A wet process equipment PTFE tank stress temperature double parameter real-time monitoring system

Abstract

The application discloses a kind of wet process equipment PTFE tank stress temperature double parameter real-time monitoring system, the system includes following unit: data acquisition unit, for collecting PTFE tank stress and temperature data, obtains original stress-temperature time series data and carries out tank area stress distribution difference analysis, obtains stress distribution difference data;According to stress distribution difference data, the time-space correlation analysis of stress gradient and temperature gradient is carried out, and stress-temperature gradient coupling distribution data is obtained, and the application relates to the technical field of wet process equipment monitoring.This kind of wet process equipment PTFE tank stress temperature double parameter real-time monitoring system, reaches the real-time accurate collection PTFE tank stress temperature data, by multistage analysis optimization, constructs the monitoring model of adaptation different environment and is deployed to cloud platform.Can effectively evaluate coupling effect, realize abnormal early warning, guarantee tank safe operation, reduce fault risk and loss.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wet process equipment monitoring, more particularly, to a wet process equipment PTFE tank stress and temperature dual-parameter real-time monitoring system. BACKGROUND

[0002] In many industrial production fields such as chemical, electronic, pharmaceutical, etc., PTFE tank in wet process equipment plays a crucial role. PTFE material is widely used in the manufacture of various tanks due to its excellent chemical stability, corrosion resistance and low friction coefficient, etc. to meet the special requirements of equipment under different process conditions. However, in the actual production process, PTFE tank is faced with complex and changeable working conditions. On the one hand, due to various chemical reactions, material conveying and other operations involved in the process, the tank will bear different degrees of stress. For example, in the stirring process, the rotation of the stirring paddle will produce mechanical stress on the tank wall; the filling and discharge of materials will also cause the change of internal pressure of the tank, and then produce stress. On the other hand, the temperature change in the production process cannot be ignored. Chemical reactions are often accompanied by heat release or heat absorption, resulting in fluctuations in tank temperature. In addition, the temperature change of the external environment will also affect the tank.

[0003] Stress and temperature are not isolated, there is a close coupling relationship between them. Excessive stress can cause deformation or even rupture of the tank structure, and abnormal temperature change will change the physical properties of PTFE material, such as elastic modulus, thermal expansion coefficient, etc., and then affect the stress bearing capacity of the tank. At the same time, uneven distribution of stress can also cause local temperature to abnormally rise or fall, both of which interact and act together, seriously threatening the safe operation of the tank.

[0004] At present, the monitoring of PTFE tank mainly has the following problems:

[0005] Single monitoring parameter: Most monitoring systems only focus on one of the stress or temperature parameters, and cannot fully understand the actual running state of the tank. The monitoring of a single parameter cannot find the potential correlation between stress and temperature, and potential fault hidden dangers are easily missed.

[0006] Lack of spatiotemporal correlation analysis: The existing monitoring methods often only measure the parameters at a certain time or a certain position, without considering the dynamic changes of stress and temperature in time and space and their mutual relationship. It is difficult to accurately grasp the evolution law of tank stress and temperature, and it is difficult to give early warning of possible abnormal situations.

[0007] Poor environmental adaptability: the operation environment of the equipment is complex and changeable, and environmental factors such as humidity, air pressure and vibration will affect the stress and temperature state of the tank. However, the current monitoring system rarely considers these environmental interference factors, which affects the accuracy and reliability of the monitoring data, and cannot adjust the monitoring strategy in time according to the actual environmental changes.

[0008] Incomplete early warning mechanism: lack of scientific and effective abnormal early warning quantitative evaluation method, it is difficult to accurately judge whether the stress and temperature of the tank body are abnormal and the severity of the abnormality. The setting of the early warning threshold is unreasonable, which is easy to cause false alarm or miss report, and cannot provide timely and accurate decision basis for the operator.

[0009] Therefore, it has important practical significance to develop a monitoring system that can monitor the stress and temperature of PTFE tank in real time, comprehensively and accurately, and has good environmental adaptability and reliable abnormal early warning mechanism. SUMMARY

[0010] In view of the problems of single monitoring parameter, lack of space-time correlation analysis, poor environmental adaptability and imperfect early warning mechanism in the prior art, the purpose of the present application is to provide a wet process equipment PTFE tank stress and temperature double parameter real-time monitoring system, which can realize real-time and accurate collection of PTFE tank stress and temperature data, multi-link analysis and optimization, construction of monitoring model suitable for different environments and deployment to cloud platform. It can effectively evaluate the coupling effect, realize abnormal early warning, ensure the safe operation of the tank body, reduce the risk and loss of failure.

[0011] To solve the above problems, the technical scheme adopted by the present application is as follows.

[0012] A wet process equipment PTFE tank stress and temperature double parameter real-time monitoring system, the system comprises the following units:

[0013] The data acquisition unit is used for acquiring PTFE tank stress and temperature data, obtaining original stress-temperature time series data and performing tank body area stress distribution difference analysis to obtain stress distribution difference data; according to the stress distribution difference data, the space-time correlation analysis of stress gradient and temperature gradient is carried out, and the stress-temperature gradient coupling distribution data is obtained;

[0014] The space-time feature analysis unit is used for marking the stress-temperature space-time motion trajectory of the key area of the tank body according to the stress-temperature gradient coupling distribution data, obtaining the stress-temperature space-time trajectory data; based on the stress-temperature space-time trajectory data, the coupling effect evaluation of local stress concentration and temperature anomaly is carried out, and the stress-temperature coupling effect evaluation data is obtained;

[0015] The environmental adaptability monitoring period optimization unit is configured to simulate a monitoring period demand according to stress-temperature coupling effect evaluation data, generate initial period benchmark data, collect equipment operating environment parameters, identify environmental interference factors for the stress-temperature coupling effect evaluation data according to the operating environment parameters, obtain associated environmental interference factors, dynamically correct the initial monitoring period benchmark data based on the associated environmental interference factors, and obtain environmental adaptability monitoring period data.

[0016] The monitoring model construction and deployment unit is configured to construct a real-time monitoring model for the environmental adaptability monitoring period data by using a policy gradient algorithm, obtain a PTFE tank stress-temperature dual-parameter real-time monitoring model, and deploy the real-time monitoring model to a cloud platform to perform dual-parameter real-time monitoring and abnormal early warning of the tank stress and temperature.

[0017] Further, the stress distribution difference data is obtained by the following steps:

[0018] The collected original stress-temperature time series data is divided according to different regions of the tank to form regional sub-data sets;

[0019] The mean and variance of each regional sub-data set are calculated to obtain regional stress statistical characteristic values;

[0020] The regional stress statistical characteristic values are compared with each other to analyze the difference degree of the stress statistical characteristics between different regions;

[0021] The regions with a larger difference degree are marked as key regions in combination with the structural characteristics and process flow of the tank;

[0022] The stress distribution difference of each region of the tank is determined according to the comparison results of the stress statistical characteristics of the key marked regions and other regions, and the stress distribution difference data is obtained.

[0023] Further, the stress-temperature gradient coupling distribution data is obtained by the following steps:

[0024] The stress distribution difference data is subjected to spatial interpolation processing to construct a stress gradient field of the tank;

[0025] A temperature gradient field is constructed based on the temperature data;

[0026] The stress gradient field and the temperature gradient field are time-space aligned to analyze the correlation between the change trends of the stress gradient and the temperature gradient at the same time-space position;

[0027] The correlation analysis results are used to establish a stress-temperature gradient coupling relationship model in combination with the elastic modulus and thermal expansion coefficient characteristics of the tank material;

[0028] According to the model, the stress-temperature gradient coupling values at different space-time positions are calculated to obtain stress-temperature gradient coupling distribution data.

[0029] Further, the stress-temperature space-time trajectory data is obtained, including the following steps:

[0030] Based on the stress-temperature gradient coupling distribution data, the key area of the tank body is selected as the research object;

[0031] Set a time interval, and record the stress and temperature values in the key area at each time point;

[0032] According to the stress-temperature values at different time points, the trajectory tracking algorithm is used to mark the movement path of stress and temperature in space-time;

[0033] Smooth the marked path to eliminate abnormal fluctuation points;

[0034] Integrate the processed path data to obtain the stress-temperature space-time trajectory data in the key area.

[0035] Further, the stress-temperature coupling effect evaluation data is obtained, including the following steps:

[0036] Based on the stress-temperature space-time trajectory data, the trajectory features of the local stress concentration area are extracted;

[0037] Analyze the trajectory features of the temperature abnormal area;

[0038] Compare the trajectory features of the local stress concentration and temperature abnormality to judge the overlap degree in space-time;

[0039] Extract cases similar to the current overlap degree from the tank failure case database, analyze the severity and frequency of local stress concentration and temperature abnormality leading to tank failure in these cases, and according to these information, according to the pre-set coupling effect intensity classification standard, the classification standard includes slight, moderate, and severe, to evaluate the coupling effect intensity of local stress concentration and temperature abnormality;

[0040] Use fuzzy comprehensive evaluation method to quantitatively evaluate the coupling effect to obtain stress-temperature coupling effect evaluation data.

[0041] Further, the initial period reference data is generated, including the following steps:

[0042] Based on the stress-temperature coupling effect evaluation data, analyze the change law of the coupling effect with time;

[0043] Determine the monitoring requirement level under different coupling effect intensities combined with the design service life and safety factor requirements of the tank.

[0044] According to the monitoring demand level, referring to industry standards and similar equipment monitoring experience, set the monitoring period range corresponding to different levels;

[0045] Using linear programming method, determine the initial monitoring period benchmark data under the condition of meeting the monitoring demand and cost control.

[0046] Further, the associated environmental interference factors are obtained, including the following steps:

[0047] Collecting equipment operating environment parameters, the environmental parameters include humidity, air pressure, vibration;

[0048] Synchronous analysis of stress-temperature coupling effect evaluation data and environmental parameters;

[0049] Using principal component analysis method, extract the environmental parameters which have significant influence on coupling effect as candidate interference factors;

[0050] Correlation test is performed on the candidate interference factors to remove redundant factors with strong correlation;

[0051] Through the establishment of multiple linear regression model, the quantitative relationship between the remaining candidate interference factors and the coupling effect is analyzed;

[0052] According to the significance test result of the regression model, the environmental interference factors associated with the stress-temperature coupling effect evaluation data are determined.

[0053] Further, the environmental adaptability monitoring period data is obtained, including the following steps:

[0054] Based on the associated environmental interference factors, the influence weight of different interference factors on the initial monitoring period benchmark data is analyzed;

[0055] Establish the mapping relationship model between environmental interference factors and monitoring period adjustment coefficient;

[0056] Real-time collection of environmental parameters, calculation of monitoring period adjustment coefficient under current environment according to the mapping relationship model;

[0057] Apply the adjustment coefficient to the initial monitoring period benchmark data for dynamic correction, and set the upper and lower limits of period adjustment, the minimum monitoring period is 5 minutes, and the maximum monitoring period is 60 minutes;

[0058] Taking the prediction accuracy of the monitoring model reaching 90% as the termination condition of iterative optimization, after multiple iterations and optimization, the environmental adaptability monitoring period data is obtained.

[0059] Further, the PTFE groove stress-temperature double parameter real-time monitoring model is obtained, including the following steps:

[0060] The monitoring cycle data of environmental adaptability is taken as input to build a monitoring dataset containing stress and temperature double parameters;

[0061] The dataset is preprocessed, including data cleaning and normalization;

[0062] The policy gradient algorithm is selected as the model construction algorithm, and the initial parameters of the algorithm are set;

[0063] The preprocessed dataset is divided into training set and test set;

[0064] The policy gradient algorithm is trained using the training set, and the algorithm parameters are continuously adjusted to optimize the model performance;

[0065] The trained model is verified using the test set to evaluate the accuracy and generalization ability of the model;

[0066] According to the evaluation results, the model is fine-tuned to obtain the PTFE tank stress and temperature double parameter real-time monitoring model.

[0067] Further, after obtaining the PTFE tank stress and temperature double parameter real-time monitoring model, in order to further quantify the evaluation of the model's early warning ability for tank stress and temperature anomalies, the abnormal early warning quantitative evaluation formula is introduced:

[0068] Let be the actual stress value at the th time point, be the stress value predicted by the monitoring model at the th time point; be the actual temperature value at the th time point, be the temperature value predicted by the monitoring model at the th time point; be the total number of selected time points for evaluation; and be the weight coefficients of stress and temperature in abnormal early warning evaluation, and , , The value is determined according to the importance of stress and temperature to the safety of the tank. If stress has a greater impact on the safety of the tank, the value of is relatively large, and vice versa. The value of is relatively large;

[0069] The stress prediction error , and the temperature prediction error

[0070] The calculation formula of the comprehensive abnormal early warning evaluation index is:

[0071]

[0072] When the comprehensive abnormal early warning evaluation index is greater than the preset early warning threshold , the system determines that the stress and temperature of the tank body are abnormal, triggers the abnormal early warning mechanism, and sends early warning information to relevant personnel to take timely measures to ensure the safe operation of the tank body;

[0073] Among them, and , and respectively represent the actual value and the predicted value, which are used to calculate the error, is the number of evaluation time points, which ensures that the evaluation is based on a certain amount of data, and are weight coefficients, which can adjust the importance of stress and temperature in evaluation according to actual situation, is the comprehensive abnormal early warning evaluation index, which is used to determine whether to trigger the early warning, is the early warning threshold, which can be determined according to the safety standard of the tank body and historical data.

[0074] The advantages of the present application are:

[0075] 1、The system can simultaneously collect stress and temperature data of PTFE tank body, realize real-time monitoring of double parameters, and obtain stress distribution difference data and stress-temperature gradient coupling distribution data through in-depth analysis of original stress-temperature time series data by the data acquisition unit, so as to comprehensively understand the stress temperature state and its change rule of different regions of the tank body, avoid the limitation of single parameter monitoring, and greatly improve the accuracy and comprehensiveness of the monitoring.

[0076] 2、The space-time feature analysis unit marks the stress-temperature space-time motion trajectory of the key area of the tank body based on the stress-temperature gradient coupling distribution data, and evaluates the coupling effect of local stress concentration and temperature anomaly. This space-time correlation analysis method can accurately grasp the dynamic changes of stress and temperature in time and space and their mutual relationship, which helps to find potential hidden troubles and predict possible problems of the tank body in advance, and provides a scientific basis for the maintenance and repair of the equipment.

[0077] 3、The environmental adaptability monitoring period optimization unit fully considers the influence of equipment operating environment parameters on stress-temperature coupling effect. By identifying the associated environmental interference factors and dynamically correcting the initial monitoring period benchmark data, the monitoring period can be automatically adjusted according to the actual environmental changes, so as to ensure that accurate and reliable monitoring data can be obtained under different environments, and the environmental adaptability and stability of the system are improved.

[0078] 4. The abnormal early warning quantitative evaluation formula is introduced, the weight coefficients of stress and temperature in the abnormal early warning evaluation are comprehensively considered, the comprehensive abnormal early warning evaluation index is calculated and compared with the pre-set early warning threshold, and whether the stress and temperature of the groove body are abnormal and the severity of the abnormality is accurately judged. This scientific early warning mechanism can effectively avoid the occurrence of false and missed reports, provide timely and accurate early warning information for the operating personnel, so as to take timely measures to ensure the safe operation of the groove body, reduce the equipment failure risk and reduce the production loss. BRIEF DESCRIPTION OF DRAWINGS

[0079] The drawings described herein are used to provide further understanding of the present application, constitute a part of the present application, the illustrative embodiments of the present application and the description thereof are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0080] Fig. 1 The system block diagram of the present application is shown in the figure;

[0081] Fig. 2 The flow chart of the environment adaptability monitoring cycle optimization unit of the present application is shown in the figure. DETAILED DESCRIPTION

[0082] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application; obviously; the described embodiments are only part of the embodiments of the present application; rather than all the embodiments. Based on the embodiments in the present application; all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0083] Embodiment:

[0084] Please refer to Figs. 1-2 The present application provides a technical solution: a wet process equipment PTFE groove body stress and temperature double parameter real-time monitoring system, which comprises the following units:

[0085] The data acquisition unit is used for acquiring PTFE groove body stress and temperature data, obtaining original stress-temperature time series data and performing groove body area stress distribution difference analysis to obtain stress distribution difference data; according to the stress distribution difference data, the space-time correlation analysis of stress gradient and temperature gradient is performed to obtain stress-temperature gradient coupling distribution data;

[0086] The original stress-temperature time sequence data is a data set composed of stress values and corresponding temperature values borne by different positions of the PTFE tank at different times recorded in time sequence by a specific sensor or other data collection equipment; the tank region stress distribution difference analysis is a comparative analysis of stress data of each different region of the PTFE tank, to find out the difference in stress size between different regions, so as to understand the distribution characteristics of the stress inside the tank; the stress distribution difference data is obtained after the tank region stress distribution difference analysis, and can reflect the stress distribution difference degree and characteristics of different regions of the PTFE tank; the stress gradient refers to the rate of change of stress in space, that is, the amount of change of stress per unit distance, which is used to describe the speed and direction of change of stress between different positions inside the tank; the temperature gradient is similar to the stress gradient, which refers to the rate of change of temperature in space, indicating the amount of change of temperature per unit distance, reflecting the change of temperature between different positions inside the tank; the stress-temperature gradient coupling distribution data is the data obtained by analyzing the common distribution characteristics of stress-temperature gradient in the tank space on the basis of considering the mutual influence and interaction of stress gradient and temperature gradient;

[0087] The space-time feature analysis unit is configured to mark stress-temperature space-time motion trajectories of key regions of the tank according to the stress-temperature gradient coupling distribution data, to obtain stress-temperature space-time trajectory data; and evaluate the coupling effect of local stress concentration and temperature anomaly based on the stress-temperature space-time trajectory data, to obtain stress-temperature coupling effect evaluation data.

[0088] The stress-temperature space-time motion trajectory mark records and identifies the changes of stress and temperature with time and the movement in spatial position in the key area of the PTFE tank, so as to intuitively present the dynamic change process of stress and temperature in the space-time dimension. The stress-temperature space-time trajectory data obtained through the stress-temperature space-time motion trajectory mark contains a data set of the change information of stress and temperature in time and space. The local stress concentration is a phenomenon that the stress value in some local area of the PTFE tank is obviously higher than that in other surrounding areas, which may be caused by factors such as tank structure and stress condition. The local stress concentration may cause damage or failure of the tank. The temperature anomaly refers to the case that the temperature of some parts of the PTFE tank deviates from the normal working temperature range, which may be caused by equipment failure, environmental factors, etc. The temperature anomaly may also have an adverse effect on the performance and service life of the tank. The coupling effect evaluation of local stress concentration and temperature anomaly: analyze the mutual influence and interaction between the two factors of local stress concentration and temperature anomaly, evaluate the comprehensive influence degree and nature of their joint action on the PTFE tank, such as whether it will accelerate the aging and damage of the tank, etc. The stress-temperature coupling effect evaluation data obtained after the coupling effect evaluation of local stress concentration and temperature anomaly can reflect the influence of the interaction of stress concentration and temperature anomaly on the tank.

[0089] The environment adaptability monitoring cycle optimization unit is used to simulate the monitoring demand cycle according to the stress-temperature coupling effect evaluation data, generate initial cycle reference data, collect equipment operating environment parameters, identify the stress-temperature coupling effect evaluation data according to the operating environment parameters, obtain the associated environmental interference factors, and dynamically correct the initial monitoring cycle reference data based on the associated environmental interference factors to obtain the environment adaptability monitoring cycle data.

[0090] The monitoring demand cycle is determined according to the actual situation of the PTFE tank body, process requirements, stress-temperature coupling effect and other factors, and is the time interval required for monitoring the stress and temperature of the tank body to meet the demand for timely discovering potential problems and ensuring safe operation of the equipment; the initial cycle reference data is initial reference data of the monitoring demand cycle simulated according to the stress-temperature coupling effect evaluation data without considering the influence of the equipment operating environment parameters; the equipment operating environment parameters refer to data of various environmental factors that affect the stress and temperature of the PTFE tank body and the monitoring effect, such as environmental temperature, humidity, pressure, chemical properties of the surrounding medium and the like; the environmental interference factor is a factor that will interfere with the stress-temperature coupling effect evaluation data of the PTFE tank body in the equipment operating environment parameters, thereby affecting the monitoring accuracy and effectiveness; the associated environmental interference factor is an environmental interference factor that has a significant influence on the monitoring data, which is determined by analyzing the relationship between the equipment operating environment parameters and the stress-temperature coupling effect evaluation data; the environmental adaptability monitoring cycle data is monitoring cycle data that is more in line with the actual equipment operating environment conditions and can improve the pertinence and effectiveness of monitoring, which is obtained by dynamically correcting the initial cycle reference data considering the influence of the associated environmental interference factor;

[0091] The monitoring model construction and deployment unit is configured to construct a real-time monitoring model for the PTFE tank body stress and temperature double parameters by using a policy gradient algorithm on the environmental adaptability monitoring cycle data, and deploy the real-time monitoring model to a cloud platform to perform real-time monitoring and abnormal early warning of the double parameters of the stress and temperature of the tank body.

[0092] The policy gradient algorithm is a reinforcement learning algorithm that finds the optimal policy by directly optimizing the parameters of the policy function. In the system, it is used to construct a real-time monitoring model based on the environmental adaptability monitoring cycle data, and continuously adjusts the model parameters to make the model better adapt to the actual monitoring demand, so as to realize accurate monitoring of the PTFE tank body stress and temperature double parameters; the PTFE tank body stress and temperature double parameter real-time monitoring model is constructed based on the environmental adaptability monitoring cycle data by using the policy gradient algorithm, and is a mathematical model or algorithm system that can perform real-time monitoring and data analysis on the stress and temperature of the PTFE tank body, and is used to determine whether the tank body is in normal state or to issue an abnormal early warning;

[0093] It should be noted that in operation, the wet process equipment refers to the equipment that uses liquid medium (such as solution, suspension, etc.) for processing, treatment or reaction in the process; the PTFE tank body is a kind of high-performance engineering plastic with excellent chemical corrosion resistance, high temperature resistance and low friction coefficient, etc. The PTFE tank body is a container made of such material, which is used to contain the liquid medium in the wet process equipment and related reaction or treatment process; the stress-temperature double parameter real-time monitoring system is a system specially designed for PTFE tank body, which simultaneously monitors the stress and temperature, two key parameters, to timely grasp the state change of the tank body in these two aspects and ensure the safe and stable operation of the equipment.

[0094] In an embodiment, obtaining stress distribution difference data includes the following steps:

[0095] The collected original stress-temperature time series data is divided according to different regions of the tank body to form regional sub-data sets;

[0096] The mean and variance of each regional sub-data set are calculated to obtain the stress statistical characteristic values of each region;

[0097] The stress statistical characteristic values of each region are compared with each other to analyze the difference degree of stress statistical characteristics between different regions;

[0098] According to the structure characteristics and process flow of the tank body, the regions with larger difference degree are marked as key regions;

[0099] According to the comparison results of the stress statistical characteristics of the key marked regions and other regions, the stress distribution difference of each region of the tank body is determined, and then the stress distribution difference data is obtained.

[0100] In this way, the stress distribution is analyzed from local to global and gradually in-depth. Through quantitative statistical characteristics and comparative analysis, the stress abnormal region can be accurately located, which provides a reliable basis for further research on the relationship between stress and temperature and the safety evaluation of the tank body, helps to find potential risk points in advance, and ensures the stable operation of the tank body.

[0101] In an embodiment, obtaining stress-temperature gradient coupling distribution data includes the following steps:

[0102] The stress distribution difference data is subjected to spatial interpolation processing to construct a stress gradient field of the tank body;

[0103] A temperature gradient field is constructed based on the temperature data;

[0104] The stress gradient field and the temperature gradient field are time and space aligned to analyze the correlation between the change trends of the stress gradient and the temperature gradient at the same time and space position;

[0105] Using the correlation analysis results, combined with the elastic modulus and thermal expansion coefficient characteristics of the groove material, a stress-temperature gradient coupling relationship model is established.

[0106] According to the model, the stress-temperature gradient coupling values at different spatial and temporal positions are calculated, and the stress-temperature gradient coupling distribution data is obtained.

[0107] This design takes into account the changes of stress and temperature in space and time as well as the material properties. By constructing the gradient field and coupling model, the internal relationship between stress and temperature gradient can be revealed in depth. The obtained coupling distribution data provides a key basis for subsequent analysis of stress-temperature space-time trajectory and coupling effect, and helps to better understand the physical state changes of the groove.

[0108] In an embodiment, the stress-temperature space-time trajectory data is obtained, including the following steps:

[0109] Based on the stress-temperature gradient coupling distribution data, the key areas of the groove are selected as the research object;

[0110] Set the time interval, and record the stress and temperature values in the key area at each time point;

[0111] According to the stress-temperature values at different time points, the trajectory tracking algorithm is used to mark the movement path of stress and temperature in space-time;

[0112] Smooth the marked path to eliminate abnormal fluctuation points;

[0113] Integrate the processed path data to obtain the space-time motion trajectory data of stress and temperature in the key area.

[0114] This design can clearly show the dynamic changes of stress and temperature in space-time by selecting key areas and recording time series data, combined with the trajectory tracking algorithm. The smoothing process improves the reliability of the data, and the obtained space-time trajectory data provides intuitive and accurate information for analyzing the stress-temperature coupling effect, which helps to discover abnormal trends in a timely manner.

[0115] In an embodiment, the stress-temperature coupling effect evaluation data is obtained, including the following steps:

[0116] Based on the stress-temperature space-time trajectory data, the trajectory features of the local stress concentration area are extracted;

[0117] Analyze the trajectory features of the temperature abnormal area;

[0118] Compare the trajectory features of local stress concentration and temperature abnormality to judge the overlap degree in space-time;

[0119] extracting cases with similar current overlap degree from the groove body failure case database, analyzing the severity and frequency of local stress concentration and temperature anomaly leading to groove body failure in these cases, and evaluating the coupling effect intensity of local stress concentration and temperature anomaly according to the pre-set coupling effect intensity classification standard, which includes slight, medium and severe;

[0120] quantitative evaluation of the coupling effect is performed by using the fuzzy comprehensive evaluation method to obtain stress-temperature coupling effect evaluation data.

[0121] In this way, the stress-temperature coupling effect can be comprehensively and objectively evaluated by using trajectory feature analysis, historical case reference and quantitative evaluation method. The classification evaluation and quantitative data provide a scientific basis for the safety evaluation and maintenance decision of the groove body, which helps to take preventive measures to prevent groove body failure in advance.

[0122] In an embodiment, the initial period reference data is generated, including the following steps:

[0123] Based on the stress-temperature coupling effect evaluation data, the variation law of the coupling effect with time is analyzed;

[0124] Combined with the design service life and safety factor requirements of the groove body, the monitoring requirement level under different coupling effect intensities is determined;

[0125] According to the monitoring requirement level, the monitoring period range corresponding to different levels is set by referring to industry standards and monitoring experience of similar equipment;

[0126] The linear programming method is used to determine the initial monitoring period reference data under the condition of meeting the monitoring requirements and cost control.

[0127] In this way, the coupling effect variation, equipment requirements and cost control are considered comprehensively. Through scientific analysis and reasonable planning, the generated initial monitoring period reference data can not only ensure effective monitoring of the groove body, but also avoid excessive monitoring and waste of resources, providing a basis for subsequent adjustment of the monitoring period according to environmental factors.

[0128] In an embodiment, the associated environmental interference factors are obtained, including the following steps:

[0129] Collecting equipment operating environment parameters, including humidity, air pressure and vibration;

[0130] Synchronous analysis of stress-temperature coupling effect evaluation data and environmental parameters is performed;

[0131] The principal component analysis method is used to extract environmental parameters that have a significant impact on the coupling effect as candidate interference factors;

[0132] Correlation test is performed on the candidate interference factors to remove redundant factors with strong correlation;

[0133] A multiple linear regression model is established to analyze the quantitative relationship between the remaining candidate interference factors and the coupling effect;

[0134] According to the significance test result of the regression model, the environmental interference factors associated with the stress-temperature coupling effect evaluation data are determined.

[0135] In this way, by collecting and analyzing environmental parameters, and using various statistical methods, the environmental interference factors associated with the stress-temperature coupling effect can be accurately identified. This helps to better understand the influence of environmental factors on the state of the tank, and provides a basis for subsequent adjustment of the monitoring period according to environmental changes, improving the pertinence and accuracy of monitoring.

[0136] In an embodiment, obtaining the environment-adaptive monitoring period data includes the following steps:

[0137] Based on the associated environmental interference factors, the influence weight of different interference factors on the initial monitoring period reference data is analyzed;

[0138] A mapping relationship model between the environmental interference factors and the monitoring period adjustment coefficient is established;

[0139] Real-time collection of environmental parameters, calculation of the monitoring period adjustment coefficient under the current environment according to the mapping relationship model;

[0140] The adjustment coefficient is applied to the initial monitoring period reference data for dynamic correction, and the upper and lower limits of the period adjustment are set, with the minimum monitoring period being 5 minutes and the maximum monitoring period being 60 minutes;

[0141] Taking the prediction accuracy of the monitoring model reaching 90% as the termination condition of iterative optimization, the environment-adaptive monitoring period data is obtained after multiple iterations and optimizations.

[0142] In this way, the dynamic changes of environmental factors are considered, and the monitoring period can better adapt to the actual environment through the establishment of a mapping relationship and real-time adjustment. The upper and lower limits and the iterative optimization mechanism ensure the rationality and effectiveness of the monitoring period, improve the prediction accuracy of the monitoring model for the stress and temperature changes of the tank, and enhance the reliability of the monitoring system.

[0143] In an embodiment, obtaining the PTFE tank stress-temperature dual-parameter real-time monitoring model includes the following steps:

[0144] Taking the environment-adaptive monitoring period data as input, a monitoring data set containing stress and temperature dual parameters is constructed;

[0145] The data set is preprocessed, including data cleaning and normalization;

[0146] The policy gradient algorithm is selected as the model building algorithm, and the initial parameters of the algorithm are set.

[0147] The preprocessed dataset is divided into a training set and a test set;

[0148] The policy gradient algorithm is trained using the training set, and the algorithm parameters are continuously adjusted to optimize model performance.

[0149] Use the test set to validate the trained model and evaluate its accuracy and generalization ability;

[0150] Based on the evaluation results, the model was fine-tuned to obtain a real-time monitoring model for the stress and temperature of the PTFE tank.

[0151] This design forms a complete process from data preparation to model building, training, and validation. Through reasonable data processing and algorithm selection, an accurate and reliable real-time monitoring model can be built, enabling real-time monitoring and early warning of anomalies in the stress and temperature parameters of the PTFE tank, thus providing strong protection for the safe operation of the tank.

[0152] In one embodiment, after obtaining the real-time monitoring model for the dual parameters of stress and temperature in the PTFE tank, an anomaly early warning quantitative evaluation formula is introduced to further quantify and evaluate the model's ability to predict anomalies in tank stress and temperature:

[0153] set up For the first Actual stress values ​​at each time point (unit: MPa). For monitoring model in the first Predicted stress values ​​at specific time points (unit: MPa); For the first The actual temperature values ​​at each time point (unit: °C). For monitoring model in the first Predicted temperature values ​​at specific times (unit: °C); To determine the total number of time points selected for evaluation, data from a continuous 24-hour period is typically chosen, with data collected every 15 minutes. ;

[0154] and These are the weighting coefficients of stress and temperature in the anomaly early warning assessment, respectively. , , Its value is determined based on the relative importance of stress and temperature to the safety of the tank. If stress has a greater impact on the safety of the tank, then... The value is relatively large, and vice versa. The value is relatively large, according to the past research on PTFE tank and practical engineering experience, when the tank is running in normal condition, the influence of stress on the integrity of the tank structure is more critical than temperature, after expert evaluation and a large number of case analysis, it is determined that , ;

[0155] Definition of stress prediction error , temperature prediction error ;

[0156] Then the calculation formula of comprehensive abnormal early warning evaluation index

[0157]

[0158] In order to determine the early warning threshold , through the simulation analysis of a large number of historical data of PTFE tank under normal and abnormal conditions, 100 groups of data in different time periods (each group for 24 hours) are calculated, the average value and standard deviation of comprehensive abnormal early warning evaluation index are obtained, after calculation (MPa and ℃ comprehensive unit, the same below), ;

[0159] Comprehensive consideration of data distribution under normal and abnormal conditions, and in order to reduce false alarm and omission as far as possible, set the early warning threshold ;

[0160] When the comprehensive abnormal early warning evaluation index is greater than the pre-set early warning threshold , the system determines that the stress and temperature of the tank are abnormal, triggers the abnormal early warning mechanism, and sends early warning information to the relevant personnel, so as to take timely measures to ensure the safe operation of the tank;

[0161] Among them, and , and respectively represent the actual value and the predicted value, which are used to calculate the error,

[0162] is the number of evaluation time points, the data in 24 consecutive hours is selected, and it is collected once every 15 minutes. This setting can cover various working condition changes of the tank in a complete running cycle, while ensuring that the data quantity is sufficient to accurately evaluate the early warning ability of the model,

[0163] and ​is a weight coefficient, according to previous research and practical engineering experience, combined with expert evaluation and a large number of case analysis, the stress weight is 0.6, the temperature weight is 0.4, reflecting the relatively more critical influence of stress on the integrity of the tank structure,

[0164] The importance of stress and temperature in evaluation can be adjusted according to actual situation, is a comprehensive abnormal early warning evaluation index, used to judge whether to trigger early warning,

[0165] is a warning threshold, through simulation analysis of a large number of historical data under normal and abnormal conditions, the average value and standard deviation under normal condition, and the average value under abnormal condition are calculated, considering the data distribution, the method of is used to set the early warning threshold, which can reduce false positives and false negatives, and discover abnormal conditions of the tank in time.

[0166] In this way, the prediction errors of stress and temperature are considered comprehensively, and the comprehensive abnormal early warning evaluation index is obtained by weighted average, making the evaluation more comprehensive and accurate.

[0167] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment methods can be completed by programs instructing related hardware, therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer usable program codes.

[0168] The above embodiments have been described in detail, and the principles and implementation modes of the present application have been described by applying specific examples; the above embodiment descriptions are only used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed; in summary, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A real-time monitoring system for dual parameters of stress and temperature in a wet PTFE tank of a processing plant, characterized in that, The system includes the following units: The data acquisition unit is used to collect stress and temperature data of the PTFE tank, obtain raw stress-temperature time series data, and perform stress distribution difference analysis in the tank area to obtain stress distribution difference data. Based on the stress distribution difference data, the spatiotemporal correlation analysis of stress gradient and temperature gradient is performed to obtain stress-temperature gradient coupled distribution data. The stress-temperature gradient coupling distribution data is obtained through the following steps: Spatial interpolation is performed on the stress distribution difference data to construct the stress gradient field of the tank. Construct a temperature gradient field based on temperature data; The stress gradient field and the temperature gradient field are spatiotemporally aligned to analyze the correlation between the changing trends of the stress gradient and the temperature gradient at the same spatiotemporal location. Using the results of correlation analysis, and combining the elastic modulus and thermal expansion coefficient characteristics of the tank material, a stress-temperature gradient coupling model was established. Based on this model, the stress-temperature gradient coupling values ​​at different spatiotemporal locations are calculated, and the stress-temperature gradient coupling distribution data are obtained. The spatiotemporal feature analysis unit is used to mark the spatiotemporal motion trajectory of stress and temperature in key areas of the tank based on the stress-temperature gradient coupling distribution data, and to obtain stress-temperature spatiotemporal trajectory data. The coupling effect of local stress concentration and temperature anomaly is evaluated based on stress-temperature spatiotemporal trajectory data, and stress-temperature coupling effect evaluation data is obtained. The stress-temperature spatiotemporal trajectory data is obtained through the following steps: Based on the stress-temperature gradient coupling distribution data, key areas of the tank were selected as the research objects. Set time intervals and record the stress and temperature values ​​in the key areas at each time point; Based on the stress-temperature values ​​at different time points, a trajectory tracking algorithm is used to mark the movement path of stress and temperature in space and time. The marked path is smoothed to eliminate abnormal fluctuations; The processed path data are integrated to obtain the spatiotemporal motion trajectory data of stress and temperature in the key area; The environmental adaptability monitoring cycle optimization unit is used to simulate the monitoring demand cycle based on the stress-temperature coupling effect assessment data and generate initial cycle baseline data. Collect equipment operating environment parameters, identify environmental interference factors based on stress-temperature coupling effect assessment data according to operating environment parameters, and obtain associated environmental interference factors; The environmental adaptability monitoring cycle data is obtained by dynamically correcting the initial monitoring cycle baseline data based on the associated environmental interference factors. The monitoring model construction and deployment unit is used to construct a real-time monitoring model for environmental adaptability monitoring cycle data through a strategy gradient algorithm, resulting in a real-time monitoring model for the stress and temperature of the PTFE tank. The real-time monitoring model is then deployed to a cloud platform to perform real-time monitoring and anomaly warning for the stress and temperature of the tank.

2. The real-time monitoring system for dual parameters of stress and temperature in a wet PTFE tank of a wet processing equipment according to claim 1, characterized in that, Obtaining stress distribution difference data includes the following steps: The collected raw stress-temperature time series data were divided into different regions of the tank to form regional subsets. The mean and variance of each regional subset are calculated to obtain the stress statistical characteristic values ​​of each region. By comparing the stress statistical characteristic values ​​of each region pairwise, the degree of difference in stress statistical characteristics between different regions is analyzed. Based on the structural characteristics of the tank and the process flow, areas with significant differences are marked as key areas; Based on the comparison of stress statistical characteristics between the key marked areas and other areas, the stress distribution differences in each area of ​​the tank are determined, and thus stress distribution difference data are obtained.

3. The real-time monitoring system for dual parameters of stress and temperature in a wet PTFE tank of a wet processing equipment according to claim 2, characterized in that, Obtaining stress-temperature coupling effect assessment data includes the following steps: Based on stress-temperature spatiotemporal trajectory data, trajectory features of local stress concentration regions are extracted; Analyze the trajectory characteristics of the temperature anomaly region; By comparing the trajectory characteristics of local stress concentration and temperature anomalies, the degree of overlap between the two in space and time can be determined. Cases with similar overlap to the current case are extracted from the tank failure case database. The severity and frequency of tank failure caused by local stress concentration and temperature anomaly in these cases are analyzed. Based on this information, the coupling effect strength of local stress concentration and temperature anomaly is evaluated according to a pre-set coupling effect strength grading standard, which includes slight, moderate and severe. The fuzzy comprehensive evaluation method was used to quantitatively evaluate the coupling effect and obtain the evaluation data of the stress-temperature coupling effect.

4. The real-time monitoring system for dual parameters of stress and temperature in a wet PTFE tank of a wet processing equipment according to claim 3, characterized in that, Generating initial periodic baseline data includes the following steps: Based on the stress-temperature coupling effect assessment data, the variation law of the coupling effect over time is analyzed; Based on the design service life and safety factor requirements of the tank, the monitoring requirement level under different coupling effect intensities is determined; Based on the monitoring requirement level and with reference to industry standards and monitoring experience with similar equipment, set the monitoring cycle range corresponding to different levels; Using linear programming, baseline data for the initial monitoring cycle are determined while meeting monitoring requirements and controlling costs.

5. The real-time monitoring system for dual parameters of stress and temperature in a wet PTFE tank of a wet processing equipment according to claim 4, characterized in that, The steps to obtain the associated environmental interference factor include: Collect equipment operating environment parameters, including humidity, air pressure, and vibration; Simultaneous analysis of stress-temperature coupling effect assessment data with environmental parameters; Principal component analysis was used to extract environmental parameters that significantly affect the coupling effect as candidate interference factors. Perform a correlation test on the candidate interference factors and remove redundant factors with strong correlations. By establishing a multiple linear regression model, the quantitative relationship between the remaining candidate interference factors and the coupling effect is analyzed. Based on the significance test results of the regression model, environmental disturbance factors associated with the stress-temperature coupling effect assessment data were identified.

6. The real-time monitoring system for dual parameters of stress and temperature in a wet PTFE tank of a wet processing equipment according to claim 5, characterized in that, Obtaining environmental adaptability monitoring cycle data includes the following steps: Based on the associated environmental interference factors, the influence weights of different interference factors on the baseline data of the initial monitoring period are analyzed. Establish a mapping relationship model between environmental disturbance factors and monitoring cycle adjustment coefficients; Real-time collection of environmental parameters, and calculation of the monitoring cycle adjustment coefficient under the current environment based on the mapping relationship model; The adjustment coefficient is applied to the initial monitoring cycle baseline data for dynamic correction, and upper and lower limits for cycle adjustment are set, with a minimum monitoring cycle of 5 minutes and a maximum monitoring cycle of 60 minutes. The prediction accuracy of the monitoring model reaches 90% as the termination condition for iterative optimization. After multiple iterations of optimization, the environmental adaptability monitoring cycle data is obtained.

7. The real-time monitoring system for dual parameters of stress and temperature in a wet PTFE tank of a wet processing equipment according to claim 6, characterized in that, The following steps are included to obtain a real-time monitoring model for the stress and temperature of the PTFE tank: Using environmental adaptability monitoring cycle data as input, a monitoring dataset containing both stress and temperature parameters is constructed; Preprocess the dataset, including data cleaning and normalization; The policy gradient algorithm is selected as the model building algorithm, and the initial parameters of the algorithm are set. The preprocessed dataset is divided into a training set and a test set; The policy gradient algorithm is trained using the training set, and the algorithm parameters are continuously adjusted to optimize model performance. Use the test set to validate the trained model and evaluate its accuracy and generalization ability; Based on the evaluation results, the model was fine-tuned to obtain a real-time monitoring model for the stress and temperature of the PTFE tank.

8. The real-time monitoring system for dual parameters of stress and temperature in a wet PTFE tank of a wet processing equipment according to claim 7, characterized in that, After obtaining the real-time monitoring model for the dual parameters of stress and temperature in the PTFE tank, in order to further quantify and evaluate the model's ability to provide early warning of anomalies in tank stress and temperature, an anomaly early warning quantitative evaluation formula is introduced: set up For the first The actual stress value at each time point For monitoring model in the first Stress values ​​predicted at specific time points; For the first The actual temperature value at each time point. For monitoring model in the first Predicted temperature values ​​at specific time points; The total number of time points selected for evaluation; and These are the weighting coefficients of stress and temperature in the anomaly early warning assessment, respectively. , , Its value is determined based on the relative importance of stress and temperature to the safety of the tank. If stress has a greater impact on the safety of the tank, then... The value is relatively large, and vice versa. The value is relatively large; Define stress prediction error Temperature prediction error ; Comprehensive anomaly early warning assessment indicators The calculation formula is: ； When comprehensive abnormal early warning assessment indicators Greater than the preset warning threshold When the system detects abnormal stress and temperature in the tank, it triggers an abnormality warning mechanism and sends warning information to relevant personnel so that timely measures can be taken to ensure the safe operation of the tank. in, and , and These represent the actual value and the predicted value, respectively, used to calculate the error. It involves assessing the number of time points to ensure the assessment is based on a sufficient amount of data. and These are weighting coefficients, which can be adjusted to reflect the relative importance of stress and temperature in the assessment, based on actual circumstances. It is a comprehensive anomaly early warning assessment indicator used to determine whether an early warning has been triggered. It is the warning threshold, which can be determined based on the tank's safety standards and historical data.

Images