Online Automatic Calibration and Control Method of Moisture Meter

Abstract

The present invention discloses an on-line automatic calibration and control method for a moisture meter, which is characterized in that it includes S1 moisture meter channel issuance, S2 comparative analysis and early warning of moisture meter parameters, S3 measurement system analysis and early warning, offset and linear analysis, repeatability and reproducibility analysis, and S4 automatic calibration of the moisture meter. The present invention conducts comparative analysis on parameters such as the zero position, span, and damping of the wire-making moisture meter, analyzes and gives early warning of the offset monitoring through regression analysis, and applies measurement system analysis such as repeatability and reproducibility to establish an on-line automatic calibration and automatic control system and method for the wire-making moisture meter, improving the accuracy of moisture meter measurement and being beneficial to improving the stability of the moisture content in the wire-making process.

Description

Technical Field

[0001] The present invention relates to the field of moisture meter calibration, and particularly to an on-line automatic calibration and control method for a moisture meter. Background Art

[0002] The moisture content of cut tobacco is a key index, which directly affects the physical and sensory qualities of cigarettes. Therefore, the control of the moisture content during the cut tobacco processing is crucial. Currently, it is mainly measured by a moisture meter and participates in the process control.

[0003] The applications of moisture meters in the cut tobacco processing include off-line moisture meters and on-line moisture meters. Among them, on-line moisture meters are mainly used to measure the moisture content of materials at different process points. The more commonly used ones are infrared moisture meters and microwave moisture meters. Moisture meters can be used for moisture control at various points in the cut tobacco process, such as the moisture control at the outlet of the feeding and humidifying section, the moisture control at the outlet of the cut tobacco humidifying section, the moisture control at the outlet of the cut tobacco drying section, etc. Therefore, in order to achieve precise control of the moisture content during the cut tobacco process, ensuring the accuracy of the moisture meter measurement is the prerequisite. However, in the actual production process, there is usually a large deviation between the measured value of the moisture meter and the actual moisture content of the material, resulting in abnormal situations. Usually, it relies on manual inspection and comparison, and manual adjustment, which leads to problems such as adjustment lag and high labor intensity.

[0004] For the above reasons, in order to improve the accuracy of moisture meter measurement, it is desirable to propose an on-line automatic calibration, automatic control system and method for cut tobacco moisture meters. Summary of the Invention

[0005] In order to overcome the above problems, the present invention provides an on-line automatic calibration and control method for a moisture meter. By real-time comparison of on-line data of the moisture meter, as well as automatic comparison and analysis of parameters, regression analysis warning and measurement system analysis, the present invention establishes an on-line automatic calibration and automatic control method for cut tobacco moisture meters to improve the accuracy of moisture meter measurement.

[0006] The technical solution adopted by the present invention to solve its technical problems is as follows:

[0007] An on-line automatic calibration and control method for a moisture meter, comprising

[0008] S1 Moisture meter channel issuance

[0009] Before the cut tobacco production, according to the brand of cigarettes to be produced, production instructions are issued through the central control system, and at the same time, moisture meter channels are issued;

[0010] Each moisture meter channel corresponds to a brand of cigarettes. The moisture meter channel includes: the zero position, span and damping parameters of each process moisture meter, so as to achieve the issuance of moisture meter channels for each on-site process and the establishment of formula parameters;

[0011] Comparative Analysis and Early Warning of S2 Moisture Meter Parameters

[0012] Automatically compare and analyze the moisture meter channels and parameters issued by the central control system, including the comparison and analysis of the grades to be produced, channel codes, zero positions, spans, and damping parameters;

[0013] If the above information is completely consistent, issue an instruction to start formal production;

[0014] If the grade, channel, or a certain parameter does not match, issue an early warning, and the workshop metrology personnel shall conduct on-site inspections and adjustments according to the early warning situation;

[0015] In this step, by automatically comparing and analyzing the moisture meter channels and parameters issued by the central control system, the anti-error method can effectively prevent the occurrence of quality problems where the moisture meter executes inconsistently with the actual situation;

[0016] S3 Measurement System Analysis and Early Warning

[0017] S3.1 Bias and Linearity Analysis

[0018] Conduct a linear regression analysis on the moisture content data detected by the moisture meter for each process;

[0019] When the P value ≥ 0.05, it indicates that the overall bias of the moisture content data is not significant and the moisture meter measures accurately;

[0020] When the P value < 0.05, it indicates that the overall moisture content data shows a bias, then issue an early warning, and correct it according to the average value of the moisture content of this batch to ensure the accuracy of the moisture meter measurement data;

[0021] S3.2 Repeatability and Reproducibility Analysis

[0022] Use the analysis of variance method to analyze the sources of moisture meter measurement fluctuations. The sources of measurement fluctuations include three categories: tobacco leaves or cut tobacco, moisture meters, and the interaction between tobacco leaves or cut tobacco and moisture meters, and the three are independent of each other;

[0023] Y ij = μ + L i + M + (LM) ij ;

[0024] In the formula:

[0025] i -- 1, 2,..., I;

[0026] j -- 1, 2,..., J;

[0027] Y ij -- The moisture content data of the jth test on the ith batch of tobacco leaf or cut tobacco samples by the moisture meter;

[0028] μ -- Unknown constant;

[0029] L i 、M、(LM) ij -- Fluctuations caused by tobacco leaves or cut tobacco, moisture meters, and the interaction between tobacco leaves or cut tobacco and moisture meters; repeatability error

[0030] Using the analysis of variance method, respectively using σ 2 M 、σ 2 L 、σ 2 LM to represent the fluctuations caused by tobacco leaves or cut tobacco, moisture meters, and the interaction between tobacco leaves or cut tobacco and moisture meters, the mathematical expressions for the repeatability error and reproducibility error of the moisture meter measurement system are respectively:

[0031] σ 2 repe =σ 2 M ,σ 2 repr =σ 2 L +σ 2 LM ;

[0032] In the formula:

[0033] σ 2 repe -- Repeatability error of the measurement system;

[0034] σ 2 repr -- Reproducibility error of the measurement system;

[0035] σ 2 MS =σ 2 repe +σ 2 repr ;

[0036] In the formula: σ MS -- Standard deviation of the measurement system fluctuation;

[0037] G = σ MS / σ T ,R = 6σ MS / (USL - LSL),

[0038] In the formula:

[0039] G-- Percentage of the measurement system fluctuation in the overall process fluctuation;

[0040] R-- Percentage of the measurement system accuracy in the tolerance;

[0041] ndc--Number of distinguishable categories of the measurement system;

[0042] σ T --Total process variation of the measurement system;

[0043] USL--Upper standard limit of the moisture content of each process;

[0044] LSL--Lower standard limit of the moisture content of each process;

[0045] When both G and R are less than 10%, and ndc ≥ 5, it indicates that the measurement system is good and has sufficient resolution;

[0046] When either G or R is ≥ 10%, or ndc < 5, it indicates that the measurement system needs to be improved or has insufficient resolution. Then, a warning is issued and calibration is performed;

[0047] S4 Automatic calibration of the moisture meter

[0048] Select at least 3 batches of data with a linear regression P-value ≥ 0.05 for the moisture content data of the same brand and the same process, both G and R less than 10%, and ndc ≥ 5 as the basic data, and take the average of the three batches of data as the basis for calibration, and automatically calculate the deviation between the two;

[0049] Adjust the zero position of the moisture meter through the central control system and reissue it. If both G and R of the measurement data of this batch are less than 10% and ndc ≥ 5, the calibration is successful;

[0050] S5 Batch similarity analysis and warning

[0051] Statistically analyze and record the moisture content change of cut tobacco in each batch, and establish a historical record database of cut tobacco moisture content by brand and process;

[0052] During batch production, compare with the historical best level and use the similarity analysis method for result analysis. If the similarity between the two is high, it indicates that the moisture meter detection of this batch of cigarettes is good and accurate; if the similarity between the two is low, it indicates that there is a certain deviation in the moisture content of this batch;

[0053] Based on the deviation degree of this batch from the historical best level, issue a pre-production warning for the next batch. The metrology personnel ensure the accuracy of the moisture meter detection in the next batch of cut tobacco production process and the stability control of the cut tobacco quality by timely adjusting other parameters such as the zero position of the moisture meter.

[0054] Beneficial effects brought by the present invention:

[0055] The present invention analyzes the comparison of parameters such as the zero position, span, and damping of the wire-making moisture meter, analyzes and warns of the deviation through regression analysis, and applies measurement system analysis means such as repeatability and reproducibility to establish an on-line automatic calibration, automatic control system and method for the wire-making moisture meter, improve the accuracy of the moisture meter measurement, and is conducive to improving the stability of the moisture content in the wire-making process. Description of the Drawings

[0056] The present invention will be further described below in conjunction with the drawings and specific embodiments:

[0057] Figure 1 It is the automatic control flow chart of the moisture meter before production of the present invention;

[0058] Figure 2 It is the automatic control flow chart of the moisture meter during the production process of the present invention;

[0059] Figure 3 It is the similarity analysis and warning flow chart of the moisture meter of the present invention. Specific Embodiments

[0060] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0061] Embodiment 1

[0062] Refer to Figures 1 to 3 , an on-line automatic calibration and control method for a moisture meter, which controls the moisture content at the outlet of the loose rewetting of a certain brand of cigarette, including

[0063] S1 Issuance of the moisture meter channel

[0064] Before the wire-making production, according to the cigarette brand to be produced, production instructions are issued through the central control system, and at the same time, the moisture meter channel code 1 is issued;

[0065] Each moisture meter channel corresponds to a cigarette brand. The moisture meter channel includes: the zero position, span, and damping parameters of each process moisture meter, so as to achieve the issuance of each process moisture meter channel on-site and the establishment of formula parameters;

[0066] For example, the zero position of the moisture meter at the outlet of the loose rewetting of a certain brand of cigarette is -17.0, the span is 25, and the damping is 13;

[0067] S2 Comparative analysis and warning of moisture meter parameters

[0068] Automatically compare and analyze with the moisture meter channels and parameters issued by the central control system. If the information of brand, channel code, zero position, span, and damping parameters is consistent, issue an instruction to start formal production;

[0069] If the brand, channel, or a certain parameter does not match, give an early warning, and the workshop metrology personnel will conduct on-site inspections and adjustments according to the warning situation;

[0070] In this step, by automatically comparing and analyzing with the moisture meter channels and parameters issued by the central control system, the anti-error method can effectively prevent the occurrence of quality problems where the moisture meter executes inconsistently with the actual situation;

[0071] S3 Measurement System Analysis and Early Warning

[0072] S3.1 Bias and Linearity Analysis

[0073] Conduct a linear regression analysis on the moisture meter detection data of the outlet moisture content of the cut tobacco for this brand during the loosening and rewetting process;

[0074] When the P-value ≥ 0.05, it indicates that the overall offset of the moisture content data is not significant, and the moisture meter measurement is accurate;

[0075] When the P-value < 0.05, it indicates that the overall moisture content data shows an offset, then give an early warning, and correct it according to the average value of the moisture content of this batch to ensure the accuracy of the moisture meter measurement data;

[0076] Linear regression analysis:

[0077] If the P-value is 0.065 and the P-value ≥ 0.05, it indicates that the overall offset of the moisture content data is not significant, and the moisture meter measurement is accurate.

[0078] S3.2 Repeatability and Reproducibility Analysis

[0079] Use the analysis of variance method to analyze the sources of measurement fluctuations of the moisture meter for the outlet moisture content of the cut tobacco during the loosening and rewetting process of this brand. The sources of measurement fluctuations include three categories: tobacco leaves or cut tobacco, moisture meter, and the interaction between tobacco leaves or cut tobacco and the moisture meter, and the three are independent of each other;

[0080] Respectively use σ 2 M 、σ 2 L 、σ 2 LM to represent the fluctuations caused by tobacco leaves or cut tobacco, moisture meter, and the interaction between tobacco leaves or cut tobacco and the moisture meter: After calculation, it is obtained that

[0081] σ 2 repe =σ 2 M =0.0009, σ 2repr = σ 2 L + σ 2 LM = 0.0007;

[0082] Where:

[0083] σ 2 repe -- Repeatability error of the measurement system;

[0084] σ 2 repr -- Reproducibility error of the measurement system;

[0085] σ 2 MS = σ 2 repe + σ 2 repr = 0.0016;

[0086] Where: σ MS -- Standard deviation of the measurement system fluctuation;

[0087] G = σ MS / σ T = 0.04 / 1 = 0.04, R = 6σ MS / (USL - LSL) = 6 * 0.04 / 3 = 0.08,

[0088] When both G and R are less than 10%, and ndc ≥ 5, it indicates that the measurement system is good and has sufficient resolution.

[0089] S4 Moisture Meter Automatic Calibration

[0090] Select 3 batches of data with a linear regression P value ≥ 0.05, both G and R less than 10%, and ndc ≥ 5 for the moisture content at the outlet of the loose rewetting of this brand as the basic data, and take the average value of the three batches of data as the basis for calibration, and automatically calculate the deviation between the two;

[0091] Adjust the zero position of the moisture meter through the central control system and re - issue it. When both G and R of the measurement data of this batch are less than 10% and ndc ≥ 5, the calibration is successful.

[0092] S5 Batch Similarity Analysis and Early Warning

[0093] Statistically analyze and record the moisture content change of cut tobacco in each batch during the process, and establish a historical record database of cut tobacco moisture content by brand and process;

[0094] During batch production, by comparing with the historical best level and using the similarity analysis method to analyze the results. If the similarity between the two is high, it indicates that the moisture meter for this batch of cigarettes is accurately detected; if the similarity is low, it indicates that there is a certain deviation in the moisture content of this batch.

[0095] Based on the deviation degree of this batch from the historical best level, a pre-production warning is given for the next batch. The metrology personnel ensure the accuracy of the moisture meter detection and the stability control of the cut tobacco quality during the next batch of cut tobacco production process by timely adjusting other parameters such as the zero position of the moisture meter.

[0096] Embodiment 2

[0097] An on-line automatic calibration and control system for a moisture meter, including a moisture meter channel issuing unit, a comparative analysis and warning unit for moisture meter parameters, a measurement system analysis and warning unit, and an automatic correction of the moisture meter, batch similarity analysis and warning.

[0098] Specifically:

[0099] The said moisture meter channel issuing unit is used to issue production instructions through the central control system according to the brand of cigarettes to be produced before the cut tobacco production, and at the same time issue moisture meter channels, where each moisture meter channel corresponds to a cigarette brand. The moisture meter channel includes: the zero position, span and damping parameters of each process moisture meter, so as to achieve the issuance of moisture meter channels for each on-site process and the establishment of formulation parameters.

[0100] The said comparative analysis and warning unit for moisture meter parameters is used to automatically compare and analyze with the moisture meter channels and parameters issued by the central control system, including the comparison and analysis of the brand to be produced, channel coding, zero position, span and damping parameters. If the above information is completely consistent, an instruction is issued for formal production; if the brand, channel or a certain parameter does not match, a warning is given, and the workshop metrology personnel conduct on-site inspections and adjustments according to the warning situation.

[0101] The said measurement system analysis and warning unit is used to evaluate the accuracy of the moisture meter measurement system through offset and linear analysis, repeatability and reproducibility analysis:

[0102] ① Offset and linear analysis

[0103] Conduct linear regression analysis on the moisture content data detected by the moisture meters of each process;

[0104] When the P value ≥ 0.05, it indicates that the overall offset of the moisture content data is not significant and the moisture meter measurement is accurate;

[0105] When the P value < 0.05, it indicates that the overall moisture content data shows an offset, then a warning is given, and correction is made according to the average value of the moisture content of this batch to ensure the accuracy of the moisture meter measurement data;

[0106] ② Repeatability and reproducibility analysis

[0107] The source of the measurement fluctuation of the moisture meter was analyzed by the analysis of variance method. The sources of the measurement fluctuation included three categories: tobacco leaves or cut tobacco, the moisture meter, and the interaction between tobacco leaves or cut tobacco and the moisture meter, and the three were independent of each other;

[0108] Y ij = μ + L i + M + (LM) ij ;

[0109] In the formula:

[0110] i--1, 2,..., I;

[0111] j--1, 2,..., J;

[0112] Y ij --The moisture content data of the jth detection of the ith batch of tobacco leaf or cut tobacco samples by the moisture meter;

[0113] μ--Unknown constant;

[0114] L i 、M、(LM) ij --The repeated errors of the interaction between tobacco leaves or cut tobacco, the moisture meter, and the interaction between tobacco leaves or cut tobacco and the moisture meter;

[0115] By using the analysis of variance method, σ 2 M 、σ 2 L 、σ 2 LM were used to represent the fluctuations caused by tobacco leaves or cut tobacco, the moisture meter, and the interaction between tobacco leaves or cut tobacco and the moisture meter respectively. The mathematical expressions of the repeatability error and reproducibility error of the moisture meter measurement system were:

[0116] σ 2 repe = σ 2 M ,σ 2 repr = σ 2 L + σ 2 LM ;

[0117] In the formula:

[0118] σ 2 repe --The repeatability error of the measurement system;

[0119] σ 2repr -- Reproducibility error of the measurement system;

[0120] σ 2 MS = σ 2 repe + σ 2 repr ;

[0121] Where: σ MS -- Standard deviation of the measurement system fluctuation;

[0122] G = σ MS / σ T , R = 6σ MS / (USL - LSL),

[0123] Where:

[0124] G--Percentage of the measurement system fluctuation in the overall process fluctuation;

[0125] R--Percentage of the measurement system accuracy in the tolerance;

[0126] ndc--Number of distinguishable categories of the measurement system;

[0127] σ T --Total fluctuation of the measurement system process;

[0128] USL--Standard upper limit of the moisture content of each process;

[0129] LSL--Standard lower limit of the moisture content of each process;

[0130] When both G and R are less than 10%, and ndc ≥ 5, it indicates that the measurement system is good and has sufficient resolution;

[0131] When one of G and R is ≥ 10%, or ndc < 5, it indicates that the measurement system needs to be improved or has insufficient resolution ability, then a warning is given and calibration is carried out.

[0132] The automatic calibration unit of the moisture meter selects at least 3 batches of data with the same brand and the same process, the linear regression P value of the moisture content data ≥ 0.05, both G and R less than 10%, and ndc ≥ 5 as the basic data, and takes the mean of the three batches of data as the basis for calibration, automatically calculates the deviation between the two, adjusts the zero position of the moisture meter through the central control system and redistributes it. When both G and R of the measurement data of this batch are less than 10% and ndc ≥ 5, the calibration is successful.

[0133] The batch similarity analysis and warning are used to statistically analyze and record the change of the moisture content of cut tobacco in each batch process, and establish a historical record database of the moisture content of cut tobacco by brand and process;

[0134] During batch production, by comparing with the historical best level and using the similarity analysis method to analyze the results, if the similarity between the two is relatively high, it indicates that the moisture meter for this batch of cigarettes detects well and accurately; if the similarity between the two is relatively low, it indicates that there is a certain deviation in the moisture content of this batch.

[0135] Based on the deviation degree of this batch from the historical best level, a pre-production warning is given for the next batch. The metrology personnel ensure the accuracy of the moisture meter detection in the cut tobacco process and the stability control of the cut tobacco quality for the next batch by timely adjusting other parameters such as the zero position of the moisture meter.

[0136] It should be noted that the above are only the preferred embodiments of the present invention and are not used 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 perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An online automatic calibration and control method for a moisture meter, characterized in that: including Issuing of moisture meter channels by the central control system Before the production of cut tobacco, according to the brand of cigarettes to be produced, production instructions are issued through the central control system, and at the same time, moisture meter channels are issued; Each moisture meter channel corresponds to a cigarette brand, and the moisture meter channel includes: The zero position, span and damping parameters of each process moisture meter are used to issue the moisture meter channels at each site and establish the formulation parameters; S2 Comparative analysis and early warning of moisture meter parameters Automatically compare and analyze with the moisture meter channels and parameters issued by the central control system, including the comparison and analysis of the brand to be produced, channel coding, zero position, span and damping parameters; If the above information is completely consistent, an instruction is issued to start formal production; If the brand, channel or a certain parameter does not match, an early warning is given, and the workshop metrology personnel conduct on-site inspections and adjustments according to the early warning situation; S3 Measurement system analysis and early warning S3.1 Offset and linearity analysis Perform linear regression analysis on the moisture content data detected by the moisture meters in each process; When the P value ≥ 0.05, it indicates that the overall offset of the moisture content data is not significant and the moisture meter measures accurately; When the P value < 0.05, it indicates that the overall moisture content data shows an offset, then an early warning is given, and corrections are made according to the average value of the moisture content of the batch to ensure the accuracy of the moisture meter measurement data; S3.2 Repeatability and reproducibility analysis Use the analysis of variance method to analyze the sources of moisture meter measurement fluctuations. The sources of measurement fluctuations include three categories: tobacco leaves or cut tobacco, moisture meters, and the interaction between tobacco leaves or cut tobacco and moisture meters, and the three are independent of each other; ； In the formula: --1，2，…， ； --1，2，…， ； -- The moisture content data detected by the moisture meter for the first batch of tobacco leaf or cut tobacco samples; The first moisture content data. -- Unknown constant; , , -- tobacco leaves or cut tobacco, moisture meters, and the repeated error of the interaction between tobacco leaves or cut tobacco and moisture meters; Using the analysis of variance method, respectively with , , represent the fluctuations caused by the interaction between tobacco leaves or cut tobacco, moisture meters, and the interaction between tobacco leaves or cut tobacco and moisture meters. The mathematical expressions for the repeatability error and reproducibility error of the moisture meter measurement system are respectively: ， ； In the formula: --Repeatability error of the measurement system; -- Reproducibility error of the measurement system; ； Where: -- Standard deviation of measurement system variation; ， ， ； In the formula: -- Measure the percentage of the measurement system variation in the overall process variation; -- Percentage of measurement system accuracy to tolerance; -- The number of distinguishable categories of the measurement system; --Measure the total process variation of the measurement system; -- Upper standard limit of moisture content for each process; -- Standard lower limit of water content for each process; When and are both less than 10%, and ≥ 5, it indicates that the measurement system is good and has sufficient resolution; When and one of them is ≥ 10%, or <5, it indicates that the measurement system needs to be improved or the resolution is insufficient, then a warning is given and calibration is carried out; S4 Automatic calibration of moisture meter Select the same brand and the same process. For the linear regression of moisture content data, the P-value ≥ 0.05, , both are less than 10%, and at least 3 batches of data with ≥ 5 are used as basic data, and the mean of the three batches of data is taken as the basis for calibration, and the deviation between the two is automatically calculated; Adjust the zero position of the moisture meter through the central control system and reissue it to make the , of the batch measurement data are all less than 10%, and ≥ 5, then the calibration is successful.

2. The online automatic calibration and control method of the moisture meter according to claim 1, characterized in that: The method further includes S5 Batch similarity analysis and early warning Statistically analyze and record the moisture content change of cut tobacco in each batch, and establish a historical record database of cut tobacco moisture content by brand and process; During the batch production process, compare with the historical best level, use the similarity analysis method to analyze the results. If the similarity between the two is relatively high, it indicates that the moisture meter detection of this batch of cigarettes is good and accurate; if the similarity between the two is relatively low, it indicates that there is a certain offset in the moisture content of this batch; Give a pre-production warning for the next batch according to the offset degree of this batch from the historical best level.

Images