Intelligent purity monitoring and feedback method and system for gas preparation

Abstract

The invention discloses an intelligent purity monitoring and feedback method and system for gas preparation, and belongs to the technical field of gas preparation monitoring. The method comprises the following steps: constructing a purity detection sensor distribution network and a data matrix of a gas container, collecting purity data at a time sequence operation node, calculating a gradient error in combination with adjacent sensor data to determine an accurate calibration range, generating a calibration matrix with a timestamp, and comparing historical and current calibration matrixes to calculate relevancy. Furthermore, the gas preparation stability is quantitatively evaluated. The problems that a traditional gas purity monitoring sensor is unreasonable in layout, low in calibration precision, lack of quantitative basis for stability evaluation and the like are solved, monitoring comprehensiveness, calibration accuracy and feedback timeliness are improved, the method is suitable for various gas preparation scenes such as industrial oxygen production and special gas synthesis, and scientific support is provided for product quality control.

Description

Technical Field

[0001] This invention relates to the field of gas preparation monitoring technology, specifically to an intelligent monitoring and feedback method and system for gas preparation purity. Background Technology

[0002] In industrial production, medical, and electronics manufacturing, the purity of gas products directly determines the quality of end products, the safety of the production process, and the stability of equipment operation. Therefore, purity monitoring and feedback during gas preparation are core control aspects. As gas preparation technology develops towards large-scale and refined operations, higher demands are placed on the accuracy, dynamism, and feedback efficiency of purity monitoring. However, existing gas purity monitoring technologies still face many critical issues that urgently need to be addressed, severely hindering the high-quality development of the gas preparation industry.

[0003] Traditional monitoring schemes often employ single-point or random sensor deployment, failing to consider the uniformity differences in gas diffusion within the gas container. This results in collected data that cannot comprehensively reflect the overall purity distribution within the container, easily leading to situations where localized purity deficiencies go undetected. While some schemes utilize multi-point deployment, they lack ordered coding and matrix management of the sensors, resulting in disorganized data storage, an inability to establish spatial relationships between collection points, and low efficiency in subsequent data processing.

[0004] Existing technologies mostly rely on fixed thresholds for purity calibration, failing to consider purity gradient changes between adjacent sampling points and ignoring the purity correlation between adjacent regions during gas diffusion. This results in significant deviations in the calibration range setting, making it impossible to accurately reflect the actual purity fluctuation range. Furthermore, the calibration process does not incorporate dynamic adjustments based on time-series data, and the fixed calibration standard is difficult to adapt to the dynamic characteristics of purity changes with reaction progress and environmental conditions during gas preparation.

[0005] Traditional methods often focus only on purity data at a single point in time, failing to establish correlation analysis between historical and current data. This makes it impossible to capture purity change trends, resulting in a lack of dynamic basis for judging the stability of the preparation process. Furthermore, stability assessments often rely on human experience and judgment, lacking quantitative calculation models. The assessment results are highly subjective and inaccurate, making it difficult to provide precise guidance for optimizing the preparation process.

[0006] The existing system's monitoring, calibration, and evaluation processes are independent of each other, and there is a lag in data transmission and processing. It cannot achieve rapid feedback from data acquisition to stability assessment and process adjustment suggestions. When abnormal fluctuations occur in purity, it is difficult to take timely control measures, which can easily lead to a large number of unqualified products and increase production costs.

[0007] The aforementioned problems result in existing gas purity monitoring systems having shortcomings such as narrow monitoring coverage, low calibration accuracy, unscientific stability assessment, and delayed feedback, failing to meet the refined purity control requirements of modern gas preparation. Therefore, developing a smart purity monitoring and feedback method and system with a reasonable sensor layout, accurate calibration, quantitative assessment, and timely feedback has become an urgent need in the current gas preparation industry. Summary of the Invention

[0008] The purpose of this invention is to provide a method and system for intelligent monitoring and feedback of gas purity in gas preparation, so as to solve the problems mentioned in the background art.

[0009] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0010] A purity intelligent monitoring and feedback system for gas preparation, comprising: a sensor network construction module, a calibration range calculation module, a current data processing module, and a stability assessment and feedback module;

[0011] The sensor network construction module is used to construct a purity monitoring sensor distribution network and a corresponding distribution network matrix for the gas container;

[0012] The calibration range calculation module is used to collect gas purity data at a set running time node, calculate gradient error and purity calibration range, and generate a purity calibration range matrix with timestamps.

[0013] The current data processing module is used to obtain the purity data at the current running time node and construct the corresponding current purity calibration range matrix;

[0014] The stability assessment feedback module is used to compare the historical and current purity calibration range matrices, calculate the purity calibration correlation, generate the purity monitoring correlation matrix, assess and feedback the gas preparation stability to the operator.

[0015] Preferably, the sensor network construction module includes a 3D model building unit, a sensor deployment coding unit, and a distribution matrix construction unit; the 3D model building unit is used to build a 3D solid image of the gas container and plan the horizontal and vertical equidistant deployment scheme of the purity monitoring sensors; the sensor deployment coding unit is used to uniformly code all deployed purity monitoring sensors and clarify the horizontal and vertical serial number identifier of each purity monitoring sensor; the distribution matrix construction unit constructs a distribution network matrix based on the sensor coding attributes and records the gas purity data collected by each purity monitoring sensor to the specified position in the distribution network matrix.

[0016] Preferably, the calibration range calculation module includes a timestamp initialization unit, a gradient error calculation unit, a calibration range determination unit, and a calibration matrix generation unit; the timestamp initialization unit is used to set a unified operating timestamp for each purity monitoring sensor and divide the operating time nodes; the gradient error calculation unit is used to calculate the gradient error of each purity monitoring sensor based on the gas purity data of adjacent purity monitoring sensors at each operating time node; the calibration range determination unit is used to determine the upper and lower limits of the purity calibration range of each purity monitoring sensor according to the gradient error; the calibration matrix generation unit is used to record the purity calibration range into the distribution network matrix and convert it into a purity calibration range matrix with timestamps.

[0017] Preferably, the current data processing module includes a current data acquisition unit and a current calibration matrix construction unit; the current data acquisition unit is used to instruct each purity monitoring sensor to acquire gas purity data at the current running time node; the current calibration matrix construction unit is used to calculate the gradient error and purity calibration range of each purity monitoring sensor at the current running time node, and generate a purity calibration range matrix for the current running time node.

[0018] Preferably, the stability assessment feedback module includes a calibration range comparison unit, a correlation calculation unit, a correlation matrix generation unit, a stability assessment unit, and a result feedback unit. The calibration range comparison unit is used to correlate and compare corresponding calibration ranges in the historical and current purity calibration range matrices. The correlation calculation unit assesses the purity calibration correlation of each sensor based on the comparison results. The correlation matrix generation unit converts the distribution network matrix into a purity monitoring correlation matrix based on the purity calibration correlation results. The stability assessment unit calculates the gas preparation stability based on the purity monitoring correlation matrix. The result feedback unit outputs the gas preparation stability assessment result.

[0019] A method for intelligent monitoring and feedback of purity in gas preparation, comprising the following steps:

[0020] Step S1: Construct the purity monitoring sensor distribution network and corresponding distribution network matrix for the gas container; next; then; finally;

[0021] Step S2: Collect gas purity data from each purity monitoring sensor at the set running time nodes, calculate the gradient error of each purity monitoring sensor and determine the corresponding purity calibration range, forming a purity calibration range matrix with running timestamp attributes.

[0022] Step S3: Obtain the gas purity data at the current running time node and generate the corresponding purity calibration range matrix for the current running time node;

[0023] Step S4: Compare the purity calibration range matrix of historical operating time nodes with that of the current operating time node, calculate the purity calibration correlation of each purity monitoring sensor and generate a purity monitoring correlation matrix, evaluate the gas preparation stability based on the purity monitoring correlation matrix and feed it back to the operator port.

[0024] Preferably, the specific implementation process of step S1 includes:

[0025] A three-dimensional model of the gas container is created, and purity monitoring sensors are set up with equal spacing in the horizontal and vertical directions to form a distribution network of purity monitoring sensors.

[0026] The deployed purity monitoring sensors are uniformly coded. The sequential number of any purity monitoring sensor in the horizontal direction is denoted as y, and its sequential number in the vertical direction is denoted as x. This purity monitoring sensor is then denoted as... ;

[0027] Based on the coding attributes of purity monitoring sensors, a distribution network matrix for purity monitoring sensors is constructed. And the row index of the distribution network matrix is ​​x, the column index is y, and the purity monitoring sensor The purity of the collected gas is recorded in the matrix position of the x-th row and y-th column of the distribution network matrix;

[0028] It should be noted that this invention establishes a three-dimensional model of the gas container, uses sensors arranged at equal intervals in the horizontal and vertical directions and performs unified coding to construct a distributed network matrix, enabling the collected data to have spatial correlation, realizing all-round, blind-angle monitoring of the purity inside the container, solving the problem of insufficient data representativeness in traditional solutions, and improving the comprehensiveness of data coverage.

[0029] Preferably, the specific implementation process of step S2 includes:

[0030] Initialize a unified operating timestamp for each purity monitoring sensor. At the t-th operating time node, instruct each purity monitoring sensor to collect gas purity data once and record it in the distribution network matrix, forming a distribution network matrix with operating timestamp attributes. The timestamp distribution network matrix generated at the t-th operating time node is denoted as... ;

[0031] In timestamp distribution network matrix In China, for purity monitoring sensors Evaluation of purity monitoring sensors gradient error In the formula, , and These are, in order, purity monitoring sensors. , and The gas purity collected at the t-th running time node;

[0032] Based on gradient error Quantitative purity monitoring sensor Purity calibration range at the t-th running time node And the lower limit of the purity calibration range Upper limit of purity calibration range ;

[0033] Purity calibration range Recorded in the distributed network matrix In the matrix position at row x and column y, the distribution network matrix is... Transform into a purity calibration range matrix, denoted as ;

[0034] It should be noted that this invention dynamically determines the purity calibration range of each sensor by calculating the purity gradient error between each sensor and its adjacent sensors. It takes into account the spatial correlation of gas diffusion, and the calibration range is more in line with the actual purity fluctuation law, so as to effectively reduce the calibration error and improve the accuracy of purity detection.

[0035] Preferably, the specific implementation process of step S3 includes:

[0036] The gas purity data collected by each purity monitoring sensor at the current operating time point is obtained and recorded in the distribution network matrix to form the timestamp distribution network matrix for the current operating time point. T is the sequence number of the current running time node;

[0037] The gradient error and purity calibration range of each purity monitoring sensor at the current operating time point are evaluated to form a purity calibration range matrix for the current operating time point. .

[0038] Preferably, the specific implementation process of step S4 includes:

[0039] Based on the purity calibration range matrix, for purity monitoring sensors The purity calibration range matrix The purity calibration range matrix in the current running time node. Comparison of purity calibration ranges:

[0040] For the purity calibration range matrix The lower and upper limits of the purity calibration range are respectively marked with source labels, and are labeled as follows: and ;

[0041] Purity calibration range matrix derived from the current running time node The lower and upper limits of the purity calibration range are respectively marked with source labels, and are labeled as follows: and ;

[0042] Based on the t-th running time node and the current running time node T, evaluate the purity monitoring sensor. Relevance of purity calibration:

[0043] ;

[0044] In the formula, max{} and min{} are the symbols for the maximum and minimum values, respectively;

[0045] If purity calibration correlation This indicates that the purity monitoring sensor at the t-th running time node and the current running time node T... There is no correlation with purity calibration;

[0046] If purity calibration correlation ,and This indicates that the purity monitoring sensor at the t-th running time node and the current running time node T... There is a purity calibration correlation. SS represents the set of sensors that make up the purity monitoring sensors, and num(SS) represents the total number of purity monitoring sensors contained in the set of sensors SS.

[0047] Based on the qualitative purity calibration correlation of the comparison results, for comparison results without purity calibration correlation, let the distribution network matrix... The matrix position in row x and column y is 0. For comparison results that are correlated with purity calibration, let the distribution network matrix... If the matrix position in row x and column y is 1, then the distribution network matrix will be... Transformed into a purity monitoring correlation matrix ;

[0048] Assess the stability of gas preparation at the current operating time point. In the formula, Represents the correlation matrix for purity monitoring Summation, ;

[0049] Feedback output monitors the stability of the gas preparation process;

[0050] It should be noted that this invention constructs a matrix of historical and current calibration ranges with timestamps, calculates the correlation between purity calibration and generates a correlation matrix, and transforms stability assessment into a quantifiable mathematical model, replacing traditional manual experience-based judgment.

[0051] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: This invention realizes intelligent monitoring and feedback of gas purity based on a collaborative mechanism of "spatial matrix layout + temporal dynamic calibration + correlation quantification evaluation". First, a sensor distribution network matrix is ​​constructed from the spatial dimension to establish spatial correlation between each collection point, ensuring the comprehensiveness and orderliness of the data. The matrix layout solves the limitations of single-point monitoring, enabling data to cover all key areas within the container. Second, temporal purity data is collected from the temporal dimension. The gradient error is calculated using the spatial correlation between adjacent sensors to dynamically determine the calibration range, making the calibration standard adaptable to the spatial and temporal fluctuation characteristics of purity. Gradient error-driven dynamic calibration avoids the rigidity of fixed thresholds and can adapt to purity fluctuations under different operating conditions. Finally, by comparing the correlation between historical and current calibration matrices, the correlation is calculated to quantify the purity stability of each collection point. Then, based on the correlation matrix, the overall preparation stability is quantitatively evaluated, forming a closed-loop logic of "collection-calibration-comparison-evaluation-feedback". Attached Figure Description

[0052] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0053] Figure 1 This is a schematic diagram illustrating the steps of a method for intelligent monitoring and feedback of gas purity in gas preparation according to the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] In this first embodiment: a purity intelligent monitoring and feedback system for gas preparation is provided. The system includes: a sensor network construction module, a calibration range calculation module, a current data processing module, and a stability evaluation feedback module.

[0056] The sensor network construction module is used to construct the purity monitoring sensor distribution network and the corresponding distribution network matrix for gas containers.

[0057] The sensor network construction module includes a 3D model building unit, a sensor deployment and coding unit, and a distribution matrix construction unit. The 3D model building unit is used to create a 3D model of the gas container and plan the horizontal and vertical equidistant deployment scheme of the purity monitoring sensors. The sensor deployment and coding unit is used to uniformly code all deployed purity monitoring sensors and clarify the horizontal and vertical serial number identifiers of each purity monitoring sensor. The distribution matrix construction unit constructs a distribution network matrix based on the sensor coding attributes and records the gas purity data collected by each purity monitoring sensor to the specified positions in the distribution network matrix.

[0058] The calibration range calculation module is used to collect gas purity data at set running time nodes, calculate gradient error and purity calibration range, and generate a purity calibration range matrix with timestamps.

[0059] The calibration range calculation module includes a timestamp initialization unit, a gradient error calculation unit, a calibration range determination unit, and a calibration matrix generation unit. The timestamp initialization unit sets a unified operating timestamp for each purity monitoring sensor and divides the operating time nodes. The gradient error calculation unit calculates the gradient error of each purity monitoring sensor based on the gas purity data of adjacent purity monitoring sensors at each operating time node. The calibration range determination unit determines the upper and lower limits of the purity calibration range for each purity monitoring sensor based on the gradient error. The calibration matrix generation unit records the purity calibration range into a distribution network matrix and transforms it into a timestamp-enabled purity calibration range matrix.

[0060] The current data processing module is used to obtain the purity data at the current running time point and construct the corresponding current purity calibration range matrix;

[0061] The current data processing module includes a current data acquisition unit and a current calibration matrix construction unit. The current data acquisition unit is used to instruct each purity monitoring sensor to acquire gas purity data at the current running time node. The current calibration matrix construction unit is used to calculate the gradient error and purity calibration range of each purity monitoring sensor at the current running time node, and generate the purity calibration range matrix for the current running time node.

[0062] The stability assessment feedback module is used to compare the historical and current purity calibration range matrices, calculate the purity calibration correlation, generate the purity monitoring correlation matrix, assess and provide feedback on the gas preparation stability to the operators.

[0063] The stability assessment feedback module includes a calibration range comparison unit, a correlation calculation unit, a correlation matrix generation unit, a stability assessment unit, and a result feedback unit. The calibration range comparison unit compares the corresponding calibration ranges in the historical and current purity calibration range matrices. The correlation calculation unit assesses the purity calibration correlation of each sensor based on the comparison results. The correlation matrix generation unit converts the distribution network matrix into a purity monitoring correlation matrix based on the purity calibration correlation results. The stability assessment unit calculates the gas preparation stability based on the purity monitoring correlation matrix. The result feedback unit outputs the gas preparation stability assessment results.

[0064] Please see Figure 1 In this second embodiment, a method for intelligent monitoring and feedback of gas purity is provided, applicable to the first embodiment above. This embodiment uses the vaporization of liquid oxygen in an industrial oxygen plant to prepare high-purity oxygen (purity requirement ≥99.5%). The gas container is a cylindrical high-pressure storage tank (3 meters in diameter and 8 meters in height). The preparation process is low-temperature vaporization of liquid oxygen - pressure regulation - purity filtration. It is necessary to monitor the oxygen purity in the storage tank in real time and evaluate the stability of the preparation process. When the stability is lower than 0.7, a process adjustment alarm is triggered.

[0065] The method includes the following steps:

[0066] Step S1: Construct the purity monitoring sensor distribution network and corresponding distribution network matrix for the gas container;

[0067] For example, a three-dimensional graphic of a gas container is created, and purity monitoring sensors are set up at equal intervals in the horizontal direction and vertical direction to form a distribution network of purity monitoring sensors.

[0068] The deployed purity monitoring sensors are uniformly coded. The sequential number of any purity monitoring sensor in the horizontal direction is denoted as y, and its sequential number in the vertical direction is denoted as x. This purity monitoring sensor is then denoted as... ;

[0069] Based on the coding attributes of purity monitoring sensors, a distribution network matrix for purity monitoring sensors is constructed. And the row index of the distribution network matrix is ​​x, and the column index is y, which will be used for the purity monitoring sensor. The purity of the collected gas is recorded in the matrix position of row x and column y of the distribution network matrix;

[0070] For example, a 3D model of the storage tank is created using CAD software, with equal spacing of 0.6 meters horizontally (around the tank circumference) and 0.8 meters vertically (up to the tank height). Five sensors are arranged horizontally (numbered y=1-5) and ten sensors are arranged vertically (numbered x=1-10), for a total of 50 purity detection sensors. Each sensor is uniformly coded (e.g., the third vertical sensor and the second horizontal sensor are denoted as S32), and a 10-row, 5-column distribution network matrix O is constructed. The x-th row and y-th column of the matrix records S. x ᵧ collected data.

[0071] Step S2: Collect gas purity data from each purity monitoring sensor at the set running time nodes, calculate the gradient error of each purity monitoring sensor and determine the corresponding purity calibration range, forming a purity calibration range matrix with running timestamp attributes.

[0072] For example, a unified operating timestamp is initialized for each purity monitoring sensor. At the t-th operating time node, each purity monitoring sensor is instructed to collect gas purity data once and record it in the distribution network matrix, forming a distribution network matrix with operating timestamp attributes. The timestamp distribution network matrix generated at the t-th operating time node is denoted as... ;

[0073] In timestamp distribution network matrix In China, for purity monitoring sensors Evaluation of purity monitoring sensors gradient error In the formula, , and These are, in order, purity monitoring sensors. , and The gas purity collected at the t-th running time node;

[0074] Based on gradient error Quantitative purity monitoring sensor Purity calibration range at the t-th running time node And the lower limit of the purity calibration range Upper limit of purity calibration range ;

[0075] Purity calibration range Recorded in the distributed network matrix In the matrix position at row x and column y, the distribution network matrix is... Transform into a purity calibration range matrix, denoted as ;

[0076] For example, the running time nodes are set to one every 15 minutes (t=1, 2, ..., 20, for a total of 5 hours); at each node from t=1 to t=19, the sensor collects purity data (e.g., S32 collects a value of 99.62% at t=5, the adjacent S33 collects a value of 99.58%, and S42 collects a value of 99.60%).

[0077] Step S3: Obtain the gas purity data at the current running time node and generate the corresponding purity calibration range matrix for the current running time node;

[0078] For example, the gas purity collected by each purity monitoring sensor at the current operating time point is obtained and recorded in the distribution network matrix to form the timestamp distribution network matrix of the current operating time point. T is the sequence number of the current running time node;

[0079] The gradient error and purity calibration range of each purity monitoring sensor at the current operating time point are evaluated to form a purity calibration range matrix for the current operating time point. .

[0080] Step S4: Compare the purity calibration range matrix of historical operating time nodes with that of the current operating time node, calculate the purity calibration correlation of each purity monitoring sensor and generate a purity monitoring correlation matrix, evaluate the gas preparation stability based on the purity monitoring correlation matrix and feed it back to the operator port;

[0081] For example, based on the purity calibration range matrix, for purity monitoring sensors The purity calibration range matrix The purity calibration range matrix in the current running time node. Comparison of purity calibration ranges:

[0082] For the purity calibration range matrix The lower and upper limits of the purity calibration range are respectively marked with source labels, and are labeled as follows: and ;

[0083] Purity calibration range matrix derived from the current running time node The lower and upper limits of the purity calibration range are respectively marked with source labels, and are labeled as follows: and ;

[0084] Based on the t-th running time node and the current running time node T, evaluate the purity monitoring sensor. Relevance of purity calibration:

[0085] ;

[0086] In the formula, max{} and min{} are the symbols for the maximum and minimum values, respectively;

[0087] If purity calibration correlation This indicates that the purity monitoring sensor at the t-th running time node and the current running time node T... There is no correlation with purity calibration;

[0088] If purity calibration correlation ,and This indicates that the purity monitoring sensor at the t-th running time node and the current running time node T... There is a purity calibration correlation. SS represents the set of sensors that make up the purity monitoring sensors, and num(SS) represents the total number of purity monitoring sensors contained in the set of sensors SS.

[0089] Based on the qualitative purity calibration correlation of the comparison results, for comparison results without purity calibration correlation, let the distribution network matrix... The matrix position in row x and column y is 0. For comparison results that are correlated with purity calibration, let the distribution network matrix... If the matrix position in row x and column y is 1, then the distribution network matrix will be... Transformed into a purity monitoring correlation matrix ;

[0090] Assess the stability of gas preparation at the current operating time point. In the formula, Represents the correlation matrix for purity monitoring Summation, ;

[0091] Feedback output monitors the stability of gas preparation.

[0092] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0093] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent monitoring and feedback of purity in gas preparation, characterized in that, The method includes the following steps: Step S1: Construct the purity monitoring sensor distribution network and corresponding distribution network matrix for the gas container; Step S2: Collect gas purity data from each purity monitoring sensor at the set running time nodes, calculate the gradient error of each purity monitoring sensor and determine the corresponding purity calibration range, forming a purity calibration range matrix with running timestamp attributes. Step S3: Obtain the gas purity data at the current running time node and generate the corresponding purity calibration range matrix for the current running time node; Step S4: Compare the purity calibration range matrix of historical operating time nodes with that of the current operating time node, calculate the purity calibration correlation of each purity monitoring sensor and generate a purity monitoring correlation matrix, evaluate the gas preparation stability based on the purity monitoring correlation matrix and feed it back to the operator port.

2. The intelligent purity monitoring and feedback method for gas preparation according to claim 1, characterized in that, The specific implementation process of step S1 includes: A three-dimensional model of the gas container is created, and purity monitoring sensors are set up with equal spacing in the horizontal and vertical directions to form a distribution network of purity monitoring sensors. The deployed purity monitoring sensors are uniformly coded. The sequential number of any purity monitoring sensor in the horizontal direction is denoted as y, and its sequential number in the vertical direction is denoted as x. This purity monitoring sensor is then denoted as... ; Based on the coding attributes of purity monitoring sensors, a distribution network matrix for purity monitoring sensors is constructed. And the row index of the distribution network matrix is ​​x, the column index is y, and the purity monitoring sensor The purity of the collected gas is recorded in the matrix position of the x-th row and y-th column of the distribution network matrix.

3. The intelligent purity monitoring and feedback method for gas preparation according to claim 2, characterized in that, The specific implementation process of step S2 includes: Initialize a unified operating timestamp for each purity monitoring sensor. At the t-th operating time node, instruct each purity monitoring sensor to collect gas purity data once and record it in the distribution network matrix, forming a distribution network matrix with operating timestamp attributes. The timestamp distribution network matrix generated at the t-th operating time node is denoted as... ; In timestamp distribution network matrix In China, for purity monitoring sensors Evaluation of purity monitoring sensors gradient error In the formula, , and These are, in order, purity monitoring sensors. , and The gas purity collected at the t-th running time node; Based on gradient error Quantitative purity monitoring sensor Purity calibration range at the t-th running time node And the lower limit of the purity calibration range Upper limit of purity calibration range ; Purity calibration range Recorded in the distributed network matrix In the matrix position at row x and column y, the distribution network matrix is... Transform into a purity calibration range matrix, denoted as .

4. The intelligent purity monitoring and feedback method for gas preparation according to claim 3, characterized in that, The specific implementation process of step S3 includes: The gas purity data collected by each purity monitoring sensor at the current operating time point is obtained and recorded in the distribution network matrix to form the timestamp distribution network matrix for the current operating time point. T is the sequence number of the current running time node; The gradient error and purity calibration range of each purity monitoring sensor at the current operating time point are evaluated to form a purity calibration range matrix for the current operating time point. .

5. The intelligent purity monitoring and feedback method for gas preparation according to claim 3, characterized in that, The specific implementation process of step S4 includes: Based on the purity calibration range matrix, for purity monitoring sensors The purity calibration range matrix The purity calibration range matrix in the current running time node. Comparison of purity calibration ranges: For the purity calibration range matrix The lower and upper limits of the purity calibration range are respectively marked with source labels, and are labeled as follows: and ; Purity calibration range matrix derived from the current running time node The lower and upper limits of the purity calibration range are respectively marked with source labels, and are labeled as follows: and ; Based on the t-th running time node and the current running time node T, evaluate the purity monitoring sensor. Relevance of purity calibration: ； In the formula, max{} and min{} are the symbols for the maximum and minimum values, respectively; If purity calibration correlation This indicates that the purity monitoring sensor at the t-th running time node and the current running time node T... There is no correlation with purity calibration; If purity calibration correlation ,and This indicates that the purity monitoring sensor at the t-th running time node and the current running time node T... There is a purity calibration correlation. SS represents the set of sensors that make up the purity monitoring sensors, and num(SS) represents the total number of purity monitoring sensors contained in the set of sensors SS. Based on the qualitative purity calibration correlation of the comparison results, for comparison results without purity calibration correlation, let the distribution network matrix... The matrix position in row x and column y is 0. For comparison results that are correlated with purity calibration, let the distribution network matrix... If the matrix position in row x and column y is 1, then the distribution network matrix will be... Transformed into a purity monitoring correlation matrix ; Assess the stability of gas preparation at the current operating time point. In the formula, Represents the correlation matrix for purity monitoring Summation, ; The feedback output monitors the stability of the gas preparation process.

6. A purity intelligent monitoring and feedback system for gas preparation, executing the purity intelligent monitoring and feedback method for gas preparation as described in any one of claims 1-5, characterized in that, The system includes: a sensor network construction module, a calibration range calculation module, a current data processing module, and a stability evaluation feedback module; The sensor network construction module is used to construct a purity monitoring sensor distribution network and a corresponding distribution network matrix for the gas container; The calibration range calculation module is used to collect gas purity data at a set running time node, calculate gradient error and purity calibration range, and generate a purity calibration range matrix with timestamps. The current data processing module is used to obtain the purity data at the current running time node and construct the corresponding current purity calibration range matrix; The stability assessment feedback module is used to compare the historical and current purity calibration range matrices, calculate the purity calibration correlation, generate the purity monitoring correlation matrix, assess and feedback the gas preparation stability to the operator.

7. The intelligent purity monitoring and feedback system for gas preparation according to claim 6, characterized in that, The sensor network construction module includes a 3D model building unit, a sensor deployment coding unit, and a distribution matrix construction unit. The 3D model building unit is used to create a 3D model of the gas container and plan the horizontal and vertical equidistant deployment scheme of the purity monitoring sensors. The sensor deployment coding unit is used to uniformly code all deployed purity monitoring sensors and clarify the horizontal and vertical serial number identifiers of each purity monitoring sensor. The distribution matrix construction unit constructs a distribution network matrix based on the sensor coding attributes and records the gas purity data collected by each purity monitoring sensor to the specified positions in the distribution network matrix.

8. The intelligent purity monitoring and feedback system for gas preparation according to claim 6, characterized in that, The calibration range calculation module includes a timestamp initialization unit, a gradient error calculation unit, a calibration range determination unit, and a calibration matrix generation unit. The timestamp initialization unit is used to set a unified operating timestamp for each purity monitoring sensor and to divide the operating time nodes; The gradient error calculation unit is used to calculate the gradient error of each purity monitoring sensor based on the gas purity data of adjacent purity monitoring sensors at each running time node; the calibration range determination unit is used to determine the upper and lower limits of the purity calibration range of each purity monitoring sensor according to the gradient error. The calibration matrix generation unit is used to record the purity calibration range into the distribution network matrix and transform it into a purity calibration range matrix with timestamps.

9. The intelligent purity monitoring and feedback system for gas preparation according to claim 6, characterized in that, The current data processing module includes a current data acquisition unit and a current calibration matrix construction unit; the current data acquisition unit is used to instruct each purity monitoring sensor to acquire gas purity data at the current running time node; The current calibration matrix construction unit is used to calculate the gradient error and purity calibration range of each purity monitoring sensor at the current running time node, and generate the purity calibration range matrix for the current running time node.

10. The intelligent purity monitoring and feedback system for gas preparation according to claim 6, characterized in that, The stability assessment feedback module includes a calibration range comparison unit, a correlation calculation unit, a correlation matrix generation unit, a stability assessment unit, and a result feedback unit. The calibration range comparison unit is used to compare the corresponding calibration ranges in the historical and current purity calibration range matrices. The correlation calculation unit evaluates the purity calibration correlation of each sensor based on the comparison results. The correlation matrix generation unit is used to convert the distribution network matrix into a purity monitoring correlation matrix based on the purity calibration correlation results. The stability assessment unit calculates the gas preparation stability based on the purity monitoring correlation matrix; The result feedback unit is used to output the gas preparation stability assessment result.

Images