Cold storage spraying polyurethane real-time monitoring system based on temperature and humidity feedback

Abstract

The invention relates to the technical field of cold storage spraying, and discloses a cold storage spraying polyurethane real-time monitoring system based on temperature and humidity feedback, and the system comprises an environment collection module, a state analysis module, a parameter regulation and control module, a dynamic fusion module, a mode matching module and a feedback execution module. The environment acquisition module acquires temperature and humidity data and sets a monitoring interval; the state analysis module divides coating detection nodes and generates feature vectors; the parameter regulation and control module is used for separating influence factors, establishing control rules and acquiring process parameters; the dynamic fusion module identifies the equipment state, performs dynamic compensation and calculates the parameter difference degree; the mode matching module deduces an optimal curing threshold value and generates a deviation sequence; and the feedback execution module analyzes the deviation sequence and generates an adjustment instruction. The system realizes real-time monitoring and accurate regulation and control of the process of spraying polyurethane on the refrigeration house, improves the quality stability of the coating and the construction efficiency, is suitable for refrigeration house construction and maintenance scenes, and has the advantages of high environmental adaptability, accurate parameter regulation and control and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of cold storage spraying, and specifically to a real-time monitoring system for polyurethane spraying in cold storage based on temperature and humidity feedback. Background Art

[0002] During the construction and maintenance of cold storage, polyurethane spraying is a key process, and the quality of its coating directly affects the heat preservation performance, service life and energy consumption level of the cold storage. Traditional cold storage spraying operations mainly rely on the experience of operators to adjust process parameters, lacking real-time dynamic monitoring of the spraying environment (such as temperature and humidity) and the coating curing process, resulting in the following prominent problems: Temperature and humidity are the core environmental factors affecting polyurethane curing. Their fluctuations will directly change the chemical reaction rate, coating thickness uniformity and adhesion of polyurethane. For example, a low-temperature environment may prolong the curing time, resulting in wrinkles or cracks on the coating surface; a high-humidity environment may cause incomplete foaming of polyurethane, forming bubbles or voids, reducing the heat preservation performance. However, in traditional processes, there is a lack of accurate acquisition and real-time feedback mechanism for temperature and humidity parameters, and it is impossible to adjust spraying parameters in a timely manner according to environmental changes, resulting in poor stability of coating quality.

[0003] The polyurethane curing process is a complex physical and chemical process with dynamic changes, involving the coordinated action of multi-dimensional parameters such as atomization pressure, spraying speed, and curing parameters (such as temperature sensitivity, humidity permeability). Traditional monitoring methods can only achieve discrete detection of a single parameter, and it is impossible to systematically extract and analyze the curing parameters of each coating detection node, making it difficult to construct a complete coating feature vector, resulting in the inability to accurately identify the root cause of coating state deviation, and the lack of a scientific basis for process adjustment.

[0004] The operating state of spraying equipment (such as spraying interval period, atomization particle size dispersion) is closely related to the coating quality. Traditional systems cannot effectively identify the dynamic relationship between the operating state of the equipment and the temperature and humidity influence factors, and it is difficult to perform real-time dynamic compensation for the temperature influence factors and humidity response factors, resulting in an increase in the difference degree of coating parameters when the equipment state fluctuates, and it is impossible to ensure the coating consistency under different spraying stages and environmental conditions.

[0005] In addition, there is a lack of in-depth analysis of the difference degree of coating parameters and a dynamic matching mechanism for the optimal curing threshold in the prior art. When the coating state deviates, it is impossible to quickly deduce the optimal curing threshold through a scientific pattern matching algorithm, and then generate accurate process adjustment instructions, resulting in a lag in spraying process adjustment and affecting the construction efficiency and the fine level of coating quality control. Summary of the Invention

[0006] The object of the present invention is to provide a real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: A real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback, the system includes: An environmental acquisition module, which is used to obtain the temperature and humidity data of the cold storage spraying operation area and set an environmental monitoring interval corresponding to the spraying stage; A state analysis module, which is used to divide multiple coating detection nodes within the environmental monitoring interval, extract the characteristics of the polyurethane curing parameters for each node, and generate a coating feature vector corresponding to the node; A parameter regulation module, which is used to separate the temperature influence factor and the humidity response factor from the coating feature vector, establish a curing control rule associated with the node, and obtain the spraying process parameters corresponding to the rule; A dynamic fusion module, which is used to identify the equipment operation state in the spraying process parameters, dynamically compensate the temperature influence factor and the humidity response factor according to the operation state, and calculate the coating parameter difference degree of each node under different compensation strategies; A mode matching module, which is used to deduce the optimal curing threshold according to the coating parameter difference degree, and generate a coating state deviation sequence by matching the current coating parameters with the optimal curing threshold; A feedback execution module, which is used to analyze the coating state deviation sequence, and based on the parameter difference distribution of the node, convert the coating state deviation sequence into a spraying process adjustment instruction.

[0008] Preferably, the implementation method of the state analysis module includes: constructing a spraying feature library corresponding to the coating detection node, and the spraying feature library includes the coating parameter vector mapped by the temperature and humidity data and the curing parameters; Perform similar spraying stage matching on the coating parameter vector, and divide the spraying clustering group of the coating parameter vector according to the matching result; extract the distribution center point of the curing parameters from the spraying clustering group, and set the distribution center point as the coating detection node.

[0009] Preferably, dividing the spraying clustering group of the coating parameter vector further includes: According to the spraying thickness and environmental parameters in the coating parameter vector, extract the atomization pressure, spraying speed and temperature and humidity change gradient, and generate a spraying feature label based on the above parameters; Associate the spraying feature label with the coating parameter vector, and screen the coating parameter vectors with a spraying similarity higher than the preset spraying threshold to form a spraying clustering group by calculating the spraying similarity between the feature labels.

[0010] Preferably, the implementation method of generating the coating feature vector corresponding to the node includes: For each coating detection node, according to the spatial position of the node in the environmental monitoring interval, obtain the curing fluctuation data of the node within a preset time window, and calculate the curing fluctuation coefficient of the node; When the curing fluctuation coefficient exceeds the first curing threshold, mark the node as an abnormal curing node, and extract its curing parameters to form a coating feature vector; when the curing fluctuation coefficient is lower than the first curing threshold, mark the node as a stable node, and perform mean fusion on the curing parameters of the adjacent nodes of the node, and reconstruct the fused data into a coating feature vector.

[0011] Preferably, the implementation manner of the parameter regulation module includes: Separate the temperature sensitivity, humidity permeability, and atomization uniformity from the coating feature vector, and generate a curing control rule for the coating detection node based on the above parameters; If the number of detection nodes covered by the current curing control rule is less than the preset coverage threshold, traverse the coating feature vectors of the adjacent nodes, and add the curing indicators not included in the control rules of the adjacent nodes to the current rule.

[0012] Preferably, the implementation manner of the dynamic fusion module includes: obtaining the cycle factor of the spraying interval and the discrete factor of the atomization particle size in the equipment operation state; Construct a parameter compensation network associated with the cycle factor and the discrete factor, and determine the coating parameter difference degree under different compensation strategies according to the compensation weights of each node in the network.

[0013] Preferably, constructing the parameter compensation network further includes: Identify the time distribution characteristics of the cycle factor. If the current time distribution characteristics exactly match the preset spraying time sequence, set the cycle factor as the initial node of the parameter compensation network; Calculate the compensation correlation degree between the cycle factor and the discrete factor, and generate the intermediate nodes and end nodes of the parameter compensation network in descending order of the correlation degree; Perform parameter reverse verification on the end nodes. When the correlation degree of the end nodes is lower than the preset verification threshold, output them as the final nodes of the parameter compensation network.

[0014] Preferably, the implementation manner of calculating the coating parameter difference degree includes: Statistically analyze the cycle factor variance and discrete factor standard deviation of each end node in the parameter compensation network, and calculate the global range of all node parameters; Subtract the cycle factor variance of a single end node from the cycle factor variance of the adjacent node, divide by the global range to obtain the cycle difference coefficient; at the same time, calculate the ratio of the discrete factor standard deviation to the global range, and sum the two weighted as the coating parameter difference degree of the node.

[0015] Preferably, the implementation manner of deriving the optimal curing threshold includes: Extract the curing mode in the historical data that is closest to the current coating parameter difference degree, calculate the Euclidean distance in the spatial distribution between the two, and use it as the first matching reference value; Statistically analyze the difference in peak point density between the current coating parameter difference degree and the historical curing mode, and use the density difference value as the second matching reference value; Based on the non-linear combination of the first matching reference value and the second matching reference value, match the best curing threshold in the preset curing threshold table.

[0016] Preferably, the implementation method of the feedback execution module includes: dividing the positive compensation interval and the negative compensation interval according to the parameter difference direction of each node in the coating state deviation sequence; Extract the adjustment amplitude of the parameter deviation in the positive compensation interval and the correction frequency of the parameter deviation in the negative compensation interval, and superimpose and combine the two according to the spatial weight of the coating detection node to generate the execution parameters of the spraying process adjustment instruction.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: The environmental acquisition module realizes the accurate acquisition and dynamic tracking of the temperature and humidity data in the cold storage spraying operation area by setting the environmental monitoring interval corresponding to the spraying stage, provides a real-time and reliable environmental parameter basis for subsequent process adjustment, and significantly improves the sensitivity and response speed of the system to environmental changes. The state analysis module realizes the systematic feature extraction of the curing parameters of each coating detection node by constructing a spraying feature library, dividing spraying clustering groups and extracting the center points of the curing parameter distribution. The generated coating feature vector can comprehensively and accurately reflect the coating state, provides a scientific analysis basis for parameter regulation, and avoids the blindness of traditional empirical adjustment.

[0018] The parameter regulation module separates the temperature influence factor and the humidity response factor from the coating feature vector, establishes a curing control rule associated with the node, and ensures the pertinence and comprehensiveness of the process parameter regulation by dynamically expanding the rule coverage range. This module can generate personalized spraying process parameters according to the real-time environment and coating state, realizes the refined control of the polyurethane curing process, and effectively improves the stability of the coating quality.

[0019] The dynamic fusion module realizes the dynamic collaborative compensation of the equipment state and the temperature and humidity factors by identifying the periodic factor and the discrete factor in the equipment operation state, constructing a parameter compensation network and calculating the coating parameter difference degree. This mechanism can offset the influence of equipment state fluctuations and environmental changes on the coating quality in real time, ensure that the coating parameters always remain within the optimal range in different spraying stages and equipment working conditions, and significantly improve the adaptability and robustness of the system.

[0020] The pattern matching module derives the optimal curing threshold based on the difference degree of coating parameters, and generates a state deviation sequence by matching the current coating parameters, providing a clear adjustment target for feedback execution. This module uses historical data and intelligent algorithms to achieve rapid matching of the optimal curing threshold, avoiding the inefficiency of traditional trial-and-error methods and improving the timeliness and accuracy of process adjustment.

[0021] The feedback execution module converts the coating state deviation sequence into specific spraying process adjustment instructions. By dividing the compensation interval, superimposing the adjustment amplitude and correction frequency, it achieves precise fine-tuning of process parameters. This module ensures a high degree of fit between the adjustment instructions and the coating state deviation, can quickly and effectively correct coating defects, and improves the construction efficiency and the refined level of coating quality control. Description of the Drawings

[0022] Figure 1 It is the working principle diagram of the real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback described in the present invention; Figure 2 It is the flow chart for the state analysis module to construct a spraying feature library and divide coating detection nodes; Figure 3 It is the flow chart for dividing the clustering group of coating parameter vectors based on spraying feature tags; Figure 4 It is the flow chart for analyzing the curing fluctuation coefficient of coating detection nodes and generating feature vectors. Detailed Embodiment

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0024] Please refer to Figures 1-4 , the real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback involved in the present invention, the system includes: an environment acquisition module, a state analysis module, a parameter regulation module, a dynamic fusion module, a pattern matching module and a feedback execution module. The specific implementation steps are as follows: First, the environmental acquisition module obtains the temperature and humidity data of the cold storage spraying operation area and sets an environmental monitoring interval corresponding to the spraying stage. Next, the state analysis module divides multiple coating detection nodes within the environmental monitoring interval, extracts the polyurethane curing parameters of each node, and generates a coating feature vector corresponding to the node. Then, the parameter regulation module separates the temperature influence factor and the humidity response factor from the coating feature vector, establishes a curing control rule associated with the node, and obtains the spraying process parameters corresponding to the rule. The dynamic fusion module identifies the equipment operation status in the spraying process parameters, dynamically compensates the temperature influence factor and the humidity response factor according to the operation status, and calculates the coating parameter difference degree of each node under different compensation strategies. Then, the pattern matching module derives the optimal curing threshold based on the coating parameter difference degree, and generates a coating status deviation sequence by matching the current coating parameters with the optimal curing threshold. Finally, the feedback execution module analyzes the coating status deviation sequence, and based on the parameter difference distribution of the nodes, converts the coating status deviation sequence into a spraying process adjustment instruction.

[0025] The technical solutions of the present invention will be further described in detail below in conjunction with specific embodiments.

[0026] Embodiment 1

[0027] The state analysis module of the system constructs a spraying feature library corresponding to the coating detection nodes. The spraying feature library contains the coating parameter vectors mapped by temperature and humidity data and curing parameters. Perform similar spraying stage matching on the coating parameter vectors, and divide the spraying clustering groups of the coating parameter vectors according to the matching results. Specifically, according to the spraying thickness and environmental parameters in the coating parameter vector, extract the atomization pressure, spraying speed, and temperature and humidity change gradient, and generate a spraying feature label based on the above parameters. Associate the spraying feature label with the coating parameter vector, and by calculating the spraying similarity between the feature labels, screen out the coating parameter vectors with a similarity higher than the preset spraying threshold to form a spraying clustering group. Extract the distribution center point of the curing parameters from the spraying clustering group, and set the distribution center point as the coating detection node.

[0028] For each coating detection node, according to the spatial position of the node in the environmental monitoring interval, obtain the curing fluctuation data of the node within the preset time window, and calculate the curing fluctuation coefficient of the node. When the curing fluctuation coefficient exceeds the first curing threshold, mark the node as an abnormal curing node, and extract its curing parameters to form a coating feature vector; when the curing fluctuation coefficient is lower than the first curing threshold, mark the node as a stable node, and perform mean fusion on the curing parameters of the adjacent nodes of the node, and reconstruct the fused data into a coating feature vector.

[0029] When building the spray coating feature library, the system first preprocesses the collected temperature and humidity data and curing parameters, removes the noise data and performs standardization processing. Then, the processed data is mapped to a high-dimensional feature space to form a coating parameter vector. Each coating parameter vector contains features in multiple dimensions, such as the temperature change rate, humidity change rate, coating thickness change rate, etc. By analyzing these features, the curing state of the polyurethane coating under different environmental conditions can be accurately grasped.

[0030] During the matching process of similar spray coating stages, the system will compare the current coating parameter vector with the spray coating stages in the historical data. By calculating the similarity between vectors, the most similar spray coating stage is found. For example, if the similarity between the current coating parameter vector and the vector of a certain spray coating stage in the historical data exceeds 80%, it is considered that they are in similar spray coating stages. This matching method can help the system quickly locate the stage where the current spray coating process is located, providing a basis for subsequent analysis and control.

[0031] When dividing the spray coating clustering groups, the system will perform clustering based on the similarity between spray coating feature labels. The calculation of similarity is based on parameters such as atomization pressure, spray speed, and temperature and humidity change gradients. For example, if the difference in atomization pressure between two coating parameter vectors is less than 5%, the difference in spray speed is less than 10%, and the difference in temperature and humidity change gradients is less than 15%, it is considered that their spray coating feature label similarity is high and they can be divided into the same clustering group. This clustering method can group the coating parameter vectors under similar spray coating conditions into one category, facilitating subsequent analysis and processing.

[0032] When extracting the distribution center point of the curing parameters, the system will perform statistical analysis on the coating parameter vectors in each spray coating clustering group. By calculating statistics such as the mean and median, the distribution center point of the curing parameters is determined. For example, for all the coating parameter vectors in a certain clustering group, calculate the mean values of their temperature sensitivity, humidity permeability, and atomization uniformity, and use these mean values as the distribution center point of this clustering group. The distribution center point represents the typical characteristics of the curing parameters in this clustering group and can be used as a reference standard for the coating detection node.

[0033] When calculating the curing fluctuation coefficient of a node, the system will analyze the change of the curing parameters of the node within a preset time window. The curing fluctuation coefficient reflects the stability of the curing process. The larger the coefficient, the more unstable the curing process. For example, if the temperature sensitivity of a certain node fluctuates greatly and the humidity permeability changes frequently within a period of time, its curing fluctuation coefficient is high. By setting the first curing threshold, abnormal curing nodes can be detected in a timely manner so as to take corresponding measures for adjustment.

[0034] For abnormal curing nodes, the system extracts their curing parameters to form a coating feature vector. The coating feature vector contains key information about the node during the curing process, such as temperature sensitivity, humidity permeability, etc. This information can help the system deeply analyze the reasons for abnormal curing and provide a basis for subsequent parameter regulation. For stable nodes, the system performs mean fusion on the curing parameters of their adjacent nodes. This fusion method can make full use of the information of adjacent nodes and improve the accuracy and reliability of the coating feature vector. For example, if the adjacent nodes of a stable node have similar characteristics in terms of temperature sensitivity and humidity permeability, then after fusing these characteristics by taking the mean, the curing state of this node can be more accurately reflected.

[0035] The state analysis module of the system can accurately capture the changes in the curing state during the spraying of polyurethane in cold storage through steps such as constructing a spraying feature library, dividing spraying clustering groups, setting coating detection nodes, and generating coating feature vectors. This analysis method not only considers the influence of environmental factors such as temperature and humidity, but also combines spraying process parameters, providing a solid foundation for subsequent parameter regulation and feedback control. By promptly detecting and handling abnormal curing nodes and reasonably optimizing stable nodes, the system can effectively improve the quality and efficiency of spraying polyurethane in cold storage and ensure that the curing effect of the coating meets the expected requirements.

[0036] Example 2

[0037] The working mode of the parameter regulation module is as follows: separate temperature sensitivity, humidity permeability, and atomization uniformity from the coating feature vector, and generate a curing control rule for the coating detection node based on the above parameters. If the number of detection nodes covered by the current curing control rule is less than the preset coverage threshold, then traverse the coating feature vectors of adjacent nodes and add the curing indicators not included in the control rules of adjacent nodes to the current rule.

[0038] When separating temperature sensitivity, humidity permeability, and atomization uniformity from the coating feature vector, the system first performs a dimensionality analysis on the coating feature vector to identify the feature dimensions related to temperature, humidity, and atomization effect. Then, through methods such as principal component analysis, these feature dimensions are reduced in dimension, and the main components that can represent temperature sensitivity, humidity permeability, and atomization uniformity are extracted. For example, temperature sensitivity may be related to features such as the temperature change rate on the coating surface and the temperature distribution gradient; humidity permeability may be related to features such as the humidity diffusion rate inside the coating and the humidity gradient change; atomization uniformity may be related to features such as the size distribution and density distribution of spraying particles.

[0039] Based on the extracted parameters such as temperature sensitivity, humidity permeability, and atomization uniformity, the system generates curing control rules for the coating detection nodes. These rules define the parameter ranges and variation trends that each coating detection node should meet in different environmental conditions to achieve the best curing effect. For example, for nodes with high temperature sensitivity, the curing control rules may stipulate that the ambient temperature should be maintained at 25±2°C within the first 30 minutes after spraying to ensure the normal curing of the polyurethane coating.

[0040] When the system determines that the number of detection nodes covered by the current curing control rules is less than the preset coverage threshold, it indicates that the current rules may not be comprehensive enough and need to be further improved. At this time, the system will traverse the coating feature vectors of adjacent nodes to check whether there are curing indicators not covered by the current rules in the control rules of adjacent nodes. If there are any uncovered indicators, the system will add these indicators to the current rules.

[0041] For example, the curing control rules of a certain node initially only considered two indicators, temperature sensitivity and atomization uniformity. However, when traversing the coating feature vectors of adjacent nodes, it is found that the control rules of adjacent nodes include the humidity permeability indicator, which is not covered by the current rules. At this time, the system will incorporate the humidity permeability into the current rules to make the rules more comprehensive.

[0042] When traversing the coating feature vectors of adjacent nodes, the system will determine the range of adjacent nodes based on the spatial position relationship and process parameter correlation between nodes. Generally, nodes that are close in distance and have similar spraying process parameters will be regarded as adjacent nodes. The system will analyze the coating feature vectors of these adjacent nodes one by one, extract the curing indicator information, and compare it with the current rules.

[0043] After adding the curing indicators not covered by the control rules of adjacent nodes to the current rules, the system will re-evaluate the coverage range and accuracy of the rules. If the newly added indicators can significantly improve the coverage of the rules for detection nodes, it indicates that the rules have been effectively improved; otherwise, if the coverage effect of the rules does not improve significantly after adding the new indicators, the system may further screen or adjust the indicators.

[0044] During the process of improving the curing control rules, the system will also consider the mutual relationship between various curing indicators. For example, there may be a certain coupling effect between temperature sensitivity and humidity permeability, and adjusting the temperature parameter may affect the humidity response. Therefore, the system will establish a parameter correlation model to analyze the interaction between various indicators to ensure that the newly added indicators can work in coordination with the original indicators to jointly optimize the curing control effect.

[0045] By dynamically improving the curing control rules, the parameter control module can continuously expand and optimize the rules based on the actual coating feature vector, better adapting them to the curing requirements of different nodes. This not only improves the system's monitoring and control accuracy of the cold storage polyurethane spraying process, but also effectively addresses various complex situations that may arise during the spraying process, ensuring the stability and consistency of coating quality.

[0046] Example 3

[0047] The dynamic fusion module obtains the periodic factor of the spraying interval and the discrete factor of the atomized particle size in the equipment operation state. A parameter compensation network associated with the periodic factor and the discrete factor is constructed. The specific steps are as follows: identify the time distribution characteristics of the periodic factor. If the current time distribution characteristics completely match the preset spray timing, the periodic factor is set as the initial node of the parameter compensation network. Calculate the compensation correlation between the periodic factor and the discrete factor, and generate the intermediate nodes and end nodes of the parameter compensation network in descending order of correlation. Perform a reverse parameter check on the end node. When the correlation of the end node is lower than the preset verification threshold, it is output as the final node of the parameter compensation network. According to the compensation weight of each node in the network, the difference in coating parameters under different compensation strategies is determined.

[0048] To determine the periodic and discrete factors associated with the equipment's operating status, the system first collects and analyzes the spray equipment's operating data in real time. The periodic factor of the spray interval reflects the temporal regularity of the spraying operation. Statistical analysis of historical spraying data reveals a standard spray timing pattern. The discrete factor of the atomized particle size describes the uniformity of the particle size distribution during the spraying process, a factor that significantly impacts coating uniformity and quality.

[0049] When constructing the parameter compensation network, the system first identifies the temporal distribution characteristics of the periodic factor. This process requires comparing the currently collected periodic factor data with the preset standard spray timing. If the two exactly match, the spraying operation is proceeding according to the predetermined time rhythm, and the periodic factor can be used as the starting point for the parameter compensation network. For example, if the preset spray timing is to spray once every 30 seconds, and the actual collected periodic factor data shows that the spraying interval is stable at around 30 seconds, the temporal distribution characteristics are considered to match.

[0050] Next, the system calculates the compensation correlation between the periodic factor and the discrete factor. This correlation reflects the mutual relationship between the two factors in affecting coating quality. For example, changes in the periodic factor may cause fluctuations in spray pressure, which in turn affects the distribution of atomized particle size. Therefore, there is a certain correlation between the two. By calculating this correlation, it is possible to determine how the two factors should be coordinated during the parameter compensation process to achieve the optimal compensation effect.

[0051] When generating the intermediate nodes and end nodes of the parameter compensation network, the system will arrange them in descending order of correlation degree. The higher the correlation degree of a factor combination, the higher the level it occupies in the network, and the greater its impact on the final compensation result. For example, if a certain group of periodic factors and discrete factors has a high correlation degree, it indicates that they have a significant impact on the coating parameters. The system will place them at a higher level in the network and assign a higher compensation weight.

[0052] Performing parameter reverse verification on the end nodes is an important step to ensure the accuracy of the compensation network. The system will start from the end nodes and reverse-derive the parameter values of each intermediate node and initial node to check whether the original input data can be restored. If the correlation degree of a certain end node is lower than the preset verification threshold, it indicates that the role of this node in the compensation process is not clear and may introduce errors. Therefore, it needs to be adjusted or removed.

[0053] When determining the coating parameter difference degree, the system will statistically analyze the variance of the periodic factors and the standard deviation of the discrete factors of each end node in the parameter compensation network, and calculate the global range of all node parameters. The variance of the periodic factors reflects the stability of the spraying interval. The larger the variance, the more unstable the spraying interval. The standard deviation of the discrete factors reflects the uniformity of the atomization particle size. The larger the standard deviation, the more uneven the particle size distribution. The global range represents the maximum difference range of all node parameters.

[0054] By subtracting the variance of the periodic factors of a single end node from the variance of the periodic factors of the adjacent node and dividing by the global range, the periodic difference coefficient can be obtained. This coefficient measures the difference degree of this node from the adjacent node in terms of the spraying cycle. At the same time, calculating the ratio of the standard deviation of the discrete factors to the global range can reflect the discrete degree of this node in terms of the atomization particle size. By weighted summing these two indicators, the coating parameter difference degree of this node can be obtained.

[0055] This calculation method comprehensively considers the factors of the spraying cycle and the atomization particle size, and can more comprehensively reflect the change of the coating parameters. By analyzing the coating parameter difference degrees under different compensation strategies, the system can select the optimal compensation strategy, thereby improving the quality and stability of spraying polyurethane in cold storage.

[0056] Example 4

[0057] The process by which the pattern matching module derives the optimal curing threshold is as follows: extract the curing pattern in the historical data that is closest to the current coating parameters, calculate the Euclidean distance in the spatial distribution between the two, and use it as the first matching reference value. Statistically analyze the difference in peak point density between the current coating parameter difference degree and the historical curing pattern, and use the density difference value as the second matching reference value. Based on the non-linear combination of the first matching reference value and the second matching reference value, match the optimal curing threshold in the preset curing threshold table.

[0058] When extracting the curing pattern in the historical data that is closest to the current coating parameters, the system first conducts a comprehensive search of the historical database. The historical database stores a large amount of spraying operation data under various conditions, including various temperature and humidity environments, spraying process parameters, and the corresponding curing effects. The system will compare the currently calculated coating parameter difference degree with the curing patterns in the historical data one by one to find the pattern with the highest similarity.

[0059] When calculating the Euclidean distance in the spatial distribution, the system regards the coating parameter difference degree and the historical curing pattern as points in a high-dimensional space. Each point is composed of parameters in multiple dimensions, such as temperature sensitivity, humidity permeability, atomization uniformity, etc. By calculating the Euclidean distance between these points, the similarity of their spatial distribution can be quantified. The smaller the Euclidean distance, the closer the positions of the two points in space, and the more similar the corresponding coating parameter difference degree and curing pattern.

[0060] When statistically analyzing the peak point density difference, the system will analyze the peak point distribution of the current coating parameter difference degree and the historical curing pattern in the parameter space. The peak point density reflects the concentration and intensity of parameter changes. If there is a large difference in peak point density between the two, it indicates that their parameter change patterns are different, and different curing thresholds may need to be adopted.

[0061] Based on the non-linear combination of the first matching reference value and the second matching reference value, the system will construct a complex matching function. This function is not a simple linear addition, but takes into account the interaction and influence between the two reference values. For example, when the first matching reference value is small, it indicates that the similarity in spatial distribution is high, and at this time, the weight of the second matching reference value may be relatively reduced; conversely, when the first matching reference value is large, the weight of the second matching reference value may be correspondingly increased.

[0062] When matching the preset curing threshold table, the system will, according to the matching result after non-linear combination, search for the closest threshold in the preset curing threshold table. The preset curing threshold table is summarized based on a large amount of experimental data and experience, and contains the optimal threshold ranges corresponding to different curing patterns. Through an accurate matching algorithm, the system can find the most suitable curing threshold for the current spraying operation from the table.

[0063] In this way, the pattern matching module can make full use of historical data, combine the spatial distribution of the current coating parameter difference degree and the peak point density characteristics, and accurately deduce the optimal curing threshold. This method not only considers the similarity of the spatial distribution of parameters, but also pays attention to the pattern and intensity of parameter changes, making the deduced threshold more in line with the actual spraying situation.

[0064] Then, by matching the current coating parameters with the optimal curing threshold, a coating state deviation sequence is generated. The system will compare each parameter of the current coating with the optimal curing threshold one by one, and calculate the deviation value of each parameter. These deviation values are arranged in a certain order to form a coating state deviation sequence. This deviation sequence can clearly reflect the difference between the current coating parameters and the optimal curing state, providing a clear adjustment direction and basis for the feedback execution module.

[0065] This implementation method of the pattern matching module, through in-depth mining and analysis of historical data and comprehensive consideration of multiple matching reference values, can provide accurate curing threshold guidance for the process of spraying polyurethane in cold storage. This helps to ensure that the spraying process can be carried out according to the expected curing requirements, improve the stability and consistency of the coating quality, and thus achieve precise control of the process of spraying polyurethane in cold storage.

[0066] Embodiment 5

[0067] The feedback execution module is implemented as follows: According to the parameter difference direction of each node in the coating state deviation sequence, the positive compensation interval and the negative compensation interval are divided. The adjustment amplitude of the parameter deviation in the positive compensation interval and the correction frequency of the parameter deviation in the negative compensation interval are extracted, and the two are superimposed and combined according to the spatial weight of the coating detection node to generate the execution parameters of the spraying process adjustment instruction.

[0068] When dividing the positive compensation interval and the negative compensation interval, the system first analyzes the parameter difference direction of each node in the coating state deviation sequence. If the parameter value of a certain node is higher than the optimal curing threshold, then this node is divided into the positive compensation interval; if the parameter value is lower than the optimal curing threshold, it is divided into the negative compensation interval. For example, when the temperature sensitivity parameter of the coating is higher than the optimal curing threshold, it indicates that the curing speed of this node may be too fast and negative adjustments such as cooling are required; conversely, if the temperature sensitivity parameter is lower than the threshold, positive adjustments such as heating are required.

[0069] When extracting the adjustment amplitude of parameter deviation within the positive compensation range, the system calculates the difference between the parameter value of each node and the optimal curing threshold. The magnitude of this difference reflects the degree of adjustment required. For example, if the humidity permeability parameter of a certain node is 15% higher than the optimal threshold, it indicates that a relatively large adjustment is needed for the humidity control of this node. Within the negative compensation range, the system counts the frequency of parameter deviation, that is, the correction frequency. If the parameter deviation of a certain node appears frequently, it means that the curing state of this node is unstable and more frequent corrections are required.

[0070] When generating the execution parameters of the spraying process adjustment instruction, the system considers the spatial weight of the coating detection nodes. The spatial weight reflects the degree of influence of nodes at different positions on the overall coating quality. For example, nodes located in the corners or edges of the cold storage, due to their more complex environmental conditions, may have a greater impact on the overall quality, so their spatial weights are higher than those of the nodes in the central area.

[0071] The system superimposes and combines the adjustment amplitude and the correction frequency according to the spatial weight to generate the final execution parameters. The specific formula is as follows:

[0072] Where: represents the execution parameter of the th coating detection node; represents the spatial weight of the th node, and the value range is and is determined by the position of the node in the cold storage; represents the adjustment amplitude of the th node within the positive compensation range, which is calculated from the parameter deviation value; represents the correction frequency of the th node within the negative compensation range, which is counted from the number of occurrences of parameter deviation.

[0073] In this way, the system can generate personalized spraying process adjustment instructions according to the specific situation of each node. For nodes with a larger spatial weight, the adjustment amplitude has a greater impact on the execution parameters; while for nodes with a smaller spatial weight, the influence of the correction frequency is relatively greater. This method of generating parameters that combines spatial position factors can more specifically adjust the coating state deviation at different positions, thereby improving the quality and efficiency of spraying polyurethane in the cold storage, ensuring that the spraying process can proceed according to the expected curing requirements, and achieving effective control and optimization of the process of spraying polyurethane in the cold storage.

[0074] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0075] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback, characterized in that, Including: An environment acquisition module, which is used to obtain the temperature and humidity data of the cold storage spraying operation area and set an environmental monitoring interval corresponding to the spraying stage; A status analysis module, which is used to divide multiple coating detection nodes within the environmental monitoring interval, extract the characteristics of the polyurethane curing parameters for each node, and generate a coating feature vector corresponding to the node; A parameter regulation module, which is used to separate the temperature influence factor and the humidity response factor from the coating feature vector, establish a curing control rule associated with the node, and obtain the spraying process parameters corresponding to the rule; A dynamic fusion module, which is used to identify the equipment operation status in the spraying process parameters, dynamically compensate the temperature influence factor and the humidity response factor according to the operation status, and calculate the coating parameter difference degree of each node under different compensation strategies; A pattern matching module, which is used to deduce the optimal curing threshold according to the coating parameter difference degree, and generate a coating status deviation sequence by matching the current coating parameters with the optimal curing threshold; A feedback execution module, which is used to analyze the coating status deviation sequence, and based on the parameter difference distribution of the node, convert the coating status deviation sequence into a spraying process adjustment instruction.

2. The real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 1, characterized in that, The implementation method of the status analysis module includes: constructing a spraying feature library corresponding to the coating detection node, and the spraying feature library includes the coating parameter vector mapped by the temperature and humidity data and the curing parameters; Performing similar spraying stage matching on the coating parameter vector, and dividing the spraying clustering group of the coating parameter vector according to the matching result; extracting the distribution center point of the curing parameters from the spraying clustering group, and setting the distribution center point as the coating detection node.

3. The real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 2, characterized in that, Dividing the spraying clustering group of the coating parameter vector further includes: According to the spraying thickness and environmental parameters in the coating parameter vector, extracting the atomization pressure, spraying speed and temperature and humidity change gradient, and generating a spraying feature label based on the above parameters; Associating the spraying feature label with the coating parameter vector, and screening the coating parameter vectors with a spraying similarity higher than the preset spraying threshold through calculating the spraying similarity between the feature labels to form a spraying clustering group.

4. The real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 1, characterized in that, The implementation method of generating the coating feature vector corresponding to the node includes: For each coating detection node, according to the spatial position of the node in the environmental monitoring interval, obtaining the curing fluctuation data of the node within the preset time window, and calculating the curing fluctuation coefficient of the node; When the curing fluctuation coefficient exceeds the first curing threshold, marking the node as an abnormal curing node, and extracting its curing parameters to form a coating feature vector; when the curing fluctuation coefficient is lower than the first curing threshold, marking the node as a stable node, and performing mean value fusion on the curing parameters of the adjacent nodes of the node, and reconstructing the fused data into a coating feature vector.

5. The real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 1, characterized in that, The implementation method of the parameter regulation module includes: Separating the temperature sensitivity, humidity permeability and atomization uniformity from the coating feature vector, and generating a curing control rule for the coating detection node based on the above parameters; If the number of detection nodes covered by the current curing control rule is less than the preset coverage threshold, traversing the coating feature vectors of the adjacent nodes, and adding the curing indexes not included in the control rules of the adjacent nodes to the current rule.

6. The real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 1, wherein The implementation method of the dynamic fusion module includes: obtaining the cycle factor of the spraying interval and the discrete factor of the atomization particle size in the equipment operation status; Construct a parameter compensation network associated with the periodic factor and the discrete factor, and determine the coating parameter difference degree under different compensation strategies according to the compensation weights of each node in the network.

7. The real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 6, characterized in that, Constructing the parameter compensation network further includes: Identifying the time distribution characteristics of the periodic factor. If the current time distribution characteristics exactly match the preset spraying time sequence, set the periodic factor as the initial node of the parameter compensation network; Calculating the compensation correlation degree between the periodic factor and the discrete factor, and generating the intermediate nodes and end nodes of the parameter compensation network in descending order of the correlation degree; Performing parameter reverse verification on the end nodes. When the correlation degree of the end node is lower than the preset verification threshold, output it as the final node of the parameter compensation network.

8. A real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 7, characterized in that, The implementation method of calculating the coating parameter difference degree includes: Statistical the variance of the periodic factor and the standard deviation of the discrete factor of each end node in the parameter compensation network, and calculate the global range of all node parameters; Subtract the variance of the periodic factor of a single end node from the variance of the periodic factor of the adjacent node, and divide by the global range to obtain the period difference coefficient; at the same time, calculate the ratio of the standard deviation of the discrete factor to the global range, and sum the two weighted as the coating parameter difference degree of this node.

9. The real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 1, characterized in that The implementation method of deriving the optimal curing threshold includes: Extract the curing mode in the historical data that is closest to the current coating parameter difference degree, and calculate the Euclidean distance in the spatial distribution between the two as the first matching reference value; Statistical the difference in peak point density between the current coating parameter difference degree and the historical curing mode, and use the density difference value as the second matching reference value; Based on the non-linear combination of the first matching reference value and the second matching reference value, match the optimal curing threshold in the preset curing threshold table.

10. A real-time monitoring system for cold storage spraying polyurethane based on temperature and humidity feedback according to claim 1, characterized in that, The implementation method of the feedback execution module includes: dividing the positive compensation interval and the negative compensation interval according to the parameter difference direction of each node in the coating state deviation sequence; Extract the adjustment amplitude of the parameter deviation in the positive compensation interval and the correction frequency of the parameter deviation in the negative compensation interval, and superimpose and combine the two according to the spatial weight of the coating detection node to generate the execution parameters of the spraying process adjustment instruction.

Images