Systems and methods for generating a design space for chemical processes

Abstract

Disclosed here are methods and systems to generate a design space for determining one or more polymeric compositions or polymerization operation settings. In an embodiment, a method may include, in response to receipt of data, determining input parameters, the input parameters to include at least a minimum, a maximum, and an increment for each polymerization operation settings and / or compositions. The method may include generating, via iterative application of the experimental data and input parameters to trained machine learning models based on the input parameters, a design space to include one or more of a plurality of polymeric formulations, associated performances, and / or polymerization operation settings. The method may include, in response to a request for a polymeric formula and / or polymerization operation settings that meet or exceed a specified performance, determining a plurality of polymeric formulas and / or polymerization operation settings that meet or exceed the specified performance based on the design space.

Description

SYSTEMS AND METHODS FOR GENERATING A DESIGN SPACE FOR CHEMICAL PROCESSESFIELD OF DISCLOSURE

[0001] The present disclosure generally relates to systems and methods for generation of a design space and, particularly, to systems and methods for generation of an interactive design space associated with chemical processes or operations and compositions and the chemical products resulting therefrom, such as one or more polymerization operations, polymeric compositions or polymeric products, and enabling chemical process or operation and composition production optimization.BACKGROUND

[0002] Experimental data may be generated based on execution of various chemical process experiments and / or normal chemical processes and / or operations. However, generation of experimental data does not include measured data points for many potential variabilities (such as varying amounts of different components in a chemical composition), parameters, and / or settings. Thus, a data set for a particular chemical process or a chemical product produced by the chemical process may lack data points for numerous scenarios. Further, such a data set would be specific to that particular chemical process and / or the chemical product produced by the chemical process, thus any such data set would simply apply to one chemical process and / or chemical composition.

[0003] In addition, first principle based models or, in other words, physics based models, are based on first principles, for example, the laws of thermodynamics. If well validated, physics based models can be used to predict quantities not present in the original data set and may allow for some extrapolation. Furthermore, physics based models are highly interpretable since the parameters have a physical meaning. However, such physics based models exhibit various drawbacks, such as the use of expert knowledge in the field of interest for upfront behavior assumptions and / or the length of time to execute such a model, for example, hours or days, to achieve meaningful results. Thus, generating data points for the data set using a first principle model would not be practical, as the length of time to fill in a small number of data points would be impractical and such an output would account for asmall number of data points and would not include many different chemical processes and / or chemical compositions.BRIEF SUMMARY

[0004] In view of the foregoing, Applicant has recognized these problems and others in the art, and has recognized a need for enhanced systems and methods for generation of a design space or interactive design space associated with a plurality of chemical processes, such as polymerization operations, and enabling generation of updated chemical processes and / or chemical formula for a plurality of chemical products.

[0005] The present disclosure generally relates to systems and methods that address the relevant issues as described above, among other issues. In particular, such systems and methods may enable a user a to, in less time than in typical experimentation, generate a plurality of alternate chemical processes or operation parameters, chemical composition amounts, and / or chemical component types to produce a particular chemical product that meet or exceed selected chemical product performances. Such a design space may be interactive, further enabling generation of highly interpretable results, charts, and graphs associated with any one of a plurality of chemical processes or operations, chemical compositions to produce a chemical product, and the performances of those chemical products, (such as a polymerization process, polymeric formula, and / or polymeric products) selected by a user and / or computing device.

[0006] Such systems and methods may capture data, in real-time and / or, in another embodiment, may acquire or capture data from previous or historical chemical processes and / or operations data (for example, from a database or other type of storage), and preprocess the captured data to form a preprocessed data set. Such preprocessing may include fitting the data to one of a plurality of trained machine learning models (in other words, selecting a model that fits based on several variables, such as the type of data and / or other factors) and / or defining one or more input parameters for different variables of the chemical process or operation and / or different amounts and / or types of components of the chemical composition. After preprocessing, the preprocessed data set may be iteratively applied to one or more selected trained machine learning models based on the input parameters (the input parameters including, for example, at least a minimum, a maximum, and an increment or increment value for each of a plurality of polymerization operation settings and each of a plurality of polymericcomposition components). In another embodiment, the preprocessed data set (or a portion of the preprocessed data set) may first be applied to one or more physics based models, to produce one or more new polymerization operation parameters, new polymeric compositions, and / or new polymerization operation results or performances, and then the updated data set may be applied to the one or more trained machine learning models. In another embodiment, preprocessing may include determining which portions of the data set should be applied to which of the one or more trained machine learning models. Such a determination may be based on one or more types of data (such as temperatures, pressure, composition amount, and / or composition type, among other factors).

[0007] Iterative application of the data to the machine learning models may produce an output. The output may include, in an embodiment, a simulation, a chart, a curve, or some other value indicating a resulting chemical product’s performance and, in some embodiments, a predicted variable, product, property or chemical process parameter and / or chemical composition and / or type. Such systems and methods may include adding or updating a design space and / or updating the data set with the and / or based on the output of such trained machine learning models, and the resulting data set may be added to or stored in a design space or design space database. The design space or design space database may include or may be connected to a user interface. A user may input a request for updated chemical processes or operations, parameters associated with chemical processes or operations, performance of a chemical product, and / or components of a selected chemical composition. The design space database may, based on a input including, at least, the selected chemical product, output one or more graphical representations and / or data relating to a plurality of chemical processes and / or chemical formulations to produce the chemical product and, in an embodiment, meets or exceeds one or more specified parameters. In an embodiment, the design space (or a system communicatively connected to the design space) may utilize a nearest neighbor search algorithm, a linear search algorithm, a binary search algorithm, a hash table search algorithm, a genetic algorithm, and / or simulated annealing to produce the varying chemical processes and / or chemical compositions in the design space.

[0008] Thus, a user may be able to quickly obtain a plurality of different options relating to adjustment of a chemical process and / or chemical composition to obtain a chemical product that meets or exceeds selected parameters or performances. Further, the output of the design space may be interactive, allowing a user to adjust one or more parameters or composition, thereby adjusting thegraphical representations in real time and allowing a user to quickly refine or adjust a chemical process and / or chemical composition.

[0009] Accordingly, an embodiment of the disclosure is directed to a method to generate a design space for determining one or more polymeric compositions or polymerization operation settings. The method may include, in response to receipt of data that corresponds to a plurality of experimental data and / or historical polymerization operations data and / or polymerization process data, determining input parameters, the input parameters to include at least a minimum, a maximum, and an increment for each of a plurality of polymerization operation settings and each of a plurality of polymeric composition components. In an embodiment, plurality of historical polymerization operations data may include data indicative of a portion of the plurality of polymerization settings, a portion of the plurality of polymeric composition components, an output of one or more of the plurality of historical polymerization operations, and associated performances of the output of one or more of the plurality of historical polymerization operations. In an embodiment, polymerization operations and / or process data may include data from the full life cycle chain of a chemical or polymer, including, but not limited to, polymerization settings (for example, including input material streams data and / or process settings), polymerization conditions (for example, including temperature, pressure, flow rates, and / or other conditions), powder properties (for example, bulk density and / or flowability, among other powder properties), process parameters, polymer material properties, formulations of compounds, compounding process settings to produce compounds, settings to produce test specimen for properties measurements, product properties (for example, density, molecular chain lengths, and / or stiffness, among other product properties), compounded product properties, data of the conversion process to produce applications (for example, such as films, pipe, and / or injection molded products), extrusion process properties and / or parameters, blend process properties and / or parameters, and / or application performance measurements. The method may include generating a design space via iterative and incremental application of the data and input parameters corresponding to a current iteration to one or more trained machine learning models. Each application to the trained machine learning model may generate a synthetic output of a polymerization operation and a performance of the synthetic output based on the input parameters of the current iteration. The design space may include data indicative of outputs or synthetic outputs and one or more of a corresponding plurality of polymerization operation settings, a corresponding plurality of polymeric compositions, or a corresponding plurality of associated performances. The method may include, in response to a requestfor one or more of a polymeric composition or polymerization operation settings that meet or exceed a specified performance, determining one or more of a plurality of polymeric compositions or a plurality of polymerization operation settings that meet or exceed the specified performance based on the design space.

[0010] In an embodiment, the method may include generating one or more visualizations based on the one or more of the plurality of polymeric compositions or the plurality of polymerization operation settings that meet or exceed the specified performance based on the design space. The one or more visualizations may comprise one or more interactive plots or a tabular list.

[0011] In another embodiment, the method may include, prior to determining the one or more of the plurality of polymeric compositions or a plurality of polymerization operation settings, determining if the request includes unknown polymeric compositions or unknown polymerization operations settings. The method may further include, in response to a determination that the request includes unknown polymeric compositions or unknown polymerization operations settings: (a) generating a prompt for additional data, (b) in response to receipt of the additional data, retraining the trained machine learning models, and (c) updating the design space via application of data corresponding to the unknown polymeric compositions or unknown polymerization operations settings to the trained machine learning models.

[0012] In another embodiment, the polymerization operation settings may comprise one or more polymerization reactor settings, process settings, catalyst type, catalyst amount, temperature, pressure settings, flow rates, residence times, concentrations of reactants (for example, alpha-olefins, ethylene, propylene, 1 -butene, 1 -hexene, hydrogen), extrusion equipment settings, blending equipment settings, or solvent type and / or amount.

[0013] Another embodiment of the disclosure is directed to a method to generate a design space for determining one or more polymeric compositions or polymerization operation settings. The method may include receiving data that corresponds to a plurality of experimental data, real-time polymerization operations data, and historical polymerization operations data. The method may include determining input parameters based on the data, the input parameters to include at least a minimum value, a maximum value, and a increment value for each of a plurality of polymerization operation settings and each of a plurality of polymeric composition components. The method may include, while the maximum value for the plurality of polymerization operation settings has not been reached, iteratively performing the following steps. The steps may include, (a) while the maximumvalue for the plurality of polymeric composition components has not been reached, iteratively, (i) generating, via application of the data and input parameters corresponding to a current iteration to one or more trained machine learning models, a simulation of a polymerization operation based on the input parameters of the current iteration, (ii) determining new entries in the design space based on the simulation, (iii) generating a new selected amount of the component of the selected polymeric composition based on the input parameters, and (iv) repeating steps (i), (ii), and (iii) for the new selected amount of the component of the selected polymeric composition. The method may further include (b) generating new selected polymerization operation settings based on the input parameters and (c) repeating steps (a) and (b) for the new selected polymerization operation settings.

[0014] In another embodiment, the method may include generating a user interface to include a user input function that allows a user to obtain (a) one or more of a plurality of polymerization operation settings or a plurality of amounts of polymeric composition components and (b) an associated performance for each of the one or more of plurality of polymerization operation settings or the plurality of amounts of polymeric composition components. The method include, in response to a selection of one of the plurality of polymerization operation settings and one of the plurality of amounts of polymeric composition components, initiating a polymerization operation based on the selection.

[0015] Another embodiment of the disclosure is directed to a system to generate a design space for determining one or more polymeric compositions or polymerization operation settings. The system may include a communications circuitry configured to receive data corresponding to one or more of continuing polymerization operations, polymerization experiments, or historical polymerization operations. The system may include a pre-processing circuitry configured to determine input parameters including at least a minimum, a maximum, and an increment for each parameter corresponding to polymerization operations in the data and for each amount of a plurality of polymeric compositions in the data. The system may include a modeling circuitry configured to iteratively generate a design space via incremental application of the data and the input parameters corresponding to a current iteration to one or more trained machine learning models. The design space may include data indicative of outputs or synthetic outputs and one or more of a plurality of corresponding polymerization operation settings, a plurality of corresponding polymeric compositions, or a plurality of corresponding associated performances. The system may include a chemical operation controller configured to, in response to receipt of a request for one or more of apolymeric composition or polymerization operation settings that meet or exceed a specified performance, determine one or more of a plurality of variable polymeric compositions or a plurality of polymerization operation settings that meet or exceed the specified performance based on the design space.

[0016] In another embodiment, the system may include a visualization circuitry. The visualization circuitry may be configured to generate one or more visualizations based on the one or more of the plurality of variable polymeric compositions or the plurality of polymerization operation settings that meet or exceed the specified performance based on the design space. The visualization circuitry may be configured to display options to a user interface allowing selection of the one or more visualizations based a type of each of the one or more visualizations. The visualization circuitry may be configured to in response to selection of the one or more visualizations, display selected visualizations to the user interface.

[0017] In another embodiment, the system may include a design space circuitry configured to process a plurality of outputs of the one or more trained machine learning models to produce data formatted for the design space.

[0018] In another embodiment, the modeling circuitry may be configured to, in response to selection of one of the one or more of the plurality of the variable polymeric compositions and one of the plurality of polymerization operation settings, retrain the one or more trained machine learning models.

[0019] In another embodiment, one of the one or more trained machine learning models may include an image-based machine learning model trained based on a series of images of polymeric products, corresponding data, and acceptance or rejection of the polymeric products. The request may include a visual aspect of a polymeric product produced based on one or more of the polymeric compositions or the polymerization operation settings.

[0020] In another embodiment, the modeling circuitry may be configured to, in response to a missing portion of data in the one or more of the plurality of polymeric compositions, the plurality of associated performances, or the plurality of polymerization operation settings, generate the missing portion of data via a physics based model and based on the data and the design space.

[0021] Another embodiment of the disclosure is directed to a controller for generating a design space for determining one or more polymeric compositions or polymerization operation settings. The controller may include an input / output in signal communication with polymerization equipment. Thecontroller may be configured to, in response to receipt of experimental data, determine input parameters, the input parameters including at least a minimum, a maximum, and an increment of process parameters in the experimental data. The controller may further be configured to generate, via iterative and incremental application of the experimental data and the input parameters corresponding to a current iteration to one or more trained machine learning models based on the input parameters, a design space to include one or more of a plurality of polymeric compositions, associated performances, or polymerization operation settings. The controller may be configured to, in response to receipt of a request for one or more of a polymeric formula or polymerization operation settings that meet or exceed a specified performance, determining one or more of a plurality of variable polymeric formulas or a plurality of polymerization operation settings that meet or exceed the specified performance based on the design space. The controller may be configured to display one or more visualizations generated based on the one or more of the plurality of variable polymeric formulas or the plurality of polymerization operation settings that meet or exceed the specified performance based on the design space.

[0022] In another embodiment, the one or more visualizations include one or more of graphical visualizations or tabular visualizations. Further, the request includes an application critical-to-quality (CTQ) input, and wherein the controller is configured to convert the Application CTQ input to a material to process CTQ.

[0023] Additional and / or alternative objects, features and advantages of the present disclosure will become apparent to the skilled artisan from the figures, detailed description, and examples herein. Applicant notes, however, that the figures, detailed description, and examples, while indicating certain embodiments of the instant disclosure, are provided for illustrative purposes only and are not intended to be limiting or to imply a particular limitation. Moreover, certain changes and modifications within the spirit and scope of the disclosed technology will become apparent to those of ordinary in the relevant art from this detailed description.BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The disclosed aspects, features and advantages of the disclosure will become better understood with regard to the following descriptions, examples, claims, and accompanying drawings.Applicant notes, however, that the drawings illustrate certain embodiments of the disclosure and should not be considered limiting with regards to the breadth and scope of the disclosure:

[0025] FIG. l is a schematic diagram of a system for generating a design space, in accordance with certain embodiments of the present disclosure;

[0026] FIG. 2 is another schematic diagram of an apparatus for generating a design space, in accordance with certain embodiments of the present disclosure;

[0027] FIG. 3 is a schematic diagram of a controller to generate a design space, in accordance with certain embodiments of the present disclosure;

[0028] FIG. 4A and FIG. 4B are illustrations of a user interface to display results generated via the design space, in accordance with certain embodiments of the present disclosure;

[0029] FIG. 5A illustrates an exemplary information flow for virtual experimentation used with extruders;

[0030] FIG. 5B is a flow diagram for generating a design space, in accordance with certain embodiments of the present disclosure; and

[0031] FIG. 6A and FIG. 6B are other flow diagrams for generating a design space, in accordance with certain embodiments of the present disclosure.DETAILED DESCRIPTION

[0032] So that the manner in which the features and advantages of the embodiments of the systems and methods disclosed herein, as well as others that will become apparent, may be understood in more detail, a more particular description of embodiments of systems and methods briefly summarized above may be had by reference to the following detailed description of embodiments thereof, in which one or more are further illustrated in the appended drawings, which form a part of this specification. It is to be noted, however, that the drawings illustrate only various embodiments of the systems and methods disclosed herein and are therefore not to be considered limiting of the scope of the systems and methods disclosed herein as it may include other effective embodiments as well.

[0033] The present disclosure generally relates to systems and methods that address generation of design spaces and other relevant issues as described herein. In particular, such systems and methods may enable faster than typical analysis, such analysis providing a plurality of options that meet or exceed specified performance levels for a chemical product. Further, such a plurality of options maybe presented in an interactive way, such that a user may adjust or optimize the chemical compositions and / or chemical processes for a particular or selected chemical product. Thus, rather than performing lengthy experimentation, the user may obtain a more efficient and optimized chemical process and chemical composition in a relatively short amount of time (as compared to the time used for experimentation).

[0034] Further, the design space may be utilized to adjust chemical operation settings and / or chemical compositions, such as, in a non-limiting example, adjustment of temperature and / or pressure within a polymerization reactor, adjustment of chemicals or polymers (such as amount or type) included in a polymeric composition or in the manufacture of the polymeric composition, and / or the type and / or the amount of catalyst utilized to produce the chemical compositions, among other adjustments to chemical operation settings.

[0035] The present disclosure generally relates to systems and methods that address the relevant issues as described above, among other issues. In particular, such systems and methods may enable a user a to, in less time than in typical experimentation, generate a plurality of alternate chemical process or operation parameters, chemical composition amounts, and / or chemical component types to produce a particular chemical product. Such a design space may be interactive, further enabling generation of highly interpretable results, charts, and graphs associated with any one of a plurality of chemical processes or operations and chemical compositions to produce a chemical product, such as a polymerization process and / or polymeric formula, selected by a user and / or computing device.

[0036] Such systems and methods may utilize data (whether captured in real-time or from an existing data source) to build a design space. Such a design space may include data indicative of a plurality of outputs and / or synthetic outputs (for example, the output of a polymerization operation, such as the output of conversion of chemicals, an extrusion operation, or blending operation) and one or more of a corresponding plurality of polymerization operation settings, a corresponding plurality of polymeric composition components, or a corresponding plurality of associated performances. The systems and methods may utilize such data to determine or define input parameters. The input parameters may include (for each of one or more different types of input parameters) a minimum value or starting value, a maximum value or end point, and / or an increment value (in other words, the value to increment a variable until a maximum is reached). The system or method may define the input parameters for a number of values based on various factors. For example, the variables selected for input parameters may be based on the performance of the end chemical product that is sought (forexample, tensile strength, modulus of elasticity, toughness, pattern or visible pattern, and / or temperature resistance or threshold, among other performance variables). In other words, the variables that may be applied to one or more trained machine learning models may include variables with or determined to provide or produce a desired performance.

[0037] As a non-limiting example, the production of a selected chemical product may include up to 28 components or ingredients (in other examples, more or less components or ingredients may be utilized). An amount of each of the components or ingredients may be varied or kept static prior to application to a machine learning model. Further, selected corresponding outputs (in other words, material properties) may be desired for such a chemical product, for example, up to 7 corresponding outputs (and, in other examples, additional corresponding outputs). The increment value for each of the amounts of the components or ingredients and the corresponding outputs may be as little as tenth of a digit, a hundredth of a digit, or an even smaller amount. Thus, thousands, and in some cases even more, applications of some variation of the components or ingredients and corresponding outputs may be applied to one or more machine learning models. In further examples, the incremental value for any of the components or ingredients and corresponding outputs may be small (such as tenths or hundredths of a digit, or even smaller) and / or random.

[0038] The different variables and / or combinations of variables may be applied to different trained machine learning models. The systems and methods may determine which model to apply the different variables and / or combinations of variables based on a number of factors, including, but not limited to, the type of variable and / or models previously used for similar variables, among other variables.

[0039] Application of such data to the one or more machine learning models may be iterative and incremental, according to the input parameters. For example, an amount of a particular component of a chemical composition may be varied or adjusted according to the increment value, then applied to one or more trained machine learning models, until the maximum value has been reached. After the maximum value has been reached, then another variable of a chemical process or operation may be incremented, and then the amount of the particular component of a chemical composition may be varied according to the increment value until the maximum value has been reached again, and so on. Such an example is not a limitation and it will be understood that any variable may be incremented, while others remain static, then applied to one or more machine learning models and incrementedagain until the incremented variable reaches the maximum value. Further, after that maximum value has been reached, then one of the other static variables may be incremented and the process repeated.

[0040] Such an iterative application of the data to the machine learning models or trained machine learning models may produce an output. The output may include, in an embodiment, a simulation, a chart, a curve, or some other value indicating a predicted variable and / or a chemical performance and, in some embodiments, predicting a chemical process parameter and / or chemical composition and / or type. Such systems and methods may include adding or updating a design space and / or updating the data set with the output of such trained machine learning models, and the resulting data set may be added to or stored in a design space or design space database. The design space or design space database may include or may be connected to a user interface. A user may input a request for updated chemical processes or operations, parameters associated with chemical processes or operations, performance of a chemical product, and / or components of a selected chemical composition. The design space database may, based on a input including at least the selected chemical product and the chemical product’s performance, output one or more graphical representations and / or data relating to a plurality of chemical processes and / or chemical formulations to produce the chemical product that meets or exceeds one or more specified parameters. In an embodiment, the design space may utilize a nearest neighbor search algorithm or other search algorithm to produce the varying chemical processes and / or chemical compositions.

[0041] Thus, a user may be able to quickly obtain a plurality of different options relating to adjustment of a chemical process and / or chemical composition to obtain a chemical product that meets or exceeds selected parameters. Further, the output of the design space may be interactive, allowing a user to adjust one or more parameters or composition, thereby adjusting the graphical representations in real time and allowing a user to quickly refine or adjust a chemical process and / or chemical composition.

[0042] The following definitions are provided for clarifying certain terms and phrases of the present disclosure and are in no way intended to unnecessarily or unduly limit any embodiments and aspects related thereto.

[0043] The use of the words “a” or “an” when used in conjunction with the term “comprising,” “including,” “containing,” or “having” in the claims or the specification may mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.”

[0044] The words “comprising” (and any form of comprising, such as “comprise” and “comprises”), “having” (and any form of having, such as “have” and “has”), “including” (and any form of including, such as “includes” and “include”) or “containing” (and any form of containing, such as “contains” and “contain”) are inclusive or open-ended and do not exclude additional, unrecited elements or method steps.

[0045] FIG. 1 is a schematic diagram of a system 100 for generating a design space, in accordance with certain embodiments of the present disclosure. Such a design space may be utilized to generate variable chemical processes or operation parameters and / or chemical composition components for a chemical product that meets or exceeds selected performances based on an input from a computing device (such as from computing device 124 A, 124B, or up to 124N via user interface (UI) 122 A, 122B, or up to 122N). The design space may be a data base that includes data indicative of a plurality of outputs (for example, physically generated outputs and synthetically generated outputs, in other words, outputs determined via application of data to a model) and data indicative of one or more of a corresponding plurality of polymerization operation settings, a corresponding plurality of polymeric composition components, or a corresponding plurality of associated performances. Such a system 100 may include a design space system 102. The design space system 102 may include a processor 104 or a plurality of processors and a memory 106. The memory 106 may store or include instructions. The instructions may include preprocessing instructions 108, design space instructions 110, machine learning models 112 (or trained machine learning models), physics based models 114, and / or visualization instructions 120. The design space system 102 may be communicatively connected to a controller 126 and / or a database 132. The controller 126 may connect to various equipment 130 and / or a polymerization reactor 128 (or other types of reactors) located at a plant. The controller 126 may further connect to various sensors, analyzers, and / or meters associated with the equipment 130 and / or polymerization reactor 128 (among other components and / or devices located at a plant or other location), thus enabling the controller 126 and the design space system 102 to capture data as polymerization operations are conducted and allowing a design space to be continuously updated and / or refined over time. In another embodiment, the design space system 102 may directly connect to the equipment 130 and polymerization reactor 128 (and / or the sensors, analyzers, and / or meters associated therewith and / or other components or devices). In such examples, the design space system 102 may be or may include the functionality of a controller (such as, for example, controlling operating parameters and / or settings of the equipment 130 and / or the polymerization reactor 128). Inan embodiment, the sensors, analyzers, and / or meters may include temperature sensors, pressure sensors, flow meters, densitometers, spectrographic analyzers, gas chromatography devices, other chemical analyzers, and / or other sensors or meters to determine or measure characteristics of the equipment 130 and / or the polymerization reactor 128 or the fluids therein. In another embodiment, the equipment 130 may include pumps, compressors, control valves, extruders, other types of reactors, furnaces, heaters, coolers, distillation columns, separation columns, heat exchangers, flash vessels, purge tanks, storage vessels, decanters, cyclones, pressure sensors, membrane separators, flow sensors, temperate sensors, composition measurement devices (such as gas chromatographs, Raman spectroscopy, NMR, or other measurement devices, as will be understood by one skilled in the art), and / or other equipment utilized at a plant configured to conduct polymerization operations. Further, the design space system 102, as described in further detail below, may connect to computing devices 124A, 124B, and up to 124N (for example, via UIs 122A, 122B, and up to 122N). The design space system 102 may generate the UIs 122 A, 122B, and up to 122N to enable the computing devices 124A, 124B, and up to 124N to interact with the design space system 102. For example, a user may request, from a computing device 124A, 124B, or up to 124N via the UI 122A, 122B, or up to 122N, a number of different compositions and / or process or operation parameters for a particular chemical product. The design space system 102 may utilize instructions (such as visualization instructions 120) stored in memory 106 to generate such a UI 122A, 122B, and up to 122N and / or other visualizations and interactive functionality, as described in further detail below.

[0046] As noted, the memory 106 may include preprocessing instructions 108. In embodiments, the preprocessing instructions 108 may be configured to, when executed by the processor 104, receive data from the controller 126 (and / or, in another embodiment, directly from the equipment 118, polymerization reactor 116, and the sensors, meters, and / or analyzers associated therewith), from the computing devices 124A, 124B, and up to 142N, and / or from the database 132, among other data sources. Prior to generation of the design space, the design space system 102 may obtain, request, and / or receive a large amount of data corresponding to a plurality of different chemical products, the chemical processes or operations to make those chemical products, and the different compositions or formula to make those chemical products, as well as the performance exhibited by the chemical product after being produced according to the chemical processes or operations and / or chemical composition. To fill in, generate, or determine the initial design space, the preprocessing instructions 108 may first determine a minimum value, maximum value, and increment value for each one of ora portion of the chemical operations parameters or settings and / or each one of or a portion of the amount of each component in the chemical composition. Such values may be utilized to initialize the design space. As noted, the design space may be updated or refined over time. As new chemical operations are conducted and / or as chemical compositions or formula are adjusted, new data associated with each may be preprocessed as well.

[0047] In an embodiment, the design space system 102 may receive data in real-time and / or continuously, as chemical operations (such as polymerization operations) are being performed, are underway, or are being executed. In another embodiment, the data may be transmitted to or stored in the database 132. In such an embodiment, the design space system 102 may periodically obtain or receive data from the database 132. The preprocessing instructions 108 may also, when executed, determine whether the data received includes known chemical operations and / or chemical compositions. The preprocessing instructions 108 may, when executed, compare the results corresponding to the received data to a current design space. In another embodiment, each data point may include a tag. The tag may indicate a particular chemical process or operation. Based on the chemical operation to be analyzed, the data points with a selected or associated tag (indicating the particular polymerization operation) may be separated and stored as a data subset. In such examples, the preprocessing instructions 108 may determine a type or types of machine learning models 112 to utilize for varying portions of the data (for example, utilizing a particular model, in addition to, in some embodiments, a physics based model, to fill in temperature data for a selected chemical operation and / or chemical product). In another example, a plurality of chemical operations may be executed over a selected time frame. During such a time frame, data received at a selected time may be related to a polymerization operation experiment, for example to determine (i) how a particular composition performs, (ii) how the composition is made, (iii) the properties exhibited by the composition, and / or (iv) how various factors affect the end product (factors such as temperature, pressure, flow, among other factors). The data indicating such an experiment, as noted, may be separated and stored as a data subset. The remaining portions of data may be stored (for example in the memory 106 or in the database 132) for later use or, in another embodiment, deleted or removed from the design space system 102.

[0048] As noted, a physics based model or a first-principle model may be utilized to predict or determine values for a particular input that has not been measured. Such an input may be utilized as an input to the machine learning models 112. For example, a measured temperature and volume maybe utilized by a physics based model to determine pressure, which may be utilized as an input to the machine learning models 112.

[0049] Once a portion of the data or a data subset is selected, the preprocessing instructions 108 may smooth the data subset. In other words, upon execution of the preprocessing instructions 108, any outlying data points or errors in the data subset may be removed to form a smoothed data subset. Such a smoothing process may include various smoothing algorithms, such as moving average smoothing, exponential smoothing, double exponential smoothing, triple exponential smoothing, among other techniques, as will be understood by one skilled in the art. In an embodiment, the smoothing algorithm may be selected by a user.

[0050] Once a dataset is available, design space instructions 110 may be executed. The design space instructions 110 may determine which portion of the data set to apply to the machine learning models 112 (and in some embodiments, to a physics based model 114). The design space instructions 110 may, when executed, apply the data set to the machine learning model 112. The design space instructions 110 may subsequently, when executed, increment a particular variable or variables for the next application applied to the machine learning model 112.

[0051] In an embodiment, the one or more machine learning models 112 may be trained based on historical data. The historical data may include prior polymerization operations, the resulting output or produced polymeric composition, any adjustments made based on data collected from such operations, and / or the performance of the resulting output chemical product. In an embodiment, one or more different machine learning models may be trained and each such machine learning model may be a different type, based on performance of such a model for such a collection or type of data. For example, a selected model may be utilized when considering temperature over time, while another type of model may be considered for pressure. In such examples, the relevant portions of a data set for a particular chemical operation or chemical composition may be applied to the respective one or more machine learning models 112. Based on the application of the data (such as the chemical process or operation parameters and / or a chemical composition) to the one or more machine learning models 112, the one or more machine learning models 112 may produce one or more charts, performance predictions, output predictions (for example a “synthetic output” or an output that is predicted rather than an output that has been physically tested), and / or other operations or process data. Such an output (in addition to, in some embodiments, measured data and / or synthetic data) may be added to the design space. After a plurality of iterative applications, the design space may include a plurality ofdifferent chemical processes or operations, chemical compositions, performances, and / or other relevant data. A computing device, controller, and / or user may utilize such a design space to quickly, and without experimentation, optimize a chemical process and / or adjust a chemical composition to produce a chemical product that meets or exceeds specified performance.

[0052] In an embodiment, the design space system 102 may include visualization instructions 120. The visualization instructions 120, when executed, may generate a UI 122A, 122B, and up to 122N or a portion of the UI 122 A, 122B, and up to 122N for one or more computing devices UI 124 A, 124B, and up to 124N. The UI 122 A, 122B, and up to 122N may include a search bar or other search function (such as a file or data upload function) to allow or enable a user to request or search for variable chemical compositions, chemical operations, and / or for a specified chemical product performance for a particular chemical product. Upon such a search or request, the visualization instructions 120 may generate results including a selected number of chemical operation parameters and / or chemical compositions meeting or exceeding specified properties. The visualization instructions 120 may utilize various search algorithms to generate such results, including, but not limited to, a nearest neighbor search algorithm or another algorithm configured to select a specified number of results matching or close to the requested chemical composition, chemical operations, and / or chemical product performance.

[0053] In an embodiment, the design space system 102 may connect to a plurality of controllers, plants (for example, connected to computing devices and / or other devices located at the plant), and / or other locations. In such embodiments, the design space system 102 may provide the optimization functionality described above and herein to each of these locations. Further, the design space system 102 may utilize data collected at each location to further refine machine learning models and / or to add data relating to new chemical processes and / or chemical compositions. In an embodiment, a polymerization operation or setting may include converting monomers / oligomers, or other chemicals, to polymers, blending chemicals or polymers, and / or extruding chemicals or polymers. In an embodiment, the polymerization operation settings may include one or more polymerization reactor settings, process settings, catalyst type, catalyst amount, co-catalyst type and amount, temperature, pressure settings, flow rates, residence times, blending time, concentrations of reactants (for example, alpha-olefins, ethylene, propylene, 1 -butene, 1 -hexene, hydrogen), or solvent type and amount. In another embodiment, polymerization operations and / or process data may include data from the full life cycle chain of a chemical or polymer, including, but not limited to, polymerization settings,process parameters, polymer material properties, formulations of compounds, compounding process settings to produce compounds, settings to produce test specimen for properties measurements, compounded product properties, data of the conversion process to produce applications (for example, such as films, pipe, and / or injection molded products), blending data, extrusion data, and / or application performance measurements.

[0054] In some examples, the design space system 102 may be a computing device. The term “computing device” is used herein to refer to any one or all of programmable logic controllers (PLCs), controller, programmable automation controllers (PACs), industrial computers, servers, virtual computing devices or environments, desktop computers, personal data assistants (PDAs), laptop computers, tablet computers, smart books, palm-top computers, personal computers, smartphones, virtual computing devices, cloud based computing devices, and similar electronic devices equipped with at least a processor and any other physical components necessarily to perform the various operations described herein.

[0055] The term “server” or “server device” is used to refer to any computing device capable of functioning as a server, such as a master exchange server, web server, mail server, document server, or any other type of server. A server may be a dedicated computing device or a server module (e.g., an application) hosted by a computing device that causes the computing device to operate as a server. A server module (e.g., server application) may be a full function server module, or a light or secondary server module (e.g., light or secondary server application) that is configured to provide synchronization services among the dynamic databases on computing devices. A light server or secondary server may be a slimmed-down version of server type functionality that can be implemented on a computing device, such as a smart phone, thereby enabling it to function as an Internet server (e.g., an enterprise e-mail server) only to the extent necessary to provide the functionality described herein.

[0056] As used herein, a “non-transitory machine-readable storage medium” or “memory” may be any electronic, magnetic, optical, or other physical storage apparatus to contain or store information such as executable instructions, data, and the like. For example, any machine-readable storage medium described herein may be any of random access memory (RAM), volatile memory, nonvolatile memory, flash memory, a storage drive (e.g., hard drive), a solid state drive, any type of storage disc, and the like, or a combination thereof. The memory may store or include instructions executable by the processor.

[0057] As used herein, a “processor” or “processing circuitry” may include, for example one processor or multiple processors included in a single device or distributed across multiple computing devices. The processor (such as, processor 104 shown in FIG. 1) may be at least one of a central processing unit (CPU), a semiconductor-based microprocessor, a graphics processing unit (GPU), a field-programmable gate array (FPGA) to retrieve and execute instructions, a real time processor (RTP), other electronic circuitry suitable for the retrieval and execution instructions stored on a machine-readable storage medium, or a combination thereof.

[0058] In an embodiment, the one or more machine learning models 112 may be a supervised or unsupervised learning model. In an embodiment, the one or more machine learning models 112 may be based on one or more of decision trees, random forest models, random forests utilizing bagging or boosting (as in, gradient boosting), K-nearest neighbors, neural network methods, support vector machines (SVM), lasso based models, other supervised learning models, other semi-supervised learning models, other unsupervised learning models, or some combination thereof, as will be readily understood by one having ordinary skill in the art.

[0059] In another embodiment, one of the one or more machine learning models may be an image and / or recognition machine learning model. Such a model may be trained using a number of images of a chemical product exhibiting a particular pattern. The design space, in an embodiment, may include image or pattern data, as well as tags or other indicators to indicate such an image or pattern, for particular chemical products. In such embodiments, a user may search for a specific chemical product exhibiting such pattern. Such a search may be text and / or image based.

[0060] FIG. 2 is another schematic diagram of a system for generating a data set associated with a polymerization operation, in accordance with certain embodiments of the present disclosure. Such a system may be comprised of a processing circuitry 202, a memory 204, a communications circuitry 206, a preprocessing circuitry 208, a design space circuitry 210, a modeling circuitry 212, a visualization circuitry 214, and a chemical operation controller circuitry 216, each of which will be described in greater detail below. While the various components are only illustrated in FIG. 2 as being connected with processing circuitry 202, it will be understood that the apparatus 200 may further comprise a bus (not expressly shown in FIG. 2) for passing information amongst any combination of the various components of the apparatus 200. The apparatus 200 may be configured to execute various operations described herein, such as those described above in connection with FIG. 1 and below in connection with FIGS. 3-6B.

[0061] The processing circuitry 202 (and / or co-processor or any other processor assisting or otherwise associated with the processor) may be in communication with the memory 204 via a bus for passing information amongst components of the apparatus. The processing circuitry 202 may be embodied in a number of different ways and may, for example, include one or more processing devices configured to perform independently. Furthermore, the processor may include one or more processors configured in tandem via a bus to enable independent execution of software instructions, pipelining, and / or multithreading.

[0062] The processing circuitry 202 may be configured to execute software instructions stored in the memory 204 or otherwise accessible to the processing circuitry 202 (e.g., software instructions stored on a separate storage device). In some cases, the processing circuitry 202 may be configured to execute hard-coded functionality. As such, whether configured by hardware or software methods, or by a combination of hardware with software, the processing circuitry 202 represents an entity (e.g., physically embodied in circuitry) capable of performing operations according to various embodiments of the present disclosure while configured accordingly. Alternatively, as another example, when the processing circuitry 202 is embodied as an executor of software instructions, the software instructions may specifically configure the processing circuitry 202 to perform the algorithms and / or operations described herein when the software instructions are executed.

[0063] Memory 204 is non-transitory and may include, for example, one or more volatile and / or non-volatile memories. In other words, for example, the memory 204 may be an electronic storage device (e.g., a computer readable storage medium). The memory 204 may be configured to store information, data, content, applications, software instructions, or the like, for enabling the apparatus 200 to carry out various functions in accordance with example embodiments contemplated herein.

[0064] The communications circuitry 206 may be any means such as a device or circuitry embodied in either hardware or a combination of hardware and software that is configured to receive and / or transmit data from / to a network and / or any other device, circuitry, or module in communication with the apparatus 200. In this regard, the communications circuitry 206 may include, for example, a network interface for enabling communications with a wired or wireless communication network. For example, the communications circuitry 206 may include one or more network interface cards, antennas, buses, switches, routers, modems, and supporting hardware and / or software, or any other device suitable for enabling communications via a network. Furthermore, the communications circuitry 206 may include the processing circuitry 202 for causing transmission of such signals to anetwork or for handling receipt of signals received from a network. The communications circuitry 206, in an embodiment, may enable reception of polymerization operation data (including, in an example, polymeric compositions or other data related to the polymerization operation) and transmission of polymerization operation settings to associated equipment and / or devices.

[0065] The apparatus 200 may include preprocessing circuitry 208 configured to preprocess received data. Preprocessing received data may include determining input parameters for received data and / or fitting the data to one or more machine learning models. The preprocessing circuitry 208 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, as described in connection with FIGS. 5-6B below. The preprocessing circuitry 208 may further utilize communications circuitry 206, as noted above, to gather data (such as real-time chemical operations data and / or historical chemical operations data) from a variety of sources (for example, one or more different components or devices at a plant or plants, such as a polymerization reactor and associated sensors or analyzers; a database; and / or other data sources). The output of the preprocessing circuitry 208 may be transmitted to other circuitry of the apparatus 200 (such as the design space circuitry 210).

[0066] In addition, the apparatus 200 further comprises the design space circuitry 210 that may apply or transmit the data, based on the input parameters, to the modeling circuitry 212. For example, the design space circuitry 210 may generate sets or subsets of data based on the input parameters and then send each set separately to the modeling circuitry 212 for application to models. The design space circuitry 210 may further receive the output from the modeling circuitry 212 and, upon reception of the output, apply or add the output to a design space database. Other information may be added the outputs added to the design space database, such as tags and / or the original corresponding data. In another embodiment, the design space circuitry 210 may apply the data to the modeling circuitry 212 iteratively. For example, the design space circuitry 210 may apply one portion of the data corresponding to a chemical operation and chemical composition to the modeling circuitry 212. The design space circuitry 210 may then increment one or more variables, according to the input parameters, in the data corresponding to the chemical operation and chemical composition, and apply the adjusted and / or incremented data to the modeling circuitry 212. The design space circuitry 210 may continue to increment variables and apply the adjusted and / or incremented data to the modeling circuitry 212 until a maximum has been reached for one of the variables, according to the input parameters. The design space circuitry 210 may then increment other variables and continue to applysuch data to the modeling circuitry 212. Thus, the design space circuitry 210 may form, generate, or produce a design space including a plurality of chemical operations, chemical compositions, and corresponding chemical product performance. The design space circuitry 210 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, as described in connection with FIGS. 5-6B below. The design space circuitry 210 may further utilize communications circuitry 206 to gather data (for example, preprocessed data and / or input parameters) from a variety of sources (such as the preprocessing circuitry 208, a controller 126, or a polymerization reactor 116 and / or equipment 118 at a plant); receive an output from the modeling circuitry 212; and / or add, postprocess and / or combine the output from the modeling circuitry 212 to a design space. The output of the design space circuitry 210 may be transmitted to other circuitry of the apparatus 200, such as the modeling circuitry 212.

[0067] The apparatus 200 further comprises the modeling circuitry 212 that may apply received data to one or more trained machine learning models based on the type of data received, an indicator transmitted along with the data, and / or a determination of which of the one or more trained machine learning models fits the data. Such an application of the data to the one or more trained machine learning models may produce an output including one or more predictions, probabilities, simulations (such as a simulated performance of a potential chemical product), and / or charts or graphs associated with a performance of a resulting chemical product (for example, the chemical product being a result of the chemical operations and the chemical composition). The modeling circuitry 212 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, as described in connection with FIGS. 5-6B below. The machine learning model circuitry 212 may further utilize communications circuitry 206 to receive data from the design space circuitry 210. The output of the modeling circuitry 212 may be transmitted to other circuitry of the apparatus 200, such as the visualization circuitry 214 and / or chemical operation controller circuitry 214.

[0068] The apparatus 200 further comprises the visualization circuitry 214 that may generate a user interface, allow a user to search for or determine a number of variable chemical operations and / or chemical compositions that meet or exceed specified performances based on data in the design space, and / or allow or enable a user to initiate the chemical operation with the chemical composition to produce a chemical product. In another embodiment, the visualization circuitry 214 may be configured to automatically initiate a polymerization operation or other chemical operation based ona search for a specified performance of particular chemical operation and / or chemical composition. In such embodiments, a user may initiate such a search and then the visualization circuitry 214 may initiate the chemical operation based on a performance match from the design space, available chemical compositions, and / or current equipment availability and / or downtime. The visualization circuitry 214 may generate a user interface that includes interactive fields to enable a user to further adjust a chemical operation and / or chemical composition and view corresponding chemical performances. Further, the visualization circuitry 214 may enable the user to initiate or begin the chemical operation to produce the chemical product. The visualization circuitry 214 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, as described in connection with FIGS. 5-6B below. The visualization circuitry 214 may further utilize communications circuitry 206 to receive data from the design space circuitry 210. In an embodiment, the visualization circuitry 214, in response to a signal received to initiate a chemical process, may transmit such an initiation signal to the chemical operation controller circuitry 216.

[0069] The apparatus 200 further comprises the chemical operation controller circuitry 216 that may control or initiate a polymerization operation based on a signal from the visualization circuitry 214. The chemical operation controller circuitry 214 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, as described in connection with FIGS. 5-6B below. The chemical operation controller circuitry 216 may further utilize communications circuitry 206 to receive an initiation signal and / or communicate with a selected plant or facility controller, such that the chemical operation controller circuitry 216 may cause the selected plant or facility to initiate a chemical operation.

[0070] Although components 202-216 are described in part using functional language, it will be understood that the particular implementations necessarily include the use of particular hardware. It should also be understood that certain of these components 202-216 may include similar or common hardware. For example, the preprocessing circuitry 208, the design space circuitry 210, the modeling circuitry 212, the visualization circuitry 214, and the chemical operation controller circuitry 216 may, in some embodiments, each at times utilize use of the processing circuitry 202, memory 204, or communications circuitry 206, such that duplicate hardware is not required to facilitate operation of these physical elements of the apparatus 200 (although dedicated hardware elements may be used for any of these components in some embodiments, such as those in which enhanced parallelism may bedesired). Use of the terms “circuitry” with respect to elements of the apparatus therefore shall be interpreted as necessarily including the particular hardware configured to perform the functions associated with the particular element being described. Of course, while the term “circuitry” should be understood broadly to include hardware, in some embodiments, the term “circuitry” may in addition refer to software instructions that configure the hardware components of the apparatus 200 to perform the various functions described herein.

[0071] Although the preprocessing circuitry 208, the design space circuitry 210, the modeling circuitry 212, the visualization circuitry 214, and the chemical operation controller circuitry 216 may utilize processing circuitry 202, memory 204, or communications circuitry 206 as described above, it will be understood that any of these elements of apparatus 200 may include one or more dedicated processors, specially configured field programmable gate arrays (FPGA), or application specific interface circuits (ASIC) to perform its corresponding functions, and may accordingly leverage processing circuitry 202 executing software stored in a memory or memory 204, communications circuitry 206 for enabling any functions not performed by special-purpose hardware elements. In all embodiments, however, it will be understood that the preprocessing circuitry 208, the design space circuitry 210, the modeling circuitry 212, the visualization circuitry 214, and the chemical operation controller circuitry 216 are implemented via particular machinery designed for performing the functions described herein in connection with such elements of apparatus 200.

[0072] In some embodiments, various components of the apparatus 200 may be hosted remotely (e.g., by one or more cloud servers) and thus need not physically reside on the corresponding apparatus 200. Thus, some or all of the functionality described herein may be provided by third party circuitry. For example, a given apparatus 200 may access one or more third party circuitries via any sort of networked connection that facilitates transmission of data and electronic information between the apparatus 200 and the third party circuitries. In turn, that apparatus 200 may be in remote communication with one or more of the other components describe above as comprising the apparatus 200.

[0073] As will be appreciated based on this disclosure, example embodiments contemplated herein may be implemented by an apparatus 200 (or by a controller 302). Furthermore, some example embodiments (such as the embodiments described for FIGS. 1 and 3) may take the form of a computer program product comprising software instructions stored on at least one non-transitory computer- readable storage medium (such as memory 204). Any suitable non-transitory computer-readablestorage medium may be utilized in such embodiments, some examples of which are non-transitory hard disks, CD-ROMs, flash memory, optical storage devices, and magnetic storage devices. It should be appreciated, with respect to certain devices embodied by apparatus 200 as described in FIG. 2, that loading the software instructions onto a computing device or apparatus produces a special-purpose machine comprising the means for implementing various functions described herein.

[0074] FIG. 3 is a schematic diagram of a controller to generate a complete or substantially complete data set associated with a polymerization operation for use in adjusting polymerization operation settings, in accordance with certain embodiments of the present disclosure. The control system, as described herein, may be a controller 302, one or more controllers, a PLC, a SC AD A system, a computing device, and / or other components to generate a design space for use in adjusting and / or initiating a chemical operation (for example, the components, devices, or apparatus described in FIGS. 1 and 2). The controller 302 may include one or more processors (e.g., processor 304) to execute instructions stored in memory 306. In an example, the memory 306 may be a machine- readable storage medium.

[0075] As used herein, “signal communication” refers to electric communication such as hard wiring two components together or wireless communication, as understood by those skilled in the art. For example, wireless communication may be or include Wi-Fi®, Bluetooth®, ZigBee, forms of near field communications, or other wireless communication methods as will be understood by those skilled in the art. In addition, signal communication may include one or more intermediate controllers, relays, or switches disposed between elements that are in signal communication with one another.

[0076] As noted, the memory 306 may store instructions executable by the processor 304, to preprocess data, such as preprocessing instructions 308. The controller 302 may connect to and / or receive data from one or more of a polymerization reactor 322, sensors or other devices to determine some characteristic of the polymerization reactor 322, equipment 324 at one or more locations (such as a plant, facility, or other location), sensors or other devices to determine some characteristic of the equipment 324, a database 326, and / or from a user interface 328. The controller 302 may obtain data from each of the one or more data sources. The preprocessing instructions 308 may define input parameters and / or preprocess data. The preprocessing instructions 308 may, in response to reception of such data, be executed to preprocess such data. The preprocessing instructions 308 may, upon execution, determine input parameters for the data. The determination of the input parameters may be based on the type of data received, the type of machine learning models available, and / or on anindication of which variables in the data that may utilize input parameters. The input parameters may include a minimum value for a corresponding one or more variable, a maximum value for the one or more corresponding variable, and / or a increment value that indicates the next value that a variable should be adjusted to when the variable and other corresponding data is applied to one or more machine learning models.

[0077] The controller 302 may include design space generation instructions 312. The design space generation instructions 312 may, upon execution, apply and then increment data, based on the input parameters to the trained machine learning models 314. The design space generation instructions 312 may, upon execution, increment variables and then re-apply the data to the machine learning models. Further, the design space generation instructions 312, upon execution, may determine which of the trained machine learning models 314 to utilize based on parts and / or types of data, available trained machine learning models 314, whether a chemical operation and / or chemical composition are unknown, and / or an input received via the user interface 328, among other factors. The design space generation instructions 312 may also receive outputs from the trained machine learning models 314 and, after reception of the output, update a design space with the output and corresponding data used to generate the output. The output may include a simulated performance of the resulting chemical composition.

[0078] The controller 302 may include data visualization instruction 316. The data visualization instructions 316 may, upon execution, generate a graphical user interface (GUI) or a portion of a GUI to be displayed via the user interface. The data visualization instructions 316 may enable the user interface to search for various chemical operations and / or chemical compositions that meet a specified performance. The data visualization instructions 316 may further generate and display one or more interactive charts, graphs, and / or tables that enable further refinement of a chemical composition and / or chemical operations.

[0079] The controller 302 may include process and / or composition adjustment and / or initialization instructions 318. The process and / or composition adjustment and / or initialization instructions 318 may, when executed, cause generation of adjustments or updates to chemical operation parameters and / or chemical compositions. In an embodiment, the adjustments or updates to chemical operation parameters and / or chemical compositions may be transmitted to the polymerization reactor 322 and / or other equipment 324, thus causing the polymerization reactor 322 and / or other equipment 324 to operate at those settings for subsequent operations. In another embodiment, a user may select achemical operation and / or chemical composition, via the GUI generated by the data visualization instructions 318, and, based on such a selection, the process and / or composition adjustment and / or initialization instructions 318 may cause the polymerization reactor 322 and / or other equipment 324 to operate at those selected settings.

[0080] Turning to FIGS. 4A and 4B, a graphical user interface (GUI) 402 is provided that illustrates the user interface corresponding to the design space or a search of the design space. As noted, a user may search for particular chemical operation and / chemical product performance. Such a GUI 402 may include a search box 404 for a user to enter the search. The GUI 402 may further include a button 406 for selecting a type of view of the data (for example, as a graph, chart, an interactive plot, and / or a tabular list or tabular visualization) and whether to download the data, a portion of the data, and / or the currently displayed chart, graph, and / or table. Such a search may generate, for example, one or more interactive graphical displays. For example, a spider graph 410 with multiple formulation or composition suggestions 412 may be generated. A user may move any point on the spider graph to adjust a desired chemical process and / or chemical composition performance. Based on the adjustment, an adjusted chemical formula and / or chemical composition may be generated. As noted, the GUI 402 may generate additional views, for example, the GUI 402 may generate a table including performance of one or more chemical compositions.

[0081] In an example, to utilize the design space, as described above, the system 100, apparatus 200, or controller 302 may receive a data set, as show below in Table 1. In further embodiments, substantially more data may be received. Such data may include various polymers in a formula or composition, modifiers added to the compound, formula, or composition, and / or measured outputs (such as ash content (measured as weight percentage), melt flow rate (measured as decigram per minute), tensile modulus (measured as megapascal), charpy impact at various temperatures (measured as kilojoule per meters squared), break type, and / or visual esthetics, among other measurable outputs). For example, measured data or sample data may include the weight percentage of a polymer (for example, polymer 1, polymer 2, and / or polymer 3) in a compound, the weight percentage of external modifiers (for example, external modifier 1 and / or external modifier 2) in the compound, the weight percentage of talc in the compound, the weight percentage of short glass in the compound, and / or the weight percentage of additives in the compound.Table 1

[0082] Once the system 100, apparatus 200, or controller 302 receives the data, the system 100, apparatus 200, or controller 302 may pre-process the data. For example, as shown in Table 2, the system 100, apparatus 200, or controller 302 may, for example, filter the data. Other pre-processing steps may be performed, as described above.Table 2

[0083] Based on the filtered data, the system 100, apparatus 200, or controller 302 may apply such data to one or more machine learning models to generate synthetic data and / or predicted data and / or performance (or, in other words, data points not measured) for various other formulations or compositions, as shown in Table 3 below.Table 3

[0084] Based on the synthetic data or predicted data and / or performance, as well as other data in the design space, the system 100, apparatus 200, or controller 302 may provide suggested formulation or compositions and / or operation or process settings and / or parameters, such as the chart illustrated in FIG. 4A and / or the table illustrated in FIG. 4B.

[0085] In another example, data for a design space may be generated for an extrusion operation. In such embodiments, the polymerization operations data may include various extrusion process parameters and / or components, including, but not limited to, feed components, recycle ingredients, the final output or product, fillers, temperature in one or more zones of an extruder, pressure in one or more zones of an extruder, barrel torque, feed quantities, ash content, product modulus or stiffness, and / or product impact. All of these data points may be obtained for a number of historical extrusion operations. As noted, the data may not be complete. In other words, many variables may not be represented by existing, actual data, including the resulting outputs for those different variables. Assuch, the systems and methods described herein may be utilized to produce a design space for extrusion operations. For example, such systems and methods may utilize the historical data to produce a number of inputs. Those inputs may include variations of the extrusion process parameters and / or components. The inputs may then be applied to the trained machine learning model to produce a synthetic output with predicted parameters or features. Each output may be added and / or stored in the design space.

[0086] In an embodiment, a portion of the data may be gathered in real-time. The systems and methods described herein may determine alternate process data based on the real-time data and apply the alternate process data to the trained machine learning model and continue to fill in the design space. Further, the systems and methods may utilize the new entries into the design space to actively adjust the extrusion process, for example, by adjusting fillers, additives, temperatures, pressures, and / or other parameters.

[0087] Prophetic Example: Virtual Experimentation - Extruder

[0088] FIG. 5 A illustrates an exemplary information flow 501 for virtual experimentation. An exemplary system may include one or more extruders 503. A “Feed 1” 505 may enter the one or more extruders 503 at a first inlet, and may be, for example, a polypropylene (“PP”) recycle feed stream. The “Feed 1” 505 may be provided by one or more suppliers 507 and may be associated with various feed data 509 including, but not limited to, post-consumer recycled (“PCR”) materials’ quality data. The PCR quality data may include, but is not limited to, a certificate of analysis, containing information like (but not limited to) melt flow index, ash content, modulus (stiffness), and / or variability characteristics. In certain embodiments, the PCR materials may have high variability.

[0089] A “Feed 2” composition 511 may be fed inline into the extruder 503. “Feed 2” 511 may include, for example, virgin polypropylene, additives, fillers, and / or elastomers. Inline processes 513 may provide various inline PCR quality data 515 including, but not limited to, viscosity (such as for the purposes of measuring for manufacturers), ethylene content, and / or ash content. Other inline processes 517 may provide various processing data 519 including, but not limited to, barrel torque, melt pressure, temperature (in some embodiments based on zones within the extruder 503), “Feed 1”quantities, and / or “Feed 2” quantities. In various embodiments, additional feeds may be added as needed to the extruder 503.

[0090] A “Final Product” 521 may exit the extruder 503. Offline 523 processes may provide offline final product quality data 525 including, but not limited to, ash content, modulus (stiffness), impact strength, melt flow rate, and / or quality / assurance testing.

[0091] A pre-trained machine-learning model 527 may be used. The model 527 may be used to determine what to add to the extruder 503 as a “Feed 2” composition 511 based on the unknown and / or variable qualities of “Feed 1” compositions 505. In certain embodiments, the model 527 may provide input on the amounts and / or types of ingredients to add as a “Feed 2” composition 511 to produce a “Final Product” 521 with desired and / or known specifications from the extruder 503. The model may be obtained using known material informatics tools. To be able to anticipate batch-to- batch variations in PCR materials, however, an in-line measurement of PCR quality may be applied. This may allow for real time adjustments of “Feed 2” compositions to ensure meeting desired performance of a “Final Product”.

[0092] Model Training

[0093] To train the model 527, training data / input 529 may be generated based on key identifiers of the PCR in “Feed 1” 505.

[0094] This training data may be split into a set of training data and validation data. The ratio of training data may vary from approximately 99: 1 to approximately 1 :99, but preferably is in a ratio of approximately 75:25. The training data may be used to train the machine learning model using a folding scheme (k-fold cross validation) or similar process. The validation data (or unseen data) may not be used for training and will may to validate the model.

[0095] To train the model 527 it may be necessary to have off-line measured performance data 525 from the “Final Product” available (e.g., via design of experiments). Utilizing PCR supplier certificates of analysis may improve model performance. The exact features (i.e., measured quantities) that may be used for obtaining optimal model performance may be obtained byinvestigating feature importance and may depend on the measurement capabilities of the exact equipment used.

[0096] Model utilization

[0097] Once the model is trained, the in-line measured properties including, but not limited to inline PCR quality 515 and inline processing data 519, together with desired Final Product performance may be used to tune the composition of “Feed 2” 511.

[0098] FIG. 5B is a flow diagram for generating a design space including a plurality of chemical operations (such as polymerization operations), chemical compositions (such as polymeric compositions), and corresponding performance of the resulting chemical product, in accordance with certain embodiments of the present disclosure. Unless otherwise specified, the actions of method 500 may be completed within system 100, apparatus 200, and / or controller 302. Specifically, method 500 may be included in one or more programs, protocols, or instructions loaded into the memory 106 of the design space system 102 and executed on the processor 104 or one or more processors. The order in which the operations are described is not intended to be construed as a limitation, and any number of the described blocks may be combined in any order and / or in parallel to implement the methods.

[0099] At block 502, the design space system 102 may receive a data set. At block 506, the design space system 102 may define input parameters for each or a portion of each variable of each chemical operation and / or chemical composition. In an embodiment, the design space system 102 may determine which variables to define input parameters based on a number of factors, such as the type of data received, the number of data points for a particular variable, and / or an indication or indicator included with the data. The input parameters may include, for example, a minimum value for a variable, a maximum value for the variable, and / or an incremental value for the variable. At block 508, the design space system 102 may iteratively apply the data to one or more machine learning models according to the input parameters. Upon application of data to the one or more machine learning models, the machine learning model may produce an output, such as a prediction, probability, and / or simulation illustrating or representing the performance or characteristics of a chemical product produced by the chemical operations and / or chemical composition. Such an iterative application of data to the machine learning models and the addition of the resulting output and other data to a design space may form the design space (in addition to, in some embodiments, measured data and / or othersynthetic data produced by one or more physics based models). Blocks 506 and 508, encompassed by block 504, may comprise a process or subprocess to generate the design space. Block 504 may include additional steps or blocks, in some embodiments, for example, the design space system 102 may preprocess the data set further, prior to iterative application to the one or more machine learning models and / or the design space system 102 may perform postprocessing on the output generated from each application of data to a machine learning model, prior to addition of the output to the design space.

[0100] Once the design space is generated, then at block 516 and / or block 510, the design space system 102 may receive an input. In particular, at block 516, the design space system 102 may receive a user constraints input. Such an input may include a type of chemical product sought, the performance of such a product, and / or the chemical operation and / or the chemical composition used to produce such a chemical product. In addition to or rather than the user constraints input, at block 510, the design space system 102 may receive an application critical to quality (CTQ) (in other words, a desired performance or other aspect of a chemical product and / or chemical process). At block 512, the design space system 102 and / or another user may convert the application CTQ to a material CTQ and / or a process CTQ. Such a conversion, if performed by the design space system 102, may include applying the application CTQ to a model or other algorithm configured to, based on the desired performance specified in the input application CTQ, produce one or more of a material CTQ (in other words, a material that may meet the desired performance) and / or process CTQ (in other words, a process that may meet the desired performance). Based on reception of either the application CTQ and / or the user input constraints, at block 518, the design space system 102 may generate multiple chemical compositions or formula that meet or exceed the user input (such as the application CTQ and / or the user constraints input). Further, the design space system 102 may generate one or more visualizations, such as, for example, the performance of the resulting chemical product and / or the chemical composition and / or the chemical operation used to produce the chemical product.

[0101] FIG. 6A and 6B are additional flow diagrams for generating a design space including a plurality of chemical operations (such as polymerization operations), chemical compositions (such as polymeric compositions), and corresponding performance of the resulting chemical product, in accordance with certain embodiments of the present disclosure. Unless otherwise specified, the actions of method 600 and method 601 may be completed within system 100, apparatus 200, and / or controller 302. Specifically, method 600 and method 601 may be included in one or more programs,protocols, or instructions loaded into the memory 106 of the design space system 102 and executed on the processor 104 or one or more processors. The order in which the operations are described is not intended to be construed as a limitation, and any number of the described blocks may be combined in any order and / or in parallel to implement the methods.

[0102] Turning first to method 600, at block 602, a design space system 102 may receive data. The data may include real-time and / or historical chemical operations data. In a further embodiment, the data may include experimental data. At block 604 the design space system 102 may define input parameters for the received data, such as a minimum value, a maximum value, and / or an incremental value for one or more variables included in the received data set. In an embodiment, design space creation may be an on-going or continuous process. For example, the design space system 102 may receive data at various times and update the design space, based on the additional data received. Such received data may include chemical operations, chemical compositions, and / or the resulting chemical products produced, along with corresponding performances of the chemical product.

[0103] At block 606, the design space system 102 may iteratively apply the data to one or more machine learning models according to the input parameters. Such an application to the one or more machine learning models may produce, as noted, a chemical operation and / or chemical composition, in addition to the performance of a resulting chemical product. The output of such machine learning models may be a set of values, a probabilities, predictions, and / or simulation results. The design space system 102 may utilize the output from the machine learning models, in addition to the received data to produce the design space.

[0104] At block 608, the design space system may determine whether input constraints and / or a CTQ has been received. If no input constraints and / or CTQ are received, the design space system 102 may wait until such an input is received and / or until new data is received to fill in the design space.

[0105] If an input is received, at block 610, the design space system 102 may determine whether the input is unknown or if relevant results are contained with the design space. If the input is unknown, the design space system 102 may determine whether the input includes additional data sufficient for addition to the design space (via application to the one or more machine learning models). For example, if an input is received that includes a chemical composition and / or chemical parameters and a resulting performance or potential performance data, then the design space system 102 may determine that such data is sufficient (further depending on whether any data points are missing).

[0106] If the data is not sufficient for addition to the design space, at block 614, the design space system 102 may prompt another computing device and / or a user to submit additional data. Once the data is received and or deemed sufficient, then the design space system 102 may define additional input parameters and iteratively apply the data to one or more of the machine learning models.

[0107] If the input was known or included relevant entries in the design space, then the design space system 102 may determine one or more formulations or compositions and / or operation settings that meet or exceed a specified performance of the resulting chemical product. At block 618, the design space system 102 may generate one or more visualizations of each of the one or more formulations or compositions and / or the operation settings. In another embodiment, at block 620, if a user selects a specific chemical composition and / or operation settings and then, after implementation of such settings, transmits the results of those operations with the selected chemical composition (for example, the performance of the resulting chemical compound), then the design space system 102 may use such data to further refine one or more of the one or more machine learning models.

[0108] Turning to FIG. 6B, method 601 illustrates another process for populating a design space. In such embodiments, the design space system may receive data at block 602 and define the input parameters for the data at block 604, as described above and herein.

[0109] At bock 622, the design space system 102 may then determine if the maximum for a variable of the operation settings has been reached. In an embodiment, such a determination may occur for a plurality of and varying combination of variables of the operation settings. If the maximum has been reached, at block 632 the design space system 102 may determine whether additional or new data has been received and if so, the process described for method 601 may be repeated, otherwise the design space system 102 may wait for such data to further populate the design space.

[0110] If the maximum has not been reached for the variables of the operations settings, the design space system 102 may determine whether a maximum for an amount of a component in a chemical composition, such as a polymeric composition, has been reached. If a maximum has been reached, at block 630 the design space system may increment one or more variables of the operation settings and back to block 622. If a maximum has not been reached, at block 626, the design space system 102 may generate a simulation of the chemical operation with the chemical composition and utilize the resulting performance data to fill the design space, then at block 628, the design space system 102may generate a new chemical composition based on the input parameters (for example, incrementing an amount of a component of the preceding chemical composition).

[0111] While particular terms and concepts are incorporated in the present disclosure, Applicant notes that the disclosed terms and concepts are exclusively utilized in a descriptive capacity and should not therefore be construed or interpreted as limiting in any way. Certain embodiments and aspects of the disclosed systems, processes and methods have been described in detail with particular reference to the illustrated embodiments. However, it will be apparent that numerous and various modifications and alterations may be made within the spirit and scope of the embodiments of systems, processes and methods described herein, and such modifications and changes are to be considered equivalents and within the breadth and scope of the disclosure.

Claims

CLAIMSWhat is claimed is:

1. A method to generate a design space for determining one or more polymeric compositions or polymerization operation settings, the method comprising: in response to receipt of a plurality of historical polymerization operations data, determining input parameters, the input parameters to include at least a minimum, a maximum, and an increment for each of a plurality of polymerization operation settings and each of a plurality of polymeric composition components, the plurality of historical polymerization operations data including data indicative of a portion of the plurality of polymerization settings, a portion of the plurality of polymeric composition components, an output of one or more of the plurality of historical polymerization operations, and associated performances of the output of one or more of the plurality of historical polymerization operations; generating the design space via iterative and incremental application of the data and input parameters corresponding to a current iteration to one or more trained machine learning models, each application to generate a synthetic output of a polymerization operation and a performance of the synthetic output based on the input parameters of the current iteration, the design space to include data indicative of outputs or synthetic outputs and one or more of a corresponding plurality of polymerization operation settings, a corresponding plurality of polymeric composition components, or a corresponding plurality of associated performances; and in response to a request for one or more of a polymeric composition or polymerization operation settings that meet or exceed a specified performance, determining one or more of the plurality of polymeric compositions or the plurality of polymerization operation settings that meet or exceed the specified performance based on the design space.

2. The method of claim 1, further comprising: generating one or more visualizations based on the one or more of the plurality of polymeric compositions or the plurality of polymerization operation settings that meet orexceed the specified performance based on the design space, and wherein the one or more visualizations comprise one or more interactive plots or a tabular list.

3. The method of claim 1, further comprising: prior to determining the one or more of the plurality of polymeric compositions or a plurality of polymerization operation settings, determining if the request includes unknown polymeric compositions or unknown polymerization operations settings; and in response to a determination that the request includes unknown polymeric compositions or unknown polymerization operations settings: generating a prompt for additional data, in response to receipt of the additional data, retraining the trained machine learning models, and updating the design space via application of data corresponding to the unknown polymeric compositions or unknown polymerization operations settings to the trained machine learning models.

4. The method of claim 1, wherein the polymerization operation settings comprise one or more polymerization reactor settings, process settings, catalyst type, catalyst amount, temperature, pressure settings, flow rates, residence times, concentrations of reactants, solvent type, extrusion equipment settings, blending equipment settings, or solvent amount.

5. A method to generate a design space for determining one or more polymeric compositions or polymerization operation settings, the method comprising: receiving data that corresponds to a plurality of experimental data, real-time polymerization operations data, and historical polymerization operations data; determining input parameters based on the data, the input parameters to include at least a minimum value, a maximum value, and a increment value for each of a plurality of polymerization operation settings and each of a plurality of polymeric composition components; and while the maximum value for the plurality of polymerization operation settings has not been reached, iteratively:(a) while the maximum value for the plurality of polymeric composition components has not been reached, iteratively:(i) generating, via application of the data and input parameters corresponding to a current iteration to one or more trained machine learning models, a simulation of a polymerization operation based on the input parameters of the current iteration,(ii) determining new entries in the design space based on the simulation,(iii) generating a new selected amount of the component of the selected polymeric composition based on the input parameters, and(iv) repeating steps (i), (ii), and (iii) for the new selected amount of the component of the selected polymeric composition,(b) generating new selected polymerization operation settings based on the input parameters, and(c) repeating steps (a) and (b) for the new selected polymerization operation settings.

6. The method of claim 5, further comprising: generating a user interface to include a user input function that allows a user to obtain (a) one or more of a plurality of polymerization operation settings or a plurality of amounts of polymeric composition components and (b) an associated performance for each of the one or more of plurality of polymerization operation settings or the plurality of amounts of polymeric composition components; in response to a selection of one of the plurality of polymerization operation settings and one of the plurality of amounts of polymeric composition components, initiating a polymerization operation based on the selection.

7. A system to generate a design space for determining one or more polymeric compositions or polymerization operation settings, the system comprising: a communications circuitry configured to:receive data corresponding to one or more of polymerization operations, polymerization experiments, or historical polymerization operations; a pre-processing circuitry configured to: determine input parameters including at least a minimum, a maximum, and an increment for each parameter corresponding to polymerization operations in the data and for each amount of a plurality of polymeric compositions in the data; a modeling circuitry configured to: iteratively generate a design space via incremental application of the data and the input parameters corresponding to a current iteration to one or more trained machine learning models, the design space including data indicative of outputs or synthetic outputs and one or more of a plurality of corresponding polymerization operation settings, a plurality of corresponding polymeric compositions, or a plurality of corresponding associated performances; and a chemical operation controller configured to: in response to receipt of a request for one or more of a polymeric composition or polymerization operation settings that meet or exceed a specified performance, determine one or more of a plurality of variable polymeric compositions or a plurality of polymerization operation settings that meet or exceed the specified performance based on the design space.

8. The system of claim 7, comprising: a visualization circuitry configured to: generate one or more visualizations based on the one or more of the plurality of variable polymeric compositions or the plurality of polymerization operation settings that meet or exceed the specified performance based on the design space; display options to a user interface allowing selection of the one or more visualizations based a type of each of the one or more visualizations; and in response to selection of the one or more visualizations, display selected visualizations to the user interface.

9. The system of claim 7, comprising:a design space circuitry configured to: process an output of the one or more trained machine learning models to produce data formatted for the design space.

10. The system of claim 7, wherein the modeling circuitry is configured to, in response to selection of one of the one or more of the plurality of the variable polymeric compositions and one of the plurality of polymerization operation settings, retrain the one or more trained machine learning models.

11. The system of claim 7, wherein the one of the one or more trained machine learning models includes an image-based machine learning model trained based on a series of images of polymeric products, corresponding data, and acceptance or rejection of the polymeric products, and wherein the request includes a visual aspect of a polymeric product produced based on one or more of the polymeric compositions or the polymerization operation settings.

12. The system of claim 7, wherein the modeling circuitry is configured to: in response to a missing portion of data in the one or more of the plurality of polymeric compositions, the plurality of associated performances, or the plurality of polymerization operation settings, generate the missing portion of data via a physics based model and based on the data and the design space.

13. A controller for generating a design space for determining one or more polymeric compositions or polymerization operation settings, the controller comprising: an input / output in signal communication with polymerization equipment, the controller configured to: in response to receipt of experimental data, determine input parameters, the input parameters including at least a minimum, a maximum, and an increment of process parameters in the experimental data; generate, via iterative and incremental application of the experimental data and the input parameters corresponding to a current iteration to one or more trainedmachine learning models, a design space to include one or more of a plurality of polymeric compositions, associated performances, or polymerization operation settings; in response to receipt of a request for one or more of a polymeric formula or polymerization operation settings that meet or exceed a specified performance, determining one or more of a plurality of variable polymeric formulas or a plurality of polymerization operation settings that meet or exceed the specified performance based on the design space; and display one or more visualizations generated based on the one or more of the plurality of variable polymeric formulas or the plurality of polymerization operation settings that meet or exceed the specified performance based on the design space.

14. The controller of claim 13, wherein the one or more visualizations include one or more of graphical visualizations or tabular visualizations.

15. The controller of claim 13, wherein the request includes an application critical -to- quality (CTQ) input, and wherein the controller is configured to convert the application CTQ input to a material to process CTQ.

Images