Ultrasonic sensor monitoring device, system and method

Abstract

A sensing assembly configured to monitor flow in a pipe, the sensing assembly comprising: an ultrasonic sensor assembly comprising an ultrasonic sensor, wherein the ultrasonic sensor assembly is configured to be physically coupled to a pipe, and wherein the ultrasonic sensor is configured to transmit and receive ultrasonic signals through the pipe; and a controller electrically coupled to the ultrasonic sensor assembly, wherein the controller is configured to: determine a time of flight (ToF) of the ultrasonic signals, determine a flow rate in the pipe based on the time of flight, store flow data comprising the determined flow rate, and use a segmentation machine learning model to recognize a single plumbing fixture associated with the flow data. In some embodiments, the controller is further configured to recognize a leak or a malfunction of a single plumbing fixture based on the flow data.

Description

Cross-reference to related applications

[0001] This application claims priority and benefit to U.S. Provisional Application No. 63 / 527,279, filed July 17, 2023, and U.S. Provisional Application No. 63 / 527,137, filed July 17, 2023, the entire contents of which are hereby incorporated by reference. Technical Field

[0002] This disclosure relates to ultrasonic monitoring devices and systems, particularly for monitoring, measuring and detecting fluid flow (e.g., water flow in buildings). Background Technology

[0003] Currently, many users are unaware of their unique water usage habits and the condition of their water supply and drainage systems at any given time. To conserve water and ensure the structural integrity of buildings, efforts are underway to develop systems and methods configured to monitor and identify leaks in water supply and drainage systems and obtain water usage data corresponding to individual sanitary fixtures. Detecting and providing users with data on leaks in their water supply and drainage systems and the water efficiency of various sanitary fixtures allows users to take steps to repair leaks and develop water-saving habits. Summary of the Invention

[0004] Systems and methods are needed to determine the water efficiency of various water-using sanitary appliances (such as household fixtures and appliances, like toilets, faucets, showers, dishwashers, etc.). To achieve this, a sensing system is required that can be connected to a single water inlet pipe in a residence, multi-family residential building, or office building.

[0005] Water inlet pipes in residences and other buildings come in a wide range of sizes and materials. A universal sensing system is needed that can be easily and non-destructively connected to water inlet pipes of different sizes and materials and configured to monitor and identify leaks in homes or buildings. A universal sensing system is also needed that can identify different water-using devices and appliances and determine their water usage efficiency.

[0006] Therefore, a sensing assembly configured to monitor material flow in a pipe is disclosed, the sensing assembly including an adjustment mechanism, a housing, and an ultrasonic sensor assembly, wherein the ultrasonic sensor assembly is at least partially located within the housing and includes an ultrasonic sensor configured to connect to the pipe, and the adjustment mechanism is configured to allow the ultrasonic sensor to be connected to pipes of various diameters. The adjustment mechanism also allows for fine-tuning of the sensor-to-pipe connection.

[0007] The sensing assembly may include a general-purpose C-clamp type shape, allowing for tool-free installation on a pipe. The housing and adjustment mechanism may be configured to properly align the pipe with the assembly, enabling optimal ultrasonic signal acquisition from the ultrasonic sensor. The ultrasonic sensor may include a first ultrasonic transducer coupled to a first upstream wedge and a second ultrasonic transducer coupled to a second downstream wedge, wherein the first and second wedges are configured to directly physically contact the pipe via lateral contact surfaces of the wedges. The wedges may comprise engineering thermoplastics. The subject of this disclosure is also the wedges and ultrasonic sensors as described herein.

[0008] Sensing components can be configured to identify and report system leaks or malfunctions. Sensing components can be configured to identify and report the use of individual water-using fixtures or appliances (e.g., toilets, showers, faucets, dishwashers, washing machines, etc.). Sensing components can be configured to determine and report leaks or malfunctions in specific (individual) water-using fixtures and appliances. Sensing components can be configured to determine and report the water efficiency of individual water-using fixtures and appliances.

[0009] A sensing assembly can be configured to monitor material flow in a pipe. The sensing assembly includes an adjustment mechanism, a controller, a housing, and an ultrasonic sensor assembly, wherein the ultrasonic sensor assembly is at least partially located within the housing and includes an ultrasonic sensor. The ultrasonic sensor assembly is electrically connected to the controller. The ultrasonic sensor is configured to connect to the pipe and to transmit and receive ultrasonic signals through the pipe. The controller is configured to determine the time-of-flight (ToF) of the ultrasonic signals and to determine the flow rate in the pipe based on the time-of-flight.

[0010] In some embodiments, a sensing component configured to monitor flow in a pipe is provided, the sensing component comprising: an ultrasonic sensor assembly including an ultrasonic sensor, wherein the ultrasonic sensor assembly is configured to be physically coupled to the pipe, and wherein the ultrasonic sensor is configured to transmit and receive ultrasonic signals through the pipe; and a controller electrically coupled to the ultrasonic sensor assembly, wherein the controller is configured to: determine the time of flight (ToF) of the ultrasonic signals, determine a flow rate in the pipe based on the time of flight, store flow data including the determined flow rate, and use a segmentation machine learning model to identify individual sanitary appliances associated with the flow data.

[0011] In some implementations, identifying a single sanitary appliance associated with the flow data includes: sampling the flow data and dividing the sampled data into multiple time-duration segments, wherein a first time-duration segment of the multiple time-duration segments is characterized by no flow rate change, and wherein a second time-duration segment of the multiple time-duration segments is characterized by having a flow rate change.

[0012] In some implementations, identifying a single sanitary appliance associated with the flow data includes analyzing one or more of the duration, volume accumulation, direction, start flow rate, and end flow rate in one or more of the plurality of time duration segments.

[0013] In some implementations, identifying individual sanitary appliances associated with the flow data includes using a state machine to apply one or more rules to the time duration segment.

[0014] In some implementations, determining the flow rate includes applying temperature-compensated data processing operations to the time of flight of the ultrasonic signal.

[0015] In some implementations, the temperature compensation data processing operation is based on a set of no-flow rates determined at various temperatures.

[0016] In some implementations, the flow data includes one or more of volume data, moving average flow rate data, and timestamps of the ultrasonic signals.

[0017] In some implementations, identifying individual sanitary appliances associated with the flow data includes determining a no-flow rate based on the flow data.

[0018] In some implementations, the controller is configured to use the no-flow rate to calibrate the ultrasonic sensor assembly.

[0019] In some implementations, the controller is configured to identify a leak or malfunction of the individual sanitary appliance based on the flow data and the no-flow rate.

[0020] In some implementations, identifying a leak or malfunction of the individual sanitary appliance via the controller includes identifying volumes above a defined threshold from the volume data.

[0021] In some implementations, the controller is configured to communicate with a cloud server, and the cloud server is configured to identify leaks or malfunctions in the individual sanitary fixture based on the flow data and the no-flow rate.

[0022] In some implementations, identifying the leak or malfunction associated with a single sanitary appliance includes transmitting the flow data and flow rate moving average data to the cloud server and receiving water usage trend analysis from the cloud server.

[0023] In some implementations, the controller is configured to determine the velocity of sound in the pipe based on the ultrasonic signal.

[0024] In some implementations, the controller is configured to transmit the streaming data to the cloud server according to a publishing protocol.

[0025] In some implementations, the publishing protocol includes transmitting the streaming data when the volume is identified as exceeding the defined threshold.

[0026] In some implementations, the publishing protocol includes publishing samples of flow data with a directly preceding timestamp when flow rate changes above a defined threshold are identified.

[0027] In some implementations, the publishing protocol includes transmitting streaming data from the controller to the cloud server at variable time intervals.

[0028] In some implementations, the publishing protocol takes into account the signal-to-noise ratio (SNR).

[0029] In some implementations, the sensing component is configured to send and receive ultrasonic signals in millisecond timescales.

[0030] In some implementations, the sensing component is configured to transmit and receive ultrasonic signals until the moving average catches up with the flow rate in the flow data.

[0031] In some implementations, the sensing component further includes an LED indicator configured to indicate the intensity of the ultrasonic signal.

[0032] In some embodiments, the ultrasonic sensor assembly is further configured to transmit a test signal through the conduit and receive a response to the test signal, the response including one or more test outputs.

[0033] In some implementations, the controller is configured to determine the material of the pipe based on the one or more test outputs.

[0034] In some implementations, the controller is configured to determine the diameter of the pipe based on the one or more test outputs.

[0035] In some embodiments, a sensing assembly configured to monitor flow in a pipe is provided, the sensing assembly comprising: an ultrasonic sensor assembly including an ultrasonic sensor, wherein the ultrasonic sensor assembly is configured to be physically coupled to the pipe, and wherein the ultrasonic sensor is configured to transmit and receive ultrasonic signals through the pipe; and a controller electrically coupled to the ultrasonic sensor assembly, wherein the controller is configured to: determine the time of flight (ToF) of the ultrasonic signals, determine a flow rate and a no-flow rate in the pipe based on the time of flight, store flow data including the determined flow rate and the no-flow rate, calibrate the ultrasonic sensor using the no-flow rate, and identify a leak or malfunction of an individual sanitary appliance based on the flow data. Attached Figure Description

[0036] This disclosure is illustrative by way of the accompanying drawings and is not restrictive. For simplicity and clarity, features shown in the drawings are not necessarily drawn to scale. For example, the dimensions of some features may be exaggerated relative to others for clarity. Furthermore, reference numerals are repeated where appropriate in the drawings to indicate corresponding or similar elements.

[0037] Figures 1A to 1J Various views of the sensing assembly and sensing assembly components are shown according to some implementation schemes.

[0038] Figure 1K Views of sensing component parts are provided according to some implementation schemes.

[0039] Figure 2A Some implementation schemes provide representations of segmented analysis.

[0040] Figure 2B Data processing pipelines are illustrated in some implementation schemes.

[0041] Figure 2C The state machine is illustrated in some implementation schemes.

[0042] Figure 3A Some implementation schemes illustrate the rejected time-segmented flow rate distribution used to determine temperature compensation.

[0043] Figure 3B Some implementation schemes illustrate the accepted time-segmented flow rate distribution used to determine temperature compensation.

[0044] Figure 3C Some implementation schemes provide graphs of regression analysis that will be applied to temperature compensation.

[0045] Figure 4A A graph of flow rate based on sample frequency is shown according to some implementation schemes.

[0046] Figure 4B A graph showing flow rate data and moving average data obtained from samples at fixed time intervals is shown according to some implementation schemes.

[0047] Figure 5A A graph showing the flow rate distribution is shown based on some implementation schemes.

[0048] Figure 6A Some implementation schemes provide a view of the user dashboard.

[0049] Figure 6B Some implementation schemes provide a portion of the user dashboard.

[0050] Figure 7A and Figure 7B Views of the user interface notification main page and notification subpages are provided according to some implementation schemes.

[0051] Figures 8A to 8D Views of an interactive water level indicator for a user interface used for consumption comparison are shown according to some implementation schemes.

[0052] Figure 9 Zero flow rate detection is illustrated in some implementation schemes.

[0053] Figure 10 The computer is illustrated according to some implementation schemes. Detailed Implementation

[0054] In some embodiments, a sensing component is provided that can be configured to monitor material flow (e.g., water flow rate) in a pipe (such as a water supply sanitary appliance). The sensing component may include an ultrasonic sensor assembly with an ultrasonic sensor. The ultrasonic sensor assembly may be configured to be physically coupled to pipes of various shapes and sizes (e.g., tubes). In some embodiments, the ultrasonic sensor may be configured to transmit and receive ultrasonic signals through the pipe. The sensing component may also include a controller electrically coupled to the ultrasonic sensor assembly. In some embodiments, the controller may be configured to determine the time of flight (ToF) of the ultrasonic signal and determine the flow rate (e.g., in gallons per minute) in the pipe based on the time of flight. In some embodiments, the controller may store water flow data including the determined flow rate and identify individual water-using sanitary appliances associated with the water flow data. The controller may also transmit the water flow data to a cloud server for further analysis and storage. The sensing component and / or the cloud server may be configured to provide users with water flow data for individual water-using sanitary appliances, allowing users to determine which sanitary appliances use the most water and adjust their habits accordingly.

[0055] In some implementations, the controller can be configured to determine a no-flow rate (e.g., the flow rate corresponding to zero water flow) instead of a flow rate, or to determine a no-flow rate in addition to a flow rate. Since ultrasonic sensors can measure temperature changes as flow rate changes, determining a no-flow rate (e.g., a true "zero") after removing temperature contributions allows for more precise detection of smaller leaks. The controller can be configured to use the no-flow rate to calibrate the ultrasonic sensor for more accurate and precise leak detection. The controller can then be able to analyze the water flow data to determine the presence of small or large leaks in pipes and / or individual water-using sanitary appliances. The controller can also be configured to transmit the no-flow rate / water flow data to a cloud server to perform leak detection. The sensing component and / or the cloud server can be configured to report the presence of a leak to the user.

[0056] In the following description of various embodiments, it should be understood that, unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” used in the following specification are intended to also include the plural forms. It should also be understood that the term “and / or” as used herein refers to and covers any and all possible combinations of one or more of the associated listed items. It should also be understood that the terms “comprising” and / or “including”, when used herein, specify the presence of the stated feature, integer, step, operation, element, component, and / or unit, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, units, and / or groups thereof.

[0057] Sensing components Figure 1A Views of a sensing assembly 100 are provided according to some embodiments. The sensing assembly 100 includes a housing 101 in which an ultrasonic sensor assembly 104 is located. The ultrasonic sensor assembly 104 includes a wedge 104w located in a wedge holder 119. The housing 101 includes an electronic port 118 configured to receive a plug / wire to provide power to components including a controller (not visible), an LED indicator (not visible), and the ultrasonic sensor assembly 104. The sensing assembly 100 includes... Figure 1BThe hanger 115 is shown separately. The hanger 115 is configured to receive and hold pipes of various diameters in physical contact with the wedge 104w. The sensing assembly 100 also includes a light tube 135 that surrounds the entire circumference of the upper edge of the housing 101 (around the edge inside the housing). The assembly 100 has a compact shape and size, allowing it to be coupled to pipes of various diameters in a variety of confined spaces. The light tube 135 is configured to be visible from virtually any angle, thus indicating whether / when the assembly 100 is correctly mounted on the pipe. Correct mounting likely occurs when the pipe is positioned on the wedge 104w, allowing for optimal reception of ultrasonic signals via the controller.

[0058] According to some implementation plans Figure 1C A side view of the sensing component 100 is shown, and Figure 1D A cross-sectional view of the sensing assembly 100 is shown. A conduit CDT is coupled to an ultrasonic sensor assembly 104 located within a housing 101 and between the housing 101 and a hanger 115. The hanger 115 includes a well 115w configured to receive the conduit CDT, and the housing 101 includes a well 101w also configured to receive the conduit CDT. The hanger 115 is configured to be adjustable relative to the housing 101 by means of a screw 102 and a knob 102k, such that the hanger well 115w and the housing well 101w can move closer and further apart along a generally straight line to accommodate conduits of different diameters. In some embodiments, the knob 102k and the screw 102 may be collectively referred to as an "adjustment mechanism". The interior of the housing 101 includes a rectangular box-shaped opening 101h configured to hold the ultrasonic sensor assembly 104.

[0059] Figure 1E and Figure 1F Further views of the sensing assembly 100 are provided according to some embodiments. Various dimensions are shown, for example in the locations shown, with the housing 101 and knob 102k together comprising a maximum length of approximately 100 mm, a width of approximately 85 mm, and a depth of approximately 70 mm. Thus, the housing 101 including the knob 102k can have a length of approximately 600 cm. 3 The total volume. It can be seen that the hanger well 115w and the shell well 101w are basically aligned, that is, their low points are basically on the same line and are configured to be adjustable relative to each other along a line. Figure 1GA sensing assembly 100 is shown attached to a conduit CDT, with the outer diameter (OD) of the conduit on the left being approximately 23 mm and the outer diameter of the conduit on the right being approximately 51 mm. The sensing assembly 100 is configured to receive and attach to conduits with a minimum outer diameter of approximately 15 mm (approximately 0.6 inches) to a maximum outer diameter of approximately 51 mm (approximately 2.0 inches). In the sensing assembly 100, adjustment mechanisms 102 / 102k are configured to “pull” the hanger 115 toward the ultrasonic sensor assembly 104, rather than “push” it toward the ultrasonic sensor assembly.

[0060] Figure 1H , Figure 1I and Figure 1J Various views of the sensing assembly 100 are provided, in which the light tube 135 can be seen. An ultrasonic sensor assembly 104 is also shown, comprising a wedge 104w located in a wedge holder 119. A wire portion 118w inserted into a power port 118 is also shown. The light tube 135 is a separate component configured to be positioned around the entire circumference of the edge of the housing 101. The light tube 135 also functions as a cover positioned above the ultrasonic sensor assembly 104. An LED (not shown) may be mounted on a controller (not shown) located inside the housing 101. The LED may be configured to flash yellow, for example, when the assembly 100 is positioned on the tube. One or more LEDs may be configured to emit a solid color, such as green, when someone adjusts the assembly 102 / 102k to optimize the tube position on the ultrasonic sensor assembly 104 between the hanger 115 and the housing 101, and the controller detects a good or optimal ultrasonic signal. The LED may be configured to emit light toward the light tube 135, and the light may be configured to illuminate the entire or approximately the entire volume of the light tube 135. The light tube 135 may be solid or have a hollow volume, and may be translucent. The light tube 135 is configured to provide a view of the LED light from virtually any view or angle. In one embodiment, the light tube 135 may comprise a thermoplastic, such as an engineering thermoplastic, like polycarbonate.

[0061] Figure 1KViews of an ultrasonic sensor assembly 104 and a wedge 104w are provided according to some embodiments. The top illustration shows the ultrasonic sensor assembly 104 having a wedge 104w located within a wedge holder 119. The center illustration shows an exploded view of the ultrasonic sensor assembly 104. The wedge has a curved contact surface 104c configured to provide contact with a high surface area of ​​the conduit. The wedge holder 119 includes a support 119b configured to receive each wedge 104w. The wedge holder 119 also includes two pairs of protrusions 119t (one of each pair visible) configured to engage with blind holes 104h located on either side of each wedge 104w. The protrusion 119t / hole 104h engagement is configured to indicate proper attachment of the wedge 104w to the holder 119 and allow a degree of rotational movement of the wedge 104w during engagement with the conduit. A certain degree of rotational motion is configured to compensate for any irregular shape of the pipe and provide good contact between the wedge 104w and the pipe.

[0062] Figure 1K The bottom illustration shows a perspective view of the wedge 104w. Visible is an inclined surface 104a configured to receive an ultrasonic transducer, which contrasts with an inclined surface 104b, a blind hole 104h, and a curved contact surface 104c. An upper flat surface 104u is configured to rest within a bracket 119b of a retainer 119. The wedge surface 104a includes an extended edge 104p configured to receive an edge of an ultrasonic sensor. The wedge 104w includes a length L1 of approximately 24.5 mm, a maximum width W1 of approximately 12 mm, a maximum height of approximately 11 mm, and a curved contact surface 104c width W2 of approximately 10 mm. The edge 104p extends outward from the surface 104a by approximately 0.5 mm. In one embodiment, the contact surface 104c may include a "rubberized" or elastic coating configured to act as a surface contact facilitator between the surface 104c and the conduit. Surface contact enhancers can improve surface-to-surface contact between surface 104c and the pipe surface, and can improve ultrasonic signal strength and quality. For example, they can eliminate any air gaps in the contact area.

[0063] Thermoplastics may include, for example, engineering thermoplastics. In some embodiments, the wedge may include a thermoplastic selected from one or more of the following: polyamide, polyester, polycarbonate, polyacetal, acrylonitrile-butadiene-styrene, poly(meth)acrylate, polyetheretherketone, polyetherketoneketone, polyketone, polyphenylene sulfide, polyphenylene ether, polysulfone, or polytetrafluoroethylene. Polysulfone includes, for example, one or more of poly(arylsulfone), poly(bisphenol A sulfone), polyethersulfone, polyphenylene sulfone, polysulfone, or VICTREX HTA. The term "polysulfone" may refer to one or more of the polysulfones listed above.

[0064] In some embodiments, the sound velocity and / or acoustic impedance of the thermoplastic used to manufacture the wedge can be close to that of water. This can reduce the adverse effects of ultrasound refraction when passing through different media. In some embodiments, the wedge can be manufactured using molding techniques (e.g., injection molding). In some embodiments, the wedge can be manufactured using 3D printing technology.

[0065] An ultrasonic sensor assembly may include a first ultrasonic transducer coupled to a first upstream wedge and a second ultrasonic transducer coupled to a second downstream wedge. The first wedge may be located on an upstream portion of the pipe, while the second wedge may be located on a downstream portion of the pipe. The upstream / downstream orientation of the wedges may be reversed as needed for positioning the sensing assembly on the material pipe. The first and second wedges may be symmetrical and interchangeable. When coupled to a pipe, the first and second wedges may be mirror images of each other. The wedges may have direct physical contact with the material pipe at their lateral contact surfaces.

[0066] As shown in the figure, an ultrasonic transducer can be connected to the upstream inclined surface of a first upstream wedge, while a second ultrasonic transducer can be connected to the downstream inclined surface of a second downstream wedge. The inclined surfaces of the wedges can be configured to optimize the angle at which ultrasonic signals or waves travel through the conduit to reach the other ultrasonic transducer. Similarly, the opposite inclined surface without an ultrasonic transducer connection can also be configured to optimize the angle at which ultrasonic signals or waves travel through the conduit to reach the other ultrasonic transducer.

[0067] In some embodiments, the sensing assembly may include a surface contact amplifier configured to contact a first wedge, a second wedge, and a conduit. The surface contact amplifier is configured to increase surface contact between the first wedge, the second wedge, and the conduit. In some embodiments, the surface contact amplifier may include an elastic strip, such as a silicone strip. In some embodiments, the surface contact amplifier may include a gel. The strip or gel may include an adhesive. In some embodiments, the surface contact amplifier may be in direct contact with each of the first wedge, the second wedge, and the conduit. In other embodiments, the surface contact amplifier may include an elastic coating on the wedge contact surface. In some embodiments, the wedge having the surface contact amplifier may still be considered to be in direct physical contact with the conduit.

[0068] Ultrasonic transducers can also be piezoelectric transducers. Ultrasonic transducers can be connected to a controller (e.g., a microcontroller or microprocessor) and a power supply via wired or wireless communication. Ultrasonic transducers can communicate with an analog-to-digital converter (ADC) via wired or wireless communication. In some embodiments, the ultrasonic transducer alternately transmits and receives ultrasonic waves through a first wedge, a conduit filled with material, and a second wedge. The acoustic data can be converted into electrical signals and then sent to the ADC. The ADC can convert the electrical signals into digital data. The controller can be configured to receive digital data from the ADC.

[0069] The controller can be configured to determine the time of flight (ToF) based on sensor data. Based on the ToF measurement, the controller is configured to determine the flow rate using an algorithm. The controller can be configured to evaluate digital flow rate data to determine water usage by individual appliances or sanitary fixtures. The controller may include algorithms configured to identify and recognize water-using sanitary fixtures and appliances by the "audible characteristics" of water usage (reflected in the digital flow rate data). Water-using fixtures and appliances may include one or more of the following: toilets, urinals, faucets, dishwashers, washing machines, shower equipment, bathtub faucets, ice makers, water heaters, etc., including any feature with a water valve, such as a sprinkler system or outdoor hose.

[0070] The controller can be configured to identify the contribution of temperature to flow rate data. For example, the sensing component may include one or more temperature sensors in electronic communication with the controller. The controller may include algorithms configured to perform temperature-compensated data processing operations. In some embodiments, temperature may be continuously monitored by the controller. In some embodiments, temperature may be determined by monitoring the relative velocity of sound waves within the material.

[0071] The controller can be configured to identify and determine the normal and abnormal states of a water supply and drainage system. For example, identifying the normal state of a water supply and drainage system may include determining "no leaks." Identifying the abnormal state of a water supply and drainage system may include determining "leaks." A leak may be a general leak somewhere in the pipes and / or a leak associated with a specific water-using appliance or fixture. A leak may include a toilet leak because the float valve is not properly seated. In some embodiments, the sensing component may be configured to identify and determine general leaks and / or leaks associated with a specific (individual) water-using appliance or fixture.

[0072] The controller can be configured to report the normal or abnormal status of the water supply and drainage system. For example, a computing device such as a smartphone or laptop can be linked to the controller. The controller can be configured to transmit data to the computing device. The computing device can have a graphical user interface configured to display data, including water usage data of water-using devices and appliances over a period of time, water efficiency data, efficiency data, etc.

[0073] In some implementations, the controller can communicate remotely with a server (e.g., a cloud server). The controller can be configured to transmit data to / from the cloud server while performing one or more of the functions described herein. In some implementations, one or more of the sanitary appliance detection, leak detection, temperature compensation data processing operations, zeroing function, and other functions described herein and further described in detail below can be performed by the server, rather than replacing or excluding the controller from performing one or more of these functions.

[0074] In some embodiments, the material conduit can be a water pipe, such as a main water supply pipe for a residence, or a main water supply pipe for a portion of a large apartment building or office building. In some embodiments, the sensing assembly can be configured to be positioned on the conduit using an adjustment mechanism or device (e.g., clamping device, screw, set screw, spring, etc.). The sensing assembly can be coupled to conduits of different sizes (e.g., diameters from about 0.5 inches to about 1.5 inches). In some embodiments, "diameter" can refer to the inner or outer diameter of the conduit. The sensing assembly can be configured for conduits of different materials, such as copper, iron, steel, PVC, or PEX (cross-linked polyethylene).

[0075] In some embodiments, the sensing component can be coupled to both a cold water inlet pipe and a hot water inlet pipe. For example, the hot water inlet pipe can be downstream of a water heater. In other embodiments, the material conduit can be a gas conduit, such as a conduit for transporting natural gas. In some embodiments, the material conduit can be a conduit configured to transport crude oil, refined oil, or gasoline. The conduit can include pipes configured to transport or convey materials, said materials including one or more of the following: gas, liquid (free-flowing or viscous liquid), particles, liquid suspension, liquid solution, etc. In some embodiments, the material can include a refrigerant, such as a refrigerant used in HVAC systems.

[0076] In some embodiments, the controller may be electrically connected to an indicator or display configured to indicate whether the sensing component is properly positioned on the conduit to receive sensing information. For example, the controller may electronically communicate with an LED indicator that may include one or more LEDs built into or attached to the housing of the sensing component. In some embodiments, the LED indicator may be configured to indicate signal quality as the user installs the sensing component. This allows the sensing component to provide feedback to the user during installation without requiring the user to install a user interface on their mobile device to obtain such feedback prior to installing the physical sensing component.

[0077] For example, an LED indicator might display a first color when no signal is received, a second color when a weak signal is received, and a third color when a strong or optimal signal is received. Users can use this color change as a guide to determine when the sensing component has been correctly positioned during installation. In some implementations, the LED indicator might be configured to blink, flash, dim, or gradually brighten / dim rather than change color to reflect the quality of the received signal. For example, a low-amplitude, low-quality signal received during installation might cause the LED indicator to dim or gradually dim to a lower brightness, or to gradually brighten / dim at an accelerated rate, potentially prompting the user to place the sensing component in a better location on a pipe or conduit. Conversely, placing the sensing component to receive a stronger, higher-quality signal might cause the LED indicator to brighten, or cause the LED to remain constantly lit instead of gradually brightening / dimming.

[0078] In some implementations, the controller can determine the pipe diameter or material when determining the flow rate or other information. The pipe diameter can be measured in any suitable manner (e.g., with a "ruler") and the data input to the controller. In some implementations, the sensing components may include a proximity sensor configured to determine the position of an adjustment mechanism. The proximity sensor, electrically connected to the controller, can be configured to transmit the position of the adjustment mechanism to the controller, and the controller can be configured to determine the pipe diameter based on the sensed position. The proximity sensor may include one or more inductive, capacitive, optical (infrared (IR), photoelectric), magnetic, ultrasonic, or rotational sensors.

[0079] One challenge of clamp-on ultrasonic flow meters is providing accurate and reliable measurements across a wide variety of pipe (e.g., tube) shapes, sizes, and materials. Existing flow meters may limit the range of compatible pipe conditions and / or may require users to provide precise pipe information to calibrate their measuring devices. As described herein, sensing components can be advantageously calibrated to accommodate a wide range of pipe shapes and sizes without user input, although user input may still be optionally provided. In some embodiments, the sensing component can be configured to automatically detect pipe characteristics and can calibrate one or more parameters of the ultrasonic sensor of the sensing component based on the detected pipe characteristics.

[0080] To detect pipe characteristics, such as pipe diameter or material composition, sensing components can be configured, after installation, to perform different test programs by sending test signals to the pipe, including inputs that may affect the behavior of the sensing component's transducers in pipes of different sizes and how they sense data. These inputs may include, for example, the frequency, pulse number, signal amplitude, or noise mask of the transmitted ultrasonic signal. Exemplary outputs of the received signal may include time of flight, signal amplitude, signal frequency, or signal phase shift. Other relevant inputs / outputs that may vary in pipes of different sizes and materials can be measured during pipe calibration.

[0081] A controller for a sensing component can be configured to match various outputs from the sensing component with a library of verified outputs from sample pipes of different sizes or material properties. For example, the verified library may include, for instance, a set of verified outputs associated with sample pipes of different material properties, where each verified output corresponds to a test plan input. The controller can determine which pipe, with a specific characteristic or set of characteristics, the sensing component should be attached to based on the closest match between the received outputs from the sample output library and the sample pipe. For example, the controller may receive measured time-of-flight values ​​from the sensing component in response to a specific test plan and may determine that the measured values ​​best match verified outputs corresponding to copper pipes of a specific size. The controller can be configured to calibrate the sensing component based on the characteristics of the pipe that best matches the observed output.

[0082] In some implementations, the controller can be configured to determine the percentage of similarity between data values ​​associated with a specific pipe in a known library and measurement outputs received from the sensing component in response to a test plan. In some implementations, the controller can determine a “match” as the pipe in the known library that has the highest percentage of similarity or score relative to the measurement data. In some implementations, the controller can determine a “match” if the similarity percentage is within a pre-configured threshold. In some implementations, multiple test plans can be executed by the sensing component, and the closest match can be identified by the pipe in the known library that has the highest average percentage of similarity. By calibrating the sensing component based on the specific characteristics of the pipe, the accuracy of flow rate determination, null determination, temperature contribution, and other measurements, which will be described in more detail, can be improved.

[0083] In some embodiments, the sensing system can be configured to be easily coupled to a pipe in various orientations without the use of any tools. In some embodiments, the sensing system can be "integrated," meaning that the component located on the pipe can be a single part that is simply coupled to the pipe. In some embodiments, the sensing system may include a housing and an adjustment mechanism. The ultrasonic sensor may be located within the housing. The housing may have an opening through which the pipe can be coupled to the ultrasonic sensor. The housing opening may have components configured adjacent to the opening to receive inclined or bent edges of pipes of different diameters.

[0084] In some embodiments, a first ultrasonic transducer may be coupled to an upstream surface of a first upstream wedge, and a second ultrasonic transducer may be coupled to a downstream surface of a second downstream wedge. In some embodiments, the ultrasonic transducer may be coupled to an inclined surface of the wedge. The inclined surface can be defined by the angle between the lateral contact surface of the wedge and the inclined surface. In some embodiments, this angle can be between about 40° and about 78°. In some embodiments, this angle can be greater than or equal to about 40°, about 42°, about 44°, about 46°, about 48°, about 50°, about 52°, about 54°, or about 56°, and any one of about 58°, about 60°, about 62°, about 64°, about 66°, about 68°, about 70°, about 72°, about 74°, or about 76°. In some implementations, the angle may be less than or equal to about 42°, about 44°, about 46°, about 48°, about 50°, about 52°, about 54° or about 56°, and to any one of about 58°, about 60°, about 62°, about 64°, about 66°, about 68°, about 70°, about 72°, about 74°, about 76° or about 78°.

[0085] In some embodiments, the wedge may include a length of about 10 mm to about 46 mm. In some embodiments, the wedge may include a length greater than or equal to any one of about 10 mm, about 12 mm, about 14 mm, about 16 mm, or about 18 mm to about 20 mm, about 22 mm, about 23 mm, about 24 mm, about 25 mm, about 26 mm, about 27 mm, about 28 mm, about 30 mm, about 32 mm, about 34 mm, about 36 mm, about 40 mm, about 42 mm, or about 44 mm. In some embodiments, the wedge may include a length less than or equal to any one of about 12 mm, about 14 mm, about 16 mm, or about 18 mm to about 20 mm, about 22 mm, about 23 mm, about 24 mm, about 25 mm, about 26 mm, about 27 mm, about 28 mm, about 30 mm, about 32 mm, about 34 mm, about 36 mm, about 40 mm, about 42 mm, about 44 mm, or about 46 mm.

[0086] In some embodiments, the wedge may include a height of about 6 mm to about 30 mm. In some embodiments, the wedge may include a height greater than or equal to about 6 mm, about 8 mm, about 10 mm, about 11 mm, about 12 mm, about 13 mm, about 14 mm, about 15 mm, about 17 mm, about 19 mm, about 21 mm, about 23 mm, about 25 mm, about 27 mm, or about 29 mm. In some embodiments, the wedge may include a height less than or equal to about 11 mm, about 12 mm, about 13 mm, about 14 mm, about 15 mm, about 17 mm, about 19 mm, about 21 mm, about 23 mm, about 25 mm, about 27 mm, about 29 mm, or about 30 mm.

[0087] In some embodiments, the wedge may include a curved lateral contact surface configured to provide contact with pipes of various sizes and / or diameters. The curved lateral surface may be configured to make direct physical contact with the pipe. Alternatively, as described above, the curved lateral surface may make direct physical contact with a surface contact facilitator, which in turn may make direct physical contact with the pipe. In some embodiments, the lateral contact surface may include a width of about 4.0 mm to about 17 mm. In some embodiments, the lateral contact surface may include a width greater than or equal to any one of about 4.0 mm, about 5.0 mm, about 6.0 mm, about 7.0 mm, or about 8.0 mm to about 9.0 mm, about 10.0 mm, about 11.0 mm, about 12.0 mm, about 13.0 mm, about 14.0 mm, about 15.0 mm, or about 16.0 mm. In some embodiments, the lateral contact surface may include a width of less than or equal to any one of about 5.0 mm, about 6.0 mm, about 7.0 mm, or about 8.0 mm to about 9.0 mm, about 10.0 mm, about 11.0 mm, about 12.0 mm, about 13.0 mm, about 14.0 mm, about 15.0 mm, about 16.0 mm, or about 17.00 mm.

[0088] In some embodiments, the wedge-shaped contact surface may have lateral grooves on both sides. The wedge-shaped contact surface and the lateral grooves may be substantially parallel to each other. The lateral grooves located next to the contact surface can prevent ultrasonic sensors from detecting ultrasonic waves that are not aligned with the pipe centerline. In some embodiments, the lateral grooves may include a width of about 0.3 mm to about 3 mm. In some embodiments, the lateral grooves may include a width of less than or equal to about 0.5 mm, about 0.7 mm, about 0.9 mm, about 1.1 mm, about 1.3 mm, about 1.5 mm, about 1.7 mm, about 1.9 mm, about 2.1 mm, about 2.3 mm, about 2.5 mm, about 2.7 mm, about 2.9 mm, or about 3 mm. In some embodiments, the lateral groove may include a width greater than or equal to about 0.3 mm, about 0.5 mm, about 0.7 mm, about 0.9 mm, about 1.1 mm, about 1.3 mm, about 1.5 mm, about 1.7 mm, about 1.9 mm, about 2.1 mm, about 2.3 mm, about 2.5 mm, about 2.7 mm, or about 2.9 mm.

[0089] In some embodiments, the side groove may include a depth of about 0.4 mm to about 3.2 mm. In some embodiments, the side groove may include a depth greater than or equal to about 0.4 mm, about 0.6 mm, about 0.8 mm, or about 1.0 mm, about 1.2 mm, about 1.4 mm, about 1.6 mm, about 1.8 mm, about 2.0 mm, about 2.2 mm, about 2.4 mm, about 2.6 mm, about 2.8 mm, or about 3.0 mm. In some embodiments, the side groove may include a depth less than or equal to about 0.6 mm, about 0.8 mm, or about 1.0 mm, about 1.2 mm, about 1.4 mm, about 1.6 mm, about 1.8 mm, about 2.0 mm, about 2.2 mm, about 2.4 mm, about 2.6 mm, about 2.8 mm, about 3.0 mm, or about 3.2 mm.

[0090] In some embodiments, the first wedge may include a downstream inclined surface, and the second wedge may include an upstream inclined surface, and the inclined surfaces may be substantially mirror images of each other. In some embodiments, the wedge may include a channel or groove located on its upper portion. The upper channel may be substantially perpendicular to the lateral contact surface and the lateral groove adjacent to the lateral contact surface. Such a channel may be configured to provide adjustment or movement of the ultrasonic sensor assembly to aid in alignment with the conduit. In some embodiments, the ultrasonic transducer may be directly bonded to the wedge surface. This can be achieved using adhesives, ultrasonic welding, etc. Adhesives may include epoxy resins, acrylic adhesives, etc.

[0091] In some implementations, the adjustment mechanism can be configured to adjust or adjustably move the hanger toward and away from the ultrasonic sensor assembly located inside the housing. Adjustable movement of the hanger toward and away from the ultrasonic sensor assembly can position the sensor assembly optimally on the pipe. When the sensor assembly is correctly positioned on the pipe, the hanger helps to maintain and retain its correct position. In some embodiments, the adjustment mechanism can be configured to "push" the hanger toward the receiving pipe towards the housing and the ultrasonic sensor assembly, or alternatively, it can be configured to "pull" the hanger toward the housing and the ultrasonic sensor assembly. The adjustment mechanism can be automated or electrically powered, for example, automatically adjusting the tightness of knobs and screws.

[0092] In some embodiments, the housing having the ultrasonic sensor assembly therein may include a first well or bend configured to receive conduit and assist in positioning the conduit on the wedges of the ultrasonic sensor assembly. The housing may optionally be configured to automatically adjust the position and distance between each wedge, or the position and distance between the wedges may be manually adjusted. In some embodiments, the hanger may include a second well or bend also configured to receive conduit and assist in positioning the conduit on the wedges of the ultrasonic sensor assembly when the conduit is coupled to the sensing assembly. The first and second wells may be configured to be adjustable relative to each other to couple the conduit to the sensing assembly. This adjustable movement may be linear. The first and second wells may each include a “low point,” that is, the deepest point along the bend in the well section. In some embodiments, the first and second well low points may lie on a straight line and be configured to be adjustable toward and away from each other along said line.

[0093] In some embodiments, the housing may include a rectangular box-shaped portion configured to receive an ultrasonic sensor assembly. In some embodiments, the housing may include a compact shape, allowing it to be positioned on a pipe within a “narrow” space, such as a pipe located near one or more walls. In some embodiments, the housing, and any knob portion of the adjustment mechanism, may have a maximum dimension of about 70 mm to about 130 mm. In some embodiments, the maximum dimension may be greater than or equal to about 70 mm, about 80 mm, or about 90 mm, and to about 100 mm, about 110 mm, or about 120 mm. In some embodiments, the maximum dimension may be less than or equal to about 80 mm, or about 90 mm, and to any one of about 100 mm, about 110 mm, about 120 mm, or about 130 mm.

[0094] In some embodiments, the housing may have a width of about 60 mm to about 110 mm. In some embodiments, the housing may have a width greater than or equal to any one of about 60 mm, about 65 mm, about 70 mm, or about 75 mm to about 80 mm, about 85 mm, about 90 mm, about 95 mm, or about 100 mm. In some embodiments, the housing may have a width less than or equal to any one of about 65 mm, about 70 mm, or about 75 mm to about 80 mm, about 85 mm, about 90 mm, about 95 mm, about 100 mm, or about 110 mm. In some embodiments, the housing may include a depth of about 45 mm to about 95 mm. In some embodiments, the housing may include a depth greater than or equal to any one of about 45 mm, about 50 mm, about 55 mm, or about 60 mm to about 65 mm, about 70 mm, about 75 mm, about 80 mm, about 85 mm, or about 90 mm. In some embodiments, the housing may include a depth of less than or equal to any one of about 50 mm, about 55 mm, or about 60 mm to about 65 mm, about 70 mm, about 75 mm, about 80 mm, about 85 mm, about 90 mm, or about 95 mm.

[0095] The total space or volume that the shell may occupy may be, for example, between 400 cm³ and 800 cm³. In some embodiments, the shell may occupy a volume greater than or equal to about 400 cm³, about 450 cm³, about 500 cm³, about 550 cm³, about 600 cm³, about 650 cm³, about 700 cm³, or about 750 cm³. In some embodiments, the shell may occupy a volume less than or equal to about 450 cm³, about 500 cm³, about 550 cm³, about 600 cm³, about 650 cm³, about 700 cm³, about 750 cm³, or about 800 cm³.

[0096] In some implementations, the sensing component can be configured to be adjustablely and preferably located on or coupled to a pipe with an outer diameter (OD) of about 15 mm to about 51 mm. In some embodiments, the sensing component can be configured to be coupled to a pipe with an outer diameter greater than or equal to any one of about 15 mm, about 18 mm, about 21 mm, about 24 mm, about 27 mm, about 30 mm, or about 33 mm to about 36 mm, about 39 mm, about 42 mm, about 45 mm, or about 48 mm. In some embodiments, the sensing component can be configured to be coupled to a pipe with an outer diameter less than or equal to any one of about 18 mm, about 21 mm, about 24 mm, about 27 mm, about 30 mm, or about 33 mm to about 36 mm, about 39 mm, about 42 mm, about 45 mm, about 48 mm, or about 51 mm.

[0097] In some implementations, the controller may be located inside the housing. The controller may be electrically connected to the ultrasonic sensor assembly and also electrically connected to one or more LED indicators. The one or more LED indicators may be configured to indicate the intensity of the ultrasonic signal received by the controller from the ultrasonic sensor. For example, the LED indicators may be configured to indicate to a user whether and when the sensing assembly is correctly connected to the pipe, such as to receive optimal ultrasonic signals via the controller. For example, the LED indicators may display or flash a first color when the sensing assembly is connected to the pipe, and display or flash a second color when the assembly is correctly installed. The LED indicators may be configured to emit either the first or second color based on instructions from the controller.

[0098] In some embodiments, one or more LED indicator lights may be located outside the housing and visible to installers. In other embodiments, one or more LED indicator lights may be located inside the housing and may be configured to emit light from inside the housing onto pipes and / or hangers. In some embodiments, the housing may include a light tube at least partially located around an edge positioned around the interior of the housing. The light tube may be positioned around the entire circumference of the housing edge. One or more LED lights located inside the housing may be configured to emit light from inside the housing onto the light tube, illuminating the light tube. This makes the indicator light visible from any angle or virtually any angle or advantageous location.

[0099] In some embodiments, the light tube may include a portion configured to cover the interior of the housing. In some cases, the light tube may include a portion configured to function as a cover to position an ultrasonic sensor assembly within the housing. In some embodiments, one or more LEDs may be configured to illuminate the entire volume of the light tube. The light tube may include a translucent thermoplastic, such as polycarbonate. The light tube may be a solid, one-piece part or may include hollow portions.

[0100] Figure 10 Various components of the computer are depicted according to different implementations. As will be understood, the sensing component 100 and / or the controller of the sensing component 100 may include one or more components that will be described with respect to the computer 1000. The computer 1000 may be a part of a sensing component configured to monitor flow in a pipe, or one or more elements of the sensing component described herein may be able to communicate remotely with one or more computers (such as computer 1000).

[0101] Computer 1000 can be a host computer connected to a network. Computer 1000 can be a client computer or a server. Figure 10 As shown, computer 1000 can be any suitable type of microprocessor-based device, such as a personal computer, workstation, server, video game console, or handheld computing device, such as a telephone or tablet. The computer may include, for example, one or more of a processor 1001, a computer input device 1002, an output device 1003, a memory 1004, and a communication device 1005. Computer input device 1002 can generally correspond to any of the aforementioned devices and can be connected to or integrated with the computer.

[0102] The computer input device 1002 can be any suitable device that provides input, such as a touch screen or monitor, keyboard, mouse, or voice recognition device. The output device 1003 can be any suitable device that provides output, such as a touch screen, monitor, printer, disk drive, or speaker.

[0103] Memory 1004 can be any suitable means of providing storage, such as electrical, magnetic, or optical memory, including RAM, cache memory, hard disk drive, CD-ROM drive, magnetic tape drive, removable storage disk, or other non-transitory computer-readable media. Memory 1004 may include one storage device or more than one storage device. As used herein, the terms memory, RAM, and / or storage medium may refer to single and / or multiple devices that can individually, redundantly, and / or collaboratively store data and / or code / instructions, for example, in local and / or cloud storage environments. Communication device 1005 may include any suitable means capable of transmitting and receiving signals over a network, such as a network interface chip or card. Components of the computer may be connected in any suitable manner, such as via a physical bus or wirelessly. Memory 1004 may be a non-transitory computer-readable storage medium including one or more programs that, when executed by one or more processors (such as processor 1001), cause the one or more processors to perform the methods described herein.

[0104] The software 1006, which may be stored in memory 1004 and executed by processor 1001, may include programming that embodies, for example, the functions of this disclosure (e.g., as embodied in systems, computers, servers, and / or devices as described above). In some embodiments, the software 1006 may be implemented and executed on a combination of servers such as application servers and database servers.

[0105] Software 1006 or portions thereof may also be stored and / or transmitted in any computer-readable storage medium for use by or in connection with an instruction execution system, device, or apparatus (such as those described above), which may retrieve and execute instructions associated with the software from and execute such instructions. In the context of this disclosure, a computer-readable storage medium may be any medium, such as memory 1004, which may contain or store programs used by or in connection with an instruction execution system, device, or apparatus.

[0106] Software 1006 can also be propagated within any transmission medium for use by or in conjunction with an instruction execution system, device, or apparatus (such as those described above), which can retrieve and execute instructions associated with the software from and execute such instructions. In the context of this disclosure, a transmission medium can be any medium capable of communicating, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, device, or apparatus. Transmission readable media can include, but are not limited to, electrical, magnetic, optical, electromagnetic, or infrared wired or wireless transmission media.

[0107] Computer 1000 can be connected to a network, which can be any suitable type of interconnected communication system. The network can implement any suitable communication protocol and can be protected by any suitable security protocol. The network can include any suitable network link that enables the transmission and reception of network signals, such as wireless network connections, T1 or T3 lines, cable networks, DSL, or telephone lines.

[0108] Computer 1000 can implement any operating system suitable for operating a network. Software 1006 can be written in any suitable programming language (such as C, C++, Java, or Python). In various implementations, the application software embodying the functionality of this disclosure can be deployed in different configurations, such as in a client / server setup or via a web browser as, for example, a web-based application or web service.

[0109] Sanitary appliance testing The systems and methods of this disclosure can be configured to identify and distinguish water-using sanitary appliances and utilities (e.g., “sanitary appliances,” also referred to herein as “sanitary appliance detection” or “endpoint detection”). Sanitary appliance detection allows users to determine the water efficiency of each individual water-using sanitary appliance in their home to identify areas where efficiency can be improved. For example, the systems and methods of this disclosure can be configured to identify and distinguish flow rates and volumes associated with specific water-using fixtures or appliances. Current systems and methods can be configured to detect water flow rates multiple times per second and transmit the data over a network to a backend or cloud server according to a publication protocol. The backend or cloud server can be configured to analyze the data and identify specific water-using sanitary appliances, such as toilets, shower units, etc., thereby determining the source of the data.

[0110] Sanitary appliance detection can be challenging because water usage can overlap, and often does. For example, a faucet may be used when a toilet tank is filling or someone is showering. Furthermore, water pressure can vary, affecting flow rates and other factors. In some embodiments, the systems and methods described herein can utilize machine learning models to identify sanitary appliances by matching the entire function signature against previously obtained training data. In some embodiments, the systems and methods described herein can use machine learning models trained from general training data to perform sanitary appliance detection. In some embodiments, training data can be obtained from regular use of a state machine in a specific location or residence, allowing the machine learning model to identify flow rate segments specific to that residence, thereby improving accuracy and precision. Training typically includes recording the time of the flow rate function signature by marking the start and end of water usage at the sanitary appliance.

[0111] In some implementations, sanitary appliance detection can be performed on a cloud server using a publishing (e.g., compression) protocol and optional training phase. In some implementations, the publishing protocol can be implemented on / at the sensing component (e.g., at the sensing component controller). Published data can be transferred from the sensing component to the cloud server. Sanitary appliance detection can then be performed at the cloud server. Sanitary appliance detection may include detection using untrained data, and the controller can be configured to self-train based on long-term analysis of the collected data. In some implementations, sanitary appliance detection can be performed in the presence of overlapping water use and can be improved by training and refining a segmentation machine learning model, as will be described.

[0112] In some implementations, current systems and methods can be configured to utilize time, cumulative volume, and flow rate to detect sanitary appliances (e.g., toilets) using a multi-stage and multi-feature process. With the aid of a distribution (e.g., compression) protocol, sampled data can be segmented into time / duration segments or horizontal / flat segments of flow rate variation. In some implementations, current systems and methods involve sampling water flow rate 10 times or more per second, thus the distribution protocol can be used to selectively transmit only the most relevant data to a cloud server, as will be described.

[0113] Several characteristics that can be derived from each segment for use in sanitary appliance detection analysis can be obtained: duration, volume accumulation, direction (up / down), and start and end flow rates. In some implementations, flow rate changes can be observed through measurements taken closely over time, while the duration of the horizontal segment may be longer. The detection of a sanitary appliance (e.g., a toilet) can then be performed by matching each segment to a validated, typical segment of the specific sanitary appliance, or, for example, to previously observed characteristics of a specific sanitary appliance in the home during a training period. For example, a toilet typically has three phases: an initiation segment with specific volume, duration, direction, and flow rate changes (e.g., indicating the start of flushing); a horizontal segment with specific flow rate, volume, and duration (e.g., indicating the toilet flush valve is open); and an end segment with duration, volume, and flow rate changes (e.g., indicating the toilet flush valve is closed). Once a segment has been associated with or "matched" to correspond to a toilet flush, these segments can be excluded from other detections / matches. The characteristics of toilet segments can be identified using heuristics, by using predetermined values, or by using machine learning (ML) techniques such as gradient boosting to determine specific toilet flow rate function behavior. A horizontal segment can indicate the accumulated water volume and also indicate whether the segment likely belongs to the toilet bowl.

[0114] For example, sanitary fixtures can be matched (e.g., detected) by analyzing features (e.g., flow rate segments) to observe how closely said features match a trained general model (e.g., specific to a particular residence / building / sanitary fixture) or a trained specific model. In some implementations, if at any point the flow rate might fall below a predefined threshold rate (e.g., if the toilet end segment no longer changes downwards sufficiently), the segment may not match the toilet, and therefore the toilet may not be associated with corresponding water usage data. In some implementations, current systems and methods allow a mode in which sanitary fixture detection is performed in a flow-based manner using optimistic detection. Optimistic detection may involve initiating latent detection when a possible toilet start segment is detected, followed by relaxing a predefined threshold indicating the start of the toilet. This helps ensure that the relative importance or weight given to the toilet does not guarantee that the sanitary fixture detected from the flow rate segment is indeed a toilet.

[0115] Figure 2A An exemplary segmentation pattern 200 is shown. In some implementations, Figure 2A The flow rate segment pattern shown can indicate the toilet. As described herein and will be further described in detail below, the sensing component can sample the flow rate in short time segments by sending and receiving ultrasonic signals in the conduit (e.g., pipe) at fixed or variable time intervals or a combination of both. For example, the sensing component can be configured to sample the conduit (e.g., send and receive ultrasonic signals) every 100 to 200 milliseconds by default. When the controller detects a positive change in flow rate (e.g., an increase), the sensing component can be configured to sample the conduit more frequently to more accurately capture water usage events (e.g., toilet flushes) and reduce the overlap between detected water usage events and concurrent background water usage / other water usage events. When a negative change in flow rate (e.g., a decrease) is detected, the controller can configure the sensing component to reduce the sampling frequency back to its default state.

[0116] The segmentation machine learning model can be configured to separate flow rate segments based on whether these sampled flow rate segments have a fixed flow rate (indicated by a straight line) or a flow rate variation (indicated by a sloping line). The machine learning model can be configured to associate specific segment patterns with specific water-using sanitary appliances. The segmented data can also be stored by a controller and / or a cloud server and can be used to train the model, allowing it to be customized to detect water usage in specific sanitary appliances within a specific household / building.

[0117] In some implementation schemes, such as Figure 2AThe segment 202 shown can indicate the start of a toilet flush. For example, the start of a toilet flush can be marked by a sharp increase in flow rate over a short period of time, as shown in segment 202. The middle portion of a toilet flush can be characterized by a brief, flat segment 204, which can indicate a period during a toilet flush when the flush valve is fully open and the flow rate has stabilized. The end of a toilet flush can be characterized by a third flat segment 206 where the flow rate is lower than in segment 204. Then, a sloping segment 208 may appear, indicating a sharp drop in flow rate, which eventually levels off as the tank refills and the flush cycle ends.

[0118] Although Figure 2A The example shown illustrates an exemplary segmentation pattern corresponding to a toilet, but similar segmentation machine learning models can also be used to segment water flow data from other sanitary appliances (such as sinks, showers, etc.), which may have their own unique, identifiable segmentation patterns. In some implementations, the process may include separating the total measured flow rate from specific identified sanitary appliances. This helps ensure that the total volume is calculated correctly and that the volume of identified sanitary appliances is not counted repeatedly. The duration of each segment may be short, thus minimizing the possibility of overlap between sanitary appliances during the beginning or end of a segment. In some implementations, overlapping water usage events may be allowed and detected individually. For example, a shower may be used while the toilet is being refilled. Intermediate segments may not match water usage events and can be excluded. The cumulative volume can be estimated based on the identified horizontal segments. For example, the volume of each segment can be calculated by integrating the flow rate map between the start and end points of a flow rate segment, and the cumulative volume can be calculated over time by adding the volumes of each segment between two different horizontal segments over time. Therefore, even if overlap exists, the volume can be calculated correctly, and the user can view the historical accumulation of their water usage volume.

[0119] Detection can be performed on a cloud server, allowing segmentation machine learning models to self-train because historical and aggregated data from various residences can be analyzed. For example, a cloud server can also be configured with segmentation machine learning models that can associate specific segment patterns with specific water-using sanitary fixtures. Furthermore, using a cloud server advantageously allows for the storage of large amounts of data. Therefore, water flow data collected by ultrasonic sensor components can be stored on a cloud server and used to train and improve the machine learning model. Self-training may take hours or days to complete, during which more sanitary fixture detections will be performed. Non-repeating features can be judged as not being toilets, for example, as false positives. Repeated matching can enhance the machine learning model's ability to perform sanitary fixture detection based on segment patterns through repeated training.

[0120] In some implementations, the controller and / or cloud server may use an instance-based rule-based machine learning model to perform sanitary appliance detection, thereby performing segmentation and sanitary appliance detection, while a state machine may be used to perform rule adjustments based on the identified sanitary appliances. Figure 2B An exemplary sanitary appliance inspection processing pipeline 210 is illustrated. Data from an ultrasonic sensor assembly (e.g., the time between transmitting and receiving ultrasonic signals) can be fed into processing pipeline 210. The initial steps 212 through 214 of the processing pipeline are described herein and in more detail in the next section. Step 212 may involve determining the contributions of temperature and background noise (e.g., background water flow) to the Time-of-Flight (ToF) data collected by the sensing assembly. These contributions can be determined to improve the accuracy of machine learning models in determining “no-flow” states or zero-flow segments.

[0121] In step 214, the controller / cloud server can determine the flow rate and its moving average based on data collected by the ultrasonic sensor over a fixed or variable time period. Step 216 of the pipeline processing may involve calculating the time and volume corresponding to each determined flow rate. In step 218, the machine learning model segments the flow data, dividing the flow dataset into horizontal (or near-horizontal) segments corresponding to constant flow rates and transitional (sloping) segments corresponding to changes in flow rate, either increasing or decreasing. The machine model may use the original flow rate and its moving average for segmentation.

[0122] Segmentation applied using machine learning models can be beneficial for detecting individual water-using sanitary appliances, such as toilets, because the controlled valve mechanism is not continuously changing, and generally, much of the dataset has zero flow rate. This allows machine learning models to reliably detect the onset of water-using events (such as toilet flushing), i.e., rapid changes in flow rate. For example, in Figure 2A In the toilet flushing example, the observed flow rate was zero for a short period after the flushing started, and then returned to zero until the toilet was flushed again.

[0123] In step 218, a segment can be defined based on its start time, end time, type (e.g., direction (up, down, or horizontal)), flow rate at the beginning, flow rate at the end, and volume. In step 220, a state machine can be used to apply rules to each segment. The state machine essentially selects rules for determining the sanitary fixture based on segmented flow rate data from the sensor component operating in different states. In some implementations, the "state" of the sensor component can be "tracking / not tracking / end of tracking," "transition," "toilet / non-toilet," etc. For example, as... Figure 2CAs shown, determining that the sensing component is in a state where the toilet is not detected can cause the state machine to apply various rules. These rules help determine whether these segments correspond to other sanitary fixtures besides the toilet, or whether the sensing component is transitioning from a tracking state to a non-tracking state. The rules applied by the state machine are typically related to changes in time, volume, or flow rate. Although Figure 2C The rules shown are related to toilets, but similar rules can be applied to other sanitary appliances. In some implementations, one or more machine learning models can be used to improve the rules applied by the state machine to increase accuracy and reduce false matches, such as... Figure 2B As shown.

[0124] Figure 2C An exemplary state machine is shown in the figure. Although Figure 2C This is an example specific to toilet detection, but other state machines tailored for other sanitary appliances (such as shower units, washing machines, etc.) can be used. Multiple instances of the state machine can run in parallel. For example, a toilet state machine can run concurrently with a shower unit state machine. Additionally, a general state machine (e.g., a state machine with general rules not specific to a particular building, plumbing system, or sanitary appliance) can run concurrently with a state machine with rules tailored to a specific plumbing system (e.g., a particular residence or building). In some implementations, each segment identified by the machine learning model as the toilet's starting segment can prompt the generation of new tracking instances of the state machine.

[0125] Some rules may apply to situations where water flow from multiple sanitary fixtures overlaps, such as when showering and toilet flushing occur simultaneously. However, simpler rule sets may not account for volume accumulation when water flow from multiple fixtures overlaps. Generally, even in cases of overlap, rules for segments and time may match, but volume accumulation may need adjustment. Therefore, in some implementations, the rules applied by the state machine can be adjusted to account for overlapping water flow from multiple fixtures. For example, the time-limiting rules used by the state machine to measure cumulative volume can increase the duration of cumulative volume calculation when overlapping water flow is detected to address periods of lower water pressure caused by the simultaneous use of multiple fixtures. To address the drop in water pressure, the time limit for each segment is extended, but the cumulative volume remains unchanged.

[0126] In some implementations, the segmented data can be validated. Laboratory data validation showed that the instance-based rule-based machine learning model described herein correctly detected toilet flushes with 97% accuracy. In some implementations, the instance-based rule-based machine learning model can be validated by manually labeling real user data based on visual inspection. In some implementations, the model achieved 85% accuracy when validated in this way using a manually tuned rule base. Manual labeling can be useful when toilets are not detected in segmented data or when the number of detected toilets is excessive compared to other sanitary fixtures.

[0127] No flow rate detection and leak detection In some implementations, the controller / cloud server can also be configured to determine the flow rate contribution due to temperature, and continuous temperature-compensated data processing operations can be applied to the flow rate samples to obtain highly accurate measurement results without knowing the actual material temperature at any point in time.

[0128] Ultrasonic flow rate measurement devices can suffer from inaccuracies because changes in flow rate may be observed with variations in water temperature, when in reality only temperature changes are being interpreted as changes in flow. Ultrasonic flow rate sensors can measure flow rate using sound and the time it takes to travel a distance. The speed of sound changes with temperature. Therefore, if the distance between the transmitter and receiver is known, a calibrated ultrasonic sensor can determine the contribution of temperature to the flow rate by measuring the speed at which sound travels through the pipe.

[0129] Water in a stationary pipe or conduit can regulate its temperature according to the surrounding environment. When water flows, its temperature can quickly change to the source temperature, which is usually the ground temperature. Therefore, the ambient temperature is often higher than the ground temperature or the source water temperature.

[0130] The controller and / or cloud server of the sensing component described herein can be configured to report temperature at various flow rates, thus enabling accurate determination of relative changes through continuous temperature-compensated data processing operations, even without knowing the actual material temperature. Additionally, the controller and / or cloud server can include algorithms for determining when there is no flow (e.g., zero flow). In some embodiments, the sensing component can be calibrated based on a zero flow rate determination. For example, because flow rate is relative, an ultrasonic sensor may report a non-zero flow rate even without water flow due to inaccurate calibration or the effect of temperature on flow rate. The controller and / or cloud server can be configured with one or more algorithms to interpret these inaccuracies during calibration and provide a “zero flow rate,” such as “no flow rate.” By comparing flow rate samples to a zero flow rate, subtle flow rate changes caused by minor leaks or problems with water sanitation fixtures can be accurately and precisely detected.

[0131] In some implementations, the algorithm can be configured to continuously search for zero flow rate, such as no flow rate, over a period of time, for example, by analyzing flow rate samples obtained from the ultrasonic sensor assembly over a two-minute time period. Figure 9 As shown. Flow can be determined by multiplying the time difference (“tdiff”) between the upstream and downstream ultrasonic signals by a constant associated with the corresponding pipe type and / or size. The duration of sample acquisition and analysis (e.g., the time length for the algorithm to search for zero) can be fixed or variable. For example, the time period may be long enough to perform frequency analysis (e.g., histogram analysis) on flow rate samples, typically hundreds of samples. The sample acquisition frequency can also be fixed or can be varied. While non-zero flow rates detected using time-of-flight data from ultrasonic sensors may not be precisely constant due to noise, periods of zero flow can be identified because they produce highly consistent flow rates. Ultrasonic flow meters can provide a normal distribution, making them probabilistic about the actual flow rate.

[0132] Figure 3A and Figure 3B An exemplary distribution of flow rate is shown in the figure. Figure 3A The time segment may be rejected as having no flow rate because it is not normally distributed, has multiple peaks, and is unevenly distributed around a single value. Figure 3B The time-segmented flow rate distribution can be accepted as having no flow rate because it is normally distributed, which indicates stable flow rate.

[0133] The time-segmented flow rate distribution can be analyzed using statistical analytics by the controller / cloud server. If the distribution is highly normal, it can be considered zero flow, e.g., no flow. A second algorithm can be used to determine if the center of the distribution is close to the lowest sample found in a given time period, thus ruling out a constant actual flow rate. For example, if the center of the normal distribution is near a sample with a higher flow rate, this can indicate that it is an actual flow rate, not a zero flow rate. When a zero flow rate is determined, the current temperature and flow rate can be recorded. This process can be repeated until many (e.g., 20 or more) temperature / flow rate readings are obtained. For example, zero detection can be performed and repeated periodically using a histogram or frequency plot of all sampled data. For example, it can be sampled once every 20 milliseconds, or 3000 times per minute.

[0134] Regression analysis of the collected data provides curves, such as those shown in Figure 3D. Such curves can be applied continuously to each sample. By measuring the flow rate and recording the temperature, the temperature can be mapped onto the curve to determine the contribution of temperature to the flow rate. This contribution can then be removed from the measured flow rate through continuous temperature-compensated data processing to compensate for any artificial variations in the measured flow rate caused by temperature.

[0135] In some implementations, temperature can be calculated using an algorithm that measures the velocity of sound passing through the pipe or material, for any specific known pipe size, and the sensing components can be calibrated in a flow-free (zero-flow) environment. In this way, it is possible to generate... Figure 3C The compensation curve is obtained and applied to flow rate determination through continuous temperature compensation data processing. Since the temperature analysis focuses on the relative change in temperature, a temperature sensor is not required to measure the actual temperature.

[0136] In some implementations, sensing components as described herein can be used to identify leaks or malfunctions, such as water leakage. Methods may include statistical analysis and real-time flow rate analysis. The source of the leak or malfunction can be determined by employing sanitary appliance detection and endpoint detection and / or advanced modeling as described in the previous section. Water leakage can include a continuous, slow, moderate, high, or very high non-stop flow of water. Typically, leakage can include a constant flow rate that may increase over time. A decrease in flow rate at the leak over time is possible but uncommon. Leaks can also include temporary leaks, such as toilet valve malfunctions. Such leaks may occur immediately after flushing the toilet.

[0137] In some implementations, leak detection may involve communicating data from the sensing component to a backend cloud server configured to perform analysis. This allows users to perform real-time monitoring and provides the ability to dynamically change leak detection parameters and employ one or more machine learning models. For example, similar machine learning models used for performing segmentation and sanitary appliance detection can be used to detect flow characteristics associated with different leaks or defects in parts. For instance, in the case of a leaking toilet float valve, a machine learning model could be configured to detect the leak based on the fact that the flow characteristics of the leaking float valve are the same as those of a normal toilet, but differ in volume, duration, and flow rate characteristics. Distribution protocols (e.g., compression protocols) can be utilized to reduce the amount of data sent to the server while maintaining high detail during water change events.

[0138] Leak detection can be performed using sensing components as described herein by measuring flow rates and using these flow rates to generate aggregated volume information. High-frequency sampling of flow rates allows for continuous calculation of total volume based on flow rate data, in parallel with segmented volumes calculated during sanitary appliance testing. In some embodiments, leaks can be detected if the flow rate is in an atypical range for a duration longer than the water usage period. Leaks can also be detected if the flow rate is above no flow (e.g., a calibrated “zero”) or above a predefined threshold for a sustained period. In some embodiments, leak detection can be performed by observing a continuous moving average of the flow rate. The moving average may be useful for detecting smaller, more continuous leaks compared to the starting point of a flow rate variation segment. The moving average can be observed over a duration of 30 seconds or more, and can be used to determine whether the flow rate is above or below no flow rate within a predetermined time period.

[0139] In some implementations, the controller of the sensing component can perform leak detection; however, the need to access high-frequency sampling data results in large amounts of data being transmitted to backend servers via networks such as WiFi, cellular networks, and LoRa. The controller's memory may be limited and unknown to the central system, so historical analysis and / or real-time tracking may not always be possible. The use of machine learning techniques is also more restricted when using the controller. Therefore, transmitting water flow data from the sensing component to a backend server (such as a cloud server) can make it easier for users to use the sensing component to track their historical water usage and / or their real-time water usage.

[0140] To reduce the amount of data to be transmitted and the associated costs of transmitting and storing data on cloud servers, sensing components can employ a "publishing" (e.g., compression) protocol. Slowly changing water flow samples may be filtered out, as they are more likely to correspond to no flow rate. In some implementations, the publishing protocol can be configured to send the total volume reading with each sample as a moving average of approximately 30 seconds. Utilizing the long moving average along with the volume, slow, persistent leaks can be observed, even without using a single flow rate data point for determination. Leak analysis can be performed on cloud servers, allowing for the tracking of very long trends, such as days, weeks, months, years, etc. Efficient leak detection can be performed while maintaining efficient data publishing during data transmission to backend servers.

[0141] In some implementations, large leaks can be observed as water usage exceeding predefined and / or configurable thresholds. Normal flow rate data can be used to detect larger leaks or sanitary appliance malfunctions compared to moving averages, because the onset of a large leak or malfunction may be large enough to trigger a release protocol that sends flow rate data to a cloud server.

[0142] Detecting leaks or malfunctions associated with water-using fixtures or appliances can be achieved by first identifying the water-using fixtures and appliances as described in the section on sanitary appliance testing, and then performing small-scale, continuous leak identification as described above. The source of the leak can be determined by linking the initiation of the leak to the water-using fixtures and appliances. This can be performed on a cloud server and can employ machine learning techniques, as described above regarding sanitary appliance testing and instance-based rule-based models.

[0143] In some implementations, current systems and methods employ data compression or distribution protocols when transmitting data from the sensing component controller to the cloud server. To observe flow rate change events, flow velocity data can be measured at a high-frequency millisecond timescale (e.g., a scale of approximately 100 ms to approximately 200 ms). Because ultrasonic sensors are probabilistic, a bell-shaped distribution centered on the actual flow rate is observed, such as... Figure 4A As shown and as mentioned above regarding Figure 3B Similarly, based on this distribution, assumptions can be made about the detected flow rate signal-to-noise ratio (SNR). For example, Figure 5A The histogram comprises 3000 data points. Assuming zero flow for at least a period of time, the lowest local maxima (e.g., the initial estimate of the signal-to-noise ratio) will be zero. Therefore, Figure 5A The graph in the graph is asymmetrical and tilted to the right. If the noise is symmetrical, the range to the left of the local maximum can be considered noise, and the standard deviation will provide a noise estimate to determine the signal-to-noise ratio.

[0144] Since the SNR is known, it can be determined whether the observed flow rate change is due to noise or an actual flow rate change caused by a leak or water usage event. In some implementations, the SNR can be used when employing a data distribution protocol. The distribution protocol may use a variable time between "distributed" samples, for example, from approximately 20 ms to approximately 30 seconds. In some implementations, "distributed" samples may refer to samples selected for transmission to the server. In some implementations, a maximum of 30 seconds, or longer or shorter time periods, can be used, allowing the cloud server to detect missing data and determine if there may be a problem with the sensing component.

[0145] In some implementations, the publishing protocol may include interpolation between data points. For example, flow data may be sampled over a 200 ms time interval. The publishing protocol may measure a moving average of the flow data, for example, every 2 seconds, 10 seconds, and 30 seconds. Each sample may include a timestamp, a flow velocity observation, and a volumetric observation. Since the time corresponding to a published sample can change, the timestamp can be utilized. Figure 4BThe illustration shows samples captured according to the publishing protocol and published samples, where circles represent samples at fixed time intervals, squares represent 2-second moving averages, and solid circles represent published samples.

[0146] In some implementations, a flow rate change can be defined if the current sample differs significantly from the 2-second moving average of all previous samples. This feature can be defined as a ratio to the signal-to-noise ratio. For example, a significant difference could be a signal-to-noise ratio difference of approximately two times between the current sample and the 2-second moving average of the previous sample. Figure 4B In this context, the moving average 450 represents the first sample that is "significantly different." When the first sample is found to be significantly different, the previous sample 451 can also be selected for publication, thus "closing" the segment before the change begins and therefore capturing the flat portion of the segment as well. If the time between previous values ​​exceeds a fixed interval (e.g., 2 seconds), the actual value can be replaced with the moving average, because the actual sample may not accurately represent a longer time interval, while the moving average provides a closer approximation.

[0147] According to the publishing protocol, when a sample is found to be significantly different from the 2-second moving average, subsequent samples can continue to be published until the 2-second moving average "catches up" with the previous moving average. While this example uses a 2-second moving average, the time period used to determine the moving average can be varied. In some implementations, the moving average can be determined over a time period of 2 to 30 seconds. In some implementations, the moving average can be determined over a time period greater than or equal to 2 seconds, 5 seconds, 10 seconds, 15 seconds, 20 seconds, or 25 seconds. In some implementations, the moving average can be determined over a time period less than or equal to 5 seconds, 10 seconds, 15 seconds, 20 seconds, 25 seconds, or 30 seconds. For publishing protocol calculations, if the flow rate and / or volume do not change significantly, it may be necessary to use the moving average corresponding to the "flat" portion of the graph as the moving average.

[0148] For example, a sensing component can sample data at a fixed time period of approximately 200 ms, but can calculate a moving average of the flow rate over a 30-second time interval. Therefore, data can only be transmitted to the cloud server every 30 seconds, achieving highly “compressed” data transmission in some implementations. For instance, if each sample has 5 flow rate values, with sampling every 200 ms, there are 1500 sampled values ​​per minute. If the water flow rate does not change significantly within 30 seconds, publishing data as a moving average every 30 seconds will produce 10 values ​​per minute, or a data compression rate of approximately 99%.

[0149] For example, a caveat when publishing data based on a 30-second moving average is that slowly changing values ​​will not trigger a publishing protocol because the flow rate changes not quickly enough to reflect a significant difference from the moving average. This can be addressed using "creep detection," which compares the latest sample to the previously published moving average. If the difference between the latest sample and the previously published moving average exceeds a predetermined ratio of the signal-to-noise ratio (SNR) (e.g., 2x SNR or higher), this can trigger a single point to be published. In this way, a gradual "climbing" or decreasing trend in the moving average (potentially indicating persistent small leaks) can be published to the server and made visible to users. In some implementations, the increase or decrease in cumulative volume over a given time period (e.g., 30 seconds) can also be used during creep detection. As long as samples are used for a sufficient period, the volume is not affected by probability values. This is because over time, over-reporting and under-reporting of volumes can cancel each other out.

[0150] The situation where rapidly changing values ​​happen to cross the moving average can be addressed by artificially extending any detected change condition. For example, when a measurement deviates from the moving average, both the current and previous samples can be published. When the current sample again aligns with the previous moving average, the next two measurements can also be published, regardless of their differences from the moving average, thus enabling the capture and publication of rapid changes in flow rate.

[0151] If there is no significant difference between the measured flow rate and the moving average, the sensing component can be configured to automatically publish the moving average every 30 seconds. If the protocol simply uses the current value, this will result in a semi-random value (a probability value around the actual value). For this, a 30-second moving average can be used. If the time since the last published point is greater than 20 seconds but less than 30 seconds, a 10-second moving average can be used. If this time is greater than 4 seconds but less than 10 seconds, a 2-second moving average can be used.

[0152] Therefore, the sensing components described herein can utilize sanitary fixture detection and leak detection to collect data on user water usage. By employing zero-detection and automatic temperature calibration as described herein, water usage events at sanitary fixtures and leaks associated with changes in water flow rate can be identified more accurately and precisely. The publishing protocol also enables a compact and accurate representation of water usage by ensuring that flow rate changes associated with actual flow (rather than noise) are reported to the server for analysis and displayed via the user interface.

[0153] In some implementations, cloud servers and / or controllers can be configured to report to users the status of the water supply and drainage system as normal or abnormal (e.g., leaks or malfunctions). For example, a computing device such as a smartphone or laptop can be linked to the controller. The controller can be configured to transmit data to the computing device. The computing device can have a graphical user interface configured to display data, including water usage data for water-using devices and appliances, water efficiency data, efficiency data over a period of time, etc. The controller can be configured to communicate bidirectionally with the cloud server, and the cloud server can be configured to communicate bidirectionally with devices such as smartphones or laptops. This allows users to monitor their water usage from a single point, such as a smartphone. The graphical user interface can be configured to provide real-time water usage, as well as historical water usage and trends, for specific water-using devices and appliances and for the entire residence. By streamlining the display of water usage information to users in a simple manner, graphical user interfaces such as those described herein can help users identify areas where water usage can be reduced or efficiency can be improved, thereby helping to encourage users to change their behavior and develop more water-saving habits.

[0154] The computing device can be configured to alert users in the event of abnormal water usage, suspected leaks, or malfunctions. In some implementations, the alerts can be adjusted by the user. For example, known problems can be muted, or the sensitivity level can be adjusted so that the user only receives the most relevant alerts. The computing device interface can be configured to allow users to set and customize water-saving goals and objectives.

[0155] Figure 6A The user interface dashboard is shown according to some implementation schemes. Item 601 may represent global time-scale navigation, initially set to "monthly". Item 602 is timeline navigation. Item 603 may indicate flow rate and can display on / off / high usage. Item 604 may display monthly water consumption (e.g., gallons, liters, etc.). Item 605 may display average monthly water consumption. Item 606 may display "water conversion," which is a contextual reference using everyday items to interpret water volume; for example, a daily view might show a water bottle, a weekly view might show a bathtub, etc. Item 607 may provide a usage graph, in which case monthly usage is displayed in weekly navigable intervals, with coded colors indicating above-average usage. Item 608 is a timeline and navigation interaction that can be used to slide to the previous or next month. Item 609 may show usage trends. Item 610 may provide a breakdown of water usage for sanitary appliances. Items 611 and 612 may be video elements. For example, item 611 could be a looping head video with its tilt angle configured to respond to gestures, while item 612 could be a looping tail video. Figure 6B An exemplary real-time view of item 607 / 608 is provided, showing the current flow rate 613, the flow rate graph 614, and the scale in minutes 615.

[0156] Figure 7A and Figure 7B Views of the main notification page and subpages are provided separately for the user interface. Item 701 can provide high-priority activity alerts, such as leaks. Item 702 can provide a list of activity alerts. Item 703 can display notifications and alerts with medium / low priority. Item 704 can be a filter that allows filtering by notification type. Item 705 can display quick details of the alert, such as water volume, flow duration, and flow rate. Item 706 can list the “severity” of the event. Item 707 can show the date / time when the component began detecting the event. Item 708 can be a “consciousness investigation.” Item 709 allows the user to turn event updates on or off. Item 710 allows the user to select the alert frequency. Item 711 “Next Steps” can provide tips on how to resolve the issue. Item 712 can provide event history and details. Figure 7B As shown, users can optionally select “tagged” events to manually monitor and report / verify detected leaks and their severity, thereby improving the accuracy of the sensing components.

[0157] In addition to displaying customizable water usage reports to users, the user interface, as described herein, can also allow users to provide feedback on the quality of leak detection / sanitary appliance detection by the sensing components. For example, after a toilet flush is detected and displayed on the user interface, the interface may prompt the user to confirm whether the toilet was actually flushed during the stated time period or whether it was a false alarm. This may occur for events that the user has "marked." User-provided feedback can be transmitted from the mobile device back to the sensing components and / or cloud servers, and user feedback can be used to validate or improve one or more machine learning models used in sanitary appliance or leak detection.

[0158] Figure 8A , Figure 8B , Figure 8C and Figure 8D Views of an interactive water level indicator for comparing water consumption are shown according to some implementation schemes. The current indicator can be configured to visually represent the user's water consumption and provide a comparison with the average consumption over a selected time period. The interface may have a video water level background with dynamically changing values. The values ​​shown are the current monthly water consumption and the monthly average water consumption. These figures can be configured to "rise" or "fall" based on the comparison between the current water consumption and the monthly average water consumption.

[0159] Figure 8A An example is shown where current water usage may be below the monthly average. Figure 8B An example is shown where current water usage may be higher than the monthly average. Figure 8CThis illustrates an example where current water usage might be the same as the monthly average. As shown, the relative positions of the numbers and the water level background can change, providing a visual representation of consumption trends. The monthly consumption value determined based on the user's water usage can drive the displayed water level, causing it to rise or fall accordingly. The water level background can include slow waves to simulate a realistic water surface.

[0160] Furthermore, the user interface can incorporate the device's motion sensing capabilities. When a user rotates their smartphone, the water surface can follow and change its angle based on the user's gestures, thus creating an interactive and immersive display. Figure 8D An example is shown where the water level can follow and change its angle based on the smartphone's location. The rising and falling water levels displayed on the user interface can be configured to directly correspond to the user's current consumption within a selected time period (such as the current month). This dynamic presentation allows users to intuitively understand their water usage patterns and easily compare them to average consumption over the selected time period. The use of video backgrounds, dynamic numerical values, and device motion sensors is configured to enhance user understanding and engagement with water consumption monitoring.

[0161] The terms “connection” or “link” can mean that one element is “attached to” or “associated with” another element. A connection or link can mean a direct link or a link via one or more other elements. An element can be connected to another element sequentially or non-sequentially via two or more other elements. Referring to the term “via an element,” the word “via” can mean “through” or “through” an element. A connection, link, or “associated with” can also mean that elements are not directly or indirectly attached, but rather operate “together,” meaning that one element can function together with another element.

[0162] The terms "upstream" and "downstream" indicate the direction of gas or fluid flow, meaning that the gas or fluid will flow from upstream to downstream.

[0163] The term “orientation” in relation to an attachment point can mean exactly at that location or point, or alternatively, it can mean closer to the point than another different point, for example, “orientation to the center” means closer to the center than the edge.

[0164] The term "like" means similar, but not necessarily identical. For example, "ring-shaped" usually means that the shape resembles a ring, but it is not necessarily a perfect circle.

[0165] The article “a / kind” in this text refers to one / kind or more (i.e., at least one / kind) grammatical objects. Any range referenced herein includes end values. The term “about” used throughout the text is used to describe and account for small fluctuations. For example, “about” can mean that a value can be modified by ±0.05%, ±0.1%, ±0.2%, ±0.3%, ±0.4%, ±0.5%, ±1%, ±2%, ±3%, ±4%, about ±5%, or ±10%. All numerical values ​​are modified with the word “about”, whether explicitly stated or not. Numerical values ​​modified by the term “about” refer to a specific identified value. For example, “about 5.0” includes 5.0.

[0166] The term “substantially” is similar to “about” because the defined term may differ from the definition, for example, by ±0.05%, ±0.1%, ±0.2%, ±0.3%, ±0.4%, ±0.5%, ±1%, ±2%, ±3%, ±4%, ±5%, or ±10%; for example, the term “substantially vertical” can mean a 90° vertical angle or “about 90°”. The term “roughly” can be equivalent to “substantially”.

[0167] Even if not explicitly stated otherwise, the features described in one embodiment of this disclosure may be used in conjunction with other embodiments.

[0168] The embodiments disclosed herein include any and all parts and / or portions of the embodiments, claims, specification, and drawings. The embodiments disclosed herein also include any and all combinations and / or sub-combinations of the embodiments.

Claims

1. A sensing assembly configured to monitor flow in a pipe, the sensing assembly comprising: An ultrasonic sensor assembly including an ultrasonic sensor, wherein the ultrasonic sensor assembly is configured to be physically coupled to the conduit, and wherein the ultrasonic sensor is configured to transmit and receive ultrasonic signals through the conduit; and A controller electrically connected to the ultrasonic sensor assembly, wherein the controller is configured to: Determine the time of flight (ToF) of the ultrasonic signal. The flow rate in the pipe is determined based on the flight time. The storage includes flow data comprising the determined flow rate, and A segmentation machine learning model is used to identify individual sanitary appliances associated with the flow data.

2. The sensing component of claim 1, wherein identifying a single sanitary appliance associated with the flow data comprises: Sampling of the flowing data, and The sampled data is divided into multiple time-duration segments. The first time duration segment of the plurality of time duration segments is characterized by no flow rate change, and the second time duration segment of the plurality of time duration segments is characterized by having a flow rate change.

3. The sensing component of claim 2, wherein identifying a single sanitary appliance associated with the flow data includes analyzing one or more of the duration, volume accumulation, direction, start flow rate, and end flow rate in one or more of the plurality of time duration segments.

4. The sensing component of claim 2, wherein identifying a single sanitary appliance associated with the flow data includes applying one or more rules to the time duration segment using a state machine.

5. The sensing component of claim 1, wherein determining the flow rate includes applying temperature-compensated data processing operations to the time of flight of the ultrasonic signal.

6. The sensing component of claim 5, wherein the temperature compensation data processing operation is based on a set of no-flow rates determined at various temperatures.

7. The sensing component of claim 1, wherein the flow data includes one or more of volume data, flow rate moving average data, and timestamps of the ultrasonic signals.

8. The sensing component of claim 1, wherein identifying a single sanitary appliance associated with the flow data includes determining a no-flow rate based on the flow data.

9. The sensing assembly of claim 8, wherein the controller is configured to use the no-flow rate to calibrate the ultrasonic sensor assembly.

10. The sensing component of claim 8, wherein the controller is configured to identify a leak or malfunction of the individual sanitary appliance based on the flow data and the no-flow rate.

11. The sensing component of claim 10, wherein identifying a leak or malfunction of the individual sanitary appliance by the controller includes identifying volumes above a defined threshold from the volume data.

12. The sensing component of claim 8, wherein the controller is configured to communicate with a cloud server, and wherein the cloud server is configured to identify a leak or malfunction of the individual sanitary appliance based on the flow data and the no-flow rate.

13. The sensing component of claim 12, wherein identifying the leak or malfunction associated with a single sanitary appliance includes transmitting the flow data and flow rate moving average data to the cloud server and receiving water usage trend analysis from the cloud server.

14. The sensing component of claim 1, wherein the controller is configured to determine the speed of sound in the pipe based on the ultrasonic signal.

15. The sensing component of claim 1, wherein the controller is configured to transmit the streaming data to the cloud server in accordance with a publishing protocol.

16. The sensing component of claim 15, wherein the publishing protocol includes transmitting the flow data when the volume above the defined threshold is identified.

17. The sensing component of claim 15, wherein the publishing protocol includes publishing a sample of flow data with a directly preceding timestamp upon identifying a flow rate change above a defined threshold.

18. The sensing component of claim 15, wherein the publishing protocol includes transmitting streaming data from the controller to the cloud server at variable time intervals.

19. The sensing component of claim 15, wherein the publishing protocol takes into account the signal-to-noise ratio (SNR).

20. The sensing component of claim 1, wherein the sensing component is configured to transmit and receive ultrasonic signals in millisecond timescales.

21. The sensing component of claim 1, wherein the sensing component is configured to transmit and receive ultrasonic signals until the moving average catches up with the flow rate in the flow data.

22. The sensing component of claim 1, further comprising an LED indicator configured to indicate the intensity of the ultrasonic signal.

23. The sensing assembly of claim 1, wherein the ultrasonic sensor assembly is further configured to transmit a test signal through the conduit and receive a response to the test signal, the response including one or more test outputs.

24. The sensing component of claim 23, wherein the controller is configured to determine the material of the pipe based on the one or more test outputs.

25. The sensing component of claim 23, wherein the controller is configured to determine the diameter of the pipe based on the one or more test outputs.

26. A sensing assembly configured to monitor flow in a pipe, the sensing assembly comprising: An ultrasonic sensor assembly including an ultrasonic sensor, wherein the ultrasonic sensor assembly is configured to be physically coupled to the conduit, and wherein the ultrasonic sensor is configured to transmit and receive ultrasonic signals through the conduit; and A controller electrically connected to the ultrasonic sensor assembly, wherein the controller is configured to: Determine the time of flight (ToF) of the ultrasonic signal. The flow rate and no flow rate in the pipe are determined based on the flight time. The storage includes flow data with the determined flow rate and no flow rate. The ultrasonic sensor is calibrated using the no-flow rate, and The flow data is used to identify leaks or malfunctions in individual sanitary appliances.

27. The sensing component of claim 26, wherein determining the flow rate includes applying temperature-compensated data processing operations to the time of flight of the ultrasonic signal.

28. The sensing component of claim 27, wherein the temperature compensation data processing operation is based on a set of no-flow rates determined at various temperatures.

29. The sensing assembly of claim 26, wherein the controller is configured to determine the speed of sound in the pipe based on the ultrasonic signal.

30. The sensing component of claim 26, wherein the controller is configured to communicate with a cloud server, and wherein the cloud server is configured to identify the leakage or malfunction of the individual sanitary appliance based on the flow data and the no-flow rate.

31. The sensing component of claim 30, wherein the cloud server identifies a leak or malfunction of the individual sanitary appliance by transmitting the flow data from the controller to the cloud server.

32. The sensing component of claim 26, wherein the flow data further includes volume data, flow rate moving average data, and timestamps of ultrasonic signals.

33. The sensing component of claim 32, wherein identifying a leak or malfunction of the individual sanitary appliance by the controller includes identifying volumes above a defined threshold from the volume data.

34. The sensing component of claim 26, wherein the controller is configured to transmit the streaming data to the cloud server in accordance with a publishing protocol.

35. The sensing component of claim 34, wherein the controller is configured to transmit the flow data according to the publishing protocol when the volume above the defined threshold is identified.

36. The sensing component of claim 34, wherein the publishing protocol includes transmitting streaming data from the controller to the cloud server at variable time intervals.

37. The sensing component of claim 34, wherein the controller is configured to transmit the flow data according to the publishing protocol when it detects a flow rate change above a defined threshold.

38. The sensing component of claim 37, wherein the publishing protocol includes publishing a sample of flow data with a directly preceding timestamp upon identifying a flow rate change above a defined threshold.

39. The sensing component of claim 34, wherein the publishing protocol takes into account the signal-to-noise ratio (SNR).

40. The sensing component of claim 26, wherein identifying the leak or malfunction associated with a single sanitary appliance includes the cloud server transmitting volume data and flow rate moving average data to the cloud server and performing water usage trend analysis.

41. The sensing assembly of claim 26, wherein the ultrasonic sensor is configured to transmit and receive ultrasonic signals in millisecond timescales.

42. The sensing assembly of claim 26, wherein the ultrasonic sensor is configured to transmit and receive ultrasonic signals within a fixed time period of about 100 milliseconds (ms) to about 200 ms.

43. The sensing assembly of claim 26, wherein the ultrasonic sensor is configured to transmit and receive ultrasonic signals until the moving average catches up with the flow rate.

44. The sensing component of claim 26, wherein the controller is further configured to use a segmentation machine learning model to identify individual sanitary appliances associated with the flow data.

45. The sensing component of claim 44, wherein identifying a single sanitary appliance associated with the flow data comprises: Sample the flowing data; The sampled data is divided into multiple time-duration segments, wherein a first time-duration segment is characterized by no flow rate change, and a second time-duration segment is characterized by flow rate change.

46. ​​The sensing component of claim 44, wherein identifying a single sanitary appliance associated with the flow data comprises analyzing one or more of the duration, volume accumulation, direction, start flow rate, and end flow rate in one or more of the plurality of time duration segments.

47. The sensing component of claim 44, wherein identifying a single sanitary appliance associated with the flow data includes applying one or more rules to the time duration segment using a state machine.

48. The sensing component of claim 26, further comprising an LED indicator configured to indicate the intensity of the ultrasonic signal.

49. The sensing assembly of claim 26, wherein the ultrasonic sensor assembly is further configured to transmit a test signal through the conduit and receive a response to the test signal, the response including one or more test outputs.

50. The sensing component of claim 49, wherein the controller is configured to determine the material of the pipe based on the one or more test outputs.

51. The sensing component of claim 49, wherein the controller is configured to determine the diameter of the pipe based on the one or more test outputs.

Images