Method for monitoring a production process

Abstract

A method for monitoring an ethanol fermentation process involves positioning an accelerometer at or above the liquid level of a process stream in a fermentation vessel. The method includes collecting vibration data from the accelerometer during fermentation, analyzing the frequency spectra of the vibration data, and determining the fermentation rate based on the analysis. The system can also incorporate temperature data to improve prediction accuracy. The method can identify amplitude ranges in the frequency spectra, correlate these to carbon dioxide bubble production, and determine parameters such as bubble size, depth of origin, and microorganism location. The system can provide alerts for deviations in fermentation rate and recommend interventions. Additionally, the system can monitor various stages of ethanol production, including liquefaction, oil separation, centrifugation, evaporation, and drying, using multiple accelerometers to determine process parameters such as viscosity, composition, and equipment performance.

Description

3097.048W01METHOD FOR MONITORING A PRODUCTION PROCESSCLAIM OF PRIORITY

[0001] This application claims the benefit of priority to U.S. Provisional Patent Application No. 63 / 722,481, filed on November 19, 2024, which is incorporated by reference herein in its entirety.TECHNICAL FIELD

[0002] The invention relates generally to systems and methods for process monitoring such as for grain processing in ethanol production. The invention relates more particularly, in various embodiments, to systems and methods of monitoring using sensed vibration spectra.BACKGROUND

[0003] Ethanol and various types of feed stock can be produced from processing of grain. During such processing, fiber, syrup, oil, and protein can be produced. Such processes can include a variety of steps, such as separation steps, at varying times during processing. In these processes, fermentable carbohydrates may be used during production of syrup, oils, and protein from the grain.

[0004] Ethanol production is a complex process that involves several key steps to convert biomass, typically corn or other starchy grains, into fuel-grade ethanol. The process begins with the preparation of the feedstock, which involves grinding the biomass into a fine meal to increase its surface area for more efficient processing.

[0005] After grinding, the ground biomass is mixed with water to create a slurry. Enzymes are added to begin breaking down the starch and other carbohydrate polymers into simpler sugars. This process occurs in liquefaction vessels, where temperature and pH are carefully controlled to optimize enzyme activity.

[0006] Following liquefaction, the mixture undergoes fermentation, which is the core of the ethanol production process. In fermentation vessels, yeast is added to the sugar-rich solution. The yeast consumes the sugars and produces ethanol and carbon dioxide as byproducts. This process typically takes3097.048W0148-72 hours and is closely monitored to ensure optimal conditions for yeast activity and ethanol production.

[0007] After fermentation, the resulting mixture is distilled. During distillation, the ethanol is separated from the water and other components based on their different boiling points. This process occurs in distillation columns and results in a more concentrated ethanol solution. Further purification and separation steps are then typically performed to remove solid residues from the process. These solid residues are often processed into valuable co-products such as distillers dried grains with solubles (DDGS), which are used as animal feed.

[0008] Ethanol has several important uses, with its primary application being as a renewable fuel additive for gasoline. When blended with gasoline, ethanol helps reduce greenhouse gas emissions and decreases reliance on fossil fuels. It is commonly used in blends such as E10 (10% ethanol, 90% gasoline) and E85 (85% ethanol, 15% gasoline) for flexible-fuel vehicles. Ethanol also finds applications in the chemical industry as a solvent and as a key ingredient in the production of various products including pharmaceuticals, cosmetics, and industrial chemicals. Its versatility and renewable nature make it an important component in efforts to reduce carbon emissions and promote sustainable industrial practices.

[0009] Monitoring of ethanol production, grain processing and feed stock production is known using sensors and sample testing. However, such sensors such as near-infrared spectroscopy and testing methods can be cost prohibitive and / or time consuming to develop models and perform. They are frequently not commercially relevant for implementing monitoring in real-time or near real-time.SUMMARY OF THE DISCLOSURE

[0010] The present application provides a novel approach to monitoring and controlling ethanol fermentation and other grain production processes using accelerometers and advanced data analysis techniques. For example, the present application discloses systems and methods of monitoring a fermentation process by positioning an accelerometer or a combination of an accelerometer and a microphone above a process stream (referred to sometimes herein as a liquid level) in a fermentation vessel. This strategic placement allows for the3097.048W01 measurement of vibrations (or sound) caused by CO2 bubble production at the surface of the process stream, which directly correlates to ethanol production.

[0011] According to one example, the methods and systems can utilize frequency spectrum analysis of vibration data collected from the accelerometer or a combination of sound data collected with the microphone and the vibration data. Specifically, such analysis with regard to vibration data can identify a first amplitude range of between: about 50 nm and about 1500 nm, about 100 nm and about 1000 nm, about 250 nm and about 800 nm, about 350 nm and about 700 nm or about 400 nm and about 600 nm. It was found that this amplitude range strongly correlates with CO2 bubble production and, consequently, the fermentation rate. The present application also considers / analyzes how the first amplitude range correlates (e.g., combines with or measures as a ratio with) other amplitude ranges. This correlation can enhance the accuracy of fermentation rate determination, can allow for determination of an average bubble size, depth of bubble origination in the fermentation vessel, bubble population determination and identification of microorganism location. This analysis enables a more precise understanding of the fermentation process in real-time but at lower cost using low-cost accelerometers, and optionally in conjunction with temperature sensor(s) in lieu of more expensive near infrared sensors or more tedious testing and analysis.

[0012] Furthermore, according to some examples, the present application incorporates artificial intelligence such as a machine learning model. This machine learning model can be trained using historical vibration data, temperature data, and known fermentation rate data (e.g., determined by sample testing, use of near-infrared spectroscopy and / or other known methods). By integrating temperature data such as from the fermentation vessel with vibration data from the accelerometer, the system and method can improve the prediction accuracy of the fermentation rate determination. Additionally, the model can be used to infer yeast activity levels. The machine learning model have been validated and demonstrate the ability to achieve prediction accuracy of 99% or higher for determining fermentation rate. Such prediction accuracy represents an advancement in process monitoring capabilities. The machine learning model can be continuously updated with new data, allowing for ongoing enhancement in prediction accuracy and adaptability to changes in the production process or3097.048W01 equipment over time. This ensures that the system remains effective and relevant even as production processes evolve or equipment ages.

[0013] The present systems and methods can provide for an ability to detect fermentation issues early, potentially within the first 6-12 hours of the fermentation process beginning. This early detection enables timely interventions, such as adding yeast or antibiotics to the fermentation vessel, which can improve the overall efficiency and yield of the fermentation process.

[0014] Other aspects of the present application extend beyond fermentation monitoring to encompass various stages of the ethanol production process. Accelerometers can be placed at different points in the production facility, including at fermentation vessel(s), at yeast / microorganism propagation vessel(s), at liquefaction vessel(s), at oil separation unit(s), centrifuge(s), evaporator(s), and drying drum(s). This monitoring approach allows for the determination of various process parameters such as but not limited to fermentation rate, starch conversion, product composition, product viscosity, slurry qualities, hydrolysis progress, oil recovery yield, equipment performance, potential fouling in evaporators, and material flowability in drying drums.

[0015] Thus, the present application provides an improved approach to monitoring and controlling ethanol fermentation and related grain production processes. Using vibration or sound sensing, advanced data analysis, and artificial intelligence, the present systems and methods offer the potential for improvements in process efficiency, early problem detection, and enhanced quality control across multiple stages of grain processing and ethanol production using low-cost sensors. The systems and methods abilities to provide real-time insights and predictive capabilities can aid in optimizing these production processes.BRIEF DESCRIPTION OF THE FIGURES

[0016] In the drawings, which are not necessarily drawn to scale, like numerals describe substantially similar components throughout the several views. Like numerals having different letter suffixes represent different instances of substantially similar components. The drawings illustrate generally, by way of example, but not by way of limitation, various examples discussed in the present document.3097.048W01

[0017] FIG. l is a schematic diagram a system and method for grain processing and ethanol production including accelerometers, processing circuitry and memory, according to an example of the present application.

[0018] FIG. 2 is a plot showing vibration data collected from an accelerometer positioned on or within a fermentation vessel over a fermentation period, the plot illustrates changes in vibration amplitude corresponding to different stages of the processing, according to an example of the present application.

[0019] FIG. 3 is a plot that includes a representation of the accelerometer signal of FIG. 2, in the plot the vertical axis represents frequency and the color represents changes in amplitude, according to an example of the present application.

[0020] FIG. 4 illustrates a machine learning engine for predicting fermentation rate and other process parameters, according to an example of the present disclosure.

[0021] FIG. 5 illustrates of a system for monitoring and controlling an ethanol production process using vibration data and optionally temperature data, including components such as a machine learning system, and user interface elements according to an example of the present application.

[0022] FIG. 6 is a flowchart of a method for monitoring an ethanol fermentation process using vibration data collected from an accelerometer positioned above the liquid level in a fermentation vessel according to an example of the present application.

[0023] FIG. 7 is a flowchart of a method for monitoring multiple aspects of an ethanol production process using accelerometers positioned at various locations throughout the production plant according to an example of the present application.

[0024] FIG. 8 is a block diagram of an example machine upon which any one or more of the techniques, apparatuses, systems or methods discussed herein may perform according to an example of the present application.

[0025] FIG. 9 is a plot of a fermentation vessel with composition of desired compounds plotted over time, the compounds determined using nearinfrared spectroscopy sensors according to an example of the present application.3097.048W01

[0026] FIG. 10 is a plot of the actual ethanol production rate (the fermentation rate) of FIG. 9 measured against instantaneous predicted rate determined using the methodology discussed herein.DETAILED DESCRIPTION

[0027] Reference will now be made in detail to certain examples of the disclosed subject matter, examples of which are illustrated in part in the accompanying drawings. While the disclosed subject matter will be described in conjunction with the enumerated claims, it will be understood that the exemplified subject matter is not intended to limit the claims to the disclosed subject matter.

[0028] The Detailed Description explains embodiments of the subject matter and the various features and advantageous details more fully with reference to non-limiting embodiments and examples that are described and / or illustrated in the accompanying figures and detailed in the following attached description. Descriptions of well-known components and processing techniques may be omitted so as to not unnecessarily obscure the embodiments of the subject matter. The examples used herein are intended merely to facilitate an understanding of ways in which the subject matter may be practiced and to further enable those of skill in the art to practice the embodiments of the subject matter. Accordingly, the examples, the embodiments, and the figures herein should not be construed as limiting the scope of the subject matter.

[0029] The present disclosure describes environments and techniques for monitoring processes that provide for the separating solids from liquids in a process stream, which may be obtained from a production facility. For instance, the production facility may include, but is not limited to, biofuels, alcohol, animal feed, oil, biodiesel, pulp and paper, textile, chemical industry, and other fields. Thus, the techniques, methods and systems discussed herein are not limited to ethanol production. The processes discussed can produce valuable feed products and co-products. The feed products may include, but are not limited to, Distiller's Dried Grains with Solubles (DDGS), Condensed Distillers Solubles (CDS), Single Cell Protein (SCP), UltraMax™, SolMax™, grain distillers dried yeast, syrup with fiber, and the like. The co-products may also3097.048W01 include, but are not limited to, corn distillers oil, clarified products, and / or concentrated products.

[0030] The above Summary of Disclosure discusses advantages of the present techniques, systems and methods. These advantages improve the rapidity and accuracy of monitoring of various process steps, while reducing costs. Discussed below in regard to FIG. 1 are an overview of a method and a system of processing grain and example produced products including ethanol from these methodologies.

[0031] A method of processing grain such as for ethanol production and other purposes can include a series of steps, including milling the grain, fiber separation, feed optimization processing, and solids separation, among other steps. The methods discussed herein can allow for production of ethanol and feed among other useful products. An example system 100 and method 101 are depicted in FIG. 1, on which an example method of processing grain for ethanol production can be accomplished.

[0032] The method 101 may receive feedstock of a grain that includes, but is not limited to, barley, beets, cassava, corn, cellulosic feedstock, grain, milo, oats, potatoes, rice, rye, sorghum grain, triticale, sweet potatoes, lignocellulosic biomass, wheat, and the like, or pulp. Lignocellulosic biomass may include corn fiber, corn stover, corn cobs, cereal straws, sugarcane bagasse and dedicated energy crops, which are mostly composed of fast growing tall, woody grasses, including, but not limited to, switch grass, energy / forage sorghum, miscanthus, and the like. Also, the feedstock may further include, grain fractions or by-products as produced by industry, such as hominy, wheat middlings, corn gluten feed, Distillers Dried Grains with Solubles, and the like. The feedstock may include, an individual type, a combined feedstock of two types or of multiple types, or any combination or blend of the above grains. The feedstock may include, but is not limited to, one to four different types combined in various percentage ranges. The feedstock may be converted into different products and co-products that may include, but is not limited to, ethanol, syrup, distillers oil, distillers dried grains, distillers dried grains with solubles, condensed distillers solubles, wet distillers grains, and the like. For brevity purposes, the method 101 of using a single stream of feedstock will be described with reference to FIG. 1.3097.048W01

[0033] The system 100 can be used to receive grain 102 and produce products such as fiber 134, syrup 189, oil 186, and high protein feed 192, in addition to sending certain separated components to produce ethanol 164. FIG. 1 depicts the system 100 and the method 101 (a process flow) that includes passing grain 102 through the system components.

[0034] The system 100 can include milling station 110, slurry tanks 114, liquefaction tanks 118, fiber processing system 200 including selective separating system 210 and fiber separation system 240, ethanol processing system 140 with fermentation vessel(s) 142, distillation apparatuses 148, and dehydration apparatuses such as an evaporator 160, feed processing system 170, separation 180 and 187, and drying drum 190.

[0035] First, the grain 102 is received. The grain is sent to the milling station 110, which may be roller mills. Grain flour 112 is produced and sent to the slurry tanks 114. Here, the grain flour 112 is mixed with cook water and combined to form a slurry 116. In an example, the method 101 adds a liquefying enzyme, such as alpha-amylase to this mixture. The alpha-amylase enzyme hydrolyzes and breaks starch polymer into short sections, dextrins, which are a mix of oligosaccharides. The method 101 maintains a temperature between about 60° C. to about 100° C. (about 140° F. to about 212° F., about 333 K to about 373 K) in the slurry tank 104 to cause the starch to gelatinize and a residence time of about 30 to about 60 minutes to convert insoluble starch in the slurry to soluble starch. The slurry may have suspended solids content of about 26% to about 40%, which includes starch, fiber, protein, and oil. Other components in the slurry tanks 114 may include, grit, salts, and the like, as is commonly present on raw incoming grain from agricultural production, as well as recycled waters that contain acids, bases, salts, yeast, and enzymes. The method 101 can adjust the pH of the slurry.

[0036] The slurry 116 is flowed to the liquefaction tanks 118, wherein liquefaction of the slurry 116 occurs. This converts the slurry to mash. The method 101 uses a temperature range of about 80° C. to about 150° C. (about 176° F. to about 302° F., about 353 K to about 423 K) to hydrolyze the gelatinized starch into maltodextrins and oligosaccharides to produce a liquefied mash. Here, the method 101 produces a mash stream, which has about 26% to about 45% total solids content. The mash may have suspended solids content3097.048W01 that includes protein, oil, fiber, grit, and the like. In embodiments, one or more liquefaction tanks 118 may be used in the method 101. The liquefaction generates a stream 120, which is sent to the fiber processing system 200. The liquefaction tanks 118 also produces a stream 122 that is sent to the ethanol processing system 140.

[0037] The stream 120 can include fiber for processing in the fiber processing system 200 for further processing and separation of fiber. The selective separating system 210 is a selective flaking system to recover starch for conversion to ethanol and allow for available oil recovery. The liquids form stream 120 can optionally be routed back to liquefaction 118 after separation. After separated at the selective separating system 210, the stream 230, which includes fiber, can be sent through the fiber separation system 240. At the fiber separation system 240, fiber is removed from the process stream 230 prior to fermentation. This can allow for more fermentable carbohydrates to be loaded into each batch for fermentation. Fiber 134 can be produced, such as in the form of a fiber cake, and sent for further processing at the dryer 190. An additional liquid stream 132 can be produced and sent to the ethanol processing system 140.

[0038] Both the second stream 122 and the additional liquid stream 132 can be sent to the ethanol processing system 140. There, the stream 122 and the stream 132 can be fermented at the fermentation vessel(s) 142. Excess carbon dioxide 144 can be released at this stage. The fermented liquid 146 can be distilled at the distillation apparatuses 148. The distilled liquid 150 can be dehydrated at the dehydration apparatuses (e.g., evaporator 160), producing ethanol 164. Excess solids 162 from the distillation apparatuses 148 can be sent to the feed optimization system 170.

[0039] At the fermentation vessel(s) 142, the method 101 adds a microorganism to the mash for fermentation in the fermentation vessel(s) 142. The method 101 may use a common strain of microorganism, such as Saccharomyces cerevisiae to convert the simple sugars (i.e., maltose and glucose) into alcohol with solids and liquids, CO2, and heat. The method 101 may use a residence time in the fermentation vessel(s) 142 as long as about 50 to about 74 hours. However, variables such as a microorganism strain being used, rate of enzyme addition, temperature profile for fermentation, targeted alcohol3097.048W01 concentration, and the like, may affect fermentation time. In embodiments, one or more fermentation vessel(s) may be used in the method 101.

[0040] The method 101 creates alcohol, solids, liquids, microorganisms, and various particles through fermentation in the fermentation vessel(s) 142. Once completed, the mash is commonly referred to as beer, which may contain about 10% to about 20% alcohol, plus soluble and insoluble solids from the grain components, microorganism metabolites, and microorganism bodies. The microorganism may be recycled in a microorganism recycling step, which is an option. The part of method 101 that occurs prior to distillation may be referred to as the “front end”, and the part of the method 101 that occurs after distillation may be referred to as the “back end”. The method 101 distills the beer to separate the alcohol from the non-fermentable components, solids and the liquids by using a distillation process, which may include one or more distillation columns, beer columns, and the like. The method 101 pumps the beer through the distillation apparatuses 148, which is boiled to vaporize the alcohol or produce concentrated stillage. The method 101 condenses the alcohol vapor in distillation apparatuses 148 where liquid alcohol exits through a top portion of the distillation apparatuses 148 at about 90% to about 95% purity ethanol, 5% water which is about 190 proof. In embodiments, the distillation columns and / or beer columns may be in series or in parallel. At the evaporator 160, the method 101 removes any moisture from the 190 proof alcohol by going through dehydration. The dehydration may include one or more drying column(s) packed with molecular sieve media to yield a product of nearly 100% alcohol. Further ethanol processing can be done using holding tanks, etc.

[0041] The feed processing system 170 can produce protein-rich feed. As shown in FIG. IB, feed processing system 170 can include multiple separations producing respective stream 172 and additional stream(s) that are not illustrated specifically in FIG. 1. The stream 172 can, for example, include liquid and solids, and be sent to the solids separation system 180 (e.g., a centrifuge 180A). The stream 174 can include, for example, solids, and be sent to drying drum 190. An example feed processing separation system is shown and discussed in U.S. Patent Application No. 16 / 875,894, which are herein incorporated in its entirety.

[0042] The solids separation system 180 can be used to separate stillage process streams, such as the stream 172, into valuable components such as3097.048W01 protein, soluble 188, and oils 186. In an example, the solids separation system 180 can include preparation technology in addition to the centrifuge 180A. The centrifuge 180A can separate the process stream into two streams 182 and 184. The first stream 182 which can include liquids, can be sent to further separation 187, evaporators (not shown) or other components to produce a syrup 189 and oil 186. The second stream 184, which can include solids, can be sent to the drying drum 190. An example solids separation system is shown and discussed in U.S. Patent Applications Nos. 17 / 672,493, 16 / 624,836, 16 / 624,831, 16 / 624,824, 17 / 683,011, and 16 / 624,811, which are herein incorporated in their entirety. Another example of a system for processing grain can be found in U.S. Provisional Patent Application Serial Nos. 63 / 444,487 and 63 / 472,57 the disclosure of each of which is incorporated herein in its entirety by reference.

[0043] The system 100 and method 101 can additionally include processing circuitry 300 and memory 302 in electronic communication with various sensors such as accelerometers 304 A, 304B, 304C, 304D, 304E, 304F and 304G and temperature sensor 305. Unless otherwise noted, the location of and components the accelerometers 304A, 304B, 304C, 304D, 304E, 304F and 304G are placed within or mounted to is purely exemplary in FIG. 1. The present application contemplates that the accelerometers 304 A, 304B, 304C, 304D, 304E, 304F and 304G can be positioned within our mounted externally (e.g., on top or side surfaces) of the various vessels shown above the processing liquid line (e.g., above the surface). Additionally, although not specifically illustrated, the accelerometers 304 A, 304B, 304C, 304D, 304E, 304F and 304G can be mounted to equipment such as shafts or piping adjacent the vessels rather than being mounted within or on surfaces of the vessels.

[0044] The processing circuitry 300 can include, for example, software, hardware, and combinations of hardware and software configured to execute several functions related to, among others, operation of the system 100. The processing circuitry 300 can be an analog, digital, or combination analog and digital controller including a number of components. As examples, the processing circuitry 300 can include integrated circuit boards or ICB(s), printed circuit boards PCB(s), processor(s), data storage devices, switches, relays, or any other components. Examples of processors can include any one or more of a microprocessor, a controller, a digital signal processor (DSP), an application3097.048W01 specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or equivalent discrete or integrated logic circuitry. Commercially available microprocessors can be configured to perform the functions of the processing circuitry 300. Various known circuits may be associated with processing circuitry 300, including power supply circuitry, signal-conditioning circuitry, actuator driver circuitry (i.e., circuitry powering solenoids, motors, or piezo actuators), and communication circuitry. In some examples, the processing circuitry 300 may be part of a control unit, computer, computer network, etc.

[0045] The processing circuitry 300 can include the memory 302 such as memory circuitry. The memory 302 may include storage media to store and / or retrieve data or other information such as, operational algorithms. Storage devices, in some examples can be a computer-readable storage medium. The data storage devices can be used to store program instructions for execution by processor(s) of the processing circuitry 300, for example. The storage devices, for example, are used by software, applications, algorithms, as examples, running on and / or executed by the processing circuitry 300. The storage devices can include short-term and / or long-term memory and can be volatile and / or nonvolatile. Examples of non-volatile storage elements include magnetic hard discs, optical discs, floppy discs, flash memories, or forms of electrically programmable memories (EPROM) or electrically erasable and programmable (EEPROM) memories. Examples of volatile memories include random access memories (RAM), dynamic random-access memories (DRAM), static randomaccess memories (SRAM), and other forms of volatile memories known in the art.

[0046] As shown in FIG. 1, the processing circuitry 300 electronically communicates with the accelerometers 304 A, 304B, 304C, 304D, 304E, 304F and 304G and temperature sensor 305 to receive data therefrom. The accelerometer 304 A can be positioned within, mounted on an outside surface or inside surface of the fermentation vessel(s) 142. According to one example, the accelerometer 304 A can be at or above a liquid level of a process stream within the fermentation vessel(s) 142. In most vessels, this can position the accelerometer 304 A above 75% to up to 100% of a height of the fermentation vessel(s) 142, for example. It has been determined through extensive testing that positioning the accelerometer 304 A above the liquid level (rather than3097.048W01 below the level or within the liquid of the process stream) of the process stream can reduce vibration interference from pumps, etc. providing for more accurate vibration data collection. According to some examples, another sensor such as a microphone 304A’ can be substituted for or combined with the accelerometer 304 A. The microphone 304 A’ can be placed within the fermentation vessel(s) 142 above the liquid level of the process stream within the fermentation vessel(s) 142.

[0047] The accelerometer 304B can be located at or adjacent the slurry tanks 114. The accelerometer 304C can be located at or adjacent the liquefaction vessel(s) 118. Thus, the method 101 includes mounting at least one accelerometer (accelerometer 304B) to a side of the liquefaction vessel(s) 118 or piping adjacent the liquefaction vessel(s) 118. The accelerometers 304D can be positioned at or adjacent the solids separation 180 (on, in or adjacent the centrifuge 180A). The accelerometer 304D can be used to detect vibrations into, within, and out of the centrifuge 180A. The accelerometer 304E can be positioned on, in or adjacent the further separation 187 (e.g., a centrifuge, decanter, disc stack separator(s), heat and hold tank, etc.). The accelerometer 304D can be used to detect vibrations into, within, and out of the further separation 187. The accelerometer 304F can be positioned on, in or adjacent the drying drum 190. The accelerometer 304G can be positioned on, in or adjacent the evaporator 160. The temperature sensor 305 can be positioned at the fermentation vessel(s) 118 such as within the process stream or at another location within the fermentation vessel(s) such as above the process stream.

[0048] The accelerometers 304 A, 304B, 304C, 304D, 304E, 304F and304G can be configured to collect / sense vibration data of the various equipment or of shafts and / or piping associated with the various equipment discussed above. This vibration data, along with temperature data gathered by the temperature sensor 305 in some examples can be used for monitoring by the system 100 (in particular the processing circuitry 300) and method 101 as further discussed subsequently. One or more of the accelerometers 304A, 304B, 304C, 304D, 304E, 304F and 304G can multiple (two or more) accelerometers mounted to the same equipment (or unit) according to some examples. These multiple accelerometers can be used in combination such as to triangulate or otherwise refine capture of the vibrations.3097.048W01

[0049] Use of accelerometers 304A, 304B, 304C, 304D, 304E, 304F and / or 304G can detect phase changes and other parameters within the system 100. For example, with regard to liquefaction, use of the accelerometer 304C mounted to the side of or within the liquefaction vessels or piping can be used to characterize different slurry qualities. The accelerometer 304C, for example, can detect changes in suspended solids size distribution as well as a degree of hydrolysis of carbohydrates to oligiomers and sugars. Many approaches have been attempted to apply real-time hydrolysis monitoring. However, previous systems struggle with accurately sampling the slurry as well as fouling of the sampler system. The accelerometer 304C can be applied to the outside of the vessel or associated piping in a non-evasive way that enables a longer duration of reliable operation as compared with prior sampling methodology.

[0050] With regard to fermentation, use of accelerometer(s) 304 A mounted on the inside or outside of a fermentation vessel 142 above the liquid level of the process stream can accurately detect the bubble activity within the vessel. This is effectively detecting CO2 bubbles on the liquid surface. The vibration frequency and amplitude are recorded and can be used to model the kinetic rate of the fermentation. Frequency characterization can be executed such that not only the population of bubbles can be estimated, but also the size (and thus the depth in the vessel where they originated, thus identifying microorganism location). Previous fermentation monitoring solutions applied a hydrophone or vibration sensor below the liquid level operating as sonar. In such prior methods, specific sound is emitted across the vessel and then when detected is correlated to the speed of sound in the fluid, which relates to liquid density. This approach is not ideal as density changes very slowly in an industrial setting. Additionally, local irregularities in bubbles can cause significant errors utilizing such methodology. In contrast, in the disclosed systems and methods, the accelerometer(s) 304 A can be placed above the liquid level of the process stream, thus reacting to the liquid surface, which nets out the entire vessel activity.

[0051] With regard to oil separation, use of accelerometer(s) 304E can be used to detect the quality of streams in, within and out of the centrifuge, as well as the centrifuge performance. The ethanol process tends to have an emulsion at the point of oil extraction. Traditional devices like turbidity probes3097.048W01 struggle with the phase interface and saturate the signal easily, making fine tuning of oil recovery yield difficult to set up with a control loop.

[0052] With regard to drying, use of accelerometer(s) 304F on or within drum dryers is able to characterize the flowability of the material as it conveys through the drum dryer. Additionally the accelerometer(s) 304F on or within drum dryers are able to characterize bulk particle size of the material in the dryer, as different sizes will generate different sound vibrations as it is conveyed through the dryer.

[0053] FIG. 2 is a plot 306 showing vibration data collected from the accelerometer 304 A positioned on or within the fermentation vessel(s) 142 of FIG. 1. The plot 306 is taken over a fermentation cycle. The plot illustrates changes in vibration amplitude corresponding to different stages of the processing for fermentation including at filling 308 and draining 310 of the fermentation vessel. The plot illustrates with numbers 1) start of filling, 2) 25% filled, 3) 50% filled and 4) 75% filled. According to the example of FIG. 2, the accelerometer was positioned above the liquid level of the process stream of the fermentation vessel such that the accelerometer is positioned above the 75% (the maximum fill mark). The accelerometer can be positioned either within or on a side or top of the vessel above this maximum fill mark.

[0054] The plot 306 illustrates the overall vibration sensed by the accelerometer. The plot 306 includes vibration that results from fermentation 312. Such vibration results, among other things, from bubble activity within the fermentation vessel. The accelerometer is listening to CO2 bubbles on the liquid surface of the process stream. The vibration frequency and amplitude are recorded by the accelerometer during the fermentation 312 including in an area of high fermentation activity 312A.

[0055] FIG. 3 is a plot 314 that includes a representation of the accelerometer signal of FIG. 2. In the plot of FIG. 3, the vertical axis represents frequency and the color represents changes in amplitude (yellow and green highlight larger changes in amplitude as compared with smaller amplitude changes indicated with blue). The plot 314 is distinctive in capturing vibration due to filling 308, draining 310, fermentation 312 and the area of high fermentation activity 312A that was also illustrated in FIG. 2. The plot 314 is able to capture these amplitude changes due to the positioning of the3097.048W01 accelerometer above the above the liquid level of the process stream of the fermentation vessel. Other accelerometer locations were tested but did not provide as clear a vibration signal with distinctive changes in amplitude due to interference.

[0056] As will be discussed in further detail subsequently, the plot 306 of FIG. 2 and the plot 314 of FIG. 3 can be analyzed to determine the fermentation rate as corresponding to carbon dioxide bubble production within the process stream of the fermentation vessel. According to one example, the vibration data such as that of plot 306 and / or 314 can be analyzed such as by analyzing a frequency spectra of the vibration data. This can allow for determining the fermentation rate based on the analyzing the frequency spectra. In particular, analyzing the frequency spectra of the plot 306 and / or the plot 314 can include identifying a first amplitude range between: about 50 nm and about 1500 nm, about 100 nm and about 1000 nm, about 250 nm and about 800 nm, about 350 nm and about 700 nm or about 400 nm and about 600 nm and correlating amplitude changes in the first amplitude range to carbon dioxide bubble production within the process stream of the fermentation vessel. According to further examples, two or more amplitude ranges can be identified in addition to the first amplitude range. The analysis can include comparing the first amplitude range to the two or more amplitude ranges and determining the fermentation rate based on the comparing the first amplitude range with the two or more amplitude ranges. The two or more amplitude ranges can optionally include at least one amplitude range below and one amplitude range above the first amplitude range.

[0057] Vibration spectra analysis on the plot 306 of FIG. 2 and the plot 314 of FIG. 3 can determine a depth within the fermentation the carbon dioxide bubble production originates based upon the correlating the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production. This can allow for identifying microorganism location based upon the depth. Furthermore, the vibration spectra analysis can determine an average size of the carbon dioxide bubble production and / or a population of the carbon dioxide bubble production based upon the correlating the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production.3097.048W01

[0058] FIG. 4 illustrates a machine learning engine 400 for predicting fermentation and other process parameters in accordance with some embodiments. A system (such as those discussed herein) may calculate one or more weightings for criteria based upon one or more machine learning algorithms. FIG. 4 shows an example machine learning engine 400 according to some examples of the present disclosure. Machine learning engine 400 may be part of the system 100 of FIG. 1 or method 101 of FIG. 1, for example. The engine 400 can be implemented using a database, a server, etc., or the machine learning system 508 of FIG. 5, described below.

[0059] Machine learning engine 400 utilizes a training engine 402 and a prediction engine 404. Training engine 402 inputs historical data 406 for historical test results on composition, sensor data (e.g., accelerometer, temperature, pH, near-infrared, infrared, etc.) and further information on fermentation or other steps in the method 101 (FIG. 1) including operational inputs from various equipment discussed in regard to FIG. 1. This historical data 406 is input into parameter determination engine 408. The historical data 406 may be labeled with an indication, such as a degree of success of an outcome of fermentation or degree of success or outcome of other processing step. This indication may include or be associated with vibration data from various locations, temperature data from various locations, bubble related data, CO2 related data, fermentation rate (actual calculated), starch conversion, oil recovery yield, centrifuge performance, evaporator fouling, product composition and product viscosity, or the like. In some examples, an outcome may be subjectively assigned to historical data, but in other examples, one or more labelling criteria may be utilized that may focus on objective outcome metrics such as those discussed above (e.g., vibration data from various locations, temperature data from various locations, bubble related data, depth of bubbles when produced, volume of CO2 produced, average size of bubbles produced, population of the carbon dioxide bubble production, fermentation rate (calculated), starch conversion, lignocellulose conversion, oil recovery yield, centrifuge performance, evaporator fouling, product composition and product viscosity or the like).

[0060] Parameter determination engine 408 determines one or more features or parameters 410 from this historical data 406. Stated generally,3097.048W01 features 410 are a set of the information input and is information determined to be predictive of a particular outcome. Example features are the parameters given above and can further include process setting(s), product stream compositions, timing of changes in process, other sensor derived input(s), etc. In some examples, the features 410 may be all the historical activity data, but in other examples, the features 410 may be a subset of the historical activity data. The machine learning algorithm 412 produces a model 420 based upon the features 410 and the labels.

[0061] In the prediction engine 404, current data 414 (e.g., action information such as sensor inputs including vibration data, vibration spectra, temperature data or the like) may be input to the parameter determination engine 416. Parameter determination engine 416 may determine the same set of features or a different set of features from the current data 414 as parameter determination engine 408 determined from historical data 406. In some examples, parameter determination engine 416 and 408 are the same engine. Parameter determination engine 416 produces feature vector 418, which is input into the model 420 to generate one or more criteria weightings 422. The training engine 402 may operate in an offline manner to train the model 420. The prediction engine 404, however, may be designed to operate in an online manner. It should be noted that the model 420 may be periodically updated via additional training or user feedback (e.g., an update to a technique or procedure).

[0062] The machine learning algorithm 412 may be selected from among many different potential supervised or unsupervised machine learning algorithms. Examples of supervised learning algorithms include artificial neural networks, Bayesian networks, instance-based learning, support vector machines, decision trees (e.g., Iterative Dichotomiser 3, C4.5, Classification and Regression Tree (CART), Chi-squared Automatic Interaction Detector (CHAD), and the like), random forests, linear classifiers, quadratic classifiers, k-nearest neighbor, linear regression, logistic regression, and hidden Markov models.Examples of unsupervised learning algorithms include expectation-maximization algorithms, vector quantization, and information bottleneck method.Unsupervised models may not have a training engine 402. In an example embodiment, a regression model is used and the model is a vector of coefficients3097.048W01 corresponding to a learned importance for each of the features in the vector of features 410, 418.

[0063] FIG. 5 illustrates a system 500 for monitoring and controlling grain processing including an ethanol production process using vibration data and optionally temperature data. The system 500 can include a process monitoring system 501, which may include the processing circuitry system 504 similar to that discussed previously with regard to FIG. 1. The processing circuitry system 504 can include inputs such as from vibration data 502, temperature data 506, database 518, a display device 510 and a machine learning system 508 such as a machine learning model or system such as the system discussed previously with regard to FIG. 4. The processing circuitry system 504 can control various functions and make various determinations with the aid of and / or based upon the inputs. The processing circuitry system 504 and other components of the process monitoring system 501 such as the vibration data 502, the temperature data 506, database 518 and display device 510 (including user interface 516) may output data to the machine learning system 508. Similarly, the processing circuitry system 504 and other components of the process monitoring system 501 such as the vibration data 502, the temperature data 506 can provide output to the display device 510, and / or a database 518. In an example, the machine learning system 508 may output information to the display device 510 and the database 518 or other components of the system 500. The display device 518 may retrieve information stored in the database 518. The display device 510 may be used to display the user interface 516. In an example, the machine learning system 508 includes a training engine 512 and a real-time feedback engine 514.

[0064] The process monitoring system 501 may be used to perform all or a portion of process monitoring within a plant such as an ethanol production plant. The processing circuitry system 504 may be coupled to memory (e.g., on the process monitoring system 501 or the machine learning system 508). The process monitoring system 501 may be used to record actions taken such as various data inputs including with vibration data 502 such as during fermentation or other portions of the grain processing using other equipment and the accelerometers discussed in regard to FIG. 1. The processing circuitry system 504 may query the database 518 to retrieve information about related3097.048W01 prior processing related data (e.g., parameters such as stream composition and applicable process rates such as fermentation rate, starch and lignocellulose conversion, oil recovery yield, centrifuge performance, evaporator fouling, product composition, product viscosity, etc.). In an example, the information may include at least one result or next action taken after the action (e.g., adjust composition, increase kinetic rate, decrease kinetic rate, take equipment off-line, intervene, make other adjustment). The processing circuitry system 504 may determine a recommended change, such as perform maintenance, shut equipment down, add yeast, add antibiotics, send alert: potential for fouling, intervention needed, etc. The recommended change may be a change performed on any of the various equipment discussed in regard to FIG. 1, for example. Yeast could be added directly to fermenter or via additional propagated yeast from the propagation system.

[0065] The machine learning system 508 may be trained using the related prior test results and sensor data (e.g., from accelerometers, pH, temperature, near-infrared, infrared, etc.) such as to determine process stream composition (not limited to fermentation compounds composition but also other data such as slurry qualities, including changes in suspended solids size distribution, a degree of hydrolysis of carbohydrates to oligomers and sugars, oil recovery yield, centrifuge performance, flowability of the process stream conveyed through the drying drum, concentration of entrained non-condensables in the process stream, starch and lignocellulose conversion, viscosity, product stream composition, etc.) and other process parameters. In an example, the processing circuitry system 504 may submit a plan to the machine learning system 508 to receive feedback preemptively or in real-time or near real-time. In an example, the machine learning system 508 may simulate at least a portion of the grain processing including ethanol production processing to determine changes. The machine learning system 508 may select the change / changes from a plurality of possible changes, such as based on outcome likelihoods (estimated using statistical analysis) of the plurality of possible changes.

[0066] The process monitoring system 501 such as via processing circuitry system 504 can collect and record the vibration data 502 such as from the various accelerometers discussed in reference to FIG. 1. The process monitoring system 501 or the processing circuitry system 504 can query the3097.048W01 database 518 to retrieve information about related prior processing parameters, the information including at least one result or next action taken after the vibration data 502 was provided to the processing circuitry system 504. The system 500 can determine, based upon the information, a next action including at least one of: implementing a change to the process, implementing an intervention such as addition or removal of equipment or compound such as yeast or water to the process, implementing an alert via the display device 510 to facilitate the intervention, etc. The system 500 can implement the next action and collect further data such as vibration data 502 and / or temperature data 506 as to results from the next action. The system 500 can take actions with or without feedback from the machine learning system 508.

[0067] Thus, according to one example the system 500 can be configured to collect data regarding at least one of: a vibration (from accelerometer such as accelerometer 304 A of FIG. 1) of and combination of vibration supplemented with a sound (via the microphone 304A’ of FIG. 1) within the fermentation vessel. The system 500 can include the processing circuitry system 504 including processing circuitry and a memory as discussed in regard to FIG. 1. The memory can include instructions, the instructions, when executed by the processing circuitry, cause the processing circuitry to: electronically communicate with the sensor to receive the data, analyze a frequency spectra of the data, and determine a fermentation rate based on the frequency spectra as analyzed. The system 500 can further include temperature data (e.g., from sensor 306 of FIG. 1 or from another sensor location). Thus, the system 500 can utilize a temperature sensor in the fermentation vessel configured to collect data indicative of a temperature of the process stream within the fermentation vessel. The processing circuitry can be further configured to: electronically communicate with the sensor to receive the data indicative of the temperature, and combine the data indicative of the temperature with the data to improve a prediction accuracy of the fermentation rate determined by the processing circuitry.

[0068] The system 500 can further comprise a machine learning engine (e.g., machine learning system 508 and / or machine learning engine 400 of FIG. 4) in electronic communication with the processing circuitry. The processing circuitry can be configured to train the machine learning engine using historical3097.048W01 vibration data, the data indicative of the temperature and known fermentation rate data gathered by sample testing. Using the machine learning engine, the processing circuitry is configured to determine a fermentation rate of one or more subsequent fermentation processes with the prediction accuracy at or above 90% or at or above 99%.

[0069] According to some examples, the processing circuitry can be configured to: identify a first amplitude range in the frequency spectra, wherein the first amplitude range is between: about 50 nm and about 1500 nm, about 100 nm and about 1000 nm, about 250 nm and about 800 nm, about 350 nm and about 700 nm or about 400 nm and about 600 nm, correlate amplitude changes in the first amplitude range to carbon dioxide bubble production, and determine the fermentation rate based on the amplitude changes in the first amplitude range correlated with the carbon dioxide bubble production. The processing circuitry can be configured to: identify two or more amplitude ranges including at least one amplitude range below and one amplitude range above the first amplitude range, analyze the first amplitude range, analyze the two or more amplitude ranges, and determine the fermentation rate based on analysis of the first amplitude range and analysis of the two or more amplitude ranges.

[0070] In further examples, the processing circuitry can be configured to: determine a depth within the fermentation vessel the carbon dioxide bubble production originates based upon correlation of the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production and identify microorganism location based upon the depth. The processing circuitry can be configured to: determine an average size of the carbon dioxide bubble production and a population of the carbon dioxide bubble production based upon the correlation of the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production. The processing circuitry can be configured to provide an alert if the fermentation rate deviates from an expected fermentation rate. The alert can include recommend an intervention action to add yeast or antibiotics to the fermentation vessel.

[0071] According to yet further examples, the system 500 can be used for monitoring further parameters of the grain processing including the ethanol production process previously discussed in FIG. 1. Thus, the system 500 can be a system for monitoring an ethanol production process, including: a plurality of3097.048W01 accelerometers (example illustrated in FIG. 1) positioned at multiple locations in an ethanol production plant. These multiple locations can be on the same unit or equipment or can be on different units such as including two or more of: a fermentation vessel, a liquefaction vessel, an oil separation unit, a centrifuge, an evaporator, or a drying drum. The system 500 can include processing circuitry and a memory that includes instructions, the instructions, when executed by the processing circuitry, cause the processing circuitry to: electronically communicate to receive vibration data from the plurality of accelerometers during the ethanol production process, analyze frequency spectra of the vibration data, and determine two or more process parameters based on the frequency spectra as analyzed. The processing circuitry can be further configured to: detect changes in viscosity or composition of process streams at different stages of the ethanol production process based on the frequency spectra as analyzed. The two or more process parameters can include correlating the frequency spectra to two or more of: fermentation rate, cell growth rate for a yeast propagation system, starch and lignocellulose conversion, oil recovery yield, centrifuge performance, evaporator fouling, product composition and product viscosity.

[0072] FIG. 7 is a flowchart of a method 600 for monitoring an ethanol fermentation process. The method 600 can include positioning 602 an accelerometer at a fermentation vessel at or above a liquid level of a process stream, collecting 604 vibration data from the accelerometer during fermentation of the process stream, analyzing 606 a frequency spectra of the vibration data, and determining 608 a fermentation rate based on the analyzing the frequency spectra.

[0073] The method 600 can optionally include the accelerometer is positioned within or on an outside surface of the fermentation vessel above the liquid level. The analyzing the frequency spectra can include: identifying a first amplitude range between: about 50 nm and about 1500 nm, about 100 nm and about 1000 nm, about 250 nm and about 800 nm, about 350 nm and about 700 nm or about 400 nm and about 600 nm, and correlating amplitude changes in the first amplitude range to carbon dioxide bubble production within the process stream of the fermentation vessel. The method 600 can optionally further include: identifying two or more amplitude ranges in addition to the first3097.048W01 amplitude range; comparing the first amplitude range to the two or more amplitude ranges, and determining the fermentation rate based on the comparing the first amplitude range with the two or more amplitude ranges. The method 600 can also include: determining a depth within the fermentation the carbon dioxide bubble production originates based upon the correlating the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production, and identifying microorganism location based upon the depth. Additionally, the method 600 can include determining an average size of the carbon dioxide bubble production and a population of the carbon dioxide bubble production based upon the correlating the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production. The method 600 can include providing an alert if the fermentation rate deviates from an expected fermentation rate. The providing the alert includes recommending an intervention action to add yeast or antibiotics to the fermentation vessel.

[0074] The method 600 can further include: collecting temperature data from the fermentation vessel, and combining the temperature data with the vibration data to determine the fermentation rate. The method 600 can also include training a machine learning model to determine the fermentation rate, wherein the training optionally includes: receiving data including composition data, temperature data and vibration data from a plurality of test samples taken during fermentation of a plurality of test batches, comparing the fermentation rate determined by the machine learning model to an actual fermentation rate calculated using the composition data, temperature data and vibration data from the plurality of test samples, evaluating a prediction accuracy of the machine learning model to determine the fermentation rate, receiving a confirmation that the machine learning model accurately determined the fermentation rate and exporting the machine learning model in response to receiving the confirmation. The method 600 can apply the machine learning model to determine a fermentation rate of one or more subsequent fermentation processes, wherein the prediction accuracy of the machine learning model is at or above 90% or at or above 99% to pass the evaluating.

[0075] FIG. 7 is a flowchart of a method 700 for monitoring an ethanol production process. The method 700 can include positioning 702 single or multiple accelerometers at multiple locations in an ethanol production plant,3097.048W01 including two or more of: a fermentation vessel, a liquefaction vessel, an oil separation unit, a centrifuge, an evaporator, and a drying drum; collecting 704 vibration data from the accelerometers during the ethanol production process; analyzing 706 frequency spectra of the vibration data from the two or more of the fermentation vessel, the liquefaction vessel, the oil separation unit, the centrifuge, the evaporator, and the drying drum; and determining 708 two or more parameters of the ethanol production process based on the analyzing the frequency spectra.

[0076] The method 700 can include positioning at least one accelerometer on the side of the liquefaction vessel or piping adjacent to it, analyzing frequency spectra and determining slurry qualities, including changes in suspended solids size distribution and degree of hydrolysis of carbohydrates to oligomers and sugars. For the oil separation unit (e.g., further separation 187 of FIG. 1), the method 700 can include mounting at least one accelerometer to detect vibrations in, within, and out of the centrifuge, analyzing the frequency spectra, and determining oil recovery yield or centrifuge performance. For the drying drum, the method 700 can include mounting at least one accelerometer on or within the drum, analyzing the frequency spectra, and characterizing the flowability of a process stream conveyed through the drying drum. The method 700 can further include using the vibration data collected from the evaporator to identify a potential for fouling and preemptively intervening in the ethanol production process based on the potential for fouling. The method 700 can also include using the vibration data to determine a concentration of entrained noncondensables in a process stream, wherein the concentration is indicative of a separation performance.

[0077] According to yet further examples, the method 700 can include tracking the vibration data over time and determining if maintenance should be performed on various components based on changes in the vibration data. The two or more parameters determined can include fermentation rate, cell growth rate for a yeast propagation system, starch and lignocellulose conversion, oil recovery yield, centrifuge performance, evaporator fouling, product composition and product viscosity. The method 700 can further include training a machine learning model using the vibration data and other known or sensed process parameter data, and using the machine learning model to monitor and determine3097.048W01 the two or more parameters in subsequent production runs, as well as to alert personnel of potential deviations from desired parameters. Additionally, the method 700 can include collecting data from temperature sensors, pH sensors, or near-infrared spectroscopy sensors in the ethanol production plant and combining this data with the vibration data to improve prediction accuracy of the determined parameters. Optionally, the method 700 can use the vibration data to detect changes in viscosity or composition of process streams at different stages of the ethanol production process.

[0078] FIG. 8 illustrates a block diagram of an example machine 800 upon which any one or more of the techniques discussed herein may perform in accordance with some embodiments. This example machine can operate some or all of the apparatus and / or system function discussed herein. In other examples, the example machine 800 is merely one of many such machines utilized. In alternative embodiments, the machine 800 may operate as a standalone device or may be connected (e.g., networked) to other machines. In a networked deployment, the machine 800 may operate in the capacity of a server machine, a client machine, or both in server-client network environments. In an example, the machine 800 may act as a peer machine in peer-to-peer (P2P) (or other distributed) network environment. The machine 800 may be a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a mobile telephone, a web appliance, a network router, switch or bridge, or any machine capable of executing instructions (sequential or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term "machine" shall also be taken to include any collection of machines that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein, such as cloud computing, software as a service (SaaS), other computer cluster configurations.

[0079] Machine (e.g., computer system) 800 may include a hardware processor 802 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), a hardware processor core, or any combination thereof), a main memory 804 and a static memory 806, some or all of which may communicate with each other via an interlink (e.g., bus) 808. The machine 800 may further include a display unit 810, an alphanumeric input device 812 (e.g., a keyboard), and a user3097.048W01 interface (UI) navigation device 814 (e.g., a mouse). In an example, the display unit 810, input device 812 and UI navigation device 814 may be a touch screen display. The machine 800 may additionally include a storage device (e.g., drive unit) 816, a signal generation device 818 (e.g., a speaker), a network interface device 820, and plurality of sensors 821, such as any of those discussed previously (e.g., an IMU, a global positioning system (GPS) sensor, compass, accelerometer, or other sensor). The machine 800 may include an output controller 828, such as a serial (e.g., Universal Serial Bus (USB), parallel, or other wired or wireless (e.g., infrared (IR), near field communication (NFC), etc.) connection to communicate or control one or more peripheral devices (e.g., a printer, card reader, etc.).

[0080] The storage device 816 may include a machine readable medium 822 on which is stored one or more sets of data structures or instructions 824 (e.g., software) embodying or utilized by any one or more of the techniques or functions described herein. The instructions 824 may also reside, completely or at least partially, within the main memory 804, within static memory 806, or within the hardware processor 802 during execution thereof by the machine 800. In an example, one or any combination of the hardware processor 802, the main memory 804, the static memory 806, or the storage device 816 may constitute machine readable media.

[0081] While the machine readable medium 822 is illustrated as a single medium, the term "machine readable medium" may include a single medium or multiple media (e.g., a centralized or distributed database, and / or associated caches and servers) configured to store the one or more instructions 824. The term "machine readable medium" may include any medium that is capable of storing, encoding, or carrying instructions for execution by the machine 800 and that cause the machine 800 to perform any one or more of the techniques of the present disclosure, or that is capable of storing, encoding or carrying data structures used by or associated with such instructions. Non-limiting machine- readable medium examples may include solid-state memories, and optical and magnetic media.

[0082] The instructions 824 may further be transmitted or received over a communications network 826 using a transmission medium via the network interface device 820 utilizing any one of a number of transfer protocols (e.g.,3097.048W01 frame relay, internet protocol (IP), transmission control protocol (TCP), user datagram protocol (UDP), hypertext transfer protocol (HTTP), etc.). Example communication networks may include a local area network (LAN), a wide area network (WAN), a packet data network (e.g., the Internet), mobile telephone networks (e.g., cellular networks), Plain Old Telephone (POTS) networks, and wireless data networks (e.g., Institute of Electrical and Electronics Engineers (IEEE) 802.11 family of standards known as Wi-Fi®, IEEE 802.16 family of standards known as WiMax®), IEEE 802.15.4 family of standards, peer-to-peer (P2P) networks, among others. In an example, the network interface device 820 may include one or more physical jacks (e.g., Ethernet, coaxial, or phonejacks) or one or more antennas to connect to the communications network 826. In an example, the network interface device 820 may include a plurality of antennas to wirelessly communicate using at least one of single-input multiple-output (SIMO), multiple-input multiple-output (MIMO), or multiple-input single-output (MISO) techniques. The term “transmission medium” shall be taken to include any intangible medium that is capable of storing, encoding or carrying instructions for execution by the machine 800, and includes digital or analog communications signals or other intangible medium to facilitate communication of such software.

[0083] Definitions

[0084] Throughout this document, values expressed in a range format should be interpreted in a flexible manner to include not only the numerical values explicitly recited as the limits of the range, but also to include all the individual numerical values or sub-ranges encompassed within that range as if each numerical value and sub-range is explicitly recited. For example, a range of “about 0.1% to about 5%” or “about 0.1% to 5%” should be interpreted to include not just about 0.1% to about 5%, but also the individual values (e.g., 1%, 2%, 3%, and 4%) and the sub-ranges (e.g., 0.1% to 0.5%, 1.1% to 2.2%, 3.3% to 4.4%) within the indicated range. The statement “about X to Y” has the same meaning as “about X to about Y,” unless indicated otherwise. Likewise, the statement “about X, Y, or about Z” has the same meaning as “about X, about Y, or about Z,” unless indicated otherwise.

[0085] In this document, the terms “a,” “an,” or “the” are used to include one or more than one unless the context clearly dictates otherwise. The term3097.048W01“or” is used to refer to a nonexclusive “or” unless otherwise indicated. The statement “at least one of A and B” has the same meaning as “A, B, or A and B.” In addition, it is to be understood that the phraseology or terminology employed herein, and not otherwise defined, is for the purpose of description only and not of limitation. Any use of section headings is intended to aid reading of the document and is not to be interpreted as limiting; information that is relevant to a section heading may occur within or outside of that particular section.

[0086] In the methods described herein, the acts can be carried out in any order without departing from the principles of the disclosure, except when a temporal or operational sequence is explicitly recited. Furthermore, specified acts can be carried out concurrently unless explicit claim language recites that they be carried out separately. For example, a claimed act of doing X and a claimed act of doing Y can be conducted simultaneously within a single operation, and the resulting process will fall within the literal scope of the claimed process.

[0087] The term “about” as used herein can allow for a degree of variability in a value or range, for example, within 10%, within 5%, or within 1% of a stated value or of a stated limit of a range, and includes the exact stated value or range.

[0088] The term “substantially” as used herein refers to a majority of, or mostly, as in at least 80%, 90%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.9%, 99.99%, or at least about 99.999% or more, or 100%.

[0089] Although particular embodiments of the present invention have been described above in detail, it will be understood that this description is merely for purposes of illustration and the above description of the invention is not exhaustive. Specific features of the invention are shown in some drawings and not in others, and this is for convenience only and any feature may be combined with another in accordance with the invention. A number of variations and alternatives will be apparent to one having ordinary skills in the art. Such alternatives and variations are intended to be included within the scope of the claims. Particular features that are presented in dependent claims can be combined and fall within the scope of the invention. The invention also encompasses embodiments as if dependent claims were alternatively written in a multiple dependent claim format with reference to other independent claims.3097.048W01

[0090] The variations are within the spirit of the present invention.Thus, while the invention is susceptible to various modifications and alternative constructions, certain illustrated embodiments thereof are shown in the drawings and have been described above in detail. It should be understood, however, that there is no intention to limit the invention to the specific form or forms disclosed, but on the contrary, the intention is to cover all modifications, alternative constructions, and equivalents falling within the spirit and scope of the invention, as defined in the appended claims.

[0091] Examples

[0092] Various examples of the present disclosure can be better understood by reference to the following Examples which are offered by way of illustration. The present disclosure is not limited to the Examples given herein.

[0093] After doing some background testing to factor out characteristic vibrations from pumps and agitation, a cleaned vibration monitoring is made for several batches where periodic samples are taken to measure ethanol or other desired compounds. FIG. 9 shows a plot of results, with composition of desired compounds plotted over time. The composition was ascertained using nearinfrared spectroscopy sensors. Once sufficient data was collected, a model for the vibration content corresponding to the kinetic rate of desired compound can be determined with statistical modeling. Vibration data was correlated to the ethanol production via near-infrared spectroscopy sensor testing with samples captured every 30 minutes. To estimate the actual bubble size population and size distribution, bench scale testing with known bubble diameters in the media / broth of the target system could be used.

[0094] In the case of ethanol fermentation, reading the CO2 evolution is a direct measurement also of the ethanol content, as both are produced in the same reaction, whereas most submerged sensing approaches attempt to get to a fluid density and then correlate that to kinetic rate but many things besides the ethanol can change the fluid density since enzymes are converting starch to sugar while yeast is converting the sugars to ethanol and CO2. It was determined by the present inventor that parsing the frequency data into spectra bands based upon amplitude as discussed herein allowed for accurate kinetic modeling of a full scale anaerobic ethanol fermentation. This accuracy was then improved when combined with temperature data. FIG. 10 shows a plot where the actual ethanol3097.048W01 production rate (the fermentation rate) of FIG. 9 was plotted against instantaneous predicted rate determined using the methodology discussed herein. This prediction was generated with the aid of a trained machine learning model with the resulting prediction accuracy when the determined ethanol production rate was compared with the actual ethanol production rate exceeding 99%. However, other models including those that do not use machine learning have been created. These models have prediction accuracy exceeding 90% on production runs using: vibration data (including frequency spectra with certain amplitude ranges as discussed herein and using data from the accelerometer placed above the liquid level of the process stream as discussed herein), temperature within the fermenter, duration (minutes) of batch fermentation, duration (minutes) within the fermenter, and other less important process parameters.

[0095] The following exemplary examples are provided, the numbering of which is not to be construed as designating levels of importance. The examples can be combined in any permutation or combination. Portions of examples can be combined with other portions of other examples.

[0096] In one example, the techniques described herein relate to a method for monitoring an ethanol fermentation process, including: positioning an accelerometer at a fermentation vessel at or above a liquid level of a process stream; collecting vibration data from the accelerometer during fermentation of the process stream; analyzing a frequency spectra of the vibration data; and determining a fermentation rate based on the analyzing the frequency spectra.

[0097] In one example, the techniques described herein relate to a method, wherein the accelerometer is positioned within or on an outside surface of the fermentation vessel above the liquid level.

[0098] In one example, the techniques described herein relate to a method, wherein analyzing the frequency spectra includes: identifying a first amplitude range between: about 50 nm and about 1500 nm, about 100 nm and about 1000 nm, about 250 nm and about 800 nm, about 350 nm and about 700 nm or about 400 nm and about 600 nm; and correlating amplitude changes in the first amplitude range to carbon dioxide bubble production within the process stream of the fermentation vessel.3097.048W01

[0099] In one example, the techniques described herein relate to a method, further including: identifying two or more amplitude ranges in addition to the first amplitude range; comparing the first amplitude range to the two or more amplitude ranges; and determining the fermentation rate based on the comparing the first amplitude range with the two or more amplitude ranges.

[0100] In one example, the techniques described herein relate to a method, further including: identifying two or more amplitude ranges including at least one amplitude range below and one amplitude range above the first amplitude range; analyzing the first amplitude range; analyzing the two or more amplitude ranges; and determining the fermentation rate based on the analyzing the first amplitude range and analyzing the two or more amplitude ranges.

[0101] In one example, the techniques described herein relate to a method, further including: determining a depth within the fermentation the carbon dioxide bubble production originates based upon the correlating the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production; and identifying microorganism location based upon the depth.

[0102] In one example, the techniques described herein relate to a method, further including: determining an average size of the carbon dioxide bubble production and a population of the carbon dioxide bubble production based upon the correlating the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production .

[0103] In one example, the techniques described herein relate to a method, further including providing an alert if the fermentation rate deviates from an expected fermentation rate.

[0104] In one example, the techniques described herein relate to a method, wherein the providing the alert includes recommending an intervention action to add yeast or antibiotics to the fermentation vessel.

[0105] In one example, the techniques described herein relate to a method, further including: collecting temperature data from the fermentation vessel; and combining the temperature data with the vibration data to determine the fermentation rate.

[0106] In one example, the techniques described herein relate to a method, further including: training a machine learning model to determine the3097.048W01 fermentation rate, wherein the training includes: receiving data including composition data, temperature data and vibration data from a plurality of test samples taken during fermentation of a plurality of test batches; comparing the fermentation rate determined by the machine learning model to an actual fermentation rate calculated using the composition data, temperature data and vibration data from the plurality of test samples; evaluating a prediction accuracy of the machine learning model to determine the fermentation rate; receiving a confirmation that the machine learning model accurately determined the fermentation rate; and exporting the machine learning model in response to receiving the confirmation.

[0107] In one example, the techniques described herein relate to a method, further including applying the machine learning model to determine a fermentation rate of one or more subsequent fermentation processes.

[0108] In one example, the techniques described herein relate to a method, wherein the prediction accuracy of the machine learning model is at or above 99% to pass the evaluating.

[0109] In one example, the techniques described herein relate to an electronically networked system for monitoring an ethanol fermentation process, including: a fermentation vessel; a sensor configured to positioned above a liquid level of a process stream in a fermentation vessel, wherein the sensor is configured to collect data regarding a vibration of or within the fermentation vessel or the vibration of or within the fermentation vessel in combination with a sound within the fermentation vessel; processing circuitry; and a memory that includes instructions, the instructions, when executed by the processing circuitry, cause the processing circuitry to: electronically communicate with the sensor to receive the data, analyze a frequency spectra of the data, and determine a fermentation rate based on the frequency spectra as analyzed.

[0110] In one example, the techniques described herein relate to an electronically networked system, wherein the sensor is an accelerometer or a combination of a microphone and the accelerometer.

[0111] In one example, the techniques described herein relate to an electronically networked system, further including a temperature sensor in the fermentation vessel configured to collect data indicative of a temperature of the process stream within the fermentation vessel; wherein the processing circuitry3097.048W01 is further configured to: electronically communicate with the sensor to receive the data indicative of the temperature, and combine the data indicative of the temperature with the data to improve a prediction accuracy of the fermentation rate determined by the processing circuitry.

[0112] In one example, the techniques described herein relate to an electronically networked system, further including a machine learning engine in electronic communication with the processing circuitry and wherein the processing circuitry is configured to train the machine learning engine using historical vibration data, the data indicative of the temperature and known fermentation rate data gathered by sample testing.

[0113] In one example, the techniques described herein relate to an electronically networked system, wherein, using the machine learning engine, the processing circuitry is configured to determine a fermentation rate of one or more subsequent fermentation processes with the prediction accuracy at or above 99%.

[0114] In one example, the techniques described herein relate to an electronically networked system, wherein the processing circuitry is configured to: identify a first amplitude range in the frequency spectra, wherein the first amplitude range is between: about 50 nm and about 1500 nm, about 100 nm and about 1000 nm, about 250 nm and about 800 nm, about 350 nm and about 700 nm or about 400 nm and about 600 nm; correlate amplitude changes in the first amplitude range to carbon dioxide bubble production; and determine the fermentation rate based on the amplitude changes in the first amplitude range correlated with the carbon dioxide bubble production.

[0115] In one example, the techniques described herein relate to an electronically networked system, wherein the processing circuitry is configured to: identify two or more amplitude ranges including at least one amplitude range below and one amplitude range above the first amplitude range; analyze the first amplitude range; analyze the two or more amplitude ranges; and determine the fermentation rate based on analysis of the first amplitude range and analysis of the two or more amplitude ranges.

[0116] In one example, the techniques described herein relate to an electronically networked system, wherein the processing circuitry is configured to: determine a depth within the fermentation vessel the carbon dioxide bubble3097.048W01 production originates based upon correlation of the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production; and identify microorganism location based upon the depth.

[0117] In one example, the techniques described herein relate to an electronically networked system, wherein the processing circuitry is configured to: determine an average size of the carbon dioxide bubble production and a population of the carbon dioxide bubble production based upon the correlation of the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production.

[0118] In one example, the techniques described herein relate to an electronically networked system, wherein the processing circuitry is configured to provide an alert if the fermentation rate deviates from an expected fermentation rate based upon historical data, wherein the alert recommends an intervention action to add yeast or antibiotics to the fermentation vessel.

[0119] In one example, the techniques described herein relate to a method for monitoring an ethanol production process, including: positioning accelerometers at multiple locations in an ethanol production plant, including two or more of on or within: a fermentation vessel, a yeast / microorganism propagation vessel, a liquefaction vessel, an oil separation unit, a centrifuge, an evaporator, and a drying drum; collecting vibration data from the accelerometers during the ethanol production process; analyzing frequency spectra of the vibration data from the two or more of the fermentation vessel, the liquefaction vessel, the oil separation unit, the centrifuge, the evaporator, and the drying drum; and determining two or more parameters of the ethanol production process based on the analyzing the frequency spectra.

[0120] In one example, the techniques described herein relate to a method, wherein positioning accelerometers at multiple locations includes mounting one or more accelerometers to a side of the liquefaction vessel or piping adjacent the liquefaction vessel, wherein analyzing frequency spectra includes analyzing frequency spectra from vibration data from at least one accelerometer, and wherein determining one of the two or more parameters includes determining one or both of: slurry qualities, including changes in suspended solids size distribution and a degree of hydrolysis of carbohydrates to oligomers and sugars.3097.048W01

[0121] In one example, the techniques described herein relate to a method, wherein positioning accelerometers at multiple locations includes mounting one or more accelerometers to detect vibrations in, within, and out of the centrifuge, wherein analyzing frequency spectra includes analyzing frequency spectra from vibration data from the at least one accelerometer, and wherein determining one of the two or more parameters includes determining one or combination of: oil recovery yield, solids separation efficiency, or centrifuge performance.

[0122] In one example, the techniques described herein relate to a method, wherein positioning accelerometers at multiple locations includes mounting one or more accelerometers on or within the drying drum, analyzing frequency spectra includes analyzing frequency spectra from vibration data from the at least one accelerometer, and wherein determining one of the two or more parameters includes characterizing a flowability of a process stream conveyed through the drying drum.

[0123] In one example, the techniques described herein relate to a method, further including: using the vibration data collected from the evaporator to identify a potential for or current degree of fouling; and preemptively intervening in the ethanol production process based on the potential for fouling.

[0124] In one example, the techniques described herein relate to a method, further including using the vibration data to determine a concentration of entrained non-condensables in a process stream, wherein the concentration is indicative of a separation performance.

[0125] In one example, the techniques described herein relate to a method, further including: tracking the vibration data over a period of time; and determining from at least a most recently collected of the vibration data if the vibration data has changed in a manner indicative that maintenance be performed one or more of: the fermentation vessel, the liquefaction vessel, the oil separation unit, the evaporator, the drying drum, or one or more shafts coupled thereto.

[0126] In one example, the techniques described herein relate to a method, wherein determining the two or more parameters includes correlating the frequency spectra to two or more of: fermentation rate, cell growth rate for a yeast propagation system, starch and lignocellulose conversion, oil recovery3097.048W01 yield, centrifuge performance, evaporator fouling, product composition and product viscosity.

[0127] In one example, the techniques described herein relate to a method, further including: training a machine learning model using the vibration data and other known or sensed process parameter data; and using the machine learning model to monitor and determine the two or more parameters in subsequent production runs.

[0128] In one example, the techniques described herein relate to a method, further including using the machine learning model to alert personnel of potential deviation of one or more of the two or more parameters from a desired parameter.

[0129] In one example, the techniques described herein relate to a method, further including: collecting data from temperature sensors, pH sensors, or near-infrared spectroscopy sensors in the ethanol production plant; and combining the collected data with the vibration data to improve a prediction accuracy of the determining two or more parameters.

[0130] In one example, the techniques described herein relate to a method, further including using the vibration data to detect at least one of: changes in viscosity or composition of one or more process streams at different stages of the ethanol production process.

[0131] In one example, the techniques described herein relate to a system for monitoring an ethanol production process, including: a plurality of accelerometers positioned at multiple locations in an ethanol production plant, including two or more of: a fermentation vessel, a liquefaction vessel, an oil separation unit, a centrifuge, an evaporator, or a drying drum; processing circuitry; and a memory that includes instructions, the instructions, when executed by the processing circuitry, cause the processing circuitry to: electronically communicate to receive a vibration data from the plurality of accelerometers during the ethanol production process, analyze frequency spectra of the vibration data, and determine two or more process parameters based on the frequency spectra as analyzed.

[0132] In one example, the techniques described herein relate to a system, wherein the processing circuitry is further configured to: detect changes3097.048W01 in viscosity or composition of process streams at different stages of the ethanol production process based on the frequency spectra as analyzed.

[0133] In one example, the techniques described herein relate to a system, wherein the two or more process parameters includes correlating the frequency spectra to two or more of: fermentation rate, cell growth rate for a yeast propagation system, starch and lignocellulose conversion, oil recovery yield, centrifuge performance, evaporator fouling, product composition and product viscosity.

[0134] In one example, the techniques described herein relate to a system or method 1-38, wherein a source of carbohydrate for the fermentation is lignocellulosic, including one or combination of corn stover, corn fiber, sugarcane bagasse, wheat, wheat midlings, soybean, soybean hulls, barley, sorghum, milo, energycane, sugarbeet, softwood, hardwood, agricultural residues or municipal solid waste.

[0135] The terms and expressions that have been employed are used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the examples of the present disclosure. Thus, it should be understood that although the present disclosure has been specifically disclosed by specific examples and optional features, modification and variation of the concepts herein disclosed may be resorted to by those of ordinary skill in the art, and that such modifications and variations are considered to be within the scope of examples of the present disclosure.

Claims

3097.048W01CLAIMSWhat is claimed is:

1. A method for monitoring an ethanol fermentation process, comprising: positioning an accelerometer at a fermentation vessel at or above a liquid level of a process stream; collecting vibration data from the accelerometer during fermentation of the process stream; analyzing a frequency spectra of the vibration data; and determining a fermentation rate based on the analyzing the frequency spectra.

2. The method of claim 1, wherein the accelerometer is positioned within or on an outside surface of the fermentation vessel above the liquid level.

3. The method of any one of claims 1-2, wherein analyzing the frequency spectra comprises: identifying a first amplitude range between: about 50 nm and about 1500 nm, about 100 nm and about 1000 nm, about 250 nm and about 800 nm, about 350 nm and about 700 nm or about 400 nm and about 600 nm; and correlating amplitude changes in the first amplitude range to carbon dioxide bubble production within the process stream of the fermentation vessel.

4. The method of claim 3, further comprising: identifying two or more amplitude ranges in addition to the first amplitude range; comparing the first amplitude range to the two or more amplitude ranges; and determining the fermentation rate based on the comparing the first amplitude range with the two or more amplitude ranges.

5. The method of claim 3, further comprising: identifying two or more amplitude ranges including at least one amplitude range below and one amplitude range above the first amplitude range;3097.048W01 analyzing the first amplitude range; analyzing the two or more amplitude ranges; and determining the fermentation rate based on the analyzing the first amplitude range and analyzing the two or more amplitude ranges.

6. The method of any one of claims 3-5, further comprising: determining a depth within the fermentation the carbon dioxide bubble production originates based upon the correlating the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production; and identifying microorganism location based upon the depth.

7. The method of any one of claims 3-6, further comprising: determining an average size of the carbon dioxide bubble production and a population of the carbon dioxide bubble production based upon the correlating the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production .

8. The method of any one of claims 1-7, further comprising providing an alert if the fermentation rate deviates from an expected fermentation rate.

9. The method of claim 8, wherein the providing the alert includes recommending an intervention action to add yeast or antibiotics to the fermentation vessel.

10. The method of any one of claims 1-9, further comprising: collecting temperature data from the fermentation vessel; and combining the temperature data with the vibration data to determine the fermentation rate.

11. The method of claim 10, further comprising: training a machine learning model to determine the fermentation rate, wherein the training includes:3097.048W01 receiving data including composition data, temperature data and vibration data from a plurality of test samples taken during fermentation of a plurality of test batches; comparing the fermentation rate determined by the machine learning model to an actual fermentation rate calculated using the composition data, temperature data and vibration data from the plurality of test samples; evaluating a prediction accuracy of the machine learning model to determine the fermentation rate; receiving a confirmation that the machine learning model accurately determined the fermentation rate; and exporting the machine learning model in response to receiving the confirmation.

12. The method of claim 11, further comprising applying the machine learning model to determine a fermentation rate of one or more subsequent fermentation processes.

13. The method of claim 12, wherein the prediction accuracy of the machine learning model is at or above 99% to pass the evaluating.

14. An electronically networked system for monitoring an ethanol fermentation process, comprising: a fermentation vessel; a sensor configured to positioned above a liquid level of a process stream in a fermentation vessel, wherein the sensor is configured to collect data regarding a vibration of or within the fermentation vessel or the vibration of or within the fermentation vessel in combination with a sound within the fermentation vessel; processing circuitry; and a memory that includes instructions, the instructions, when executed by the processing circuitry, cause the processing circuitry to: electronically communicate with the sensor to receive the data, analyze a frequency spectra of the data, and3097.048W01 determine a fermentation rate based on the frequency spectra as analyzed.

15. The electronically networked system of claim 14, wherein the sensor is an accelerometer or a combination of a microphone and the accelerometer.

16. The electronically networked system of any one of claims 14-15, further comprising a temperature sensor in the fermentation vessel configured to collect data indicative of a temperature of the process stream within the fermentation vessel; wherein the processing circuitry is further configured to: electronically communicate with the sensor to receive the data indicative of the temperature, and combine the data indicative of the temperature with the data to improve a prediction accuracy of the fermentation rate determined by the processing circuitry.

17. The electronically networked system of claim 16, further comprising a machine learning engine in electronic communication with the processing circuitry and wherein the processing circuitry is configured to train the machine learning engine using historical vibration data, the data indicative of the temperature and known fermentation rate data gathered by sample testing.

18. The electronically networked system of claim 17, wherein, using the machine learning engine, the processing circuitry is configured to determine a fermentation rate of one or more subsequent fermentation processes with the prediction accuracy at or above 99%.

19. The electronically networked system of any one of claims 14-18, wherein the processing circuitry is configured to: identify a first amplitude range in the frequency spectra, wherein the first amplitude range is between: about 50 nm and about 1500 nm, about 100 nm and about 1000 nm, about 250 nm and about 800 nm, about 350 nm and about 700 nm or about 400 nm and about 600 nm;3097.048W01 correlate amplitude changes in the first amplitude range to carbon dioxide bubble production; and determine the fermentation rate based on the amplitude changes in the first amplitude range correlated with the carbon dioxide bubble production.

20. The electronically networked system of claim 19, wherein the processing circuitry is configured to: identify two or more amplitude ranges including at least one amplitude range below and one amplitude range above the first amplitude range; analyze the first amplitude range; analyze the two or more amplitude ranges; and determine the fermentation rate based on analysis of the first amplitude range and analysis of the two or more amplitude ranges.

21. The electronically networked system of any one of claims 19-20, wherein the processing circuitry is configured to: determine a depth within the fermentation vessel the carbon dioxide bubble production originates based upon correlation of the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production; and identify microorganism location based upon the depth.

22. The electronically networked system of any one of claims 19-21, wherein the processing circuitry is configured to: determine an average size of the carbon dioxide bubble production and a population of the carbon dioxide bubble production based upon the correlation of the amplitude changes in at least the first amplitude range to the carbon dioxide bubble production.

23. The electronically networked system of any one of claims 14-22, wherein the processing circuitry is configured to provide an alert if the fermentation rate deviates from an expected fermentation rate, wherein the alert recommends an intervention action to add yeast or antibiotics to the fermentation vessel.3097.048W0124. A method for monitoring an ethanol production process, comprising: positioning accelerometers at multiple locations in an ethanol production plant, including two or more of on or within: a fermentation vessel, a yeast / microorganism propagation vessel, a liquefaction vessel, an oil separation unit, a centrifuge, an evaporator, and a drying drum; collecting vibration data from the accelerometers during the ethanol production process; analyzing frequency spectra of the vibration data from the two or more of the fermentation vessel, the liquefaction vessel, the oil separation unit, the centrifuge, the evaporator, and the drying drum; and determining two or more parameters of the ethanol production process based on the analyzing the frequency spectra.

25. The method of claim 24, wherein positioning accelerometers at multiple locations includes mounting one or more accelerometers to a side of the liquefaction vessel or piping adjacent the liquefaction vessel, wherein analyzing frequency spectra includes analyzing frequency spectra from vibration data from the at least one accelerometer, and wherein determining one of the two or more parameters includes determining one or both of: slurry qualities, including changes in suspended solids size distribution and a degree of hydrolysis of carbohydrates to oligomers and sugars.

26. The method of any one of claims 24-25, wherein positioning accelerometers at multiple locations includes mounting one or more accelerometers to detect vibrations in, within, and out of the centrifuge, wherein analyzing frequency spectra includes analyzing frequency spectra from vibration data from the at least one accelerometer, and wherein determining one of the two or more parameters includes determining one or combination of: oil recovery yield, solids separation efficiency, or centrifuge performance.

27. The method of any one of claims 24-26, wherein positioning accelerometers at multiple locations includes mounting one or more accelerometers on or within the drying drum, analyzing frequency spectra includes analyzing frequency spectra from vibration data from the at least one3097.048W01 accelerometer, and wherein determining one of the two or more parameters includes characterizing a flowability of a process stream conveyed through the drying drum.

28. The method of any one of claims 24-27, further comprising: using the vibration data collected from the evaporator to identify a potential for or current degree of fouling; and preemptively intervening in the ethanol production process based on the potential for fouling.

29. The method of any one of claims 24-28, further comprising using the vibration data to determine a concentration of entrained non-condensables in a process stream, wherein the concentration is indicative of a separation performance.

30. The method of any one of claims 24-29, further comprising: tracking the vibration data over a period of time; and determining from at least a most recently collected of the vibration data if the vibration data has changed in a manner indicative that maintenance be performed one or more of: the fermentation vessel, the liquefaction vessel, the oil separation unit, the evaporator, the drying drum, or one or more shafts coupled thereto.

31. The method of claim 24, wherein determining the two or more parameters comprises correlating the frequency spectra to two or more of: fermentation rate, cell growth rate for a yeast propagation system, starch and lignocellulose conversion, oil recovery yield, centrifuge performance, evaporator fouling, product composition and product viscosity.

32. The method of any one of claims 24-31, further comprising: training a machine learning model using the vibration data and other known or sensed process parameter data; and using the machine learning model to monitor and determine the two or more parameters in subsequent production runs.3097.048W0133. The method of claim 32, further comprising using the machine learning model to alert personnel of potential deviation of one or more of the two or more parameters from a desired parameter.

34. The method of any one of claims 24-33, further comprising: collecting data from temperature sensors, pH sensors, or near-infrared spectroscopy sensors in the ethanol production plant; and combining the collected data with the vibration data to improve a prediction accuracy of the determining two or more parameters.

35. The method of claim 24, further comprising using the vibration data to detect at least one of: changes in viscosity or composition of one or more process streams at different stages of the ethanol production process.

36. A system for monitoring an ethanol production process, comprising: a plurality of accelerometers positioned at multiple locations in an ethanol production plant, including two or more of: a fermentation vessel, a liquefaction vessel, an oil separation unit, a centrifuge, an evaporator, or a drying drum; processing circuitry; and a memory that includes instructions, the instructions, when executed by the processing circuitry, cause the processing circuitry to: electronically communicate to receive a vibration data from the plurality of accelerometers during the ethanol production process, analyze frequency spectra of the vibration data, and determine two or more process parameters based on the frequency spectra as analyzed.

37. The system of claim 36, wherein the processing circuitry is further configured to: detect changes in viscosity or composition of process streams at different stages of the ethanol production process based on the frequency spectra as analyzed.3097.048W0138. The system of claim 36, wherein the two or more process parameters comprises correlating the frequency spectra to two or more of: fermentation rate, cell growth rate for a yeast propagation system, starch and lignocellulose conversion, oil recovery yield, centrifuge performance, evaporator fouling, product composition and product viscosity.

39. The system or method of any one or combination of claims 1-38, wherein a source of carbohydrate for the fermentation is lignocellulosic, including one or combination of corn stover, corn fiber, sugarcane bagasse, wheat, wheat midlings, soybean, soybean hulls, barley, sorghum, milo, energycane, sugarbeet, softwood, hardwood, agricultural residues or municipal solid waste.

Images