Multi-sensor fusion-based welding wire storage environment real-time monitoring method and system

Abstract

The application provides a multi-sensor fusion welding wire storage environment real-time monitoring method and system, relates to the welding wire environment monitoring technical field, and the method comprises the following steps: collecting the environment state data of the storage box in real time through a multi-source sensor array for time sequence alignment; performing feature analysis based on the synchronous time sequence data set, extracting a plurality of feature vectors for multi-source data fusion; performing interference analysis based on the comprehensive environment state evaluation value, locking the environment interference parameter, extracting the sudden change data, combining the comprehensive environment state evaluation value to predict the welding wire storage; performing environment intelligent regulation and control on the storage box according to the early warning instruction, generating the environment regulation and control result to perform real-time closed-loop monitoring on the storage box. Through the application, the problem of single environment monitoring data in the prior art can be solved, dynamic monitoring and intelligent regulation and control based on multi-source sensors are realized, the welding quality stability is improved under the best storage condition of the welding wire, and the technical effects of operation safety and reliability are ensured.

Description

Technical Field

[0001] This application relates to the field of welding wire environmental monitoring technology, and in particular to a method and system for real-time monitoring of welding wire storage environment using multi-sensor fusion. Background Technology

[0002] As the impact of welding wire storage environment on welding quality becomes increasingly recognized, especially during the storage process where environmental factors such as humidity and temperature directly determine the quality and lifespan of the welding wire, accurate monitoring and control of the welding wire storage environment has become a crucial step in ensuring welding quality and safety. However, in practical applications, traditional environmental monitoring methods often have many shortcomings and are unable to effectively address the dynamic and complex changes in the environment during welding wire storage.

[0003] Currently, existing monitoring systems for welding wire storage environments mostly rely on single sensors or limited monitoring data sources, primarily focusing on real-time measurement and recording of temperature and humidity. These systems typically cannot achieve comprehensive analysis of multi-dimensional data and lack flexible response mechanisms to sudden environmental changes. Furthermore, existing monitoring equipment generally lacks accurate identification of environmental disturbances and the location of disturbance sources, resulting in slow responses to environmental disturbances or emergencies. For example, when environmental factors such as humidity and temperature change abruptly, existing systems cannot promptly identify and take effective measures, potentially leading to quality problems such as welding wire dampness and corrosion, and in severe cases, even affecting the normal progress of welding operations. Additionally, traditional environmental control methods are usually based on fixed control strategies, failing to adjust control strategies in real time according to different environmental changes or the storage status of the welding wire. This static control method makes the system lack adaptability and flexibility, making it difficult to dynamically adjust according to actual environmental conditions, resulting in unsatisfactory control effects. For example, when factors such as desiccant effectiveness decay or abnormal temperature and humidity fluctuations occur, existing systems lack timely response mechanisms and cannot accurately determine the safety of the welding wire storage environment, thus increasing the risk of welding wire quality damage.

[0004] In summary, existing technologies suffer from technical problems due to the limited availability of environmental monitoring data and the rigidity of control strategies. This results in insufficient precision and flexibility in the control of the welding wire storage environment, further impacting the safety of welding wire storage and the stability of welding quality. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for real-time monitoring of welding wire storage environment using multi-sensor fusion, in order to solve the technical problem in the prior art that the single nature of environmental monitoring data and the rigidity of control strategies lead to insufficient precision and flexibility in the control of welding wire storage environment, which further affects the safety of welding wire storage and the stability of welding quality.

[0006] In view of the above problems, this application provides a method and system for real-time monitoring of welding wire storage environment using multi-sensor fusion.

[0007] Firstly, this application provides a method for real-time monitoring of the welding wire storage environment using multi-sensor fusion, implemented through a real-time monitoring system for the welding wire storage environment using multi-sensor fusion. The method includes: acquiring environmental status data of the storage box in real time using a multi-source sensor array and performing time-series alignment to generate a synchronous time-series dataset; performing feature analysis based on the synchronous time-series dataset, extracting multiple feature vectors, and fusing multi-source data to generate a comprehensive environmental status assessment value; performing interference analysis based on the comprehensive environmental status assessment value to generate environmental interference parameters; extracting sudden change data by locking the environmental interference parameters and combining it with the comprehensive environmental status assessment value to predict the availability of the welding wire storage, generating predicted remaining availability information, which includes an early warning instruction; and performing intelligent environmental control of the storage box according to the early warning instruction, generating an environmental control result that is fed back to the early warning instruction for real-time closed-loop monitoring of the storage box.

[0008] Preferably, the multi-sensor fusion method for real-time monitoring of the welding wire storage environment further includes: traversing the storage box to analyze the storage environment and locating multiple environmental monitoring points based on the analysis results; performing sensor type analysis based on the multiple environmental monitoring points to determine multiple sensor types; matching multiple sensors to the multiple environmental monitoring points according to the multiple sensor types and deploying them to the multiple environmental monitoring points to construct a multi-source sensor array; activating the multi-source sensor array to collect data in real time from the multiple environmental monitoring points to obtain multiple environmental state raw data: S1: Any sensor in the multi-source sensor array collects data in real time from any environmental monitoring point to generate a raw data packet; S2: The main control unit matches the raw data packet according to the real-time clock module to generate a first timestamp; S3: The first timestamp is bound to the raw data packet, and this process is iterated until the data collection of multiple environmental monitoring points is completed to obtain multiple environmental state raw data; the multiple environmental state raw data are sorted according to the timestamp order to generate the synchronous time-series dataset.

[0009] Preferably, the multi-sensor fusion method for real-time monitoring of welding wire storage environment further includes: defining a time reference axis, mapping the multiple environmental state raw data to the time reference axis, determining multiple target time points, wherein the multiple target time points correspond to the multiple environmental state raw data; traversing the multiple target time points and timestamps in sequence for matching and searching: S1: extracting a first target time point based on the target time point matching the timestamp, and determining the first environmental state raw data based on the first target time point; S2: when the multiple target time points do not match the timestamps, calculating the distance between the timestamps and the multiple target time points, generating an adjacent distance parameter set, extracting the minimum distance value based on the adjacent distance parameter set, determining adjacent time points, and determining the second environmental state raw data based on the adjacent time points; S3: when the adjacent distance parameter set is empty, estimating based on the multiple environmental state raw data to generate the third environmental state raw data; S4: sorting the first environmental state raw data, the second environmental state raw data, and the third environmental state raw data according to the time reference axis to generate the synchronous time series dataset.

[0010] Preferably, the multi-sensor fusion method for real-time monitoring of the welding wire storage environment further includes: performing multi-dimensional feature analysis on the synchronous time-series dataset to generate an original feature set; performing min-max normalization on the original feature set to generate feature-normalized data; performing correlation calculation on the original feature set based on the feature-normalized data to generate multiple correlation coefficients; traversing the original feature set according to the multiple correlation coefficients to perform correlation filtering processing to generate multiple feature parameters; arranging the multiple feature parameters in chronological order to construct a core feature vector; constructing a welding wire storage prior knowledge graph to assign weights to the core feature vector and determine multiple initial weight coefficients; multiplying the multiple initial weight coefficients with the core feature vector to obtain multiple feature weighted scores; and pre-setting an evaluation value range, mapping the multiple feature weighted scores to the evaluation value range to generate the comprehensive environmental status evaluation value.

[0011] Preferably, the multi-sensor fusion-based real-time monitoring method for welding wire storage environment further includes: performing serialized fluctuation analysis based on the comprehensive environmental state assessment value to determine the fluctuation pattern; identifying abnormal fluctuations in the comprehensive environmental state assessment value according to the fluctuation pattern, and locking environmental interference parameters based on the identification results, wherein the environmental interference parameters include environmental interference event types and key interference parameters; associating interference features based on the environmental interference event types and the key interference parameters to determine the environmental interference event set; constructing an environmental state time series curve based on the comprehensive environmental state assessment value, mapping the environmental interference event set to the environmental state time series curve for fluctuation abrupt change identification and division, and generating multiple feature abrupt change data segments; and performing availability prediction based on the environmental interference event types, the key interference parameters, the multiple feature abrupt change data segments, and the comprehensive environmental state assessment value to generate the predicted remaining availability information.

[0012] Preferably, the multi-sensor fusion method for real-time monitoring of the welding wire storage environment further includes: segmenting and matching the multiple characteristic abrupt change data segments based on the environmental interference event type and the key interference parameters to obtain multiple types of characteristic abrupt change events, including storage and retrieval operation interference events, desiccant effectiveness decay events, and thermal environment disturbance events; performing humidity intrusion analysis based on the storage and retrieval operation interference events to generate additional wet load parameters; performing adsorption decay calculation based on the desiccant effectiveness decay events to generate adsorption capacity decay rate; performing joint analysis on the additional wet load parameters and the adsorption capacity decay rate according to the thermal environment disturbance events to generate joint analysis results; predicting the availability of the welding wire based on the joint analysis results to generate predicted desiccant remaining effective life information and welding wire remaining safe storage time information; issuing a life warning for the welding wire based on the predicted desiccant remaining effective life information and welding wire remaining safe storage time, generating a warning command, and adding the warning command to the predicted remaining availability information.

[0013] Preferably, the multi-sensor fusion method for real-time monitoring of the welding wire storage environment further includes: mapping and converting the comprehensive environmental state assessment value into equivalent average humidity data; introducing welding wire material information to perform moisture absorption and deterioration calculations, setting a critical humidity threshold for deterioration and performing time-series cumulative analysis to construct a time-humidity cumulative effect relationship; calculating the critical cumulative moisture absorption time of the welding wire based on the equivalent average humidity data according to the time-humidity cumulative effect relationship; using the critical cumulative moisture absorption time of the welding wire as a basic safe time to perform over-limit calculations on multiple characteristic sudden change data segments to generate an over-limit cumulative duration; performing a moisture damage simulation on the welding wire based on the over-limit cumulative duration weighted and superimposed on the basic safe time to generate simulated moisture damage cumulative data; and subtracting the basic safe time based on the simulated moisture damage cumulative data to generate the remaining safe storage time information of the welding wire.

[0014] Preferably, the multi-sensor fusion method for real-time monitoring of welding wire storage environment further includes: performing a lifespan criticality calculation based on the predicted remaining effective lifespan information of the desiccant, and setting a lifespan warning threshold; performing a lifespan behavior impact analysis based on the predicted remaining effective lifespan information of the desiccant, and setting a lifespan behavior threshold; generating a first warning instruction when the predicted remaining effective lifespan information of the desiccant is lower than the lifespan behavior threshold, and generating a second warning instruction when the predicted remaining effective lifespan information of the desiccant is lower than the lifespan warning threshold and higher than the lifespan behavior threshold; performing a time criticality calculation based on the basic safety time, and setting a safety time warning threshold; performing a time behavior impact analysis based on the basic safety time, and setting a safety time action threshold; generating a third warning instruction when the remaining safe storage time of the welding wire is lower than the safety time action threshold, and generating a fourth warning instruction when the remaining safe storage time of the welding wire is lower than the safety time action threshold and higher than the safety time action threshold.

[0015] Preferably, the multi-sensor fusion method for real-time monitoring of the welding wire storage environment further includes: traversing the first warning command, the second warning command, the third warning command, and the fourth warning command for priority analysis to construct multiple command priorities; constructing a control strategy library based on historical control logs, matching the multiple command priorities with the first warning command, the second warning command, the third warning command, and the fourth warning command, mapping the matching results to the control strategy library for retrieval, and generating environmental control operation commands, wherein the environmental control operation commands correspond to the first warning command, the second warning command, the third warning command, and the fourth warning command; issuing the environmental control operation commands to the storage box to drive the actuator to perform environmental control and generate environmental control results; feeding back the environmental control results to the warning command to activate the enhanced monitoring mode for high-frequency data acquisition, updating the warning command to construct a monitoring optimization strategy for real-time closed-loop monitoring of the storage box.

[0016] Secondly, this application also provides a multi-sensor fusion real-time monitoring system for welding wire storage environment, used to execute the multi-sensor fusion real-time monitoring method for welding wire storage environment as described in the first aspect, including: a synchronous time-series dataset generation module, used to collect environmental state data of the storage box in real time through a multi-source sensor array and perform time-series alignment to generate a synchronous time-series dataset; a comprehensive environmental state assessment value generation module, used to perform feature analysis based on the synchronous time-series dataset, extract multiple feature vectors for multi-source data fusion, and generate a comprehensive environmental state assessment value; a predicted remaining available information generation module, used to perform interference analysis based on the comprehensive environmental state assessment value, generate environmental interference parameters, extract sudden change data by locking the environmental interference parameters and combine them with the comprehensive environmental state assessment value to predict the availability of welding wire storage, and generate predicted remaining available information, wherein the predicted remaining available information includes an early warning instruction; and an environmental control result generation module, used to perform intelligent environmental control of the storage box according to the early warning instruction, generate environmental control results and feed them back to the early warning instruction for real-time closed-loop monitoring of the storage box.

[0017] The technical solution provided in this application has at least the following technical effects or advantages: by realizing the technical goal of dynamic monitoring and intelligent control based on multi-source sensor data, it achieves real-time and accurate multi-dimensional monitoring of the welding wire storage environment, can flexibly respond to environmental changes and automatically adjust the control strategy, and ensure that the welding wire is always kept under the best storage conditions, thereby improving the stability of welding quality and ensuring the safety and reliability of welding operations.

[0018] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the real-time monitoring method for the welding wire storage environment using multi-sensor fusion based on this application.

[0021] Figure 2This is a schematic diagram of the multi-sensor fusion real-time monitoring system for welding wire storage environment according to this application.

[0022] Figure labeling: Synchronous time series dataset generation module 1, comprehensive environmental state assessment value generation module 2, prediction of remaining available information generation module 3, environmental control result generation module 4. Detailed Implementation

[0023] This application provides a method and system for real-time monitoring of welding wire storage environment using multi-sensor fusion. This addresses the technical problems in existing technologies where the limited availability of environmental monitoring data and rigid control strategies lead to insufficient precision and flexibility in controlling the welding wire storage environment, further impacting the safety of welding wire storage and the stability of welding quality. The method achieves dynamic monitoring and intelligent control based on multi-source sensor data, enabling real-time, precise, and multi-dimensional monitoring of the welding wire storage environment. It can flexibly respond to environmental changes and automatically adjust control strategies, ensuring the welding wire is always kept under optimal storage conditions. This improves the stability of welding quality and guarantees the safety and reliability of welding operations.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a method for real-time monitoring of welding wire storage environment based on multi-sensor fusion, which is applied to a real-time monitoring system for welding wire storage environment based on multi-sensor fusion, and specifically includes the following steps:

[0026] The environmental status data of the storage box is collected in real time by a multi-source sensor array and time-series aligned to generate a synchronized time-series dataset.

[0027] Furthermore, this application also includes: traversing the storage box to perform storage environment analysis, and locating multiple environmental detection points based on the analysis results; performing sensor type analysis based on the multiple environmental detection points to determine multiple sensor types; matching multiple sensors to the multiple environmental detection points according to the multiple sensor types and deploying them to the multiple environmental detection points to construct a multi-source sensor array; activating the multi-source sensor array to collect data from the multiple environmental detection points in real time, obtaining multiple environmental state raw data: S1: Any sensor in the multi-source sensor array collects data from any environmental detection point in real time, generating a raw data packet; S2: The main control unit matches the raw data packet according to the real-time clock module to generate a first timestamp; S3: Binding the first timestamp to the raw data packet, and iterating until the data collection of multiple environmental detection points is completed, obtaining multiple environmental state raw data; sorting the multiple environmental state raw data according to the timestamp order to generate the synchronous time-series dataset.

[0028] Furthermore, this application also includes: defining a time reference axis, mapping the multiple environmental state raw data to the time reference axis, determining multiple target time points, wherein the multiple target time points correspond to the multiple environmental state raw data; traversing the multiple target time points and timestamps in sequence for matching and searching: S1: extracting a first target time point based on the target time point matching the timestamp, and determining the first environmental state raw data based on the first target time point; S2: when the multiple target time points do not match the timestamps, calculating the distance between the timestamps and the multiple target time points, generating an adjacent distance parameter set, extracting the minimum distance value based on the adjacent distance parameter set, determining adjacent time points, and determining the second environmental state raw data based on the adjacent time points; S3: when the adjacent distance parameter set is empty, estimating based on the multiple environmental state raw data to generate the third environmental state raw data; S4: sorting the first environmental state raw data, the second environmental state raw data, and the third environmental state raw data according to the time reference axis to generate the synchronous time series dataset.

[0029] Specifically, a comprehensive analysis of the storage box environment is conducted, and inspections are carried out based on environmental characteristics and conditions. This may include the measurement values ​​of multiple environmental parameters such as temperature, humidity, air circulation, light, and vibration. The distribution and changing trends of environmental parameters and their relationship with the safety of welding wire storage are analyzed to identify multiple key locations that may affect the storage status of welding wire, which are the test points for subsequent sensor deployment and data acquisition.

[0030] After locating the points to be monitored, the next step is to select and analyze the types of sensors. Different environments may require different sensors to accurately monitor parameters such as temperature, humidity, and pressure. Therefore, by analyzing the characteristics of the points to be monitored, the types of sensors that need to be deployed are determined to ensure that the required environmental data can be effectively acquired.

[0031] Based on the identified sensor types, the corresponding sensor devices will be deployed to various monitoring points to construct a multi-source sensor array. This ensures that necessary environmental data can be acquired in real time and comprehensively under different environmental conditions. The multi-source sensor array improves the accuracy and reliability of monitoring through the collaborative work of different types of sensors.

[0032] Once the multi-source sensor array is deployed, each sensor can be activated to collect real-time data from multiple points to be tested, including various parameters related to the welding wire storage environment such as temperature, humidity, and air pressure, which will serve as the basis for subsequent analysis.

[0033] During this process, each sensor in the multi-source sensor array acquires real-time data at the designated detection point. Each sensor generates a raw data packet based on its set acquisition range and acquisition method, recording the environmental parameters of that point as the raw data source for subsequent processing.

[0034] After data acquisition is complete, the main control unit coordinates the operation of all sensors and timestamps each raw data packet using a real-time clock module. To ensure data consistency, the main control unit generates a first timestamp based on the acquisition time, recording the exact time point of each data packet acquisition.

[0035] The main control unit binds timestamps to the raw data packets to ensure that each data packet has an accurate time identifier. The entire data acquisition process is iterative until data acquisition is completed at all test points, ultimately obtaining multiple sets of raw environmental state data.

[0036] A unified time reference framework is established by defining a time baseline. Then, the collected raw environmental state data are mapped onto this time baseline, with each data point having a defined time coordinate. Next, based on the time baseline, multiple target time points are determined. These target time points are specific time points corresponding to each raw environmental state data point, used to establish the correlation between data and time.

[0037] Next, a traversal matching process is performed based on the relationship between the target time point and the timestamps of the actual collected data. The target time point is the key element for matching with the timestamps of the collected data, ensuring that each data point can accurately correspond to a position on the timeline, thereby ensuring the temporality and consistency of the data. First, a target time point is matched based on the timestamps, and the raw environmental state data associated with the target time point is extracted. Based on the target time point, the first set of raw environmental state data is determined as preliminary data for subsequent processing, for further analysis and fusion.

[0038] If the target time point and the timestamp do not perfectly match, a set of adjacent distance parameters is generated by calculating the distance between the timestamp and the target time point. Based on the adjacent distance parameter set, the smallest distance value is selected to determine the closest time point. Based on the adjacent time points, the original data of the second environmental state is determined to fill in missing data points or supplement time series information.

[0039] If no adjacent distance parameters are available for reference, new data will be generated by estimating multiple raw environmental state data. The estimation process is based on existing raw data to fill in gaps and ensure data continuity and integrity. Specifically, the estimation process first performs time-series analysis on the existing raw data to identify trends and patterns in data changes. Then, interpolation calculations are performed based on environmental state data at adjacent time points, using methods such as linear interpolation and curve fitting to extrapolate the data values ​​for missing time points from existing data points. For example, linear interpolation calculates the estimated value of missing data points proportionally based on known data from preceding and following time points, ensuring a smooth data transition. Alternatively, based on the slope or rate of change of preceding and following data points, the possible values ​​of missing points can be inferred, thus filling data gaps. Furthermore, the estimation process also considers the physical characteristics of environmental changes and system operating characteristics to ensure that the generated data conforms to reality. This effectively fills data gaps and maintains the continuity of the data sequence, thereby meeting the integrity requirements of environmental monitoring data.

[0040] Finally, all collected raw environmental state data, including the raw data for the first, second, and third environmental states obtained through matching and estimation, will be sorted according to the time baseline. This sorting process generates a complete synchronous time-series dataset that accurately reflects the changing environmental states and provides a data foundation for subsequent analysis and prediction.

[0041] Feature analysis is performed on the synchronous time-series dataset to extract multiple feature vectors for multi-source data fusion, generating a comprehensive environmental state assessment value.

[0042] Furthermore, this application also includes: performing multi-dimensional feature analysis on the synchronous time-series dataset to generate an original feature set; performing min-max normalization on the original feature set to generate feature-normalized data; performing correlation calculation on the original feature set based on the feature-normalized data to generate multiple correlation coefficients; traversing the original feature set according to the multiple correlation coefficients to perform correlation filtering processing to generate multiple feature parameters; arranging the multiple feature parameters in temporal order to construct a core feature vector; constructing a prior knowledge graph of welding wire storage to assign weights to the core feature vector and determine multiple initial weight coefficients; multiplying the multiple initial weight coefficients with the core feature vector to obtain multiple feature weighted scores; and pre-setting an evaluation value range, mapping the multiple feature weighted scores to the evaluation value range to generate the comprehensive environmental state evaluation value.

[0043] Specifically, by performing multi-dimensional feature analysis on the synchronous time-series dataset, including humidity, temperature, and event dimensions, physically meaningful feature parameters are extracted. Each dimension contains key indicators of the relevant environmental state. Through data extraction and analysis, a raw feature set is formed, providing basic data for subsequent processing and optimization.

[0044] To ensure consistent numerical ranges for different feature parameters for subsequent processing and comparison, a min-max normalization method is used to normalize each feature in the original feature set. All feature values ​​are adjusted to a uniform range of 0 to 1, generating normalized feature data and eliminating biases caused by different units between features.

[0045] After obtaining the normalized data, correlation analysis is performed on the feature normalized data to quantitatively analyze the relationships between feature data, revealing the degree of association between different features. Multiple correlation coefficients are then calculated to reflect the strength of the correlation between different features. The calculation of correlation coefficients helps identify the intrinsic relationships between features, thus providing guidance for feature selection and subsequent optimization.

[0046] Based on correlation coefficients, the original feature set is filtered. By traversing each feature, feature parameters highly relevant to the target task are selected. This filtering process helps remove redundant or irrelevant features, ensuring that the final features effectively reflect changes in the environmental state and play a crucial role in subsequent analysis.

[0047] After feature selection, the selected feature parameters are arranged in chronological order to form a core feature vector. This feature vector contains important information reflecting the environmental conditions of the welding wire storage environment. The chronological order ensures the temporal consistency of the data, enabling subsequent analysis to accurately capture the dynamic changes over time.

[0048] By constructing a prior knowledge graph storing the welding wire, weights are assigned to each feature in the core feature vector. Based on practical experience and domain knowledge, the prior knowledge graph assigns an initial weight coefficient to each feature, reflecting the importance of each feature in environmental state assessment. This helps improve the accuracy and reliability of the assessment model.

[0049] After determining the initial weighting coefficients, these coefficients are multiplied by each feature in the core feature vector to generate a weighted score for each feature. Combining the importance of different features with their numerical values ​​provides weighted information for subsequent comprehensive evaluation, ensuring higher accuracy in assessing the environmental state.

[0050] Finally, based on the actual needs and experience of the welding wire storage environment, an evaluation value range is preset. This range represents the quantitative range of the overall environmental condition, including certain upper and lower limits, such as 0 to 100. The setting of the evaluation value range is based on the environmental factors and standards involved in the welding wire storage process, ensuring that the range accurately reflects the quality of the environmental condition. Weighted scores are then mapped to the evaluation value range. The mapping process converts the weighted scores into values ​​that match the range. For example, linear, nonlinear, or piecewise mapping methods may be used to adjust the feature scores to the defined range, thereby generating a comprehensive environmental condition evaluation value. This value reflects the overall condition of the welding wire storage environment, providing a basis for subsequent intelligent control and early warning systems, and enabling real-time monitoring and optimization of the storage environment.

[0051] Interference analysis is performed based on the comprehensive environmental condition assessment value to generate environmental interference parameters. By locking the environmental interference parameters and extracting sudden change data, the availability of welding wire storage is predicted in combination with the comprehensive environmental condition assessment value, and the predicted remaining availability information is generated, which includes early warning instructions.

[0052] Furthermore, this application also includes: performing serialized fluctuation analysis based on the comprehensive environmental state assessment value to determine the fluctuation pattern; identifying abnormal fluctuations in the comprehensive environmental state assessment value according to the fluctuation pattern, and locking environmental interference parameters based on the identification results, wherein the environmental interference parameters include environmental interference event types and key interference parameters; performing interference feature association based on the environmental interference event types and the key interference parameters to determine the environmental interference event set; constructing an environmental state time series curve based on the comprehensive environmental state assessment value, mapping the environmental interference event set to the environmental state time series curve for fluctuation sudden change identification and division, and generating multiple feature sudden change data segments; and performing availability prediction based on the environmental interference event types, the key interference parameters, the multiple feature sudden change data segments, and the comprehensive environmental state assessment value to generate the predicted remaining availability information.

[0053] Furthermore, this application also includes: segmenting and matching the multiple characteristic abrupt change data segments based on the environmental interference event type and the key interference parameters to obtain multiple types of characteristic abrupt change events, including access operation interference events, desiccant effectiveness decay events, and thermal environment disturbance events; performing humidity intrusion analysis based on the access operation interference events to generate additional wet load parameters; performing adsorption decay calculation based on the desiccant effectiveness decay events to generate adsorption capacity decay rate; performing joint analysis on the additional wet load parameters and the adsorption capacity decay rate according to the thermal environment disturbance events to generate joint analysis results; predicting the availability of the welding wire based on the joint analysis results to generate predicted desiccant remaining effective life information and welding wire remaining safe storage time information; issuing a life warning for the welding wire based on the predicted desiccant remaining effective life information and welding wire remaining safe storage time, generating a warning command, and adding the warning command to the predicted remaining availability information.

[0054] Furthermore, this application also includes: mapping the comprehensive environmental condition assessment value into equivalent average humidity data; introducing welding wire material information to calculate moisture absorption and deterioration, setting a critical humidity threshold for deterioration and performing time-series cumulative analysis to construct a time-humidity cumulative effect relationship; calculating the critical cumulative moisture absorption time of the welding wire based on the equivalent average humidity data according to the time-humidity cumulative effect relationship; using the critical cumulative moisture absorption time of the welding wire as a basic safe time to perform over-limit calculations on multiple characteristic sudden change data segments to generate an over-limit cumulative duration; performing a moisture damage simulation on the welding wire based on the over-limit cumulative duration weighted by time and superimposed on the basic safe time to generate simulated moisture damage cumulative data; and reducing the basic safe time based on the simulated moisture damage cumulative data to generate the remaining safe storage time information of the welding wire.

[0055] Furthermore, this application also includes: performing a lifespan criticality calculation based on the predicted remaining effective lifespan information of the desiccant, and setting a lifespan warning threshold; performing a lifespan behavior impact analysis based on the predicted remaining effective lifespan information of the desiccant, and setting a lifespan behavior threshold; generating a first warning instruction when the predicted remaining effective lifespan information of the desiccant is lower than the lifespan behavior threshold, and generating a second warning instruction when the predicted remaining effective lifespan information of the desiccant is lower than the lifespan warning threshold and higher than the lifespan behavior threshold; performing a time criticality calculation based on the basic safety time, and setting a safety time warning threshold; performing a time behavior impact analysis based on the basic safety time, and setting a safety time action threshold; generating a third warning instruction when the remaining safe storage time of the welding wire is lower than the safety time action threshold, and generating a fourth warning instruction when the remaining safe storage time of the welding wire is lower than the safety time action threshold and higher than the safety time action threshold.

[0056] Specifically, a sequential fluctuation analysis is conducted based on the comprehensive environmental status assessment values. By analyzing the changing trends and fluctuation characteristics of the assessment values, fluctuation patterns of the environmental status over a certain period can be identified. These fluctuation patterns reflect the normal fluctuation patterns of the environmental status and help in further analyzing abnormal changes in the environmental status.

[0057] After identifying abnormal fluctuations in the environmental status assessment values ​​based on the fluctuation patterns, further analysis and identification of environmental interference parameters that cause these abnormal fluctuations are performed. This includes environmental interference event types such as rapid changes in temperature and humidity, as well as key interference parameters such as the amplitude or frequency of environmental changes. By locking in the interference parameters, the specific factors affecting the welding wire storage environment can be located.

[0058] After identifying the types of environmental interference events and key interference parameters, a correlation analysis of the interference characteristics was conducted. By analyzing the relationships between different interference parameters, a set of environmental interference events was constructed, including various environmental interference events that may affect the storage status of the welding wire, providing data support for subsequent interference analysis.

[0059] Using comprehensive environmental state assessment values, a time-series curve of the environmental state is constructed, and the identified set of environmental disturbance events is mapped onto the time-series curve. Sudden fluctuations and abrupt changes are marked on the time-series curve to identify sudden changes caused by environmental disturbances, generating multiple characteristic abrupt change data segments to capture the impact of environmental changes on welding wire storage.

[0060] By analyzing and segmenting multiple characteristic abrupt change data segments, and combining the types of environmental interference events and key interference parameters, different characteristic abrupt change events were identified, including storage and retrieval operation interference events, desiccant effectiveness decay events, and thermal environment disturbance events, which respectively represent different environmental interference factors and their impact on welding wire storage conditions.

[0061] When analyzing interference events during access operations, a humidity intrusion analysis is further performed. By calculating the humidity intrusion, additional moisture load parameters are generated, reflecting the additional impact of humidity on the welding wire storage environment during access operations, which helps predict the degradation and moisture damage risk of the welding wire.

[0062] For events related to desiccant effectiveness degradation, adsorption decay calculations are required. By analyzing the decline in the desiccant's adsorption capacity during storage, its decay rate can be determined. This provides insights into the changing trend of desiccant effectiveness, helping to assess its sustainability and effectiveness in the storage environment.

[0063] By combining additional wet load parameters with the adsorption capacity decay rate for joint analysis, especially under the influence of thermal environmental disturbances, a more accurate comprehensive assessment of the impact of environmental changes on welding wire storage can be obtained, providing data support for subsequent disturbance prediction and environmental control.

[0064] During the environmental condition assessment process, the comprehensive environmental condition assessment value is mapped to equivalent average humidity data. This simplifies the complex environmental conditions into a humidity-based representation, facilitating subsequent analysis of welding wire storage conditions and prediction of moisture damage.

[0065] By incorporating the material information of the welding wire and combining it with the influence of humidity, moisture absorption degradation calculations are performed. A critical humidity threshold for degradation is set for time-series cumulative analysis, and a time-humidity cumulative effect relationship is constructed to describe the deterioration effect of humidity on the welding wire material.

[0066] By using equivalent average humidity data and calculating the time-humidity cumulative effect relationship, the critical cumulative moisture absorption time of the welding wire is obtained. This time represents the time when the cumulative effect of humidity on the welding wire reaches the critical point during storage, which is of great significance for predicting the deterioration and failure of the welding wire.

[0067] Based on the critical cumulative moisture absorption time of the welding wire, it is used as the basic safe time. The over-limit calculation of multiple characteristic sudden change data segments is performed to obtain an over-limit cumulative time, which represents the cumulative time of the welding wire exceeding the safe range. This helps to determine the safe storage period of the welding wire.

[0068] Based on the cumulative duration of exceeding the limit, the data is weighted and added to the basic safe time to simulate the wet damage of the welding wire. By simulating the cumulative wet damage, the long-term deterioration process of the welding wire under the influence of humidity can be evaluated, generating simulated cumulative wet damage data.

[0069] Based on the simulated cumulative wet damage data, the basic safe storage time is reduced to obtain the remaining safe storage time information of the welding wire, which reflects the remaining safe storage time of the welding wire under the current storage environment and provides data support for the early warning system.

[0070] By predicting the remaining effective lifespan of the desiccant, its usage status in the current environment is assessed. Based on the prediction results, a critical lifespan calculation is performed, and a critical threshold is set to determine the risk of insufficient effectiveness of the desiccant when its remaining effective lifespan reaches this threshold. The lifespan warning threshold is set to identify the potential desiccant failure point in advance, ensuring the stability of the welding wire storage environment.

[0071] Next, by analyzing the predicted remaining effective lifespan of the desiccant, the impact of changes in desiccant behavior on the overall storage environment is assessed. Based on the analysis results, a lifespan behavior threshold is set, indicating when the desiccant's effectiveness begins to decline to a certain extent, affecting the humidity control or drying effect of the storage environment. This lifespan behavior threshold is used to further identify early signs of desiccant effectiveness degradation, allowing for proactive countermeasures.

[0072] The condition of the desiccant is assessed based on changes in its remaining effective lifespan. When the remaining lifespan falls below the lifespan behavior threshold, a first warning instruction is generated, prompting immediate replacement of the desiccant. Conversely, when the remaining lifespan is below the lifespan warning threshold but still above the lifespan behavior threshold, a second warning instruction is generated, suggesting preparations for desiccant replacement. This ensures effective early warning before the desiccant's effectiveness declines, preventing instability in the storage environment.

[0073] In addition to the desiccant's lifespan, the safe storage time of the welding wire also needs to be considered. Based on a baseline safe storage time, the safe storage period of the welding wire under current environmental conditions is calculated, and a safe storage time warning threshold is set to identify the risk that the welding wire's storage time is about to exceed the safe range, alerting users to the need for protective measures.

[0074] Next, we analyze the impact of the baseline safe storage time on the behavior of welding wire storage, and assess the potential impact of the storage environment on the safety of the welding wire when it approaches the critical point of the safe storage time. We set a safe storage time action threshold so that when the remaining safe storage time of the welding wire approaches this threshold, we can issue an early warning and initiate emergency measures.

[0075] Based on the remaining safe storage time of the welding wire, an early warning instruction is generated when the storage time approaches the safe storage threshold. When the remaining storage time falls below the safe time action threshold, a third early warning instruction is generated, prompting emergency measures. When the remaining storage time is below this threshold but still above the action threshold, a fourth early warning instruction is generated, suggesting preparation for action. This early warning mechanism ensures the safe storage of the welding wire and prevents it from being affected by environmental factors. Incorporating the early warning instructions into the predicted remaining usable information provides a reference for subsequent environmental control and decision-making.

[0076] The storage box is intelligently environmentally controlled according to the warning command, and the environmental control results are fed back to the warning command to perform real-time closed-loop monitoring of the storage box.

[0077] Furthermore, this application also includes: traversing the first warning instruction, the second warning instruction, the third warning instruction, and the fourth warning instruction for priority analysis to construct multiple instruction priorities; constructing a control strategy library based on historical control record logs, matching the multiple instruction priorities with the first warning instruction, the second warning instruction, the third warning instruction, and the fourth warning instruction, mapping the matching results to the control strategy library for retrieval, generating environmental control operation instructions, wherein the environmental control operation instructions correspond to the first warning instruction, the second warning instruction, the third warning instruction, and the fourth warning instruction; issuing the environmental control operation instructions to a storage box to drive the actuator to perform environmental control and generate environmental control results; feeding back the environmental control results to the warning instruction to activate the enhanced monitoring mode for high-frequency data collection, updating the warning instruction to construct a monitoring optimization strategy for real-time closed-loop monitoring of the storage box.

[0078] Specifically, the four warning instructions are prioritized. The priority of each warning instruction is analyzed and evaluated based on its urgency and the significance of its impact on the welding wire storage environment. For example, when the remaining desiccant life is very low, the first warning instruction might be given the highest priority, while when the remaining safe storage time for the welding wire is long, the fourth warning instruction might be given a lower priority. This priority analysis ensures that the optimal response can be made according to the urgency of each warning instruction.

[0079] Next, historical control records and a control strategy library are used to match strategies based on the priority of early warning commands. The historical control logs provide past environmental control data and strategies, while the control strategy library contains control methods to be adopted under different circumstances. Based on priority ranking and the specific content of the early warning commands, the most appropriate control operation command is retrieved from the control strategy library, thereby ensuring effective environmental control. There is a one-to-one correspondence between each control operation command and its corresponding early warning command, ensuring accurate response to each early warning command during control implementation.

[0080] Then, environmental control operation instructions are issued to the storage box based on the matching and retrieval results. The instructions include operations such as adjusting temperature, humidity, and air circulation. The actuator adjusts the environment inside the storage box according to these instructions and generates corresponding environmental control results based on the control effect. This feedback is used to determine whether the storage environment has met the expected safety standards, ensuring that the welding wire storage conditions always meet the requirements.

[0081] Finally, once the environmental control results are generated, they are fed back to the early warning command, activating the enhanced monitoring mode. This leads to a more frequent data acquisition mode, performing high-frequency monitoring of the storage environment to obtain more accurate and real-time environmental data. Based on the updated data, the early warning command will be readjusted, and the monitoring strategy will be optimized according to the new data, thereby achieving real-time closed-loop monitoring of the storage container environment and ensuring that environmental control can continuously and effectively respond to various changes.

[0082] In summary, the multi-sensor fusion real-time monitoring method for welding wire storage environment provided in this application has the following technical effects: by realizing the technical goal of dynamic monitoring and intelligent control based on multi-source sensor data, it achieves real-time and accurate multi-dimensional monitoring of the welding wire storage environment, can flexibly respond to environmental changes and automatically adjust the control strategy, and ensure that the welding wire is always kept under the best storage conditions, thereby improving the stability of welding quality and ensuring the safety and reliability of welding operations.

[0083] Example 2: Based on the same inventive concept as the multi-sensor fusion real-time monitoring method for welding wire storage environment in the foregoing examples, this application also provides a multi-sensor fusion real-time monitoring system for welding wire storage environment. Please refer to the appendix. Figure 2The system includes: a synchronous time-series dataset generation module 1, used to collect environmental status data of the storage box in real time through a multi-source sensor array and perform time-series alignment to generate a synchronous time-series dataset; a comprehensive environmental status assessment value generation module 2, used to perform feature analysis based on the synchronous time-series dataset, extract multiple feature vectors, perform multi-source data fusion, and generate a comprehensive environmental status assessment value; a predicted remaining usable information generation module 3, used to perform interference analysis based on the comprehensive environmental status assessment value, generate environmental interference parameters, extract sudden change data by locking the environmental interference parameters and combine them with the comprehensive environmental status assessment value to predict the availability of the welding wire storage, and generate predicted remaining usable information, which includes early warning instructions; and an environmental control result generation module 4, used to perform intelligent environmental control of the storage box according to the early warning instructions, generate environmental control results and feed them back to the early warning instructions for real-time closed-loop monitoring of the storage box.

[0084] Furthermore, the multi-sensor fusion welding wire storage environment real-time monitoring system is also used for: traversing the storage box to analyze the storage environment, locating multiple environmental monitoring points based on the analysis results; performing sensor type analysis based on the multiple environmental monitoring points to determine multiple sensor types; matching multiple sensors to the multiple environmental monitoring points according to the multiple sensor types and deploying them to the multiple environmental monitoring points to construct a multi-source sensor array; activating the multi-source sensor array to collect data on the multiple environmental monitoring points in real time, obtaining multiple environmental state raw data: S1: Any sensor in the multi-source sensor array collects data on any environmental monitoring point in real time, generating a raw data packet; S2: The main control unit matches the raw data packet according to the real-time clock module to generate a first timestamp; S3: Binding the first timestamp to the raw data packet, and iterating until the data collection of multiple environmental monitoring points is completed, obtaining multiple environmental state raw data; sorting the multiple environmental state raw data according to the timestamp order to generate the synchronous time-series dataset.

[0085] Furthermore, the multi-sensor fusion welding wire storage environment real-time monitoring system is also used for: defining a time reference axis, mapping the multiple environmental state raw data to the time reference axis, determining multiple target time points, and having a corresponding relationship between the multiple target time points and the multiple environmental state raw data; traversing the multiple target time points and timestamps in sequence for matching and searching: S1: extracting a first target time point based on the target time point matching the timestamp, and determining the first environmental state raw data based on the first target time point; S2: when the multiple target time points do not match the timestamps, calculating the distance between the timestamps and the multiple target time points, generating an adjacent distance parameter set, extracting the minimum distance value based on the adjacent distance parameter set, determining adjacent time points, and determining the second environmental state raw data based on the adjacent time points; S3: when the adjacent distance parameter set is empty, estimating based on the multiple environmental state raw data to generate the third environmental state raw data; S4: sorting the first environmental state raw data, the second environmental state raw data, and the third environmental state raw data according to the time reference axis to generate the synchronous time series dataset.

[0086] Furthermore, the multi-sensor fusion-based real-time monitoring system for welding wire storage environment is also used for: performing multi-dimensional feature analysis on the synchronous time-series dataset to generate an original feature set; performing min-max normalization on the original feature set to generate feature-normalized data; performing correlation calculation on the original feature set based on the feature-normalized data to generate multiple correlation coefficients; traversing the original feature set according to the multiple correlation coefficients to perform correlation filtering processing to generate multiple feature parameters; arranging the multiple feature parameters in chronological order to construct a core feature vector; constructing a welding wire storage prior knowledge graph to assign weights to the core feature vector and determine multiple initial weight coefficients; multiplying the multiple initial weight coefficients with the core feature vector to obtain multiple feature weighted scores; and pre-setting an evaluation value range, mapping the multiple feature weighted scores to the evaluation value range to generate the comprehensive environmental status evaluation value.

[0087] Furthermore, the multi-sensor fusion-based real-time monitoring system for the welding wire storage environment is also used for: performing serialized fluctuation analysis based on the comprehensive environmental state assessment value to determine the fluctuation pattern; identifying abnormal fluctuations in the comprehensive environmental state assessment value according to the fluctuation pattern, and locking environmental interference parameters based on the identification results, wherein the environmental interference parameters include environmental interference event types and key interference parameters; associating interference features based on the environmental interference event types and the key interference parameters to determine the environmental interference event set; constructing an environmental state time series curve based on the comprehensive environmental state assessment value, mapping the environmental interference event set to the environmental state time series curve for fluctuation abrupt change identification and division, and generating multiple feature abrupt change data segments; and performing availability prediction based on the environmental interference event types, the key interference parameters, the multiple feature abrupt change data segments, and the comprehensive environmental state assessment value to generate the predicted remaining availability information.

[0088] Furthermore, the multi-sensor fusion-based real-time monitoring system for the welding wire storage environment is also used for: segmenting and matching multiple characteristic abrupt change data segments based on the type of environmental interference event and the key interference parameters to obtain multiple types of characteristic abrupt change events, including storage and retrieval operation interference events, desiccant effectiveness decay events, and thermal environment disturbance events; performing humidity intrusion analysis based on the storage and retrieval operation interference events to generate additional wet load parameters; performing adsorption decay calculation based on the desiccant effectiveness decay events to generate adsorption capacity decay rate; performing joint analysis on the additional wet load parameters and the adsorption capacity decay rate according to the thermal environment disturbance events to generate joint analysis results; predicting the availability of the welding wire based on the joint analysis results to generate predicted desiccant remaining effective life information and welding wire remaining safe storage time information; issuing a life warning for the welding wire based on the predicted desiccant remaining effective life information and welding wire remaining safe storage time, generating a warning command, and adding the warning command to the predicted remaining availability information.

[0089] Furthermore, the multi-sensor fusion-based real-time monitoring system for welding wire storage environment is also used for: mapping and converting comprehensive environmental status assessment values ​​into equivalent average humidity data; introducing welding wire material information to perform moisture absorption and deterioration calculations, setting a critical humidity threshold for deterioration, performing time-series cumulative analysis, and constructing a time-humidity cumulative effect relationship; calculating the critical cumulative moisture absorption time of the welding wire based on the equivalent average humidity data according to the time-humidity cumulative effect relationship; using the critical cumulative moisture absorption time of the welding wire as a basic safe time to perform over-limit calculations on multiple characteristic sudden change data segments to generate an over-limit cumulative duration; performing a moisture damage simulation on the welding wire based on the over-limit cumulative duration weighted and superimposed on the basic safe time to generate simulated moisture damage cumulative data; and subtracting the basic safe time based on the simulated moisture damage cumulative data to generate the remaining safe storage time information for the welding wire.

[0090] Furthermore, the multi-sensor fusion-based real-time monitoring system for welding wire storage environment is also used for: performing lifespan criticality calculations based on the predicted remaining effective lifespan information of the desiccant and setting a lifespan warning threshold; performing lifespan behavior impact analysis based on the predicted remaining effective lifespan information of the desiccant and setting a lifespan behavior threshold; generating a first warning instruction when the predicted remaining effective lifespan information of the desiccant is lower than the lifespan behavior threshold, and generating a second warning instruction when the predicted remaining effective lifespan information of the desiccant is lower than the lifespan warning threshold and higher than the lifespan behavior threshold; performing time criticality calculations based on the basic safety time and setting a safety time warning threshold; performing time behavior impact analysis based on the basic safety time and setting a safety time action threshold; generating a third warning instruction when the remaining safe storage time of the welding wire is lower than the safety time action threshold, and generating a fourth warning instruction when the remaining safe storage time of the welding wire is lower than the safety time action threshold and higher than the safety time action threshold.

[0091] Furthermore, the multi-sensor fusion-based real-time monitoring system for the welding wire storage environment is also used for: traversing the first, second, third, and fourth early warning commands for priority analysis to construct multiple command priorities; constructing a control strategy library based on historical control logs, matching the multiple command priorities with the first, second, third, and fourth early warning commands, mapping the matching results to the control strategy library for retrieval, and generating environmental control operation commands, wherein the environmental control operation commands correspond to the first, second, third, and fourth early warning commands; issuing the environmental control operation commands to the storage box to drive the actuator to perform environmental control and generate environmental control results; feeding back the environmental control results to the early warning command to activate the enhanced monitoring mode for high-frequency data acquisition, updating the early warning command to construct a monitoring optimization strategy for real-time closed-loop monitoring of the storage box.

[0092] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The multi-sensor fusion welding wire storage environment real-time monitoring method and specific examples in the aforementioned embodiment one are also applicable to the multi-sensor fusion welding wire storage environment real-time monitoring system of this embodiment. Through the foregoing detailed description of the multi-sensor fusion welding wire storage environment real-time monitoring method, those skilled in the art can clearly understand the multi-sensor fusion welding wire storage environment real-time monitoring system of this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0093] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily 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 this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0094] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for real-time monitoring of welding wire storage environment using multi-sensor fusion, characterized in that, The method includes: The environmental status data of the storage box is collected in real time by a multi-source sensor array and time-series aligned to generate a synchronous time-series dataset. Based on the synchronous time-series dataset, feature analysis is performed to extract multiple feature vectors for multi-source data fusion, generating a comprehensive environmental state assessment value. The method includes: Perform multi-dimensional feature analysis on the synchronous time-series dataset to generate an original feature set; Based on the original feature set, minimum-maximum normalization is performed to generate feature-normalized data; Based on the feature normalization data, the correlation of the original feature set is calculated to generate multiple correlation coefficients; Based on the multiple correlation coefficients, the original feature set is traversed to perform correlation filtering processing, generating multiple feature parameters; The multiple feature parameters are arranged in chronological order to construct a core feature vector; A prior knowledge graph of welding wire storage is constructed to assign weights to the core feature vectors and determine multiple initial weight coefficients. The multiple initial weight coefficients are multiplied by the core feature vector to obtain multiple feature weighted scores; A preset evaluation value range is used to map the weighted scores of the multiple features to the evaluation value range, thereby generating the comprehensive environmental status evaluation value. Based on the comprehensive environmental status assessment value, interference analysis is performed to generate environmental interference parameters. By locking the environmental interference parameters and extracting sudden change data, combined with the comprehensive environmental status assessment value, the availability of welding wire storage is predicted, and the predicted remaining availability information is generated. The predicted remaining availability information includes early warning instructions. The storage box is intelligently environmentally controlled according to the warning command, and the environmental control results are fed back to the warning command to perform real-time closed-loop monitoring of the storage box.

2. The multi-sensor fusion method for real-time monitoring of welding wire storage environment as described in claim 1, characterized in that, The environmental status data of the storage box is collected in real time by a multi-source sensor array and time-series aligned to generate a synchronized time-series dataset. The method includes: The storage boxes are traversed to analyze the storage environment, and multiple environmental monitoring points are located based on the analysis results. Based on the multiple environmental detection points, sensor type analysis is performed to determine multiple sensor types. Multiple sensors are matched to the multiple sensing types and deployed to multiple environmental detection points to construct a multi-source sensor array; Activate the multi-source sensor array to collect real-time data from multiple environmental monitoring points, obtaining raw environmental status data from multiple locations: S1: Any sensor in the multi-source sensor array collects data in real time at any environmental point to be detected, generating a raw data packet; S2: The main control unit matches the original data packet according to the real-time clock module to generate a first timestamp; S3: Bind the first timestamp to the original data packet, and iterate until the data collection of multiple environmental detection points is completed, and obtain multiple environmental state original data; The original environmental state data are sorted according to timestamp order to generate the synchronous time-series dataset.

3. The multi-sensor fusion method for real-time monitoring of welding wire storage environment as described in claim 2, characterized in that, The method for sorting the multiple raw environmental state data according to timestamp order to generate the synchronized time-series dataset includes: Define a time reference axis, map the multiple environmental state raw data to the time reference axis, determine multiple target time points, and there is a correspondence between the multiple target time points and the multiple environmental state raw data; The search is performed by iterating through the multiple target time points and timestamps in sequence. S1: Extract the first target time point based on the target time point matched with the timestamp, and determine the original data of the first environmental state based on the first target time point; S2: When the multiple target time points do not match the timestamp, calculate the distance between the timestamp and the multiple target time points, generate an adjacent distance parameter set, extract the minimum distance value based on the adjacent distance parameter set, determine the adjacent time points, and determine the original data of the second environmental state based on the adjacent time points. S3: When the adjacent distance parameter set is empty, the third environmental state raw data is generated by estimating based on the multiple environmental state raw data. S4: Sort the original data of the first environment state, the original data of the second environment state, and the original data of the third environment state according to the time reference axis to generate the synchronous time series dataset.

4. The multi-sensor fusion method for real-time monitoring of welding wire storage environment as described in claim 1, characterized in that, Based on the comprehensive environmental condition assessment value, interference analysis is performed to generate environmental interference parameters. By locking the environmental interference parameters and extracting sudden change data, combined with the comprehensive environmental condition assessment value, the availability of welding wire storage is predicted to generate predicted remaining availability information. The method includes: Based on the comprehensive environmental state assessment values, serialized fluctuation analysis is performed to determine the fluctuation patterns. According to the fluctuation pattern, abnormal fluctuations are identified in the comprehensive environmental state assessment value. Based on the identification results, environmental interference parameters are locked. The environmental interference parameters include environmental interference event types and key interference parameters. Based on the environmental interference event types and the key interference parameters, interference features are correlated to determine the environmental interference event set. Based on the comprehensive environmental state assessment value, an environmental state time series curve is constructed. The environmental disturbance event set is mapped to the environmental state time series curve for fluctuation and sudden change identification and division, generating multiple characteristic sudden change data segments. Based on the environmental interference event type, the key interference parameters, the multiple sudden change data segments, and the comprehensive environmental state assessment value, availability prediction is performed to generate the predicted remaining availability information.

5. The multi-sensor fusion method for real-time monitoring of welding wire storage environment as described in claim 4, characterized in that, Based on the environmental disturbance event type, the key disturbance parameters, the multiple sudden change data segments, and the comprehensive environmental state assessment value, availability prediction is performed to generate the predicted remaining availability information. The method includes: Based on the environmental interference event type and the key interference parameters, the multiple feature sudden change data segments are segmented and matched to obtain multiple types of feature sudden change events, including access operation interference events, desiccant efficiency decay events, and thermal environment disturbance events. Based on the access operation interference events, humidity intrusion analysis is performed to generate additional humidity load parameters; Based on the desiccant efficiency decay event, adsorption decay calculation is performed to generate the adsorption capacity decay rate. A joint analysis of the additional wet load parameters and the adsorption capacity decay rate is performed based on the thermal environment disturbance event to generate joint analysis results. Based on the joint analysis results, the availability of the welding wire is predicted, and information on the remaining effective life of the desiccant and the remaining safe storage time of the welding wire is generated. Based on the predicted remaining effective lifespan of the desiccant and the remaining safe storage time of the welding wire, a lifespan warning is issued for the welding wire, a warning instruction is generated, and the warning instruction is added to the predicted remaining available information.

6. The multi-sensor fusion method for real-time monitoring of welding wire storage environment as described in claim 5, characterized in that, Based on the joint analysis results, the availability of the welding wire is predicted, generating information on the remaining effective life of the desiccant and the remaining safe storage time of the welding wire. The method includes: The comprehensive environmental condition assessment values ​​are mapped and converted into equivalent average humidity data; We incorporate welding wire material information to calculate moisture absorption and deterioration, set a critical humidity threshold for deterioration, conduct time-series cumulative analysis, and construct a time-humidity cumulative effect relationship. Based on the equivalent average humidity data, the critical cumulative moisture absorption time of the welding wire is calculated according to the time-humidity cumulative effect relationship. The critical cumulative moisture absorption time of the welding wire is used as the basic safety time to perform over-limit calculations on multiple characteristic abrupt change data segments, generating the over-limit cumulative duration. Based on the cumulative duration of exceeding the limit, the welding wire is subjected to wet damage simulation by time-weighted summation to the basic safe time, generating simulated wet damage cumulative data. Based on the simulated cumulative wet damage data, the basic safe time is reduced to generate the remaining safe storage time information for the welding wire.

7. The multi-sensor fusion method for real-time monitoring of welding wire storage environment as described in claim 6, characterized in that, Based on the predicted remaining effective lifespan of the desiccant and the remaining safe storage time of the welding wire, a lifespan warning is issued for the welding wire, and a warning command is generated. The method includes: Based on the predicted remaining effective lifespan information of the desiccant, a lifespan criticality calculation is performed, and a lifespan warning threshold is set. Based on the predicted remaining effective lifespan information of the desiccant, an impact analysis on lifespan behavior is performed, and a lifespan behavior threshold is set. When the predicted remaining effective lifespan of the desiccant is lower than the lifespan behavior threshold, a first warning instruction is generated; when the predicted remaining effective lifespan of the desiccant is lower than the lifespan warning threshold and the predicted remaining effective lifespan of the desiccant is higher than the lifespan behavior threshold, a second warning instruction is generated. Based on the aforementioned basic safety time, time criticality calculations are performed, and a safety time early warning threshold is set. Based on the aforementioned basic safety time, a time behavior impact analysis is performed, and a safety time action threshold is set. When the remaining safe storage time of the welding wire is lower than the safe time action threshold, a third warning instruction is generated; when the remaining safe storage time of the welding wire is lower than the safe time action threshold and the remaining safe storage time of the welding wire is higher than the safe time action threshold, a fourth warning instruction is generated.

8. The multi-sensor fusion method for real-time monitoring of welding wire storage environment as described in claim 7, characterized in that, The method includes: intelligent environmental control of the storage box based on the warning command, generating environmental control results and feeding them back to the warning command for real-time closed-loop monitoring of the storage box; The first warning command, the second warning command, the third warning command, and the fourth warning command are traversed for priority analysis to construct multiple command priorities; A control strategy library is constructed based on historical control records. The multiple instructions are matched with the first warning instruction, the second warning instruction, the third warning instruction, and the fourth warning instruction according to their priority. The matching results are mapped to the control strategy library for retrieval, and environmental control operation instructions are generated. The environmental control operation instructions correspond to the first warning instruction, the second warning instruction, the third warning instruction, and the fourth warning instruction. The environmental control operation command is sent to the storage box to drive the actuator to perform environmental control and generate environmental control results. The environmental control results are fed back to the early warning command to activate the enhanced monitoring mode for high-frequency data collection, and the early warning command is updated to construct a monitoring optimization strategy for real-time closed-loop monitoring of the storage box.

9. A real-time monitoring system for welding wire storage environment based on multi-sensor fusion, characterized in that, The steps for implementing the real-time monitoring method for the welding wire storage environment using multi-sensor fusion as described in any one of claims 1 to 8 include: The synchronous time series dataset generation module is used to collect environmental status data of the storage box in real time through a multi-source sensor array, perform time series alignment, and generate a synchronous time series dataset. The comprehensive environmental status assessment value generation module is used to perform feature analysis based on the synchronous time-series dataset, extract multiple feature vectors, perform multi-source data fusion, and generate a comprehensive environmental status assessment value. The module for predicting remaining available information is used to perform interference analysis based on the comprehensive environmental state assessment value, generate environmental interference parameters, extract sudden change data by locking the environmental interference parameters and combine them with the comprehensive environmental state assessment value to predict the availability of welding wire storage, and generate predicted remaining available information, which includes early warning instructions. The environmental control result generation module is used to perform intelligent environmental control on the storage box according to the early warning command, generate environmental control results and feed them back to the early warning command for real-time closed-loop monitoring of the storage box.

Images