System and method for predictive irrigation system maintenance

Abstract

A predictive maintenance system includes an irrigation system including a plurality of components and configured to irrigate an area of tillage. The predictive maintenance system includes a sensor configured to generate a signal indicative of a condition of at least one of the plurality of components of the irrigation system based on network power quality. The sensor is disposed at a central pivot of the irrigation system or a main break of a utility. A memory includes instructions that, when executed by a processor, cause the predictive maintenance system to receive the sensed signal, determine a change in the condition of the at least one component, and predict a maintenance need of the at least one component based on predetermined data.

Description

[0001] Cross-references to related applications

[0002] This application claims the benefit of U.S. Provisional Patent Application No. 62 / 990,737, filed March 17, 2020, and U.S. Provisional Patent Application No. 63 / 002,930, filed March 31, 2020, the entire contents of each of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to irrigation systems, and more specifically, to structures and methods for enabling predictive maintenance of irrigation systems. Background Technology

[0004] Irrigation systems such as pivot systems, lateral movement systems, and drip irrigation systems experience an average of three failures per year over 40 uses. These failures occur during critical growth stages and, in many cases, in the middle of the field. Summary of the Invention

[0005] To limit delays, increased costs, and other problems associated with irrigation system failures, this disclosure details a solution that includes digitally observing the irrigation system during normal operation and setting parameters to indicate abnormal operation. To observe these operational anomalies, sensors can be added to the irrigation system to provide data for algorithms to be processed. These algorithms can be logic-based or analysis-based. In some cases, existing operational data from "off-the-shelf" sources can be used. In other respects, other data sources can be external to the system, such as meteorological and geomorphological maps, soil moisture data, etc., or combinations thereof from the National Oceanic and Atmospheric Administration (NOAA).

[0006] According to one aspect of this disclosure, a predictive maintenance system for an irrigation system includes an irrigation system, sensors, a processor, and a memory. The irrigation system is configured to irrigate a cultivated area and includes multiple components. The sensors are configured to generate signals indicating the condition of at least one of the multiple components of the irrigation system based on network power quality. The sensors are positioned at a central pivot of the irrigation system or at a major point of interruption in the utility. The memory includes instructions stored thereon that, when executed by the processor, cause the predictive maintenance system to receive the sensed signals, determine changes in the condition of at least one component, and predict maintenance needs for at least one component based on predetermined data.

[0007] In one aspect of this disclosure, the sensor may include an encoder, a pressure sensor, a flow meter, a current sensor, a power sensor, a voltage sensor, or a combination thereof.

[0008] In another aspect of this disclosure, the various components of the irrigation system may include pumps, pivots, towers, terminal towers, corner towers, air compressors, end guns, or combinations thereof.

[0009] In another aspect of this disclosure, when executed by a processor, the instructions may further cause the predictive maintenance system to transmit an indication of predicted maintenance needs to a user equipment for display; and to display the indication of predicted maintenance needs on a display of the user equipment.

[0010] In another aspect of this disclosure, when executed by the processor, the instructions may further enable the predictive maintenance system to predict the unexpected downtime of at least one component based on predetermined data; and to display the predicted unexpected downtime of at least one component on a display of the user equipment.

[0011] In another aspect of this disclosure, determining a change in the condition of at least one component may include comparing a sensed signal with predetermined data.

[0012] In one aspect of this disclosure, the instructions, when executed by a processor, may further enable the predictive maintenance system to receive data from weather stations, field soil moisture sensors, topographic and soil maps, temperature sensors, National Oceanic and Atmospheric Administration (NOAA) meteorological data, or a combination thereof.

[0013] In another aspect of this disclosure, when executed by a processor, the instructions may further enable the predictive maintenance system to improve the determined condition changes of the at least one component based on the received data; and to improve the prediction of maintenance needs for the at least one component based on the improved determined changes.

[0014] In another aspect of this disclosure, when the instructions are executed by the processor, the predictive maintenance system may further enable the system to display an improved prediction of maintenance needs on a display.

[0015] In another aspect of this disclosure, the prediction can be based on a comparison of power sensed by a sensor with a predicted power, said predicted power being based on at least one of the following: directly measured soil moisture, soil moisture inferred from meteorological data from field and / or regional weather stations, topographic maps, soil maps, motor speed, gearbox speed ratio, tower weight, span weight, operating conditions of said at least one component, or a combination thereof.

[0016] According to another aspect, a computer-implemented method for predictive maintenance of an irrigation system includes: receiving a signal sensed by a sensor located at a central pivot of the irrigation system or a major point of interruption of a utility, the signal indicating the condition of at least one of a plurality of components of the irrigation system based on network power quality, the irrigation system being configured to irrigate a cultivated area and including the plurality of components; determining a change in the condition of at least one component; and predicting the maintenance needs of at least one component based on predetermined data.

[0017] In another aspect of this disclosure, the sensor may include an encoder, a pressure sensor, a flow meter, a current sensor, a power sensor, a voltage sensor, or a combination thereof.

[0018] In one aspect of this disclosure, the irrigation system comprises multiple components including pumps, pivots, towers, terminal towers, corner towers, air compressors, tail nozzles, or combinations thereof.

[0019] In another aspect of this disclosure, the computer-implemented method may further include transmitting an indication of the predicted maintenance needs to a user equipment for display, and displaying the indication of the predicted maintenance needs on a display of the user equipment.

[0020] In another aspect of this disclosure, the computer-implemented method may further include predicting the unexpected downtime of at least one component based on predetermined data and displaying the predicted unexpected downtime of at least one component on a display of a user equipment.

[0021] In another aspect of this disclosure, determining a change in the condition of at least one component may include comparing a sensed signal with predetermined data.

[0022] In another aspect of this disclosure, the computer-implemented method may also include receiving data from weather stations, field soil moisture sensors, topographic and soil maps, temperature sensors, meteorological data from the National Oceanic and Atmospheric Administration (NOAA), or a combination thereof.

[0023] In one aspect of this disclosure, the computer-implemented method may further include improving the determined condition changes of the at least one component based on received data; improving the prediction of maintenance needs for the at least one component based on the improved determined changes; and displaying the improved prediction of maintenance needs on a display.

[0024] In another aspect of this disclosure, the prediction is based on a comparison of the power sensed by the sensor with a predicted power, the predicted power being based on at least one of the following: directly measured soil moisture, soil moisture inferred from meteorological data from field and / or regional weather stations, topographic map, soil map, motor speed, gearbox speed ratio, tower weight, span weight, operating condition of the at least one component, or a combination thereof.

[0025] According to one aspect, this disclosure relates to a non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform a computer-implemented method for predictive maintenance of an irrigation system. The computer-implemented method includes receiving a signal sensed by a sensor indicating the condition of at least one of a plurality of components of the irrigation system; determining a change in the condition of the at least one component; and predicting maintenance needs for said at least one component based on predetermined data. The irrigation system is configured to irrigate a cultivated area and includes a plurality of components.

[0026] According to one aspect, this disclosure relates to a machine learning-based predictive maintenance system, comprising an irrigation system configured to irrigate a cultivated area, multiple components, and sensors arranged at a central pivot of the irrigation system or at a major point of interruption of a utility. The sensors are configured to generate signals indicating the condition of at least one of the multiple components of the irrigation system. The system also includes a processor and a memory. The memory includes instructions that, when executed by the processor, cause the predictive maintenance system to receive the sensed signals, determine an abnormal operation of at least one component, and predict maintenance needs for at least one component based on the determined abnormal operation using a machine learning model.

[0027] In another aspect of this disclosure, the instructions, when executed by a processor, can cause the predictive maintenance system to display on a display the predicted maintenance requirements of at least one component.

[0028] In another aspect of this disclosure, the irrigation system comprises multiple components including pumps, pivots, towers, terminal towers, corner towers, air compressors, tail nozzles, or combinations thereof.

[0029] In another aspect of this disclosure, signals of abnormal operation may include an increase in the energy required by the mobile irrigation system, changes in system speed, changes in tower movement sequence, tail nozzle rotation frequency, and / or power quality indicators.

[0030] In one aspect of this disclosure, power quality metrics may include phase balance, inrush current, and / or power factor.

[0031] In another aspect of this disclosure, the sensor may include an encoder, a pressure sensor, a flow meter, a current sensor, a power sensor, a voltage sensor, or a combination thereof.

[0032] In another aspect of this disclosure, when executed by a processor, the instructions may further cause the predictive maintenance system to transmit an indication of the predicted maintenance needs to a user equipment for display, and to display the indication of the predicted maintenance needs on a display of the user equipment.

[0033] In another aspect of this disclosure, the machine learning model is based on deep learning networks, classical machine learning models, or a combination thereof.

[0034] In another aspect of this disclosure, when executed by a processor, the instructions may further enable the predictive maintenance system to receive data from at least one of the following: a weather station, a field soil moisture sensor, a topographic and soil map, a temperature sensor, a meteorological report from the National Oceanic and Atmospheric Administration (NOAA), or a combination thereof.

[0035] In one aspect of this disclosure, the prediction can be based on a comparison of power sensed by a sensor with a predicted power, said predicted power being based on at least one of the following: directly measured soil moisture, soil moisture inferred from meteorological data from field and / or regional weather stations, topographic maps, soil maps, motor speed, gearbox speed ratio, tower weight, span weight, operating conditions of said at least one component, or a combination thereof.

[0036] According to one aspect, this disclosure relates to a computer-implemented method for predictive maintenance of an irrigation system. The computer-implemented method includes receiving signals sensed by sensors located at a central pivot of the irrigation system or at a major point of interruption of the utility, the signals indicating the condition of at least one of a plurality of components of the irrigation system based on network power quality. The irrigation system is configured to irrigate a cultivated area and includes a plurality of components. The computer-implemented method further includes determining abnormal operation of at least one component and predicting maintenance needs for at least one component based on the determined abnormal operation using a machine learning model.

[0037] In another aspect of this disclosure, when executed by the processor, the instructions may further cause the predictive maintenance system to display on a display the predicted maintenance requirements of at least one component.

[0038] In another aspect of this disclosure, the various components of the irrigation system may include pumps, pivots, towers, terminal towers, corner towers, air compressors, tail guns, or combinations thereof.

[0039] In another aspect of this disclosure, signals of abnormal operation include an increase in the energy required by the mobile irrigation system, changes in system speed, changes in tower movement sequence, tail nozzle rotation frequency, power quality indicators, or combinations thereof.

[0040] In another aspect of this disclosure, power quality metrics may include phase balance, inrush current, power factor, or combinations thereof.

[0041] In one aspect of this disclosure, the sensor may include an encoder, a pressure sensor, a flow meter, a current sensor, a power sensor, a voltage sensor, or a combination thereof.

[0042] In another aspect of this disclosure, the irrigation system comprises at least one of the following: a pump, a pivot, a tower, a terminal tower, an angle tower, an air compressor, a tail nozzle, or a combination thereof.

[0043] In another aspect of this disclosure, the machine learning model is based on deep learning networks, classical machine learning models, or a combination thereof.

[0044] In another aspect of this disclosure, the prediction is based on a comparison of the power sensed by the sensor with a predicted power, the predicted power being based on directly measured soil moisture, soil moisture inferred from meteorological data from field and / or regional weather stations, topographic maps, soil maps, motor speed, gearbox speed ratio, tower weight, span weight, operating conditions of the at least one component, or a combination thereof.

[0045] According to one aspect, this disclosure relates to a non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform a computer-implemented method for predictive maintenance of an irrigation system. The computer-implemented method includes: receiving signals sensed by sensors located at a central pivot of the irrigation system or at a major point of failure of a utility, the signals indicating the condition of at least one of a plurality of components of the irrigation system based on network power quality; determining abnormal operation of the at least one component; and predicting maintenance requirements for the at least one component based on the determined abnormal operation using a machine learning model. The irrigation system is configured to irrigate cultivated areas and includes a plurality of components.

[0046] Other aspects, features, and advantages will become apparent from the following description, drawings, and claims. Attached Figure Description

[0047] The accompanying drawings, which are included in and form part of this specification, illustrate various aspects of this disclosure and, together with the general description of the disclosure given above and the detailed description given below, explain the principles of the disclosure, wherein:

[0048] Figure 1 This is a diagram illustrating a predictive maintenance system.

[0049] Figure 2 It is configured to be used with Figure 1 A block diagram of the controller used in predictive maintenance systems;

[0050] Figure 3 It shows the configuration used with Figure 1 A diagram of machine learning models used in predictive maintenance systems;

[0051] Figure 4A An exemplary flowchart of typical tillage operations is shown;

[0052] Figure 4B An exemplary flowchart of a tillage operation including a predictive maintenance system according to the principles of this disclosure is shown;

[0053] Figure 5 It shows the use of Figure 1 The data science workflow for various models of the predictive maintenance system shown;

[0054] Figures 6-8 yes Figure 1 A diagram showing an example hardware interface and instrumentation for a predictive maintenance system;

[0055] Figure 9 yes Figure 1 A perspective view of a portion of an exemplary pivot of a predictive maintenance system; and

[0056] Figure 10 yes Figure 1 Another exemplary pivotal view of an air compressor instrument in a predictive maintenance system. Detailed Implementation

[0057] The various aspects of the disclosed predictive maintenance system are described in detail with reference to the accompanying drawings, wherein in each of the several views, the same reference numerals denote the same or corresponding elements. Furthermore, directional terms such as front, back, top, bottom, and top are used for descriptive convenience only and are not intended to limit the appended disclosure.

[0058] In the following description, well-known functions or structures are not described in detail in order to avoid confusing this disclosure with unnecessary details.

[0059] Currently, for potato or vegetable cultivation, simple watering timers are used, and if they are not reset within approximately 30-minute intervals, the irrigation system is presumed to have a problem. Each tower also has a safety microswitch that opens if the tower falls too far behind due to a drive system failure. At the control box, there may be an encoder that provides angular or linear position in the case of a linear system. The encoder can also be used in conjunction with a drip irrigation system. There are also voltage measurements and monitorable circuit breakers at the controller. This disclosure adds these measurements to provide more data to the algorithm for better prediction of system health.

[0060] Advantageously, the disclosed system predicts common unexpected downtime, rather than providing after-the-fact notification. The disclosed system offers better insight than a team that drives around observing operations, which can be subjective. Current technology only notifies of failures after they occur, while the disclosed system predicts maintenance before failures happen.

[0061] Other diagnostic health measurements are ex post facto, logic-based, and do not attempt to assign system health. This system predicts failures before they occur, for example, by checking a car's engine lights or the digital twin of a connected device. Furthermore, while the system disclosed herein relates to irrigation for potato or vegetable farming, it can be modified for any suitable farming operation requiring irrigation and may include drip irrigation systems, linear pivot systems, and / or central pivot systems.

[0062] refer to Figure 1 and Figures 7-9 A predictive maintenance system 100 is provided. Typically, the predictive maintenance system 100 includes an irrigation system 106 and a controller 200 configured to execute instructions controlling the operation of the predictive maintenance system 100. The irrigation system 106 may include a pump 10 (e.g., a compressor), a pivot 20, one or more towers 30, a terminal tower 40, a corner tower 50, an air compressor 60, and a tail nozzle 70. The pump 10 may include one or more current sensors 102 and a wireless communication device 104 configured to wirelessly transmit data (e.g., sensed current data) to the controller 200. The pivot 20 may include one or more sensors 102 and a wireless communication device 104 configured to wirelessly transmit data to the controller 200. Each tower 30, corner tower 50, and terminal tower 40 may include one or more sensors 102 and a wireless communication device 104 configured to wirelessly transmit data to the controller 200. Wireless communication devices may include, for example, 3G, Long Term Evolution (LTE), 4G, 5G, Bluetooth, and / or Wi-Fi. Sensor 102 may include at least one of the following: a current sensor, a voltage sensor, and / or a power sensor, which are configured to sense, for example, current, voltage, and / or power, respectively.

[0063] In some respects, one or more sensors 102 may include any suitable sensor, such as an encoder (e.g., an angle encoder), a pressure sensor, a flow meter, or a combination thereof. An angle encoder is a position sensor that measures the angular position of a rotating shaft.

[0064] In some respects, one or more sensors may be connected (e.g., directly connected) and / or may be independent components that can be connected via a wide area network (WAN). In some respects, one or more sensors may be aggregated in the cloud based on configuration settings. In some respects, one or more sensors may include, for example, low-power wide area network (LPWAN) technology, which may be long-range (LoRa).

[0065] In some respects, the controller 200 can determine the condition change of at least one component based on comparing the sensed signal with predetermined data.

[0066] Controller 200 is configured to receive data from sensor 102 and from external data sources (e.g., weather station 82, field soil moisture sensor 86, topographic and soil map 88, temperature sensor 89, and / or NOAA weather 84) to make and / or improve predictions indicating the condition of at least one of a plurality of components of irrigation system 106 (e.g., pivot 20, tail nozzle 70, tower 30, etc.). This prediction enables controller 200 to determine changes in the condition of at least one component and predict maintenance needs for at least one component based on predetermined data (e.g., historical data). For example, the prediction may be based on comparing the determined changes in the condition of at least one component of irrigation system 106 with predetermined data. For example, sensor 102 of tower 30 may sense the typical current consumption of tower 30. The sensed current consumption can then be compared by controller 200 with historical and / or typical tower current consumption. The controller can determine that, under a specific set of conditions (sunny weather, dry soil, etc.), the sensed current consumption of the tower 30 is higher than the historical current consumption by a predetermined amount (e.g., approximately 30%). Based on this determination, the controller 200 can predict that the tower 30 requires maintenance. Furthermore, it can predict the specific type of maintenance. For example, if the motor current of the tower 30 is high, it may indicate a tire leak. The system 100 can also predict the number of hours typically required to repair such an event. In another example, the system can sense a low current on the pump 10 via sensor 102 and predict a pump 10 failure accordingly.

[0067] Data from external data sources can be used to improve model predictions. For example, by processing data (e.g., the motor in tower 30 uses higher power due to muddy fields caused by recent rains), this processed data can be used to improve model predictions. The predictive maintenance system 100 can display field maps of terrain, soil type, etc., which help the model interpret changes in power usage. Predictions can be transmitted to user equipment 120 via controller 200 for display and / or further analysis.

[0068] In some respects, data and / or forecasts can be handled by a data visualization system 110. Data visualization is the graphical representation of information and data. By using visualization elements such as charts, graphs, and maps, data visualization tools provide a convenient way to view and understand trends, outliers, and patterns in data.

[0069] In some respects, the predictive maintenance system 100 can be implemented in the cloud. For example, Linux running Python scripts can be used to implement predictions.

[0070] Figure 2 The controller 200 is shown to include a processor 220 connected to a computer-readable storage medium or memory 230. The computer-readable storage medium or memory 230 may be a volatile type of memory, such as RAM, or a non-volatile type of memory, such as flash memory, disk media, etc. In various aspects of this disclosure, the processor 220 may be another type of processor, such as a digital signal processor, microprocessor, application-specific integrated circuit (ASIC), graphics processing unit (GPU), field-programmable gate array (FPGA), or central processing unit (CPU). In some aspects of this disclosure, network inference may also be implemented in a system where weights are implemented as memristors, chemical, or other inference computations, rather than in a processor.

[0071] In some aspects of this disclosure, memory 230 may be random access memory, read-only memory, disk storage, solid-state storage, optical disk storage, and / or other types of memory. In some aspects of this disclosure, memory 230 may be separable from controller 200 and may communicate with processor 220 via a communication bus on a circuit board and / or via a communication cable (e.g., a serial ATA cable or other type of cable). Memory 230 includes computer-readable instructions executable by processor 220 to run controller 200. In other aspects of this disclosure, controller 200 may include network interface 240 for communicating with other computers or servers. Storage device 210 may be used to store data.

[0072] The disclosed method can be run on the controller 200 or on a user device, including, for example, on a mobile device, an Internet of Things (IoT) device, or a server system.

[0073] Figure 3 The basic machine learning model 300 and data flow / storage / feedback of the pivot predictive maintenance system are illustrated. Model 300 can be hosted at pivot 20 or in the cloud (e.g., a remote server). Machine learning model 300 may include one or more convolutional neural networks (CNNs).

[0074] In machine learning, Convolutional Neural Networks (CNNs) are a class of artificial neural networks (ANNs) most commonly used for analyzing visual images. The convolutional aspect of a CNN involves applying matrix processing operations to local portions of an image; the result of these operations (which can involve dozens of different parallel and serial computations) is a set of many features used to train the neural network. CNNs typically include convolutional layers, activation function layers, and pooling (usually max pooling) layers to reduce dimensionality without losing too much feature. Additional information can be included in the operations that generate these features. The unique information that enables the production of features that inform the neural network can be used to ultimately provide an aggregation method to distinguish different data input to the neural network. In some aspects, machine learning model 300 may include a combination of one or more deep learning networks (e.g., CNNs) and classical machine learning models (e.g., Support Vector Machines (SVMs), decision trees, etc.). For example, machine learning model 300 may include two deep learning networks.

[0075] In some respects, two methods of labeling training data can be used: one based on a connection to a computer maintenance system (CMMS), and the other based on user input. In some respects, users can be prompted to label data, or they can provide data manually (e.g., when a service incident occurs).

[0076] As mentioned above, Figure 4A An exemplary flowchart 400a of typical farming operations is shown. Typically, at step 410, pre-season maintenance is performed on the irrigation equipment. Next, at step 420, the irrigation equipment is used during the season. At step 440, if a malfunction is determined, the equipment is sent for repair at step 430. At the end of the season (step 450), post-season maintenance is performed (step 460).

[0077] Figure 4B An exemplary flowchart 400b of a tillage operation including a pivot predictive maintenance system 100 according to the principles of this disclosure is shown. Typically, at step 410, pre-season maintenance is performed on the irrigation equipment. Next, the predictive maintenance system 100 predicts whether specific parts of the irrigation equipment require maintenance. If maintenance is predicted at step 415, the equipment is inspected and repaired at step 430. Next, at step 420, the irrigation equipment is used during the season. At step 440, if a malfunction is determined, the equipment is sent for repair at step 430. At the end of the season (step 450), post-season maintenance is performed (step 460).

[0078] Figure 5 It shows the use of Figure 4B The data science workflow for various models of the pivot predictive maintenance system is shown.

[0079] These five models include tail gun prediction model 502, tower drive prediction model 504, sequence prediction model 506, air compression prediction model 508, and electrical prediction model 510. These models can be implemented through logic and / or machine learning.

[0080] Tail-mounted gun prediction model 502:

[0081] The tail gun prediction model can calculate the tail gun 70 ( Figure 1 The number of times from left to right and then back to left. The expected time to travel left and right may be based on pressure, bearing condition, tension, or a combination thereof.

[0082] The tail gun prediction model 502 can consider expected power based on soil moisture, topographic maps, soil maps, motor speed, gearbox speed ratio, tower weight, span weight, operating conditions, or combinations thereof, directly measured or inferred from meteorological data from field and / or regional weather stations. The tail gun 70 includes instruments that can measure each cycle using proximity switches, encoders, capacitance, and / or imaging systems. Various aspects of the predictive maintenance system 100 can be mounted on or outside the irrigation system 106, such as a humidity sensor that can record as water distributed to the field splashes away. If an electron gun is used, energy utilization and duty cycle can be used.

[0083] Tower-driven prediction model 504 (see Figure 9 ):

[0084] From the center, the tower drive prediction model 504 can predict which tower is moving based on power surge sequences. Measurements at each tower, motor, segment, etc., and / or combinations thereof provide deeper insights into the operation of individual towers 30, 40, 50. Power can be measured using average current transformers, high-bandwidth current transformers, variable frequency drive (VFD) communication, smart contactors / circuit breakers / relays, Hall effect, etc., and / or combinations thereof.

[0085] Component temperatures, such as those of contactors, motors, and gearboxes, can be monitored and compared to expected temperatures. Expected temperatures can be derived from speed, ambient temperature, component type, component combination, location within the system, wire temperature, and / or combinations thereof. A position sensor can be installed on each of the 30, 40, and 50 towers to measure ground velocity. Position sensors can include Global Positioning System (GPS), ultra-wideband, and / or Real-time Kinematic (RTK). Accelerometers can be used to measure the grinding operation of gears in motor shaft bearings. Tire pressure sensors can directly measure tire pressure. An angle sensor between the towers can be used to monitor the relative position between the towers. A tilt sensor can be used to measure the tilt between the towers, indicating a low tire. A speed sensor can be placed on each tire to indicate drive problems.

[0086] Ordinal prediction model 506:

[0087] The sequence is used to segment the water distribution at the corner tower 50. The example system has thirteen solenoid valves that open as the span extends from the terminal tower 40 to the corner tower 50. When the corner tower 50 begins to expand first, one solenoid valve opens until it has fully expanded, at which point all thirteen valves are open. Solenoid valves can malfunction, leading to over-watering when they fail to close and under-watering when they fail to open.

[0088] Monitoring pressure, flow rate, current, transient voltage, water circuit conductivity, nozzle condition / status measurements, visual / infrared systems for observing spray, nozzle temperature, vibration, and / or combinations thereof can be used to provide insights into operations.

[0089] Air compressor predictive model 508:

[0090] The air compressor predictive model 508 can monitor duty cycle, pressure, vibration, motor temperature and / or electrical power in on / off states, which can provide insights into compressor health.

[0091] Monitoring output parameters, such as tail nozzle timing, flow rate, and / or pressure, can also help infer the health of the air compressor.

[0092] The condition of one or more components of the irrigation system can also be inferred from power signals at the central pivot 20 (and / or the main point of interruption of utility 22) without the need for power sensors on individual towers. This offers the advantage of using fewer sensors and simpler setup in the field. In some respects, the controller 200 can monitor the network power quality for signals that indicate the irrigation system 106 (based on power network analysis) Figure 1 The condition of at least one component among multiple components of a network. Network power quality includes the degree to which the voltage, current, power, frequency, and / or waveform of the power supply system conform to specifications. Power network analysis can be performed through machine learning and / or analysis. Power network analysis analyzes network power quality over time to determine the condition of components electrically connected to the network. Specifications can be determined in advance based on the analysis of historical network power data. For example, typically, Tower 30 ( Figure 1 The system can use approximately 2 kW of power. In an irrigation system 106 with three towers operating, the controller might measure approximately 6 kW of power used at a specific time, measured at the central pivot 20. However, if the controller 200 is measuring 7 kW, it can determine that one of the towers 30 is malfunctioning (e.g., a flat tire making the middle tower 30 more difficult to move, thus causing that tower 30 to use more power). In some respects, sensors 102 (e.g., vibration sensors connected to each drivetrain of each tower 30) can be used to determine which tower 30 is operating and when, thus providing additional data to the controller 200.

[0093] Electrical Instruments:

[0094] The system can also monitor contactor, commutator ring, motor winding, electrical connection, and / or wiring faults. Monitoring electrical transients or power metrics, such as total harmonic distortion (THD), power factor, and current balance, can help infer the health of the electrical system.

[0095] Monitoring the temperature of the components listed above can also help infer the health of the electrical system.

[0096] Furthermore, the disclosed structure may include any suitable mechanical, electrical, and / or chemical components for operating the disclosed pivot predictive maintenance system or its components. For example, such electrical components may include, for instance, any suitable electrical and / or electromechanical and / or electrochemical circuitry that may include or be coupled to one or more printed circuit boards. As used herein, the term "controller" includes terms such as "processor," "digital processing device," etc., and is used to denote a microprocessor or central processing unit (CPU). A CPU is an electronic circuit within a computer that executes instructions of a computer program by performing basic arithmetic, logic, control, and input / output (I / O) operations specified in the instructions, and by non-limiting example, includes server computers. In some aspects, the controller includes an operating system configured to execute executable instructions. An operating system is, for example, software comprising programs and data that manages the hardware of the disclosed binding device and provides services for executing applications used with the disclosed binding device. Those skilled in the art will recognize, by way of non-limiting example, that suitable server operating systems include FreeBSD, OpenBSD, NetBSD®, Linux, Apple® Mac OS X Server®, Oracle® Solaris®, Windows Server®, and Novell® NetWare®. In some respects, the operating system is provided by cloud computing.

[0097] In some respects, the term "controller" can be used to refer to a device that controls the transmission of data from a computer or computing device to a peripheral or stand-alone device and vice versa, and / or mechanical and / or electromechanical devices that mechanically operate and / or actuate peripheral or stand-alone devices (e.g., levers, knobs, etc.).

[0098] In some aspects, the controller includes storage and / or memory devices. Storage and / or memory devices are one or more physical devices used for temporary or permanent storage of data or programs. In some aspects, the controller includes volatile memory and requires power to maintain the stored information. In various aspects, the controller includes non-volatile memory and retains the stored information when no power is applied. In some aspects, non-volatile memory includes flash memory. In some aspects, non-volatile memory includes dynamic random-access memory (DRAM). In some aspects, non-volatile memory includes ferroelectric random-access memory (FRAM). In various aspects, non-volatile memory includes phase-change random access memory (PRAM). In some aspects, the controller is a storage device, including, by non-limiting examples, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), flash memory devices, disk drives, tape drives, optical disc drives, and cloud-based storage. In all respects, storage and / or memory devices are combinations of devices such as those disclosed herein.

[0099] In some aspects, the controller includes a display that sends visual information to the user. In various aspects, the display is a cathode ray tube (CRT). In various aspects, the display is a liquid crystal display (LCD). In some aspects, the display is a thin-film transistor liquid crystal display (TFT-LCD). In some aspects, the display is an organic light-emitting diode (OLED) display. In some aspects, the OLED display is a passive-matrix organic light-emitting diode (PMOLED) or an active-matrix organic light-emitting diode (AMOLED) display. In some aspects, the display is a plasma display. In some aspects, the display is a video projector. In various aspects, the display is interactive (e.g., having a touchscreen or sensors such as a camera, three-dimensional (3D) sensor, LiDAR, radar, etc.) capable of detecting user interactions / gestures / responses, etc. In some aspects, the display is a combination of devices such as those disclosed herein.

[0100] The controller may include or be coupled to a server and / or a network. As used herein, the term "server" includes terms such as "computer server," "central server," and "master server" to refer to a computer or device on a network that manages external binding devices, their components, and / or their resources. As used herein, the term "network" may include any network technology, including, for example, cellular data networks, wired networks, fiber optic networks, satellite networks, and / or IEEE 802.11a / b / g / n / ac wireless networks.

[0101] In various aspects, the controller can be coupled to a mesh network. As used in this article, a “mesh network” is a network topology in which each node relays data within the network. All mesh nodes cooperate in data distribution within the network. It can be applied to both wired and wireless networks. A wireless mesh network can be viewed as a type of “Wireless Ad Hoc” network. Therefore, wireless mesh networks are closely related to Mobile Ad Hoc Networks (MANETs). While MANETs are not limited to a specific mesh network topology, wireless ad hoc networks or MANETs can adopt any form of network topology. Mesh networks can use flooding or routing techniques to relay messages. Through routing, messages propagate along paths by hopping from one node to another until they reach their destination. To ensure that all their paths are available, the network must allow continuous connectivity and must use self-healing algorithms, such as shortest path bridging, to reconfigure itself around broken paths. Self-healing allows route-based networks to function even when nodes fail or connections become unreliable. Therefore, networks are generally very reliable because there are often multiple paths between sources and destinations within the network. This concept can also be applied to wired networks and software interactions. A mesh network in which all nodes are interconnected is a fully connected network.

[0102] In some respects, a controller may include one or more modules. As used herein, the term "module" and similar terms are used to refer to a self-contained hardware component of a central server, which in turn includes software modules. In software, a module is part of a program. A program consists of one or more independently developed modules that are not combined before the program is linked. A single module may contain one or more routines, or program segments that perform a specific task.

[0103] As used herein, the controller includes software modules for managing various aspects and functions of the disclosed external binding device or its components.

[0104] The disclosed architecture can also utilize one or more controllers to receive various types of information and transform the received information to generate output. The controller can include any type of computing device, computing circuit, or any type of processor or processing circuit capable of executing a series of instructions stored in memory. The controller can include multiple processors and / or multi-core central processing units (CPUs), and can include any type of processor, such as a microprocessor, digital signal processor, microcontroller, programmable logic device (PLD), field-programmable gate array (FPGA), etc. The controller can also include memory for storing data and / or instructions, which, when executed by one or more processors, cause one or more processors to execute one or more methods and / or algorithms.

[0105] Any method, program, algorithm, or code described herein can be converted into or represented in a programming language or computer program. The terms "programming language" and "computer program" as used herein include any language used to specify computer instructions, and include (but are not limited to) the following languages ​​and their derivatives: assembler, Basic, batch files, Basic Combined Programming Language (BCPL), C, C++, Delphi, Fortran, Java, JavaScript, machine code, operating system command languages, Pascal, Perl, PL1, scripting languages, Visual Basic, meta-languages ​​that specify the program itself, and all first-, second-, third-, fourth-, fifth-, or next-generation computer languages. Databases and other data schemas are also included, as well as any other meta-languages. There is no distinction between interpreted, compiled, or both compiled and interpreted languages. There is no distinction between a compiled version of a program and a source version. Therefore, where a programming language can exist in multiple states (e.g., source, compiled, object, or linked), a reference to a program is a reference to any and all such states. References to a program may contain actual instructions and / or the intent of those instructions.

[0106] For complex faults, machine learning (ML) models may be most effective. However, basic logic can be applied to simpler fault modes. Possible signals of anomalous operation can come from increased energy requirements of the mobile irrigation system, changes in system speed, changes in tower movement sequence, tail nozzle rotation frequency, or power quality indicators such as phase balance, inrush current, power factor, and THD. Because these variables vary over complex inference spaces, ML can help predict anomalous operation and simplify input for users and subject matter experts by providing simple labeling methods.

[0107] It is understood that any component of the disclosed device can be secured using known fastening techniques (such as welding, crimping, bonding, fastening, etc.).

[0108] Those skilled in the art will understand that the structures and methods specifically described herein and illustrated in the accompanying drawings are non-limiting exemplary aspects, and that the descriptions, disclosures, and figures should be interpreted as exemplary of a particular aspect only. Therefore, it should be understood that this disclosure is not limited to the explicitly described aspects, and that various other changes and modifications can be made by those skilled in the art without departing from the scope or spirit of this disclosure. Furthermore, it is contemplated that elements and features shown or described in relation to one exemplary aspect may be combined with elements and features of another exemplary aspect without departing from the scope of this disclosure, and such modifications and variations are also intended to be included within the scope of this disclosure. In fact, any combination of any disclosed elements and features is within the scope of this disclosure. Therefore, the subject matter of this disclosure is not limited to what is specifically shown and described.

Claims

1. An irrigation system for irrigating farmland, the irrigation system comprising: Multiple components; Sensors, located only at the central pivot of the irrigation system or at major points of interruption of the utility, and not at each individual component, are configured to generate signals indicating the condition of at least one of the multiple components of the irrigation system based on network power quality. processor; as well as The memory includes instructions stored thereon that, when executed by the processor, cause the irrigation system to: Receive the sensed signal; The network power quality is analyzed over time using a machine learning model based on multiple predetermined data corresponding to the multiple components, in order to determine the status changes of at least one component. Predicting the maintenance requirements of at least one component, wherein the machine learning model comprises at least two of the following: The tail nozzle prediction model is configured to model the motion of the tail nozzle of the irrigation system relative to a predetermined position. A tower-driven prediction model is configured to predict which of a plurality of towers in the irrigation system will move based on a power surge sequence; or A sequential prediction model is configured to model the state and pressure of the solenoid valves in the irrigation system; as well as In response to the prediction, repair of the at least one component is carried out.

2. The irrigation system according to claim 1, wherein, The sensors include: encoders, pressure sensors, flow meters, current sensors, power sensors, voltage sensors, or combinations thereof.

3. The irrigation system according to claim 1, wherein, The irrigation system comprises multiple components including pumps, pivots, towers, terminal towers, corner towers, air compressors, tail nozzles, or combinations thereof.

4. The irrigation system according to claim 1, wherein, When the instructions are executed by the processor, the irrigation system is further configured to: Transmit the predicted maintenance requirements to the user equipment for display; and The predicted maintenance requirements are displayed on the user equipment's screen.

5. The irrigation system according to claim 1, wherein, When the instructions are executed by the processor, the irrigation system is further configured to: Predict the unexpected downtime of at least one component based on predetermined data; as well as The predicted unexpected downtime of the at least one component is displayed on the user equipment's monitor.

6. The irrigation system according to claim 1, wherein, Determining a change in the condition of at least one component includes comparing the sensed signal with predetermined data.

7. The irrigation system according to claim 1, wherein, When executed by the processor, the instructions further cause the irrigation system to receive data from weather stations, field soil moisture sensors, topographic and soil maps, temperature sensors, National Oceanic and Atmospheric Administration (NOAA) meteorological data, or a combination thereof.

8. The irrigation system according to claim 7, wherein, When the instructions are executed by the processor, the irrigation system is further configured to: The condition changes of the at least one component determined based on the received data are improved; and The prediction of maintenance requirements for at least one component is improved based on the determined changes.

9. The irrigation system according to claim 8, wherein, When executed by the processor, the instructions further cause the irrigation system to display an improved prediction of the maintenance requirements on a display screen.

10. The irrigation system according to claim 1, wherein, The prediction is based on a comparison of the power sensed by the sensor with the expected power, which is based on: directly measured soil moisture, soil moisture inferred from meteorological data from field and / or regional weather stations, topographic maps, soil maps, motor speed, gearbox speed ratio, tower weight, span weight, operating status of at least one component, or a combination thereof.

11. A computer-based method for predictive maintenance of an irrigation system, the computer-based method comprising: The system receives signals sensed by sensors indicating the condition of at least one of a plurality of components of an irrigation system based on network power quality. The sensors are located only at the central pivot of the irrigation system or at a major point of interruption of the utility, and not at each individual component. The irrigation system is configured to irrigate a cultivated area and includes the plurality of components. The network power quality is analyzed over time using a machine learning model based on multiple predetermined data corresponding to the multiple components, in order to determine the status changes of at least one component. Predicting the maintenance requirements of at least one component, wherein the machine learning model comprises at least two of the following: The tail nozzle prediction model is configured to model the motion of the tail nozzle of the irrigation system relative to a predetermined position. A tower-driven prediction model is configured to predict which of a plurality of towers in the irrigation system will move given a power surge sequence; or A sequential prediction model is configured to model the state and pressure of the solenoid valves in the irrigation system; as well as In response to the prediction, repair of the at least one component is carried out.

12. The computer implementation method according to claim 11, wherein, The sensors include encoders, pressure sensors, flow meters, current sensors, power sensors, voltage sensors, or combinations thereof.

13. The computer implementation method according to claim 11, wherein, The irrigation system comprises multiple components including pumps, pivots, towers, terminal towers, corner towers, air compressors, tail nozzles, or combinations thereof.

14. The computer implementation method according to claim 11, further comprising: The predicted maintenance needs are transmitted to the user equipment for display. as well as The predicted maintenance requirements are displayed on the user equipment's screen.

15. The computer implementation method according to claim 11, further comprising: Predict the unexpected downtime of at least one component based on predetermined data; as well as The predicted unexpected downtime of the at least one component is displayed on the user equipment's monitor.

16. The computer implementation method according to claim 11, wherein, Determining a change in the condition of at least one component includes comparing the sensed signal with predetermined data.

17. The computer implementation method according to claim 11, further comprising: It receives data from weather stations, field soil moisture sensors, topographic and soil maps, temperature sensors, meteorological data from the National Oceanic and Atmospheric Administration (NOAA), or a combination thereof.

18. The computer implementation method according to claim 17, further comprising: The condition changes of the at least one component are improved based on the received data; The prediction of maintenance requirements for at least one component is improved based on the determined changes; And display the improved prediction of the maintenance requirements on the monitor.

19. The computer implementation method according to claim 11, wherein, The prediction is based on a comparison of the power sensed by the sensor with the expected power, which is based on directly measured soil moisture, soil moisture inferred from meteorological data from field and / or regional weather stations, topographic maps, soil maps, motor speed, gearbox speed ratio, tower weight, span weight, operating status of at least one component, or a combination thereof.

20. A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform a method for predictive maintenance of an irrigation system, the method comprising: The system receives signals sensed by sensors indicating the condition of at least one of a plurality of components of an irrigation system based on network power quality. The sensors are located only at the central pivot of the irrigation system or at a major point of interruption of the utility, and not at each individual component. The irrigation system is configured to irrigate a cultivated area and includes the plurality of components. The network power quality is analyzed over time using a machine learning model based on multiple predetermined data corresponding to the multiple components, in order to determine the status changes of at least one component. Predicting the maintenance requirements of at least one component, wherein the machine learning model comprises at least two of the following: The tail nozzle prediction model is configured to model the motion of the tail nozzle of the irrigation system relative to a predetermined position. A tower-driven prediction model is configured to predict which of a plurality of towers in the irrigation system will move given a power surge sequence; or A sequential prediction model is configured to model the state and pressure of the solenoid valves in the irrigation system; as well as In response to the prediction, repair of the at least one component is carried out.

Images