A low-cost, non-invasive flow measurement and leakage detection device for closed conduits

Abstract

The present invention provides a low-cost, non-invasive device (100) for measuring flow and leakage detection of fluids within closed conduits. The invention utilizes affordable a Micro-Electro Mechanical System (MEMS) based Inertial Measurement Units (IMUs) sensor (102) in the device to non-invasively measure fluid flow and leakage of fluids which exhibit turbulence or flow-induced-vibrations within a conduit. This device fin directly estimates flow rate or velocity based on the concept of Reynolds number in fluid dynamics. Further, this device is easy to install on pipes of all dimensions, requires minimal calibration, has no moving parts in contact with the fluid, can be integrated with Geographic Information System (GIS) software, and possesses advanced metering infrastructure to support remote reading and automated bill generation.

Description

A LOW-COST, NON-INVASIVE FLOW MEASUREMENT AND LEAKAGE DETECTION DEVICE FOR CLOSED CONDUITS

[0001] The field of invention generally relates to Smart Metering and The Internet of Things (IoT). More specifically, it relates to a low-cost, non-invasive flow measurement and leakage detection device for closed conduits.

[0002] Flow measurement is a critical process that extends its significance across various sectors, playing a pivotal role in ensuring safety, regulatory compliance, and operational success. In sectors such as chemical processing, pharmaceuticals, and food and beverage, precise flow measurements are imperative for maintaining consistent product quality and upholding safety standards. In HVAC systems, flow measurement aids in ensuring comfort and safety, while also ensuring energy efficiency. In agriculture, it contributes to enhanced irrigation management thereby ensuring optimal resource utilization. Furthermore, the water sector benefits significantly from flow measurement by enabling leak detection, optimizing resource utilization, and ensuring compliance with regulations. However, present flow metering technologies either suffer from accuracy issues in the presence of impurities (dissolved salts, particulate matter etc.) have limited compatibility & versatility or are exorbitantly priced, rendering them economically infeasible for several applications. This particularly holds for the water sector.

[0003] Flow meters such as the Positive displacement flow meter, and the Turbine flow meter possess moving parts that come in contact with the fluid. The impurities in the form of dissolved salts or particulate matter can lead to scale buildup or directly damage the moving parts resulting in the loss of accuracy in volume measurements. Often, such meters require the installation of sophisticated filters which increases the setup cost. Also, the wear on the moving parts needs to be periodically monitored and if necessary repairs or recalibration must be carried out leading to downtime and additional cost. Moreover, the positive displacement flow meter is better suited for fluids with high viscosity as low-viscosity fluids lead to more slippage within the internal components resulting in a loss of accuracy.

[0004] The vortex flow meter does not possess any moving parts and is highly accurate but is extremely expensive and is not suitable at low flow rates rendering it ineffective as a leakage detection system. Vortex flow meters also require straight upstream sections, are not compatible with pipes of large diameters, and are affected by external vibration.

[0005] Like the Vortex flow meter, Ultrasonic (transit-time) flow meters, Electromagnetic flow meters, and Coriolis flow meters do not possess any moving parts that are in contact with the fluid. The electromagnetic flow meter however has limited compatibility given that it only works with conductive fluids. It is also not suitable for fluids flowing at a low velocity hence rendering it ineffective as a leakage detection system.

[0006] Coriolis flow meters are not suitable for pipes with large diameters used in water distribution networks. They are also extremely expensive making them economically infeasible for use in residential buildings. Ultrasonic flow meters, too, are not economically feasible for the residential sector. They also require periodic calibration to ensure high levels of accuracy. This leads to increased downtime and cost. They also require the pipe to be filled for accurate measurements which is unlikely in the case of water distribution networks. In residential pipes which are of smaller dimensions, ultrasonic flow meters struggle to recognise and measure low flow rendering them ineffective as a leakage detection system.

[0007] The central focus on the water sector is inevitable in the context of metering due to the urgency, criticality, and pervasiveness of water scarcity as a socio-economic issue and the fact that metering has been proven to be an effective solution. A UNICEF report stated that over 4 billion people already experience severe water scarcity for at least 1 month in a year and going forward this situation is likely to worsen due to climate change. The biggest drivers of water scarcity, after climate change & pollution, are injudicious consumption by households, and poor maintenance of the water infrastructure. Injudicious consumption by households, especially in urban areas, can primarily be attributed to the low price of piped water and a lack of consumption monitoring (metering). In the case of high-rises in water-stressed urban areas, even though the price of tanker water is sufficiently high to discourage excessive use, water expenses are hidden within the maintenance amount and hence the high price fails to act as a motivator. Moreover, the practice of splitting all water-related expenses evenly amongst all households erodes any incentive to adopt water-saving measures.

[0008] The pipes that carry treated water from the treatment plant to different parts of the city are laid underground. Over time, such pipes corrode due to the chemical interactions with the soil and water, resulting in leaks. Since leakage detection systems / flow meters are not installed at regular intervals, it becomes impossible to locate and seal the leaks immediately resulting in exorbitant water losses. Such losses account for over 30% of total input volume globally i.e. 126 billion m³ / year and conservatively cost about $40 billion. However the inter-regional differences are vast and in some regions, this figure could be as high as 65%. Further, water losses due to poor maintenance of infrastructure also plague developed nations.

[0009] Metering can address such inefficiencies and proves to be an effective solution based on the fundamental premise that what gets measured gets managed. In the case of excessive residential / domestic consumption, behavioural science has demonstrated that individual metering & billing of water in high-rises reduces consumption by up to 25%. This happens in two ways. By monitoring each household’s consumption and generating individual bills, households are incentivized to reduce their consumption as they now have to pay for their consumption. This effect can further be amplified by setting individual targets for each household. Another way through which consumption is bound to decrease is the high price of water is now explicitly mentioned in the water bill as opposed to being hidden with the maintenance amount paid in high-rises. In the case of water distribution networks, by installing meters / leak detection systems at regular intervals, the utility can pinpoint the exact location of the leak to take immediate corrective action. Smart meters can reduce the duration between occurrence of leaks and detection of leaks potentially saving billions of dollars globally.

[0010] Thus, in light of the above discussion, it is implied that there is need for a low-cost, non-invasive flow measurement and leakage detection device for closed conduits, which is reliable and does not suffer from the problems discussed above.Object of Invention

[0011] The principal object of this invention is to provide a low-cost, non-invasive flow measurement and leakage detection device for closed conduits.

[0012] Another object of the invention is to provide a low-cost, non-invasive device for measuring fluid flow and leakage detection of fluids within closed conduits using MEMS based IMU sensors in areas comprising but not limited to water distribution networks, water pipes in residential buildings, offices, schools, hotels, and malls.

[0013] Another objective of the invention is to provide a low-cost, non-invasive device for measuring fluid flow and leakage detection of fluids within closed conduits using MEMS based IMU sensors as a part of smart metering systems with advanced metering infrastructure (AMI) for water, gas and other fluids in various settings.

[0014] This invention is illustrated in the accompanying drawings, throughout which, like reference letters indicate corresponding parts in the various figures.

[0015] The embodiments herein will be better understood from the following description with reference to the drawings, in which:Fig. 1

[0016] depicts / illustrates a schematic representation of a device for non-invasive measurement of fluid flow and detection of fluid leakage in a closed conduit, in accordance with an embodiment;Fig. 2

[0017] depicts / illustrates a block diagram of a microcontroller, in accordance with an embodiment;Fig. 3

[0018] depicts / illustrates a prototype of the disclosed invention, showing the mounting position of the MPU6050 and how the MPU6050 and ESP8266 are connected using wires and placed within the weatherproof enclosure, in accordance with an embodiment;Fig. 4

[0019] depicts / illustrates a setup of the prototype wherein components (MPU6050, ESP8266, wires) are mounted appropriately and protected by the weatherproof enclosure, in accordance with an embodiment;Fig. 5

[0020] depicts / illustrates a dashboard from the Blynk IoT Cloud tool that shows graphical representation of real-time flow rate and hourly consumption data, in accordance with an embodiment;Fig. 6

[0021] depicts / illustrates a tabular representation comparing flow rates obtained through the disclosed invention against the actual flow rates measured with a standard water flow sensor, in accordance with an embodiment; and;Fig. 7

[0022] illustrates a method for non-invasive measurement of fluid flow and detection of fluid leakage in a closed conduit, in accordance with an embodiment;Statement of Invention

[0023] The present invention discloses a device for non-invasive measurement of fluid flow and detection of fluid leakage in a closed conduit. The device comprises a MEMS-based inertial measurement unit (IMU) sensor mounted externally on the closed conduit, configured to capture flow-induced vibrations caused by turbulent fluid flow within the conduit. The device further comprises a microcontroller operatively connected to the MEMS-based IMU sensor. The microcontroller is configured to process vibration data received from the MEMS-based IMU sensor by applying noise-reduction filters, comprising a recursive Gaussian filter, and extracting one or more spectral features, such as spectral centroid, spectral kurtosis, and spectral variance.

[0024] The microcontroller is further configured to estimate the flow rate of the fluid using a machine learning model based on a decision tree classification algorithm. The device detects fluid leakage by identifying flow rates below a predefined threshold or by recognizing sustained low flow patterns indicative of leakage. Additionally, the microcontroller computes the total fluid consumption by multiplying the estimated flow rate with the duration of fluid flow. The processed data, comprising the estimated flow rate, leakage status, and consumption data, is transmitted to a cloud platform via a communication unit.

[0025] The cloud platform is configured to store the estimated flow rate and leakage status, generate consumption reports comprising hourly and daily flow data, and display the stored data and reports on a user-accessible dashboard. The cloud platform also provides real-time alerts for abnormal flow rates or leakage, enabling proactive measures to address potential issues. The device is housed in a weatherproof enclosure made of materials such as polycarbonate, ABS, or phenolic resins to ensure durability in various environmental conditions. The invention offers a cost-effective, reliable, and efficient solution for monitoring fluid flow and detecting leaks in various applications, such as residential plumbing, industrial fluid management, and water distribution systems.Detailed Description

[0026] The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and / or detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.

[0027] Given the importance of metering for the various sectors in general and the water sector in particular, and the problems with the existing metering technologies, there is a strong necessity for a new type of flow meter that is affordable, easy to install on pipes of all dimensions, requires minimal calibration, has no moving parts in contact with the fluid, can be integrated with GIS software, possesses advanced metering infrastructure to support remote reading and automated bill generation. Such a meter can be designed by exploiting a characteristic of virtually all closed-conduit flows: Flow-Induced Vibration.

[0028] When a fluid flows through a pipe, even at nominal flow rates, it exhibits turbulence, which refers to the chaotic and irregular motion of a fluid within a pipe. Turbulence is commonly characterized by mixing, dispersion, swirling, and fluctuations in velocity and pressure. In such fluctuations the fluid molecules, on average, all travel in the direction of flow, and many collide with the pipe wall. According to the first law of thermodynamics, most of this kinetic energy is converted into potential energy in the form of pressure and some of it is dissipated as heat. The pressure causes the pipe to deform, converting the potential energy to kinetic energy and back to potential energy as the deformation is completed. Such energy conversion cycles induce vibrations that are proportional to the average flow rate and are called flow-induced vibrations. The key insight leading to the invention described herein is that the turbulence caused by the flow of fluids in closed conduits is measurable, providing an indirect way of characterizing the flow itself: enabling a cost-effective method for measuring the flow rate.

[0029] These vibrations can then be captured and analyzed using affordable, and commercially available Micro-Electro-Mechanical System (MEMS) based Inertial Measurement Units (IMUs) to estimate flow rate and subsequently volumetric flow. IMUs are compact electronic devices that are typically used to measure and report specific movements, orientations, and accelerations of an object in three-dimensional space. A combination of sensors, such as accelerometers, gyroscopes, and sometimes magnetometers. Although they are primarily used to capture and analyze the exaggerated movement of machines, automobiles, people, mobile phones etc, they can be calibrated to capture and analyze faint vibrations such as flow-induced vibrations by adjusting sensitivity digitally and installing them such that they are firmly bound to the conduit. This process can be better understood by the formula stated below:

[0030] Re = ρVD / μ …(1)

[0031] The formula stated in equation 1 is the formula used to calculate the Reynolds number, where ρ is fluid density in kg / m³, μ is the dynamic viscosity of the fluid in kg / m / s, V is fluid velocity is m / s, and D is pipe diameter in m. The formula states that by measuring the extent of turbulence using MEMS based IMUs, we can make inferences on the fluid velocity, provided we have information about the fluid and pipe characteristics. This forms the theoretical basis of the disclosed invention.

[0032] The present invention discloses a low-cost, non-invasive flow measurement and leakage detection device for closed conduits. The present invention provides a low-cost, non-invasive device for closed conduits which is designed for flow measurement and leakage detection of fluids exhibiting primarily turbulent flow in closed conduits by utilizing affordable, commercially available MEMS-based IMUs, microcontrollers, and proprietary machine learning algorithms.

[0033] The low-cost, non-invasive flow meter for closed conduits comprises a MEMS based IMU sensor, a microcontroller with a power source and WiFi module, a cloud platform that comprises a database and a dashboard, a mobile phone / tablet / computer with internet access, Internet connection(s), a weatherproof enclosure. This flow meter can be fitted to a closed conduit of any kind, comprising but not limited to rigid, non-rigid, pressurized, unpressurized, completely filled, or partially filled.

[0034] The MEMS based IMU sensor of the flow meter could comprise a single or multi axis accelerometer, a single or multi axis gyroscope, a single or multi axis magnetometer or any combination thereof. This could be attached to the external surface of any portion of the conduit using cable ties, straps, velcro, adhesive strips, and / or clamps for capturing flow-induced vibrations for various flow rates.

[0035] depicts / illustrates a device 100 comprising a MEMS-based inertial measurement unit IMU sensor 102, a microcontroller 104, a communication network 112, and a cloud platform 106.

[0036] In an embodiment, the MEMS-based IMU sensor 102 is configured to capture flow-induced vibrations caused by turbulent fluid flow within the closed conduit. The IMU sensor 102 is mounted externally on the closed conduit and operates without making direct contact with the fluid. The IMU sensor 102 generates vibration data corresponding to the dynamics of the fluid flow and transmits this data to the microcontroller 104 for processing.

[0037] In an embodiment, the microcontroller 104 is operatively connected to the IMU sensor 102 and processes the vibration data received from the IMU sensor 102. The processing involves applying noise-reduction filters, such as a recursive Gaussian filter, to eliminate unwanted noise, and extracting spectral features comprising at least one of spectral centroid, spectral kurtosis, and spectral variance. The microcontroller 104 further estimates the flow rate of the fluid using a machine learning model based on a decision tree classification algorithm. Additionally, the microcontroller 104 detects fluid leakage by analyzing the flow rate and identifying rates below a predefined threshold or sustained low flow patterns that indicate potential leakage. The microcontroller 104 transmits the processed data, comprising the estimated flow rate and leakage status, to the cloud platform 106 via the communication network 112.

[0038] In an embodiment, the communication network 112 is configured to enable data transfer between the microcontroller 104 and the cloud platform 106. The communication network 112 supports wireless communication protocols, such as Wi-Fi, Bluetooth, or LoRa, ensuring reliable and secure data transmission. It acts as an intermediary to transmit the processed data for further storage and analysis on the cloud platform 106.

[0039] In an embodiment, the cloud platform 106 is configured to store the estimated flow rate and leakage status received from the microcontroller 104. The cloud platform 106 generates consumption reports comprising hourly and daily flow data and displays the stored data on a user-accessible dashboard. Additionally, the cloud platform 106 provides real-time alerts to notify users of abnormal flow rates or potential leakage, enabling prompt corrective actions.

[0040] depicts / illustrates a block diagram of a microcontroller 104 and its operational connections to the MEMS-based inertial measurement unit IMU sensor 102 and the cloud platform 106 in a device for non-invasive measurement of fluid flow and detection of fluid leakage in a closed conduit. The microcontroller 104 comprises a central processing unit 212, a memory unit 214, at least one general-purpose input / output GPIO pin 216, and a communication unit 218.

[0041] In an embodiment, the MEMS-based IMU sensor 102 is configured to capture flow-induced vibrations caused by turbulent fluid flow within the closed conduit. The IMU sensor 102 transmits the vibration data to the microcontroller 104 through the GPIO pin 216. This connection enables the microcontroller 104 to receive raw vibration data for further processing. The IMU sensor 102 communicates using standard protocols, such as I²C or SPI, ensuring compatibility with the GPIO pin 216.

[0042] In an embodiment, the microcontroller 104 is configured to process the received vibration data using the central processing unit 212. The central processing unit 212 is programmed to apply noise-reduction filters, such as a recursive Gaussian filter, to eliminate unwanted noise from the vibration data. It further extracts spectral features from the filtered data, comprising at least one of spectral centroid, spectral kurtosis, and spectral variance, among others. These features are then analyzed by a machine learning model based on a decision tree classification algorithm, which is stored in the memory unit 214. The central processing unit 212 utilizes the extracted features to estimate the fluid flow rate and detect fluid leakage by identifying flow rates below a predefined threshold or sustained low flow patterns indicative of leakage.

[0043] In an embodiment, the memory unit 214 is configured to store program instructions, vibration data, and machine learning models. The memory unit 214 enables the microcontroller 104 to process data in real time while also maintaining historical data for reference and analytics. The memory unit 214 may comprise RAM for temporary storage and flash memory for long-term storage of program instructions and datasets.

[0044] In an embodiment, the communication unit 218 is configured to transmit processed data, comprising the estimated flow rate and leakage status, to the cloud platform 106. The communication unit 218 supports wireless communication protocols such as Wi-Fi, Bluetooth, or LoRa, enabling secure and efficient data transfer. Alternatively, wired communication protocols, such as Ethernet, may be used if required. The communication unit 218 utilizes the MQTT protocol for transmitting data, ensuring minimal latency and efficient bandwidth usage.

[0045] In an embodiment, the cloud platform 106 is configured to receive and store the processed data transmitted by the communication unit 218. The cloud platform 106 generates detailed consumption reports and displays the stored data on a user-accessible dashboard. It further provides real-time alerts to notify users of abnormal flow rates or potential leakage.

[0046] depicts / illustrates a prototype of the disclosed invention. It illustrates the mounting position of the MPU6050 and how the MPU6050 and ESP8266 are connected using wires and placed within the weatherproof enclosure.

[0047] In the flow meter, the MEMS based IMU sensor is connected to the microcontroller using physical wires and transfers data using at least one of the following communication protocols: Inter-Integrated-Circuit (I2C), Serial Peripheral Interface (SPI), Universal Asynchronous Receiver / Transmitter (UART), Controller Area Network (CAN), Universal Serial Bus (USB).

[0048] The microcontroller comprises at least the following components: CPU, RAM, flash memory, 4+ GPIO pins, a WiFi or Power over Ethernet (PoE) or Zigbee or Z-Wave or LoRa or Bluetooth module. The said microcontroller and MEMS based IMU sensor could be powered by batteries, mains electricity, any wireless power transfer method, or Power over Ethernet (PoE) technology.

[0049] The microcontroller is programmed to read and process the vibration data acquired from the said MEMS based IMU sensor’s gyroscope, accelerometer and / or magnetometer, which comprises filtering as well as feature extraction from either time-domain, frequency domain, and / or time-frequency domain, before running a machine learning model to estimate the flow rate from the processed vibration data.

[0050] The microcontroller uses its WiFi or Power over Ethernet (PoE) or Zigbee Z-Wave or LoRa or Bluetooth module to connect to the said internet connection directly or indirectly through a gateway in order to transmit data to the cloud platform. The mobile phone / tablet / computer with internet access is used to view the cloud platform’s database and dashboard which displays information comprising but not limited to present flow rate, total consumption, and hourly & daily consumption. Wherein the internet connection could be of any type, comprising but not limited to Dial-up connection, Digital Subscriber Line (DSL), Cable TV connection, Satellite Internet connection, wireless internet connection, cellular network.

[0051] depicts / illustrates a setup of the prototype wherein all the components (MPU6050, ESP8266, wires) are mounted appropriately and protected by the weatherproof enclosure.

[0052] The weatherproof enclosure could be made from pDCPD, phenolic resins, polycarbonate or ABS.

[0053] The proposed device could also be used as a standalone device for fluid leakage detection or as a leakage detection system with additional flow meters of its kind or any other kind.

[0054] In a preferred embodiment, the present invention provides a device for flow rate measurement of water in residential buildings. A pipe with an outer diameter of 0.75 inch / 19.2 mm is chosen to demonstrate the working of the invention. For this demonstration, the MEMS-based IMU sensor chosen is the MPU6050, and the microcontroller selected is the ESP8266. The MPU6050 is MEMS based IMU manufactured by TDK InvenSense that combines a 3-axis accelerometer and 3-axis gyroscope on a single chip and is used to capture and analyze flow-induced vibration in the disclosed invention. The biggest advantages of the MPU6050 over other IMUs are: low cost, easy compatibility with microcontrollers through the I2C communication protocol, customizability in terms of adjusting sensitivity digitally, easy integration as libraries are readily available in programming languages such as Python and C, integrated analog-to-digital converters that reduce computation load on the microcontroller. It is mounted to the pipe using cable ties such that the y-axis of the sensor is parallel to the direction of flow.

[0055] The MPU6050 is interfaced with the microcontroller ESP8266 via the Inter-Integrated Circuit (I2C) communication protocol. The Blue wire (Serial Clock pin used for providing clock pulse ) and the Green wire (Serial Data pin used for data transfer) facilitate the I2C protocol while the Red wire is used to power the MPU6050 and the Black wire is used to connect to the ground pin. The ESP8266 is a microcontroller designed and manufactured by Espressif Systems and was the preferred choice for this application due to its: low cost, integrated WiFi module that facilitates wireless connectivity, integrated 32-bit microprocessor with in-built RAM and memory, low power consumption, multiple GPIO pins, and flexibility in terms programming languages supported (MicroPython, C++, Arduino IDE etc). The ESP8266 is what performs the actual data collection & computation in the disclosed invention and is programmed using MicroPython for this specific application due to its readily available libraries.

[0056] Finally, both the MPU6050 and the ESP8266 are enclosed within a case to make the setup weatherproof. The enclosure itself is a one-piece, hinged plastic box with a snap-fit design that comprises cutouts for the pipe and power supply. Even the cutouts can be sealed using gaskets to ensure that the enclosure remains weatherproof. Figures 3 and 4 represent the complete setup with the enclosure in open and closed positions respectively.

[0057] Although the MPU6050 sensor comprises a 3-axis accelerometer and a 3-axis gyroscope, only the data from the y-axis of the gyroscope is used to estimate the flow rate. This was primarily done to reduce the computational load on the ESP8266. The gyroscope in the MPU6050 provides data related to angular velocity or rotational rate. It measures how fast the MPU6050 sensor is rotating or changing its orientation. The entire code for the disclosed invention is written in MicroPython due to the readily available libraries.

[0058] depicts / illustrates a dashboard from the Blynk IoT Cloud tool that shows a graphical representation of real-time flow rate and hourly consumption data.

[0059] The steps involved in flow rate measurement are as follows:

[0060] A sample of 512 observations from the y-axis of the gyroscope is collected at a sampling rate of 1024 samples per second.

[0061] The collected data sample is then filtered using a recursive Gaussian filter with a sigma value of 0.8 to reduce noise.

[0062] The filtered signal is then Fourier transformed and the following features were extracted: (i) sum of the product of the frequencies and their respective amplitudes, (ii) spectral centroid, (iii) spectral kurtosis, and (iv) spectral variance.

[0063] A tree / rule-based machine learning classification model, specifically the simple Decision tree model based on the Gini information gain criteria, is then applied to the extracted features to predict the flow rate.

[0064] This flow rate data is then published to the cloud using the MQTT communication protocol which is processed further by multiplying it with the duration to prepare detailed consumption reports that are then viewed on a readily available Blynk IoT Cloud tool dashboard as shown in.

[0065] Since the present invention disclosed also makes use of microcontrollers, advanced metering infrastructure (AMI) can be implemented to give the flow meter ‘smart’ capabilities. Additionally, in the setting of water distribution networks, the meter can be geo-tagged so that it is possible to pinpoint the location of leaks.

[0066] depicts / illustrates a tabular representation comparing the flow. rates obtained through the disclosed invention against the actual flow rates measured with a standard water flow sensor.

[0067] In the case of multiple meters along a lengthy conduit, leaks can be identified by detecting significant discrepancies beyond a certain threshold in flow rate measurements of consecutive meters. Conversely, for a single meter along a shorter conduit, leaks can be detected through the observation of flow rates over extended periods.FluidFlow rateDiameter of pipe (mm)Reynolds numberLPG (Butane+Propane)10 l / min19.053676Air (HVAC)6 gpm12.72836Methane2 m3 / hr19.052584Iso-octane (Petrol)5 l / min50.82890Milk10 l / min50.82151Sulphuric acid30 l / min25.42194Water2 l / min19.052223

[0068] Although the emphasis has been on flow measurement of water, the disclosed invention can be used to measure flow rates of various other fluids in different settings. This is possible as the invention measures the extent of turbulence-generated flow-induced vibrations to estimate flow rate and virtually all fluids exhibit turbulence when flowing within closed conduits. The invention can be used to measure flow rates of fluids comprising, but not limited to, LPG in kitchens, natural gas and oils in refineries, milk, sulphuric acid etc. The table given below provides a non-exhaustive list of fluids whose flow rates can be measured based on the fluid characteristics by the present invention.

[0069] The invention would be ideal for settings that require affordability, non-contact with fluids as in the case of corrosive fluids, remote reading, automated bill generation, geo-tagging, low maintenance, minimal calibration, compatibility with pipes of all dimensions. Another crucial aspect of the disclosed innovation is its capacity to monitor low flow conditions, comprising flows as low as 0.1 liters per minute as demonstrated in our experiment with water in a 19.05 mm pipe. Although these flows may not be theoretically classified as turbulent, this capability makes it well-suited for leak detection and a wide range of flow monitoring scenarios.

[0070] illustrates a method 700 for non-invasive measurement of fluid flow and detection of fluid leakage in a closed conduit. The method begins with capturing flow-induced vibrations caused by turbulent fluid flow within the closed conduit using a MEMS-based inertial measurement unit (IMU) sensor, as depicted at step 702. Subsequently, the method 700 discloses transmitting the vibration data from the MEMS-based IMU sensor to a microcontroller, as depicted at step 704. Thereafter, the method 700 discloses processing the vibration data in the microcontroller by applying noise-reduction filters and extracting one or more spectral features, as depicted at step 706. Thereafter, the method 700 discloses estimating flow rate of the fluid using a machine learning model based on a decision tree classification algorithm, as depicted at step 708. Thereafter, the method 700 discloses detecting fluid leakage by identifying flow rates below a predefined threshold and sustained low flow patterns indicative of leakage, as depicted at step 710.

[0071] The advantages of the current invention include:

[0072] Non-Invasive Measurement: The invention enables accurate measurement of fluid flow and detection of fluid leakage without direct contact with the fluid. This reduces the risk of contamination, corrosion, or wear associated with invasive systems.

[0073] Cost-Effectiveness: By using a MEMS-based IMU sensor mounted externally on the conduit, the device eliminates the need for complex or expensive installation procedures, making it a cost-effective solution for fluid monitoring.

[0074] Leakage Detection: The device is configured to detect fluid leakage by analyzing flow rates below a predefined threshold or recognizing sustained low flow patterns, enabling early detection and minimizing water wastage or fluid loss.

[0075] Real-Time Data Processing: With an integrated microcontroller, the device processes vibration data in real-time, ensuring prompt estimation of flow rates and immediate identification of anomalies or leaks.

[0076] IoT Integration: The device communicates processed data, comprising flow rates and leakage status, to a cloud platform. This enables remote monitoring, data storage, and analysis, making the device ideal for modern IoT-based systems.

[0077] Customizable Thresholds: The predefined thresholds for leakage detection are adjustable, enabling the device to adapt to varying fluid types, conduit sizes, and application requirements.

[0078] Comprehensive Data Analytics: The cloud platform generates detailed consumption reports comprising hourly and daily flow data. This provides users with actionable insights into fluid usage trends and helps optimize fluid management.

[0079] User Notifications: The cloud platform is capable of sending real-time alerts to users in case of abnormal flow rates or detected leakage, enabling timely corrective actions to prevent further loss or damage.

[0080] Durability and Versatility: The weatherproof enclosure protects the device from environmental factors such as moisture, dust, and temperature extremes, ensuring reliable operation in diverse conditions, comprising residential, industrial, and outdoor applications.

[0081] Scalability and Compatibility: The device is compatible with a variety of communication protocols, comprising Wi-Fi, Bluetooth, and LoRa, enabling integration into different network environments. It is also scalable for use in large distribution networks.

[0082] Energy Efficiency: The device uses efficient data processing algorithms and lightweight communication protocols such as MQTT, ensuring minimal power consumption, making it suitable for battery-powered applications.

[0083] Ease of Installation: The device is mounted externally using mechanisms such as cable ties, clamps, straps, or adhesive strips, eliminating the need for pipe cutting or invasive modifications during installation.

[0084] Adaptability to Multiple Applications: The invention is versatile and can be employed in various applications, comprising residential plumbing, water distribution systems, industrial fluid management, and agricultural irrigation monitoring.

[0085] Enhanced Accuracy: The use of noise-reduction filters, such as recursive Gaussian filters, and machine learning algorithms ensures high accuracy in flow rate estimation and leakage detection.

[0086] Environmental Sustainability: By enabling efficient fluid monitoring and reducing fluid waste through early leakage detection, the invention contributes to resource conservation and environmental sustainability.

[0087] Applications of the current invention include:

[0088] Residential Plumbing Systems: Monitoring water usage in residential homes to ensure efficient consumption and detect leaks in plumbing systems to prevent water wastage.

[0089] Industrial Fluid Management: Measuring and monitoring fluid flow in industrial pipelines, comprising oil, gas, and chemical processing systems, to optimize operations and detect potential leaks.

[0090] Municipal Water Distribution Systems: Monitoring water flow in municipal water distribution networks to detect leakages, reduce non-revenue water (NRW), and improve overall water management efficiency.

[0091] Agricultural Irrigation Systems: Measuring water flow in irrigation systems to optimize water distribution and detect leaks, ensuring efficient water use in agricultural applications.

[0092] Smart Water Meters: Integration into smart water metering systems for accurate real-time monitoring of water usage and leakage detection in urban and rural areas.

[0093] Oil and Gas Pipelines: Non-invasive monitoring of oil and gas pipelines to measure flow rates and detect leakage, reducing the risk of environmental contamination and economic losses.

[0094] HVAC Systems: Monitoring the flow of coolants or refrigerants in HVAC (Heating, Ventilation, and Air Conditioning) systems to ensure proper operation and detect potential leaks.

[0095] Hydraulic Systems: Monitoring fluid flow and leakage in hydraulic systems used in heavy machinery, manufacturing equipment, and transportation systems.

[0096] Drinking Water Infrastructure: Ensuring the integrity of drinking water pipelines in urban infrastructure to minimize water loss and maintain supply reliability.

[0097] Fire Suppression Systems: Monitoring fluid flow in fire suppression systems, such as sprinkler systems, to ensure proper operation and detect leaks that could compromise system functionality.

[0098] Chemical Processing Plants: Monitoring fluid flow and detecting leakage in pipelines and storage tanks in chemical processing plants to ensure operational safety and minimize material loss.

[0099] Pharmaceutical Manufacturing: Non-invasive monitoring of fluid flow in pharmaceutical manufacturing pipelines, ensuring precise flow rates and detecting leaks to maintain quality control.

[0100] Desalination Plants: Monitoring water flow and detecting leaks in desalination plants to optimize water treatment processes and prevent resource loss.

[0101] Wastewater Treatment Plants: Monitoring the flow of wastewater in treatment facilities to ensure efficient operation and detect leaks or blockages in the pipeline system.

[0102] Beverage and Food Processing Industries: Measuring and monitoring fluid flow in pipelines used for transporting liquids, such as juices, milk, or other beverages, in food and beverage processing industries.

[0103] Energy Sector: Monitoring and detecting leaks in pipelines transporting thermal fluids, such as in geothermal plants or district heating systems, to ensure energy efficiency.

[0104] Remote Monitoring Systems: Integration with IoT-enabled systems for remote monitoring and management of fluid flow in remote or hard-to-reach locations.

[0105] Research and Development: Use in laboratory setups for research and development involving fluid dynamics, non-invasive monitoring techniques, or pipeline system optimization.

[0106] Marine Applications: Monitoring the flow of fluids, such as water or fuel, in marine vessels to ensure optimal operation and detect potential leaks.

[0107] Mining Operations: Monitoring fluid transport pipelines in mining operations to detect leaks and ensure efficient material transport in harsh environmental conditions.

[0108] The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and / or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the scope of the embodiments as described here.

Claims

A method for non-invasive measurement of fluid flow and detection of fluid leakage in a closed conduit, comprising:capturing flow-induced vibrations caused by turbulent fluid flow within the closed conduit using a MEMS-based inertial measurement unit (IMU) sensor (102);transmitting vibration data from the MEMS-based IMU sensor (102) to a microcontroller (104);processing the vibration data in the microcontroller (104) by applying noise-reduction filters and extracting one or more spectral features;estimating flow rate of fluid using a machine learning model based on a decision tree classification algorithm; anddetecting fluid leakage by identifying flow rates below a predefined threshold and sustained low flow patterns indicative of leakage.The method as claimed in claim 1, comprising configuring the microcontroller (104) comprises:processing vibration data by using a central processing unit (CPU) (212);storing at least one of program instructions, vibration data, and machine learning models, by using a memory unit (214);providing an interface with the MEMS-based IMU sensor (102), by using at least one general-purpose input / output (GPIO) pin (216);communicating processed data to the cloud platform (106), by using the communication unit (218).The method as claimed in claim 1, comprising configuring the microcontroller (104) is configured for:applying a recursive Gaussian filter with a sigma value of 0.8 to reduce noise from the vibration data;computing total fluid consumption by multiplying the estimated flow rate with duration of fluid flow; andextracting spectral features comprises at least one of spectral centroid, spectral kurtosis, and spectral varianceThe method as claimed in claim 1, comprising estimating the flow rate is a decision tree classification algorithm based on the Gini information gain criteria, by using the machine learning model .The method as claimed in claim 1, comprising configuring a cloud platform (106) for:storing the estimated flow rate and leakage status; andgenerating consumption reports, comprising hourly and daily flow data.A device for non-invasive measurement of fluid flow and detection of fluid leakage in a closed conduit, comprising:a MEMS-based inertial measurement unit (IMU) sensor (102) connected to the closed conduit, configured to capture flow-induced vibrations caused by turbulent fluid flow within the closed conduit;a microcontroller (104) operatively connected to the MEMS-based IMU sensor (102), configured to:process vibration data received from the MEMS-based IMU sensor (102) by applying noise-reduction filters and extracting one or more spectral features;estimate flow rate of fluid using a machine learning model based on a decision tree classification algorithm; anddetect fluid leakage by identifying flow rates below a predefined threshold and sustained low flow patterns indicative of leakage.The device (100) as claimed in claim 6, wherein the microcontroller (104) comprises:a central processing unit (CPU) (212) configured to process vibration data;a memory unit (214) configured to store at least one of program instructions, vibration data, and machine learning models;at least one general-purpose input / output (GPIO) pin (216) configured to provide an interface with the MEMS-based IMU sensor (102); anda communication unit (218) configured to communicate the estimated flow rate and leakage status to a cloud platform (106).The device (100) as claimed in claim 6, wherein the microcontroller (104) is configured to:apply a recursive Gaussian filter with a sigma value of approximately 0.8 to reduce noise from the vibration data;compute total fluid consumption by multiplying the estimated flow rate with duration of fluid flow; andextract spectral features comprises at least one of spectral centroid, spectral kurtosis and spectral variance.The device (100) as claimed in claim 6, wherein the machine learning model is used to estimate the flow rate is a decision tree classification algorithm based on a Gini information gain criteria.The device (100) as claimed in claim 6, comprising a cloud platform (106) configured to:store the estimated flow rate and leakage status; andgenerate consumption reports, comprising hourly and daily flow data.The device (100) as claimed in claim 6, comprising a weatherproof enclosure (302) configured to house the MEMS-based IMU sensor (102) and the microcontroller (104), wherein the enclosure is made from a material comprising at least one of polycarbonate, ABS, or phenolic resins.The device (100) as claimed in claim 6, wherein the predefined threshold to detect fluid leakage is a flow rate of approximately 0.1 liters per minute, and wherein the predefine threshold determined based on vibration patterns of the closed conduit.The device (100) as claimed in claim 6, wherein the MEMS-based inertial measurement unit (IMU) sensor (102) comprises at least one of a single or multi axis accelerometer, a single or multi axis gyroscope and a single or multi axis magnetometer.The device (100) as claimed in claim 6, wherein the MEMS-based inertial measurement unit (IMU) sensor (102) connected to the closed conduit using at least one of cable ties, straps, velcro, adhesive strips, and / or clamps.The device (100) as claimed in claim 6, wherein the MEMS-based inertial measurement unit (IMU) sensor (102) and the microcontroller (104) are powered by at least one of batteries, mains electricity, any wireless power transfer method, or Power over Ethernet (PoE) technology.

Images