Methods for producing chemical products using regression models
By generating a combined regression model and performing data cleaning, the instability of properties caused by climatic factors in polyurethane foam production is solved, and precise control of the properties of polyurethane foam and the stability of the production process is achieved.
Patent Information
- Application Number
- CN202080055733.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-08-05
- Filing Date
- 2020-07-28
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2040-07-28
AI Technical Summary
In the production of polyurethane foam, climate factors change lead to instability in the properties of the final product, lack of effective data cleaning and preprocessing, lack of high-order regression sub-participation and automation solution deployment, resulting in difficult formulation correction and affecting production efficiency and products. quality.
By generating a combined model based on regression model, using data sets of multiple production instances, data cleaning and preprocessing are carried out, interactive higher-order regressors are introduced, and the composition and process conditions of the reaction mixture under climatic conditions are automatically calculated to achieve the polyurethane foam. Accurate adjustment of properties.
Accurate control of the properties of polyurethane foam is achieved, the stability and efficiency of the production process is improved, the generation of unqualified products is reduced, and the ability to respond to climate factors is enhanced.
Smart Images

Figure CN114144734B_ABST
Abstract
Description
[0001] The present invention relates to a method for producing a chemical product from a reaction mixture. In the method, a combined transformed regression model is used. This allows chemical or process parameters to be adjusted in response to local climate parameters such as humidity, temperature or air pressure. The present invention also relates to a system configured for producing a chemical product and a computing platform configured for producing a chemical product.
[0002] In the production of external polyurethane foams, climatic factors can lead to changes in the properties of the final product, for example with respect to raw density and compression deflection. If the process is run over a long period of time where weather conditions may change, there is often no choice but to dispose of the unqualified material.
[0003] Before the advent of regression models, heuristic rules were used based on the experience of plant personnel. However, this individual experience could prove to be too inaccurate to account for the many climate factors that change simultaneously and their effects on the reaction mixture formulation and process parameters.
[0004] The publication "Flexible Polyurethane Slabstock Foam: The Influence of Formulation, Climatic Conditions and Storage Conditions on Foam Properties" by R. Schiffauer and C. den Heijer, Journal of Cellular Plastics Jan / Feb 1983, 61-64 reports:
[0005] “The following climate variables were found to have significant effects on foam properties: atmospheric pressure during production and absolute humidity during production. Absolute humidity during storage may also have a significant effect, but due to the high intercorrelation with absolute humidity during production, this effect may not be estimated separately. No significant effects of climate variables were found for compression set, cell count, and air flow.
[0006] The effects of these two main climate variables on the remaining foam properties are: higher (lower) atmospheric pressure leads to higher (lower) density; higher (lower) hardness; lower (higher) elongations at break; and lower (higher) tensile strength.
[0007] Higher (lower) absolute humidity does not lead to a change in density; it leads to a lower (higher) hardness; it leads to a higher (lower) elongation at break; it leads to a higher (lower) tensile strength.
[0008] The similarity of the coefficients for the different grades justifies the construction of a combined model using both formulation and climate variables. Thus, the most important foam properties can be calculated as follows:
[0009] ln(properties) = C0 + C1 ln(water) + C2 ln(foaming) + C3 ln(TDI index) + C4 ln(atmospheric pressure) + C5 ln(absolute humidity)".
[0010] The publication "Mathematical Property Prediction Models for Flexible Polyurethane Foams - A Comparison between Conventional Slabstock, High Resilience Slabstock and High Resilience Molded Foams" by R. Schiffauer, Journal of Cellular Plastics 1996, 32, pages 318-354, reports a group of foams prepared by varying a series of formulation parameters and:
[0011] "The data were analyzed mathematically by multiple linear regression techniques using various transformations of the original variables to obtain equations relating the properties to the predictor variables. As a first approach, all data were regressed in their original linear format. However, in most cases a better fit was obtained by transforming the predictor variables or the resulting properties, or both, from linear to logarithmic format. Further improvements in the resulting equations were achieved by using interaction terms between some of the predicted properties. When laboratory and production data were analyzed together, so-called "dummy variables" were assigned (zero for laboratory data and one for production results) in order to calculate the effect of scaling up on various properties."
[0012] The publication "Innovative Produktionsoptimierung für die Blockschaum-Fertigung (innovative production optimization for flexible foam production)" by Hubert Ehbing, Karl-Heinz Dörner, Hans-Friedrich Walter, Bolko Raffel and R. Näscher in the conference proceedings "PUR 2002-Automobil-Comfort-Struktur-Kühlen-Bauen", October 2002, pages 192-144 discusses a regression model for accounting for the so-called summer / winter effect in flexible foam production. The regression model is:
[0013]
[0014] Where y = compression hardness; x 1= NCO index; x2 = water; x3 = absolute humidity; x4 = T65 content; x5 = nozzle pressure; x6 = block height; x7 = belt speed.
[0015] The publications mentioned above fail to address the following four points: first, there is no data cleaning process and preprocessing before applying the regression model; second, there is a lack or missing involvement of higher-order regressors in the regression model; third, there is a lack or missing possibility to automatically and symbolically solve the formula for generating each recipe component affected by climate conditions from the regression model; and fourth, there is a lack or missing deployment of these formulas in a user interface or embedding in an FPGA in a production plant.
[0016] Without the first two points, the influence of climatic conditions may have an adverse effect on the recipe correction in the production plant. The last two points are useful for either giving advice to the plant technician to correct the recipe, or for automatically performing the correction.
[0017] The present invention has the object of at least partially overcoming the drawbacks of the prior art. In particular, the present invention has the object of providing the production of chemical products from reaction mixtures, wherein the influence of atmospheric parameters on the final product can be compensated by varying the composition of the reaction mixture in a simplified manner.
[0018] This object is achieved by a method according to claim 1, a system according to claim 13, and a computing platform according to claim 14. Advantageous embodiments are the subject matter of the dependent claims. Unless the context clearly indicates otherwise, they can be combined freely. It is also within the scope of the invention that the subject matter of the embodiments described in conjunction with the method, in particular the subject matter described in the dependent method claims, can also be applied to the system and computing platform according to the invention. This also extends to combinations of embodiments.
[0019] Thus, a method for producing a chemical product from a reaction mixture comprising at least two components comprises:
[0020] I) providing a data set comprising a plurality of production instances, each production instance comprising data on: (a) a composition and / or process conditions of the reaction mixture, (b) at least one climate parameter at a production site of the product, and (c) at least one target physical property of the product;
[0021] II) generating a regression model for each target physical property of the product using the data set, wherein in the regression model, the corresponding target physical property is a dependent variable and at least a portion of other data in the data set are independent variables, the other data excluding any remaining target physical properties;
[0022] III) transforming each regression model to obtain a transformed regression model, wherein a) the composition of the reaction mixture and / or b) the process conditions of the reaction depend on at least one climatic parameter at the production site of the product, and each target physical property is assigned an expected value;
[0023] IV) combining the transformed regression models to obtain a combined regression model;
[0024] V) measuring at said production site at least one climate parameter used in said combined regression model;
[0025] VI) using the combined regression model and at least one measured climate parameter to calculate a) a composition of the reaction mixture and / or b) a process condition of the reaction;
[0026] VII) composing said reaction mixture and / or process conditions according to said calculation; and
[0027] VIII) reacting the reaction mixture to obtain the chemical product.
[0028] The method according to the invention may be at least partly a computer-implemented method.Preferably, steps I to VII are computer-implemented steps.
[0029] The data set in step I) of the method comprises (represents) a plurality of production instances and may be obtained from data collected as a routine process and product monitoring. A production instance may be a single batch or a single campaign in the production of a chemical product. In the case of a continuous production process, a production instance may be considered as a time span in the continuous process. In general, the quality of the data set will improve with more details about the production instances.
[0030] In addition to data on the composition of the reaction mixture and / or process conditions applicable to the respective production run, a production instance also comprises data on at least one (preferably two or more) climate parameters at the production site of the product during the respective production run. Since this climate data is attributed to a production instance, it can also be described as the respective climate parameter recorded during the production of the product in the individual instance, such as air pressure, etc. In view of the recorded climate parameter(s), the production of the chemical product is preferably a process in which the reaction mixture or one or more components of the reaction mixture are at least partially in contact with the atmosphere.
[0031] Finally, a production instance includes data on at least one (preferably two or more) target physical properties of the product for the corresponding production run. Such target physical properties in a production instance and therefore in the data set can be considered as properties that are part of the specification of the product, which have been determined by measurements to populate the data set and will be the target to be achieved in future production of the chemical product.
[0032] In step II), a regression model is generated for each target physical property of the product using the data set. Thus, the target physical property is the output and the other data in the data set are the inputs. If other target physical properties exist in the data set, they are ignored. Examples of regression models include least squares sum and maximum likelihood.
[0033] In step III), each regression model obtained in step II) is transformed into a model in which at least one climate parameter at the production site of the product is input. It goes without saying that the at least one climate parameter is also included in the data set as described in step I). In addition, each target physical property is assigned a value, thereby reflecting the desired result of the chemical reaction. It also goes without saying that one or more target physical properties are included in the data set as described in step I). The output of each transformed model is the composition of the reaction mixture and / or the process conditions of the reaction necessary to achieve the target physical properties. In summary, step III) n The regression model is transformed inton A transformed regression model.
[0034] In addition, the involvement of higher-order regressors with interactions can also improve the fit of the regression model. Such a regression model can have the following form:
[0035]
[0036] Using the fitted regression model for each property y i = f ( x j ), a symbolic solver can be applied to receive the parameters for the correction recipe x 1 and x 2's solution.
[0037] It is advantageous to use data cleaning and preprocessing techniques to enhance the data quality of historical data and improve the fit of the regression model(s). Data cleaning techniques preferably focus on finding outliers and impute these outliers either using default values or using statistical descriptors such as the mean and median of the corresponding parameters.
[0038] Therefore, in one embodiment, before step I), statistical methods are used to process the data set to remove outliers. Examples of suitable statistical methods include analysis of variance, mutual information, and Z-scores. Outliers can be removed based on values such as median or expected values derived from the above methods. The size of the feature space created by the properties of the manufacturing process details, atmospheric conditions, process details, and the final product can also be reduced to basic features. Statistical analysis also allows, for example, identification of common features or clusters in the data by using scatter plots or principal component methods.
[0039] In another embodiment, step II) comprises:
[0040] Divide the data set into a training data set and a test data set,
[0041] Each regression model is generated using the training dataset, and
[0042] The regression model is validated against the test data set.
[0043] The coefficient of determination R 2 , mean absolute error (MAE), or root mean square error (RMSE) to cross-validate the candidates for the regression model validated for the test data set. Then, the candidate with the highest R 2 , the regression model for each target physical property with the lowest MAE and RMSE values.
[0044] In another embodiment, the data sets are grouped into clusters, and each cluster is divided into a training data set and a test data set for generating the regression model.
[0045] In another embodiment, at least one of steps I) to VI) comprises: providing interaction with a user via a graphical user interface. For example, the calculated recipe for recipe parameters can be integrated in a user interface, where a plant technician inserts his standard recipe. At the same time, climate conditions are measured by sensors implemented in the production plant and in the foam storage area. The recipe consumes the measured information of the standard recipe with the plant technician, and then calculates a climate-corrected recipe. The climate-corrected recipe is visualized in the user interface, where the plant technician can manually correct the set points.
[0046] In another scenario, the calculated recipe is implemented in an FPGA that controls the setpoints of a machine. The factory technician inserts his standard recipe in the user interface. The FPGA consumes the inserted recipe with the measured climate information from the above sensors. The FPGA can now change the setpoints of the machine without interaction with the factory technician.
[0047] In another embodiment, steps V) through VIII) occur in 60 seconds or less.
[0048] In another embodiment, at least two target physical properties are considered.
[0049] In another embodiment, at least two climate parameters are considered.
[0050] In another embodiment, the at least one climate parameter at the production site is air pressure, air temperature, relative humidity of the air, or a combination thereof.
[0051] In another embodiment, the at least one target physical property is the original density according to DIN EN ISO 845 and / or the compression load deflection at 40% compression according to EN ISO 3386.
[0052] In another embodiment, the chemical product is polyurethane foam, and the reaction mixture comprises polyisocyanates, polyisocyanate-reactive compounds and blowing agents. The example of suitable polyisocyanates comprises TDI, MDI, H12-MDI, XDI, HDI, IPDI and derivatives thereof, such as oligomers, NCO-terminated prepolymers, biuret, uretdione, allophanate and isocyanurate. The example of suitable polyol comprises polyester polyol, polyester polyol, polycarbonate polyol, polyether ester polyol and polyester carbonate polyol. The blowing agent can be a physical blowing agent such as cyclopentane, or a chemical blowing agent such as water or formic acid.
[0053] In another embodiment, forming the reaction mixture in step VII) comprises: starting from a predefined standard reaction mixture composition,
[0054] (a) increasing or decreasing the molar ratio of isocyanate groups to isocyanate-reactive groups;
[0055] (b) increasing or decreasing the amount of blowing agent;
[0056] (c) increasing or decreasing the relative amounts of physical and chemical blowing agents with respect to each other; or
[0057] (d) combinations thereof.
[0058] In another embodiment, the method further comprises:
[0059] IX) measuring at least one climate parameter used in the combined regression model at the production site after step VIII has been initiated;
[0060] X) calculating the composition of the reaction mixture or the process conditions of the reaction using the combined regression model and at least one of the measured climate parameters of step IX);
[0061] XI) adjusting the reaction mixture or process conditions based on the calculation in step X, or deciding not to adjust the reaction mixture or process conditions; and
[0062] XII) after making an adjustment to the reaction mixture or process conditions, or a decision not to adjust the reaction mixture or process conditions in step XI, reacting the reaction mixture to obtain the chemical product.
[0063] Preferably, steps IX to XII are repeated at least once every 60 seconds, at least once every 30 seconds, or continuously.
[0064] In another embodiment, the method further comprises:
[0065] IX') measuring at least one climate parameter used in said combined regression model at said production site after step VIII has been initiated;
[0066] X') determining whether the measurement result in step IX' is different from a previous measurement result of said at least one climate parameter;
[0067] XI') calculating the composition of the reaction mixture or the process conditions of the reaction using the combined regression model and at least one of the measured climate parameters of step IX' if it has been determined that the measurement results are different;
[0068] XII') if such calculation in step XI' is made, adjusting the reaction mixture or process conditions according to said calculation in step XI'; and
[0069] XIII') If the calculation was made in step XI, then after the adjustments to the reaction mixture or process conditions were made in step XII', reacting the reaction mixture to obtain the chemical product.
[0070] Preferably, steps IX' to XI' are repeated at least once every 60 seconds, at least once every 30 seconds, or continuously.
[0071] It is also preferred that at least two climate parameters are measured in steps V and IX'.
[0072] The present invention furthermore relates to a system configured for producing a chemical product from a reaction mixture comprising at least two components, the system comprising:
[0073] One or more hardware processors configured by machine-readable instructions to:
[0074] providing a data set comprising a plurality of production instances, each production instance comprising data regarding: a composition of the reaction mixture, at least one climate parameter at a production site of the product, and at least one target physical property of the product;
[0075] generating a regression model for each target physical property of the product using the data set, wherein in the regression model, the corresponding target physical property is a dependent variable and other data in the data set are independent variables;
[0076] transforming each regression model to obtain a transformed regression model, wherein the composition of the reaction mixture and / or the process conditions of the reaction depend on at least one climatic parameter at the production site of the product, and each target physical property is assigned an expected value;
[0077] combining the transformed regression models to obtain a combined regression model;
[0078] measuring at least one climate parameter used in the combined regression model at the production site;
[0079] calculating a composition of the reaction mixture and / or a process condition of the reaction using the combined regression model and at least one measured climate parameter; and
[0080] composing the reaction mixture and / or process conditions according to the calculation;
[0081] and a reactor for reacting the reaction mixture to obtain the chemical product.
[0082] Another aspect of the invention is a computing platform configured to produce a chemical product from a reaction mixture comprising at least two components, the computing platform comprising:
[0083] a non-transitory computer-readable storage medium having executable instructions embodied thereon; and
[0084] One or more hardware processors configured to execute the instructions to:
[0085] providing a data set comprising a plurality of production instances, each production instance comprising data regarding: a composition of the reaction mixture, at least one climate parameter at a production site of the product, and at least one target physical property of the product;
[0086] generating a regression model for each target physical property of the product using the data set, wherein in the regression model, the corresponding target physical property is a dependent variable and other data in the data set are independent variables;
[0087] transforming each regression model to obtain a transformed regression model, wherein the composition of the reaction mixture and / or the process conditions of the reaction depend on at least one climatic parameter at the production site of the product, and each target physical property is assigned an expected value;
[0088] combining the transformed regression models to obtain a combined regression model;
[0089] measuring at least one climate parameter used in the combined regression model at the production site;
[0090] calculating a composition of the reaction mixture and / or a process condition of the reaction using the combined regression model and at least one measured climate parameter; and
[0091] The reaction mixture and / or process conditions are composed according to the calculations.
[0092] Preferably, the computing platform is in data communication with a reactor for reacting the reaction mixture to obtain the chemical product.
[0093] The present invention will be further described with reference to the following drawings, without wishing to be limited by them.
[0094] Figure 1
[0013] Illustrated is a system configured for producing a chemical product from a reaction mixture comprising at least two components, according to one or more implementations.
[0095] Figure 2
[0013] Methods for producing a chemical product from a reaction mixture comprising at least two components, according to one or more implementations, are illustrated.
[0096] Figure 1 A system 100 configured for producing a chemical product from a reaction mixture comprising at least two components is illustrated according to one or more implementations. In some implementations, the system 100 includes one or more servers 102. The server(s) 102 are configured to communicate with one or more client computing platforms 104 according to a client / server architecture and / or other architectures. The client computing platform(s) 104 are configured to communicate with other client computing platforms via the server(s) 102 and / or according to a peer-to-peer architecture and / or other architectures. A user can access the system 100 via the client computing platform(s) 104.
[0097] The server(s) 102 are configured by machine readable instructions 106. The machine readable instructions 106 include one or more instruction modules. The instruction modules include computer program modules. The instruction modules include one or more of the following: data providing module 108, regression model generation module 110, regression model transformation module 112, regression model combination module 114, composition calculation module 118, reaction mixture composition module 120, reaction mixture reaction module 122, climate parameter measurement module 124, reaction mixture process condition adjustment module 126, reaction mixture process condition adjustment module 128, measurement determination module 130 and / or other instruction modules.
[0098] The data providing module 108 is configured to provide a data set including a plurality of production instances. As a non-limiting example, each production instance includes data on the composition of the reaction mixture, at least one climate parameter at the production site of the product, and at least one target physical property of the product. Before providing the data set, a statistical method is used to process the data set to remove outliers. The at least one target physical property is an original density according to DIN EN ISO 845 and / or a compression load deflection under 40% compression according to EN ISO 3386.
[0099] The regression model generation module 110 is configured to generate a regression model for each target physical property of the product using the data set. Generating a regression model includes dividing the data set into a training data set and a test data set. Generating a regression model includes generating each regression model using the training data set. Generating a regression model includes validating the regression model against the test data set. The data set is grouped into clusters, and each cluster is divided into a training data set and a test data set for generating a regression model.
[0100] In a regression model, the corresponding target physical property is the dependent variable, and the other data in the dataset are the independent variables.
[0101] The regression model transformation module 112 is configured to transform each regression model to obtain a transformed regression model, wherein the composition of the reaction mixture and / or the process conditions of the reaction depend on at least one climate parameter at the production site of the product, and each target physical property is assigned an expected value. As a non-limiting example, the at least one climate parameter at the production site is air pressure and / or air temperature.
[0102] The regression model combination module 114 is configured to combine the transformed regression models to obtain a combined regression model.
[0103] The composition calculation module 118 is configured to calculate the composition of the reaction mixture and / or the process conditions of the reaction using the combined regression model and at least one measured climate parameter.
[0104] The composition calculation module 118 is alternatively configured to calculate the composition of the reaction mixture or the process condition of the reaction using the combined regression model and at least one measured climate parameter when measuring at least one climate if the measurements have been determined to be different.
[0105] The reaction mixture composition module 120 is configured to compose a reaction mixture and / or process conditions based on the calculations.
[0106] The reaction mixture reaction module 122 is configured to react the reaction mixture to obtain the chemical product after the adjustment is made. In adjusting the reaction, the reaction mixture or process conditions may adjust the reaction mixture if a calculation is made.
[0107] The climate parameter measurement module 124 is configured to measure at least one climate parameter used in the combined regression model at the production site before, during, and / or after reaction of the reaction mixture has begun.
[0108] The reaction mixture process condition adjustment module 126 is configured to adjust the reaction mixture or process conditions based on the calculations made when calculating the composition.
[0109] The measurement determination module 130 is configured to determine whether a measurement result when measuring at least one climate parameter is different from a previous measurement result of the at least one climate parameter.
[0110] In some implementations, at least two target physical properties are considered. In some implementations, at least two climate parameters are considered. In some implementations, air temperature, air pressure or relative air humidity, or a combination thereof are considered. In some implementations, as a non-limiting example, the chemical product is a polyurethane foam, and the reaction mixture includes polyisocyanates, polyisocyanate-reactive compounds and a blowing agent. In some implementations, forming the reaction mixture includes: starting from a predefined standard reaction mixture composition, increasing or decreasing the molar ratio of isocyanate groups to isocyanate-reactive groups.
[0111] In some implementations, composing the reaction mixture includes: starting from a predefined standard reaction mixture composition, increasing or decreasing the amount of blowing agent. In some implementations, composing the reaction mixture includes: starting from a predefined standard reaction mixture composition, increasing or decreasing the relative amounts of physical and chemical blowing agents with respect to each other.
[0112] In some implementations, server(s) 102, client(s) computing platform(s) 104, and / or external resources 132 are operably linked via one or more electronic communication links. For example, such electronic communication links are established at least in part via a network, such as the Internet and / or other networks. It will be appreciated that this is not intended to be limiting, and the scope of the present disclosure includes implementations in which server(s) 102, client(s) computing platform(s) 104, and / or external resources 132 are operably linked via some other communication medium.
[0113] A given client computing platform 104 includes one or more processors configured to execute computer program modules. The computer program modules are configured to enable an expert or user associated with a given client computing platform 104 to interface with the system 100 and / or external resources 132, and / or to provide other functionality attributed herein to the client computing platform(s) 104. As non-limiting examples, a given client computing platform 104 includes one or more of a desktop computer, a laptop computer, a handheld computer, a tablet computing platform, a netbook, a smart phone, a game console, and / or other computing platforms.
[0114] External resources 132 include information sources external to system 100, external entities participating in system 100, and / or other resources. In some implementations, some or all of the functionality attributed herein to external resources 132 is provided by resources included in system 100.
[0115] Server(s) 102 include electronic storage 134, one or more processors 136, and / or other components. Server(s) 102 include communication links or ports to enable information exchange with a network and / or other computing platforms. Figure 1 The illustration of the server(s) 102 in FIG. 1 is not intended to be limiting. The server(s) 102 include multiple hardware, software, and / or firmware components that operate together to provide the functionality attributed herein to the server(s) 102. For example, the server(s) 102 are implemented by a cloud as a computing platform on which the server(s) 102 operate together.
[0116] Electronic storage 134 may include non-transitory storage media that electronically stores information. The electronic storage media of electronic storage 134 includes one or both of a system storage device and / or a removable storage device that is provided integrally with (i.e., substantially non-removable) the server(s) 102 and / or a removable storage device that is removably connected to the server(s) 102 via, for example, a port (e.g., a USB port, a FireWire port, etc.) or a drive (e.g., a disk drive, etc.). Electronic storage 134 includes one or more of the following: optically readable storage media (e.g., optical disks, etc.), magnetically readable storage media (e.g., magnetic tape, magnetic hard drive, floppy disk drive, etc.), charge-based storage media (e.g., EEPROM, RAM, etc.), solid-state storage media (e.g., flash drives, etc.), and / or other electronically readable storage media. Electronic storage 134 includes one or more virtual storage resources (e.g., cloud storage, virtual private networks, and / or other virtual storage resources). Electronic storage 134 may store software algorithms, information determined by processor(s) 136, information received from server(s) 102, information received from client computing platform(s) 104, and / or other information that enables server(s) 102 to function as described herein.
[0117] Processor(s) 136 are configured to provide information processing capabilities in server(s) 102. As such, processor(s) 136 include one or more of: a digital processor, an analog processor, a digital circuit designed to process information, an analog circuit designed to process information, a state machine, and / or other mechanisms for electronically processing information. Figure 1 136 is shown as a single entity, but this is for illustrative purposes only. In some implementations, (one or more) processors 136 include multiple processing units. These processing units are physically located within the same device, or (one or more) processors 136 can represent the processing functions of multiple devices operating in coordination. (One or more) processors 136 are configured to: execute modules 108, 110, 112, 114, 118, 120, 124, 126 and / or 130 and / or other modules through software, hardware, firmware, some combination of software, hardware and / or firmware; and / or other mechanisms for configuring the processing capabilities on (one or more) processors 136. As used herein, the term "module" may refer to any component or collection of components that perform the functions attributed to the module. This includes one or more physical processors, processor-readable instructions, circuits, hardware, storage media, or any other components during the execution of processor-readable instructions.
[0118] It should be appreciated that although modules 108, 110, 112, 114, 118, 120, 124, 126, and / or 130 are Figure 1 130 are illustrated as being implemented within a single processing unit, but in implementations where the processor(s) 136 include multiple processing units, one or more of the modules 108, 110, 112, 114, 118, 120, 124, 126, and / or 130 are implemented remotely from the other modules. The description of functionality provided by different modules 108, 110, 112, 114, 118, 120, 124, 126, and / or 130 described below is for illustrative purposes and is not intended to be limiting, as any of the modules 108, 110, 112, 114, 118, 120, 124, 126, and / or 130 may provide more or less functionality than described. For example, one or more of the modules 108, 110, 112, 114, 118, 120, 124, 126, and / or 130 are eliminated and some or all of their functionality is provided by other modules of the modules 108, 110, 112, 114, 118, 120, 124, 126, and / or 130. As another example, the processor(s) 136 are configured to execute one or more additional modules that may perform some or all of the functionality attributed below to one of the modules 108, 110, 112, 114, 118, 120, 124, 126, and / or 130.
[0119] Figure 2 A method 200 for producing a chemical product from a reaction mixture comprising at least two components according to one or more implementations is illustrated. The operations of the method 200 presented below are intended to be illustrative. In some implementations, the method 200 is implemented with one or more additional operations not described and / or without one or more of the operations discussed.
[0120] In some implementations, method 200 is implemented in one or more processing devices (e.g., digital processors, analog processors, digital circuits designed to process information, analog circuits designed to process information, state machines, and / or other mechanisms for electronically processing information). The one or more processing devices include one or more devices that perform some or all of the operations of method 200 in response to instructions stored electronically on an electronic storage medium. The one or more processing devices include one or more devices configured by hardware, firmware, and / or software that are specifically designed to perform one or more operations of method 200.
[0121] Operation 202 includes providing a data set including a plurality of production instances. Each production instance includes data on the composition of a reaction mixture, at least one climate parameter at a production location of the product, and at least one target physical property of the product. According to one or more implementations, operation 202 is performed by one or more hardware processors configured by machine-readable instructions, the machine-readable instructions including a module that is the same or similar to data providing module 108.
[0122] Operation 204 includes: using the data set to generate a regression model for each target physical property of the product. In the regression model, the corresponding target physical property is a dependent variable, and other data in the data set are independent variables. According to one or more implementations, operation 204 is performed by one or more hardware processors configured by machine-readable instructions, and the machine-readable instructions include modules that are the same or similar to the regression model generation module 110.
[0123] Operation 206 includes: transforming each regression model to obtain a transformed regression model, wherein the composition of the reaction mixture and / or the process conditions of the reaction depend on at least one climate parameter at the production site of the product, and each target physical property is assigned an expected value. According to one or more implementations, operation 206 is performed by one or more hardware processors configured by machine-readable instructions, the machine-readable instructions including a module that is the same as or similar to the regression model transformation module 112.
[0124] Operation 208 includes combining the transformed regression models to obtain a combined regression model. According to one or more implementations, operation 208 is performed by one or more hardware processors configured by machine-readable instructions, which include modules that are the same or similar to the regression model combination module 114.
[0125] Operation 210 includes measuring at least one climate parameter used in the combined regression model at the production site. According to one or more implementations, operation 210 is performed by one or more hardware processors configured by machine-readable instructions including modules that are the same or similar to measurement module 116.
[0126] Operation 212 includes calculating the composition of the reaction mixture and / or the process conditions of the reaction using the combined regression model and at least one measured climate parameter. According to one or more implementations, operation 212 is performed by one or more hardware processors configured by machine-readable instructions, which include modules that are the same or similar to the composition calculation module 118.
[0127] Operation 214 includes composing a reaction mixture and / or process conditions based on the calculation. According to one or more implementations, operation 214 is performed by one or more hardware processors configured by machine-readable instructions including modules that are the same or similar to reaction mixture composition module 120.
[0128] Operation 216 includes reacting the reaction mixture to obtain the chemical product. According to one or more implementations, operation 216 is performed by one or more hardware processors configured by machine-readable instructions including modules that are the same or similar to the reaction mixture reaction module 122 .
[0129] The present invention will be further described in the following section without wishing to be limited by the process outlined therein. A data set having the structure given below can be used. Each row represents data for one production example of polyurethane foam. The foam can be produced by reacting a polyol component with a polyisocyanate component. The data can be normalized to a standard score .
[0130] Water: amount of water added to the polyol component as a chemical blowing agent [g]; ISO: NCO index of the reaction mixture; AbsHum_Mean: mean value of the absolute air humidity at the production site [g / m 3 ]; p: air pressure at the production site [kPa]; RD: original density of polyurethane foam [kg / m 3 ]; CH40: compressive load deflection of polyurethane foam at 40% compression [kPa].
[0131] .
[0132] In accordance with the method according to the invention, each row of the table represents a production example. The composition of the reaction mixture as related to the method of the invention is recorded in the columns "ISO" and "water". Similarly, the climate parameters are "AbsHum_Mean" and "p", and the target physical properties are "RD" and "CH40".
[0133] For the original density "RD", the following regression model can be generated:
[0134] RD = constant_01
[0135] + constant_02∙X1
[0136] + constant_03∙X2
[0137] + constant_04∙X4
[0138] According to the method according to the invention, the original density is the dependent variable and part of the other data (ISO, water and air pressure) are independent variables. Absolute humidity is not considered. Furthermore, also according to the method according to the invention, the compressive load deflection has been excluded as other target physical properties.
[0139] For the compression hardness "CH40", the following regression model was generated:
[0140] CH40 = constant_05
[0141] + constant_06∙X1
[0142] + constant_07∙X2
[0143] + constant_08∙X3
[0144] + constant_09∙X4
[0145] + constant_10∙X1 2
[0146] + constant_11∙X2 2
[0147] + constant_12∙X1∙X2
[0148] According to the method according to the invention, the compression load deflection is the dependent variable and the other data (ISO, water, absolute humidity and air pressure) are the independent variables. Furthermore, also according to the method according to the invention, the original density has been excluded as the other target physical property.
[0149] The regression model is transformed as follows. The subscript "_nr" means "new recipe", i.e. it is related to the recipe of the polyurethane foam after taking into account the climate correction (absolute humidity and air pressure). The constants constant_01 to constant_33 can be positive or negative and have small values, usually from −2 to 2.
[0150] RD − RD_nr = 0; in expanded form:
[0151] (constant_01
[0152] + constant_02∙X1
[0153] + constant_03∙X2
[0154] + constant_04∙X4)
[0155] − (constant_01
[0156] + constant_02∙X1_nr
[0157] + constant_03∙X2_nr
[0158] + constant_04∙X4_nr)
[0159] = 0
[0160] Here RD_nr has new climate correction components X1_nr, X2_nr and X4_nr.
[0161] CH40 − CH40_nr = 0; in expanded form:
[0162] (constant_05
[0163] + constant_06∙X1
[0164] + constant_07∙X2
[0165] + constant_08∙X3
[0166] + constant_09∙X4
[0167] + constant_10∙X1 2
[0168] + constant_11∙X2 2
[0169] + constant_12∙X1∙X2)
[0170] − (constant_05
[0171] + constant_06∙X1_nr
[0172] + constant_07∙X2_nr
[0173] + constant_08∙X3_nr
[0174] + constant_09∙X4_nr
[0175] + constant_10∙X1_nr 2
[0176] + constant_11∙X2_nr 2
[0177] + constant_12∙X1_nr∙X2_nr)
[0178] = 0
[0179] Among them, CH40_nr also has new climate correction components X1_nr, X2_nr and X4_nr.
[0180] The transformed regression model is composed as follows (sqrt: square root):
[0181] X1_nr = constant_13
[0182] + constant_14∙X2
[0183] + constant_14∙X4_nr
[0184] + constant_15∙X1
[0185] − constant_14∙X4
[0186] + constant_16∙sqrt(
[0187] constant_17
[0188] + constant_18∙X4_nr
[0189] +constant_19∙X2
[0190] + constant_20∙X3_nr
[0191] − constant_18∙X4
[0192] − constant_21∙X1
[0193] − constant_20∙X3
[0194] + constant_22∙X2∙X4_nr
[0195] + constant_23∙X1∙X4_nr
[0196] − constant_22∙X2∙X4
[0197] − constant_23∙X1∙X4
[0198] − constant_24∙X4∙X4_nr
[0199] − constant_25∙X1∙X2
[0200] + constant_26∙X1 2
[0201] + constant_27∙X4 2
[0202] + constant_27∙X4_nr 2
[0203] + constant_28∙X2 2 )
[0204] X2_nr = − constant_29
[0205] + constant_30∙X2
[0206] + constant_31∙X1
[0207] + constant_32∙X4_nr
[0208] − constant_32∙X4
[0209] − constant_33∙sqrt(
[0210] constant_17
[0211] + constant_18∙X4_nr
[0212] + constant_19∙X2
[0213] + constant_20∙X3_nr
[0214] − constant_18∙X4
[0215] − constant_21∙X1
[0216] − constant_20∙X3
[0217] + constant_22∙X2∙X4_nr
[0218] + constant_23∙X1∙X4_nr
[0219] − constant_22∙X2∙X4
[0220] − constant_23∙X1∙X4
[0221] − constant_24∙X4∙X4_nr
[0222] − constant_25∙X1∙X2
[0223] + constant_26∙X1 2
[0224] + constant_27∙X4 2
[0225] + constant_27∙X4_nr 2
[0226] + constant_28∙X2 2 )
[0227] X3_nr and X4_nr are the new measured absolute humidity and air pressure values. X3 and X4 are standard laboratory climate conditions.
Claims
1. A method for producing a chemical product from a reaction mixture comprising at least two components, comprising: I) providing a data set comprising a plurality of production instances, each production instance comprising data on: (a) a composition and / or process conditions of the reaction mixture, (b) at least one climate parameter at a production site of the product, and (c) at least one target physical property of the product; II) generating a regression model for each target physical property of the product using the data set, wherein in the regression model, the corresponding target physical property is a dependent variable and at least a portion of other data in the data set are independent variables, the other data excluding any remaining target physical properties; III) transforming each regression model to obtain a transformed regression model, wherein a) the composition of the reaction mixture and / or b) the process conditions of the reaction depend on at least one climatic parameter at the production site of the product, and each target physical property is assigned an expected value; IV) combining the transformed regression models to obtain a combined regression model; V) measuring at said production site at least one climate parameter used in said combined regression model; VI) using the combined regression model and at least one measured climate parameter to calculate a) a composition of the reaction mixture and / or b) a process condition of the reaction; VII) formulating the reaction mixture and / or process conditions according to the calculation; as well as VIII) reacting the reaction mixture to obtain the chemical product.
2. The method according to claim 1, wherein before step I), the data set is processed using statistical methods to remove outliers.
3. The method according to claim 1, wherein step II) comprises: Divide the data set into a training data set and a test data set, Each regression model is generated using the training dataset, and The regression model is validated against the test data set. 4 . The method according to claim 3 , wherein the data set is grouped into clusters, and each cluster is divided into a training data set and a test data set for generating the regression model.
5. The method of claim 1, wherein the at least one climate parameter at the production site is air pressure, air temperature, relative humidity of air, or a combination thereof.
6. The method according to claim 1, wherein the at least one target physical property is the original density according to DIN EN ISO 845 and / or the compressive load deflection at 40% compression according to EN ISO 3386.
7. The method of claim 1, wherein the chemical product is a polyurethane foam and the reaction mixture comprises a polyisocyanate, a polyisocyanate-reactive compound, and a blowing agent.
8. The method according to claim 1, wherein the step VII) of forming the reaction mixture comprises: Starting with a predefined standard reaction mixture composition, (a) increasing or decreasing the molar ratio of isocyanate groups to isocyanate-reactive groups; (b) increasing or decreasing the amount of blowing agent; (c) increasing or decreasing the relative amounts of physical and chemical blowing agents with respect to each other; or (d) combinations thereof.
9. The method according to claim 1, further comprising: IX) measuring at least one climate parameter used in the combined regression model at the production site after step VIII has been initiated; X) calculating the composition of the reaction mixture or the process conditions of the reaction using the combined regression model and at least one of the measured climate parameters of step IX); XI) adjusting the reaction mixture or process conditions based on the calculation in step X, or deciding not to adjust the reaction mixture or process conditions; and XII) after making an adjustment to the reaction mixture or process conditions, or a decision not to adjust the reaction mixture or process conditions in step XI, reacting the reaction mixture to obtain the chemical product.
10. The method of claim 9, wherein steps IX to XII are repeated at least once every 60 seconds, at least once every 30 seconds, or continuously.
11. The method according to claim 1, further comprising: IX') measuring at least one climate parameter used in said combined regression model at said production site after step VIII has been initiated; X') determining whether the measurement result in step IX' is different from a previous measurement result of said at least one climate parameter; XI') calculating the composition of the reaction mixture or the process conditions of the reaction using the combined regression model and at least one of the measured climate parameters of step IX' if it has been determined that the measurement results are different; XII') if such calculation in step XI' is made, adjusting the reaction mixture or process conditions according to said calculation in step XI'; and XIII') If the calculation was made in step XI, then after the adjustments to the reaction mixture or process conditions were made in step XII', reacting the reaction mixture to obtain the chemical product.
12. The method of claim 11, wherein steps IX' to XI' are repeated at least once every 60 seconds, at least once every 30 seconds, or continuously.
13. A system configured to produce a chemical product from a reaction mixture comprising at least two components, the system comprising: One or more hardware processors configured by machine-readable instructions to: providing a data set comprising a plurality of production instances, each production instance comprising data regarding: a composition of the reaction mixture, at least one climate parameter at a production site of the product, and at least one target physical property of the product; generating a regression model for each target physical property of the product using the data set, wherein in the regression model, the corresponding target physical property is a dependent variable and other data in the data set are independent variables; transforming each regression model to obtain a transformed regression model, wherein the composition of the reaction mixture and / or the process conditions of the reaction depend on at least one climatic parameter at the production site of the product, and each target physical property is assigned an expected value; combining the transformed regression models to obtain a combined regression model; measuring at least one climate parameter used in the combined regression model at the production site; calculating a composition of the reaction mixture and / or a process condition of the reaction using the combined regression model and at least one measured climate parameter; as well as composing the reaction mixture and / or process conditions according to the calculation; and a reactor for reacting the reaction mixture to obtain the chemical product.
14. A computing platform configured to produce a chemical product from a reaction mixture comprising at least two components, the computing platform comprising: a non-transitory computer-readable storage medium having executable instructions embodied thereon; as well as One or more hardware processors configured to execute the instructions to: providing a data set comprising a plurality of production instances, each production instance comprising data regarding: a composition of the reaction mixture, at least one climate parameter at a production site of the product, and at least one target physical property of the product; generating a regression model for each target physical property of the product using the data set, wherein in the regression model, the corresponding target physical property is a dependent variable and other data in the data set are independent variables; transforming each regression model to obtain a transformed regression model, wherein the composition of the reaction mixture and / or the process conditions of the reaction depend on at least one climatic parameter at the production site of the product, and each target physical property is assigned an expected value; combining the transformed regression models to obtain a combined regression model; measuring at least one climate parameter used in the combined regression model at the production site; calculating a composition of the reaction mixture and / or a process condition of the reaction using the combined regression model and at least one measured climate parameter; as well as The reaction mixture and / or process conditions are composed according to the calculations.
15. The platform of claim 14, wherein the computing platform is in data communication with a reactor for reacting the reaction mixture to obtain the chemical product.
