System and method for detecting water in a mixture

Abstract

A sensing system for a mixture, comprising a sensing region containing the mixture. Within the sensing region is a sensor, which includes a printed circuit board (PCB), a transmitting electrode configured to generate an electric field, and a receiving electrode. The transmitting and receiving electrodes are configured to be in direct contact with the mixture. The sensor is configured to detect electrical parameter values ​​of the mixture. The system further includes a signal analyzer that determines the water content of the mixture based on the detected electrical parameters.

Description

Background Art

[0001] Many products require mixtures in their manufacture - for example, commercial or industrial paints, adhesives, resins, and the like. Many mixtures contain different materials that can settle or separate over time. It may not be easy for the user of the mixture to recognize that its composition has become non-uniform.

Summary of the Invention

[0002] The present disclosure relates to a sensing system for a mixture comprising a sensing region containing the mixture. The sensor is within the sensing region and includes a printed circuit board (PCB), a transmitting electrode configured to generate an electric field, and a receiving electrode. The transmitting electrode and the receiving electrode are configured to directly contact the mixture. The sensor is configured to sense an electrical parameter value of the mixture. The system also includes a signal analysis device configured to determine the moisture content of the mixture based on the sensed electrical parameter.

[0003] Systems and methods including a sensor as described herein enable direct contact between a fluid flowing through a dispensing device and the sensor, and the sensors herein have low manufacturing costs and can be discarded after use. The systems and methods herein also enable collection of a plurality of sensor signals over the entire fluid flow, providing real-time information regarding the material flowing into and out of the mixing region. The systems and methods herein also enable detection and removal of air bubbles. The systems and methods herein also enable the dispensing system and its operator to change operating parameters to address problems occurring or likely to occur during operation, resulting in reduced waste of materials and more accurate dispensing.

[0004] The above summary of this disclosure is not intended to describe all embodiments or examples of this disclosure. The following description provides more specific illustrative embodiments. Guidelines are provided in several places throughout the application through lists of examples, which may be used in various combinations. In each case, the lists provided are merely representative and should not be construed as exclusive lists. Accordingly, the scope of this disclosure should not be limited to the specific exemplary structures described herein, but rather to the structures described by the language of the claims and their equivalents. Any element that is positively described as an alternative herein may, at will, be expressly included in or excluded from the claims. Although various theories or possible mechanisms may be discussed herein, in no case should such discussions be construed as limiting the subject matter of the claims. [Brief explanation of the drawing]

[0005] [Figure 1A] Figure 1A shows systems and apparatus that can benefit from embodiments disclosed herein. [Figure 1B] Figure 1B shows systems and apparatus that can benefit from embodiments disclosed herein.

[0006] [Figure 2A] Figure 2A shows an electrical parameter sensor according to an embodiment described herein. [Figure 2B] Figure 2B shows an electrical parameter sensor according to an embodiment described herein. [Figure 3A] Figure 3A shows an electrical parameter sensor according to an embodiment described herein. [Figure 3B] Figure 3B shows an electrical parameter sensor according to an embodiment described herein.

[0007] [Figure 4A] Figure 4A shows a material measurement flow sensor used according to embodiments described herein. [Figure 4B] Figure 4B shows a material measurement flow sensor used according to embodiments described herein. [Figure 4C] Figure 4C shows a material measurement flow sensor used according to embodiments described herein. [Figure 4D] Figure 4D shows a material measurement flow sensor used according to embodiments described herein.

[0008] [Figure 5A] Figure 5A shows a material flow sensor according to an embodiment described herein. [Figure 5B] Figure 5B shows a material flow sensor according to an embodiment described herein. [Figure 5C] Figure 5C shows a material flow sensor according to an embodiment described herein.

[0009] [Figure 6A] Figure 6A shows the electrical signal values ​​and moisture content values ​​measured using the system described herein. [Figure 6B] Figure 6B shows the electrical signal values ​​and moisture content values ​​measured using the system described herein.

[0010] [Figure 7A] Figure 7A shows the predicted mixing ratio and water content produced by the systems and methods described herein. [Figure 7B] Figure 7B shows the predicted mixing ratio and water content produced by the systems and methods described herein.

[0011] [Figure 8] Figure 8 shows a method for controlling the quality of a material dispensing system according to an embodiment described herein.

[0012] [Figure 9] Figure 9 shows a quality control system according to an embodiment in this specification.

[0013] [Figure 10A] Figure 10A shows a stacked electrical parameter sensor according to an embodiment in this specification. [Figure 10B] Figure 10B shows a stacked electrical parameter sensor according to an embodiment in this specification.

[0014] [Figure 11A] Figure 11A shows a sensor according to an embodiment in this specification. [Figure 11B] Figure 11B shows a sensor according to an embodiment in this specification.

[0015] [Figure 12A] Figure 12A shows a sensing system according to an embodiment in this specification. [Figure 12B] Figure 12B shows a sensing system according to an embodiment in this specification.

[0016] [Figure 13A] Figure 13A shows a conductivity measurement system in an exemplary network architecture. [Figure 13B] Figure 13B shows a conductivity measurement system in an exemplary network architecture. [Figure 13C] Figure 13C shows a conductivity measurement system in an exemplary network architecture. [Figure 13D] Figure 13D shows a conductivity measurement system in an exemplary network architecture.

[0017] [Figure 14] Figure 14 shows an exemplary computing device that can be used in an embodiment in this specification. [Figure 15]Figure 15 shows an exemplary computing device that can be used in embodiments of this specification. [Figure 16] Figure 16 shows an exemplary computing device that can be used in embodiments of this specification.

[0018] [Figure 17] Figure 17 shows the predicted and measured results for estimating the water content, which will be further explained in the example. [Figure 18A] Figure 18A shows the predicted and measured results for estimating the moisture content, which will be further explained in the example. [Figure 18B] Figure 18B shows the predicted and measured results for estimating the moisture content, which will be further explained in the example. [Figure 18C] Figure 18C shows the predicted and measured results for estimating the moisture content, which will be further explained in the example. [Figure 18D] Figure 18D shows the predicted and measured results for estimating the water content, which will be further explained in the example. [Figure 18E] Figure 18E shows the predicted and measured results for estimating the moisture content, which will be further explained in the example. [Figure 18F] Figure 18F shows the predicted and measured results for estimating the moisture content, which will be further explained in the example. [Figure 19A] Figure 19A shows the predicted and measured results for estimating the moisture content, which will be further explained in the example. [Figure 19B] Figure 19B shows the predicted and measured results for estimating the moisture content, which will be further explained in the example. [Figure 19C] Figure 19C shows the predicted and measured results for estimating the water content, which will be further explained in the example. [Modes for carrying out the invention]

[0019] This disclosure relates to a system and method including a sensor for determining the properties of a fluid in situ. The disclosure also relates to a dataset obtained by such a sensor and a method for analyzing the properties of the fluid using such data. Using the systems and methods described herein, it may be possible to adjust the operating conditions of a mixture (e.g., pressure, temperature, mixing ratio, etc.) or to improve the uniformity of the composition before or during operation (e.g., remixing, degassing, etc.). Furthermore, this disclosure may be useful in providing quality indicators for conditions important for quality control, namely, entrained air or moisture content of the mixture.

[0020] Many industrial processes utilize mixtures of liquids, such as liquid adhesives, liquid food components, liquid coolants, or liquid reaction products. The specific properties of such liquids can change over time. Specifically, dispersions or emulsions may separate, oils may decrease in viscosity with rising temperatures, coolants may degrade over time and have a lower heat capacity than initially possessed, and water may evaporate or be absorbed. As a result, the performance of the product during use may be reduced, for example, in materials with excessive or insufficient moisture. Troubleshooting these problems requires detailed chemical knowledge, time, and the elimination of other causes. In many cases, troubleshooting incurs an unacceptable time cost.

[0021] The pending international application IB2021 / 056362, filed on 14 July 2021, discloses a characteristic sensor for determining the characteristic values ​​of a liquid, comprising two printed circuit boards (PCBs) defining channels through which the liquid flows. While this configuration allows for direct contact between the sensor and the fluid, there is still a need for a cost-effective sensor that can provide more contextual information about the mixture of materials. Embodiments of this specification provide systems and methods for effectively and accurately measuring material information for quality control of mixtures.

[0022] The pending international application PCT / US2022 / 052343, filed on 9 December 2022, discloses sensors and sensor systems used for measuring the electrical properties of fluids. Generally, electrical property sensors function by having a transmitting electrode receive a voltage, thereby generating an electric field. As a fluid flows between the transmitting and receiving electrodes, the fluid conducts an electric current to the receiving electrode. However, while the embodiments herein describe applied voltages, in some embodiments it is explicitly assumed that an electric current is applied to the transmitting electrode and a voltage is received by the receiving electrode.

[0023] Furthermore, U.S. Provisional Applications No. 63 / 386,715, filed on 9 December 2022, No. 63 / 486,631, filed on 23 February 2023, and No. 63 / 507,662, filed on 12 June 2023, disclose characteristic sensors for determining material property values. The characteristic sensors described herein are configured to measure the electrical properties of a fluid and can be applied whether the fluid is flowing or stationary.

[0024] As used herein, the term "sensor" can refer to either the physical sensor itself that provides a sensor signal indicating conduction current, or to a "sensor system" that includes a processor that calculates the electrical properties of a fluid based on said sensor signal.

[0025] As used herein, the term “electrical properties” is intended to broadly refer to any electrical properties of a fluid that can be derived based on impedance measurements of a sensor. For the sake of understanding the embodiments, examples of impedance measurements are used herein. However, it is explicitly assumed that other electrical properties may be calculated and relevant to the embodiments herein. For example, conductivity measurements or dielectric constants may also be determined from impedance measurements. Either conductivity or dielectric constants may be relevant to determining the relevant functionality of a distribution system or the quality of the fluid flowing within it, as shown herein.

[0026] As described herein, sensors are described as measuring the electrical properties of a “fluid.” The term “fluid” as used herein is intended to be interpreted broadly and is intended to include low-viscosity liquids, high-viscosity liquids, semi-solid materials, suspensions, molten materials, or other fluid materials.

[0027] As used herein, “electrical parameters” can be detected by an electrode pair. A fluid may flow between or near the electrode pair. The transmitting electrode may generate an electric field when a voltage or current is applied, while the receiving electrode receives a current or voltage. The detected electrical parameters may be conductivity, relative permittivity, or impedance. The terms “relative permittivity” and “permittivity constant” are used synonymously herein.

[0028] Sensors as described herein are described as having one or more "apertures" within a "printed circuit board." These terms are intended to be interpreted broadly. For example, an aperture may extend along part or all of the length of the sensor, through its entire thickness. An aperture may have a chamfer on part or all of its perimeter. An aperture may be elongated, like a slot, or it may be circular or elliptical, like a hole. An aperture may have one or more corners or edges, or it may have curvature on part or all of its perimeter. As used herein, "printed circuit board" refers to a laminated sandwich structure of conductive and insulating layers. A printed circuit board (PCB) as described herein may include any number of terminals and conductors that enable the application of a voltage to a transmitting electrode and the conduction of current from a receiving electrode. However, in some embodiments, current may be applied to the transmitting electrode and a voltage may be detected at the receiving electrode. PCBs may be manufactured using conventional printed circuit board manufacturing techniques or additive manufacturing techniques. As used herein, the term PCB is intended to encompass any number of layers, with or without edge connectors. Any suitable conductive metal may be used to form the conductive layer. Any suitable insulating material may be used to form the insulating layer.

[0029] Any suitable technology, including those known to a person skilled in the art in the interest of the disclosure, can be used to form the sensors of this disclosure. In some embodiments, the sensors may be formed using three-dimensional electronic circuit printing technology (e.g., electronic circuit printing on a three-dimensional substrate or fully additive manufacturing). Non-limiting examples of fully additive manufacturing include fused filament fabrication, stereolithography, or inkjet printing. In a fully additive manufacturing approach, both the electronic and structural elements of the sensor may be manufactured using one or more additive manufacturing techniques, and this approach may also be described as a fully integrated manufacturing approach. Generally, the structural elements of the sensor may be described as a substrate material, and the electronic elements of the sensor may be described as a functional material.

[0030] Suitable substrate materials usable in three-dimensional electronic circuit printing technology may include polymers or ceramics. The substrate may include flexible materials such as polyimide or polyethylene terephthalate (PET). The substrate material may be selected to have good adhesion properties to the functional material used, withstand the curing or sintering used in the printing process, and have sufficient durability for the intended application of the sensor. The substrate material may be used to form the structural elements of the sensor (e.g., the dielectric substrate of a printed circuit board) by any suitable technique, including additive manufacturing techniques.

[0031] Suitable functional materials usable in three-dimensional electronic circuit printing technology may include conductive inks, dielectric inks, hybrid inks, or other functional inks. In some embodiments, conductive inks may be used to print conductive patterns, electrodes, and interconnects. Conductive inks generally contain conductive materials such as silver nanoparticles, graphene, or nanotubes dispersed in a liquid medium. Dielectric inks may be used to print electrically insulating structures. Dielectric inks may contain polymers or ceramics dispersed in a liquid medium. Hybrid inks can combine multiple functions into a single ink composition.

[0032] The characteristic sensors described herein can be used to sense the properties of a fluid obtained as a result of a mixing process. They can also be used to sense the properties of input fluids in a mixing process or an industrial manufacturing process. Advantageously, a separate characteristic sensor corresponding to each input fluid is placed immediately before the mixer. Data from these characteristic sensors measuring the input fluids can be processed together with data from characteristic sensors measuring the mixed fluid (e.g., within an integrated material property monitoring system). For example, if a fluid composition is mixed from three input fluids, three characteristic sensors can be used at the outlet of each of the three containers containing the three input fluids to determine the properties of each fluid before mixing. This contributes to quality control and reduces waste that may occur if one of the input fluids is outside of its specified properties.

[0033] The characteristic sensors described herein may also be used to sense the properties of a fluid in its bulk state or as the fluid passes through or flows near the sensor. The sensors described herein can determine various properties of a fluid, such as the mixing ratio of a two-component adhesive, the curing state of a curable composition, or the degree of degradation of a material over time. The sensors described herein can also detect the water content in a mixture and determine when and how the water content changes. The number of characteristics varied to establish a calibration dataset representing calibration impedance responses previously measured at different characteristic values ​​determines the number of characteristics that can then be determined by the characteristic sensor. A pre-stored calibration dataset, representing calibration impedance responses previously measured at different characteristic values ​​of the fluid's properties at one or more sensing frequencies, forms or represents a fluid-specific multidimensional data field. This data field allows the characteristic value derivation unit to determine the fluid's characteristic values ​​from the actually measured response impedance.

[0034] Fluids possess many properties. These include viscosity, density, color, content of volatile components, water content, chemical composition, boiling point, as well as the curing state if the fluid is a curable composition, and the mixing ratio if the fluid is a mixture.

[0035] Furthermore, since the specific properties of a particular fluid change with time and / or other parameters, the response impedance in the characteristic sensors described herein also changes with time and / or other parameters. The values ​​of these properties can be derived through the sensors and systems described herein. Changes with time also include changes in properties between different manufacturing lots of a fluid. Therefore, the characteristic sensors described herein can also be used to detect differences between later and earlier manufacturing lots for specific properties (e.g., chemical composition) of a suitable fluid.

[0036] The term “fluid properties” as used in this disclosure is not limited to the examples discussed herein. For example, one property of interest in the embodiments herein is the mixing ratio of two or more components of the fluid. In some of these embodiments, the fluid is a two-component adhesive, and the fluid property is the mixing ratio of the components. In other embodiments, the property of interest is the degree of curing or curing state. In some of these embodiments, the fluid is a curable composition, and the fluid property is the degree of curing of the composition.

[0037] In other embodiments, the characteristic of interest is the degree of degradation or degradation state. In some of these embodiments, the fluid is a degraded fluid, i.e., a fluid whose specific characteristics change over time after it is created. A characteristic sensor may determine the change in the response impedance of the degraded fluid after degradation by comparing it with the response impedance of the same fluid recorded at specific points in time before and after degradation. This allows the characteristic sensor to determine the degree of degradation or degradation state of the fluid. The characteristics of the fluid can take on different values.

[0038] For example, the kinematic viscosity of the fluid "water" can take values ​​such as 1.30 mPa·s or 0.31 mPa·s. Such values ​​are referred to herein as "characteristic values." Certain properties may not be related solely to numerical characteristic values. For example, the degree of hardening can take characteristics such as "unhardened," "partially hardened," or "fully hardened." Similarly, the hardening state can take characteristics such as "unhardened" or "fully hardened." The fluids according to this disclosure may be viscous fluids. Regardless of their viscosity, the fluids may be fluids that have flow properties. The fluids may be fluids that flow continuously.

[0039] As used herein, the terms “fluid” or “fluid mixture” broadly refer to a composition comprising two or more components. Both components may be liquids or particles in a liquid. Generally, “fluid” or “fluid mixture” refers to a substance that has fluidity. The systems and methods herein are useful for a range of fluid system applications, including but not limited to paints, resins (including adhesives and others), field-curing gaskets, adhesives or other coating materials, dental impression materials, void fillers, sealants, engineered fluids, thermally conductive interface materials, precursors to any of these, or emulsions and other materials that may lose stability over time.

[0040] The systems and methods described herein are particularly useful for determining whether and how much water is present in a mixture. For example, water contamination in monomers can cause numerous defects and performance problems in acrylate adhesives. For instance, in a premixture for an optically transparent adhesive, if the water content exceeds a certain threshold, it can cause gelation defects. Gelation defects render the optically transparent adhesive unusable and are difficult to detect. Therefore, water contamination is a significant cost and waste factor in the production of optically transparent adhesives. Water also reduces the production efficiency of acrylate adhesives.

[0041] Figure 1A shows a device having a touchscreen. The touchscreen comprises multiple layers, each layer of which must be properly bonded to adjacent layers in order for the device to function and meet quality control processes. One or more glass layers 120 are placed on top of the display layer 140. A touchscreen device 130 is provided between the glass layers 120. The touchscreen device 130 may include glass or other suitable devices having a transparent conductive coating made of metal or other suitable material. The top layer 110 may include a transparent coating and / or an oleophobic coating. Other suitable components may also be used.

[0042] Figure 1B shows several examples of defective samples in optically transparent adhesives. Image 150 in Figure 1B shows a micrograph of a die-cut defect sample caused by moisture contamination in the optically transparent adhesive layer of a device such as device 100.

[0043] Optically transparent adhesives are an example of adhesives where moisture content can be a quality control specification. Adhesives and sealants can absorb water after manufacturing, depending on the type of packaging and storage environment. While moisture content can be measured at the manufacturing site, it can change by the time the material is used at the customer's site. Therefore, it is desirable to measure the moisture content during the dispensing operation.

[0044] In some materials (e.g., urethanes and sealants), the large amount of water present in the bulk mixture can alter material properties (e.g., degree of curing and / or mechanical properties). Understanding the water content at the time of dispensing allows for adjustments to be made during processing, resulting in more consistent and predictable adhesive use.

[0045] Currently, moisture content is measured off-line using the Kari Fisher method, infrared analysis (IR), or gas chromatography. Therefore, there is a need for systems and methods that can measure moisture content in-line, in real-time or substantially in real-time, thereby reducing defect rates and improving productivity.

[0046] As used herein, the term "real-time" refers to data being processed in milliseconds and becoming available as feedback virtually immediately. Even if some processing delay is unavoidable, "real-time" is intended to encompass systems and methods in which data is collected or entered and users can interact with it without significant delay. For example, when a user enters data into a system, that data entry becomes available for viewing or editing virtually immediately.

[0047] In-situ detection of moisture content has proven to be a challenging problem due to the diversity of data and the difficulties in processing it using conventional curve fitting techniques. One approach attempted involved individually correcting temperature, mixing ratio, and dielectric constant at specific frequencies and correlating these with conductivity at those frequencies. However, this method has not been successful due to the number of processes involved and the required precision of the calculations, as some variables are intricately interrelated due to multicollinearity.

[0048] The system described herein measures the electrical properties (e.g., impedance, conductivity, or dielectric constant) of a liquid in contact with or moving through a sensor. The electrical properties change with respect to changes in water content. A change in water content results in a change in the signal response. This change in signal response is also a function of the frequency of the input voltage. Temperature also affects the electrical properties and can be measured in some embodiments. The mixing ratio of a mixture also affects the electrical properties.

[0049] As used herein, the term "curing" is intended to broadly encompass the change of a material from a first state to a second state. For example, a liquid may curing into a solid. A mixture may undergo a crosslinking reaction. A mixture may undergo prepolymerization. A mixture may undergo a transformation. Detecting these and similar changes in state is explicitly assumed in the embodiments herein.

[0050] The systems and methods described herein save resources through a high level of automation and can generate data histories for different monomer materials without requiring an in-line spectrometer. The sensors described herein may be used in in-line operation or in a bypass loop. The sensors described herein may be communicatively connected to or include a programmable logic controller, processing circuit, or other computing device. The measurement modes of the sensors described herein may be based on controlled bi-frequency dielectric constant and / or conductivity detection. The sensors described herein may have electrodes that are in direct contact with the mixture using a surface contact or flow-through architecture.

[0051] Figures 2 and 3 show an electrical parameter sensor according to an embodiment of this specification. Figure 2A shows an extended sensor according to an embodiment of this specification. In Figure 2A, the sensor 200 is shown having a separation length 230 between the electrode portion 220 and the edge connector 210. The edge connector 210 must not come into contact with the mixture. Therefore, providing a separation portion 230 between the edge connector 210 and the electrode portion 220 improves the flexibility of use of the conductivity sensor. For example, the sensor 200 can be used in deeper containers, ensuring consistency across the entire depth of the container. The sensor 200 can be immersed in a liquid and used to stir a mixture without the edge connector coming into contact with the fluid (and without causing a short circuit). This allows for real-time monitoring and visualization of material property data (conductivity, temperature, dielectric constant).

[0052] Figure 2B shows the sensor configuration 240, in which the electrode section 220 is communicatively connected to the sensor reader 250, for example, using an edge connector 210. The sensor reader 250 may be connected to an appropriate computing system so that the signals from the sensor 220 can be analyzed in real time or substantially in real time.

[0053] The systems and methods described herein can be used to sense electrical parameter values ​​in a variety of applications, such as adhesive dispensers, spray guns, bulk material storage tanks, and pump systems.

[0054] Figures 3A and 3B illustrate a material measurement flow sensor according to an embodiment of this specification. Figure 3A shows a PCB material measurement flow sensor 300. As shown in Figure 3A, the sensing system 300 includes a PCB substrate 302 having one or more grounds 330 and TX contacts 440. The TX contacts supply a transmit signal to each transmit electrode 310. Four RX contacts (not shown) are located on the back surface of the PCB and receive an impedance index detected from each electrode pair. The potential of each receiving electrode 320 is individually electronically controlled relative to the ground potential. In some embodiments, the control operation of each receiving electrode is interpreted as an impedance signal to each electrode pair. In the illustrated embodiment, there are four independent measurement channels, each providing information via separate TX contacts 340 and RX contacts (not shown).

[0055] In the illustrated embodiment, the sensing system 300 has four electrode pairs, with each of the four transmitting electrodes 310 paired with one of the four receiving electrodes 320. However, it is explicitly assumed that the number of electrode pairs may be greater or less, depending on the available space on the PCB substrate and the sensing requirements.

[0056] Each electrode pair is isolated from adjacent electrode pairs, thereby providing four independent conductivity measurements (from electrode pairs 310 and 320, respectively). In some embodiments, the sensing system 300 is positioned perpendicular to the material flow, with a first sensing region 352 receiving the first portion of the material flow, a second sensing region 354 receiving the second portion, a third sensing region 356 receiving the third portion, and a fourth sensing region 358 receiving the fourth portion. Thus, the system 300 can generate four different signals simultaneously for a single material flow, providing a better understanding of whether the mixing ratio (or other measurement parameter) is consistent across the sensing regions.

[0057] Compared to conventional sensing systems, conductivity measurement required both positive and negative electrodes, necessitating two PCBs for each electrode pair. In contrast, System 300 enables four measurements to be performed simultaneously on a single PCB. It also provides a larger surface area for material flow through a shorter sensor distance.

[0058] Figure 3A shows an embodiment in which each electrode pair is part of slots 352, 354, 356, and 358. However, it is conceivable that the structure may not be closed on both sides, but rather the sensing region may include a pair of electrodes on a protrusion or within an opening in a comb-like structure. However, particularly in the case of high-viscosity fluids, it is desirable from a structural standpoint that both ends are closed.

[0059] Furthermore, as will be described later, electrodes 310 and 320 may be formed by metallizing the inner surfaces of slots 352, 354, 356, and 358, for example, using copper. The metallization process may result in electrode 320 being connected to electrode 410. Therefore, a decoupling or cutting process is required. This can be done, for example, by drilling holes at positions 350A and 350B shown in the figure, or by punching, milling, nibbling, etching, laser cutting, or other appropriate methods of perforating the parts.

[0060] The systems and methods described herein can be used for a variety of materials to be distributed. PCB substrates typically have a maximum operating temperature of less than 170°C, which limits the material temperature that can be distributed via the sensing system 300. The material is, for example, about 10 5 The viscosity range may be up to Pa·s. If the viscosity is higher, the distribution pressure required to pass through slots 352-358 may not be sufficient without damaging the sensor. However, higher viscosity materials can be accommodated by widening the width of slots 352-358. However, this may reduce the sensitivity of the sensing system 300. Similarly, for materials containing particles, such as suspensions, the particle size must be smaller than the width of slots 352-358. Furthermore, the system according to this specification may be limited to solvents that do not corrode or damage PCB 302 or electrodes 310, 320.

[0061] Figure 3B shows another sensing system 360 according to an embodiment of this specification. The sensing system 360 includes a built-in temperature sensor 370. The temperature sensor 370 is located in a slot and has a connection point 372 for a ground signal and a connection point 374 for a temperature signal. The ground signal connection point 372 is connected to a ground signal transmitter 382. The temperature signal connection point 374 is connected to a temperature signal transmitter 376. Similar to the embodiment in Figure 3A, there are also four impedance or conductivity sensor slots 380, each connected to a ground signal 382. However, it should be noted that in the embodiment of Figure 3B, there are two different spacings between the slots. A first spacing 362 exists between the first and second slots 380 and between the third and fourth slots 380, and a second spacing 364 exists between the second and third slots 380. The wider spacing 364 may have the effect of improving shielding against interference between electromagnetic fields generated by each electrode pair.

[0062] Many mixing processes are at least partially temperature-dependent, and material properties such as viscosity change with temperature. Externally inserted temperature sensors are often fragile and need to be located in the center of the flow of the material under test. In the embodiment shown in Figure 3B, the temperature sensor is sealed within a housing, thereby isolating it from the material. The sealing layer may be, for example, a varnish layer, which can improve thermal contact compared to other housing materials. As shown, the temperature sensor is connected via contact 382 on an edge connector.

[0063] Figures 3A and 3B show embodiments in which slots 352-358, 370, and 380 are oval-shaped, with a generally straight body and rounded ends. However, other configurations are possible. For example, electrodes 310 and 320 may be curved or have other shapes to fit the available space in the distribution system.

[0064] Figures 3A and 3B show sensors that may be used according to embodiments of this specification. However, it is explicitly assumed that other sensor configurations are also applicable. For example, other configuration examples are described in U.S. Provisional Patent Application No. 63 / 486631, filed on 23 February 2023. Other configuration examples are also shown in Figures 9-11, 13-14, and 16-18 of U.S. Provisional Patent Application No. 63 / 507662, filed concurrently with this application. These Figures 9-11, 13-14, and 16-18 and their descriptions are incorporated herein by reference.

[0065] Figures 4A to 4D show material measurement flow sensors used according to embodiments of this specification. In some embodiments, the sensors are used in an inline configuration, and the fluid flows through the sensors before being distributed.

[0066] Figure 4A shows a sensor 410 positioned within the conduit 400. The sensor 410 has four electrode slots, and is configured so that the mixture passes through the slots of the sensor 410 as it passes through the conduit (i.e., through the electric field). Measurements of electrical parameters are transmitted to a control system, for example, via an edge connector 412. If differences are observed in conductivity measurements between electrode slots, these differences may indicate a change in the quality consistency of the mixture.

[0067] Figure 4B shows a perspective view 400 of the conduit 422. The conduit 422 may be connected to other parts of the distribution system or fluid transport system. The conduit 422 may be connected to other parts of the fluid flow system using a threaded section 426 or other suitable fastening mechanism.

[0068] Figures 4C and 4D show cross-sectional views of the conduit. In Figure 4C, overmolded plastics 444, 440, and 460 are used as seals to hold the PCB sensor in place. Such seals may include end stops to ensure the sensor is in the correct position. However, other seal configurations and positioning means (e.g., snaps or clips) are also possible. The illustrated seal may include barbs to maintain the connection.

[0069] Figure 4D shows a different sealing configuration in which an O-ring can be used. By machining a corresponding recess in the conduit to receive the O-ring 464, the sensor can be stabilized against the pressure of the fluid flow.

[0070] The illustrated conduit may be replaceable, in which case the sensing assembly becomes a disposable assembly. In other embodiments, the sensor may be removable, and the PCB sensor may be a disposable sensor.

[0071] The embodiments shown in Figures 4A to 4D relate to a PCB-based impedance sensor that can be connected to a static mixer and, using an adapter or other connection mechanism, provides real-time mixing ratio information. By using an adapter that can accept a PCB unit, the PCB sensor can be made compatible with multiple distribution systems.

[0072] Figures 5A to 5C show sensor configurations particularly useful for detecting the aggregation of second-phase bubbles or droplets that form before phase separation occurs, according to embodiments of this specification.

[0073] Figure 5A shows an immersion sensor particularly useful for detecting bubbles or droplets in a mixture. The sensor includes four pairs of electrodes arranged in four slots 502, 504, 506, and 508. Slot 502 is wider than slot 504, slot 504 is wider than slot 506, and slot 506 is wider than slot 508. Slots 502-508 are arranged in order from thickest to thinnest, although other arrangements are explicitly assumed to be possible. Similarly, all four pairs of electrodes are arranged coplanar and are all substantially the same distance from the edge connector 514.

[0074] Slots 502–508 are designed to detect bubbles or droplets and indicate their size. Generally, a uniform mixture without bubbles or droplets exhibits an insulating effect, maintaining constant conductivity across all electrode pairs. When a droplet reaches the width of a slot, it connects both sides of the electrode, causing a detectable change in conductivity.

[0075] In the design shown in Figure 5A, the slot width increases linearly (e.g., 1 mm, 2 mm, 3 mm, and 4 mm). The linear increase in diameter corresponds to a cubic increase in the volumetric flow rate passing through the opening. Such a configuration allows for a good understanding of how quickly phase separation will occur or how stable the mixture will be. For example, if phase separation is expected to occur more than an hour later, it may still be possible to distribute the mixture without taking corrective action.

[0076] However, in some embodiments, it is explicitly assumed that smaller or larger slot sizes will be required. For example, the thinnest slots may be less than 100 μm, or less than 150 μm, less than 200 μm, less than 300 μm, or less than 400 μm. One or more slots may be less than 500 μm, and one or more slots may be less than 1 mm. Larger slot sizes may also be required for other applications, such as when used as a stirring stick in large-scale measurement operations, such as checking the quality of a mixture in a 50-gallon drum.

[0077] In addition to variations in width, slots can also be modified in length to suit specific applications. For example, when verifying the shelf life of large containers, the entire sensor needs to be longer, for instance, up to 1 meter or more. In such cases, the opening needs to be enlarged to increase signal strength and ensure sufficient flow. By extending the length, signal strength can be increased while the width is appropriately selected to avoid obstructing flow and strike a balance without sacrificing signal strength.

[0078] The sensor configurations described so far have shown a single row of parallel electrodes (e.g., the horizontal configuration in Figures 3A and 3B, or the vertical configuration in Figure 11, described later). However, arrangements combining features of both configurations are also explicitly envisioned. For example, a grid arrangement of electrode pairs may be useful for simultaneously detecting consistency and mixture quality (or the presence of droplets / bubbles) at multiple depths. Furthermore, while the embodiments herein show four sets of electrode pairs in different configurations, it is explicitly envisioned that more or fewer sets of electrode pairs may be provided in either vertical or horizontal arrangements.

[0079] Figures 6A and 6B show electrical signal values ​​and moisture content values ​​measured using the system described herein, and the measurement results are for the commercial product DP460 epoxy, available from 3M® in Minnesota. By using the signals received from the sensors described herein, the moisture content of a mixture can be calculated using a machine learning algorithm trained on data obtained from mixtures with known moisture content percentages. Graph 600 shown in Figure 6A shows conductivity over time and the measured moisture content at the same time. The conductivity signal 610 and moisture content 620 are recorded by the electrical parameter sensor during the distribution operation. The sensor may be located in-line within the distribution device, may also measure the material source supplied to the distribution device, or may be located in other suitable locations. This ensures that the moisture content 620 is measured close enough to the distribution time to allow time to take necessary actions to prevent defects.

[0080] Figure 6B shows Graph 650 over time for conductivity 660 and moisture content 670, obtained on different days. A comparison of Figure 6A and Figure 6B shows that the curve shape for conductivity remains the same, while the moisture content decreases. The moisture content is consistent throughout the entire distribution operation in both Figure 6A and Figure 6B. As shown in Figures 6A and 6B, the systems and methods described herein can accurately calculate the moisture content of a mixture.

[0081] The electrical parameters of a liquid change based on its water content. That is, measurements of conductivity, impedance, and relative dielectric constant can be correlated with the percentage of water content. As described above, there are numerous variables that can be correlated to detect water content.

[0082] The systems and methods described herein utilize machine learning models trained on known data. Conventionally, methods that attempt to fit data by analytically applying multi-stage curve approximation methods without using machine learning have not been successful due to the involvement of numerous variables. Similarly, methods that individually correct for different variables such as temperature, mixing ratio, and dielectric constant and compare them to conductivity have not been successful due to the complexity of the procedure and the high level of precision required. Controlled and calculated variables have complex relationships due to multicollinearity. Machine learning models can be trained using data including, but not limited to, mixing ratio, conductivity, impedance, dielectric constant, temperature, frequency, applied voltage, flow rate, material elapsed time, material viscosity, resistivity, conductance, admittance, susceptance, reactance, current, phase shift, different lots, type of distribution valve, type of adhesive packaging, ambient humidity, ambient temperature, adhesive temperature, back pressure, presence of bubbles, adhesive filler material, filler particle size, filler filling amount, etc. In some embodiments, the machine learning model is trained using data of a single type of mixture. For example, one possible configuration might involve using a model specifically for DP460 data, and using a different model for other materials.

[0083] Figures 7A and 7B show the predicted mixing ratio and moisture content generated by the system and method described herein. Figures 7A and 7B show graph 700, which represents the mixing ratio, and graph 750, which represents the moisture content, respectively. In addition to the mixing ratio and moisture content shown in Figures 7A and 7B, it is explicitly assumed that the system and method described herein are also useful for calculating other material properties and for application to other materials.

[0084] Figure 8 shows a method for quality control of a material distribution system according to an embodiment of this specification. Method 800 can be used with the distribution apparatus described herein and can also be used for bulk material characterization, spray or coating systems, or other suitable applications.

[0085] In block 810, one or more components are supplied to a sensing area. The sensing area may be before a material dispenser, a transport line to a material dispenser, a nozzle, an atomizer, or other transport mechanism or container in the fluid system. The one or more components may include a premix, components to be mixed, a mixture, or bulk material. For example, the material dispenser may dispense liquid 812, suspended or other form of particles 814. The material may also be a mixture of materials 816, such as an emulsion or other mixture of components A and B. The emulsion must be dispensed as a stable emulsion, and the reactive components A:B should be supplied in a desired mixing ratio. Other components 818 may also be supplied to the sensing area for characterization and / or quality control.

[0086] In block 820, one or more components come into contact with a sensing system. Contact with the sensing system may include, for example, passing through one or more openings on the PCB before distribution, storage, or removal from storage. Contact with the sensing system may also include the mixture coming into contact with the surface of the sensor, i.e., a surface that constitutes part of the PCB. In the case of an electrical parameter sensor, direct contact between the material and the electrode pair ensures accurate measurement. The sensing system may be a bulk sensing system or a surface sensing system. The sensing system may receive signals from one, two, three, four, or more individual electrode pairs.

[0087] In block 830, electrical parameter measurements are received from a sensing system. The sensing system may have multiple sensors, for example, multiple electrode pairs, which detect the electrical parameters of the material when a sufficient voltage is applied to them. Based on the sensor signals, a number of things can be determined about the material. In the case of a mixture, the mixing ratio can be determined. In the case of a curable material, the progress of curing can be detected. The moisture content of the material can be determined. Aging can also be detected, and differences between material lots can also be detected. Indicators of instability such as trapped air, impending phase separation, and contamination can also be detected. Conductivity measurements may be performed sequentially, for example, every second, or at a higher frequency. Conductivity measurements may be performed in parallel from each of the multiple electrode pairs. In some embodiments, the electrode pairs may be coplanar with each other.

[0088] In block 840, feedback is provided based on conductivity measurements. The feedback may include material characterization, as shown in block 832. For example, mixing ratio, entrained air, single-component fluid pockets, time-series information, or other parameters of interest may be calculated and provided. Predictions may also be provided, as shown in block 834. For example, it is possible to predict the future behavior of the material under measurement based on trends in previous conductivity sensor readings. Other characterization information 838 may also be provided. For example, if the conductivity readings are skewed in one direction, it may indicate that the mixing ratio is approaching the edge of the acceptable range and therefore the mixing rate should be changed, or that instability is increasing toward phase separation. Similarly, the conductivity readings may indicate that a curable component is curing.

[0089] The feedback may also indicate that corrective action is required. For example, moisture content may be detected, as shown in block 842. Based on the percentage of moisture content, corrective action or parameter adjustment may be taken. The feedback may also include alerts, as shown in block 844, for example, if the detected moisture content is too high or too low, or if corrective action is required. The feedback may also indicate that purging of one component, multiple components, or a mixture is required. Other feedback may predictively indicate that the sensor needs to be replaced, as shown in block 846, if the material has corrosive effects or hardens over time. Other predictive information may also be provided, as shown in block 838, which may trigger other actions, as shown in block 848.

[0090] In some embodiments, as described herein, providing feedback may include providing electrical parameter readings, material properties, or predictions to the user or controller of the dispensing device. Useful information such as the material source, lot number, material name, dispensing temperature, dispensing pressure, material concentration, mixing ratio, or other information may also be provided. Sensed or calculated parameter values ​​may be provided to a storage device or historical data repository for later retrieval.

[0091] Figure 9 shows a quality control system according to an embodiment of this specification. The quality control system 950 may be used to identify and correct any detected inconsistencies in a mixture. The quality control system 950 may be implemented in a static environment, i.e., as an immersion rod or other analytical tool for a sealed fluid, or in a dynamic environment, i.e., in a fluid conduit through which a fluid passes or through an electrode pair.

[0092] Several systems and methods described herein can benefit from using relative thresholds instead of absolute thresholds. Measuring a reference level may be important to obtain more accurate relative thresholds. For example, if a conductivity measurement falls below a proportionality constant relative to the reference level (e.g., 50% of the reference level), it can be determined that a mismatch exists. This could be a concentration gradient indicating poor mixing, droplets indicating phase separation, or entrained air. Relative thresholds can help reduce material waste due to incorrect purging and wasted time attempting to correct mismatches at levels that do not exist or do not require correction. Similarly, relative thresholds can also be useful in determining whether a detected moisture content or mixing ratio is too high or too low, or whether the rate of change is too rapid.

[0093] The quality control system 950 may be implemented by a suitable computing device that communicates with the sensing system 910. This could be a compact signal receiving and analysis system, as shown in Figures 14A-D below, or a more conventional setup that transmits signals to the computing system using multiple cables. The quality control system 950 is illustrated with multiple functionalities integrated within a single architecture 950. However, it is explicitly assumed that the illustrated functionalities may be performed by separate processing devices, processors, or processing circuits at multiple locations.

[0094] The sensing system 910 may be any suitable sensing system that senses electrical parameters by directly contacting the material. The sensing system 910 may include one or more electrode pairs 912 that are in direct contact with the material flow. The sensing system 910 may also include a temperature sensor 914. The electrode pairs 912 may be, for example, part of a printed circuit board, or formed in an opening machined into or incorporated into the printed circuit board. The opening may be closed at both ends, or open at one end, for example, in a comb-like structure. In some embodiments, the temperature sensor 914 may be shielded from direct contact with the material flow. The sensing system 910 may include other features 916.

[0095] Sensor signals from the sensing system 910 are received by the quality control system 950 using an active signal acquisition device 952. The active signal acquisition device 952 may receive signals from the sensing system 910 periodically or continuously during operation. The received sensor signals may be electrical parameter values ​​related to the use case, including, but not limited to, impedance signals, conductivity signals, dielectric constant signals, or combinations thereof. In embodiments where conductivity values ​​are used to detect mismatches.

[0096] The history signal acquisition device 954 can communicate with a data store to acquire previously acquired signal values ​​or to transmit acquired signals at the time of acquisition. Past signal values ​​of interest may include signal values ​​acquired in recent periods from the same batch or the same material mixture. For example, values ​​acquired in the last few seconds or minutes may be important. In some embodiments, signal values ​​may drift over longer periods due to temperature changes, material degradation over time, fluctuations in mixing ratios, etc. However, mismatches may be detectable as abrupt changes in conductivity or discrepancies between conductivity measurements within the sensing system. In some applications, thresholds can be acquired to indicate when corrective actions should be taken. For example, these may be the time or amount of curing that indicates when purging should be performed, a range of acceptable mixing ratios, an acceptable amount of moisture in the mixture, etc.

[0097] In some embodiments, the threshold generator 960 periodically or continuously generates relative thresholds based on past signals. The relative threshold may be an absolute value that specifies, for example, that an increase or decrease of X% over Y time indicates an inconsistency. If the fluctuation of the received signal value is large, the threshold change value may be larger, but if the conductivity value does not fluctuate much, the threshold change value may be smaller.

[0098] The signal analyzer 962 compares the received signal or the calculated conductivity with a threshold, and if a deviation outside the acceptable threshold is detected, the command generation device 964 generates a command, which is then communicated using the command communication device 966.

[0099] The signal analysis may include a mixing ratio calculator 956 that calculates the mixing ratio of materials based on the received electrical parameter signals. The signal analysis may also include a moisture content calculator 958 that calculates the moisture content of the materials.

[0100] Based on the received signal, a comparison with a threshold, or calculated parameter values, feedback may be generated by the feedback generator 972. The feedback may include alarms, received values, or other communications. The feedback communication device 974 may communicate the feedback.

[0101] Commands or feedback communicated to the device 920 may, in some embodiments, include a display component, and the generated commands may be updates to a graphical user interface using the graphical user interface generation device 980, which may be displayed on the display component. In some embodiments, the device 920 may include feedback components such as acoustic, visual, or tactile feedback indicating to the control device that air bubbles have been detected. The device 920 may also be a correction mechanism, and the command generation device 964 may generate commands to execute a correction mechanism selected based on the detected mismatch, i.e., a purge valve, a remixing command, a degassing command, etc.

[0102] System 950 may include other features 976.

[0103] In some embodiments, the signal analyzer 962 includes a machine learning model for predicting parameter values ​​from either historical signal data, current signal data, or a combination thereof. This prediction may include so-called confidence intervals. Training may be performed in advance on a reference dataset in which no quality control concerns have been detected or which have quantified quality control concerns. The signal analyzer 962 then compares the received signal to determine whether it is within or outside the confidence interval.

[0104] As described herein, measuring electrical parameter values ​​may be used to determine the current mixing ratio and / or current moisture content of a material. However, the systems and methods herein are also explicitly intended to determine other information about the material. For example, as described herein and in the Examples section of PCT / US2022 / 52343, conductivity measurements may be used to determine lot-to-lot variability, trapped air, droplet formation, degradation over time, concentration gradients, and consistency issues due to dispersion separation or emulsion separation.

[0105] The sensor systems described so far are based on a single PCB board. Such systems are relatively inexpensive and therefore cost-effective in use and replacement. However, one drawback of the designs described so far is the large leakage field compared to the main field present between each electrode pair. The leakage field effect arises from the short distance between the input and output of the material flow, i.e., the thickness of the PCB. One way to reduce the leakage field effect is to solder multiple PCBs with openings containing electrodes together to form a PCB stack.

[0106] Figures 10A and 10B illustrate stacked electrical parameter sensors according to embodiments of this specification. Figure 10A shows a stacked PCB sensor described in Figure 10B of U.S. Provisional Patent Application No. 63 / 386,715 (filed December 9, 2022) and No. 63 / 486,631 (filed February 23, 2023). Figure 10B shows a stacked PCB sensor on a PCB substrate similar to Figures 5A to 5C. As shown, in one embodiment, the sensor stack 1000 may include four PCB sensors comprising one four-layer PCB 1010, two stacking PCBs 1020 provided to obtain the required sensitivity by increasing the electrode surface area, and a top PCB 1030. In some embodiments, the flow through the sensor stack 1000 is indicated by arrow 1040. Both embodiments in Figures 10A and 10B show a 4-layer sensor stack, but it is explicitly assumed that fewer or more PCB sensors may be connected. For example, it may consist of as few as 2 PCBs, or as many as 5, 6, 7, 8, 9, 10 or more PCBs.

[0107] The stacked sensor 1000 offers the advantages of a single PCB sensor with reduced leakage field effects. Its compact design also improves shielding of the sensitive electrode, and the sensitive region can be internally sealed, allowing it to be used as an electrode cartridge without the need for additional housing. In some embodiments, the sensitive region is internally sealed by soldering, allowing it to withstand pressure applied from the material sensor without the need for additional housing.

[0108] Furthermore, the stacked sensor 1000 can utilize smaller electrodes, and the sensor stack 1000 can be integrated into an active or passive mixing nozzle not only in the material input section but also in the material output section. The sensor stack has only one electrode 1010 having an edge connector configured to connect to a lead wire. As shown in Figures 10A to 10B, the stacked sensor may include a temperature sensor, and may also include an elongated section in embodiments where a stirring rod is a suitable medium for detecting conductivity. However, it is explicitly assumed that any configuration of the stacked sensor may be suitable for placement in a conduit as shown in Figures 4A to 4D without including an elongated section.

[0109] Figures 11A and 11B show sensors according to embodiments of this specification. Here, it is shown that multiple electrode slots may be arranged in a row, with each slot being approximately the same distance from the edge connector. It is also shown that multiple electrode slots may be arranged in a row, with each slot being at a different distance from the edge connector. Furthermore, in some embodiments, it is explicitly assumed that the electrode slots may be arranged in both rows and columns.

[0110] Figure 11A shows a sensing configuration 1100 comprising a sensor 1110 partially immersed in solution 1120. The sensor 1110 includes a first size 1102 electrode slot and a second size 1104 electrode slot. The electrode slots are located in both row 1108 and column 1108. Arranging the electrode slots in both rows and columns provides additional insight with respect to the material.

[0111] Figure 11A shows a homogeneous solution 1120, while Figure 11B shows a solution 1150 that has undergone sedimentation. Sedimentation may be an indication of material degradation over time. Sensor 1140 may provide 12 different sensor signals for analysis, providing one signal from each electrode pair through which the material can flow. The difference between the signals from electrode slots 1142 and 1144 may indicate degradation over time. The difference between the signals from electrode slots 1144 and 1146 may indicate the viscosity of solution 1150.

[0112] In many production facilities, raw materials are stored in large containers such as drums. Separation of materials results in a lighter phase at the top and a heavier phase at the bottom. As separation progresses, quality can deteriorate.

[0113] Furthermore, it is desirable to have sensors capable of processing materials with a wider viscosity range. Narrow electrode slots may not be able to process high-viscosity materials well, while wide electrode slots may result in lower accuracy for low-viscosity materials. Sensors 1110 and 1140 can provide signals along the depth of the material container while processing a wider viscosity range. Although only four pairs of electrodes are illustrated, it is explicitly assumed that in other embodiments, more rows may be present to accommodate the depth of the container. Also, although only three rows are illustrated, it is explicitly assumed that additional rows with wider or narrower electrode slots are also possible.

[0114] The arrangement of sensors stacked vertically along the PCB allows each electrode slot to be positioned at a different depth within the container. For example, electrode 1144 at the lowest depth measures the first conductivity. Electrode pair 1152 at a different depth measures the second conductivity. Due to some sedimentation or separation, the received electrical parameter values ​​will be different.

[0115] Figures 11A and 11B show a sensor 1100 having 12 electrical pairs arranged in a grid on a PCB, but it is explicitly assumed that there may be different numbers of electrode pairs. For example, there may be more than 4 rows, i.e., 5, 6, 8, 10 or more rows, and more than 3 rows, i.e., 4, 5, 6, 8, 10 rows. Furthermore, the spacing between electrodes may be longer or shorter than that shown.

[0116] Figures 12A and 12B illustrate another embodiment of a system in which the embodiments described herein may be useful. In the example shown in Figure 12A, some items are similar to those shown in the previous figure. Figure 12A specifically shows that the sensing system 1210 may be located at the remote server location 1202. Thus, the computing device 1220 accesses those systems via the remote server location 1202. The user 1250 can also access the user interface 1222 using the computing device 1220. For example, the user 1250 can interact with an application on the user interface 1222 of a smartphone 1220 or laptop 1220 or other computing device 1220 to receive information from the dispensing system or quality control system.

[0117] Figure 12A also shows that it is assumed that some elements of the system described herein are located at the remote server location 1202, while others are not. For example, data stores 1230, 1240, and / or 1260 may be located elsewhere than location 2002 and accessed through the remote server at location 1202. Wherever they are located, they may be accessed directly by the computing device 1220, accessed via a network (either a wide area network or a local area network), hosted or provided as a service at the remote site by a service, or accessed by a connectivity service located at the remote location. Also, data may be stored in virtually any location and accessed or transferred intermittently by interested parties. For example, a physical carrier may be used instead of, or in addition to, an electromagnetic carrier. This allows user 1250 to interact with system 1210 via computing device 1260 and initiate the seal check process.

[0118] It should also be noted that elements or parts thereof of the systems described herein may be deployed on a wide variety of different devices. Some of these devices include servers, desktop computers, laptop computers, embedded computers, industrial controllers, tablet computers, or other mobile devices (such as palmtop computers, mobile phones, smartphones, multimedia players, and personal digital assistants).

[0119] The conductivity measurement system may be any suitable system configured to collect conductivity measurements, perform analysis, and provide the analysis to a receiving device, storage, or graphical user interface generating device using the systems and methods described herein. Figure 19 of PCT / US2022 / 52343 illustrates the operation of such a system and is incorporated herein by reference.

[0120] System 1210 receives conductivity measurements from one or more sensors 1270. Each sensor may include one or more electrode pairs on a PCB. The electrodes are coplanar and, in some embodiments, may be spaced similarly apart from one end of the PCB, or may be aligned in a straight line along the length of the PCB. The sensors may be formed by metallization or other processes. The sensors 1270 are isolated from each other, and independent conductivity signals are received from each sensor. Each sensor 1270 may include positive and negative electrodes that are isolated from each other.

[0121] The electrical parameter measurement system 1210 may receive the sensor signal as a conductivity signal or a dielectric constant signal, or it may receive it as an impedance signal. In embodiments where the received signal is an impedance signal, the conductivity value may be calculated based on the impedance signal. Similarly, the dielectric constant may be calculated based on the received impedance signal. Based on the received sensor signal, calculations and / or predictions may be made (as shown, for example, in Figure 12). The mixing ratio may also be calculated based on calibration data stored in the data store 1260, which may show conductivity data from pure components and / or mixtures of known components. As described above, sensors may be placed at both the inlet and outlet of the sensing area, and so the system 1210 can receive sensor signals from all sensors associated with the material dispensing system. In some embodiments, the system 1210 may be configured to compensate for a time delay between the acquisition and analysis of the sensor signal. In other embodiments, compensation may not be necessary, especially if trend information is particularly relevant.

[0122] The systems and methods described herein utilize machine learning algorithms. Machine learning models are sometimes preferred because they can better handle noisy data, predict future signal trends, and allow adjustments to be made before significant changes in mixing quality occur. The systems and methods described herein can calculate mixing ratios in real time. Machine learning techniques allow mixing ratios to be predicted in advance. This enables faster adjustments and allows the mixing ratio to be kept closer to the target value for longer periods. In some current dispensers, large amounts of material are introduced into a static mixer, so by the time a change in mixing ratio is detected, the material in the mixer already has an incorrect mixing ratio for at least one mixer load of adhesive; therefore, early identification of mixing ratio problems can save material and potential purging.

[0123] Similarly, as described herein, a machine learning model can receive information from multiple systems, including multiple sensors within a dispensing system (such as conductivity sensors, temperature sensors, motor speed signals, and material information). In some embodiments, multiple machine learning models are used simultaneously, each used by a separate system, so that the model for each system can learn and the overall model can be improved. However, it is explicitly assumed that non-machine learning models may also be used.

[0124] The sensing systems described herein are said to have the ability to receive and transmit information that can be communicated with other devices. This may be done, for example, via an application program interface. In this case, system 1210 can communicate with a pump controller, a line pressure sensor, an operating controller for a part of a dispensing system, a temperature sensor, a heating element, or a data store having information about the material to be dispensed or the mixture to be produced.

[0125] In embodiments where machine learning models are used, the data store may also include an analyzer that learns the usage behavior of a particular dispensing system to improve its behavior and predictions. Similarly, dispensing frequency and patterns can provide information about curing and improve the mixed model. For example, usage data such as dispensing frequency, purging frequency, dispensing patterns, and sensor replacements can be collected and used to train the model to more accurately predict trends and provide corrective actions.

[0126] Similarly, as described herein, the display device 1260 displays a GUI created by the generator 1220, which is periodically updated with information collected by the system 1210 and / or any data stores 1230-1260. The information may be updated passively or provided with alerts or notifications when updated. For example, current status information may be displayed, and an alert (visual, auditory, or tactile) may be provided when the mixing ratio is approaching an acceptable range. In addition, or alternatively, notifications may be provided when a device command is generated or when operator intervention is required.

[0127] In some embodiments, the signal encoder and regressor may operate locally using, for example, a computer processing unit associated with a material dispensing system. Alternatively, the encoder or the regressor, or both, may be deployed on a cloud-based storage system.

[0128] The encoder output may be used directly to modify the dispensing parameters of one or more material components so that the mixture satisfies a predefined mixing ratio. For example, if there is too much part A in the mixed material, the flow rate from the cartridge containing part A may be reduced and the flow rate from the cartridge containing part B may be increased. For example, the operating parameters may include changing the speed or output of an extruder, progressive cavity pump, gear pump, or other suitable positive displacement pump.

[0129] Next, the regressor takes in the encoded signal and generates a mixing ratio signal. The regressor may be a machine learning-based algorithm that can be trained in any suitable way.

[0130] The first training option is a segregated training option in which the encoder-decoder model is trained on signal sets of Part A, Part B, and various parts of a diverse mixture. A machine learning regressor is then trained in a subsequent second step on the encoded signals and corresponding mixing ratios.

[0131] The second training option is the alternating training option, where one batch of signals is used for one training step in the encoder-decoder and then for one training step in the encoder-machine learning regressor part. The training step consists of a forward pass of the data in the batch, gradient calculation, and gradient application to optimize the weights in the model.

[0132] The third training option is a coupled training option that simultaneously optimizes a tripartite set of encoder-decoder pairs and a machine learning model. This means that batches are forwarded through the encoder, and the resulting representations are forwarded through the decoder and the machine learning regressor. The gradients computed on both outputs are then applied in a backward pass as a weighted coupler.

[0133] Alternating or coupled training can offer the advantage of learning how signal representations positively impact the regressor's performance, thereby reducing errors in estimating mixing ratios. By learning signal representations for various materials and mixing ratios, the model can be used for unidentified materials within the same chemical family.

[0134] Unlike systems that use only a single signal from the mixed materials, this new approach allows for adaptation to lot-to-lot variations in raw materials; that is, a change in one part can cause a change in the mixing signal for the same mixing ratio. It also allows tracking the mixing of new materials of the same family by learning to fuse the signals of two parts to generate a mixed signal.

[0135] Data traces collected from the sensor system may be processed to provide other information, as described herein. For example, the sensor may provide a signal indicating that corrective action is required.

[0136] As described herein, in some embodiments, the sensor includes four electrode pairs. By analyzing the time series of conductivity from the four sensor capacitors, it is possible to determine when the corrective action was successful, i.e., when remixing is complete, when phase separation is reversed, or when the mixture has again reached stability.

[0137] For example, mixing (or remixing) may take time to reach a steady state. For instance, when initiating a mixing operation, back pressure and the differing viscosities of the components may cause the mixing to start unsettled and gradually stabilize. The same dispersion can be used to track stabilization and indicate when the dispenser can dispense the material onto a workpiece or receiving container. The dispersion trend can be analyzed against a threshold, which is unique to each material. However, instead of determining a threshold, the signal can be tested for steadyness using the extended Dickey-Fuller test. The advantage of this is that manual thresholds often need to be adjusted for each new batch, whereas the ADF test is adaptable.

[0138] Non-uniformity can also be detected using the sensors described herein. The four electrode pairs should also record similar readings. While a certain offset is possible due to manufacturing tolerances, in a stable mixing process, the fluctuations of the four signals should be synchronized. After each signal stabilizes, the four sensors should have high covariance. Negative covariance indicates persistent anticorrelated behavior and signifies spatial non-uniformity.

[0139] Similarly, a single component of a mixture can also be heterogeneous, for example, due to sedimentation in the barrel or insufficient mixing during manufacturing. The extended Dickey-Fuller test can be used again to confirm stability over a longer period of time. The relevant time frame is determined by the time required to empty the container.

[0140] Figure 12A shows a concentration profile simulation system architecture 1200. However, architecture 1200 represents one embodiment of an implementation of a conductivity sensing system 1210. As an example, architecture 1200 can provide computation, software, data access, and storage services that do not require end-user knowledge of the physical location or configuration of the system providing the service. In various embodiments, remote servers can provide services over a wide area network such as the Internet using appropriate protocols. For example, remote servers can provide applications over a wide area network, which can be accessed through a web browser or other computing components. The software or components and corresponding data shown or described in Figures 1 to 20 may be stored on servers located at remote locations. Computing resources within a remote server environment may be aggregated at remote data center locations or distributed. The remote server infrastructure can provide services over a shared data center, even though it appears to the user as a single point of access. Thus, the components and functions described herein may be provided from remote servers located at remote locations using a remote server architecture. Alternatively, they may be provided by conventional servers, installed directly on client devices, or provided in other ways.

[0141] It should be noted that dispensing, coating, or spraying operations may be performed in parallel or at multiple different locations on the manufacturing floor. Multiple users may need to view information on multiple production lines simultaneously. Figure 12B shows one configuration of a system that can provide such functionality.

[0142] Figure 12B shows a signal analysis system communicating with multiple devices using a cloud-based network. The signal analysis system 2100 may communicate with a local analysis system 2140. The signal analysis system 2100 can receive multiple sensor signal data 2110 from multiple dispensing operations such as a pilot line 2104, any operational line 2102 and / or laboratory setup 2106. The sensor signals 2100 may be digital signals, analog signals, conductivity (or impedance or dielectric constant) measurement signals, pressure signals, or other signal information. For example, a low reservoir detection signal, a valve switch indication, or any other detectable indication from any of systems 2102-2106.

[0143] The signal analysis system 2100 may perform analysis on the received sensor signal information 2100. For example, any suitable analysis tool such as a lookup table, comparison threshold, and / or machine learning algorithm may be used to detect parameter trend information that indicates a problem or requires action such as purging or adjusting the mixing ratio.

[0144] The signal analysis system 2100 may provide output displays 2120 to multiple suitable devices 2150. The signal analysis system 2100 may provide output information 2120 continuously or in response to request information 2134. Request 2130 may be a one-time request for current status information, or a request for continuous updates in the future.

[0145] Figures 13A–13D illustrate sensing systems according to embodiments of this specification. Current sensing configurations include components from different manufacturers, and data preparation and processing are performed using separate computing devices. However, there is a need for a robust and compact system that can process data rapidly with limited downtime or startup time so that quality concerns regarding materials can be detected quickly.

[0146] In some embodiments of this specification, the sensor is a "smart" sensor, for example, one that includes signal preparation and processing within a single housing. Such a smart sensor includes a processing component—for example, a microprocessor, microcontroller, digital signal processor, or other processing circuitry. In some embodiments, the sensor also includes one or more standardized interfaces for interfacing with other systems—for example, a fieldbus system, a sensor network, an input / output link, etc. In some embodiments of this specification, sensor signal processing is completed without an external computer. The sensing systems of this specification offer decentralization, improved reliability, cost reduction, increased flexibility, and simplification.

[0147] In some embodiments, the sensor system herein includes a concentrator that integrates electronic components into a single housing. In some embodiments, all electronic components are located on a single PCB. In some embodiments, an analog front end with signal conversion (e.g., AD converter, DA converter, or both) is connected to a microcontroller that performs signal conversion, processing, and provides an output signal. The sensing system herein may also incorporate operating circuits including power supply, input / output protection circuits, signal regulation, reset management, and / or debugging circuits and interfaces. In some embodiments herein, the concentrator includes user interface components such as LED signals, UART, USB, wireless interfaces (e.g., Bluetooth®, WiFi, Zigbee®, cellular network), dot matrix or alphanumeric displays, industrial bus systems, and / or tactile interface components such as push buttons, switches, and touchscreens.

[0148] The systems described herein may include user-accessible data such as signal values, pass / fail (e.g., "yes" or "no", "forward" or "stop"), etc. The systems described herein may provide quality or quantity indications. The systems described herein may provide data streams with time and / or frequency-dependent data for storage and / or further processing. The systems described herein may include algorithms and / or calibrations necessary for data manipulation.

[0149] Figure 13A shows a schematic diagram of a sensing system according to an embodiment of this specification. The sensing system 1300 may be used with, for example, the sensors described herein, or other suitable sensors. A sensor signal reader 1302 is connected to the sensor, for example, an edge connector on a PCB board containing one or more electrode pairs. In some embodiments, a transimpedance amplifier is present to convert current measurements into voltages. A concentrator 1310 receives the sensor signal, processes the sensor signal, and provides an output. The output may be provided using an I / O device 1306 and / or other wired or wireless communication protocol 1308. A power supply 1312 may supply power to the concentrator 1310. Although a wired power supply 1312 is illustrated, power may be supplied wirelessly, or the concentrator 1310 may be integrated into a material dispensing system that draws power.

[0150] Figure 13B shows an interface 1320 of one example of a concentrator that can receive sensor signals using one or more sensor signal receiving ports 1324. Other data or inputs may be received through other receivers 1322 in some embodiments.

[0151] Figure 13C shows another interface 1330 that can receive connections to input / output devices. Power may be supplied, for example, using port 1334. Data may be communicated from the concentrator using computer link 1336.

[0152] Figure 13D shows a component diagram of a sensing system 1340 according to embodiments of this specification. One or more sensors 1342 provide sensor signals, which are received by one or more receivers 1344 coupled to or included in the housing 1370. In some embodiments, the system 1340 includes an analog front end, including a filter 1348 and / or an analog multiplexer 1346. Converters such as a DA or DC converter 1349 may be present. The concentrator 1350 may include a non-volatile memory 1352, a flash memory 1354, or other suitable information storage device. A temperature sensor 1356 may be incorporated into the concentrator 1350 or may receive a temperature signal from a temperature sensor. The concentrator 1350 may include a clock 1358. The concentrator 1362 may also include a reset function 1362.

[0153] The sensor analyzer 1370 may include calibration data and / or functions 1372. A real-time operating system 1373 can manage the functions. The sensor analyzer 1370 may include a Fourier transformer 1376. The sensor analyzer 1370 may include a waveform generator 1376. The sensor analyzer may include other applications 1375 that provide other functions, such as detecting material properties such as mixing ratio, material degradation over time, and hardening progression. The sensor analyzer 1370 may also include an identifier 1374 that identifies the type of sensor.

[0154] The concentrator 1350 may include a power supply 1366 or a power management system 1360 that accesses it. Power quality 1368 may be monitored. Energy consumption 1369 may be tracked. Conversion input and output ranges 1364 may be stored. Symmetric voltage 1367 may be used.

[0155] Figure 14 shows a dispensing system according to an embodiment of this specification. Many dispensing operations are performed with portable handheld systems. Dispensing errors or adhesive defects can occur if the quality of the materials or machine settings are incorrect. For example, incorrect mixing ratios or incorrect pressure settings may result in unacceptable products. A handheld dispensing system that can provide the user with real-time sensing and feedback is desired.

[0156] Figure 14 shows an example of a system capable of receiving and processing sensor signals without a separate computing device. This specification describes sensors of many embodiments that can be used with dispensers. This specification also describes a system for measuring pressure within a dispensing system.

[0157] System 1400 includes a dispenser 1410. Although dispenser 1410 is illustrated as an adhesive dispenser 1410, the system described herein is also useful for other dispensers. Dispenser 1410 includes an inline sensor 1430 that senses the electrical properties of the material being dispensed. A pressure sensor 1440 is incorporated into dispenser 1410 to monitor the pressure within the dispenser.

[0158] The dispenser 1410 also includes a signal processing system 1420. A signal receiver receives parameter signals sensed from the sensor 1430. The processing unit, which may include any suitable processor or processing circuitry, processes the sensed signals. Memory may store calibration data, history signals, etc. A display 1450 presents the processed information received from the signal processing system 1420 to the user, for example using a communication module. The display 1450 may be integrated into the dispenser 1410, or it may be another display visible to the dispenser operator—for example, a mobile computer or a work site display. Although the display 1450 is illustrated to convey processed information to the operator, in some embodiments of this specification it is explicitly assumed that the output from the signal processing system 1420 is presented as audio or haptic feedback.

[0159] Based on the sensing signal, the signal processing system 1420 may activate changes to the dispensing parameters. For example, a mixing ratio that deviates from a specified mixing ratio may be sensed. Based on the sensed drift in the mixing ratio, the signal processing system 1420 may adjust the mixing ratio by changing the pump speed of one of the components. The signal processing system 1420 may directly or indirectly control the pump speed and send a command to the pump controller to change the pump speed. The signal processing system 1420 may also communicate the drift in the mixing ratio, for example, via the display 1450. In some embodiments, the signal processing system 1420 may detect sensed material problems—e.g., mixing ratio, degradation over time, hardening, pressure, etc.—and the operator may need to manually take action to resolve the problem. However, in some embodiments of this specification, it is explicitly assumed that the dispenser parameters are automatically adjusted in real time based on signals from sensors 1430 and 1440.

[0160] In some embodiments, the dispensing system 1400 includes a material inventory system 1460. The material inventory system 1460 may store physical materials 1462 and dispensers 1464 (e.g., different types of static mixers for placement within the dispensing system 1410) available to the operator. However, in some embodiments of this specification, it is explicitly assumed that the material inventory system 1460 stores only information regarding materials 1462 and dispensers 1464.

[0161] Materials 1462 may include information related to the dispensing operation in which they are used. For example, a dispensing cartridge may include an RFID tag, an NFC tag, or other wirelessly accessible data storage. The information may also be transmitted using a printed code—for example, a barcode or a QR code. In some embodiments, a printed RFID label is applied to the dispensing cartridge.

[0162] When a cartridge of material 1462 enters a data transfer range (such as an RFID range or NFC range), dispensing information can be acquired by the dispensing system 1410 or the material inventory system 1460. The dispensing information may include dispensing parameters 1466, such as the operating pressure of one or both components, and / or recommended values ​​1468 for use with material 1462. Other information may also be provided.

[0163] Based on the received information, the system 1410 may display operational guidance to the operator, for example, on a display device 1450. For example, in some embodiments, the dispenser receives expected process parameters from a material information system 1460 based on the identification of the material 1462 from an NFC tag, RFID tag or other information storage system on the material to be dispensed.

[0164] By using sensors such as those described herein, or other suitable sensors, the dispensing system 1410 can receive sensed electrical parameter values ​​from which it can determine ongoing process conditions—such as mixing ratio, material degradation over time, and curing. Based on the sensed process values, guidance may be provided, or the system 1410 may automatically take action to correct any discrepancies.

[0165] Figure 15 provides a general block diagram of the components of a mobile cellular device 1516 that can perform some of the components shown and described herein. The mobile cellular device 1516 interacts with, or performs some of, and interacts with some of them. Within the device 1516, a communication link 1513 is provided, which enables the handheld device to communicate with other computing devices and, in some embodiments, provides a channel for automatically receiving information, such as by scanning. Examples of the communication link 1513 include those that enable communication via one or more communication protocols, such as a wireless service used to provide cellular access to a network, as well as protocols that provide a local wireless connection to a network.

[0166] In other examples, the application may be received on a removable secure digital (SD) card connected to interface 1515. Interface 1515 and communication link 1513 communicate with processor 1517 (which may also be an embodiment of the processor) along bus 1519, which is also connected to memory 1521 and input / output (I / O) components 1523, as well as clock 1525 and position system 1527.

[0167] In one embodiment, the I / O component 1523 is provided to facilitate input and output operations, and the device 1516 may include input components such as buttons, touch sensors, optical sensors, microphones, touchscreens, proximity sensors, accelerometers, and attitude sensors, as well as output components such as display devices, speakers, and / or printer ports. Other I / O components 1523 can be used in a similar manner.

[0168] Clock 1525, as an example, constitutes a real-time clock component that outputs the time and date. It can also provide timing capabilities for processor 1517.

[0169] Exemplary, the position system 1527 includes a component that outputs the current geographical location of the device 1516. This may include, for example, a Global Positioning System (GPS) receiver, a LORAN system, a dead reckoning system, a cellular triangulation system, or other positioning systems. It may also include mapping or navigation software that generates desired maps, navigation routes, and other geographical functions.

[0170] Memory 1521 stores the operating system 1529, network settings 1531, applications 1533, application configuration settings 1535, data store 1537, communication drivers 1539, and communication configuration settings 1541. Memory 1521 can include all types of tangible volatile and non-volatile computer-readable memory devices. It can also include computer storage media (as described below). Memory 1521 stores computer-readable instructions that, when executed by processor 1517, cause the processor to perform computer implementation steps or functions according to those instructions. Processor 1517 can also be activated by other components to facilitate their functions. Although the physical memory store 1521 is illustrated as part of the device, it is explicitly assumed that cloud computing options are available in which some data and / or processing are performed using remote services.

[0171] Figure 16 shows that the device may be a smartphone 1671. The smartphone 1671 has a touch-sensitive display 1673 that displays icons or tiles or other user input mechanisms 1675. The mechanisms 1675 can be used by the user to run applications, make calls, or perform data transfer operations. Generally, the smartphone 1671 is built on a mobile operating system and provides more advanced computing power and connectivity than a feature phone. Note that other forms of the device are possible.

[0172] However, while Figure 16 illustrates an embodiment in which the device 1600 is a smartphone 1671, it is explicitly assumed that the display may be presented on another computing device.

[0173] Figure 17 shows an example of a computing environment in which elements, or parts thereof (for example), of the systems and methods described herein can be deployed. Referring to Figure 17, an exemplary system for implementing some embodiments includes a general-purpose computing device in the form of a computer 1710. The components of the computer 1710 include, but are not limited to, a processing unit 1720 (which may include a processor), system memory 1730, and a system bus 1721 that connects the system memory to the processing unit 1720. The system bus 1721 may be any of several types of bus structures, such as a memory bus or memory controller, peripheral bus, and local bus using various bus architectures. The memory and programs described with respect to the systems and methods described herein can be deployed in any suitable configuration.

[0174] Computer 1710 typically includes a variety of computer-readable media. Computer-readable media are any available media accessible by computer 1710, and include both volatile / non-volatile media and removable / non-removable media. For example, but without limitation, computer-readable media may include computer storage media and communication media. Computer storage media, unlike modulated data signals or carrier waves, do not include them. This includes both volatile / non-volatile and removable / non-removable hardware storage media implemented in any way or technique for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other media that store desired information and can be accessed by computer 1710. Communication media include any information transmission media that embodies computer-readable instructions, data structures, program modules, or other data within a transport mechanism. The term "modulated data signal" refers to a signal in which one or more of its properties are set or modified in order to encode information within the signal.

[0175] The system memory 1730 includes computer storage media in the form of volatile and / or non-volatile memory, such as read-only memory (ROM) 1731 and random-access memory (RAM) 1732. The basic input / output system (BIOS) 1733 contains basic routines that help transfer information between elements within the computer 1710, is used at startup, etc., and is usually stored in ROM 1731. RAM 1732 is usually immediately accessible by the processing unit 1720 or contains data and / or program modules currently being operated on. As an example, but not limited to, Figure 17 shows an operating system 1734, an application program 1735, other program modules 1736, and program data 1737.

[0176] Computer 1710 may also include other removable / non-removable and volatile / non-volatile computer storage media. For example only, Figure 17 shows a hard disk drive 1741, a non-volatile magnetic disk 1752, an optical disk drive 1755, and a non-volatile optical disk 1756 that read and write non-removable non-volatile magnetic media. The hard disk drive 1741 is typically connected to the system bus 1721 via a non-removable memory interface such as interface 1740, and the optical disk drive 1755 is typically connected to the system bus 1721 via a removable memory interface such as interface 1750.

[0177] Alternatively, or in addition, the functions described herein may be performed, at least in part, by one or more hardware logic components. For example, without limit, exemplary types of usable hardware logic components include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standards (ASSPs), systems-on-a-chip (SOCs), and complex programmable logic devices (CPLDs).

[0178] The drives and associated computer storage media discussed above and illustrated in Figure 17 provide storage for computer-readable instructions, data structures, program modules, and other data for computer 1710. In Figure 17, for example, a hard disk drive 1741 is illustrated as storing the operating system 1744, application programs 1745, other program modules 1746, and program data 1747. Note that these components may be identical or different from the operating system 1734, application programs 1735, other program modules 1736, and program data 1737.

[0179] The user can input commands and information to the computer 1710 through input devices such as a keyboard 1762, a microphone 1763, and a pointing device 1761 such as a mouse, trackball, or touchpad. Other input devices (not shown) may include joysticks, gamepads, satellite receivers, scanners, etc. These and other input devices are often connected to the processing unit 1720 via a user input interface 1760 coupled to the system bus, but may be connected by other interfaces and bus structures. A visual display 1791 or other type of display device is also connected to the system bus 1721 via an interface such as a video interface 1790. In addition to the monitor, the computer may also include other peripheral output devices such as speakers 1797 and a printer 1796, which may be connected via an output peripheral interface 1795.

[0180] Computer 1710 operates in a network environment, connecting to one or more remote computers, such as remote computer 1780, using a logical connection such as a local area network (LAN) or wide area network (WAN). When used in a LAN network environment, computer 1710 connects to LAN 1771 via a network interface or adapter 1770. When used in a WAN network environment, computer 1710 typically includes other means of establishing communication via modem 1772 or WAN 1773 (e.g., the Internet). In a network environment, program modules may be stored in remote memory storage devices. Figure 17 shows, for example, that a remote application program 1785 may reside on remote computer 1780.

[0181] In the detailed description of preferred embodiments of the present invention, reference is made to the accompanying drawings, which illustrate specific embodiments in which the invention may be carried out. The illustrated embodiments are not intended to encompass all embodiments of the invention. It should be understood that other embodiments may be used and structural or logical modifications may be made without departing from the scope of the invention. Accordingly, the following detailed description should not be taken as restrictive, and the scope of the invention is defined by the appended claims.

[0182] Unless otherwise explicitly stated, all numerical values ​​representing characteristic dimensions, quantities, and physical properties used in the specifications and claims should be understood in all cases as being modified by the word "approximately." Therefore, unless otherwise indicated, the numerical parameters described in the specifications and accompanying claims are approximations that may vary depending on the desired properties sought by those skilled in the art utilizing the teachings disclosed herein.

[0183] As used in this specification and the attached claims, the singular forms “a,” “an,” and “the” encompass embodiments having multiple references unless the context explicitly indicates otherwise. As used in this specification and the attached claims, the term “or” is used generally to mean “and / or” unless the context explicitly indicates otherwise.

[0184] When spatial relation terms such as “proximal,” “distal,” “lower,” “upper,” “below,” “downward,” “above,” and “on,” are used herein, without limitation, for convenience to describe the spatial relationship between one element and another. Such spatial relation terms encompass different orientations of the device during use or operation, in addition to the specific orientations shown in the figures and described herein. For example, if an object shown in the figures is turned over or inverted, a part previously described as being below or below other elements would be located above or above those other elements.

[0185] In this specification, where it is stated, for example, that an element, component, or layer “forms a matching interface” or that it is “on,” “connected,” “joined,” “laminated,” or “in contact” with another element, component, or layer, it may be directly on, directly connected, directly joined, directly laminated, or directly in contact with that particular element, component, or layer, or an intervening element, component, or layer may be on, connected, joined, or in contact with that particular element, component, or layer. For example, where it is said that an element, component, or layer is “directly on,” “directly connected,” “directly joined,” or “in direct contact” with another element, there is no intervening element, component, or layer.

[0186] The technologies described herein can be implemented in a wide range of computer devices, including servers, laptop computers, desktop computers, notebook computers, tablet computers, handheld computers, and smartphones. Any component, module, or unit is described to highlight its functional aspects and does not necessarily need to be implemented by different hardware units. The technologies described herein can also be implemented in hardware, software, firmware, or any combination thereof. Any feature described as a module, unit, or component may be implemented together within an integrated logic unit or separately as individual, interoperable logic units. In some cases, various features may be implemented as integrated circuit devices, such as integrated circuit chips or chipsets. Furthermore, numerous independent modules are described throughout this specification, many of which perform their own functions, but all functions of all modules may be integrated into a single module or further divided into additional modules. The modules described herein are described illustratively only for the sake of clarity.

[0187] When implemented in software, these techniques may be implemented at least in part by a computer-readable medium containing instructions that, when executed within a processor, perform one or more of the methods described above. The computer-readable medium may include a tangible computer-readable storage medium and may constitute part of a computer program product that includes packaging materials. The computer-readable storage medium may include random access memory (RAM) such as synchronous dynamic random access memory (SDRAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory, magnetic or optical data storage media, etc. The computer-readable storage medium may also include non-volatile storage devices such as hard disks, magnetic tapes, compact discs (CDs), digital versatile discs (DVDs), Blu-ray discs, holographic data storage media, or other non-volatile storage devices.

[0188] As used herein, the term “processor” may refer to any of the structures described above, or any other structure suitable for implementing the technologies described herein. Furthermore, in some embodiments, the functions described herein may be provided within a dedicated software module or hardware module configured to perform the technologies of this disclosure. Even when implemented in software, these technologies may use a processor to run the software, or hardware such as memory to store the software. In such cases, the computer described herein may define a specific machine capable of performing the specific functions described herein. These technologies may also be implemented entirely within one or more circuits or logic elements, which can also be considered processors.

[0189] example Example 1: Predicting the mixing ratio The two-part adhesive 3M® Scotch-Weld® Epoxy Adhesive DP460 was dispensed through two progressive cavity pumps, one for each part of the adhesive. After pumping, the two parts were combined in a static mixer to mix the adhesives. The sensor described herein was mounted at the end of the static mixer so that the mixed adhesive would pass through the sensor and an electrical signal would be obtained as described above. The sensor collected dielectric constant data at two frequencies, along with the temperature of the adhesive. The following variables were used to create the dataset: flow rate, temperature, moisture content, frequency, dielectric constant, and mixing ratio. Sensor data was collected continuously for approximately two hours.

[0190] The obtained labeled data was split into 70 / 30 training / test datasets. The training dataset was used to train a machine learning model to predict the mixing ratio of the adhesive. The test dataset was then used to score the model, demonstrating that it correctly identified the mixing ratio using only the sensor data. Figure 18 shows how the model predicted the mixing ratio under each of the conditions described above. The line graph shows the electrical signals during the dispensing process. Actual and predicted results are illustrated. In this example, the difference between the actual and predicted mixing ratios was within 3%.

[0191] Example 2: Predicting moisture content The two-part adhesive 3M® Scotch-Weld® Epoxy Adhesive DP460 was dispensed according to Example 1. The output data was relabeled with moisture content as the output variable, and a machine learning model was trained using a 70 / 30 training / test dataset split. Figure 19 shows the model's ability to predict moisture content as the moisture content steadily increases from 0% to 1%.

Claims

1. A sensing system for mixtures, A sensing area containing a mixture, A sensor arranged within the aforementioned sensing area, Printed circuit board (PCB) and, A transmitting electrode configured to generate an electric field, Receiving electrode and, Includes, The transmitting electrode and the receiving electrode are configured to be in direct contact with the mixture. The sensor is configured to detect the electrical parameter values ​​of the mixture, A signal analyzer that determines the water content of the mixture based on the detected electrical parameters, A sensing system equipped with [the following features].

2. The sensing system according to claim 1, further comprising a communication component configured to communicate the detected moisture content.

3. The sensing system according to claim 1 or 2, wherein the signal analysis device includes a machine learning algorithm.

4. The sensing system according to claim 3, wherein the machine learning algorithm determines the moisture content based on a control value.

5. The sensing system according to claim 4, wherein the control value includes the detected electrical parameter and temperature.

6. The sensing system according to claim 4, wherein the material is a mixture and the control value includes the mixing ratio.

7. The sensing system according to claim 4, wherein the control value includes the viscosity of the material.

8. The sensing system according to claim 4, wherein the control value includes the elapsed time, curing state, or curing rate of the material.

9. The sensing system according to any one of claims 1 to 8, wherein the determined moisture content is the predicted moisture content.

10. The sensing system according to any one of claims 1 to 9, wherein the sensor further includes an edge connector, and the signal analyzer is configured to receive the detected electrical parameters using a signal reader that interfaces with the edge connector.

11. The sensing system according to claim 10, wherein the signal analyzer and the signal reader are integrated into a single processing unit.

12. The sensing system according to claim 11, wherein the single processing unit further includes a display device, and the communication component communicates the detected moisture content or the detected electrical parameter value to the display device.

13. The sensing system according to claim 11, wherein the electrical parameter value is a first electrical parameter value, and the processing unit is configured to receive a second detected electrical parameter value.

14. The sensing system according to claim 13, wherein the transmitting electrode is a first transmitting electrode, the receiving electrode is a first receiving electrode, and the sensor further includes a second transmitting electrode and a second receiving electrode, and the second detected electrical parameter is received from the second receiving electrode.

15. The system according to claim 14, wherein the printed circuit board (PCB) includes a first opening and a second opening, the first opening includes a first transmitting electrode and a receiving electrode, the second opening includes a second transmitting electrode and a receiving electrode, and the first detected electrical parameter value is different from the second detected electrical parameter value.

16. The system according to any one of claims 1 to 15, wherein the printed circuit board (PCB) includes a plurality of openings, each opening includes a transmitting unit and a receiving unit, and each opening generates a different sensing electrical parameter value.

17. The system according to claim 16, wherein each of the plurality of openings has a length, width, and thickness.

18. The system according to claim 17, wherein the plurality of openings are arranged in a grid format.

19. The system according to claim 17, wherein the plurality of openings are aligned in a row and each length is parallel to the adjacent length.

20. The system according to claim 17, wherein the plurality of openings are arranged in a row, each opening is spaced apart from the edge connector, the first opening is at a first distance from the edge connector, the second opening is at a second distance from the edge connector, and the second opening is spaced further apart from the edge connector by a length greater than that of the first opening.

21. The system according to claim 17, wherein the sensor includes a temperature sensor.

22. The system according to claim 21, wherein the sensing area is a container for containing the mixture.

23. The sensor according to any one of claims 1 to 22, wherein the detected electrical characteristic value is a current flow measurement or a voltage measurement.

24. The sensor according to any one of claims 1 to 23, wherein the detected electrical characteristic value is impedance.

25. The sensor according to any one of claims 1 to 23, wherein the detected electrical characteristic value is the dielectric constant.

26. The sensor according to claim 21, wherein the temperature sensor is electrically insulated from the material.

27. A method for detecting the moisture content of a material, wherein the method is A step of bringing the material into contact with a sensor, wherein the contact includes direct contact between the sensor and the material, A printed circuit board (PCB) including a first surface and a second surface, wherein the first and second surfaces are separated by a thickness, A sensing region provided on the PCB and configured to detect electrical parameter values, An edge connector configured to communicate signals from the sensing area to a signal processor, The process of bringing the items into contact, A step of receiving an electrical parameter signal from the sensor, wherein the electrical signal indicates the electrical parameter value, A step of detecting the moisture content based on the received signal, wherein the detection is performed in real time by the signal processor, The process of communicating the aforementioned moisture content, A method for detecting moisture content, including the following.

28. The method according to claim 27, wherein the signal processor receives the detected electrical parameter value, detects the moisture content, and communicates the detected material state.

29. The method according to claim 28, wherein the signal processor receives the electrical parameter signal from a receiver coupled to the edge connector.

30. The method according to any one of claims 27 to 29, wherein the sensor is configured to receive a flow of material through a conduit.

31. The method according to any one of claims 27 to 30, wherein the sensor is configured to contact the material while the material is stationary.

32. The method according to any one of claims 27 to 31, wherein the step of detecting the moisture content includes the execution of a machine learning algorithm by the signal processor.

33. The method according to claim 32, wherein the machine learning algorithm detects the moisture content based on a control value.

34. The method according to claim 33, wherein the control value includes the detected electrical parameter and temperature.

35. The method according to claim 33, wherein the material is a mixture and the control value includes the mixing ratio.

36. The method according to claim 33, wherein the control value includes the viscosity of the material.

37. The method according to claim 33, wherein the control value includes the elapsed time, curing state, or curing rate of the material.

38. The method according to any one of claims 27 to 37, wherein the determined moisture content is the predicted moisture content.

39. The method according to any one of claims 27 to 38, wherein the communication step includes the signal processor communicating the detected material state to a graphical user interface generating device that generates a graphical user interface for a device having a screen.

40. The method according to claim 39, wherein the device includes a signal receiver coupled to the edge connector.

41. The method according to claim 39, wherein the apparatus includes the signal processor.

42. The method according to any one of claims 27 to 41, wherein the detected electrical characteristic value is a current flow measurement.

43. The method according to any one of claims 27 to 42, wherein the detected electrical characteristic value is impedance.

44. The method according to any one of claims 27 to 43, wherein the detected electrical characteristic value is the dielectric constant.

45. The method according to any one of claims 27 to 44, wherein the sensing region includes a transmitting electrode and a receiving electrode.

46. The method according to claim 45, wherein the sensing region has a single opening including the transmitting electrode and the receiving electrode.

47. The method according to claim 45, wherein the receiving step includes receiving using a first frequency and receiving using a second frequency.

48. The method according to claim 45, wherein the sensing region includes a second transmitting electrode and a second receiving electrode, the first and second receiving electrodes are decoupled, and the detected electrical characteristic value from the first receiving electrode is different from the detected electrical characteristic value from the second receiving electrode.

49. The method according to any one of claims 27 to 48, further comprising a temperature sensor.

50. The method according to claim 49, wherein the temperature sensor is electrically insulated from the fluid flow.

51. The method according to any one of claims 27 to 50, wherein the sensing area includes a sensing length and a sensing width, and the length of the PCB exceeds twice the sensing length.

52. A moisture content detection system, A sensor configured to contact a material and to detect the electrical parameter values ​​of the material based on the contact, A signal processor that determines the moisture content of the material based on the detected parameter values, A communication component configured to communicate the determined moisture content, Equipped with, A moisture content detection system in which the system operates substantially in real time, the signal processor receives the detected parameter value, determines the moisture content, and communicates the determined moisture content in less than 10 seconds.

53. The moisture content detection system according to claim 52, wherein the signal processor receives the detected parameter value, determines the moisture content, and communicates the determined moisture content in less than one second.

54. The moisture content detection system according to claim 52 or 53, wherein the system includes an apparatus, and the apparatus includes the signal processor and the display device.

55. The moisture content detection system according to claim 54, wherein the device is configured to receive the detected electrical parameter value from a signal reader, and the signal reader is configured to communicate with the sensor.

56. The moisture content detection system according to claim 55, wherein the sensor includes an edge connector and the signal reader accepts the edge connector.

57. The moisture content detection system according to any one of claims 52 to 56, wherein the sensor is in direct contact with the material.

58. The moisture content detection system according to any one of claims 52 to 57, wherein the sensor includes an opening, the opening includes a transmitting electrode and a receiving electrode, the transmitting electrode transmits an electric field, and the receiving electrode receives a signal indicating the detected electrical parameter value.

59. The moisture content detection system according to claim 58, wherein the sensor includes an edge connector at its first end, the opening is a first opening, and the sensor includes a second opening, the first opening being closer to the edge connector than to the second opening.

60. The moisture content detection system according to claim 59, wherein the first opening is arranged in the same straight line as the second opening.

61. The moisture content detection system according to claim 58, wherein the sensor includes an edge connector at its first end, the opening is a first opening, the sensor includes a second opening, the first opening is spaced a first distance from the edge connector, the second opening is spaced a second distance from the edge connector, and the first and second distances are substantially equivalent.

62. The moisture content detection system according to claim 58, wherein the opening is a first opening and the sensor includes four openings.

63. The moisture content detection system according to claim 62, wherein the four openings are arranged parallel to each other along the length or width of the transmitting electrode.

64. The moisture content detection system according to claim 62, wherein the four openings form a grid.

65. The moisture content detection system according to any one of claims 52 to 64, wherein the signal analysis device includes a machine learning algorithm.

66. The moisture content detection system according to claim 65, wherein the machine learning algorithm determines the moisture content based on a control value.

67. The moisture content detection system according to claim 66, wherein the control value includes the detected electrical parameter and temperature.

68. The moisture content detection system according to claim 66, wherein the material is a mixture and the control value includes the mixing ratio.

69. The moisture content detection system according to claim 66, wherein the control value includes the viscosity of the material.

70. The moisture content detection system according to claim 66, wherein the control value includes the elapsed time, curing state, or curing rate of the material.

71. A moisture content detection system according to any one of claims 52 to 70, wherein the determined moisture content is the predicted moisture content.

72. The moisture content detection system according to any one of claims 52 to 71, wherein the electrical parameter value is a first electrical parameter value, and the processing unit is configured to receive a second detected electrical parameter value.

73. The moisture content detection system according to any one of claims 52 to 72, wherein the PCB includes a plurality of openings, each opening includes a transmitting unit and a receiving unit, and each opening generates a different detection electrical parameter value.

74. The moisture content detection system according to any one of claims 52 to 73, wherein each of the plurality of openings has a length, width, and thickness.

75. The moisture content detection system according to any one of claims 52 to 74, wherein the sensor includes a temperature sensor.

76. The moisture content detection system according to any one of claims 52 to 75, wherein the sensing area is a container for containing the mixture.

77. The moisture content detection system according to any one of claims 52 to 76, wherein the detected electrical characteristic value is a current flow measurement.

78. The moisture content detection system according to any one of claims 52 to 77, wherein the detected electrical characteristic value is impedance.

79. The moisture content detection system according to any one of claims 52 to 78, wherein the detected electrical characteristic value is the dielectric constant.

80. The moisture content detection system according to claim 79, wherein the temperature sensor is electrically insulated from the material.

81. A material quality control system, A sensor configured to directly contact the material, wherein the sensor includes a transmitting electrode and a receiving electrode, and the sensor is configured to detect an electrical parameter when a current or voltage is applied to the transmitting electrode, A signal processing system including a processing circuit, wherein the processing circuit is Upon receiving the detected electrical parameter value, Based on the detected electrical parameter values ​​and known control variables, a machine learning algorithm is activated to detect the state of the material. A signal processing system configured as follows: A communication component configured to communicate the detected state to a device having a display device, A quality control system equipped with the following features.

82. The system according to claim 81, wherein the device includes the signal processing system.

83. The system according to claim 81, wherein the device includes a signal reader that interfaces with the sensor.

84. The system according to claim 83, wherein the signal reader accepts the edge connector of the sensor.

85. The system according to any one of claims 81 to 84, wherein the sensor includes a PCB.

86. The system according to claim 85, wherein the PCB includes an opening, and the opening includes the transmitting electrode and the receiving electrode.

87. The system according to claim 86, wherein the opening is a first opening and includes two openings in which the PCBs are spaced apart.

88. The system according to claim 87, wherein the PCB has a length and a width, the length being greater than the width, and the two openings being parallel to each other and aligned along the length.

89. The system according to claim 87, wherein the PCB has a length and a width, the length being greater than the width, and the two openings being parallel to each other and aligned along the width.

90. The system according to claim 86, wherein the PCB includes four openings, the openings being arranged in a grid pattern.

91. The system according to any one of claims 81 to 90, wherein the detected state is the moisture content of the material.

92. The system according to claim 91, wherein the known control variable includes temperature.

93. The system according to any one of claims 81 to 92, wherein the known control variable is flow rate, viscosity of the material, elapsed time of the material, curing rate of the material, chemical composition of the material, curing state, applied voltage, frequency of applied voltage, or temperature.

94. The system according to any one of claims 81 to 93, wherein the material is a mixture and the detected state is a mixing ratio.

95. The system according to claim 94, wherein the known control variable is flow rate, viscosity of the material, elapsed time of the material, curing rate of the material, applied voltage, frequency of applied voltage, or temperature.

96. The material is a mixture, and the processing circuit further The system is configured to determine the mixing ratio based on the detected electrical parameters. The system according to any one of claims 81 to 95, wherein the detected state is the water content of the mixture, and the machine learning algorithm detects the water content based on the mixing ratio.

97. The system according to claim 96, wherein the known control variable is flow rate, viscosity of the material, elapsed time of the material, curing rate of the material, applied voltage, frequency of applied voltage, or temperature.

98. The system according to any one of claims 81 to 98, wherein the control value is detected by a second sensor.

99. The system according to claim 98, wherein the second sensor is a temperature sensor.

100. The system according to claim 99, wherein the temperature sensor is electrically insulated from the material.

101. The system according to any one of claims 81 to 100, wherein the detected state is the predicted state of the material.

102. The system according to any one of claims 81 to 101, wherein the electrical parameter value is a first electrical parameter value, the signal processing system is configured to receive a second detected electrical parameter value, and the detected state is based on the first and second detected electrical parameter values.

103. The system according to any one of claims 81 to 102, wherein the detected electrical characteristic value is a current flow measurement.

104. The system according to any one of claims 81 to 103, wherein the detected electrical characteristic value is impedance.

105. The system according to any one of claims 81 to 104, wherein the detected electrical characteristic value is the dielectric constant.

Images