Fluid sensing device and control system

Abstract

A system for determining characteristics of milk or other fluids flowing through an in-line sensor system includes a laser engine that emits laser radiation through the milk as it flows through the sensing system, a laser detector that receives the laser after it passes through the milk and generates corresponding laser readings, one or more processors, and computer memory storing computer-readable instructions that cause the processor to receive laser readings from the laser detector, identify spectral data representative of physical properties of the fluid from the laser readings as the fluid flows through the sensing system, and determine one or more fluid measurements of corresponding one or more components of the fluid using the spectral data and reference data defining one or more reference spectra for each possible component of the fluid.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 414,065, filed October 7, 2022, the disclosure of which is incorporated herein by reference in its entirety. [Technical Field]

[0002] This disclosure describes automatic control and sensing using lasers. [Background technology]

[0003] A tunable laser is a laser with an operating wavelength that can be changed in a controlled manner. Nearly all laser gain media allow for small shifts in output wavelength, and some allow for continuous tuning over a wider wavelength range. Gas, liquid, and solid-state lasers exist. Some examples include excimer lasers, gas lasers, dye lasers, solid-state lasers, semiconductor crystal and diode lasers, and free-electron lasers.

[0004] An industrial control system (ICS) includes electronic control systems and associated instrumentation that can be used to control industrial processes. Control systems vary in size from a few modular panel-mounted controllers to large interconnected, interactive distributed control systems. Some control systems are implemented with supervisory control and data acquisition (SCADA) and / or programmable logic controllers (PLCs).

[0005] Dairy farms involve lactating mammals, such as cows, goats, sheep, etc., for the production of milk, which is then used to make a variety of other dairy products, such as fluid milk, anhydrous milkfat, whole milk powder, lactose, cheese, butter, yogurt, cream, and kefir. Milk and milk products are an essential part of the global food sector, and dairy farms are responsible for producing the raw materials. Production efficiency, sustainability, and ultimately profitability depend on several factors, but key are efficient herd health and herd culling management. Summary of the Invention

[0006] In-line electro-optical sensors for real-time composition analysis may be incorporated into a variety of industrial, agricultural, and biological environments where there is flow through liquid, solid, or gas phase materials. In particular, there is in-line real-time monitoring of milk composition in individual animal milking lines, combined with methods that use the measured composition data to aggregate and build short-term and long-term data trends and build process-wide optimization models to enhance farm efficiency in terms of production, minimal animal rest periods, and controlled herd culling. Some embodiments of the present invention use electro-optical in-line sensors integrated into the milking line to monitor and aggregate real-time or near-real-time milk composition data of every individual milking process for every animal in the herd. The data aggregated over time can be used by control systems to proactively provide rapid warning indications of animal health, resulting in opportunities for timely treatment and potentially significant savings in both medication and animal rest periods. Trending the data, such as post-treatment animal production and milk composition, can provide information about the effectiveness of the treatment, which in turn helps farmers determine the most efficient way to treat any one animal's health problem compared to other available pharmaceutical options. Additionally, monitoring individual animals within a herd provides direct data for herd selection based on specific desirable traits, such as high protein and / or fat, disease resistance, etc., resulting in controlled processes, improved efficiency, as well as business profitability.

[0007] One or more computer systems may be configured to perform specific operations or actions by installing software, firmware, hardware, or a combination thereof on the system that, when operational, causes the system to perform a process. One or more computer programs may be configured to perform specific operations or actions by including instructions that, when executed by a data processing device, cause the device to perform the operation. One general aspect includes a sensing system for detecting a physical property of a fluid. The system includes a laser engine configured to emit spectrally tunable laser radiation through the fluid as the fluid flows through the sensing system. The sensing system also includes a laser detector configured to receive the laser radiation after it passes through the fluid and generate a corresponding laser reading. The system also includes one or more processors. The system also includes a computer memory storing computer-readable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations including receiving laser readings from the laser detector, identifying from the laser readings spectral data representative of physical properties of the fluid as it flows through the sensing system, and determining one or more fluid measurements of corresponding one or more components of the fluid using the spectral data and reference data defining one or more reference spectra for each possible component of the fluid. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the operations of the method. Implementations may include one or more of the following features. The system may further include a housing housing the laser engine, the laser detector, one or more processors, and a computer memory. The system may further include a network interface, and the operations may further include transmitting the one or more fluid measurements via the network interface. The system may further include a housing housing the laser engine and the laser detector, and one or more computing devices including one or more processors and a computer memory. The system may further include a fluid channel between the laser engine and the laser detector through which a fluid flows, the laser engine and the laser detector being fixedly held at least partially within the fluid channel such that a portion of the fluid flows between the laser engine and the laser detector. The laser engine and the laser detector are fixedly held apart by a distance of less than 10 mm. The laser engine is coupled to a photon-transparent sheath configured to prevent contact of the laser engine by the fluid. The system may further include a contact sensor fixedly held at least partially within the fluid channel, an optical emitter fixedly held at least partially within the fluid channel, and a color sensor fixedly held at least partially within the fluid channel. The system may further include a contact sensor configured to contact the fluid as it flows through the sensing system and detect one or more contact events of the fluid to generate corresponding contact readings, and determining the one or more fluid measurements may further include using the contact readings. The system may further include a light emitter configured to emit incoherent light and a color sensor configured to receive the incoherent light after it passes through the fluid and generate corresponding incoherent readings, and determining the one or more fluid measurements may further include using the incoherent readings. The system may further include an automated valve configured to selectively direct fluid flow to the plurality of output channels, and operating may include actuating the automated valve based on at least one of the fluid measurements.Actuating the automatic valve may include operating the automatic valve to direct the fluid to a waste tank in response to determining that the contaminant fluid measurement value is greater than a threshold value. Actuating the automatic valve based on at least one of the fluid measurements may include operating the automatic valve to direct the fluid to a waste tank before the contaminated fluid reaches the automatic valve. The fluid is raw milk obtained from livestock. The laser engine may include a III-V semiconductor-based laser. The III-V semiconductor-based laser is configured to emit light through an optical interface, and a laser detector is disposed opposite the optical interface. The III-V semiconductor-based laser is a tunable laser. The III-V semiconductor-based laser is configured to emit light over a range of wavelengths through the optical interface. The laser engine may be a III-V / IV semiconductor-based laser. The III-V / IV semiconductor-based laser is configured to emit light through an optical interface, and a laser detector is disposed opposite the optical interface. The III-V / IV semiconductor-based laser is a tunable laser. The III-V / IV semiconductor-based laser is configured to emit light over a range of wavelengths through an optical interface. The optical interface is positioned in a flow path of a fluid flowing through the sensing system. The optical interface includes a tube extending from the laser engine into a channel through which the fluid flows through the sensing system, the tube extending toward a laser detector, the distance from the distal end of the glass tube to the surface of the laser detector being between 0.5 mm and 10 mm, and the tube including an optically transparent portion. The fluid measurements are measurements of each of the components at a particular time, and the operations may further include aggregating the fluid measurements for a single milking session into a set reflecting changes to the fluid measurements throughout the milking session. The operations may further include storing the fluid measurements for a single milking session indexed by a unique identifier for the single milking session for each individual animal.The operations may further include storing fluid measurements from one milking session with other fluid measurements from other milking sessions to generate historical data for a particular animal. The operations may further include generating a herd-specific index for multiple animals using the historical data for each animal in the herd. The operations may further include combining the herd-specific index with third-party data for at least one of the following groups, which may include: 1) herd health data, 2) herd culling data, and 3) herd management data. The generation of the individual or herd-specific index includes the use of third-party data and the historical data for each animal in the herd. The fluid physical properties include fluid composition, electrical properties, temperature, flow, color, quantified constituent concentration levels, and the presence or absence of one or more constituents. Implementations of the described techniques may include hardware, methods or processes, or computer software on a computer-accessible medium.

[0008] Other features, aspects, and potential advantages will be apparent from the accompanying detailed description and drawings. [Brief explanation of the drawings]

[0009] [Figure 1A] FIG. 1A shows an example of a system for obtaining milk from livestock using an electro-optical in-line sensor with a tunable laser spectral sensor. [Figure 1B] FIG. 1B shows an example of a system for obtaining milk from livestock using an electro-optical in-line sensor with a tunable laser spectral sensor. [Figure 2A] FIG. 2A shows an example of a device for sensing a fluid phenomenon. [Figure 2B] FIG. 2B shows an example of a device for sensing a fluid phenomenon. [Figure 3A] FIG. 3A shows an example of a device for sensing a fluid phenomenon. [Figure 3B] FIG. 3B shows an example of a device for sensing a fluid phenomenon. [Figure 4] FIG. 4 shows an example of a fluid channel. [Figure 5] FIG. 5 shows a hardware architecture that may be used. [Figure 6] FIG. 6 illustrates the process for generating a unique animal's milk composition record. [Figure 7] Figure 7 shows the model. [Figure 8A] FIG. 8A shows the generation of data from raw sensor data. [Figure 8B] FIG. 8B shows the generation of data from raw sensor data. [Figure 8C] FIG. 8C shows the generation of data from raw sensor data. [Figure 8D] FIG. 8D shows the generation of data from raw sensor data. [Figure 9] FIG. 9 is a swimlane diagram of an example process for sensing a fluid. [Figure 10] FIG. 10 shows data illustrating the absorption spectra of milk measured between the optical interface and the laser detector at various transmission path distances. [Figure 11] FIG. 11 shows an example of data that can be used to sense fluids and control automated devices. [Figure 12] FIG. 12 shows a swim-lane diagram of an example process for detecting available fluids in a milk acquisition system. [Figure 13] FIG. 13 shows a flow chart of an example process for diverting contaminated fluid from a storage tank to a waste tank. [Figure 14] FIG. 14 is a schematic diagram illustrating an example of a computing device and a mobile computing device.

[0010] Reference numbers in the various figures indicate like elements. DETAILED DESCRIPTION OF THE INVENTION

[0011] In-line sensors can use one or more sensors to detect one or more properties of a fluid as it passes through the in-line sensor. For example, an electro-optical sensor in an in-line sensor can be used to detect properties of raw milk as it is obtained from cows or other livestock. Based on the sensing performed by the in-line sensor, one or more automated devices can be activated under computer control. For example, an automatic valve in a fluid line can direct the flow of raw milk to a waste tank if a contaminant (e.g., blood, elevated somatic cell count, etc.) is identified in the milk. Information from an in-line sensor can also be used to perform compositional analysis to determine other properties of the milk. The determined properties can be used to make decisions about ongoing milk collections (e.g., high-fat milk being directed to a particular tank, identifying misplaced milking equipment, etc.) and can be aggregated with other historical and third-party information to identify trends regarding animals within a herd and the production of a particular farm. Additionally, the aggregated data can provide information about herd health and production that can be used to make decisions about herd health, culling, management, and care.

[0012] As shown in FIG. 1A, a typical dairy farm consists of tens to thousands of animals, resulting in a large number of animals being milked simultaneously several times a day (usually 2-3 times a day). A milking station designed to milk N cows simultaneously may have N, 2N, or 4N sensors (depending on the configuration of the milking system, the milking system may collect milk from all four udder sections simultaneously, two at a time, or all four). As shown in FIG. 5, data from each electro-optical sensor 2, including milk composition, conductivity, temperature, color, flow, etc., is sent to a local network device, router 5. Router 5 is hardware (or software, such as a virtual machine) installed locally on each farm that collects data and sends it to a local data server 6. Server 6 includes an operating system 61, such as, but not limited to, Ubuntu®. The operating system includes a solution environment 62, a virtualized software environment with module runtime functionality 620. The module runtime is a platform for running different modules. Modules are custom code applications designed to perform specific tasks, such as spectrum capture 621, which connects to sensors, captures data strings, and performs data pre-processing such as jitter correction, filtering, averaging, valid data selection, and data conversion to calibrated units (e.g., time to frequency / wavelength conversion). Other specific tasks include local data storage 622, where data from sensors is saved, such as file reports, logs, processed data, partially processed data, data from third-party sources 8, and data processing 623, where live measurements from (electrical, optical, spectral) sensors are fed into a data model and aggregated with farm data, such as unique animal IDs, milking times, historical data, etc., to produce a record of the milk composition of a unique animal.Further modules include Sensor Management 624, which is responsible for connecting to each of the sensors and monitoring sensor status, performance, and configuration. Another module is Telemetry 625, which is responsible for sending and receiving data to and from the cloud infrastructure of the sensor system 7. The cloud infrastructure of the described embodiment acts as a centralized infrastructure and provides resources, components, and solutions. For example, there are components such as Data Ingestion 71, which is responsible for receiving data from servers of other farms, third-party data sources, and storing them neatly in the cloud data warehouse 72 as a database and / or file storage 74, typically used for large binary files, such as spectra, images, external data inputs, etc. Data Processing and Computation 76 provides cloud-based computing resources for various stages of data processing, which depend on the configuration of the local farm server 6 and the level of local data processed on the farm. The processed data is then stored again in the data warehouse 72 and / or file storage 74, and data subsets are then provided to user applications 73, which in turn provide the data to end users 11 in the form of web applications, native mobile applications, desktop applications, etc. The data provided to the end users is adapted to the user in terms of information content, which can be thought of as various selectable views, or interface layers. Interface layers may, for example, be a farmer interface that focuses more on herd productivity, farm profitability, etc.; a veterinary interface that focuses more on individual animal health, treatment effects, historical data, etc.; a livestock technical interface that focuses on herd culling, etc., allowing the farmer to determine the desired traits and select the correct individual animals for herd culling to enhance desired characteristics, such as higher fat and / or protein content, disease resistance, or milk yield, etc.The interfaces are interconnected and allow detailed monitoring across multiple farms, from individual animals to herds, and the application of best practices in terms of treatment, culling, nutrition, etc. based on actual data. Provisioning and Management 75 acts as an administrator console for monitoring and configuring the remote farm's servers, sensors, and the entire Internet-of-Things (IoT) infrastructure.

[0013] Real-time in-line milk monitoring in dairy farms poses a rather challenging environment because the milk flowing through the milking line is not continuous, is usually turbulent, and contains air bubbles. Furthermore, the milk composition changes dynamically during the milking cycle, and the sensors must keep up with these changes. For spectral measurement of milk composition, spectrally tunable laser radiation of known intensity and wavelength profile is transmitted through the milk flow in a sensor flow-through compartment via an optical interface and collected by detectors via their respective optical interfaces. The detectors are photodetectors, thus converting the optical signal into electricity. The detectors record a time-domain signal (photovoltage or photocurrent). This signal must be processed into a usable absorbance spectrum that can be used by further data processing to convert the spectral data into component concentration levels, or in other words, composition. The process of preprocessing the raw data is called spectral capture and is shown at 621 in Figure 5. However, in certain configurations, this module may also be executed locally on the CPU module of each sensor (e.g., processor 212 shown in FIG. 2A), or in yet another scenario, on the data processing unit 76 of the cloud environment as per FIG. 5.

[0014] 6 shows the spectrum capture module 621 in more detail. Here, the electrical signal from the detector 204 is first amplified by an amplifier 6211, and the current signal is converted to a voltage. The amplified signal is then sent to an analog-to-digital converter (ADC) 6212 and digitized. The signal from the detector, amplifier, and ADC is a time signal, i.e., a function of intensity over time. This signal is then sent to a processing device 6214. The processing device 6214 may be a central processing unit (CPU), a microcontroller unit (MCU), a field programmable gate array (FPGA), a complex programmable logic device (CPLD), or a combination thereof, and performs further signal processing such as jitter correction, filtering, selection of relevant data based on predefined rules, averaging, and finally time-to-wavelength / frequency conversion. This data is then sent via a data link 6215 to the data processing module 622, which is essentially an Ethernet cable. The conversion of time to wavelength always requires knowledge of the absolute wavelength emitted by the laser through the milk, which is used for the conversion. Architectures for frequency tunable lasers that track and determine absolute wavelength are described in US1177630 and US11202453.

[0015] 1B shows an example of a system for obtaining milk from livestock using an electro-optical in-line sensor with a tunable laser spectral sensor. In system 100, cows 102a-d are milked to obtain milk. The milk is conveyed through an in-line sensor 104 in a pipe 106. Depending on the nature of the milk detected by in-line sensor 104, for example, the presence or absence of contaminants such as blood in the milk, a corresponding valve 108 conveys the milk to a collection tank 110 or a waste tank 112.

[0016] Cows 102a-d represent milking lines through which milk from one or more farm animals flows during the milking process. A typical dairy farm consists of tens to hundreds of animals, resulting in a large number of animals being milked simultaneously several times a day (usually 2-3 times a day). A milking station designed to milk N cows simultaneously will have N, 2N, or 4N sensors (depending on the configuration of the milking system, the milking system can collect milk from all four udder sections simultaneously, two at a time, or all four). Milking lines can be connected to the animals manually or via a milking robot. In-line sensors 104 can be integrated into the milking lines or added retroactively to the milking lines between the cows 102a-d and the collection tank 110 and waste tank 112. The in-line sensors 104 can have one or more adapters that facilitate the mechanical connection between the milking lines and the in-line sensors 104.

[0017] Data from the in-line sensor 104 is collected across one or more data networks by one or more computing devices 114, such as cloud services or server processes running on physical or virtual servers. The device 114 can transmit data or graphical elements across a network, such as the Internet 116, for display on a user device 118, store the data for long-term storage, or use the data for other automated processes that translate computer-controlled hardware into computer-readable instructions based on the data. The in-line sensor 104 provides electro-optical milk composition sensing in combination with additional physical parameters. As the milk flows through the in-line sensor 104, the in-line sensor 104 performs analysis (e.g., composition, color, electrical parameters) using optical spectroscopic sensing, color sensing, electrical conductivity and temperature monitoring, and milk volume and milk flow rate measurements. Although a single device 114 is shown, the system 100 can use multiple devices operated as the device 114. For example, one or more local servers may be operated at the milking location, one or more remote servers may be operated in data centers in various geographic locations, etc. In some embodiments, the local server receives data, pre-processes the data received from the inline sensors, and transmits the pre-processed data to an external server (e.g., a cloud-based server). The data is further processed on the external server, which provides feedback to specific farms based on the data and uses the data to aggregate with additional historical or third-party data to build data models as described further below. The herd data models are used to understand herd health and inform decisions about herd culling, veterinary care, and herd economic management. Additionally, data processing on the external server can determine recommendations to farmers based on the output of the herd data models (e.g., nutrition, treatment effectiveness, disease or health trends, and culling recommendations).

[0018] The system 100 may enable automated data collection and control of hardware related to the milk acquisition process. For example, an in-line sensor 104 may detect milk (or another related fluid) as it is transported through a pipe. As will be appreciated, the milk spends a short time (e.g., a few seconds) in transit between the in-line sensor 104 and the valve 108. In a time shorter than the time it takes to transport the milk that distance, the in-line sensor 104, possibly in cooperation with a computing device 114, can analyze the milk's composition. As the milk flows past the sensor, light from the sensor—tunable laser radiation from the spectral sensor, continuous RGB light from the color sensor—shines through the flowing milk. The spectral sensor responsible for spectroscopic milk composition analysis records the real-time milk absorption spectrum, which is later reconstructed into real-time concentration data. When the milk composition is not constant over the milking cycle, in many cases, such as in farm economic (yield) management, the real-time concentration level is averaged across a single milking event to provide a value for the average concentration level of milk recovered from the cow. At the same time, high-resolution temporal data of concentration levels and other physical parameters such as temperature, conductivity, contaminants, etc., can reveal valuable information, such as improper attachment of the milking system in the event of a sudden blood or other contaminant leak. Both high-resolution temporal and average data are collected and stored for further processing and aggregation. In case of contaminant detection, the system 100 can divert the contaminated milk to a waste tank 112 to prevent the contaminant from reaching the collection tank 110 and contaminating already collected milk, and / or simply activate the corresponding valve 108 in time to indicate the danger to the operator and suggest the action that needs to be taken. This is particularly beneficial when the contaminant is difficult or impossible to detect by human inspection and when the control of the valve needs to be faster than human reaction is possible.In some instances, the milk travels through the pipe 106 in less than two seconds, which is sufficient for automated analysis and operation, but less than it would take a human to notice blood contaminating the milk, much less the additional time required to reach in and switch a valve to divert the contaminated milk.

[0019] The ability to detect contaminants in flowing milk in real time allows automated systems to respond immediately to contaminated milk and provide timely warnings to non-automated systems. If left undetected, a blood leak into the milking line caused by one animal could contaminate the entire collection tank, resulting in loss of milk and farm revenue. Timely detection can automatically shut off or divert milk from the bleeding animal from the tank to waste without affecting milk collected from other farm animals. In farms where operation is manual, the sensor may provide an immediate sound or light signal for the operator so that milk from the bleeding animal does not reach the milk collection tank.

[0020] Similarly, the use of an in-line sensor that can detect the properties and contaminants in the flowing milk can be used to identify contaminants as they begin to contaminate, as opposed to sensors that require test samples to be taken and separated from the normal flow. For example, consider a system that draws a test sample from a pipe every five seconds. During this five-second window, a new contaminant may appear in the milk flow and take four seconds to reach the valve, longer than the two seconds required. In this way, the in-line sensor 104 can provide continuous sensing and faster automation than periodic sensing, protecting milk already collected in the recovery tank 110 in scenarios where periodic sampling introduces contaminants into the recovery tank, contaminating all collected milk and necessitating the disposal of the contaminated milk in the recovery tank, resulting in overall lower efficiency due to avoidable wasted materials. Furthermore, periodic sampling is labor-intensive and wasteful because it is an offline measurement, as opposed to optical sensing, which is not intrusive in-line.

[0021] 2A and 2B show examples of devices for detecting fluid phenomena. In FIG. 2B, a schematic diagram of an in-line sensor 104 is shown. As shown, the in-line sensor 104 comprises a fluid channel 106 that may couple with a pipe 106, allowing fluid to pass through the in-line sensor 104. The channel 106 may comprise a coupler 200 (e.g., a threaded or tapped connector) for connecting the pipe 106 in-line with a region of another pipe. A fluid (such as milk) flows through the in-line sensor 104. As will be appreciated, there may be a certain amount of turbulence in the flow.

[0022] The in-line sensor 104 comprises several sensing elements that are used to detect physical phenomena of the fluid, such as optical spectrum in transmittance, reflectance, temperature, electrical properties, flow velocity, and the like.

[0023] The laser engine 202 emits laser radiation through the fluid as it flows through the sensing system. The laser engine 202 may comprise a spectrally tunable semiconductor laser or laser array, a semiconductor laser based on an external cavity spectrally tunable laser, and / or a hybrid III-V / IV semiconductor-based laser or laser-on-a-chip spectrometer. The laser engine 202 is coupled to a probe or sheath (described below in FIGS. 3 and 4 ) that forms an optical interface. The optical interface directs the light emitted by the laser to illuminate the milk flowing past the sheath or probe. The optical interface may be an optical fiber, an optical lens, a system of lenses and optical mirrors, an optical window, a hollow tube with an optical window, a glass tube, or a combination thereof. The laser engine 202 provides a laser beam that passes through the optical interface. The laser beam may be collimated, focused, or diverged, depending on the optical design of the system. One example of a laser engine may be an external cavity tunable laser based on III-V semiconductors, such as a gallium antimonide gain chip emitting in the 1900-2400 nm band, which encompasses the molecular absorption spectra of lactose, milk fat, and milk proteins. Another example of a laser engine may be a hybrid gallium antimonide and group IV semiconductor integrated optical circuit-on-a-chip based laser spectrometer, consisting of one or more widely tunable hybrid III-V / IV lasers or laser arrays. For liquid substances other than milk, other III-V semiconductor-based tunable lasers or laser arrays may be considered, depending on where the absorption characteristics of the target analytes lie in the electromagnetic spectrum. For example, gallium arsenide may be the material platform of choice for wavelengths from 800 to 1100 nm, indium phosphide from 1300 to 1700 nm, and GaSb for wavelengths above 1700 nm.

[0024] In some embodiments, the laser engine 202 is a solid-state laser-based device with a solid-state gain medium, based on a laser that emits widely tunable light. In some embodiments, the laser engine also includes a wavelength-shift tracking device for tracking the wavelength shift of the emitted light. In some embodiments, the laser engine also includes an internal absolute wavelength reference with a known calibrated range (e.g., an absolute wavelength etalon). In some embodiments, the laser engine 202 performs a wavelength sweep and uses the wavelength-shift tracking device in combination with the absolute wavelength reference to determine the absolute wavelength of the emitted range during the sweep and calibrate the emitted range. This wavelength calibration may also include additional components of the laser engine 202, including optical elements such as reflectors, mirrors, prisms, etc., that enable tunable laser emission, pointing, and spatial stability of the output beam, if required for the application. In some embodiments, the laser engine 202 provides internal wavelength calibration for spectroscopic measurements. Further details of laser engine components, laser engine manufacturing, and wavelength calibration hardware that may be used in the sensor systems described herein can be found in U.S. patent application Ser. No. 16 / 609,355, filed May 21, 2018, U.S. patent application Ser. No. 16 / 965,867, filed January 31, 2019, and now-issued U.S. Patent No. 11,177,630, the contents of which are incorporated herein by reference. The laser engine provides tunable laser radiation whose internal wavelength calibration ensures that it is always known during a spectral sweep across the laser output bandwidth. This information is electronically read and used for time-to-frequency conversion, i.e., time-to-frequency conversion. After pre-processing such as jitter correction, filtering, averaging, and time-to-frequency conversion, these spectra can be used for spectroscopic composition analysis by data algorithms.

[0025] Light emitted from the laser engine 202 and transmitted through the optical interface illuminates the milk flowing past the probe or sheath and is collected by a laser detector 204 located in the channel opposite the optical interface. The laser detector 204 receives the laser radiation after it passes through the fluid and generates a corresponding laser reading. The laser detector 204 may include a photodiode, a photodetector, or a photodiode or photodetector array, using an appropriate optical interface, which may be an optical window, lens, fiber, or the like. The photodetector or photodiode is typically a semiconductor-based component selected to be spectrally sensitive to the appropriate range of laser radiation emitted by the laser engine. In some embodiments, the detector is an AlGaInAsSb / GaSb-based PIN, pBp, nBn, or superlattice-based detector, particularly for wavelengths greater than 1700 nm. The detector may also be an extended GaInAs / InP photodetector. At shorter wavelengths, Si, GaInAs / InP, or Ge may also be used.

[0026] The distance between the end of the optical interface and the laser detector 204 defines the optical path and is selected to give an optimal signal-to-noise ratio, which depends on the optical configuration (i.e., the spectral range of the laser, the laser power, the properties of the fluid being measured (e.g., absorption coefficient) and the sensitivity of the detector, etc.). As explained below in Figures 3 and 4, the optical path is defined such that, as the milk flows through the pipe, the laser optical interface 300 and the detector optical interface define a gap of known dimension (the thickness of the liquid it is passing through, in particular the milk) through which the light passes before being detected by the detector. The interface is designed so that its mechanical implementation does not obstruct the flow of milk, or obstructs it minimally. For example, it can be realized as a glass tube, a hollow tube with a window, or other variations of a lens.

[0027] The use of a laser engine 202 with an optical interface positioned within the milk's flow path through the channel may require periodic cleaning of the optical interface to ensure light transmittance through the optical interface. Fat residue from the milk may be left behind on the optical interface (as well as the sensor, described below). The buildup of fat residue may, in some cases, be monitored with a sensor and removed or offset in the collected data. The sensor system is sealed and resistant to milking system cleaning cycles, including high temperature cycles of water, low pH, and high pH cleaning agents. Additionally, the system may include a warning when residue left behind on the optical interface reaches a threshold that threatens the accuracy of the measurement. In some embodiments, one or more sensor systems, including redundant laser engines, laser detectors, and other sensors, may be included inline to ensure accurate readings and measurements.

[0028] The contact sensor group 206 contacts the fluid as it flows through the in-line sensor 104 and detects one or more contact events of the fluid to generate a corresponding contact reading. The contact sensor group 206 may include a temperature sensor for measuring the temperature of the milk and / or a pair or array of electrodes for measuring the electrical conductivity of the milk. The value and change in electrical conductivity correlates with the concentration of somatic cells present in the milk flowing through the sensor. Furthermore, the conductivity, temperature, or a combination of both may be used to evaluate the flow of milk past the sensor and may be utilized by the laser engine 202 for quantitative assessment of the constituent concentration levels of the milk. In some embodiments, the contact sensor group 206 includes multiple separate sensors spaced apart from one another within or adjacent to the fluid channel. In other embodiments, the contact sensor group 206 includes a single sensor capable of detecting multiple contact events, or alternatively includes multiple sensors positioned adjacent to one another.

[0029] An optical emitter within the color sensor 208 emits incoherent light into the fluid, and reflected light from the fluid is collected by RGB detectors forming the color sensor. The color sensor 208 can receive the incoherent light after it passes through the fluid to generate a corresponding incoherent reading. The optical emitter of 208 can emit broadband light into the milk flowing past the emitter. The light is reflected from the milk flow and collected by the color sensor 208. The color sensor 208 can include an optical window, a wavelength filter, an optical lens, an optical prism, a diffraction grating, or a combination thereof. The presence of blood in the milk affects the color of the milk, which in turn changes the reflectance spectrum of the light, which is detected by the color sensor 208. In some embodiments, milk fat content may be identified by changes in reflectance range, particularly color, as collected by the color sensor 208, which may be used in combination with spectral engine data or alone to assess milk fat levels.

[0030] The in-line sensor 104 may include computing hardware such as one or more processors 212, memory 214, and other electronic components 216. Examples of processors, memories, and other electronic components are described in the detailed description below, for example, with respect to FIG.

[0031] 2B shows another example of an in-line sensor 218. In an in-line sensor, some or all of the computations performed by the processor and memory are performed by a computing device 220, which may include one or more servers, desktop computers, etc.

[0032] 3A and 3B show an example of an in-line sensor 104 for detecting a phenomenon in a fluid. Shown here are an isometric view and a cross-sectional view of the in-line sensor 104. The housing 302 shown houses components (shown in FIG. 2B) including a laser engine 202, a laser detector 204, a contact sensor group 206, an optical emitter, a color sensor 208, and electronic components such as a processor 212, memory, and other electronic components 216.

[0033] In cross section, contact sensor group 206, laser engine 202, laser detector 204, and light emitter and color sensor 208 are shown. As shown, laser engine 202 includes a sheath 300 that encases the optical interface of laser 212 to prevent fluid from coming into contact with laser engine 202. For example, sheath 300 may be constructed of glass, plastic, stainless steel, or other suitable material that allows laser radiation of laser engine 202 to pass through the optical interface within sheath 300 and reach laser detector 204. Depending on the material (e.g., of an opaque material), sheath 300 may include one or more optical windows of a transparent material. A sealing structure (e.g., an adhesive layer, slip fit, gasket) may be used to provide a liquid / water-tight seal from external moisture. The contact sensor group 206, laser engine 202, laser detector 204, and light emitter and color sensor 208 sense multiple parameters of the flowing fluid, including its temperature, its conductivity for composition determination, and its color based on reflectance mode detection. The laser engine shown in Figure 3A is an external cavity laser with a rotating mirror for wavelength tuning, using a III-V semiconductor gain chip integrated in a Metcalf-Littman external cavity configuration. Other variations in the cavity configuration include Littrow or micromechanical membrane mirrors (MEMS) instead of the rotating mirror. Figure 3B shows another possibility, in which the laser engine 202 is understood as a hybrid III-V / IV laser spectrometer on a chip, where wavelength discrimination is achieved electronically, without moving parts, using, for example, Vernier filtering techniques.

[0034] FIG. 4 shows an example of a fluid channel 106 in a side view. As shown here, a sheath 300 is shown disposed inside the fluid channel 106. The fluid channel 106 is circular in cross section, although other shapes are possible. The circular cross section of the fluid channel 106 promotes laminar flow of fluid through the channel. In some embodiments, the fluid channel 106 is 20 mm, 30 mm, 40 mm, 50 mm, 60 mm, or 80 mm in diameter, or any suitable diameter. As described above, the optical interface sheath 300 is formed as a needle that does not obstruct the flow of fluid through the fluid channel 106. In some embodiments, the optical interface sheath includes a window, and the window material transmits spectrally tunable laser radiation of known intensity and wavelength profile through the window and the flow of milk within the channel 106 to a laser detector. In some embodiments, the optical interface sheath is formed as a needle that enters the fluid channel using an opaque material for the sheath body and a laser radiation transparent window. In some embodiments, the sheath of the optical interface has a tube / needle that extends longer or shorter within the channel, a rod that crosses the channel, or other configurations such as a window in the sidewall of the channel, or other suitable configurations.

[0035] The laser engine 202 and the laser detector 204 are at least partially stationary held within the fluid channel 106 such that a portion of the fluid flows between the laser engine 202 and the laser detector 204. In particular, a portion of the fluid flows between the optical interface within the sheath 300 and the laser detector 204. In this example, the laser engine 202 (i.e., the optical interface within the sheath 300) and the biconvex lens 400 are held at a distance of 0.6 mm from each other to allow fluid flow therebetween for detection. The lens 400 may be used, for example, to remove coherence from the laser emission before it reaches the laser detector 204. The end of the sheath 300 may have a width determined by the type of fluid being sensed, with larger widths being used for more transparent fluids. As will be appreciated, the distance may depend on the type of fluid (and its absorption characteristics) and the laser power. A longer distance may allow for more interactions with the target analyte molecules, resulting in smaller concentrations / greater precision. In some embodiments, the end of the sheath 300 for milk sensing may be 0.25 mm, 0.5 mm, 1 mm, 1.5 mm, 2 mm, 2.5 mm, 3 mm, 3.75 mm, 8 mm, 10 mm, 12 mm, 15 mm, or any other suitable distance from the laser detector 204. Other distances are possible, including longer or shorter distances. Because milk is primarily water, measuring fat, protein, and lactose requires knowledge of the precise transmission distance between the optical interface of the laser 300 and the optical interface of the laser detector 204, particularly the thickness of the milk flowing through the laser's optical path. If the gap between the components is too large, the signal from the water component of the milk will dominate the detected spectrum, and valuable modulation of the laser spectrum by the target components (lactose, fat, and protein) will be buried in noise. On the other hand, if the gap between the components is too small, absorbing components such as lactose and protein will be small and difficult to measure. This is because the modulation of the laser intensity by the components of milk is directly proportional to the molar extinction coefficient of the target molecule, thereby modulating the laser spectrum.A very small gap results in a small amount of milk between the laser and the detector, and therefore a very small modulation of the signal detected by the detector. The optimal gap depends on the power level of the laser, the nature of the flowing fluid, and the sensitivity of the detector, and is therefore selected for an optimal signal-to-noise ratio based on the performance of the components comprising the system. This step is performed during assembly and factory calibration.

[0036] The distance between the end of the sheath 300 and the laser detector 204 is directed towards the bottom of the fluid channel 106, and the portion of the fluid channel 106 between the end of the sheath 300 and the laser detector is always filled with fluid due to gravity. Any fluid disturbances caused by air bubbles or other artifacts can be detected from the light detection data at the laser detector 204 and removed from the data using appropriate algorithms during signal processing.

[0037] Real-time in-line milk monitoring in dairy farms poses a challenging environment because the milk flowing through the milking line is not continuous, is usually turbulent, and contains air bubbles, and the milk composition changes dynamically during the milking cycle, requiring sensors to keep up. For spectral measurement of milk composition, laser engine 202 emits spectrally tunable laser radiation of known intensity and wavelength profile, which is transmitted through the milk flow in channel 106 via the optical interface of sheath 300 and collected by laser detector 204 via its respective optical interface. Laser detector 204, which may be a photodetector, converts the received optical signal into an electrical signal and records a time-domain signal (photovoltage or photocurrent). The signal may be processed into a usable absorbance spectrum or may receive further data processing to convert the spectral data into constituent concentration levels or milk composition.

[0038] For example, the electrical signal from the laser detector 204 is first amplified by an amplifier, and the current signal is converted to a voltage. The amplified signal is then sent to an analog-to-digital converter (ADC) and digitized. The signal from the detector, amplifier, and ADC is a time-domain signal, i.e., an intensity function of time. This signal is then sent to a processing device (e.g., a central processing unit (CPU), a microcontroller unit (MCU), a field programmable gate array (FPGA), a complex programmable logic device (CPLD), or a combination of any of these or other devices) to perform further signal processing such as jitter correction, filtering, selection of a relevant spectrum based on predefined rules, averaging, and finally time-to-wavelength / frequency conversion, and this data is then sent to a data processing module via a data link (e.g., an Ethernet cable). The time-to-wavelength conversion always requires knowledge of the absolute wavelength emitted by the laser radiation passing through the milk, and this information comes from the laser engine 202 and is used for the conversion.

[0039] Figure 5 shows an IT hardware and software architecture that may be used. For example, the hardware architecture shown in Figure 5 may be used in a milk acquisition operation using an in-line sensor. The elements of Figure 5 are described above.

[0040] 6 shows a process 622 for generating a record of the composition of a unique animal's milk. Input 6215, including data from spectral capture and other electro-optical sensors, is received and used as live (real-time) measurements 6220. This input 6215 may include data 6221, including temperature, flow rate, color, conductivity, and other physical measurements. This input 6215 may include spectral data 6222.

[0041] The process 622 also uses static model input 6230 along with live measurements 6220 in the model 6240. The static model input 6230 may include model reference parameters 6222 and reference data 6231, which may include a water reference spectrum, molecular reference spectra (e.g., fat, lactose, protein, urea, etc.), sensor reference spectra, and component density data.

[0042] The model combines data 6220 and 6230 to produce model output 6250. Model output 6250 may include constituent concentration levels from spectrum 6251 and other relevant parameters 6252, such as color (e.g., from blood detection), somatic cell levels, and / or other diagnostic parameters.

[0043] Aggregation 6270 may use model output 6250, live measurements 6220, and farm data 6260 to generate a unique animal milk composition record 6271 that is sent to local storage 624. Farm data 6260 may include data related to farm activity (or other operation activity for other operation types), including milking times 6261, animal identifiers 6262, and data 6263, such as historical health and / or treatment data.

[0044] 7 shows a more detailed example of a model 6240 for component analysis. In this example, the model 6240 operates to generate a model output 6250.

[0045] The laser intensity value detected by the photodetector is converted to absorbance in step 6241. The system background is then subtracted in step 6242. As an example, the spectrum of the system background can be as given from a static model. The system background is all optical properties of the system, i.e., laser, photodetector, optical elements, mirrors, etc., excluding milk or other liquid flowing through the sensor.

[0046] A subtraction of the system background is performed in 6242, and the resulting spectrum is then primarily the optical spectrum of the fluid flowing through the spectrum. For example, milk is a substance made up of several different molecules: water, proteins, fat, lactose, metabolites, and different analytes. The next step is baseline calibration and subtraction of the water residual in step 6243. Water is a dominant part of the primary baseline, meaning it needs to be subtracted to reveal the other components of the fluid.

[0047] After baseline calibration and subtraction of water residuals are performed in 6243, the concentration of the target component is derived using, for example, a Beer-Lambert absorption model in a nonlinear regression framework in step 6244. For greater accuracy, reference spectra of the components are provided from a library in the static model. The absorption spectra change in terms of peak width, slope, etc. as concentration and temperature change. The static model input also minimizes the prediction error. Additional parameters and parameter ranges of the nonlinear model may include slope, path length, offset, detector background correction, dark current correction, etc. The output of step 6244 is already a quantitative concentration level of the target component or set of components. The value is sent to model output 6250 and included in a single milking dataset with additional parameters from other sensors and third-party data.

[0048] 8A-8D illustrate the generation of data from raw sensor data. Sensor data 621 is processed using the raw sensor data as described above. This processing 621 can produce spectral capture (e.g., in time-to-frequency format), color sensor data, temperature, flow, and other parameters. Third-party data 6260, such as animal identifiers, milking start / stop times, and veterinary records, can be incorporated into the data processing using models 622, which can include spectra to determine composition and other physical parameters such as flow, temperature, contaminants, etc.

[0049] Animal and herd data aggregation and calculations are performed at 624. This may include aggregating data for individual animals (e.g., cows) and / or herds of animals. Data may be aggregated across multiple farms. This may be used to provide indicators of the health of individual milkings, individual cows, farms, or groups of farms. Veterinary, animal husbandry, and economic indicators may also be generated.

[0050] In one example shown in Figure 8A, operations 621-624 are performed in a cloud computing architecture. In one example shown in Figure 8B, a raw spectrum cache is used to cache raw sensor data along with temporary storage of raw spectra and forwarding of raw spectra. In one example shown in Figure 8C, operation 621 is performed on a local server (e.g., physically located on the farm) and operations 622 and 624 are performed in a cloud architecture. In one example shown in Figure 8D, operations 621 and 622 are performed on a local server, while operation 624 is performed in the cloud.

[0051] Figure 9 is a swim-lane diagram of an example process for sensing fluid. In the process, a processing unit receives readings and determines fluid parameters from sensors such as electrical property sensors, optical color sensors, and temperature flow sensors. The processing unit also generates spectra including filtering, jitter correction, and time-to-frequency transformation. A local server performs data processing including obtaining concentration levels and averaging levels for milking cycles. This can generate aggregate parameters (e.g., for all cows on a farm) and individual milking records for one milking session of a single cow.

[0052] Cloud data processing can include collecting individual milking records and aggregating individual milking records with historical data. Health, veterinary, culling, and economic models can be provided for individual and herd-by-herd analysis.

[0053] Farm equipment can take action and activate valves or alarms based on optical color and / or temperature sensor data and / or electrical property sensors. The software user interface can be used to show individual milking data, show data trends, and / or provide health / veterinary / economic / animal husbandry data and advice.

[0054] Figure 10 shows data illustrating the absorption spectra of milk measured between the optical interface and the laser detector at various transmission path distances. Figure 10 shows six spectral absorption signals spanning various wavelengths of light in the transmission path between the optical interface and the photodetector. From the bottom line to the top line, absorption spectra are shown for transmission paths of 0.2 mm, 0.4 mm, 0.66 mm, 0.82 mm, 1.11 mm, and 1.63 mm. Figure 10 shows spectral data for various optical path lengths (i.e., gaps between the optical sheath interface and the laser detector) at a fixed laser power. The peak-to-valley ratio of the milk spectrum is related to the specificity of the measurement, and the optical path length (gap size) is chosen to maximize the peak-to-valley ratio. If the gap between the optical interface (or the end of the sheath 300 in Figures 3 and 4) and the laser detector 204 is very small, the absolute absorption of the milk fat will also be small, resulting in a weakly modulated laser signal that can be seen at gaps of 0.2 and 0.4 mm for a given laser power and spectral range. The spectral data for milk show an optimum with a gap of 0.6 to 1 mm, where the modulation of the laser signal due to molecular absorption is highest. This can be seen from the peak-to-valley ratio of the two fat peaks. At larger gap widths at a given laser power, water absorption dominates and the signal is lost. For different applications and implant configurations, the gap can vary from tens of microns to several millimeters, and in some cases even several centimeters or meters, depending on the laser power used in the sensor and the optical properties of the fluid in the line.

[0055] 11 shows an example of data 1100 that can be used to sense fluids and control automated devices. In this example, the data is maintained in various data stores 1102-614 and is accessible across a network 1116. The data stores 1102-614 may, in some cases, be implemented in one or more computing elements (e.g., servers, virtual machines, desktop computers, handheld devices) and one or more data structures (e.g., relational databases, files, data messages, memory entries) on the one or more computing elements. The network 1116 may include an internal data bus, an external data network, etc.

[0056] The live (real-time) laser detector parameters 1102 may include received and historically stored data generated by the laser detector 204. For example, the laser detector 204 may generate analog data that is converted to digital data, conditioned (e.g., by removing or subtracting jitter, averaging, aggregating, or converting to a smaller data format), and sent to storage over the network 1116. The laser parameters 1102 may include time series data including spectral values ​​and corresponding timestamp data, or other suitable formats.

[0057] The live (real-time) optical detector parameters 1104 may include received and historically stored data generated by the color sensor 208. For example, the color sensor 208 may generate analog data that is converted to digital data, conditioned (e.g., by removing or subtracting jitter, averaging, aggregating, or converting to a smaller data format), and transmitted to storage over the network 1116. The optical parameters 1104 may include time series data including light intensity values, color values, or other values ​​and corresponding timestamp data, or other suitable formats.

[0058] Live (real-time) contact sensor parameters 1106 may include received and historically stored data generated by contact sensor group 206. For example, contact sensor group 206 may generate analog data that is converted to digital data, conditioned (e.g., by removing or subtracting jitter, averaging, aggregating, or converting to a smaller data format), and sent to storage over network 1116. Contact parameters 1106 may include time series data and corresponding timestamp data, including temperature, conductivity values, flow rate, or other values, or other suitable formats.

[0059] Fluid reference parameters 1108 may include a library of reference values ​​for a particular fluid (e.g., milk) or for various fluids. For example, reference parameters 1108 for a fluid having various components (e.g., milk) may be stored for each component or possible contaminant (e.g., water, milk fat, lactose, urea, blood). These reference parameters may be used when generating quantified outputs, such as concentration levels of milk components derived from spectral measurements and other related parameters, such as temperature, milk volume, milk conductivity, somatic cell count, blood content, etc., and may be stored as fluid measurements 1110.

[0060] Fluid measurements 1110 may include both individual and aggregated measurements. For example, individual measurements may include measurements of a fluid at a specific point in time or within a short time frame (e.g., less than 0.5 seconds). Individual measurements may provide a "snapshot" of the fluid at a moment in time. On the other hand, aggregated measurements may be generated to capture fluid properties over an extended period of time. For example, a milking session for an animal may have a single set of aggregated measurements 1110. Aggregated measurements may be generated by aggregating individual measurements. For example, hundreds or thousands of fluid measurements recording milk fat content may be aggregated to an average fat content for the milking session. As will be appreciated, this aggregation may include all individual measurements in time or may include a sub-sampling of all individual measurements. The use of both individual and aggregated measurements may provide several advantages. For example, milk fat content may be expected to change over the span of a milking session. Fat content at a specific point in time can be of limited value; therefore, averaging milk fat content measurements can provide a more useful measurement for dairy operations when aggregated over longer periods of time, i.e., weeks or months of milking. Trends in composition can reveal valuable insights into individual animal health, lactation, nutrition, and recovery (treatment effectiveness), enabling early action on disease onset, nutritional changes, etc. In some instances, hundreds or thousands of individual measurements are generated per second, and aggregating them into a few aggregated measurements allows for rational understanding. Meanwhile, many automated systems can operate at hundreds or thousands of cycles per second, and having individual measurements in addition to aggregated measurements allows for accurate and responsive control systems. In some cases, maintaining hundreds or thousands of individual measurements per second can identify the time when a fluid change is first detected, which can be useful for troubleshooting a faulty system (e.g., a cow's milking cup becoming dislodged). Similarly, time series data of fluid measurements can be used to identify environmental conditions that affect fluid production. For example, changes to light, sound, temperature, or other environmental factors may be recorded by light, sound, temperature, or other environmental sensors around the cow being milked.These environmental factors can be compared with individuals and / or populations to determine improved environmental conditions for milk production (e.g., modifying light according to sunlight intensity, reducing sudden noise, increasing heat with heaters, etc.).

[0061] Fluid system data 1112 may include information collected or generated during operation of a fluid system (e.g., system 100). Data 1112 may include computer readable instructions for various parts of the fluid system that may be compiled or used to operate automated devices (e.g., valve 108). Additionally, records of historical operation (e.g., information about milk expression including volume, collection timeline, operation of valve 108, and current volume in collection tank 110) may be stored in data 1112.

[0062] Animal data 1114 can record information about animals (e.g., cows 102) used within the fluid system. This animal data 1114 can record information about individual animals (e.g., indexed by a unique identifier for each animal) and aggregate information about the herd, such as historical health data, lactation data, and treatment data. Animal data 1114 can also be used to provide timely alerts to herd management professionals regarding health issues present in individual animals before the health issues spread throughout the herd. Timely health management can minimize animal rest periods and minimize the use of "harsh" medications, such as antibiotics, and instead maximize production of both individual animals and the herd if the onset of disease is diagnosed early enough to be treated with low cost and short rest periods. Additionally, animal data 1114 can provide insight into treatment effectiveness and be used as a reference for establishing "standard operating procedures" when treating specific health issues by selecting medications and / or procedures to be administered with maximum efficiency based on historical herd and / or farm data.

[0063] The animal data 1114, which records information about the milk production, content, and characteristics of individual animals, can be used in other aspects of herd management and culling. Knowing the milk composition data of all herd animals is an invaluable asset for controlled herd culling, and only herd members with the desired composition as aggregated in the milking data can be used in herd culling. For example, an animal that produces a small amount of milk per year but milk that is rich in fat and protein may be more valuable compared to another member of the herd that produces a larger amount of milk per year, because in some countries surpluses are paid for excess fat and protein.

[0064] Animal data 1114 may be reported and displayed in various interfaces and on various client devices as described herein. Information may be formatted or filtered for the specific use of each interface in the program for storing and visualizing the information, such as a farmer interface that focuses more on herd productivity, farm profitability, etc.; a veterinary interface that focuses more on individual animal health, treatment effectiveness, historical data, etc.; and a livestock technical interface that focuses on herd culling, allowing farmers to determine desirable traits and select the correct individual animals for herd culling to enhance desirable characteristics, such as high fat and / or protein content, or disease resistance or milk yield. The interfaces are interconnected, allowing detailed monitoring across multiple farms, from individual animals to herds, and applying best practices in terms of treatment, culling, nutrition, etc., based on actual data. Provisioning and management acts as an administrator console for monitoring and configuring remote farm servers, sensors, and the entire farm Internet of Things (“IoT”) infrastructure.

[0065] 12 shows a swimlane diagram of an example process 1200 for detecting fluids available for a milk acquisition system. Process 1200 may be performed, for example, by system 100 and is therefore described with reference to elements of system 100. However, another system may be used to perform process 1200 or another similar process.

[0066] The laser detector 204 generates a laser reading 1202. For example, the laser detector 204 receives the laser radiation to generate a corresponding laser reading after the laser radiation passes through the fluid. In this manner, various properties of the fluid affect the laser radiation as it passes through the fluid, resulting in computer-readable data recorded by the laser reading. In particular, the laser signal is modulated by absorption of specific molecules of the milk constituents as it passes through the fluid. The modulation is later recovered by signal processing and data models and transformed to quantify constituent concentration levels.

[0067] The contact sensor group 206 detects contact phenomena, which may include electrical and thermal phenomena 704. For example, the contact sensor group 206 may detect one or more contact phenomena of the fluid to generate corresponding contact readings, and determining 1212 one or more fluid measurements may further include using the contact readings. In this manner, various properties of the fluid affect the sensing elements of the contact sensors (e.g., semiconductor thermistors change resistance strongly dependent on temperature) and are recorded by laser readings of computer-readable data.

[0068] The color sensor 208 generates non-coherent readings 1206. For example, the color sensor 208 can generate corresponding non-coherent readings, and determining the one or more fluid measurements further includes using the non-coherent readings. In this manner, various properties of the fluid affect the sensing elements of the color sensor (e.g., based on color, intensity, or other optical properties), resulting in non-coherent readings of computer-readable data being recorded. Examples of color sensors are standard commercially available RGB color sensors, such as, but not limited to, Hamamatsu's S9706 or AMS AG's AS73211.

[0069] The processor 212 receives the readings 1208. For example, the processor 212 may utilize laser readings, contact readings, and / or non-coherent readings. The processor 212 identifies spectral data 1210. For example, the spectral data may be identified as representing physical properties of the fluid as it flows through the sensing system.

[0070] The processor 212 determines the fluid measurement values ​​1212. For example, using the spectral data and reference data defining one or more reference spectra for each possible component of the fluid, the processor 212 can determine various values ​​for the fluid measurement values. For example, the processor 212 can utilize a library of spectra, such as temperature-dependent water and component reference spectra, which may be stored as part of the sensor's firmware. The library may also include a system background spectrum, a set of fitting parameters, parameter fitting ranges, etc. In some implementations, the parameter set included in the library includes one or more of the optical path length, offset, tilt, and dark current correction associated with an in-line sensor installed in the milking system for use in interpreting collected data. The static model can be updated and corrected with the system background spectrum and water reference spectrum during operation in the field when the milking machine is being rinsed with cleaning water. This allows any deviations from the original library data to be corrected in a timely manner, providing high measurement accuracy. This data is used by the model to process live measurement data. Here, the collected spectral data undergoes pre-processing, and the data, in the form of intensity as a function of wavelength or frequency, undergoes further processing until the output can provide a value for the estimated concentration level. In an example of a processing operation, the intensity of the collected spectral data is first converted to absorbance, followed by subtraction of the system background. Here, the system background spectrum is provided from a static model. This allows for system-related nonlinearities and spectral artifacts to be separated from the object under investigation, either milk or broadly. Once the system background has been subtracted, the next operation is baseline calibration and subtraction of the dominant residual of the primary baseline. In the case of milk, the dominant portion of the primary baseline comes from water. The water reference spectrum is provided from the static model and fitted with the best fit and the residual subtracted.Finally, the spectra are processed in some embodiments using a Beer-Lambert model in a nonlinear regression framework. In some implementations, other regression models are used to process the data. Here, a static model provides a reference spectrum of the component under investigation, such as milk fat, lactose, protein, or moisture, in combination with a set of model configuration parameters, such as offset range, slope, and path length, which are used with the reference spectral data to provide a best fit and recover the estimated concentration levels of the component, which are then sent to the model output.

[0071] The controllable device 108 actuates 1214. For example, the controllable device 108 may include an automated valve configured to selectively direct fluid flow to multiple output channels. The automated valve may be actuated by computer readable instructions based on at least one fluid measurement.

[0072] In some cases, actuating the automatic valve to direct the fluid to the waste tank is in response to determining that the contaminant fluid measurement value is greater than a threshold. For example, the threshold may be set so that very low contaminant values ​​that are more likely to be the result of sensing system noise than actual contaminants do not result in actuation. In some cases, actuating the automatic valve based on the at least one fluid measurement value includes actuating the automatic valve to direct the contaminated fluid to the waste tank before the contaminated fluid reaches the automatic valve.

[0073] Additionally, or alternatively, the automatic valves can be used to sort milk by desired composition (rather than or in addition to contaminants). Monitoring the composition of milk allows for sorting milk by desired composition, such as high fat, high protein content, etc., thereby allowing farms to maximize business profits, as in many countries milk with excess fat, protein, and / or lactose commands a price premium.

[0074] 13 shows a flow chart of an example process 1300 for diverting contaminated fluid from a storage tank to a waste tank. For example, the process 1300 may be used to activate 714 the controllable device 108.

[0075] A property of the milk is sensed at 1302. For example, the level of contaminants can be determined, the fat content may be determined, or other suitable parameters may be determined as described above.

[0076] If the properties are within specifications, a switch is activated to direct the milk to a storage tank at 1304. For example, valve 108 may be switched to a first position to direct the milk to recovery tank 110.

[0077] If the properties are out of specification (e.g., too many contaminants, too little milk protein, improper temperature, etc.), the same switch is activated at 1306 to direct the milk to waste tank. For example, valve 108 may be switched to a second position to direct the milk to waste tank 112.

[0078] When the milk is out of specification, an alert may be generated at 1308. The alert may include a physical event such as a light being turned on by computer control, an audible alert being played, etc. The alert may include a push notification on an application, a data message being sent across a network such as an email being sent, or stored in memory.

[0079] 14 illustrates an example computing device 1400 and an example mobile computing device that may be used to implement the techniques described herein. Computing device 1400 is intended to represent various types of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Mobile computing device is intended to represent various types of mobile devices, such as personal digital assistants, mobile phones, smartphones, and other similar computing devices. The components, connections and relationships, and functionality illustrated herein are intended to be examples only and are not intended to limit the embodiments of the invention described in the detailed description and / or claims herein.

[0080] Computing device 1400 includes processor 1402, memory 1404, storage device 1406, high-speed interface 1408 connected to memory 1404 and multiple high-speed expansion ports 1410, and low-speed interface 1412 connected to low-speed expansion port 1414 and storage device 1406. Each of processor 1402, memory 1404, storage device 1406, high-speed interface 1408, high-speed expansion port 1410, and low-speed interface 1412 may be interconnected using various buses and may be mounted on a common motherboard or otherwise as desired. Processor 1402 can process instructions executing within computing device 1400, including instructions stored in memory 1404 or storage device 1406, to display graphical information for a graphical user interface (GUI) on an external input / output device, such as a display 1416 connected to high-speed interface 1408. In other implementations, multiple processors and / or multiple buses may be used, along with multiple memories and types of memory, as desired. Additionally, multiple computing devices may be connected, with each device providing a portion of the required operations (eg, a bank of servers, a group of blade servers, or a multi-processor system).

[0081] The memory 1404 stores information within the computing device 1400. In some implementations, the memory 1404 is a volatile memory unit. In some implementations, the memory 1404 is a non-volatile memory unit. The memory 1404 may also be another form of computer-readable medium, such as a magnetic or optical disk.

[0082] The storage device 1406 can provide mass storage for the computing device 1400. In some implementations, the storage device 1406 can be or include a computer-readable medium, such as a floppy disk device, a hard disk device, an optical disk device, or a tape device, a flash memory or other similar solid-state memory device, or an array of devices, including devices in a storage area network or other configuration. A computer program product can be embodied in an information medium. The computer program product can also include instructions that, when executed, perform one or more of the methods described above, for example. The computer program product can be embodied in a computer- or machine-readable medium, such as memory 1404, storage 1406, or memory of the processor 1402.

[0083] High-speed interface 1408 manages bandwidth-intensive operations of computing device 1400, while low-speed interface 1412 manages less bandwidth-intensive operations. This allocation of functionality is merely exemplary. In some implementations, high-speed interface 1408 is coupled to memory 1404, a display 1416 (e.g., via a graphics processor or accelerator), and a high-speed expansion port 1410, which can accept various expansion cards (not shown). In such implementations, low-speed interface 1412 is coupled to storage device 1406 and low-speed expansion port 1414. Low-speed expansion port 1414, which may include various communication ports (e.g., USB, Bluetooth, Ethernet, wireless Ethernet), may be coupled to one or more input / output devices, such as a keyboard, pointing device, scanner, or a networking device, such as a switch or router, via a network adapter.

[0084] Computing device 1400 may be implemented in several different forms, as shown. For example, it may be implemented as a standard server 1420, or multiple times in a group of such servers. Additionally, it may be implemented in a personal computer, such as a laptop computer 1422. It may also be implemented as part of a rack server system 1424. Alternatively, components from computing device 1400 may be combined with other components in a mobile device (not shown), such as mobile computing device 1450. Each such device may include one or more computing devices 1400 and 1450, and the overall system may be made up of multiple computing devices communicating with each other.

[0085] The mobile computing device 1450 includes, for example, a processor 1452, memory 1464, and input / output devices such as a display 1454, a communication interface 1466, and a transceiver 1468, among other components. The mobile computing device 1450 may also be provided with a storage device, such as a microdrive or other device, that provides additional storage. Each of the processor 1452, memory 1464, display 1454, communication interface 1466, and transceiver 1468 may be interconnected using various buses, and several components may be mounted on a common motherboard or otherwise as desired.

[0086] The processor 1452 can execute instructions within the mobile computing device 1450, including instructions stored in the memory 1464. The processor 1452 can be implemented as a chipset of chips including separate analog and digital processors. The processor 1452 can provide for the coordination of other elements of the mobile computing device 1450, such as control of a user interface, applications running on the mobile computing device 1450, and wireless communication by the mobile computing device 1450.

[0087] The processor 1452 can communicate with a user via a control interface 1458 and a display interface 1456 coupled to a display 1454. The display 1454 can be, for example, a TFT (thin film transistor liquid crystal display) display, an OLED (organic light emitting diode) display, or other suitable display technology. The display interface 1456 can include appropriate circuitry for operating the display 1454 to present graphical and other information to the user. The control interface 1458 can receive instructions from the user and convert them for provision to the processor 1452. Additionally, an external interface 1462 can provide communication with the processor 1452 to enable short-range communication between other devices and the mobile computing device 1450. The external interface 1462 can, for example, provide wired communication in some implementations or wireless communication in other implementations, and multiple interfaces can be used.

[0088] Memory 1464 stores information on mobile computing device 1450. Memory 1464 may be embodied as one or more computer-readable media, volatile memory units, or non-volatile memory units. Expansion memory 1474 may be provided and connected to mobile computing device 1450 via expansion interface 1472, which may include, for example, a SIMM (single in-line memory module) card interface. Expansion memory 1474 may provide extra storage space for mobile computing device 1450 or may also store applications or other information for mobile computing device 1450. Specifically, expansion memory 1474 may include instructions for performing or supplementing the processes described above, and may also include security information. Thus, for example, expansion memory 1474 may be provided as a security module for mobile computing device 1450 and may be programmed with instructions that enable secure use of mobile computing device 1450. Additionally, security applications may be provided via a SIMM card, along with additional information, such as identification information, posted on the SIMM card in an unhackable manner.

[0089] The memory may include, for example, flash memory and / or NVRAM memory (non-volatile random access memory), as described below. In some implementations, a computer program product may be embodied on an information medium. The computer program product includes instructions that, when executed, perform one or more methods, such as those described above. The computer program product may be a computer- or machine-readable medium, such as memory 1464, expansion memory 1474, or memory on processor 1452. In some implementations, the computer program product may be received, for example, in a signal propagated over transceiver 1468 or external interface 1462.

[0090] Mobile computing device 1450 can communicate wirelessly via communication interface 1466, which may include digital signal processing circuitry as needed. Communication interface 1466 can provide communications under various modes or protocols, such as GSM voice (Global System for Mobile Communications), SMS (Short Message Service), EMS (Enhanced Message Service), or MMS messaging (Multimedia Message Service), CDMA (Code Division Multiple Access), TDMA (Time Division Multiple Access), PDC (Personal Digital Cellular), WCDMA (Wideband Code Division Multiple Access), CDMA2000, or GPRS (General Packet Radio Service). Such communications can occur, for example, via transceiver 1468 using radio frequencies. Additionally, short-range communications can occur using Bluetooth, WiFi, or other such transceivers (not shown). Additionally, a GPS (Global Positioning System) receiver module 1470 can provide additional guidance and location related wireless data to the mobile computing device 1450, which can be used as needed by applications running on the mobile computing device 1450.

[0091] The mobile computing device 1450 can also communicate audibly using an audio codec 1460 that can receive spoken information from a user and convert it into usable digital information. The audio codec 1460 can also generate sounds that are heard by the user, for example, through a speaker in the handset of the mobile computing device 1450. Such sounds may include sounds from a voice call, recorded sounds (e.g., voice messages, music files, etc.), and sounds generated by applications running on the mobile computing device 1450.

[0092] The mobile computing device 1450 may be implemented in several different forms, as shown in the figure, such as a mobile phone 1480, or as part of a smartphone 1482, personal digital assistant, or other similar mobile device.

[0093] Various implementations of the systems and techniques described herein may be realized in digital electronic circuitry, integrated circuits, specially designed ASICs (application-specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs executable and / or interpretable by a programmable system including at least one programmable processor coupled to receive and transmit data and instructions from and to a storage system, at least one input device, and at least one output device, for either a special or general purpose.

[0094] These computer programs (also known as programs, software, software applications, or code) include machine instructions for a programmable processor and may be implemented in a high-level procedural and / or object-oriented programming language, and / or assembly / machine language. As used herein, the terms machine-readable medium and computer-readable medium refer to any computer program product, apparatus, and / or device (e.g., magnetic disk, optical disk, memory, programmable logic device (PLD)) used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term machine-readable signal refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0095] To provide for interaction with a user, the systems and techniques described herein may be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user, and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices may also be used to provide for interaction with a user. For example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback), and input from the user may be received in any form, including sound, voice, or tactile input.

[0096] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., data servers), or that includes middleware components (e.g., application servers), or that includes front-end components (e.g., client computers having a graphical user interface or web browser through which a user can interact with an implementation of the systems and techniques described herein), or any combination of these back-end, middleware, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communications network). Examples of communications networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0097] A computing system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

Claims

1. 1. A sensing system for detecting a physical property of a fluid, comprising: a laser engine configured to emit spectrally tunable laser radiation through the fluid as the fluid flows through the sensing system; a laser detector configured to receive the laser radiation after it has passed through the fluid and to generate a corresponding laser reading; one or more processors; a computer memory storing computer readable instructions; Equipped with The instructions, when executed by the one or more processors, cause the one or more processors to: receiving a laser reading from the laser detector; and identifying spectral data from the laser readings that are indicative of physical properties of the fluid as it flows through the sensing system; determining one or more fluid measurements of one or more corresponding components of the fluid using the spectral data and reference data defining one or more reference spectra for each potential component of the fluid; performing an action including Sensing system.

2. the system further comprising a housing containing the laser engine, the laser detector, one or more processors, and the computer memory. The system of claim 1 .

3. The system further comprises a network interface; the operations further include transmitting the one or more fluid measurements via the network interface. The system of claim 2 .

4. The system further comprises: a housing that houses the laser engine and the laser detector; one or more computing devices including the one or more processors and the computer memory; Equipped with The system of claim 1 .

5. the system further comprising a fluid channel between the laser engine and the laser detector through which the fluid flows; the laser engine and the laser detector are at least partially fixedly held within the fluid channel such that a portion of the fluid flows between the laser engine and the laser detector. The system of claim 1 .

6. The system further comprises: a contact sensor fixedly held at least partially within the fluid channel; a light emitter fixedly held at least partially within the fluid channel; a color sensor fixedly held at least partially within the fluid channel; Equipped with The system of claim 5.

7. The laser engine and the laser detector are fixedly held apart by a distance of less than 10 mm. The system of claim 5.

8. the laser engine is coupled to a photon-transparent sheath configured to prevent contact of the laser engine with the fluid; The system of claim 5.

9. The system further comprises: a contact sensor configured to contact the fluid as it flows through the sensing system and to detect one or more contact events of the fluid to generate a corresponding contact reading; determining one or more fluid measurements further includes using the contact readings; The system of claim 1 .

10. The system further comprises: a light emitter configured to emit incoherent light; a color sensor configured to receive the non-coherent light after it passes through the fluid and to generate a corresponding non-coherent reading; Equipped with determining one or more fluid measurements further includes using the non-coherent readings; The system of claim 1 .

11. the system further comprising an automated valve configured to selectively direct the fluid flow to a plurality of output channels; the action includes actuating the automated valve based on at least one of the fluid measurements. The system of claim 1 .

12. and actuating the automatic valve includes, in response to determining that the contaminant fluid measurement value is greater than a threshold, actuating the automatic valve to direct the fluid to a waste tank. The system of claim 11.

13. and actuating the automatic valve based on at least one of the fluid measurements includes actuating the automatic valve to direct the contaminated fluid to the waste tank before the contaminated fluid reaches the automatic valve. The system of claim 12.

14. The fluid is raw milk obtained from livestock. The system of claim 1 .

15. the laser engine includes a III-V semiconductor-based laser; The system of claim 1 .

16. the III-V semiconductor-based laser is configured to emit light through an optical interface, and the laser detector is positioned opposite the optical interface; The system of claim 15.

17. the III-V semiconductor-based laser is a tunable laser; The system of claim 15.

18. the III-V semiconductor-based laser is configured to emit light over a range of wavelengths through an optical interface; The system of claim 15.

19. the laser engine is a III-V / IV semiconductor-based laser; 20. The system of claim 18.

20. the III-V / IV semiconductor-based laser is configured to emit through an optical interface, and the laser detector is positioned opposite the optical interface; 20. The system of claim 19.

21. the III-V / IV semiconductor-based laser is a tunable laser; 21. The system of claim 20.

22. the III-V / IV semiconductor-based laser is configured to emit the light over a range of wavelengths through the optical interface; 22. The system of claim 21.

23. the optical interface is positioned in a flow path of the fluid flowing through the sensing system; 22. The system of claim 21.

24. the optical interface includes a tube extending from the laser engine into a channel through which the fluid flows through the sensing system, the tube extending toward the laser detector, a distance from a distal end of the glass tube to a surface of the laser detector being between 0.5 mm and 10 mm, and the tube including an optically transparent portion; 20. The system of claim 19.

25. the fluid measurements are measurements of each of the components at a particular time; The operations further include aggregating the fluid measurements for a single pumping session into a set that reflects changes to the fluid measurements throughout the pumping session. The system of claim 1 .

26. The operations further include storing the fluid measurements for a single milking session indexed by a unique identifier for that single milking session for an individual animal. The system of claim 1 .

27. The operations may further include storing the fluid measurements from one milking session with other fluid measurements from other milking sessions to generate historical data for a particular animal.

27. The system of claim 26.

28. The operations further include generating a herd index for the plurality of animals using the historical data for each animal in the herd.

28. The system of claim 27.

29. The operations further include combining the herd-specific metrics with third-party data for at least one of the following groups: 1) herd health data; 2) herd culling data; and 3) herd management data.

29. The system of claim 28.

30. generating an individual or herd index includes using said third party data and historical data for each animal in the herd; 30. The system of claim 29.

31. The fluid physical properties include fluid composition, electrical properties, temperature, flow, color, quantified component concentration levels, and the presence or absence of one or more components; The system of claim 1 .

Images