KR20230007583A - Recording medium storing computer program and system of predicting, diagnosing and monitoring for energy storage system accident. - Google Patents

Abstract

The present invention relates to a recording medium-stored computer program and system for predicting, diagnosing, and monitoring an accident in an energy storage system (ESS) and, more specifically, to a recording medium-stored computer program and system for predicting, diagnosing, and monitoring an accident in an ESS, which can diagnose and predict accidents in an ESS through a failure diagnosis model learned by using electrical and environmental data of the ESS and can visualize and output the electrical data, environmental data, and fault diagnosis results, thereby enabling an administrator to effectively monitor the status of ESS. The recording medium-stored computer program comprises a data collection and storage means, a failure diagnosis means, a visualization means, and a notification means.

Description

Recording medium storing computer program and system of predicting, diagnosing and monitoring for energy storage system accident.}

The present invention relates to a computer program and system capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium, and more particularly, using electrical data and environmental data of an energy storage system (ESS). It is possible to diagnose and predict accidents of the energy storage system (ESS) through a pre-learned failure diagnosis model, and visualize and output the electrical data, environmental data, and failure diagnosis results so that the administrator can effectively A computer program and system capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium capable of monitoring the state of an energy storage system (ESS).

As the use of fossil fuels causes serious climate change due to warming caused by the emission of a large amount of CO2, restrictions on the use of fossil fuels are becoming increasingly severe.

Accordingly, major developed countries are creating new markets and industries such as renewable energy, energy storage devices, and virtual power plants by strengthening regulations and support at the national level. In particular, energy storage devices (ESS) that can maximize energy use efficiency :Energy Storage System) occupies an important place.

At this time, the Energy Storage System (ESS) integrates various components such as the Battery Management System (BMS) and the Power Conditioning System (PCS) into one device according to the purpose for integrated management. As a device that controls and controls energy consumption, it is mainly used for reducing power costs and efficient energy use because it can store energy in a battery such as a secondary battery when power consumption is low and supply it when power consumption is high.

Accordingly, the frequency of using an energy storage system (ESS) is increasing in places with high energy consumption density, such as companies, schools, homes, factories, and the like.

However, energy storage systems (ESS: Energy Storage System) can cause various accidents due to risk factors that may occur in the configuration itself, risk factors that may occur during operation, and failures. As cases leading to accidents increase, user confidence in energy storage systems (ESS) is declining.

Accordingly, in the related art, a manager periodically inspects an energy storage system (ESS) or directly compares and analyzes measured data to diagnose and deal with failures.

However, it is cumbersome and difficult to diagnose failures by individually checking and comparing and analyzing vast amounts of data accumulated in real time, and it is difficult to grasp various data at a glance, so it takes a lot of time to accurately diagnose failures.

Therefore, even if the manager does not perform periodic inspections, it is possible to compare and analyze the data measured by the energy storage system (ESS: Energy Storage System) to diagnose failures, and to analyze various data related to the energy storage system (ESS: Energy Storage System). It is necessary to have a technology that can provide monitoring services so that you can understand at a glance.

Republic of Korea Patent Publication No. 10-2017-0114087 (2017.10.13)

The present invention has been made to solve the above problems, and an object of the present invention is to diagnose and predict accidents based on electrical data and environmental data received in real time from an energy storage system (ESS). It is to provide a system capable of predicting, diagnosing, and monitoring accidents of an energy storage system in which a manager can effectively monitor the state of an energy storage system (ESS) because it can visually express various types of measured data.

In order to achieve the above object, the present invention receives state data having electrical characteristic information and environmental data having internal environmental characteristic information from a computer, energy storage system (ESS) in real time, and stores the data in a database. data collection and storage means; a failure diagnosis means for diagnosing and predicting a failure of the energy storage system (ESS) by inputting the received state data and environmental data into a failure diagnosis model learned in advance; a visualization means for visualizing and outputting the received state data and environment data and a failure diagnosis result of the energy storage system (ESS) from the failure diagnosis means; and a notification unit for transmitting the diagnosis result to the outside when the failure diagnosis unit diagnoses the energy storage system (ESS) as a failure.

In a preferred embodiment, the data collection and storage unit interworks with the Meteorological Administration server to receive real-time weather data of an area where the Energy Storage System (ESS) is located and further store it in the database, and the visualization unit visualizes the weather data and provides it.

In a preferred embodiment, the failure diagnosis model is a machine learning model trained with an LSTM algorithm.

In a preferred embodiment, the failure diagnosis and prediction means: includes a data extraction means for extracting data of items used for learning the failure diagnosis model from among the received state data and environment data; data normalization means for normalizing data of the extracted items; and a diagnosis result output means for diagnosing a failure of the energy storage system (ESS) by inputting the normalized data into the failure diagnosis model and outputting a diagnosis result.

In a preferred embodiment, the failure diagnosis and prediction unit may classify and output the diagnosis result into safety, caution, and failure stages through the failure diagnosis model, and if the diagnosis result is a warning or failure, the corresponding log may be output. record in database

In addition, the present invention may provide a server storing a computer program capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium.

In addition, the present invention can receive data from an energy storage system (ESS) through a communication network, can diagnose a failure based on the received data, and can diagnose the received data and the energy storage system (ESS). A server storing a computer program capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium that visualizes and provides failure diagnosis results of the Energy Storage System; And a display device outputting visualized information through the server; it is possible to provide a system capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium, characterized in that it includes.

The present invention has the following excellent effects.

According to the computer program and system capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium of the present invention, real-time status information and environmental information for an energy storage device (ESS: Energy Storage System) can be visualized and provided There is an advantage in that an administrator can grasp the state of an energy storage system (ESS) at a glance.

In addition, according to the computer program and system capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium of the present invention, data having electrical information and environmental information is stored in real time from an energy storage system (ESS). Since accidents can be diagnosed and predicted by inputting them into a pre-learned failure diagnosis model, managers can quickly and easily check and respond to abnormal signs.

1 is a diagram for explaining an energy storage device accident prediction, diagnosis, and monitoring system according to an embodiment of the present invention;
2 is a diagram for explaining a computer program capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium according to an embodiment of the present invention;
3 is a diagram for explaining a failure diagnosis means in a computer program capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium according to an embodiment of the present invention.

The terms used in the present invention have been selected from general terms that are currently widely used as much as possible, but in certain cases, there are terms arbitrarily selected by the applicant. Therefore, its meaning should be understood.

Hereinafter, the technical configuration of the present invention will be described in detail with reference to preferred embodiments shown in the accompanying drawings.

However, the present invention is not limited to the embodiments described herein and may be embodied in other forms. Like reference numbers indicate like elements throughout the specification.

1 is a diagram for explaining an energy storage device accident prediction, diagnosis, and monitoring system according to an embodiment of the present invention.

Referring to FIG. 1 , the present invention is a system 100 capable of diagnosing and predicting failures using data received from an energy storage device 10 and visualizing the received data and diagnosis results, a server 110 ) and a display device 120.

Here, the energy storage device 10 is a device capable of storing and outputting energy, and may include a battery management system (BMS) and a power storage system (PCS). Depending on the purpose of use, it can be configured as one device by interlocking with various components.

In addition, the energy storage device 10 is equipped with various sensors capable of measuring electrical information and environmental information of the energy storage device 10 .

The server 110 is connected to the energy storage device 10 through a communication network and can receive the electrical information and environmental information measured by the centers installed in the energy storage device 10 .

In addition, the server 110 is interlocked with the Korea Meteorological Administration server through a communication network, and may further receive weather information.

In addition, the server 110 stores a computer program 1000 capable of diagnosing and predicting failure of the energy storage device 10 and monitoring it.

In addition, the present invention may separately provide only the server 110 in which the computer program 1000 is stored.

Meanwhile, the computer program 1000 may be stored and provided in various computers other than the server 110 .

Here, the computer is a computer in a broad sense that includes not only a general personal computer, but also a server computer, a cloud system, smart devices such as smart phones and tablets, and embedded systems that can be accessed through a communication network.

In addition, the computer program 1000 may be stored and provided in a separate recording medium, and the recording medium may be specially designed and configured for the present invention or known and usable to those skilled in the computer software field. it could be

For example, the recording medium includes magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CDs and DVDs, magneto-optical recording media capable of both magnetic and optical recording, ROM, RAM, and flash memory. etc., alone or in combination, may be a hardware device specially configured to store and execute program instructions.

In addition, the computer program 1000 may be a program composed of program instructions, local data files, local data structures, etc. alone or in combination, and can be programmed into a computer using an interpreter or the like as well as machine code generated by a compiler. It can be a program written in high-level language code that can be executed by

In addition, the display device 120 is a device capable of outputting visually expressed data, and may be a computer monitor, a smartphone liquid crystal, etc., and the type is not limited.

That is, in the present invention, data collection and failure of the energy storage device 10 can be diagnosed and predicted through the computer program 1000 stored in the server 110, and the display device 120 interlocked with the communication network It is a system 100 capable of visualizing and outputting various data and diagnosis results of the energy storage device 10.

Hereinafter, a computer program 1000 capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium according to the present invention will be described in detail.

2 is a diagram for explaining a computer program capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium according to an embodiment of the present invention.

Referring to FIG. 2, a computer program 1000 capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium includes a data collection and storage means 1100, a failure diagnosis means 1200, a visualization means 1300, and a notification. Means 1400 are made.

First, the data collection and storage means 1100 performs a function of receiving data from the energy storage device 10 in real time and storing them in the database 1 .

The database 1 is created in a physical server to store and manage various data, but may also be created in a cloud-based virtual server capable of storing and managing various data by transmitting them to the outside.

Meanwhile, the data are state data and environment data of the energy storage device 10 .

In detail, the state data is data having electrical characteristic information of the energy storage device 10, and includes the voltage, current, and power of the battery management system (BMS) and the power conversion device (PCS: Power Storage System). System) voltage and current.

In addition, the environmental data is data having environmental characteristic information of the energy storage device 10, and may include temperature, humidity, and fine dust within the energy storage device 10.

In addition, the data collection and storage unit 1100 may receive and store weather data from the Korea Meteorological Administration server linked to a communication network in addition to the state data and environment data.

In this case, the weather data is weather information data of an area where the energy storage device 10 is located, and may be information such as precipitation, wind, and sunlight.

3 is a diagram for explaining a failure diagnosis means in a computer program capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium according to an embodiment of the present invention.

Referring to FIG. 3 , the failure diagnosis unit 1200 performs a function of diagnosing and predicting a failure by inputting the data received in real time from the energy storage device 10 into a failure diagnosis model, and the data extraction unit 1210 ), a data normalization unit 1220 and a diagnosis result output unit 1230.

The failure diagnosis model is a machine learning model having an artificial neural network structure, and the failure-related data of the energy storage device 10 collected in the database 1 is of a Long Short-Term Memory (LSTM) structure. It is a model built by learning with an algorithm.

Here, the LSTM (Long Short-Term Memory) is a type of artificial neural network having an internal circulation structure capable of repeatedly learning data changing over time.

On the other hand, the failure diagnosis model can build a machine learning model using an algorithm having a structure such as a recurrent neural network (RNN), a convolutional neural network (CNN), and a residual network (ResNet) in addition to long short-term memory (LSTM). can

The data extractor 1210 extracts data of main items from among the data received from the energy storage device 10 .

The data of the main items are the data of items used for learning to build the failure diagnosis model, and when the failure diagnosis model is built, the data are compared and analyzed to have a fatal effect on the failure of the energy storage device 10 These are the data of items selected as elements.

That is, the data extractor 1210 extracts data of the same items used when learning the failure diagnosis model to be input to the failure diagnosis model to diagnose failure of the energy storage device 10 .

The data normalization unit 1220 sets the range and size of the data of the extracted items in different units and because the accuracy of the diagnosis result through the failure diagnosis model may be lowered due to the difference in the range and size of the measured value. Normalize to be unified.

In this case, the normalization may use algorithms such as Standardsaler, Rovustscaler, and Minmaxscaler, and the type is not limited.

In addition, the normalization is equally performed on data used when constructing the failure diagnosis model.

The diagnosis result output unit 1230 outputs a failure diagnosis result of the energy storage device 10 by inputting the normalized data to the failure diagnosis model.

At this time, the failure diagnosis means 1200 may classify and output safety, caution, and failure stages through the failure diagnosis model, and record the corresponding log in the database 1 when the diagnosis result is caution or failure. do.

The visualization unit 1300 performs a function of visually expressing the received data so that it can be grasped at a glance.

In detail, the visualization unit 1300 receives electrical information and environmental information received from the energy storage device 10, weather information of the area where the energy storage device 10 is located, and the failure diagnosis unit. It performs the function of visualizing and providing the received failure diagnosis result.

At this time, the visualization unit 1300 visualizes and provides the data in the form of characters, graphs, diagrams, charts, figures, etc., but is not limited thereto and may visualize and provide the data in various forms that can be visually expressed.

The notification means 1400 performs a function of transmitting a notification to the outside so that a manager can recognize an anomaly when an anomaly is detected.

In detail, when the energy storage device 10 is diagnosed as a caution or failure in the failure diagnosis means 1200, the notification unit 1400 transmits a speaker and monitor in the management room and a smart device linked to the manager through a communication network. Sends a warning or failure notification.

Therefore, the present invention can visualize and provide real-time state information and surrounding environment information on the energy storage device 10, so that a manager can grasp the state of the energy storage device 10 at a glance and learn failures in advance. It has the advantage of being able to diagnose and predict failures by inputting data received in real time into the diagnosis model so that managers can quickly and conveniently check and deal with abnormal signs.

As described above, the present invention has been shown and described with preferred embodiments, but is not limited to the above embodiments, and to those skilled in the art within the scope of not departing from the spirit of the present invention Various changes and modifications will be possible.

1: database 10: energy storage device
100: failure diagnosis and monitoring system of energy storage device
110: server 120: display device

Claims (7)

computer,
Data collection and storage means for receiving in real time state data having electrical characteristic information and environmental data having internal environmental characteristic information from an energy storage system (ESS) and storing them in a database;
a failure diagnosis means for diagnosing and predicting a failure of the energy storage system (ESS) by inputting the received state data and environmental data into a failure diagnosis model learned in advance;
a visualization means for visualizing and outputting the received state data and environment data and a failure diagnosis result of the energy storage system (ESS) from the failure diagnosis means; and
When the failure diagnosis means diagnoses the energy storage system (ESS: Energy Storage System) as a failure, a notification means for transmitting the failure diagnosis result to the outside; predicting an energy storage device accident stored in a recording medium , a computer program capable of diagnosing and monitoring.
According to claim 1,
The data collection and storage unit interworks with the server of the Korea Meteorological Administration to receive real-time weather data of an area where the Energy Storage System (ESS) is located and further store the data in the database;
The computer program capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium, characterized in that the visualization unit visualizes and provides the weather data.
According to claim 1,
The failure diagnosis model is a machine learning model learned with an LSTM algorithm.
According to claim 3,
The failure diagnosis and prediction means: silver
a data extraction means for extracting data of items used for learning the failure diagnosis model from among the received state data and environment data;
data normalization means for normalizing data of the extracted items; and
and a diagnosis result output means for diagnosing a failure of the energy storage system (ESS) by inputting the normalized data into the failure diagnosis model and outputting a diagnosis result. A computer program capable of predicting, diagnosing, and monitoring storage device accidents.
According to claim 4,
The failure diagnosis and prediction unit may classify and output the diagnosis result into safety, caution, and failure stages through the failure diagnosis model,
A computer program capable of predicting, diagnosing and monitoring an energy storage device accident stored in a recording medium, characterized in that for recording a corresponding log in the database when the diagnosis result is a warning or failure.
A server storing a computer program capable of predicting, diagnosing, and monitoring an energy storage device accident stored in a recording medium according to any one of claims 1 to 5.
Data may be received from an energy storage system (ESS) through a communication network, failure may be diagnosed based on the received data, and the received data and the energy storage system (ESS: Energy Storage System) may receive data. The server of claim 6 that visualizes and provides a failure diagnosis result of; and
An energy storage device accident prediction, diagnosis, and monitoring system comprising a; display device outputting visualized information through the server.

Images