Method and device for detecting abnormalities in windows-based electrodialysis equipment

Abstract

A method for detecting abnormalities in window-based electrodialysis equipment, according to one embodiment, comprises the steps of: setting an input window range for a control variable of electrodialysis equipment; converting time-series data of the control variable into continuous section unit data corresponding to the input window range; predicting a management variable corresponding to the control variable through an artificial intelligence model trained on the basis of the continuous section unit data; and comparing the predicted management variable to a reference value so as to detect whether there are abnormalities in the electrodialysis equipment.

Description

Method and device for detecting abnormalities in Windows-based electrodialysis equipment

[0001] The present invention relates to a method and device for detecting abnormalities in a Windows-based electrodialysis equipment, and more particularly, to a method and device for detecting abnormalities in a Windows-based electrodialysis equipment for optimal automated operation of a BPED (BiPolar ElectroDialysis) process.

[0002] Bipolar Electrodialysis (BPED) is an electrochemical separation process that uses ion exchange materials to separate or concentrate specific ions from a solution. BPED is gaining attention as a particularly important water separation and regeneration technology.

[0003] BPED operational data may include various data collected during the operation of the technology. BPED operational data may include, for example, current and voltage data, material concentration data, temperature and pressure data, and time data.

[0004] One embodiment of the present invention provides a window-based electrodialysis equipment abnormality detection method and device, which is a preprocessing process specialized for operation data, sets the size and ratio of an appropriate input window and a prediction window, converts continuous interval unit data, and then applies a time series model to predict management variables of each stage of BPED (BiPolar ElectroDialysis).

[0005] Among the embodiments, a window-based electrodialysis equipment abnormality detection method includes a step of setting an input window range for a control variable of the electrodialysis equipment, a step of converting time series data of the control variable into continuous interval unit data corresponding to the input window range, a step of predicting a management variable corresponding to the control variable through an artificial intelligence model learned based on the continuous interval unit data, and a step of comparing the predicted management variable with a reference value to detect whether the electrodialysis equipment is abnormal.

[0006] The above control variables may include the input flow rate of the solution of each stage of the electrodialysis equipment, the circulation flow rate, and the voltage of the rectifier, and the above management variables may include the conductivity of each stage and the stack voltage of the rectifier.

[0007] The step of setting the input window range may include a step of setting the input window range by reflecting the propagation time for the solution injected into each stage of the electrodialysis equipment to be ion-exchanged and then passed to the next stage.

[0008] The step of setting the input window range may include a step of controlling the propagation time by adjusting the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate among the control variables.

[0009] The method may further include a step of setting a prediction window range for a management variable based on the set input window range, and the step of setting the prediction window range may include a step of setting the prediction window range such that a ratio of the prediction window range and the input window range is 1 to 1.

[0010] The step of converting into the above continuous section unit data may include a step of performing data cleaning and data interpolation preprocessing on the continuous section data within the input window range.

[0011] The step of predicting the above management variable may include a step of calculating the degree of matching of the learned artificial intelligence model and resetting the input window range and the prediction window range based on the calculated degree of matching.

[0012] The step of resetting the input window range and the prediction window range may include a step of controlling the propagation time by adjusting the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate.

[0013] The step of predicting the above management variable may include a step of matching a target time interval between a target time of the prediction window range and a start point of the input window range with the input window range.

[0014] The above artificial intelligence model may include a deep learning model for time series data processing, including an informer model and a DLinear model.

[0015] Among the embodiments, a window-based electrodialysis equipment abnormality detection device includes a preprocessing unit that sets an input window range for a control variable of the electrodialysis equipment, a data conversion unit that converts time series data of the control variable into continuous section unit data corresponding to the input window range, a prediction unit that predicts a management variable corresponding to the control variable through an artificial intelligence model learned based on the continuous section unit data, and an abnormality detection unit that compares the predicted management variable with a reference value to detect whether the electrodialysis equipment is abnormal.

[0016] The above control variables may include the input flow rate of the solution of each stage of the electrodialysis equipment, the circulation flow rate, and the voltage of the rectifier, and the above management variables may include the conductivity of each stage and the stack voltage of the rectifier.

[0017] The above preprocessing unit may include an input window setting unit that sets the input window range by reflecting the propagation time for the solution injected into each stage of the electrodialysis equipment to be ion-exchanged and passed to the next stage.

[0018] The above input window setting section can control the propagation time by adjusting the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate among the above control variables.

[0019] The above preprocessing unit may further include a prediction window setting unit that sets the prediction window range based on the input window range so that the ratio of the prediction window range and the input window range is 1 to 1.

[0020] The above data conversion unit can perform data cleaning and data interpolation preprocessing on continuous section data within the input window range.

[0021] The above prediction unit calculates the matching degree of the learned artificial intelligence model, and the preprocessing unit can reset the input window range and the prediction window range based on the matching degree until the matching degree becomes higher than a reference level.

[0022] The above input window setting unit can control the propagation time by adjusting the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate, and can reset the input window range by reflecting the propagation time.

[0023] The above prediction unit can adjust the target time interval between the target time of the prediction window range and the start point of the input window range to match the input window range.

[0024] The above artificial intelligence model may include a deep learning model for time series data processing, including an informer model and a DLinear model.

[0025] A method and device for detecting abnormalities in a window-based electrodialysis facility according to one embodiment of the present invention is a preprocessing process specialized for operation data, which sets the size and ratio of an appropriate input window and a prediction window, converts continuous interval unit data, and then applies a time series model to predict management variables of each stage of BPED (BiPolar ElectroDialysis).

[0026] FIG. 1 is a drawing showing the configuration of an electrodialysis facility according to one embodiment of the present invention.

[0027] FIG. 2 is a drawing showing one end of an electrodialysis facility according to one embodiment of the present invention.

[0028] FIG. 3 is a block diagram showing a preprocessing process of a method for detecting abnormalities in a Windows-based electrodialysis equipment according to one embodiment of the present invention.

[0029] FIGS. 4A to 4D are schematic diagrams showing an input window and a prediction window set according to one embodiment and a comparative example of the present invention.

[0030] FIG. 5 is a drawing showing an example of a method for detecting abnormalities in a window-based electrodialysis facility according to one embodiment of the present invention.

[0031] FIG. 6 is a flowchart of a method for detecting abnormalities in a window-based electrodialysis equipment according to one embodiment of the present invention.

[0032] FIG. 7 is a block diagram of a window-based electrodialysis equipment abnormality detection device according to one embodiment of the present invention.

[0033] Figures 8a to 8g are drawings showing prediction results according to one embodiment and a comparative example of the present invention.

[0034] FIG. 9 is a drawing for explaining a computing device according to one embodiment of the present invention.

[0035] Below, with reference to the attached drawings, embodiments of the present invention are described in detail so that those skilled in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In addition, in the drawings, parts irrelevant to the description are omitted for clarity of description, and similar parts are designated with similar reference numerals throughout the specification.

[0036] Throughout the specification and claims, whenever a part is referred to as "comprising" a component, this does not exclude other components, but rather includes other components, unless otherwise stated. Terms including ordinal numbers, such as "first," "second," etc., may be used to describe various components, but these components are not limited by these terms. These terms are used solely to distinguish one component from another.

[0037] Terms such as “part,” “unit,” and “module” described in the specification may mean a unit capable of processing at least one function or operation described in the specification, which may be implemented by hardware or a circuit, software, or a combination of hardware or a circuit and software.

[0038] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0039] Fig. 1 is a drawing showing the configuration of an electrodialysis facility according to one embodiment of the present invention. Fig. 2 is a drawing showing one end of an electrodialysis facility according to one embodiment of the present invention.

[0040] In FIG. 1 and FIG. 2, the electrodialysis equipment may be a BPED (Bipolar ElectroDialysis) equipment. That is, the electrodialysis equipment may be a bipolar electrodialysis equipment. The electrodialysis equipment (BPED) may be equipment that converts a lithium sulfate aqueous solution into lithium hydroxide and sulfuric acid. Here, lithium sulfate is Li2SO4, lithium hydroxide is LiOH, and sulfuric acid is H2SO4. The electrodialysis equipment (BPED) may be an aqueous solution treatment equipment that simultaneously performs water decomposition / ion separation using an electrodialysis membrane in an electric field.

[0041] Referring to FIGS. 1 and 2, the electrodialysis device (BPED) may include a cation exchange membrane (CEM), an anion exchange membrane (AEM), and a bipolar membrane (BPM).

[0042] A cation exchange membrane (CEM) has an internal anionic group, allowing only cations (e.g., Li+) to pass through. An anion exchange membrane (AEM) allows only anions (e.g., SO42-) to pass through due to its internal cation group. A bipolar membrane (BPM) consists of a cation membrane and an anion membrane overlapping each other with a water-splitting catalyst in between. A bipolar membrane (BPM) can decompose water in an electric field to produce hydrogen ions (H+) and hydroxide ions (OH-).

[0043] That is, the electrodialysis equipment (BPED) may be an aqueous solution treatment equipment that simultaneously performs water splitting (decomposition into H+, OH-) / ion separation (ion separation of Li+, SO42-) using an electrodialysis membrane (cation dialysis membrane, anion dialysis membrane, bipolar membrane) in an electric field.

[0044] In Fig. 1, in a BPED device, the LS solution can transfer Li and SO4 ions to the LH solution and sulfuric acid (H2SO4) solution through a three-stage process (Press) including the first to third stages. For example, in the BPED device, deionized water (DI water) can be converted into the LH solution and sulfuric acid solution by contacting the LS solution in a counterflow manner. Here, the LS solution is lithium sulfate (Li2SO4), and the LH solution is lithium hydroxide (LiOH).

[0045] The first to third stages may each include a salt room, an acid room, a base room, a salt tank, an acid tank, and a base tank.

[0046] In each stage, the Salt chamber supplies lithium sulfate (Li2SO4) to produce desalted water after the reaction. The Acid chamber supplies deionized water (DI water) to produce sulfuric acid (H2SO4) after the reaction. The Base chamber supplies deionized water (DI water) to produce lithium hydroxide (LiOH) after the reaction.

[0047] The salt tank stores the produced desalinated water. The acid tank stores the produced sulfuric acid. The base tank stores the produced lithium hydroxide.

[0048] The production volume of the BPED (Bioelectrodialysis Device) is determined by the discharge flow rates of sulfuric acid and lithium hydroxide. The production volume can be determined by controlling the input flow rates of water (H2O) and lithium sulfate, the circulation flow rate, the circulation pressure, the current and voltage of the rectifier, and the management of the pH and conductivity within each chamber. The input flow rate of the solution, the current and voltage of the rectifier, the pH, the conductivity, the circulation flow rate, and the circulation pressure, which are the control and management variables, can be detected through internal sensors installed in each chamber. The control and management variables can be operational data that influence the concentration of the lithium hydroxide and sulfuric acid produced.

[0049] Figure 3 is a block diagram illustrating a preprocessing process of a Windows-based electrodialysis equipment abnormality detection method according to one embodiment of the present invention. The Windows-based electrodialysis equipment abnormality detection method can be performed by a Windows-based electrodialysis equipment abnormality detection device.

[0050] In FIG. 3, the method for detecting abnormalities in a Windows-based electrodialysis facility may include a preprocessing step (310), a data conversion step (320), and a learning, evaluation, and prediction step (330) using an artificial intelligence model.

[0051] In the preprocessing step (310), the window-based electrodialysis equipment abnormality detection device can select a control variable, select a management variable, and set an input window range and a prediction window range.

[0052] A Windows-based electrodialysis equipment abnormality detection device can select control and management variables for optimal automated operation of electrodialysis equipment. In other words, the Windows-based electrodialysis equipment abnormality detection device can select control and management variables comprising key factors affecting optimal automated operation of electrodialysis equipment.

[0053] For example, the control variables may include the input flow rate, circulation flow rate, and voltage of the rectifier of each solution in each stage of the electrodialysis equipment. The management variables may be changed by the control variables. The management variables may include the conductivity, pH, and substack voltage of the rectifier in each stage of the electrodialysis equipment.

[0054] A Windows-based electrodialysis equipment anomaly detection device monitors for outliers by analyzing control and management variables containing time-series data. When analyzing time-series data, the Windows-based electrodialysis equipment anomaly detection device can appropriately select the range or size of the input window and prediction window through appropriate window preprocessing.

[0055] An input window can be a subset of time-series continuous control variables. An input window can be defined as a unit of control variables that are input and processed by an AI model at a time. The input window range can refer to the size of one unit of control variables. For example, the input window range can be determined as a specific time range, such as 1 to 24 hours in time series.

[0056] A prediction window can be a subset of a continuous control variable. The prediction window can correspond to a unit of one control variable, which corresponds to a unit of one control variable. The prediction window range can also be determined to a specific time range, such as 1 hour to 24 hours. The prediction window range can be determined based on the input window range.

[0057] In the data conversion step (320), the window-based electrodialysis equipment abnormality detection device can clean continuous interval data within a set input window range. The continuous interval data may be a set of continuous time series data within the input window range. Cleaning of the continuous interval data may utilize data cleaning techniques such as removing missing values ​​and outliers and scaling.

[0058] A window-based electrodialysis equipment anomaly detection device can interpolate cleaned continuous interval data. The window-based electrodialysis equipment anomaly detection device can then transform the preprocessed time-series data into continuous interval-based data within the input window range. Conversion of continuous interval-based data can refer to the process of dividing continuous time-series data into intervals and converting them into new categories or values ​​for each interval. This transformation can be used to simplify data or restructure it for specific analysis purposes.

[0059] A continuous interval unit may correspond to an input window range. That is, continuous interval unit data may include data on a control variable included in a continuous interval corresponding to the input window range.

[0060] The Windows-based electrodialysis equipment anomaly detection device can set the input time interval for the AI ​​model. The model input time interval represents the time interval between consecutive data points provided to the model. For example, if data is measured daily, the input time interval could be one day.

[0061] In the learning, evaluation, and prediction step (330) using an artificial intelligence model, a Windows-based electrodialysis equipment abnormality detection device can train an artificial intelligence model based on continuous interval unit data. The Windows-based electrodialysis equipment abnormality detection device performs an evaluation on the trained artificial intelligence model. The Windows-based electrodialysis equipment abnormality detection device can predict management variables within the prediction window range using the trained artificial intelligence model. The Windows-based electrodialysis equipment abnormality detection device can compare the predicted management variables with actual management variables to determine the consistency of the prediction.

[0062] A Windows-based electrodialysis equipment abnormality detection device can detect BPED abnormalities by comparing predicted management variables with preset reference values.

[0063] FIGS. 4A to 4D are schematic diagrams showing an input window and a prediction window set according to one embodiment and a comparative example of the present invention.

[0064] The range occupied by the input window on the input variable (X) axis is the input window range. The range occupied by the prediction window on the target variable (Y) axis is the prediction window range.

[0065] In Figures 4a to 4d, the target time is the starting point of the prediction window range. In other words, the target time is the reference time for the prediction range according to the prediction window range, and the AI ​​model predicts the management variable within the prediction range set after the target time. The target time interval is the interval from the start point of the input window to the start point of the prediction window. The target time interval may be the interval between the start point of the input window and the target time.

[0066] An AI model predicts a target variable (Y) within a prediction window using input variables (X) within the input window range. If the target time interval matches the input window range, there is no gap time in predicting time series data. The input variables can be control variables, and the target variable can be a managed variable.

[0067] Figure 4a shows an input window and a prediction window set according to one embodiment.

[0068] In Fig. 4a, the sizes or intervals of the set input window range and the prediction window range can be identical. That is, the ratio of the input window range and the prediction window range can be 1 to 1. In the prediction window range, the target variable (Y) can be predicted through an artificial intelligence model trained with the input variable (X) in the input window range. The input window range and the target time interval can be matched.

[0069] Figure 4b shows an input window and a prediction window set according to the first comparative example. In the first comparative example, the size of the prediction window range is larger than the size of the input window range.

[0070] In Figure 4b, the input window range and the target time interval can be aligned. That is, there is no gap between the target time and the input window range. Therefore, real-time prediction of time series data can be possible.

[0071] However, the sizes of the input window range and the prediction window range are different. For example, the ratio of the input window range to the prediction window range can be 1:2. If the input window range and the prediction window range are not 1:1, the prediction agreement or predictive power may be poor. For example, if the input window range is 2 hours and the prediction window range is 4 hours, if you predict 2 hours based on the learned window within 2 hours (assuming an agreement of 90%) and then predict 2 hours again, the error will be multiplied, resulting in a low agreement (90% X 90% = 81%).

[0072] Figure 4c shows an input window and a prediction window set according to a second comparative example. In the second comparative example, the size of the prediction window range is smaller than the size of the input window range.

[0073] In Figure 4c, the ratio of the input window range to the prediction window range may be 2:1. If the input window range and the prediction window range are not 1:1, the prediction accuracy or predictive power may decrease. For example, if the input window range is 2 hours and the prediction window range is 1 hour, overfitting may occur, reducing the predictive power.

[0074] Figure 4d shows the input window and prediction window set according to the third comparative example. The third comparative example is a case where there is a gap between the target time and the input window range.

[0075] In Fig. 4d, the sizes or intervals of the set input window range and the prediction window range can be the same. That is, the ratio of the input window range and the prediction window range can be 1 to 1. In the prediction window range, the target variable (Y) can be predicted through an artificial intelligence model trained with the input variable (X) in the input window range.

[0076] At this time, a certain gap exists between the target time and the input window range. In other words, a certain amount of time gap exists in time-series data prediction. This gap hinders real-time prediction. Here, the learning window can refer to the set of input variables (Y) learned within the input window range.

[0077] Figure 5 is a diagram illustrating an example of predicting management variables according to each step of a method for detecting abnormalities in a window-based electrodialysis facility. Figure 5 illustrates an input window and a prediction window set according to one embodiment.

[0078] In Fig. 5, the window-based electrodialysis equipment abnormality detection device can set the input window range and the prediction window range to the same 2 hours in the preprocessing step (510) of time series data for learning an artificial intelligence model. The window-based electrodialysis equipment abnormality detection device can reflect the propagation time for the solution injected into each stage of the electrodialysis equipment to be ion-exchanged and move to the next stage when setting the input window range. The window-based electrodialysis equipment abnormality detection device can set the prediction window range that matches the input window range set in this way. For example, the window-based electrodialysis equipment abnormality detection device can set an input window range of 2 hours from 0 o'clock to 2 o'clock for control variables X1 to Xn.

[0079] A window-based electrodialysis equipment abnormality detection device can convert control variables (X) within an input window range into continuous interval unit data. The continuous interval unit can be a 2-hour unit corresponding to the input window range.

[0080] The Windows-based electrodialysis equipment abnormality detection device can train a deep learning model for time series processing based on continuous interval data in the artificial intelligence model learning step (520). The deep learning model for time series processing may include an informer model and a DLinear model. The deep learning model for time series processing may include at least one of the informer model and the DLinear model.

[0081] The Informer model is a deep learning-based predictive model for time series data, primarily targeting high-performance predictions for long time series data sequences. The DLinear model combines a time series decomposition method with linear layers. The DLinear model first creates a moving average and then removes it to decompose the data into trend and periodic data for learning. The DLinear model then applies a single linear layer to each component for training, and the two are combined to compute the final prediction.

[0082] In the step (530) of predicting a management variable, the window-based electrodialysis equipment abnormality detection device can predict a management variable (Y) within a prediction window range using a trained time-series processing deep learning model. For example, the time-series processing deep learning model can predict a management variable (Y) within a prediction window range set to 2 hours. The window-based electrodialysis equipment abnormality detection device can predict a management variable (Y) within a prediction window range between 2 o'clock and 4 o'clock. That is, the target time may be 2 o'clock and the target time interval may be 2 hours. Therefore, in one embodiment, the target time interval may match the input window range. That is, the window-based electrodialysis equipment abnormality detection device can predict a management variable in real time without a gap between the input window and the prediction window.

[0083] According to one embodiment, a window-based electrodialysis equipment abnormality detection device can set the ratio of the input window range to the prediction window range to 1:1. In addition, the window-based electrodialysis equipment abnormality detection device can match the target time interval to the input window range so that there is no gap between the input window and the prediction window. Accordingly, the window-based electrodialysis equipment abnormality detection device can improve the prediction consistency through real-time prediction of management variables.

[0084] A Windows-based electrodialysis equipment anomaly detection device can compare predicted management variables with actual management variables to calculate a match rate to improve prediction consistency. If the calculated match rate falls below a certain standard, the Windows-based electrodialysis equipment anomaly detection device can reset the input window range and the predicted window range.

[0085] The window-based electrodialysis equipment abnormality detection device can control the propagation time by adjusting the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate among the control variables when resetting the input window range and the predicted window range. The window-based electrodialysis equipment abnormality detection device can reset the input window range by reflecting the propagation time and reset the predicted window range based on this.

[0086] Fig. 6 is a flowchart of a method for detecting abnormalities in a Windows-based electrodialysis equipment according to one embodiment of the present invention. The method for detecting abnormalities in a Windows-based electrodialysis equipment of Fig. 6 can be performed through a Windows-based electrodialysis equipment abnormality detection device.

[0087] In Fig. 6, the window-based electrodialysis equipment abnormality detection device can set the input window range by reflecting the propagation time for the solution injected into each stage of the electrodialysis equipment to be ion-exchanged and then transferred to the next stage (step S100). In one embodiment, the window-based electrodialysis equipment abnormality detection device can control the propagation time by adjusting the first stage salt injection flow rate, the third stage acid injection flow rate, and the third stage base injection flow rate among the control variables. Since the input window range is determined by reflecting the propagation time, the input window range may also change if the propagation time changes.

[0088] A window-based electrodialysis equipment abnormality detection device can set a prediction window range for a management variable based on a set input window range (step S200). The window-based electrodialysis equipment abnormality detection device can provide feedback on the size and ratio of the window based on the alignment of the prediction window.

[0089] For example, a window-based electrodialysis equipment abnormality detection device can set the prediction window range so that the ratio of the input window range to the prediction window range is 1 to 1. That is, if the input window range is 2 hours, the window-based electrodialysis equipment abnormality detection device can also determine the prediction window range to be 2 hours.

[0090] In the case of a window-based electrodialysis equipment anomaly detection device, even if the ratio of the input window range and the prediction window range is 1:1, if the window size is too large, learning cannot proceed, so the window size can be appropriately determined based on the matching degree of the prediction window.

[0091] Since the prediction window range is determined based on the input window range, the prediction window range can also change depending on the propagation time. The window-based electrodialysis equipment anomaly detection device can eliminate the gap in the target time interval between the input window range and the prediction window range. In other words, the window-based electrodialysis equipment anomaly detection device can match the target time interval, which is the interval between the starting point of the input window and the starting point of the prediction window, to the input window range.

[0092] A window-based electrodialysis equipment anomaly detection device can convert time-series data of control variables into continuous interval unit data corresponding to the input window range (step S300). The continuous interval unit data is data defined as units of continuous intervals corresponding to the input window range. The continuous interval unit data can be acquired by processing the control variables within the input window range through preprocessing.

[0093] A window-based electrodialysis equipment abnormality detection device can learn an artificial intelligence model based on continuous interval data and predict management variables based on control variables through the learned artificial intelligence model (step S400). The artificial intelligence model may include a deep learning model for time series processing. The window-based electrodialysis equipment abnormality detection device can match the target time, which is the start point of the prediction window range, with the end point of the input window. In other words, the window-based electrodialysis equipment abnormality detection device can predict management variables within the prediction window range in real time based on the control variables within the input window range. In other words, the blank time between the input window range and the prediction window range can be eliminated.

[0094] In addition, the Windows-based electrodialysis equipment abnormality detection device can determine whether the electrodialysis equipment is abnormal by comparing the predicted management variable with a preset reference value (step S400). For example, the rectifier of each stage is composed of three sub-stacks. The stack voltage, which is the voltage of the sub-stack, corresponds to the management variable. The limit voltage of the stack voltage may be 90 V. The Windows-based electrodialysis equipment abnormality detection device can detect that an abnormality has occurred in the electrodialysis equipment if any one of the plurality of stack voltages is predicted to be 90 V or higher.

[0095] FIG. 7 is a block diagram of a window-based electrodialysis equipment abnormality detection device according to one embodiment of the present invention.

[0096] Referring to Fig. 7, a window-based electrodialysis equipment abnormality detection device (100) may include a preprocessing unit (110, 120), a data conversion unit (130), a prediction unit (140), and an abnormality detection unit (150). The preprocessing unit may include an input window setting unit (110) and a prediction window setting unit (120).

[0097] The preprocessing unit can perform preprocessing on control variables, which are input variables for a deep learning model for time-series processing. Preprocessing may include data cleaning, data interpolation, and scaling. The preprocessing unit can set input window ranges for control variables and prediction window ranges for management variables.

[0098] The input window setting unit (110) can set the input window range by reflecting the propagation time for the solution injected into each stage of the electrodialysis equipment to undergo ion exchange and move to the next stage. The input window setting unit (110) can control the propagation time by adjusting the first stage salt injection flow rate, the third stage acid injection flow rate, and the third stage base injection flow rate among the control variables.

[0099] The prediction window setting unit (120) can set the prediction window range based on the input window range. The prediction window setting unit (120) can set the prediction window range so that the ratio of the prediction window range and the input window range is 1 to 1.

[0100] The data conversion unit (130) can convert the time series data of the control variable into continuous interval unit data corresponding to the input window range. The data conversion unit (130) can perform data cleaning and data interpolation preprocessing on the continuous interval data within the input window range. The continuous interval data can be converted into continuous interval unit data after the preprocessing.

[0101] The prediction unit (140) can predict the management variable corresponding to the control variable through an artificial intelligence model learned based on continuous interval unit data. The prediction unit (140) can calculate the degree of matching of the learned artificial intelligence model. Based on the degree of matching calculated by the prediction unit (140), the preprocessing unit (110, 120) can reset the input window range and the prediction window range until the degree of matching becomes higher than a reference level. The preprocessing unit (110, 120) can reset the input window range through propagation time control and reset the prediction window range based on the input window range.

[0102] The prediction unit (140) can adjust the target time interval between the target time of the prediction window range including the management variable and the starting point of the input window range through an artificial intelligence model to match the input window range. In other words, the gap time between the input window and the prediction window can be eliminated and the management variable can be predicted in real time.

[0103] The abnormality detection unit (150) can detect abnormalities in the electrodialysis equipment by comparing predicted management variables with reference values. For example, the reference values ​​may include 90 V, which is the monitoring limit voltage of the stack voltage.

[0104] Figures 8a to 8g are drawings showing prediction results according to one embodiment and a comparative example of the present invention.

[0105] Fig. 8a is a diagram showing a conductivity prediction result according to one embodiment of the present invention. Fig. 8a is a graph showing the conductivity that appears when the ratio of the input window range to the prediction window range is set to 1:1.

[0106] FIGS. 8B and 8C are diagrams showing prediction results of stack voltages according to one embodiment of the present invention. FIGS. 8B and 8C are graphs showing stack voltages that appear when the ratio of the input window range to the prediction window range is set to 1:1.

[0107] In FIGS. 8a to 8c, it can be seen that the graphs of the actual values ​​(Real) of the conductivity and stack voltage for each solution and the graphs of the predicted values ​​(prediction) appear similar to each other.

[0108] Figures 8d and 8e are drawings showing prediction results according to the first comparative example. The first comparative example is a case where the size of the prediction window range is larger than the size of the input window (learning window).

[0109] In Fig. 8d, it can be seen that the predicted and actual result graphs showing lithium production, current efficiency, and first-stage acid purity are different from each other.

[0110] In Fig. 8e, it can be seen that there is a difference between the prediction and the ground truth results in the graphs for the purity of the first-stage Base product (LiOH), the purity of the second-stage Base product (LiOH), and the purity (S / Li ratio) of the third-stage Base product (LiOH).

[0111] Figure 8f is a diagram showing the prediction results according to the second comparative example. Figure 8f is a graph showing the conductivity for each solution in the first stage. The second comparative example is a case where the size of the prediction window range is smaller than the size of the input window (training window).

[0112] In Fig. 8f, it can be seen that the predicted and actual values ​​(groundtruth) in the graphs for the first-stage salt conductivity, the first-stage acid conductivity, and the first-stage base conductivity are not similar to each other. Although not shown, the solution conductivities in the second and third stages also differ in the predicted and actual values ​​in the second comparative example.

[0113] Figure 8g is a diagram showing prediction results according to another comparative example. The comparative example in Figure 8f shows a case where the ratio of the input window range to the prediction window range was set to 1:1, but each size was set excessively large. For example, Figure 8g shows the results when the input window size and the prediction window size were each set to 24 hours.

[0114] In Figure 8g, this comparative example demonstrates that learning virtually does not progress. Learning does not progress when the prediction window size and learning time are too large.

[0115] The first graph (LS1) is the loss calculated with evaluation data for each epoch, and the second graph (LS2) is the loss calculated with training data for each epoch.

[0116] The second graph (LS2) shows that the loss value is the same at 1.06 regardless of the epoch because learning was not done well.

[0117] Here, epoch represents the cycle in which the entire training dataset passes through the model once.

[0118] FIG. 9 is a drawing for explaining a computing device according to one embodiment of the present invention.

[0119] Referring to FIG. 9, a method and device for detecting abnormalities in a Windows-based electrodialysis facility according to embodiments can be implemented using a computing device (900).

[0120] The computing device (900) may include at least one of a processor (910), a memory (930), a user interface input device (940), a user interface output device (950), and a storage device (560) that communicate via a bus (920). The computing device (900) may also include a network interface (970) electrically connected to a network (90). The network interface (970) may transmit or receive signals to or from other entities via the network (90).

[0121] The processor (910) may be implemented in various types such as an MCU (Micro Controller Unit), an AP (Application Processor), a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an NPU (Neural Processing Unit), etc., and may be any semiconductor device that executes instructions stored in a memory (930) or a storage device (960). The processor (910) may be configured to implement the functions and methods described above with respect to FIGS. 1 to 8.

[0122] The memory (930) and storage device (960) may include various types of volatile or non-volatile storage media. For example, the memory may include read-only memory (ROM) (931) and random access memory (RAM) (932). In the present embodiment, the memory (930) may be located inside or outside the processor (910), and the memory (930) may be connected to the processor (910) via various known means.

[0123] In some embodiments, at least some components or functions of the Windows-based electrodialysis equipment abnormality detection method and device according to the embodiments may be implemented as a program or software running on a computing device (900), and the program or software may be stored on a computer-readable medium.

[0124] In some embodiments, at least some components or functions of the Windows-based electrodialysis equipment abnormality detection method and device according to the embodiments may be implemented using hardware or circuitry of the computing device (900), or may be implemented as separate hardware or circuitry that can be electrically connected to the computing device (900).

[0125] Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements made by a person of ordinary skill in the art to which the present invention pertains using the basic concept of the present invention defined in the following claims also fall within the scope of the present invention.

[0126] [Explanation of symbols]

[0127] 100: Windows-based electrodialysis equipment abnormality detection device

[0128] 110: Input window settings section

[0129] 120: Prediction window setting section

[0130] 130: Data conversion unit

[0131] 140: Prediction Department

[0132] 150: Anomaly detection unit

Claims

1. A step for setting an input window range for control variables of an electrodialysis facility; A step of converting the time series data of the above control variable into continuous interval unit data corresponding to the input window range; A step of predicting a management variable corresponding to the control variable through an artificial intelligence model learned based on the above continuous interval unit data; and A window-based electrodialysis equipment abnormality detection method comprising a step of detecting whether the electrodialysis equipment is abnormal by comparing the predicted management variable with a reference value.

2. A window-based electrodialysis equipment abnormality detection method in the first paragraph, wherein the control variables include the input flow rate of the solution of each stage of the electrodialysis equipment, the circulation flow rate, and the voltage of the rectifier, and the management variables include the conductivity of each stage and the stack voltage of the rectifier.

3. A method for detecting abnormalities in electrodialysis equipment based on a window, wherein the step of setting the input window range in the first paragraph includes the step of setting the input window range by reflecting the propagation time for the solution injected into each stage of the electrodialysis equipment to be ion-exchanged and then passed to the next stage.

4. In the third paragraph, the step of setting the input window range includes the step of controlling the propagation time by adjusting the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate among the control variables. A window-based electrodialysis equipment abnormality detection method.

5. In the fourth paragraph, the method further includes a step of setting a prediction window range for a management variable based on the set input window range, A method for detecting abnormalities in a window-based electrodialysis facility, wherein the step of setting the prediction window range includes the step of setting the prediction window range such that a ratio of the prediction window range and the input window range is 1 to 1.

6. A method for detecting abnormalities in a window-based electrodialysis facility, wherein the step of converting into continuous section unit data in the fifth paragraph includes the step of performing data cleaning and data interpolation preprocessing on continuous section data within the input window range.

7. A method for detecting abnormalities in electrodialysis equipment based on a window, wherein the step of predicting the management variable in the 6th paragraph includes the step of calculating the degree of matching of the learned artificial intelligence model and resetting the input window range and the prediction window range based on the calculated degree of matching.

8. In the 7th paragraph, the step of resetting the input window range and the prediction window range is a window-based electrodialysis equipment abnormality detection method including the step of controlling the propagation time by controlling the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate.

9. A method for detecting abnormalities in electrodialysis equipment based on a window, wherein the step of predicting the management variable in the 6th paragraph includes the step of matching a target time interval between a target time of the prediction window range and a start point of the input window range with the input window range.

10. A method for detecting abnormalities in a Windows-based electrodialysis facility, wherein the artificial intelligence model comprises a deep learning model for time series data processing, including an informer model and a DLinear model.

11. Preprocessing unit that sets the input window range for the control variables of the electrodialysis equipment; A data conversion unit that converts the time series data of the above control variable into continuous interval unit data corresponding to the input window range; A prediction unit that predicts a management variable corresponding to the control variable through an artificial intelligence model learned based on the above continuous interval unit data; and A window-based electrodialysis equipment abnormality detection device including an abnormality detection unit that detects whether the electrodialysis equipment is abnormal by comparing the predicted management variables with reference values.

12. A window-based electrodialysis equipment abnormality detection device in the 11th paragraph, wherein the control variables include the input flow rate of the solution of each stage of the electrodialysis equipment, the circulation flow rate, and the voltage of the rectifier, and the management variables include the conductivity of each stage and the stack voltage of the rectifier.

13. In the 11th paragraph, the preprocessing unit is a window-based electrodialysis equipment abnormality detection device including an input window setting unit that sets the input window range by reflecting the propagation time for the solution injected into each stage of the electrodialysis equipment to be ion-exchanged and passed to the next stage.

14. In the 13th paragraph, the input window setting unit is a window-based electrodialysis equipment abnormality detection device that controls the propagation time by adjusting the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate among the control variables.

15. A window-based electrodialysis equipment abnormality detection device in claim 14, wherein the preprocessing unit further includes a prediction window setting unit that sets the prediction window range based on the input window range so that the ratio of the prediction window range and the input window range is 1 to 1.

16. In the 15th paragraph, the data conversion unit is a window-based electrodialysis equipment abnormality detection device that performs data cleaning and data interpolation preprocessing for continuous section data within the input window range.

17. In paragraph 16, the prediction unit calculates the consistency of the learned artificial intelligence model, The above preprocessing unit is a window-based electrodialysis equipment abnormality detection device that resets the input window range and the prediction window range until the matching degree becomes equal to or higher than a reference level based on the matching degree.

18. In the 17th paragraph, the input window setting unit is a window-based electrodialysis equipment abnormality detection device that controls the propagation time by adjusting the first stage salt input flow rate, the third stage acid input flow rate, and the third stage base input flow rate, and resets the input window range by reflecting the propagation time.

19. A window-based electrodialysis equipment abnormality detection device in the 17th paragraph, wherein the prediction unit adjusts the target time interval between the target time of the prediction window range and the start point of the input window range to match the input window range.

20. In the 19th paragraph, the artificial intelligence model is a Windows-based electrodialysis equipment abnormality detection device including a deep learning model for time series data processing including an informer model and a DLinear model.

Images