Modular model

Abstract

The invention discloses a system and method for spectral modeling. A system may include a data processing system including one or more processors coupled to a memory to receive a request to determine a material or a parameter of the material. The one or more processors may construct a model to determine the material or the parameter based on spectral data of the spectral sensor indicative of an interaction of the material with light in the full wavelength spectral range. The one or more processors may deploy the model to a second data processing system, the model executed by the second data processing system based on the spectral data to determine the material or the parameter of the material.

Description

Cross-referencing of related patent applications

[0001] This application claims the benefit and priority of U.S. Provisional Patent Application No. 63 / 509,270, filed June 20, 2023, the entire contents of which are incorporated herein by reference. introduction

[0002] The production system can produce products. These products can be liquid, gel, or solid materials. Summary of the Invention

[0003] At least one aspect of this disclosure relates to a system. The system may include a data processing system comprising one or more processors coupled to a memory to receive a request to determine a material or parameters thereof. The one or more processors may construct a model to determine the material or parameters based on spectral data from a sensor, such as a spectral sensor, which indicates the interaction of the material with light across the full wavelength spectrum. The one or more processors may deploy the model to a second data processing system, which executes the model based on the spectral data to determine the material or parameters thereof.

[0004] At least one aspect of this disclosure is a method. The method may include receiving a request from a data processing system to determine a material or parameters thereof, the data processing system including one or more processors coupled to a memory. The method may include constructing a model by the data processing system to determine the material or the parameter based on spectral data from a sensor, such as a spectral sensor, the spectral data indicating the interaction of the material with light across the full wavelength spectrum. The method may include deploying the model by the data processing system to a second data processing system, the model being executed by the second data processing system based on the spectral data to determine the material or the parameter thereof.

[0005] At least one aspect of this disclosure relates to one or more storage media storing instructions that, when executed by one or more processors, cause the one or more processors to perform an operation. The operation may include receiving a request to determine a material or parameters of the material. The operation may include constructing a model to determine the material or parameters based on spectral data from a spectral sensor, the spectral data indicating the interaction of the material with light across the full wavelength spectrum. The operation may include deploying the model to a second data processing system, the model being executed by the second data processing system based on the spectral data, to determine the material or the parameters of the material.

[0006] These and other aspects and implementations are discussed in detail below. The foregoing information and the following detailed description include illustrative examples of various aspects and implementations, and provide an overview or framework for understanding the nature and characteristics of the claimed aspects and implementations. The accompanying drawings provide illustrations and further understanding of the various aspects and implementations, and are incorporated into and constitute a part of this specification. The foregoing information and the following detailed description and drawings include illustrative examples and should not be considered limiting. Attached Figure Description

[0007] The accompanying drawings are not intended to be drawn to scale. In the various drawings, similar reference numerals and designations indicate similar elements. For clarity, not every component is labeled in every drawing. In the drawings: Figure 1 This is an example system for determining the material or parameters of a liquid using spectral modeling.

[0008] Figure 2 This is an example system for determining complex parameters from product classification and product concentration.

[0009] Figure 3 It is an example graphical user interface that includes spectral modeling results.

[0010] Figure 4 This is an example method for spectral modeling.

[0011] Figure 5 This is an example data processing system. Detailed Implementation

[0012] The following is a more detailed description of various concepts and implementations of methods, apparatuses, and systems for spectral modeling. The various concepts introduced above and discussed in more detail below can be implemented in any of a variety of ways.

[0013] Systems may require the ability to measure various parameters of fluids within a production system. This production system could produce food, chemicals, oils, cleaning products, hygiene products, pharmaceuticals, etc. In addition to or as a supplement to production systems, the techniques described herein can be applied to heating, ventilation, or air conditioning (HVAC) systems, engines (e.g., engines that incorporate lubricants or hydraulic fluids), ships, or material containers. However, some conventional sensors deployed in production systems may be unable to measure certain parameters. For example, measuring parameters such as chemical oxygen demand (COD) (e.g., the amount of oxygen used to react with all substances in the system) or biochemical oxygen demand (BOD) (e.g., the amount of oxygen used by bacteria to digest organic matter in a system) may require extensive chemical and laboratory testing. Therefore, a grab sample of the liquid may be taken, transferred to a laboratory, and then tested for COD or BOD levels. This testing process can be lengthy, such as three to five days, and may not be suitable for making real-time or near-real-time operational decisions for the production system. For example, to measure BOD levels, it may be necessary to feed the bacteria a sample and monitor them over a five-day period. Therefore, given the required five-day chemical testing, there are no deployable sensors available for manufacturing environments capable of measuring BOD. Complex parameters such as BOD and COD can be important in wastewater from production systems because BOD and COD measurements allow the system to understand the cleanliness of the wastewater and how it needs to be treated before disposal.

[0014] Some optical sensors can be deployed in manufacturing systems to measure how materials, compounds, or molecules interact with light (e.g., reflect, absorb, scatter, etc.). These measurements can be used to detect the material or parameters of a liquid. However, some materials may only interact with light at specific wavelengths. Therefore, optical sensors can only measure the wavelengths of light required to measure a specific material (e.g., one or two wavelengths). For example, a COD sensor can measure a wavelength (e.g., 254 nanometers (nm)) to detect COD, a sugar sensor can measure a wavelength to detect sugar concentration, and an oil sensor can measure a wavelength to detect oil concentration. This can lead to the deployment of many different types of sensors in a manufacturing environment. Furthermore, because a sensor may only measure a single wavelength, different samples or materials may have different optical properties at that wavelength, potentially introducing noise and inaccuracies in the readings at that wavelength.

[0015] To address these and other technical challenges, the system can utilize optical, spectroscopic, or spectral sensors that measure a range of wavelengths and use at least one model to detect materials or parameters in a liquid based on the measured wavelengths. The sensor may include at least one multicolor or monochromatic light source. The spectral sensor can measure how materials interact with light across the entire wavelength spectrum (e.g., reflected light, absorbed light, scattered light, etc.). Because spectral sensors can measure wavelength spectra, a single sensor may be sufficient to detect and measure a variety of different materials or parameters.

[0016] A spectral sensor measures or captures a wavelength spectrum that can provide a fingerprint of a sample, which can be broadband (e.g., multiple wavelengths, a set of wavelengths, a wavelength spectrum). Because different materials or products absorb, reflect, or scatter light in different ways, the system can use the amount of light absorbed, reflected, or scattered across the entire wavelength range to identify materials or products, determine their concentration, and determine whether they meet specifications (e.g., degree of mixing, degree of emulsification, specific color, concentration of specific components, particle size, etc.).

[0017] At least one model or classifier can receive spectral data as input and output product classification, material classification, composition classification, component concentration, material concentration, COD level, or BOD level. The model can be designed to identify materials or parameters based on spectral fingerprints provided by a spectral sensor. The model can directly determine materials or parameters from spectral data.

[0018] Furthermore, this model can indirectly determine materials or parameters. For example, instead of directly testing complex parameters (such as COD, BOD, oil concentration, fat concentration, etc.), the system can receive, store, or maintain data in a database indicating the complex parameters of a specific product and product concentration being measured by a spectral sensor. For instance, laboratory tests can be run on a material or product at full concentration to determine the complex parameters at that full concentration. The database data can be used to construct mappings, models, relationships, or functions that indicate the level of complex parameters of at least one product at various concentrations by scaling the complex parameters at full concentration to other concentration levels.

[0019] For example, the system can use spectral measurements to classify products and determine their concentration levels. The system can use any type of input data, including but not limited to spectral measurements. The system can utilize existing or third-party sensors. For instance, the system can classify products using a first model based on spectral data and determine the product concentration levels using a second model based on the spectral data and classification. Using the classified products and identified concentrations, the system can use this mapping to determine the corresponding complexity parameter levels. These complexity parameter levels can be estimates of the complex parameters of the fluid being measured by the spectral sensor, determined in real-time or near real-time, for example, within hours, minutes, or even less. Such indirect measurement of complex parameters provides a more efficient and rapid means of understanding the complex parameters of fluids compared to using sample grabbing and laboratory testing, especially on production lines where the material or product passing through at one time is of a single type or a few types.

[0020] The system can collect data from the production line or upstream pipelines of a production system and share the collected data or measured materials or parameters with systems, software modules, or sensors that measure downstream materials or parameters (e.g., waste pipeline treatment). Measurements of materials or parameters from upstream pipelines can be used to determine downstream pipeline measurements or parameters by modeling the time it takes for materials to move from upstream to downstream pipelines or by modeling the processes applied to the materials. Therefore, waste streams leaving the production line can be accurately measured or sensed for proactive treatment.

[0021] refer to Figure 1 Along with other accompanying figures, an example system 100 is shown that uses modeling (e.g., spectral modeling) to determine the material or parameters of a liquid. Figure 1The modeling described herein can be applied to spectral sensing, flow rate sensing, temperature sensing, acid or alkali sensing, pH sensing, conductivity sensing, impedance spectral sensing, capacitance sensing, or any other type of sensing. System 100 may include at least one production system 105. Production system 105 may be a system for manufacturing or producing products (e.g., food). Production system 105 may manufacture or produce food, beverages, or any other substance. Production system 105 may manufacture condiments (e.g., ketchup, mayonnaise, vegetable oil, olive oil, mustard), desserts (e.g., ice cream, sorbet, yogurt), foods (e.g., yogurt, cream cheese, soy sauce, powders), beverages (e.g., soft drinks, cola, wine, beer, spirits, energy drinks, vitamins, coffee, purified water, milk), chemicals, ingredients, oils, pharmaceuticals, cleaning products, and hygiene products (e.g., shampoo, toothpaste, soap, mouthwash). The product may be a solid, such as a pharmaceutical product, and gateway 125 may determine whether the pharmaceutical product is genuine or counterfeit. The product may be a gas, and gateway 125 may determine the amount of gas atomization. The product may be a powder. The product can be a liquid, solid, gel, semi-liquid, or any other composition. Production system 105 can receive one or more ingredients, mixed ingredients, emulsified ingredients, cooking ingredients, cooling ingredients, boiling ingredients, or perform various other production steps to produce the product. Production system 105 may include, but is not limited to, mixing equipment, heating equipment, cooling equipment, tanks, reactors, or presses.

[0022] Production system 105 may include at least one pipeline, tank, pool, or fluid holding device 110. Pipeline 110 may be a conduit, pipe, cavity, tank, ditch, or other area for conveying liquids (e.g., products, components used to manufacture products). Device 110 may be a pipeline for moving liquids or any other device for holding liquids. Pipeline 110 may be a pipeline for delivering liquids into or out of production system 105. Pipeline 110 may convey waste products out of production system 105 for disposal. The product or material in pipeline 110 may be at least partially mixed with or suspended in water or non-aqueous materials (e.g., cleaning products, disinfectants, product transfer agents). Sensor 115 (e.g., a spectral sensor) may be disposed in or at least partially immersed in the fluid within pipeline, tank, or fluid holding device 110. For example, sensor 115 may be placed in a tank of production system 105 and at least partially immersed in the liquid in the tank. System 100 can be applied to non-production systems, such as vehicles or devices for conveying, moving, or transporting products. For example, System 100 can be implemented as a truck (e.g., tanker), rail vehicle, transport ship, container, mixer truck, etc., for transporting products. Furthermore, System 100 can be implemented in water treatment plants, cleaning filters, etc.

[0023] System 100 may include at least one sensor 115. Sensor 115 may be an optical or light sensor. Sensor 115 may be a broadband spectral sensor. Sensor 115 may be an optical, spectroscopic, or spectral sensor that measures a range of wavelengths. Sensor 115 may include a light source that generates light, which can be measured by the sensor of spectral sensor 115 after interacting with a material. The light source may be a monochromatic light source. The light source may be a multicolor light source that emits multiple wavelengths to obtain a wavelength spectrum. Sensor 115 can measure how the material interacts with light across the entire wavelength spectrum (e.g., reflected light, absorbed light, scattered light, etc.). The wavelength spectrum may include visible light, ultraviolet (UV), and infrared (IR) wavelengths. The wavelength may be from 395 nm to 955 nm. The wavelength may be less than 395 nm. The wavelength may be greater than 955 nm. The wavelength may be from 200 nm to 1000 nm. The wavelength may be from 1000 nm to 3000 nm or longer. Sensor 115 may include a light source, such as one or more light-emitting diodes (LEDs), which generate light that is reflected, absorbed, or scattered by the liquid in conduit 110 and measured by sensor 115. Sensor 115 can be any type of sensor, such as a camera, flow sensor, impedance sensor for impedance spectroscopy, temperature sensor, pH sensor, conductivity sensor, pressure sensor, gene sequencing device for rapid nanopore genetic RNA or DNA sequencing, viscosity sensor, etc. Impedance spectroscopy sensor 115 may have a frequency range from 1 Hz to 100 Hz, 1 Hz to 1 MHz, 1 Hz to 10 MHz, 100 Hz to 1 MHz, 100 Hz to 10 MHz, 1 MHz to 5 MHz, 1 MHz to 10 MHz, etc. The data received and manipulated by gateway 125 may be process data, such as control data from controller 165, indicating which process in production system 105 is running, starting, or about to start. Gateway 125 can combine measurements from multiple different sensors to infer or determine measurements; for example, combining measurements from first and second sensors to infer a state, such as viscosity.

[0024] Sensor 115 (e.g., a spectral sensor) can generate or acquire data measurements 120 (e.g., spectral measurements) of the fluid in pipeline 110 and provide these data measurements 120 to gateway 125. Data measurements 120 can be at least one signal, data, dataset, at least one data frame, or at least one data packet. Data measurements 120 can indicate the reflectance, absorbance, or scattering level of wavelengths across the full spectral range at a specific resolution. For example, the resolution can be per nanometer, per half-nanometer, or per picometer (pm). Sensor 115 can transmit or send data measurements 120 to gateway 125 via at least one network, cable, or communication medium.

[0025] Gateway 125 may be a device, apparatus, or system that receives data measurements 120 from sensor 115. Gateway 125 may be any computing system, such as a desktop computer or server within a factory connected to a factory network. Gateway 125 may be outside the factory or facility or in the cloud (e.g., server system 170). The computer or server may have software installed to run components of gateway 125, or launch a virtual machine (VM) or container that can run the code or instructions of gateway 125 to run a model of gateway 125. In some implementations, gateway 125 is integrated with sensor 115. For example, sensor 115 may include a computer, microprocessor, or other system capable of running instructions of data processing system 130. Gateway 125 may be a system deployed on-premises within the environment where production system 105 is located (e.g., in the same building or room where production system 105 is located, or in a nearby building or room). Gateway 125 may include at least one data processing system 130. Data processing system 130 may include one or more processors that execute instructions stored on one or more storage devices or storage media. The data processing system 130 can store and execute at least one direct model 135, at least one indirect model 140, and at least one interface manager 145. The direct model 135, the indirect model 140, and the interface manager 145 can be or include scripts, code, executable files, instructions, data structures, data files, modules, circuits, or hardware.

[0026] Gateway 125 may not include user interface 155 or interface manager 145. Gateway 125 may generate or transmit control commands 160 with or without displaying information on user interface 155. Furthermore, graphical user interface 150 may be a software application running on a computer or mobile device; graphical user interface 150 may be a software application, web page, web application, or other interface. In some embodiments, gateway 125 may send text messages, generate application notifications, make phone calls, or send emails instead of displaying graphical user interface 150 on user interface 155, or as a supplement to it. In some embodiments, gateway 125 may generate sounds or illuminate lights to indicate status (e.g., cleaning cycle completed, conversion completed, etc.). For example, gateway 125 may generate notifications about what process to perform next or whether a process is complete. For example, if gateway 125 is deployed in a facility without a controller for sending control commands, gateway 125 may generate notifications for the operator to manually start a process, stop a process, press a button, activate a switch, etc. The model executed by the data processing system 130 can directly output control commands 160, such as Boolean results, control settings, commands to start the next process, and commands to stop the current process. Models 140 or 135 executed by the data processing system 130 can output data or values ​​160, which inform the control, such as the mass value of materials, the amount of contaminants in the cleaning water, the amount of a certain type of material in the system, contaminant concentration, COD or BOD value, etc. Direct models can be used to determine protein concentration and diatomaceous earth concentration.

[0027] Data processing system 130 can execute at least one direct model 135. Direct model 135 can receive data measurements 120 as input and output data to interface manager 145. Direct model 135 can identify or classify product type, material type, product concentration, material concentration, material color, product color, emulsification level, mixing level, components (e.g., sugar, cream, milk, water), component concentration, molecules in a fluid, whether the fluid is pure water, water purity, and detect the state or stage of production system 105 (e.g., transition between different products, cleaning cycle, or start-up cycle). Concentration can be parts per million (PPM), a percentage of water, or a percentage of material. For example, direct model 135 can classify products being manufactured by production system 105 and determine product concentration using at least one direct model 135 as products flow through pipeline 110. Direct model 135 can indicate complex parameters such as BOD or COD.

[0028] Direct model 135 can be a machine learning model, such as some type of neural network, regression, decision tree, ladder logic, or threshold comparison. Data processing system 130 can run one or more direct models 135 together, in parallel, or sequentially. For example, data processing system 130 can run a first model identifying a first material, a second model identifying a second material, and a third model identifying a third material in parallel or sequentially. Direct model 135 can identify one product flowing through pipeline 110 (e.g., alone or in water) or multiple products flowing through pipeline 110 simultaneously. Data processing system 130 can execute models using data from a single sensor 115 or multiple different sensors 115 (e.g., each model can be executed using input measurements from different sensors 115). Multiple sensors 115 can be of the same type or different types, or combinations of types, such as spectra, temperature, conductivity, flow rate, pH, impedance, etc. Models 140 or 135 executed on sensor data can be executed only on spectral sensor data, or on another type of sensor data, or a combination of sensor data types. Data processing system 130 can use data from the sensors to determine results or outputs. The data processing system 130 can execute on process data, control data, and / or data input by the user through the user interface 155. The data processing system 130 can compare upstream data from the production system 105 with downstream data from various sensors to determine whether the product is within or exceeds specifications.

[0029] For example, multiple sensor data 120 can be used in models 135 or 140 for cleaning optimization. For instance, spectral, temperature, flow rate, conductivity, process, and control data can be fed into models 135 or 140 to determine whether a cleaning step is complete or ready to proceed to the next step in the cleaning process (e.g., from a water cleaning step to a chemical cleaning step, and then to another water cleaning step). As another example, multiple sensor data types 120 can be used for initiation optimization, where spectral, pressure, flow rate, process, and control data can be fed into models 135 or 140 to determine whether the product is ready for use.

[0030] Since both model 135 and indirect modeling system 140 can operate on the same data measurement 120, it may only be necessary to deploy sensor 115 to measure or estimate multiple different materials or parameters of the liquid in pipeline 110. Furthermore, if production system 105 switches from producing a first product to producing a second product, data processing system 130 can execute model 135 for any indications, detections, or determinations received by data processing system 130 regarding the product produced by production system 105. Data processing system 130 can execute model 135 to determine when a process (e.g., fermentation) is complete. Data processing system 130 can generate a control command 160 indicating that fermentation is complete and that production system 105 should stop the fermentation process. If production system 105 sets fermentation to a predefined time length (e.g., based on a timer), fermentation may run longer or shorter than necessary. These techniques can be applied to processes such as mixing, stirring, reacting, and ripening.

[0031] For example, the controller 165 of production system 105 can run a timer to indicate the set time for fermentation. However, in the middle or during the set time, if the data processing system 130 determines that fermentation has been completed based on data measurement 120, the data processing system 130 can generate a control command 160 to cause the controller 165 to prematurely exit the fermentation process. This can reduce production time, ensure product quality assurance / quality control (QA / QC), and reduce the power or energy requirements of production system 105.

[0032] Models 135 or 140 can detect the presence of microorganisms or predict the growth of microorganisms in production system 105. For example, models 135 or 140 can predict fouling that may result from biological growth, or predict the growth or presence of biofilms. Models 135 or 140 can predict microbial growth and identify microbial contamination. Models 135 or 140 can also detect or predict leaks, filtration problems, or control problems in production system 105. For example, models 135 or 140 can be implemented for QC or QA processes, such as determining whether a batch of products is being manufactured correctly. For example, models 135 or 140 can detect the level of ferric phosphate in coatings or other materials (e.g., spray materials) to determine how much ferric phosphate is in the material. Similarly, models 135 or 140 can determine the level or activity of active ingredients, such as the caramelization of sugars, to determine the degree of heat treatment of the product. Models 135 or 140 can, for example, determine the level or quality of inactive ingredients. Model 135 or 140 can predict or identify the degree of digestion, cooking, or mixing of a material, such as the degree of roasting of soybeans. Model 135 or 140 can infer, predict, or determine the concentration of a material, such as chemical concentration, acid concentration, surfactant concentration, disinfectant concentration, or caustic soda concentration.

[0033] The direct model 135 can identify products, materials, or molecules in the liquid of pipeline 110 by analyzing a subset of wavelengths from the data measurements 120. For example, different product materials or molecules may react with certain wavelengths of light but not others. For instance, milk fat, milk fat concentration, dairy fat content, or dairy fat concentration may reflect light more strongly in the infrared range and less strongly in other wavelengths. In this regard, filters can filter out or remove wavelengths from the data measurements 120, so that the direct model 135 receives only wavelengths that are relevant to or important for identifying milk fat or milk fat concentration. In some embodiments, unimportant wavelengths may be discarded, ignored, or treated with low weight by the direct model 135.

[0034] Data processing system 130 can execute at least one indirect model 140 or a set of models to identify complex parameters of the liquid in pipeline 110. Indirect model system 140 may not directly predict or classify complex parameters (e.g., BOD, COD), but can instead use a mapping between directly predictable parameters and complex parameters. For example, the mapping can be constructed based on chemical tests of complex parameters of known products at known concentration levels. Indirect model system 140 can execute at least one model to identify the products in pipeline 110 and the concentrations of those products. Indirect model system 140 can use the mapping between product type, product concentration, and complex parameters to determine estimates of the complex parameters. For example, indirect model system 140 can execute a first model to classify products based on data measurement 120. Indirect model system 140 can execute a second model to identify product concentrations based on data measurement 120 and the classified products. Using the identified product type and product concentration, relationships, functions, or mappings can estimate the values ​​of the complex parameters. This mapping can scale known complex parameter levels of a particular product at known concentration levels to other concentration levels of that product.

[0035] Indirect model 140 can be in the same system, processing unit, computing unit, server, or network as direct model 135, or it can be in two separate systems and networks. Output from direct model 135 can be fed into and used by indirect model 140, or vice versa, or they can be independent of each other.

[0036] The data processing system 130 can execute the direct model 135 or the indirect model system 140 when receiving data measurement values ​​120 or when the production system 105 is producing products. In this respect, the delay time between the measurement of sensor 115 and the output of the direct model 135 or the indirect model system 140 can be real-time or sufficiently low (e.g., less than ten minutes, less than one minute, less than one second, less than one millisecond, less than half a millisecond) such that the output of the direct model 135 is suitable for real-time control of the production system 105. In this respect, the determination results of the data processing system 130 can be used to generate control commands 160 to control the production system 105. For example, the data processing system 130 can generate control commands 160 to control the production system 105 such that the product achieves a predefined color, a specific degree of emulsification or mixing, a predefined concentration, a predefined component concentration, or the presence of specific molecules at a specific concentration, based on the determination results using the direct model 135 or the indirect model 140. This control can effectively operate the production system 105, keeping the product within specific specifications or meeting standards, thereby enabling continuous product production. Furthermore, this control can avoid waste or the use of timers to guess or estimate when a product is ready, when a product should be stopped and discarded, etc. Control commands 160 can be commands for opening or closing valves, opening valves by a certain amount, changing flow paths, closing valves by a certain amount, running motors at a specific speed, stopping or starting motors, starting or stopping mixers, temperature setpoints, controlling pumps, and starting or stopping processes.

[0037] Data processing system 130 can receive data and use it to determine which models 135 to execute. For example, data received from controller 165, time received from a clock, data received from user interface 155, data received from sensor 115 (e.g., a spectral sensor), or data from any other sensor can be used to determine which models 135 to execute, which sensors 115, data inputs 120, or data types the models 135 use to generate their outputs, and how the data measurements 120 are processed. In some embodiments, the output of direct model 135 can be used to select which indirect model 140 to execute. For example, if direct model 135 detects a specific type of material, indirect model 140, which is configured or trained to execute in the presence of that material in production system 105, can be selected to execute. For example, data can indicate that production system 105 is producing a specific product or is in a specific cycle. For example, a schedule can indicate a transition or production of a specific material at a planned time. In response to the current time reaching the planned time, data processing system 130 can cause direct model 135, specifically designed for detecting the material or determining its concentration, to be executed. In this regard, all models 135 can be executed simultaneously, or some models 135 can be executed without executing others. For example, during production, the product quality model 135 can be executed. For example, during rinsing, the rinsing optimization model 135 can be run. Furthermore, during startup, the startup optimization model 135 can be run.

[0038] Control command 160 can be generated by data processing system 130 to initiate a cleaning cycle. For example, based on measured contamination, data processing system 130 can determine when cleaning is necessary, predict when future cleaning will be necessary, determine what chemicals or chemical concentrations are needed for cleaning, and determine how long the cleaning cycle should last. Data processing system 130 can transmit control command 160 to cause production system 105 to perform cleaning.

[0039] Production system 105 may include at least one controller 165. Controller 165 may be a programmable logic controller (PLC), microprocessor, computer, inverter, distributed control system (DCS), building management system (BMS), supervisory control and data acquisition (SCADA) system, or any other device capable of controlling actuators in production system 105 to control product production. For example, controller 165 may open or close valves based on control command 160. Controller 165 may start or stop fans, or control fan speed, based on control command 160. Controller 165 may control heating equipment or cooling systems to achieve a certain temperature based on control command 160. Controller 165 may start or stop mixers by operating motors based on control command 160.

[0040] Data processing system 130 can provide control command 160 as input to model 135 or 140. For example, data processing system 130 can provide model 135 or 140 with an input combining data measurement value 120 and control command 160. In this respect, data processing system 130 can predict the results produced by different control commands 160. For example, data processing system 130 can change the value of control command 160 input to model 135 or 140 and predict BOD, COD, microbial growth, efficiency, product yield, material properties, etc., based on the environment in production system 105 created by control command 160, or based on the operations performed by production system 105 based on control command 160. Data processing system 130 can run optimization algorithms (e.g., linear programming algorithms) to identify control command 160 that optimizes parameters, such as maximizing product yield, minimizing microbial growth, maximizing fermentation, etc.

[0041] Interface manager 145 can generate a user interface or graphical user interface 150 based on the determination results of direct model 135 or indirect model 140. Graphical user interface 150 may include graphics, text, graphs, charts, numerical values, interactive elements, icons, buttons, and switches. Graphical user interface 150 may include values ​​or graphs of BOD, COD, product classification, product strength, ingredient presence, ingredient strength, or emulsification. Graphical user interface 150 may also include control data, such as the status of production system 105, timers running on production system 105, or settings of controller 165 of production system 105. Interface manager 145 can display graphical user interface 150 on user interface 155 (e.g., a display, interface, input device, or screen).

[0042] System 100 may include at least one server system 170. Server system 170 may be removed from or detached from gateway 125. For example, server system 170 may be located outside the environment or far from production system 105. However, the functionality of server system 170 may be performed by gateway 125, and therefore may be performed locally in the environment of production system 105, such as through local deployment. Server system 170 may be or include at least one data processing system. Server system 170 may implement at least one model developer 175 or at least one model deployer 180. Model developer 175 or model deployer 180 may be instructions, code, executable files, scripts, hardware circuits, or logic circuits. Server system 170 may receive requests to build model 185 and may build and deploy model 185 locally without deploying model 185 to gateway 125, or in any other way than deploying model 185 to gateway 125. Gateway 125 can build and deploy model 185 locally. For example, gateway 125 may include training data 190, model developer 175, or model deployer 180.

[0043] Model developer 175 can generate at least one model 185. Model 185 can be deployed as a direct model 135 or an indirect model 140. Model developer 175 can generate model 185 based on training data 190. Training data 190 can be or include data measurements 120 measured from sensor 115 or multiple identical or different types of sensors 115, product indications 195 received from scheduling system 197 (e.g., a scheduling server or system storing production plans for different products in different production systems 105), or laboratory data 187 received from laboratory system 183. Data measurements 120 can be timestamped by sensor 115 or gateway 125 and stored as training data 190. Product indications 195 can be stored as indications of product types being produced at different timestamps or of components flowing through pipeline 110 at different timestamps. In this regard, the model developer 185 can use the timestamp of the data measurement value 120 and the timestamp of the product indication 195 to associate the data measurement value 120 with the product indication 195, for example, to determine a set of data measurement values ​​120 for a specific product at a specific concentration.

[0044] In addition, training data 190 may include laboratory data 187. Laboratory data 187 may be laboratory data indicating product type, product specifications, product color, ingredients, ingredient concentration, BOD level, and COD level. For example, a technician may take samples of fluid from pipeline 110 at different times, timestamp the samples, and bring the fluid to the laboratory for chemical testing to determine the material, parameters, or complex parameters of the tested fluid. Laboratory data 187 may indicate the results of the chemical tests, as well as the timestamp of when the samples were taken. The laboratory system or server, namely laboratory system 183, may transmit laboratory data 187 to server system 170. Model developer 175 may associate laboratory data 187 with data measurements 120 and product indications 195 based on the timestamps. In some examples, an operator may press a button or otherwise input commands or values ​​to gateway 125 in lieu of or as a supplement to laboratory data 187. For example, an operator may manually enter data to train models 135 and 140, or execute models 135 or 140. For example, manually entered data can be used to form at least a portion of training data 190. For example, manually entered data can be fed as input into model 135 or 140. Gateway 125 can transmit manually entered data to server system 170, or a client device can receive manually entered data and transmit it to server system 170.

[0045] Model developer 175 can receive requests from client devices or user devices (e.g., via user interface 155, via smartphone, via laptop, via tablet, etc.) to identify specific materials or parameters of the fluid in pipeline 110. This request can be received by server system 170 via at least one communication network. In response to receiving the request, model developer 175 can design, construct, generate, build, or train model 185 according to the request. Model 185 can be constructed or trained based on training data 190. For example, based on training data 190, model developer 175 can train a direct model 135 or a model for indirect model system 140. Requests received from client devices can indicate a target to be estimated. Model developer 175 can build a model to determine that target.

[0046] For example, model developer 175 can construct, build, or train model 185 to identify or classify products flowing through pipeline 110. For instance, model developer 175 can associate product indications 195 with data measurements 120 and generate a model that predicts product type based on a training dataset of at least one product and the corresponding data measurements 120 acquired as the at least one product flows through pipeline 110. Model 185 can be generated to identify a single product or to identify a product from a plurality of possible products. Model developer 175 can train model 185 based on the training dataset to determine products based on the data measurements 120 as input (e.g., outputting values ​​indicating product type or product presence).

[0047] For example, model developer 175 can construct, build, or train model 185 to determine fluid parameters based on laboratory data 187 and data measurements 120. For example, model developer 175 can generate a training dataset for predicting parameters (e.g., BOD, COD) measured in laboratory data 187. Model developer 175 can construct the training dataset by associating timestamps of laboratory data 187 with timestamps of data measurements 120 to determine the laboratory measurement parameters corresponding to data measurements 120 acquired at a given time. Model developer 175 can train model 185 based on the training dataset to determine parameters (e.g., output the value of the parameter) based on the data measurements 120 as input.

[0048] Furthermore, model developer 175 can generate a mapping. This mapping can be a data structure such as a table, relation, function, or equation. The mapping can indicate the level of complex parameters (e.g., BOD or COD) based on product type and product concentration. For example, laboratory data 187 can indicate complex parameters of a product sample, where the sample has a known product type and a known product concentration. Product type and product concentration can be included in laboratory data 187. Model developer 175 can generate or construct this mapping using the known product type, known product concentration, and measured complex parameters. Model deployer 180 can deploy, transmit, send, or transmit this mapping to gateway 125 or data processing system 130.

[0049] In response to model developer 175 building a new model 185, model deployer 180 can transmit, send, or deliver the model 185 to gateway 125 for deployment and operation on gateway 125. Data processing system 130 can receive model 185 and deploy it on data processing system 130 (e.g., unpack the model, save the model, store the model). In response to model deployment, data processing system 130 can begin running the new model 185 based on data measurement 120. In this respect, direct, indirect, and mapped models can be built and deployed to the system to allow the system to implement models modularly. By allowing models to be built and deployed, various different types of models can be developed to implement various different use cases, all using the same data measurement 120 from sensor 115 (or the same spectral fingerprint read from the sensor in real time). For example, for new or different deployments, or as a need for new or different implementations arises, a request can be made to server system 170 to allow models to be built and deployed to gateway 115 in a modular manner, thereby flexibly adapting system 100. One or more modular models (e.g., direct model 135, indirect model 140, mapping model) can be deployed at once or over time. Modular models can be implemented in real time or near real time and read data measurements 120 from the same sensor 115 in different ways to implement different use cases or determine different values ​​or parameters.

[0050] Model 185 can be a containerized model, algorithm, software application, or software module. Containers can be deployed to run on server system 170, gateway 125, or production system 105. A container can include all software dependencies required to run on the system, such as a DOCKER container, a KUBERNETES container, or any other type of standard software unit package, including a package containing all its own dependencies. Furthermore, in some implementations, model 185 can be or is configured to run on a virtual machine.

[0051] Gateway 125 can use indirect modeling system 140 to determine complex parameters through indirect mapping, such as BOD, COD, phosphorus, nitrogen, ammonia, sugar, sugar concentration, fat, fat concentration, oil, grease, active ingredient, inactive ingredient, protein, yeast, enzyme, antibody, active pharmaceutical ingredient (API), enzyme identification, enzyme concentration, enzyme reaction rate or reaction state, antibody, acid, microorganism, virus, metal, ion, salt, lead, sodium, calcium or magnesium.

[0052] Gateway 125 can determine the concentration of aluminum and other metals, material compounds, composition, chemical properties, or characteristics used to clean metal products (e.g., aluminum packaging). Aluminum concentration and other specific metals can be detected using direct model 135 based on wavelengths captureable in the spectral fingerprint. Gateway 125 can use direct model 135 to determine particle size, particle density, particle quantity, enzymatic reactions, fermentation, or reaction optimization. For example, model 135 can predict or identify stages in the brewing or production of soy sauce, monosodium glutamate (MSG), plastics, etc. Furthermore, the measured spectral fingerprint may change once the reaction is complete, indicating that the reaction has finished. For example, once concrete or other materials have hardened or reached a certain degree of hardening, the spectral fingerprint may become identifiable. Gateway 125 can use direct model 135 to detect the spectral fingerprint indicating that the reaction or hardening has been completed, or the reaction rate, or the hardening rate. Model 135 or 140 can use viscosity measurements or other data measurements 120 to determine the effectiveness of cleaning equipment, parts, or components (e.g., automotive equipment or CNC milled parts).

[0053] Gateway 125 can determine whether a product (e.g., a pharmaceutical product) is genuine or counterfeit. For example, a pharmaceutical product may have a defined spectral fingerprint, and deviations from that fingerprint may indicate that the product is counterfeit. Gateway 125 can use direct model 135 to determine whether a product (e.g., a pharmaceutical product or ingredient) is genuine or counterfeit. Gateway 125 can use direct model 135 to determine two or three phases, such as how much of a material is in a gas phase, liquid phase, or particulate phase. For example, different spectral fingerprints can be identified by direct model 135, which identifies different phases. Gateway 125 can use direct model 135 to determine whether centrifugation (separation of two or more materials) is complete or to determine the state or rate of centrifugation. Furthermore, gateway 125 can use direct model 135 to determine the degree of homogeneity of a fluid or product, such as the level of homogeneity. Additionally, gateway 125 can use direct model 135 to determine particle size, density, or viscosity.

[0054] Gateway 125 can use Direct Model 135 to determine whether a cleaning cycle has been completed. Gateway 125 can determine contamination levels, such as aluminum concentration or contamination detection in another material (e.g., oil), based on Direct Model 135. Gateway 125 can use Direct Model 135 for conversion optimization to detect when a conversion is complete, for example, switching from producing one product (e.g., vanilla ice cream) to producing another product (e.g., chocolate ice cream), to reduce product loss and production downtime. Gateway 125 can use Direct Model 135 to determine turbidity or haze levels. Gateway 125 can use Direct Model 135 to determine biomass or cell counts. For example, different cell counts may absorb light differently, therefore, biomass or cell counts can be detected by spectral measurements. Furthermore, Gateway 125 can use Direct Model 135 to determine dehydration, water content, maturation, cell decay, disease, and allergens.

[0055] Gateway 125 can use Direct Model 135 for quality assurance or quality control. For example, Gateway 125 can use Direct Model 135 to detect quality problems in the manufactured product. Direct Model 135 can indicate the amount of active ingredient to be added to the product to correct any quality control issues. Direct Model 135 can indicate mixing ratios and mixing amounts to determine if materials are adequately mixed at the correct concentration. Gateway 125 can send control commands 160 to the controller to alert the user or proactively control the amount of ingredients added to the product to correct quality control problems. Gateway 125 can use Direct Model 135 to determine the color, haze, protein levels, enzymes, antibody concentrations, active yeast, cell counts, sugar concentrations, protein, or cleaning optimization of the pharmaceutical product.

[0056] For example, gateway 125 can implement mixing optimization, where measurements of a mixture of multiple streams are determined. Production system 105 may include a complex valve matrix with multiple input and output streams, such as materials from a first holding tank, a second holding tank, a third holding tank, etc., mixed together according to the degree of opening or closing of valves in the valve matrix. Model 135 or 140 can acquire data from sensor 120 (such as spectral and flow rate sensors 120) to determine the mixing proportions from the individual output streams. Model 135 or 140 can use the determined mixing proportions (e.g., a mixture between a first material, a second material, a third material, etc.) to output control decisions to optimize or bring the mixing proportions to or towards a desired level. These techniques can also be applied to industries such as beer, beverages, detergents, and liquid soap.

[0057] refer to Figure 2Along with other accompanying figures, an indirect modeling system 140 is illustrated to determine complex parameters from product classification and product concentration. The indirect modeling system 140 can receive data measurements 120 from a spectral sensor and provide the data measurements 120 to at least one product model 205. The product model 205 can be a model such as a machine learning model, for example, some type of neural network, regression, decision tree, ladder logic, or threshold comparison. The product model 205 can determine product classification 210 from the data measurements 120, for example, identifying the type of product (e.g., chocolate ice cream, vanilla ice cream, mayonnaise). The indirect modeling system 140 can apply one or more product models 205 to the data measurements 120. A single product model 205 can classify products from a variety of possible product types. The product model 205 can identify the presence of a single product, for example, by providing a binary output indicating the presence of a specific product in pipeline 110.

[0058] The indirect modeling system 140 may include at least one product concentration model 215. The product concentration model 215 may determine the concentration level 220 of the product based on product classification 210 and data measurements 120. The indirect modeling system 140 may select one product concentration model 215 from a set of product concentration models 215 for the product in pipeline 110 based on product classification 210. For example, if product model 205 indicates that the product is chocolate ice cream, the indirect modeling system 140 may select the model 215 indicating the concentration of chocolate ice cream from a set of models including chocolate ice cream concentration models, vanilla ice cream concentration models, and strawberry ice cream concentration models. In some embodiments, a single model 215 may identify the concentration of multiple different product types.

[0059] The indirect modeling system 140 may include at least one mapping 225. In some embodiments, the indirect modeling system 140 may be implemented without mapping 225. Component 225 may be a mapping, model, relation, or function that indicates the level of a complexity parameter 230 of at least one product at various concentrations of that product. For example, based on product classification 210 and concentration 220, mapping 225 may indicate the value or level of complexity parameter 230. The indirect modeling system 140 may include one mapping 225 for each of a plurality of different product types. The indirect modeling system 140 may include one mapping 225 (or multiple mappings 225, each mapping a different product type of complexity parameter 230) that maps a plurality of different product types to complexity parameter 230. Complex parameter 230 may be a parameter that cannot be directly or easily measured without chemical or laboratory testing. For example, complex parameter 230 may be BOD, COD, fat, oil or fat (FOGs), nitrogen, lipids, proteins, active pharmaceutical ingredients, enzymes, or antibodies.

[0060] For example, laboratory data 187 could indicate a BOD level of 500,000 PPM for a chocolate ice cream product of pure concentration (e.g., 100% chocolate ice cream). BOD levels can also be measured in milligrams per liter (mg / L). Mapping 225 can be generated for various concentrations 220 to scale (e.g., linearly or non-linearly) the BOD level of the pure concentration to the BOD level of concentration 220. For example, if product concentration model 215 determines that the concentration level 220 of the chocolate ice cream product is half product and half water (e.g., 50% water), the corresponding BOD estimate for that ice cream product can be determined via mapping 225.

[0061] refer to Figure 3 The accompanying figures, along with other figures, show an example graphical user interface 150 that includes spectral modeling results. Figure 3 The graphical user interface 150 can indicate the use of data measurements 120 to estimate the BOD of the water heading to the wastewater treatment plant during the washing process. The graphical user interface 150 can include trends in supply sensors, return sensors, the flow rate of the supply sensors, the flow rate of the return sensors, or any other type of sensor deployed in the production system 105. For example, the graphical user interface 150 can include trends in both the supply and return sensors. The graphical user interface 150 can also include estimated BOD levels, such as BOD density or BOD volume. Estimates can be for one minute, one day, one week, one month, or one year. For example, the graphical user interface 150 can indicate a daily average BOD estimate of 0.12 pounds per gallon (lbs / gallon). The graphical user interface 150 can indicate a daily BOD volume estimate of 206 pounds.

[0062] The graphical user interface 150 can also indicate time savings, energy savings, product loss savings, etc., when operating the production system 105 using control commands 160 determined by data measurements 120, compared to timers or conventional non-data-driven control methods. The graphical user interface 150 can indicate a daily release time of sixteen minutes. The graphical user interface 150 can indicate a daily water saving of 527 gallons. The graphical user interface 150 can indicate a daily product loss of twenty-nine gallons. Savings can indicate discharge flushing time savings (e.g., four minutes and twenty-six seconds), flushing recirculation time savings (e.g., one minute and three seconds), water savings during flushing (e.g., fifty-six gallons), and water savings during recirculation (e.g., eighty-four gallons). Savings can include total time savings, total water savings, and total product savings. Furthermore, the graphical user interface 150 can indicate the operating duration (e.g., forty-eight minutes), the type of tank in the production system 105 (e.g., tank 08), and the type of product being produced by the production system 105 (e.g., chocolate). The graphical user interface 150 can indicate a total time saving of five minutes and twenty-eight seconds. The graphical user interface 150 can indicate a total water saving of 139 gallons. The graphical user interface 150 can indicate a total product loss of six gallons.

[0063] Gateway 125 can execute a direct model 135 to determine product or product concentration using spectral measurements 120 from spectral sensor 115. Utilizing product identification and concentration, an indirect model 140 can be executed using the flow rate from flow sensor 115, which determines how much identified product is flowing through pipeline 110. In this respect, direct model 135 can be executed on a first measurement 120 of a first type from the first sensor 115, while indirect model 140 can be executed on a second measurement 120 of a second type from the second sensor 115. Indirect model 140 can output product loss, water consumption, and energy consumption. This data can be displayed within a graphical user interface 150 by interface manager 145. Direct model 135 can be deployed locally on the factory floor to determine product type and concentration using spectral measurements 120, while indirect model 140 can be deployed off-premises on server system 170 and executed using flow measurements 120.

[0064] refer to Figure 4Along with other accompanying figures, an example method 400 for spectral modeling is illustrated. At least a portion of method 400 may be performed by a gateway 125, a data processing system 130, a server system 170, a controller 165, a sensor 115, a laboratory system 183, and a scheduling system 197. Method 400 may include a step 405 of receiving spectral measurements. Method 400 may include a step 410 of determining materials or parameters. Method 400 may include a step 415 of generating output data.

[0065] In step 405, method 400 may include receiving data measurements 120, such as spectral measurements, by data processing system 130. Data processing system 130 may receive data from sensor 115 (e.g., a spectral sensor measuring fluid in pipeline 110 of production system 105). The spectral or other measurements measured by sensor 115 may indicate the fluid's reflectance, absorptivity, transmittance, fluorescence, emissivity, or scattering rate across a broad spectrum of light wavelengths. Data processing system 130 may receive data messages, data packets, data frames, or other data from sensor 115.

[0066] In step 410, method 400 may include determining materials or parameters by data processing system 130. Data processing system 130 may use data measurements 120 to determine materials or parameters of the liquid. For example, data processing system 130 may use data measurements 120 to determine the product in pipeline 110, the concentration of the product in pipeline 110, the components in pipeline 110, BOD, and COD. Data processing system 130 may execute a direct model 135 based on data measurements 120. Data processing system 130 may execute an indirect model system 140 based on data measurements 120.

[0067] In step 415, method 400 may include generating output data by data processing system 130. Data processing system 130 may generate output data based on materials or parameters determined in step 410. For example, data processing system 130 may generate graphical user interface 150 based on the determined materials or parameters. For example, data processing system 130 may generate control commands 160 based on the determination results of direct model 135 or indirect model system 140. Gateway 125 may transmit control commands 160 to controller 165 to operate production system 105 and reduce production time, increase efficiency, reduce waste, and reduce wastewater. Time and resource savings may be included in graphical user interface 150.

[0068] Figure 5An example block diagram of a data processing system 130 is depicted. The data processing system 130 may include or be used to implement a data processing system or components thereof (e.g., gateway 125, server system 170, controller 165, laboratory system 183, scheduling system 197). The data processing system 130 may include at least one bus 525 or other communication components for transmitting information, and at least one processor 530 or processing circuitry coupled to the bus 525 for processing information. The data processing system 130 may include one or more processors 530 or processing circuitry coupled to the bus 525 for processing information. The data processing system 130 may include at least one main memory 510, such as random access memory (RAM) or other dynamic storage device, coupled to the bus 525 for storing information and instructions to be executed by the processor 530. The main memory 510 can be used to store information during the execution of instructions by the processor 530. The data processing system 130 may also include at least one read-only memory (ROM) 515 or other static storage device coupled to the bus 525 for storing static information and instructions for the processor 530. Storage device 520, such as a solid-state device, disk or optical disk, can be coupled to bus 525 to continuously store information and instructions.

[0069] Data processing system 130 can be coupled to display 155, such as a liquid crystal display or an active matrix display, via bus 525. Display 155 can display information to users (e.g., operators, technicians, or users of production system 105). Input device 155, such as a keyboard or voice interface, can be coupled to bus 525 for transmitting information and commands to processor 530. Input device 155 may include a touchscreen of display 155. Input device 155 may include a cursor controller, such as a mouse, trackball, or cursor arrow keys, for transmitting directional information and command selection to processor 530 and for controlling cursor movement on display 155.

[0070] The processes, systems, and methods described herein can be implemented by a data processing system 130 in response to a processor 530 executing an instruction arrangement included in main memory 510. Such instructions may be read into main memory 510 from another computer-readable medium, such as storage device 520. Execution of the instruction arrangement included in main memory 510 causes the data processing system 130 to perform the illustrative processes described herein. One or more processors in a multiprocessor arrangement may be employed to execute the instructions included in main memory 510. Hardwired circuitry systems may be used in place of or in combination with software instructions with the systems and methods described herein. The systems and methods described herein are not limited to any particular combination of hardware circuitry and software.

[0071] although Figure 5An example computing system is described, but the subject matter including the operations described in this specification can be implemented in other types of digital electronic circuit systems, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations thereof.

[0072] Some descriptions herein emphasize the structural independence of aspects of system components or the grouping of the operations and responsibilities of these system components. Other groupings performing similar overall operations are within the scope of this application. Modules may be implemented in hardware or as computer instructions on a non-transient computer-readable storage medium, and modules may be distributed across various hardware or computer-based components.

[0073] The aforementioned system can provide multiple components or each of these components, and these components can be provided on a standalone system or on multiple instances of a distributed system. Furthermore, the aforementioned system and method can be provided as one or more computer-readable programs or executable instructions embodied in or contained in one or more artifacts. The artifact can be cloud storage, a hard disk, a CD-ROM, a flash memory card, a PROM, RAM, ROM, or magnetic tape. Typically, the computer-readable program can be implemented in any programming language, such as LISP, PERL, C, C++, C#, PROLOG, or any bytecode language such as JAVA. The software program or executable instructions can be stored as object code on or contained in one or more artifacts.

[0074] Examples and non-limiting module implementation elements include: sensors that provide any value as defined herein, sensors that provide any value as a precursor to the value defined herein, data link or network hardware (including communication chips, oscillating crystals, communication links, cables, twisted pairs, coaxial cables, shielded wires, transmitters, receivers, or transceivers), logic circuits, hardwired logic circuits, reconfigurable logic circuits configured according to the module specification and in a specific non-transient state, any actuator (including at least electric, hydraulic, or pneumatic actuators), solenoids, operational amplifiers, analog control elements (springs, filters, integrators, adders, dividers, gain elements), or digital control elements.

[0075] The subject matter and operations described in this specification can be implemented in digital electronic circuits, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations thereof. The subject matter described in this specification can be implemented as one or more computer programs, such as one or more computer program instruction circuits, encoded on one or more computer storage media for execution by or control of the operation of a data processing device. Alternatively or additionally, program instructions can be encoded on artificially generated propagating signals, such as machine-generated electrical, optical, or electromagnetic signals generated to encode information for transmission to a suitable receiver device for execution by the data processing device. The computer storage medium can be or is included in a computer-readable storage device, a computer-readable storage matrix, a random or serial access memory array or device, or a combination thereof. While the computer storage medium is not a propagating signal, it can be a source or destination of computer program instructions encoded in artificially generated propagating signals. The computer storage medium can also be or be included in one or more separate components or media (e.g., multiple CDs, disks, or other storage devices including cloud storage). The operations described in this specification can be implemented as operations performed by a data processing apparatus on data stored on one or more computer-readable storage devices or received from other sources.

[0076] The terms "computing device," "component," or "data processing apparatus," or similar terms, cover a wide range of devices, apparatuses, and machines for processing data, including, for example, programmable processors, computers, systems-on-a-chip, or a combination thereof. The device may include special-purpose logic circuitry, such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits). In addition to hardware, the device may also include code that creates an execution environment for the computer program in question, such as code constituting processor firmware, protocol stacks, database management systems, operating systems, cross-platform runtime environments, virtual machines, or combinations thereof. The device and execution environment can implement a variety of different computing model infrastructures, such as network services, distributed computing, and grid computing infrastructures.

[0077] Computer programs (also known as programs, software, software applications, applications, scripts, or code) can be written in any programming language, including compiled or interpreted languages, declarative or procedural languages, and can be deployed in any form, including as standalone programs or as modules, components, subroutines, objects, or other units suitable for use in a computing environment. A computer program may correspond to a file in a file system. A computer program may be stored as a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), as a single file dedicated to the program in question, or as multiple coordinated files (e.g., a file storing portions of one or more modules, subroutines, or code). A computer program may be deployed on a single computer or executed on multiple computers located at a single site or distributed across multiple sites and interconnected via a communication network.

[0078] The processes and logic flows described in this specification can be executed by one or more programmable processors, which execute one or more computer programs to perform actions by manipulating input data and generating output. These processes and logic flows can also be executed by special-purpose logic circuitry, and the apparatus can be implemented as special-purpose logic circuitry, such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits). Suitable devices for storing computer program instructions and data can include non-volatile memory, media, and storage devices, including, for example, semiconductor storage devices such as EPROMs, EEPROMs, and flash memory devices; magnetic disks, such as internal hard disks or removable disks; magneto-optical disks; and CD-ROMs and DVD-ROMs. Processors and memory can be supplemented or incorporated into special-purpose logic circuitry.

[0079] The subject matter described herein can be implemented in a computing system that includes backend components, such as a data server; middleware components, such as an application server; or frontend components, such as a client computer with a graphical user interface or web browser through which a user can interact with embodiments of the subject matter described herein; or combinations of one or more of such backend, middleware, or frontend components. The components of the system can be interconnected via any form of digital data communication medium, such as a communication network. Examples of communication networks include local area networks (“LANs”) and wide area networks (“WANs”), interconnected networks (such as the Internet) and peer-to-peer networks (such as self-organizing peer-to-peer networks).

[0080] Although the operations are depicted in a specific order in the accompanying drawings, such operations do not need to be performed in the specific order shown or in a sequential order, and it is not necessary to perform all the operations described herein. The steps described herein can be performed in a different order.

[0081] Having described some illustrative embodiments, it is clear that the foregoing is illustrative rather than limiting, and has been presented by way of example. In particular, although many of the examples presented herein relate to specific combinations of method steps or system elements, these steps and elements can be combined in other ways to achieve the same objective. Steps, elements, and features discussed in connection with one embodiment are not intended to exclude similar effects in other embodiments.

[0082] The wording and terminology used herein are for descriptive purposes and should not be considered limiting. The use of “comprising,” “including,” “having,” “containing,” “involving,” “characterized by,” “featured in,” and variations thereof is intended to cover the items listed thereafter, their equivalents, and additional items, as well as alternative implementations consisting only of the items listed thereafter. In one implementation, the systems and methods described herein include one, a plurality of each combination, or all of the described elements, steps, or components.

[0083] Any singular reference to any implementation, element, or step of the system and method herein may also include implementations that include multiple such elements, and any plural reference to any implementation, element, or step herein may also include implementations that include only a single element. References in either the singular or plural form are not intended to limit the currently disclosed system or method, its components, steps, or elements to a singular or plural configuration. Any reference to a step or element based on any information, step, or element may include an implementation of that step or element that is at least partially based on that information, step, or element.

[0084] Any embodiment disclosed herein may be combined with any other embodiment or example, and references to "implementation," "some embodiments," "one embodiment," or similar references are not necessarily mutually exclusive and are intended to indicate that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment or example. Such terms used herein do not necessarily refer to the same embodiment. Any embodiment may be combined with any other embodiment in any manner consistent with the aspects and embodiments disclosed herein, whether inclusive or exclusive.

[0085] A reference to "or" can be understood as inclusive, so any term described using "or" can indicate any one, multiple, or all of the terms described. A reference to at least one of the terms in a connected list can be understood as an inclusive "or" to indicate any one, multiple, or all of the terms described. For example, a reference to "at least one of 'A' and 'B'" can include only 'A', only 'B', and both 'A' and 'B'. Such references used in conjunction with "include" or other open-ended terms can include additional items.

[0086] When reference numerals follow technical features in the drawings, detailed embodiments, or any claims, these reference numerals are included to enhance the comprehensibility of the drawings, detailed embodiments, and claims. Therefore, the presence or absence of reference numerals does not limit the scope of any claim element.

[0087] Modifications to the described elements and steps may be made, such as changes in the size, dimensions, structure, shape, and proportions of various elements, as well as changes in parameter values, installation arrangements, material usage, color, and orientation, without substantially departing from the teachings and advantages of the subject matter disclosed herein. For example, an element shown as a single unit may be composed of multiple parts or components, the positions of elements may be reversed or otherwise changed, and the nature or number of discrete elements or positions may be altered or varied. Other substitutions, modifications, changes, and omissions may also be made in the design, operating conditions, and arrangement of the disclosed elements and operations without departing from the scope of this disclosure.

Claims

1. A system comprising: A data processing system, comprising one or more processors coupled to a memory, to: Receive a request to determine the material or parameters of the material; A model is constructed to determine the material or the parameters based on spectral data from a spectral sensor, the spectral data indicating the interaction of the material with light across the entire wavelength spectrum; as well as The model is deployed to a second data processing system, which executes the model based on the spectral data to determine the material or the parameters of the material.

2. The system according to claim 1, comprising: The data processing system, in order to: Receive a request to determine a second material or a second parameter of the second material; A second model is constructed to determine the second material or the second parameter based on the same spectral data from the same spectral sensor; as well as The second model is deployed to the second data processing system, and the second model is executed by the second data processing system based on the same spectral data to determine the second material or the second parameter of the material.

3. The system according to claim 1, comprising: The spectral sensor is used to measure the interaction between the material and light within a wavelength spectral range; as well as The second data processing system, in order to: Receive the spectral data from the spectral sensor; and The model is executed based on the spectral data to determine the material or the parameters of the material.

4. The system according to claim 1, wherein: The parameter of the material is at least one of biochemical oxygen demand or chemical oxygen demand.

5. The system according to claim 1, comprising: The spectral sensor is arranged in the pipeline of the production system; as well as The second data processing system, in order to: The model is executed based on the spectral data to classify the products of the production system in the pipeline.

6. The system according to claim 1, comprising: The spectral sensor is arranged in the pipeline of the production system; as well as The second data processing system, in order to: The model is executed based on the spectral data to identify the concentration of the product of the production system in the pipeline.

7. The system according to claim 1, comprising: The spectral sensor is arranged in the pipeline of the production system; as well as The second data processing system, in order to: The model is executed based on the spectral data to identify whether the product meets specifications; The specification mentioned therein is at least one of emulsification degree, component concentration, molecular concentration, or product color.

8. The system according to claim 1, comprising: The second data processing system, in order to: Control commands are determined based on the material or parameters of the liquid; as well as The control commands are transmitted to the controller, which controls the production system.

9. The system according to claim 1, comprising: The second data processing system, in order to: The first model is executed to classify the products based on the spectral data; The second model is executed to identify the concentration of the product based on the spectral data; as well as The product and its concentration are mapped to biochemical oxygen demand (BOD) levels.

10. The system according to claim 1, comprising: The data processing system, in order to: Receive product schedules, which identify various products produced by the production system at multiple points in time. Receive the spectral data at the plurality of time points; and The model is constructed based on the product plan and the spectral data.

11. The system according to claim 1, comprising: The data processing system, in order to: Receive laboratory test data, which indicates the material or parameters of the liquid at multiple time points; Receive the spectral data at the plurality of time points; and The model is constructed based on the laboratory test data and the spectral data.

12. The system according to claim 1, comprising: The data processing system, in order to: Receive laboratory test data, which indicates the biochemical oxygen demand level of the product at the identified concentration; A mapping is generated that maps multiple concentration levels of the product to multiple biochemical oxygen demand (BOD) levels of the product. as well as The mapping is then transmitted to the second data processing system.

13. The system according to claim 1, comprising: The spectral sensor is arranged in the upstream pipeline of the production system, which includes the upstream pipeline and the downstream pipeline. as well as The second data processing system, in order to: A first model is executed based on the spectral data to determine the material or parameters of the liquid in the upstream pipeline; as well as A second model is executed to determine the material or parameters of the liquid in the downstream pipeline, the second model modeling the time it takes for the material to flow from the upstream pipeline to the downstream pipeline.

14. A method comprising: A request to determine a material or parameters of the material is received by a data processing system, the data processing system including one or more processors coupled to a memory; The data processing system constructs a model to determine the material or the parameters based on spectral data from a spectral sensor, the spectral data indicating the interaction of the material with light across the entire wavelength spectrum. as well as The model is deployed to a second data processing system by the data processing system, and the model is executed by the second data processing system based on the spectral data to determine the material or the parameters of the material.

15. The method of claim 14, comprising: The data processing system receives a request to determine a second material or a second parameter of the second material; The data processing system constructs a second model to determine the second material or the second parameter based on the same spectral data from the same spectral sensor. as well as The second model is deployed to the second data processing system by the data processing system, and the second model is executed by the second data processing system based on the same spectral data to determine the second material or the second parameter of the material.

16. The method of claim 14, comprising: The data processing system receives laboratory test data, which indicates the biochemical oxygen demand level of the product at the identified concentration. The data processing system generates a mapping that maps multiple concentration levels of the product to multiple biochemical oxygen demand (BOD) levels of the product. as well as The data processing system transmits the mapping to the second data processing system.

17. The method of claim 14, comprising: The second data processing system executes a first model based on the spectral data to determine the material or parameters of the liquid in the upstream pipeline of the production system. as well as The second data processing system executes a second model to determine the material or parameters of the liquid in the downstream pipeline of the production system. The second model models the time it takes for the material to flow from the upstream pipeline to the downstream pipeline.

18. One or more storage media storing instructions that, when executed by one or more processors, cause the one or more processors to perform an operation, including: Receive a request to determine the material or parameters of the material; A model is constructed to determine the material or the parameters based on spectral data from a spectral sensor, the spectral data indicating the interaction of the material with light within a wavelength spectral range; as well as The model is deployed to a second data processing system, which executes the model based on the spectral data to determine the material or the parameters of the material.

19. The operation of one or more storage media according to claim 18, wherein the operation comprises: Receive a request to determine a second material or a second parameter of the second material; A second model is constructed to determine the second material or the second parameter based on the same spectral data from the same spectral sensor; as well as The second model is deployed to the second data processing system, and the second model is executed by the second data processing system based on the same spectral data to determine the second material or the second parameter of the material.

20. One or more storage media according to claim 18, comprising: Receive laboratory test data, which indicates the material or parameters of the liquid at multiple time points; Receive the spectral data at the multiple time points; as well as The model is constructed based on the laboratory test data and the spectral data.

Images