KR20240034977A - Method for Managing Underwater Pump Failure - Google Patents

Abstract

A method for managing the presence or absence of submersible pump failure is disclosed. A computing device that integrates and manages at least one submersible pump according to an embodiment of the present specification may store a neural network model learned in advance to determine whether the submersible pump is broken. At least one sensor data is collected from at least one submersible pump, and the collected sensor data can be applied to a neural network model to predict the possibility of failure.

Description

{Method for Managing Underwater Pump Failure}

This specification relates to a method for managing the presence or absence of a submersible pump failure.

In general, an underwater pump is a water purification plant that purifies water supplied from a water source and discharges it into the water supply, a sewage and wastewater treatment plant that treats sewage and wastewater discharged from agricultural irrigation, factories or homes, and rainwater. It is used for various purposes, including drainage of sewage and sewage water and supply of water such as cooling water in various places such as drainage pumping stations, subway construction sites, tunnel construction sites, and building construction sites.

The use of submersible pumps is gradually increasing for reasons such as convenience of use and prevention of damage from flooding, but breakdowns occur due to mechanical and electrical defects, hydraulic defects, structural defects, and excessive operating load due to rapid environmental changes. However, due to the special nature of the installation location, inspection was difficult.

In order to solve the above-mentioned problems, the purpose of this specification is to provide a submersible pump failure management method that can more efficiently and accurately predict the possibility of submersible pump failure in advance.

The technical problems to be achieved by the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clear to those skilled in the art from the detailed description of the invention below. It will be understandable.

The submersible pump failure management method according to an embodiment of the present specification may be performed on a computing device (server) that integrates and manages at least one submersible pump.

The submersible pump failure management method includes: storing a pre-trained neural network model to determine whether the submersible pump is malfunctioning; Collecting at least one sensor data from at least one submersible pump; applying the collected sensor data to the neural network model to determine whether the collected sensor data is included in a data pattern that is likely to cause a failure; And when it is determined that there is a possibility of failure of the submersible pump as a result of the prediction of the neural network model, transmitting an alarm to a predetermined external terminal; storing the neural network model includes, Obtaining as learning data a first data set at the time when it was determined to be overloaded and converted to a failure state, and a second data set collected for a predetermined period of time retroactively based on the time when the overload was determined; and training the neural network model to classify overload data patterns based on the first and second data sets.

The first data corresponds to sensor data at the time of determining the overload of the submersible pump, and the second data corresponds to sensor data collected cumulatively from a predetermined period of time to the time of determining the overload. It could be data.

The submersible pump failure management method further includes extracting an overload pattern through an increase/decrease pattern of the sensor value during the predetermined period, wherein the overload pattern includes a current sensor value in a range where the water discharge amount of the submersible pump is the same, At least one of a pattern in which the voltage sensor value, rotation sensor value, pressure sensor value, and temperature sensor value increase, and a pattern in which the ratio of the current sensor value and voltage sensor value to the water discharge amount outside the normal range increases during the predetermined period. may include.

When it is determined that the sensor data collected from the submersible pump is included in the overload pattern as a result of applying the sensor data to the neural network model, predicting and providing information on a future time when the overload failure state will be determined from the current point of time may be further included. there is.

Sensor data used as the learning data may include water discharge amount data of the submersible pump, voltage sensor data, current sensor data, rotation sensor data, pressure sensor data, and temperature sensor data.

According to an embodiment of the present specification, the possibility of failure of a submersible pump can be predicted more efficiently and accurately in advance. The above prediction has the effect of confirming in advance whether to replace parts of the target submersible pump or replace the submersible pump itself.

According to an embodiment of the present specification, by predicting failure in advance before the submersible pump actually goes into a failure state, the pump lifespan can be maximized through stable operation of the submersible pump, and operation control tailored to the field situation is possible.

The effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below. .

The accompanying drawings, which are included as part of the detailed description to aid understanding of the present specification, provide embodiments of the present specification and, along with the detailed description, also explain technical features of the present specification.
1 is an example of a submersible pump control system according to an embodiment of the present specification.
Figure 2 is a block diagram of a computing device on which a submersible pump management method according to an embodiment of the present specification is operated.
Figure 3 is a diagram for explaining an AI device (module) applied to a submersible pump management method according to an embodiment of the present specification.
Figure 4 is a flowchart of a submersible pump management method according to an embodiment of the present specification.
FIG. 5 is a flowchart to explain S400 of FIG. 4 in more detail.
Figure 6 is a diagram for explaining how a submersible pump is managed through a submersible pump control system according to an embodiment of the present specification.
The accompanying drawings, which are included as part of the detailed description to aid understanding of the present invention, provide embodiments of the present invention, and together with the detailed description, explain technical features of the present invention.

Hereinafter, embodiments disclosed in the present specification will be described in detail with reference to the attached drawings. However, identical or similar components will be assigned the same reference numbers regardless of reference numerals, and duplicate descriptions thereof will be omitted. The suffixes “module” and “part” for components used in the following description are given or used interchangeably only for the ease of preparing the specification, and do not have distinct meanings or roles in themselves. Additionally, in describing the embodiments disclosed in this specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in this specification, the detailed descriptions will be omitted. In addition, the attached drawings are only for easy understanding of the embodiments disclosed in this specification, and the technical idea disclosed in this specification is not limited by the attached drawings, and all changes included in the spirit and technical scope of the present invention are not limited. , should be understood to include equivalents or substitutes.

Terms containing ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

When a component is said to be "connected" or "connected" to another component, it is understood that it may be directly connected to or connected to the other component, but that other components may exist in between. It should be. On the other hand, when it is mentioned that a component is “directly connected” or “directly connected” to another component, it should be understood that there are no other components in between.

Singular expressions include plural expressions unless the context clearly dictates otherwise.

In this application, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

1 is an example of a submersible pump control system according to an embodiment of the present specification.

Referring to FIG. 1, the submersible pump control system 1 according to an embodiment of the present specification is a system for controlling a plurality of submersible pumps, and includes a plurality of submersible pumps, a submersible pump controller 200, and a server 100. It can be included.

The plurality of submersible pumps may include submersible pumps installed in at least one area (or site). For example, the plurality of submersible pumps manages the N submersible pumps (P11, P12,..., P1n) installed at the first site as a first group, and the N submersible pumps (P21, P22,...,P2n) can be managed as the second group, and n water pumps (PN1, PN2,...,PNn) installed at the N-th site can be managed as the N-th group. Individual submersible pumps included in each group may include at least one sensor (Sensor 1, Sensor 1,..., Sensor n).

Meanwhile, the submersible pump according to an embodiment of the present specification describes an example in which a plurality of submersible pumps are grouped and managed by region and/or site where they are installed, but the present specification is not limited to this. For example, it goes without saying that the submersible pump control method according to an embodiment of the present specification can be applied to a single submersible pump or individually to specific submersible pumps installed in a specific area.

A submersible pump is installed under water and has an intake port for sucking in fluid and an outlet for discharging the sucked fluid. In addition, the submersible pump includes a motor that generates rotational power with power supplied from the outside, and an impeller connected to the motor to suction and discharge fluid by centrifugal force. The specific configuration and functions of the known configuration of the submersible pump are omitted.

At least one sensor included in the submersible pump includes a voltage/current sensor that detects the voltage and current applied to the submersible pump, a temperature sensor that detects the temperature of the submersible pump, and a sensor that detects the rotation of the impeller through the driving part and shaft of the submersible pump. It may include a rotation sensor that detects a rotation sensor, a water level sensor that determines pump operation according to the water level and water quantity, etc. In addition, it may include various sensors such as water leak detection sensor, reverse rotation sensor, and rotation speed sensor.

The submersible pump controller 200 may control the operation of the submersible pump and obtain sensor data from at least one sensor provided in the submersible pump through a communication unit. The submersible pump controller 200 may convert sensor data collected from at least one submersible pump into standardized sensor data according to a predetermined data structure. The collected sensor data may be transmitted to the server 100 through the communication unit.

The submersible pump controller 200 can be implemented as an individual controller by matching each submersible pump depending on the implementation method. The submersible pump controller 200 may be implemented as a group controller that controls a plurality of submersible pumps installed in one group-managed area. Additionally, the submersible pump controller 200 may be implemented to control a plurality of submersible pumps for each region installed in a plurality of regions, as shown in FIG. 1.

Additionally, in FIG. 1, the submersible pump controller 200 may be configured separately from the server 100, but may also be implemented as an integrated device as a component of the server 100.

The server 100 may perform the function of an integrated management server. The server 100 may determine whether at least one specific submersible pump has a failure, the possibility of failure, the need for replacement, etc., from sensor data acquired from the submersible pump controller 200.

Figure 2 is a block diagram of a computing device in which the submersible pump control method according to an embodiment of the present specification operates. Referring to FIG. 2, the submersible pump control method according to an embodiment of the present specification may be implemented under the control of a processor of a computing device (eg, server).

The computing device 100 may include a sensor data collection unit 110, a failure prediction unit, a memory 130, a communication unit 140, an AI processor 150, and a processor 160.

The sensor data collection unit 110 may collect sensor data from at least one sensor provided in at least one submersible pump. The sensor data collection unit 110 may collect the sensor data from the submersible pump controller 200 shown in FIG. 1 through the communication unit 140.

The collected sensor data can be used as learning data used to learn a neural network model to predict whether the submersible pump is broken, and after learning of the neural network model is completed, the neural network model is used based on the collected sensor data. It can be used as basic data to predict whether a submersible pump will fail. The neural network model may be controlled to be learned through the processor 160, or may be learned through an independently provided AI processor 150.

The learning data may include data that can determine the overload of the submersible pump.

The learning data is a first data set (a data set including water discharge amount data, voltage sensor data, current sensor data, rotation sensor data, pressure sensor data, and temperature sensor data) at the time when the submersible pump was judged to be overloaded and switched to a failure state. ) may include. In addition, the learning data is a second data set (water discharge amount data, voltage sensor data, current sensor data, rotation sensor data, pressure sensor data, and temperature sensor) collected from a predetermined period of time to the overload judgment time based on the overload judgment time of the submersible pump. A data set containing data) may be included.

The second data set may be data having a pattern that can predict failure of the submersible pump due to overload through an increase/decrease pattern of sensor values during the predetermined period. For example, assuming that the computing device (server) collects submersible pump sensor data once a day, submersible pump sensor data can be cumulatively collected for two weeks from two weeks before the overload determination time. The cumulatively collected data has a certain pattern in which the submersible pump progresses to an overload state. During the above two weeks, in the same range of water discharge amount, the electrical characteristic values (current sensor data and voltage sensor data), rotation characteristic value, pressure characteristic value, and temperature characteristic value may show a pattern of gradually increasing. In addition, for example, a pattern in which the degree to which the electrical characteristics (current sensor data and voltage sensor data), rotation characteristic values, pressure characteristic values, and temperature characteristic values deviate from the normal range relative to the water discharge amount gradually increases during the two-week period. I can show you.

The processor 160 may control learning a neural network model based on the first data set and the second data set. The neural network model may be a model learned to determine whether the collected sensor data is included in a data pattern that may cause overload of the submersible pump. Also, for example, the neural network model may be a model learned to determine whether sensor data collected during a predetermined period shows a data pattern that may cause overload. Additionally, for example, the neural network model may be a model learned to output time information that will be determined to be in a failure state due to overload from the current point in time when it is determined that the collected sensor data represents a pattern in which overload may occur.

In other words, the current state of the submersible pump is overloaded and is not judged to be in a malfunction state, but considering the trend or trend of sensor data to be collected in the future, there is a possibility that the submersible pump will be judged to be in a malfunction state due to overload within at least a few days (or weeks). In addition, the submersible pump management method according to an embodiment of the present specification can apply the above-described neural network model to predict and provide time information to be determined to be in a failure state in advance. Accordingly, the submersible pump manager can replace some parts of the submersible pump in advance before the submersible pump goes into a failure state due to overload.

The failure prediction unit 120 may predict the possibility of failure of a specific submersible pump based on sensor data obtained from the submersible pump based on the learned neural network model.

The communication unit 140 includes a wireless communication unit, and the wireless communication unit may include a wireless communication interface such as LTE, WiFi, or 5G system.

Accordingly, according to an embodiment of the present specification, the possibility of failure of the submersible pump can be predicted in advance. The above prediction has the effect of confirming in advance whether to replace parts of the target submersible pump or replace the submersible pump itself.

Figure 3 is a diagram for explaining an AI device (module) applied to a submersible pump control method according to an embodiment of the present specification. In other words, the neural network model learning described above can be learned through the data learning process of the AI device in FIG. 3.

Referring to FIG. 3 , the AI device 20 may include an electronic device including an AI module capable of performing AI processing or a server including an AI module. In addition, the AI device 20 may be included as at least a portion of the submersible pump system and/or the submersible pump control device and may be equipped to perform at least part of the AI processing together.

AI processing may include all operations related to the control unit and/or processor of the submersible pump control device. For example, it is possible to predict failure of a specific submersible pump and determine the need for replacement.

The AI device 20 may be a client device that directly uses AI processing results, or it may be a device in a cloud environment that provides AI processing results to other devices. The AI device 20 is a computing device capable of learning neural networks, and may be implemented as various electronic devices such as servers, desktop PCs, laptop PCs, and tablet PCs.

The AI device 20 may include an AI processor 21, memory 25, and/or a communication unit 27.

The AI processor 21 can learn a neural network using a program stored in the memory 25. In particular, the AI processor 21 uses a neural network to acquire data necessary for submersible pump management, such as whether a specific submersible pump is broken and the need for replacement, from data collected from at least one submersible pump installed in at least one site (or region). You can learn.

Here, a neural network for acquiring data necessary for submersible pump management may be designed to simulate the human brain structure on a computer and may include a plurality of network nodes with weights that simulate neurons of a human neural network. You can. Multiple network modes can exchange data according to each connection relationship to simulate the synaptic activity of neurons sending and receiving signals through synapses. Here, the neural network may include a deep learning model developed from a neural network model. In a deep learning model, multiple network nodes are located in different layers and can exchange data according to convolutional connection relationships. Examples of neural network models include deep neural networks (DNN), convolutional deep neural networks (CNN), Recurrent Boltzmann Machine (RNN), Restricted Boltzmann Machine (RBM), and deep trust. It includes various deep learning techniques such as deep belief networks (DBN) and Deep Q-Network, and can be applied to fields such as computer vision, speech recognition, natural language processing, and voice/signal processing.

Meanwhile, the processor that performs the above-described functions may be a general-purpose processor (e.g., CPU), or may be an AI-specific processor (e.g., GPU) for artificial intelligence learning.

The memory 25 can store various programs and data necessary for the operation of the AI device 20. The memory 25 can be implemented as non-volatile memory, volatile memory, flash-memory, hard disk drive (HDD), or solid state drive (SDD). The memory 25 is accessed by the AI processor 21, and reading/writing/modifying/deleting/updating data by the AI processor 21 can be performed. Additionally, the memory 25 may store a neural network model (eg, deep learning model 26) generated through a learning algorithm for data classification/recognition according to an embodiment of the present invention.

Meanwhile, the AI processor 21 may include a data learning unit 22 that learns a neural network to acquire data necessary for managing the submersible pump. The data learning unit 22 can learn standards for what learning data to use to determine whether the submersible pump is broken and/or replaced, and how to classify and recognize the data using the learning data. The data learning unit 22 may acquire learning data to be used for learning and create a machine learning model using the acquired learning data.

The data learning unit 22 may be manufactured in the form of at least one hardware chip and mounted on the AI device 20. For example, the data learning unit 22 may be manufactured in the form of a dedicated hardware chip for artificial intelligence (AI), or may be manufactured as part of a general-purpose processor (CPU) or a graphics processor (GPU) to be used in the AI device 20. It may be mounted. Additionally, the data learning unit 22 may be implemented as a software module. When implemented as a software module (or a program module including instructions), the software module may be stored in a non-transitory computer readable recording medium that can be read by a computer. In this case, at least one software module may be provided by an operating system (OS) or an application.

The data learning unit 22 may include a learning data acquisition unit 23 and a model learning unit 24.

The learning data acquisition unit 23 may acquire learning data required for a neural network model for classifying and recognizing data.

The model learning unit 24 can use the acquired training data to train the neural network model to have a judgment standard on how to classify certain data. At this time, the model learning unit 24 can learn a neural network model through supervised learning that uses at least some of the learning data as a judgment standard. Alternatively, the model learning unit 24 can learn a neural network model through unsupervised learning, which discovers a judgment standard by learning on its own using training data without guidance. Additionally, the model learning unit 24 can learn a neural network model through reinforcement learning using feedback on whether the result of the situational judgment based on learning is correct. Additionally, the model learning unit 24 may learn a neural network model using a learning algorithm including error back-propagation or gradient descent.

When the neural network model is learned, the model learning unit 24 may store the learned neural network model in memory. The model learning unit 24 may store the learned neural network model in the memory of a server connected to the AI device 20 through a wired or wireless network.

The data learning unit 22 further includes a learning data preprocessing unit (not shown) and a learning data selection unit (not shown) to improve the analysis results of the recognition model or save the resources or time required for generating the recognition model. You may.

The learning data preprocessor may preprocess the acquired data so that the acquired data can be used for learning to determine the situation. For example, the learning data preprocessor may process the acquired data into a preset format so that the model learning unit 24 can use the acquired learning data for training for image recognition.

In addition, the learning data selection unit may select data necessary for learning among the learning data acquired in the learning data acquisition unit 23 or the learning data preprocessed in the preprocessor. The selected learning data will be provided to the model learning unit 24. You can.

Additionally, the data learning unit 22 may further include a model evaluation unit (not shown) to improve the analysis results of the neural network model.

The model evaluation unit inputs evaluation data into the neural network model, and when the analysis result output from the evaluation data does not satisfy a predetermined standard, the model learning unit 22 can perform re-training. In this case, the evaluation data may be predefined data for evaluating the recognition model. As an example, the model evaluation unit may evaluate the evaluation data as not meeting a predetermined standard if the number or ratio of inaccurate evaluation data exceeds a preset threshold among the analysis results of the learned recognition model for the evaluation data. there is.

The communication unit 27 can transmit the results of AI processing by the AI processor 21 to an external electronic device. For example, the external electronic device may include a server that integrates and manages the submersible pump. Meanwhile, the AI device 20 shown in FIG. 3 has been described as functionally divided into an AI processor 21, a memory 25, a communication unit 27, etc., but the above-described components are integrated into one module to form an AI module. Please note that it may also be referred to as .

Figure 4 is a flowchart of a submersible pump management method according to an embodiment of the present specification.

The submersible pump management method can be implemented through the processor 160 of the server 100 described in FIGS. 1 and 2.

Referring to FIG. 4, the processor 160 may learn a neural network model that determines whether the submersible pump is broken and store it in advance in memory (S400). Step S400 is explained in more detail in FIG. 5.

The processor 160 collects at least one sensor data from at least one submersible pump (S410).

The processor 160 may receive the sensor data through the submersible pump controller 200 shown in FIG. 1. According to one embodiment, the submersible pump controller 200 may be an integrated controller that collects a plurality of sensor data from a plurality of submersible pumps installed in each area (site), as shown in FIG. 1. According to one embodiment, unlike shown in FIG. 1, the submersible pump controller 200 may be a controller assigned to each individual submersible pump, or may be a controller that manages a plurality of submersible pumps managed as a group. The submersible pump controller 200 collects sensor data of each submersible pump, performs a data processing process to correspond to standard specifications so that the collected sensor data can be integrated and managed, and then sends the integrated sensor data to the server 100. can be transmitted to.

The submersible pump failure management method according to an embodiment of the present specification may transmit sensor data to the submersible pump controller 200 through a PLC panel corresponding to each submersible pump. In addition, for example, a separate PLC panel that performs the function of integrating sensor data collected for each submersible pump may be embedded in the submersible pump controller 200 and transmit the integrated sensor data to the server 100.

The processor 160 may apply the collected sensor data to the neural network model (S420) and determine whether the collected sensor data is included in a data pattern that is likely to cause a failure (S430).

If it is determined that there is a possibility of failure of the submersible pump as a result of the prediction of the neural network model, the processor 160 may transmit an alarm to a predetermined external terminal (S440). The predetermined external terminal may include a terminal of a manager of the submersible pump, a submersible pump control center terminal, etc.

FIG. 5 is a flowchart to explain S400 of FIG. 4 in more detail.

The processor 160 acquires learning data to learn the neural network model. The learning data may include at least one sensor data that can indicate the operating state during the operation of the submersible pump. A neural network model that has the function of diagnosing a failure of the submersible pump is learned by using sensor data collected during the actual operation of the submersible pump as learning data.

The processor 160 uses a first data set at the time when the submersible pump is judged to be overloaded and converted to a failure state, and a second data set collected for a predetermined period of time retroactively based on the time of overload determination, as learning data. It can be obtained (S500).

For detailed descriptions of the first and second data sets, refer to the description of the sensor data collection unit 110 of FIG. 2.

Here, the first data set is sensor data at the time an overload occurs, and has the meaning of reference data for determining the presence or absence of a failure. Meanwhile, the second data set is sensor data collected before it is determined whether the submersible pump is broken due to overload, and the second data set is accumulated in a pattern that converges to the characteristics of the first data set. It can mean data.

The processor 160 trains the neural network model to classify overload data patterns based on the first and second data sets (S510).

The processor 160 may extract an overload pattern through an increase/decrease pattern of sensor values while the second data set is collected for a predetermined period in order to determine whether a failure exists through the neural network model. The overload pattern is a first pattern in which the current sensor value and voltage sensor value increase in the same range as the water discharge amount of the submersible pump, and the ratio of the current sensor value and voltage sensor value to the water discharge amount during the predetermined period is outside the normal range. It may include at least one pattern among the increasing second patterns. The processor 160 trains the neural network model to recognize the extracted first pattern and/or second pattern as an overload pattern.

The processor 160 may store the learned neural network model in memory (S520).

The learned neural network model may be a neural network model corresponding to a specific submersible pump. Accordingly, the processor 160 can learn a neural network model for determining whether a failure exists corresponding to each of the plurality of submersible pumps and store it in memory. Alternatively, the above-described failure was limited to an overload state, but in this specification, a neural network model corresponding to various failure states that can occur in a submersible pump can be learned and stored.

Alternatively, according to one embodiment, rather than a neural network model matched to each submersible pump, a neural network model commonly applicable to a plurality of submersible pumps may be learned.

Figure 6 is a diagram for explaining how a submersible pump is managed through a submersible pump control system according to an embodiment of the present specification.

The submersible pump control device can collect sensor data from the submersible pump and transmit it to the server (S600).

The server can determine whether the submersible pump is broken through AI processing operations. According to one embodiment, the server inputs sensor data into a previously learned neural network model (S610), and determines whether the sensor data input to the neural network model is sensor data with an overload pattern through the results of the neural network model (S610). 620).

Here, in order to determine whether the sensor data has an overload pattern, the sensor data may include the collection period and sensor data collected cumulatively so that the collection time can be maintained for at least a certain section.

If the server determines that the sensor data indicates an overload pattern as a result of AI processing, it may generate an alarm signal (S630). The server can transmit a failure alarm to the submersible pump control device (S640).

The submersible pump control device may analyze the received alarm signal (650) and determine whether to replace at least some of the parts of the target submersible pump or the submersible pump itself (S660). In addition to a simple alarm signal, the alarm signal may include information indicating the type of failure, and information calculated as a probability of the failure occurring within a specific time in the future based on the current time. Accordingly, the manager who manages the submersible pump control device can predict and prepare for failure in advance based on the current operating state of the submersible pump.

The above-mentioned specification can be implemented as computer-readable code on a program-recorded medium. Computer-readable media includes all types of recording devices that store data that can be read by a computer system. Examples of computer-readable media include HDD (Hard Disk Drive), SSD (Solid State Disk), SDD (Silicon Disk Drive), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. There is. Accordingly, the above detailed description should not be construed as restrictive in all respects and should be considered illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

Claims (6)

In a method of managing whether a submersible pump is broken in a computing device that integrates and manages at least one submersible pump,
Saving a previously learned neural network model to determine whether a submersible pump is broken;
Collecting at least one sensor data from at least one submersible pump;
applying the collected sensor data to the neural network model to determine whether the collected sensor data is included in a data pattern that is likely to cause a failure; and
As a result of the prediction of the neural network model, if it is determined that there is a possibility of failure of the submersible pump, transmitting an alarm to a predetermined external terminal;
The step of saving the neural network model is,
Acquiring as learning data a first data set at the time when the submersible pump was determined to be overloaded and converted to a failure state, and a second data set collected for a predetermined period of time retroactively based on the time when the overload was determined; and
training the neural network model to classify overloaded data patterns based on the first and second data sets;
A submersible pump failure management method comprising: According to claim 1,
The first data corresponds to sensor data at the time of determining overload of the submersible pump,
The second data is a submersible pump failure management method, characterized in that it corresponds to sensor data collected cumulatively from a predetermined period of time based on the overload determination time until the overload judgment time. According to claim 2,
Further comprising: extracting an overload pattern through an increase/decrease pattern of the sensor value during the predetermined period,
The overload pattern is,
A pattern in which the current sensor value, voltage sensor value, rotation sensor value, pressure sensor value, and temperature sensor value increase in the same range as the water discharge amount of the submersible pump, and the current sensor value, voltage sensor value, and rotation relative to the water discharge amount during the predetermined period. A submersible pump failure management method comprising at least one of the following patterns in which the ratio of sensor values, pressure sensor values, and temperature sensor values deviates from the normal range increases. According to claim 3,
When it is determined that the sensor data collected from the submersible pump is included in the overload pattern as a result of applying the sensor data to the neural network model, predicting and providing information on a future time when the overload failure state will be determined from the current time;
A submersible pump failure management method further comprising: According to claim 2,
The sensor data used as the learning data is,
A method for managing the presence or absence of a failure of a submersible pump, comprising the water discharge amount data, voltage sensor data, current sensor data, rotation sensor data, pressure sensor data, and temperature sensor data of the submersible pump. According to claim 1,
As a result of the prediction of the neural network model, when it is determined that there is a possibility of failure of the submersible pump, transmitting an alarm to a predetermined external terminal through a communication unit;
A submersible pump failure management method further comprising: