construction quality specifications of photovoltaic components. In this embodiment, the quantified quality indicators include positional accuracy indicators, mechanical condition indicators, and electrical condition indicators. Positional accuracy indicators characterize the deviation of the installation position, tilt angle, or azimuth angle of the photovoltaic module or support from the design value. Mechanical condition indicators characterize the stress, strain, or installation torque of the support, clamps, or connecting components. Electrical condition indicators characterize the operating temperature or open-circuit voltage of the photovoltaic module during installation. Specifically, quantified quality indicators refer to objective parameters whose values ​​can be directly or indirectly obtained through sensors or measuring tools and compared with preset standards to determine whether the construction quality is qualified. The construction quality specifications for photovoltaic components are a collection of technical standards from multiple fields, with the core objective of ensuring the safe, reliable, and efficient operation of photovoltaic power plants. These include national or industry-issued photovoltaic engineering construction quality acceptance specifications, which stipulate the process requirements and allowable deviations for each construction stage; product technical specifications provided by photovoltaic module manufacturers, which clearly define the electrical and physical parameter limits for modules during installation and testing; and structural design specifications for photovoltaic supports, which specify the mechanical performance requirements of the support system under different operating conditions.

[0020] In this embodiment, based on the construction quality specifications for photovoltaic components, at least one quantitative quality indicator corresponding to the target construction stage is determined, including: From the photovoltaic engineering construction quality acceptance specifications, photovoltaic module product technical specifications, and photovoltaic support structure design specifications, extract the allowable deviation thresholds, safe load limits, or normal operating parameter ranges associated with the target construction stage, and define the allowable deviation thresholds, safe load limits, or normal operating parameter ranges as corresponding quantitative quality indicators.

[0021] Specifically, based on the current target construction stage, the platform automatically retrieves all relevant specifications and clauses applicable to that stage from its knowledge base. For example, for this stage, the system will simultaneously call upon clauses in the acceptance specifications regarding component installation flatness, clauses in the product specifications regarding component operating temperature ranges, and clauses in the bracket design specifications regarding bolt tightening torque. The platform then parses these clauses, identifying all numerical requirements or limits. Finally, these numerical requirements are categorized and instantiated into monitorable quantifiable quality indicators. For example, the clause regarding component installation flatness, specifying an elevation error between adjacent components not exceeding 2mm, is instantiated as a positional accuracy indicator; the bolt tightening torque of XX N·m is instantiated as a mechanical condition indicator; and the component operating temperature not exceeding 85°C is instantiated as an electrical condition indicator. These instantiated indicators, along with their normal ranges, constitute the set of quality monitoring targets for the current construction stage.

[0022] S3. Based on quantitative quality indicators, and combined with the IoT sensor resources and image acquisition equipment resources deployed at the construction site for the target construction phase, a quality verification task model is constructed. In this embodiment, the quality verification task model encapsulates data acquisition rules, fusion analysis rules, and anomaly response rules. The data acquisition rules specify the sensor types, sampling frequencies, and image acquisition perspectives and time points involved in acquiring multidimensional sensor data and image data. The fusion analysis rules specify the logic and methods for correlating, comparing, or calculating multidimensional sensor data and image data to verify quantitative quality indicators. The anomaly response rules specify the warning level, notification method, and event information to be recorded when the verification results do not meet the quantitative quality indicators.

[0023] Specifically, data acquisition rules define the precise technical parameters of the data source to ensure that the acquired data meets the quality requirements of subsequent analysis. For example, to verify the positional accuracy index of component installation flatness, the rule might specify the use of a LiDAR device with a 3D coordinate accuracy better than ±5mm, scanning the corner points of the target component at a frequency of 1Hz; simultaneously, it might specify that a PTZ camera deployed at a high position acquires high-resolution images of the component array from a vertical top-down perspective at the synchronous time point of the radar scan. Fusion analysis rules define the processing logic and correlation methods between multi-source data. For example, for the same component, the rule might require first aligning the timestamps of the LiDAR data with the timestamps of the image frames, then extracting the outline of the component from the image, matching and verifying it with the radar point cloud projection, and finally comparing the matched actual coordinates with the theoretical coordinates in the engineering blueprint point by point to calculate the positional deviation. Anomaly response rules define the classification and handling process for quality events. Warning levels are typically divided into multiple levels, such as alert, warning, and severe, based on the degree of deviation and the possible consequences. Notification methods may include a combination of platform interface pop-ups, SMS, and application push notifications. The event information to be recorded is structured to include the component's unique ID, the time of the anomaly, the associated sensor data snapshot, image evidence, and preliminary judgment results.

[0024] Please see Figure 2 Based on quantitative quality indicators and combined with IoT sensor resources and image acquisition equipment resources deployed at the construction site, a quality verification task model is constructed, including: S31. Determine the data types and analytical objectives required to verify the quantitative quality indicators; The platform analyzes each quantitative quality indicator, breaking it down into data requirements that can be directly or indirectly observed by sensing devices. For example, the positional accuracy indicator directly corresponds to three-dimensional spatial coordinate data and component posture data, and its analysis objective is to calculate the spatial deviation between the actual value and the design value. The installation torque in the mechanical condition indicator corresponds to stress torque sensing data or indirect data calculated through image recognition of screwdriver angle and time; its analysis objective is to determine whether the tightening force has reached and stabilized within a preset safety threshold range. The electrical condition indicator corresponds to infrared thermal imaging temperature distribution data or contact temperature sensing data sequences; its analysis objective is to monitor whether the temperature is within the allowable operating range and analyze whether its changing trend is abnormal. The output of this step is a clearly defined list of data requirements and their corresponding analytical objectives.

[0025] S32. Based on the data type, match the sensors and devices that can provide the corresponding data from the IoT sensor resources and image acquisition device resources to form a resource set; All IoT sensor resources and image acquisition equipment resources at the construction site are registered on the platform, and their metadata is maintained in a resource directory. The platform performs matching queries in this resource directory based on the data requirement list generated in step S31. The matching process considers not only whether the device type can provide the required data but also verifies spatial accessibility and coverage. For example, for a specific photovoltaic support structure that needs monitoring, the system will filter out LiDAR and cameras installed at locations where the support structure is visible, as well as stress sensors attached to the support structure, combining them into a resource set serving the verification task. This set ensures that multi-dimensional data acquisition of the target component can be completed collaboratively in a physically coordinated manner.

[0026] S33. Based on the analysis objectives and resource set, generate a quality verification task model that includes data acquisition rules, fusion analysis rules, and anomaly response rules.

[0027] 1. When the quantitative quality index is the position accuracy index, the fusion analysis rule is configured as follows: perform Kalman filtering noise reduction on the position data collected by the lidar, and compare the processed data with the design coordinates from the engineering blueprint; 2. When the quantitative quality index is a mechanical condition index, the fusion analysis rule is configured as follows: compare the real-time stress data collected by the stress sensor with the preset safe load threshold, and combine the analysis results of image recognition to make a comprehensive judgment on the compliance of the installation action; 3. When the quantitative quality indicator is the electrical condition indicator, the fusion analysis rule is configured as follows: perform trend analysis on the temperature data sequence collected by the temperature sensor and compare it with the nominal temperature range of the current model of photovoltaic module under the same environmental conditions.

[0028] Based on the analysis objectives in S31 and the resource set in S32, the platform calls pre-set rule templates to generate specific task instructions. Data acquisition rules set optimized parameters according to the specific models and performance of each device in the resource set, such as setting the scanning frequency and resolution for a selected LiDAR. Fusion analysis rules integrate specific analysis algorithms and judgment thresholds. These thresholds are derived from values ​​extracted from specifications in step S2; for example, the safety load threshold is the maximum allowable stress of the specific type of support found in the support design specifications. For different types of indicators, the platform calls different analysis logic modules for configuration: for location data, a Kalman filter algorithm is configured to remove noise from the raw LiDAR data; for scenarios requiring image-assisted judgment, a pre-trained convolutional neural network model is configured to identify whether construction actions are compliant. Anomaly response rules are configured according to the importance of the construction stage and the criticality of the indicators; for example, in the electrical connection stage, unqualified insulation resistance will trigger a severe warning and simultaneously notify the site supervisor and supervising engineer. Finally, all these rules, parameters, thresholds, and resource binding information are encapsulated into a complete quality verification task model that can be independently executed at the edge.

[0029] S4. The quality verification task model is sent to the corresponding edge smart gateway at the construction site. The edge smart gateway drives the IoT sensors and image acquisition devices to execute the quality verification task model, so as to collaboratively collect multi-dimensional sensing data and image data of the target photovoltaic components, and perform fusion analysis based on the collected data. In this embodiment, based on the data acquisition rules in the quality verification task model, acquisition instructions for IoT sensors and image acquisition devices are generated and issued, so that the sensors and devices can acquire data from the same target photovoltaic component according to the specified time point or event triggering conditions.

[0030] Specifically, the data acquisition rules are based on the edge intelligent gateway's parsing and instruction conversion of the received quality verification task model. The edge intelligent gateway first parses the data acquisition rules section of the model to obtain the data types to be collected, the corresponding device resource identifiers, sampling parameters, and triggering conditions. Subsequently, the gateway generates standardized, executable acquisition instructions for each specific device. The gateway ensures consistency in the time base of all instructions through a precise time synchronization protocol, or sets a specific event as a unified triggering condition, thereby enabling coordinated acquisition of the same target component by different devices in time and space.

[0031] Please see Figure 3 Based on the collected data, a fusion analysis is performed, including: S41. Spatiotemporally align and correlate the multidimensional sensing data stream containing timestamps and component identifiers collected by IoT sensors with the image data stream containing corresponding timestamps and viewpoint information collected by the image acquisition device; S42. Execute the fusion analysis rules defined in the quality verification task model to perform cross-validation and comprehensive analysis on the spatiotemporally aligned multidimensional sensor data and image data to generate verification conclusions for the quantitative quality indicators.

[0032] Specifically, the edge intelligent gateway establishes a unified spatiotemporal reference system. For time alignment, it receives timestamp data from various sensors and image streams, aligning data streams of different frequencies to a unified time series by selecting the closest sampling points. For spatial correlation, it utilizes pre-calibrated spatial relationships between sensors and cameras to map and correlate component outlines or feature points identified in the images with 3D coordinates in the LiDAR point cloud, ensuring that each set of analyzed data points points to the same physical component. After completing data alignment and correlation, the edge intelligent gateway calls and executes the pre-defined fusion analysis rules in the quality verification task model. This process involves deep cross-validation and joint inference; for example, when verifying the mechanical state of bracket installation, the system not only compares whether stress sensor data exceeds the safe load threshold but also retrieves image analysis results from correlated time points to check whether the tightening sequence and tool usage by construction personnel are standardized. If the stress value is normal but the image shows a serious error in the installation action, the comprehensive judgment logic in the model may still determine it as abnormal. For positional accuracy, the system compares the actual coordinates, smoothed by Kalman filtering, with the design coordinates, calculates the deviation value, and determines whether the deviation exceeds the allowable error range. Finally, the system generates comprehensive verification conclusions.

[0033] S5. When the results of the fusion analysis trigger abnormal conditions, generate and send early warning information associated with the target construction stage and quantitative quality indicators, and record a structured quality event containing abnormal component identifiers, timestamps and associated data snapshots.

[0034] When the fusion analysis results trigger anomalies, the system, based on the quality verification task model that triggered the anomaly, retrospectively determines the target construction stage and specific quantitative quality indicators to which it belongs. The abnormal data is compared with preset anomaly thresholds to calculate the severity of the deviation. During this process, the system acquires the pre-bound early warning strategy configuration information associated with the current target construction stage. This configuration determines the early warning generation logic and its destination. Ultimately, the early warning information generated by the system will clearly include the aforementioned stage, indicators, abnormal components, specific deviation values, and recommended measures, thereby achieving precise and actionable early warnings.

[0035] The process of generating and sending early warning information associated with the target construction stage and quantitative quality indicators is as follows: obtaining early warning strategy configuration information bound to the target construction stage; generating and sending early warning information associated with the target construction stage and quantitative quality indicators, including: generating early warning information with specific levels, content formats and target recipients based on the fusion analysis results and early warning strategy configuration information, and sending it.

[0036] Specifically, the system extracts the core features of anomalies from the fusion analysis results, including the type of anomaly indicator and the specific value or proportion of deviation from the threshold. It then queries the early warning strategy configuration library bound to the current construction phase and, based on the rules in this library, maps the anomaly features to specific early warning levels. Simultaneously, the system determines the appropriate content format for this early warning and the preset list of target recipients based on the configuration. Finally, it calls the corresponding communication interface to send the formatted early warning information to all designated recipients.

[0037] Please see Figure 4 The early warning strategy configuration information includes a tiered response strategy, which defines the early warning level, notification method, and recipient list corresponding to different anomaly severity levels. Based on the fusion analysis results and the early warning strategy configuration information, early warning information is generated and sent, including: S51. Determine the severity of the anomalies indicated by the fusion analysis results; S52. Match the corresponding early warning level, notification method, and recipient list according to the graded response strategy; S53. Send warning information to targets in the recipient list according to the matching notification method.

[0038] Specifically, the early warning strategy configuration is stored in the construction quality monitoring platform in the form of a database or configuration file, supporting independent configuration according to different construction stages. Its core is a tiered response strategy, a set of rules that dynamically correlates the severity of the anomaly, the early warning level, the notification method, and the recipient list. The severity of the anomaly is calculated based on the deviation of a quantitative indicator, the frequency of the anomaly, and the weight of the indicator's impact on subsequent procedures or structural safety. Correspondingly, different early warning levels may trigger different visual indicators and handling priorities. The notification method is a combination of communication channels configured for different levels; for example, a reminder level may only display on the platform interface; a warning level may add app push notifications; and a severe level may simultaneously trigger SMS, telephone notifications, or even on-site audible and visual alarms. The recipient list is closely related to the construction stage and the type of anomaly; for example, abnormal stress in bracket installation must notify the structural engineer, while abnormal electrical wiring temperature must notify the electrical engineer. Through this configurable and refined strategy, the system ensures that the right information is delivered to the right people at the right time in the right way.

[0039] It is understood that those skilled in the art can combine various implementation methods in the above embodiments under the guidance of the above examples to obtain technical solutions with multiple implementation methods.

[0040] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for real-time monitoring of photovoltaic construction quality based on the Internet of Things, characterized in that, The method, applied to a construction quality monitoring platform, includes: Obtain the construction plan for the target photovoltaic construction project; Based on the construction plan, determine the current target construction stage, and based on the construction quality specifications for photovoltaic components, determine at least one quantitative quality indicator corresponding to the target construction stage. Based on the quantitative quality indicators, and combined with the IoT sensor resources and image acquisition equipment resources deployed at the construction site for the target construction phase, a quality verification task model is constructed. The quality verification task model is sent to the edge smart gateway corresponding to the construction site. The edge smart gateway drives the IoT sensors and image acquisition devices to execute the quality verification task model in order to collaboratively collect multi-dimensional sensing data and image data of the target photovoltaic components, and perform fusion analysis based on the collected data. When the results of the fusion analysis trigger abnormal conditions, an early warning message associated with the target construction stage and quantitative quality indicators is generated and sent, and a structured quality event containing abnormal component identifiers, timestamps and associated data snapshots is recorded.

2. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to claim 1, characterized in that, The quantitative quality indicators include at least one of the following: Position accuracy indicators are used to characterize the deviation of the installation position, tilt angle, or azimuth angle of photovoltaic modules or brackets from the design value; Mechanical condition indicators are used to characterize the stress, strain, or installation torque of supports, clamps, or connecting components; Electrical condition indicators are used to characterize the operating temperature or open-circuit voltage of photovoltaic modules during installation.

3. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to claim 2, characterized in that, The determination of at least one quantitative quality indicator corresponding to the target construction stage based on the construction quality specifications of photovoltaic components includes: From the photovoltaic engineering construction quality acceptance specifications, photovoltaic module product technical specifications, and photovoltaic support structure design specifications, extract the allowable deviation threshold, safe load limit, or normal working parameter range associated with the target construction stage, and define the allowable deviation threshold, safe load limit, or normal working parameter range as the corresponding quantitative quality indicators.

4. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to claim 1, characterized in that, The content encapsulated in the quality verification task model includes: Data acquisition rules are used to specify the sensor types, sampling frequencies, and image acquisition angles and time points involved in acquiring multidimensional sensor data and image data. Fusion analysis rules are used to specify the logic and methods for correlating, comparing, or calculating multidimensional sensor data and image data to verify the quantitative quality indicators. Anomaly response rules are used to specify the warning level, notification method, and event information to be recorded when the verification result does not meet the quantitative quality indicators.

5. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to claim 4, characterized in that, Based on the quantitative quality indicators, and combined with the IoT sensor resources and image acquisition equipment resources deployed at the construction site, a quality verification task model is constructed, including: Based on the quantitative quality indicators, determine the data types and analysis objectives required to verify the indicators; Based on the data type, sensors and devices that can provide corresponding data are matched from the IoT sensor resources and image acquisition device resources to form a resource set; Based on the analysis objectives and resource set, a quality verification task model is generated that includes the data acquisition rules, fusion analysis rules, and anomaly response rules.

6. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to claim 5, characterized in that, The generation of fusion analysis rules based on the analysis objectives and resource set includes: When the quantitative quality index is the position accuracy index, the fusion analysis rule is configured as follows: perform Kalman filtering noise reduction processing on the position data collected by the lidar, and compare the processed data with the design coordinates from the engineering blueprint; When the quantitative quality index is a mechanical condition index, the fusion analysis rule is configured to: compare the real-time stress data collected by the stress sensor with the preset safe load threshold, and combine the analysis results of image recognition to make a comprehensive judgment on the compliance of the installation action; When the quantitative quality index is an electrical condition index, the fusion analysis rule is configured to: perform trend analysis on the temperature data sequence collected by the temperature sensor and compare it with the nominal temperature range of the current model photovoltaic module under the same environmental conditions.

7. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to claim 1, characterized in that, The quality verification task model, driven by an edge intelligent gateway to utilize IoT sensors and image acquisition devices, collaboratively collects multi-dimensional sensing data and image data of the target photovoltaic component, including: According to the data acquisition rules in the quality verification task model, acquisition instructions are generated and issued for the IoT sensors and image acquisition devices, so that the sensors and devices can acquire data from the same target photovoltaic component at specified time points or event triggering conditions.

8. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to claim 7, characterized in that, The fusion analysis based on the collected data includes: The multi-dimensional sensing data stream containing timestamps and component identifiers collected by the IoT sensor is spatiotemporally aligned and correlated with the image data stream containing corresponding timestamps and viewpoint information collected by the image acquisition device. The fusion analysis rules defined in the quality verification task model are executed to perform cross-validation and comprehensive analysis on the spatiotemporally aligned multidimensional sensing data and image data, so as to generate verification conclusions for the quantitative quality indicators.

9. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to any one of claims 1-8, characterized in that, The method further includes: obtaining early warning strategy configuration information bound to the target construction stage; generating and sending early warning information associated with the target construction stage and quantitative quality indicators, including: Based on the fusion analysis results and the early warning strategy configuration information, early warning information with specific levels, content formats, and target recipients is generated and sent.

10. The method for real-time monitoring of photovoltaic construction quality based on the Internet of Things according to claim 9, characterized in that, The early warning strategy configuration information includes a graded response strategy, which defines the early warning level, notification method, and recipient list corresponding to different degrees of anomaly severity. The step of generating and sending early warning information based on the fusion analysis results and the early warning strategy configuration information includes: Determine the severity of the anomalies indicated by the fusion analysis results; The corresponding early warning level, notification method, and recipient list are matched according to the tiered response strategy. According to the matching notification method, a warning message is sent to the targets in the recipient list.

Images