A method for predicting the coating rate of finished products of reconstituted tobacco leaves on papermaking machines

Abstract

The present invention relates to a method for predicting the coating rate of finished products of papermaking reconstituted tobacco leaves on a machine, belonging to the technical field of papermaking reconstituted tobacco leaves. The present invention establishes a physical index prediction model for coating liquid based on the "Record Database of Physical Indexes Such as Sugar Content and Refractive Index of Coating Liquid Additives and Concentrates" of papermaking reconstituted tobacco leaves. Then, based on the absolute dry weight design value of the sheet base before coating, the coating rate design value, the sugar content and refractive index of the coating liquid, and other physical indicators, a model for predicting the coating rate of finished products on a machine is established, thereby obtaining a method for predicting the coating rate of finished products of papermaking reconstituted tobacco leaves on a machine. The present invention can greatly reduce the energy consumption of concentration, reduce the time for debugging the coating rate on the machine, and save production and pilot costs. At the same time, once the method model is established and accurate, it has the advantages of simple calculation, convenience and speed.

Description

Technical Field

[0001] The invention belongs to the technical field of papermaking reconstituted tobacco, and in particular relates to a method for predicting the coating rate of finished products of papermaking reconstituted tobacco on a machine. Background Art

[0002] Papermaking reconstituted tobacco uses tobacco stems, tobacco dust, and shredded tobacco as raw materials from the cigarette manufacturing process. Tobacco extract is obtained through solid-liquid separation and extraction. After multiple layers of purification and concentration, various additives are added in the batching process to create a coating solution. This solution is then accurately, evenly, and stably applied to the reconstituted tobacco base according to the designed coating rate. After drying and slicing, it is ready for cigarette production. The reconstituted tobacco market is rapidly expanding, and product categories are evolving rapidly. While established products require constant process adjustments to adapt to the market, a growing number of new products are being developed and require pilot production. Due to the varying formulations and process requirements of various products, accurately preparing a coating solution with appropriate physical properties, such as sugar refractive index, to meet the designed coating rate during initial production runs is challenging. Typically, a coating solution with higher physical properties, such as sugar refractive index, is prepared based on historical experience. During pilot production, if the coating rate is too high, water is added for gradual dilution, and the coating roller pressure is adjusted to approach the target coating rate. This traditional method is time-consuming, energy-intensive, and costly. Furthermore, coating solutions with high physical indicators like sugar refractive index (RDI) cannot effectively complete coating due to poor fluidity, resulting in poor permeability and even equipment shutdowns. If the prepared coating solution's RDI (RDI) is too low to achieve the target coating rate, the coating solution and concentrate must be discarded and the next batch of concentrate with higher RDI (RDI) levels must be prepared, resulting in material waste. Therefore, it is necessary to establish a method for predicting the coating rate of finished papermaking reconstituted tobacco leaves on a machine. This method has important guiding significance for on-machine testing of new papermaking reconstituted tobacco products and for formula adjustment testing of conventional products. Summary of the Invention

[0003] In order to overcome the problems existing in the background technology, the present invention provides a method for predicting the coating rate of the finished product of reconstituted tobacco leaves on the papermaking machine, which can accurately control the target indicators of the coating liquid, greatly reduce the energy consumption of concentration, reduce the time for debugging the coating rate on the machine, and save production and pilot costs.

[0004] To achieve the above object, the present invention is implemented through the following technical solutions:

[0005] The method for predicting the coating rate of finished product of reconstituted tobacco leaves on a papermaking machine specifically comprises the following steps:

[0006] S1. Based on the production record database, a correlation analysis was conducted on the key physical indicators of the coating solution. It was found that the indicators with strong correlation with the coating rate were viscosity, solid content, sugar refractive index, and suspended matter.

[0007] S2. Based on the production record database, prediction models for coating liquid viscosity, solid content, sugar refractive index and suspended matter were established respectively, and the above four coating liquid index prediction models were verified and calibrated to finally determine the prediction model; the coating liquid viscosity prediction model obtained by analysis is: coating liquid viscosity / pa·s=a1×concentrate viscosity / pa·s+b1×additive 1 amount+c1×additive 2 amount+d1; the coating liquid solid content prediction model is: coating liquid solid content / %=a2 × solid content of concentrated liquid / % + b2 × amount of additive 1 + c2 × amount of additive 2 + d2; the prediction model of sugar content refractive index of coating liquid is: sugar content refractive index of coating liquid / % = a3 × sugar content refractive index of concentrated liquid / % + b3 × amount of additive 1 + c3 × amount of additive 2 + d3; the prediction model of suspended matter of coating liquid is: suspended matter % of coating liquid = a4 × suspended matter % of concentrated liquid + b4 × amount of additive 1 + c4 × amount of additive 2 + d4; where a, b, c, and d are all coefficients.

[0008] S3. Establish a coating rate prediction model based on the production record statistical database. The coating rate prediction model is analyzed and obtained as follows: coating rate / % = a×absolute dry weight of the film base before coating + b×refraction of sugar content of the coating liquid + c×coating liquid viscosity / pa·s + d×solid content of the coating liquid + e×coating liquid suspension + f, where a, b, c, d, e, and f are coefficients. The coating rate prediction model is verified and corrected using the four coating liquid index prediction models established in step S2.

[0009] S4. Confirm the coating rate prediction model and predict the coating rate using the confirmed coating rate prediction model.

[0010] Furthermore, the key physical indicators of the coating solution in step S1 include temperature, viscosity, solid content, sugar refractive index, suspended matter and density.

[0011] Furthermore, based on the viscosity test values ​​of the coating liquid, the viscosity test values ​​of the concentrate corresponding to the coating liquid, and the viscosity test values ​​of additives 1 and 2, data with gradients are selected for regression analysis to establish the coating liquid viscosity prediction model described in step 2.

[0012] Furthermore, based on the solid content detection value of the coating liquid, the solid content detection value of the concentrate corresponding to the coating liquid, and the solid content detection value database of additives 1 and additives 2, data with gradients are selected for regression analysis to establish the coating liquid solid content prediction model described in step 2.

[0013] Furthermore, based on the sugar refractive index detection value of the coating liquid, the sugar refractive index detection value of the concentrate corresponding to the coating liquid, and the sugar refractive index detection value database of additives 1 and additive 2, data with a gradient are selected for regression analysis to establish the sugar refractive index prediction model for the coating liquid described in step 2.

[0014] Furthermore, based on the suspended matter detection values ​​of the coating liquid, the suspended matter detection values ​​of the concentrate corresponding to the coating liquid, and the suspended matter detection values ​​of additives 1 and 2, data with gradients are selected for regression analysis to establish the coating liquid suspended matter prediction model described in step 2.

[0015] Furthermore, the verification and correction of the four coating liquid index prediction models in step S2 specifically include the following steps:

[0016] A. Based on the coating liquid viscosity prediction model, calculate the coating liquid viscosity prediction value under a given concentrate viscosity and coating formula; based on the coating liquid solid content prediction model, calculate the coating liquid solid content prediction value under a given concentrate solid content and coating formula; based on the coating liquid sugar refractive index prediction model, calculate the coating liquid sugar refractive index prediction value under a given concentrate sugar refractive index and coating formula; based on the coating liquid suspended matter prediction model, calculate the coating liquid suspended matter prediction value under a given concentrate suspended matter and coating formula;

[0017] B. preparing a coating solution using the concentrate and coating formula given by the four physical index prediction models of the coating solution described in step A, and testing the viscosity, solid content, sugar refractive index, and suspended matter of the coating solution;

[0018] C. performing deviation analysis on the predicted values ​​of the physical indicators of the coating liquid obtained in step A and the detected values ​​of the physical indicators of the coating liquid obtained in step B;

[0019] D. Statistically analyze the deviation analysis results. If the deviations of the coating solution's viscosity, solid content, sugar refractive index, and suspended solids are within 0±1%, the prediction model can be determined by performing at least two validation cycles. Otherwise, the prediction model needs to be corrected and then validated at least twice more before the prediction model can be determined.

[0020] Furthermore, regression analysis is performed on data in the database of production record statistics of the absolute dry weight of the film base before coating, production record statistics of the viscosity of the coating liquid, production record statistics of the sugar refractive index of the coating liquid, production record statistics of the solid content of the coating liquid, production record statistics of the suspended matter of the coating liquid and production record statistics of the coating rate to establish a coating rate prediction model.

[0021] Furthermore, the coating rate prediction model verification and correction described in step S3 specifically includes the following steps:

[0022] a. The four physical indicators required for the concentrate are calculated based on the four coating liquid index prediction models obtained in step S2, the coating rate prediction model obtained in step S3, and the absolute dry quantitative design value of the new product base before coating, the coating rate design value, and the physical index comparison table of the concentrate;

[0023] b. Complete the preparation of the coating solution using the concentrate obtained in step a for the new product formulation and test the viscosity, solids content, sugar refractive index, and suspended matter of the coating solution;

[0024] c. Record the actual coating rate after coating on the machine and perform deviation analysis against the designed coating rate;

[0025] d. Perform statistics on the deviation analysis. If the deviation is within 0±1%, the coating rate prediction model can be determined by at least two cycles of verification. Otherwise, the coating rate prediction model needs to be calibrated and then at least two cycles of verification are required to determine the coating rate prediction model.

[0026] Beneficial effects of the present invention:

[0027] The present invention establishes a physical index prediction model for coating liquid based on the "Record Database of Physical Indexes Such as Sugar Content and Refractive Index of Coating Liquid Additives and Concentrates" of papermaking-processed reconstituted tobacco leaves; and then establishes a coating rate prediction model for finished products on the machine based on the absolute dry basis weight design value of the sheet base before coating, the coating rate design value, the sugar content and refractive index of the coating liquid, and other physical indicators, thereby obtaining a method for predicting the coating rate of finished products on the machine of papermaking-processed reconstituted tobacco leaves. This method can be used to reversely infer the physical indicators such as sugar refractive index of the concentrated solution to be prepared based on the target coating rate, the absolute dry quantitative design value of the base, the coating liquid physical indicator prediction model, the coating rate prediction model and the "Concentrated Solution Physical Indicator Comparison Table", providing a data basis for the online production and pilot test of new products. Before going on the machine, the physical indicators of the concentrated solution corresponding to the coating solution are predicted and the coating solution preparation is accurately completed, so that the target coating rate is accurately achieved after the machine coating, avoiding the situation where the coating rate does not meet the standard after the machine coating and a large amount of coating solution and concentrate are discarded due to large errors in the physical indicators of the prepared coating solution. At the same time, for the preparation section of the concentrated solution, there are more accurate control target indicators, which can greatly reduce the concentration energy consumption, reduce the time for debugging the coating rate on the machine, and save production and pilot test costs. At the same time, once the model of this method is established and accurate, it has the advantages of simple calculation, convenience and speed. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and beneficial effects of the present invention clearer, the preferred embodiments of the present invention will be described in detail below to facilitate understanding by technicians.

[0029] The method for predicting the coating rate of finished product of reconstituted tobacco leaves on a papermaking machine specifically comprises the following steps:

[0030] S1. Based on the production record database, a correlation analysis was conducted on the key physical indicators of the coating liquid. In the industry, the key physical indicators of the coating liquid are temperature, viscosity, solid content, sugar refractive index, suspended matter and density. It was found that the indicators with a strong correlation with the coating rate are viscosity, solid content, sugar refractive index and suspended matter.

[0031]

[0032]

[0033] The results of regression analysis are as follows:

[0034] Related: Temperature °C, Viscosity pa·s, Solids %, Sugar content Brix%, Suspended matter%, Density kg·m 3 , coating rate %

[0035]

[0036] From the above correlation analysis, it can be seen that the four indicators that are strongly correlated with the coating rate are viscosity pa·s, solid content %, sugar refractive index Brix %, and suspended matter %.

[0037] S2. Based on the production record database, prediction models for coating liquid viscosity, solid content, sugar refractive index and suspended matter are established respectively, and the above four coating liquid index prediction models are verified and calibrated to finally determine the prediction model.

[0038] (1) Establishing a coating fluid viscosity prediction model

[0039] Based on the viscosity test values ​​of the coating liquid in the production record database, the viscosity test values ​​of the concentrate corresponding to the coating liquid, and the viscosity test values ​​of additives 1 and 2, data with gradients were selected for regression analysis to establish a coating liquid viscosity prediction model.

[0040]

[0041]

[0042] The analysis results are as follows:

[0043] Regression analysis: coating solution viscosity in pa·s vs. concentrate viscosity in pa·s, additive 1 viscosity, additive 2 viscosity

[0044] Analysis of variance

[0045]

[0046] Model Summary

[0047]

[0048] coefficient

[0049]

[0050] Regression equation

[0051] Coating solution viscosity Pa·s = -320 + 1.4320 Concentrate viscosity Pa·s + 1.41 Additive 1 viscosity + 2.43 Additive 2 viscosity

[0052] (2) Establishing a coating liquid solid content prediction model

[0053] Based on the solid content test values ​​of the coating liquid in the production record database, the solid content test values ​​of the concentrate corresponding to the coating liquid, and the solid content test values ​​of additives 1 and additive 2, data with gradients were selected for regression analysis to establish a coating liquid solid content prediction model.

[0054]

[0055]

[0056] The analysis results are as follows:

[0057] Regression analysis: coating liquid solid content % vs. coating liquid concentrate solid content %, additive 1 solid content %, additive 2 solid content %

[0058] Analysis of variance

[0059]

[0060] Model Summary

[0061]

[0062] coefficient

[0063]

[0064] Regression equation

[0065] Coating liquid solid content% = 6.11 + 1.4313 coating liquid corresponding to the concentrate solid content% - 1.156 additive 1 solid content% + 2.05 additive 2 solid content%

[0066] (3) Establishing a prediction model for the refractive index of sugar in coating liquid

[0067] Based on the sugar refractive index detection values ​​of the coating liquid, the sugar refractive index detection values ​​of the concentrate corresponding to the coating liquid, and the sugar refractive index detection values ​​of additives 1 and 2 in the production record database, data with gradients were selected for regression analysis to establish a prediction model for the sugar refractive index of the coating liquid.

[0068]

[0069]

[0070] The analysis results are as follows:

[0071] Regression analysis: Brix% of sugar content in coating solution and Brix% of sugar content in concentrate corresponding to coating solution, Brix% of sugar content in additive 1

[0072] Analysis of variance

[0073]

[0074] Model Summary

[0075]

[0076] coefficient

[0077]

[0078] Regression equation

[0079] The Brix% of sugar in the coating solution is 34.144 + 0.5653. The Brix% of sugar in the concentrate corresponding to the coating solution is -0.1409. The Brix% of sugar in additive 1 is -0.0731. The Brix% of sugar in additive 2 is

[0080] (4) Establishing a prediction model for suspended solids in coating liquid

[0081] Based on the suspended matter detection values ​​of the coating liquid in the production record database, the suspended matter detection values ​​of the concentrate corresponding to the coating liquid, and the suspended matter detection value database of additives 1 and additives 2, data with gradients are selected for regression analysis to establish a coating liquid suspended matter prediction model.

[0082]

[0083]

[0084] The analysis results are as follows:

[0085] Regression analysis: coating liquid suspended solids % vs. coating liquid concentrate suspended solids %, additive 1 suspended solids %, additive 2 suspended solids %

[0086] Analysis of variance

[0087]

[0088] Model Summary

[0089]

[0090] coefficient

[0091]

[0092] Regression equation

[0093] Coating liquid suspended matter% = -55.9 + 3.70 Coating liquid corresponding concentrate suspended matter% + 4.10 Additive 1 suspended matter% + 0.299 Additive 2 suspended matter%

[0094] (5) Verification of coating fluid viscosity prediction model

[0095] Based on the coating liquid viscosity prediction model, the coating liquid viscosity prediction value is calculated under the given concentrated liquid viscosity and coating formula, the coating liquid is prepared using the concentrated liquid and coating formula given by the coating liquid viscosity prediction model, and the viscosity of the coating liquid is tested. The deviation analysis of the coating liquid viscosity prediction value and the coating liquid viscosity test value is performed, and the deviation analysis results are statistically analyzed. If the viscosity deviation of the coating liquid is within 0±1%, the prediction model can be determined by performing two cycles of verification; otherwise, the prediction model needs to be corrected, and then the prediction model can be determined by performing two cycles of verification.

[0096] Verification of coating fluid viscosity prediction model:

[0097]

[0098]

[0099] Verification results: As shown in the deviation results in the table above, the deviation between the predicted value and the measured value of the coating liquid viscosity is within the range of 8±12%, which has exceeded the tolerance range of 0±1%. It is necessary to re-perform regression analysis to calibrate the prediction model.

[0100] The results of recalibrating the coating fluid viscosity prediction model are as follows:

[0101] Regression analysis: coating liquid viscosity pa·s test value and concentrate viscosity pa·s, additive 1 viscosity, additive 2 viscosity

[0102] Analysis of variance

[0103]

[0104] Model Summary

[0105]

[0106] coefficient

[0107]

[0108] Regression equation

[0109] Coating liquid viscosity Pa·s test value = -326.6 + 1.4392 Concentrate viscosity Pa·s + 1.4517 Additive 1 viscosity + 2.657 Additive 2 viscosity

[0110] Verify the coating fluid viscosity prediction model again:

[0111]

[0112]

[0113] Verification results: As shown in the table above, the deviation between the predicted value and the measured value of the viscosity of the coating liquid is within 0±1%, which is within the allowable range.

[0114] Secondary verification of the coating fluid viscosity prediction model:

[0115]

[0116]

[0117] Verification results: As shown in the table above, the deviation between the predicted value and the measured value of the viscosity of the coating liquid is within 0±1%, which is within the allowable range.

[0118] After two cycles of verification, it can be confirmed that the coating liquid viscosity prediction model is: coating liquid viscosity pa·s = -326.6 + 1.4392×concentrate viscosity pa·s + 1.4517×additive 1 viscosity + 2.657×additive 2 viscosity.

[0119] (6) Verification of coating liquid solid content prediction model

[0120] Based on the coating liquid solid content prediction model, the predicted value of the coating liquid solid content under a given concentrated liquid viscosity and coating formula is calculated, the coating liquid is prepared using the concentrated liquid and coating formula given by the coating liquid solid content prediction model, and the solid content of the coating liquid is detected. The deviation analysis is performed on the predicted value of the coating liquid solid content and the detected value of the coating liquid solid content, and the deviation analysis results are statistically analyzed. If the solid content deviation of the coating liquid is within 0±1%, the prediction model can be determined by performing two cycles of verification; otherwise, the prediction model needs to be corrected, and then the prediction model can be determined by performing two cycles of verification.

[0121] Verification of coating liquid solid content prediction model:

[0122]

[0123]

[0124] Verification results: As shown in the table above, the deviation between the predicted value and the detected value of the solid content of the coating liquid is within 0±1%, which is within the allowable range.

[0125] Secondary verification of the coating liquid solid content prediction model:

[0126]

[0127]

[0128]

[0129] Verification results: As shown in the table above, the deviation between the predicted value and the detected value of the solid content of the coating liquid is within 0±1%, which is within the allowable range.

[0130] After two cycles of verification, it can be confirmed that the prediction model of the solid content of the coating liquid is: coating liquid solid content % = 6.11 + 1.4313 × the solid content of the concentrate corresponding to the coating liquid % - 1.156 × the solid content of additive 1 % + 2.05 × the solid content of additive 2.

[0131] (7) Verification of the sugar refractive index prediction model of the coating liquid

[0132] Based on the sugar refractive index prediction model of the coating liquid, the predicted value of the sugar refractive index of the coating liquid under a given concentrated liquid viscosity and coating formula is calculated, the coating liquid is prepared using the concentrated liquid and coating formula given by the sugar refractive index prediction model of the coating liquid, and the sugar refractive index of the coating liquid is detected. The deviation analysis is performed on the predicted value of the sugar refractive index of the coating liquid and the detected value of the sugar refractive index of the coating liquid, and the deviation analysis results are statistically analyzed. If the deviation of the sugar refractive index of the coating liquid is within 0±1%, the prediction model can be determined by performing two cycles of verification; otherwise, the prediction model needs to be corrected, and then the verification cycle is repeated twice to determine the prediction model.

[0133] Verification of the refractive index prediction model for the sugar content in the coating solution:

[0134]

[0135]

[0136] Verification results: As shown in the table above, the deviation between the predicted value and the measured value of Brix% of the sugar content in the coating solution is within 0±1%, which is within the allowable range.

[0137] Secondary verification of the refractive index prediction model for the sugar content in the coating solution:

[0138]

[0139]

[0140]

[0141] Verification results: As shown in the table above, the deviation between the predicted value and the measured value of Brix% of the sugar content in the coating solution is within 0±1%, which is within the allowable range.

[0142] After two cycles of verification, it can be confirmed that the prediction model of the sugar refractive index of the coating liquid is: Brix% of sugar refractive index of the coating liquid = 34.144 + 0.5653 × Brix% of sugar refractive index of the concentrate corresponding to the coating liquid - 0.1409 × Brix% of sugar refractive index of additive 1 - 0.0731 × Brix% of sugar refractive index of additive 2.

[0143] (8) Verification of coating liquid suspended solids prediction model

[0144] Based on the coating liquid suspended matter prediction model, the predicted value of the coating liquid suspended matter under a given concentrated liquid viscosity and coating formula is calculated, the coating liquid is prepared using the concentrated liquid and coating formula given by the coating liquid suspended matter prediction model, and the suspended matter of the coating liquid is detected. The deviation analysis of the coating liquid suspended matter prediction value and the coating liquid suspended matter detection value is performed, and the deviation analysis results are statistically analyzed. If the deviation of the coating liquid suspended matter is within 0±1%, the prediction model can be determined by performing two cycles of verification; otherwise, the prediction model needs to be corrected, and then the prediction model can be determined by performing two cycles of verification.

[0145] Verification of coating fluid suspended solids prediction model:

[0146]

[0147]

[0148] Verification results: As shown in the table above, the deviation between the predicted value and the measured value of the suspended matter % of the coating liquid is within the range of 0±7%, which has exceeded the tolerance range of 0±1%. The prediction model needs to be recalibrated.

[0149] The results of recalibrating the coating fluid suspended solids prediction model are as follows:

[0150] Regression analysis: Coating liquid suspended matter % detection value and concentrate suspended matter %, additive 1 suspended matter %_1, additive 2 suspended matter %_1

[0151] Analysis of variance

[0152]

[0153] Model Summary

[0154]

[0155] coefficient

[0156]

[0157] Regression equation

[0158] Coating liquid suspended matter% detection value = -28.53 + 2.1921 concentrate suspended matter% + 2.532 additive 1 suspended matter%_1 + 0.468 additive 2 suspended matter%_1

[0159] Verify the coating liquid suspended solids prediction model again:

[0160]

[0161]

[0162] Verification results: As shown in the table above, the deviation between the predicted value and the detected value of the suspended matter % of the coating liquid is within 0±1%, which is within the allowable range.

[0163] Secondary verification of the coating liquid suspended solids prediction model:

[0164]

[0165]

[0166] Verification results: As shown in the table above, the deviation between the predicted value and the detected value of the suspended matter % of the coating liquid is within 0±1%, which is within the allowable range.

[0167] After two cycles of verification, it can be confirmed that the prediction model of coating liquid suspended matter is: coating liquid suspended matter % =

[0168] -28.53 + 2.1921 × concentrate suspended matter % + 2.532 × additive 1 suspended matter % + 0.468 × additive 2 suspended matter %.

[0169] S3. Establish a coating rate prediction model based on the production record statistical database, and use the four coating liquid index prediction models established in step S2 to verify and calibrate the coating rate prediction model.

[0170] (1) Establishing a coating rate prediction model

[0171] A coating rate prediction model was established by performing regression analysis on the data in the database of the production record statistics of the absolute dry weight of the film base before coating, the production record statistics of the viscosity of the coating liquid, the production record statistics of the sugar refractive index of the coating liquid, the production record statistics of the solid content of the coating liquid, the production record statistics of the suspended matter of the coating liquid and the production record statistics of the coating rate.

[0172]

[0173]

[0174] The analysis results are as follows:

[0175] Regression analysis: coating rate and absolute dry weight of film base before coating, viscosity of coating liquid, solid content of coating liquid, refractive index of sugar content of coating liquid

[0176] Analysis of variance

[0177]

[0178] Model Summary

[0179]

[0180] coefficient

[0181]

[0182] Regression equation

[0183] Coating rate = -8.0 + 0.234 absolute dry weight of the film base before coating + 0.00414 viscosity of the coating liquid + 0.0926 solid content of the coating liquid + 0.5321 sugar content of the coating liquid + 0.1708 suspended matter in the coating liquid

[0184] (2) Multi-formula verification coating rate prediction model

[0185] The viscosity, solid content, sugar refractive index and suspended matter index of the corresponding concentrated liquid are calculated based on the obtained coating liquid viscosity prediction model, coating liquid solid content prediction model, coating liquid sugar refractive index prediction model and coating liquid suspended matter prediction model and coating rate prediction model, the film base absolute dry weight design value before coating of the new product, the coating rate design value, and the concentrated liquid physical index comparison table; the concentrated liquid with the indicators is used to complete the preparation of the coating liquid under the new product formula and test the viscosity, solid content, sugar refractive index and suspended matter of the coating liquid; the actual coating rate obtained after coating on the machine is recorded and compared with the designed coating rate for deviation analysis; the deviation analysis is statistically analyzed, and if the deviation is within 0±1%, the coating rate prediction model can be determined by at least two cycles of verification; otherwise, the coating rate prediction model needs to be corrected and then at least two cycles of verification are required to determine the coating rate prediction model.

[0186]

[0187]

[0188]

[0189]

[0190]

[0191]

[0192] Verification results: As shown in the table above, the deviation between the predicted value and the tested value of the physical index of the coating liquid prepared as required is within 0±1%; the deviation between the predicted value and the tested value of the coating rate of the new product is within 0±1%, which is within the allowable range.

[0193] S4. Confirm the coating rate prediction model and predict the coating rate using the confirmed coating rate prediction model.

[0194] The confirmed coating rate prediction model is: coating rate % = -8 + 0.234 × absolute dry weight design value of the base before coating g / m2 + 0.00414 × (-326.6 + 1.4392 × corresponding concentrate viscosity pa·s + 1.4517 × additive 1 viscosity pa·s + 2.657 × additive 2 viscosity pa·s) + 0.0926 × (6.11 + 1.4313 × corresponding concentrate solid content % - 1.156 × additive 1 solid content % + The prediction model is used to predict the coating rate.

[0195] The verified coating rate prediction model is generally accurate and can be applied to coating rate prediction for similar new products, pilot products, process test products, and other similar products. Before the coating is applied to the machine, the physical indicators of the concentrate corresponding to the coating liquid are predicted and the coating liquid is accurately prepared, so that the target coating rate is accurately achieved after the coating is applied on the machine. This avoids the situation where the coating rate does not meet the target after the coating is applied on the machine due to large errors in the physical indicators of the prepared coating liquid, resulting in a large amount of coating liquid and concentrate being wasted. At the same time, for the preparation stage of the concentrate, more precise control target indicators can be obtained, which can greatly reduce the energy consumption of concentration, shorten the time for debugging the coating rate on the machine, and save production and pilot costs.

[0196] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. A method for predicting the coating rate of finished products of reconstituted tobacco leaves for papermaking, characterized by: The specific steps include: S1. Based on the production record database, a correlation analysis was conducted on the key physical indicators of the coating solution. It was found that the indicators with strong correlation with the coating rate were viscosity, solid content, sugar refractive index, and suspended matter. S2. Based on the production record database, prediction models for coating liquid viscosity, solid content, sugar refractive index, and suspended matter are established respectively, and the above four coating liquid index prediction models are verified and calibrated to finally determine the prediction model; S3. Establishing a coating rate prediction model based on the production record statistical database, and using the four coating liquid index prediction models established in step S2 to verify and calibrate the coating rate prediction model; S4, confirming the coating rate prediction model and using the confirmed coating rate prediction model to predict the coating rate; the coating rate prediction model verification and correction described in step S3 specifically include the following steps: a. The four coating liquid index prediction models obtained in step S2, the coating rate prediction model obtained in step S3, and the new product base before coating are calculated from the absolute dry quantitative design value, the coating rate design value, and the four physical indicators required for the concentrate are calculated from the comparison table of physical indicators of the concentrate; b. Complete the preparation of the coating solution using the concentrate obtained in step a for the new product formulation and test the viscosity, solids content, sugar refractive index, and suspended matter of the coating solution; c. Record the actual coating rate after coating on the machine and perform deviation analysis against the designed coating rate; d. Perform statistics on the deviation analysis. If the deviation is within 0±1%, the coating rate prediction model can be determined by at least two cycles of verification. Otherwise, the coating rate prediction model needs to be calibrated and then at least two cycles of verification are required to determine the coating rate prediction model.

2. The method for predicting the coating rate of finished product of reconstituted tobacco leaves in papermaking process according to claim 1, characterized in that: The key physical indicators of the coating solution in step S1 include temperature, viscosity, solid content, sugar refractive index, suspended matter and density.

3. The method for predicting the coating rate of finished product of reconstituted tobacco leaves in papermaking process according to claim 1, characterized in that: According to the viscosity test values ​​of the coating liquid, the viscosity test values ​​of the concentrate corresponding to the coating liquid, and the viscosity test values ​​of additives 1 and 2, data with gradients are selected for regression analysis to establish the coating liquid viscosity prediction model described in step 2.

4. The method for predicting the coating rate of finished product of reconstituted tobacco leaves in papermaking process according to claim 1, characterized in that: According to the solid content test values ​​of the coating liquid, the solid content test values ​​of the concentrate corresponding to the coating liquid, and the solid content test values ​​of additives 1 and additives 2, data with gradients are selected for regression analysis to establish the coating liquid solid content prediction model described in step 2.

5. The method for predicting the coating rate of finished product of reconstituted tobacco leaves in papermaking process according to claim 1, characterized in that: According to the sugar refractive index detection value of the coating liquid, the sugar refractive index detection value of the concentrate corresponding to the coating liquid, and the sugar refractive index detection value database of additives 1 and additives 2, data with a gradient are selected for regression analysis to establish the sugar refractive index prediction model of the coating liquid described in step 2.

6. The method for predicting the coating rate of finished product of reconstituted tobacco leaves in papermaking process according to claim 1, characterized in that: According to the suspended matter detection value of the coating liquid, the suspended matter detection value of the concentrate corresponding to the coating liquid, and the suspended matter detection value database of additives 1 and additives 2, data with gradients are selected for regression analysis to establish the coating liquid suspended matter prediction model described in step 2.

7. The method for predicting the coating rate of finished product of reconstituted tobacco leaves in papermaking process according to claim 1, characterized in that: The verification and correction of the four coating liquid index prediction models in step S2 specifically include the following steps: A. Based on the coating liquid viscosity prediction model, calculate the coating liquid viscosity prediction value under a given concentrate viscosity and coating formula; based on the coating liquid solid content prediction model, calculate the coating liquid solid content prediction value under a given concentrate solid content and coating formula; based on the coating liquid sugar refractive index prediction model, calculate the coating liquid sugar refractive index prediction value under a given concentrate sugar refractive index and coating formula; based on the coating liquid suspended matter prediction model, calculate the coating liquid suspended matter prediction value under a given concentrate suspended matter and coating formula; B. preparing a coating solution using the concentrate and coating formula given by the four physical index prediction models of the coating solution described in step A, and testing the viscosity, solid content, sugar refractive index, and suspended matter of the coating solution; C. performing deviation analysis on the predicted values ​​of the physical indicators of the coating liquid obtained in step A and the detected values ​​of the physical indicators of the coating liquid obtained in step B; D. Statistically analyze the deviation analysis results. If the deviations of the coating solution's viscosity, solid content, sugar refractive index, and suspended solids are within 0±1%, the prediction model can be determined by performing at least two validation cycles. Otherwise, the prediction model needs to be corrected and then validated at least twice more before the prediction model can be determined.

8. The method for predicting the coating rate of finished product of reconstituted tobacco leaves in papermaking process according to claim 1, characterized in that: A coating rate prediction model was established by performing regression analysis on the data in the database of the production record statistics of the absolute dry weight of the film base before coating, the production record statistics of the viscosity of the coating liquid, the production record statistics of the sugar refractive index of the coating liquid, the production record statistics of the solid content of the coating liquid, the production record statistics of the suspended matter of the coating liquid and the production record statistics of the coating rate.

Images