Data visualization device, data visualization method, and program

Abstract

To assist in identifying a cause of an abnormality.SOLUTION: A data visualization system in accordance with an embodiment includes a storage unit that stores measured values concerning process variables of a batch process, statistical quantities calculated from the measured values according to a technique of multivariate statistical process control, and degrees of contribution of the process variables to the statistical quantities; a first display unit that displays a first screen image which selectively visualizes as a trend graph maximum values of the statistical quantities in one or more batch processes respectively in relation to the one or more batch processes selected by a user; a second display unit that displays a second screen image which selectively visualizes in predetermined order a total of degrees of contribution in the batch process(s) selected by the user through the first screen image; and a third display unit that displays a third screen image which visualizes as a trend graph measured values concerning process variables associated with the degrees of contribution selected by the user through the second screen image.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] This invention relates to a data visualization device, a data visualization method, and a program. [Background technology]

[0002] A method called Multi-Variate Statistical Process Control (MSPC) has been known for some time (for example, Patent Documents 1-4). In MSPC, an abnormality (or sign of an abnormality) in the process is often diagnosed when an indicator value, such as the Q value or Q statistic, exceeds a threshold. In such cases, for example, a person in charge of monitoring the process identifies the parameter that contributed to the abnormality from among the parameters such as process variables, and further narrows down those parameters to identify the cause of the abnormality. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] International Publication No. 2020 / 014881 [Patent Document 2] Japanese Patent Publication No. 2018-173948 [Patent Document 3] Japanese Patent Publication No. 2019-53537 [Patent Document 4] Japanese Patent Publication No. 2017-215908 [Overview of the project] [Problems that the invention aims to solve]

[0004] However, traditionally, identifying the cause of an anomaly required a great deal of effort and time. This is because, generally, the number of parameters, such as process variables, is often enormous, and even if an anomaly is diagnosed in a certain process, it is difficult to find and narrow down which parameters are contributing to that anomaly.

[0005] One embodiment of the present invention has been made in view of the above points and aims to support the identification of the cause of an anomaly. [Means for solving the problem]

[0006] To achieve the above objective, a data visualization device according to one embodiment includes: a storage unit configured to store measured values ​​relating to process variables of a batch process, statistical quantities calculated from the measured values ​​by a multivariate statistical process management method, and the contribution of the process variables to the statistical quantities; a first display unit configured to display a first screen that visualizes the maximum value of the statistical quantities in each of the one or more batch processes selected by the user as a trend graph; a second display unit configured to display a second screen that visualizes the sum of the contributions in the batch processes selected by the user on the first screen in a predetermined order; and a third display unit configured to display a third screen that visualizes the measured values ​​relating to the process variables corresponding to the sum of the contributions selected by the user on the second screen as a trend graph. [Effects of the Invention]

[0007] It can help identify the cause of the abnormality. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows an example of the overall configuration of the data visualization system according to this embodiment. [Figure 2] This figure shows an example of a screen transition. [Figure 3] This figure shows an example of the hardware configuration of the data visualization device according to this embodiment. [Figure 4] This figure shows an example of the functional configuration of the data visualization device according to this embodiment. [Figure 5] This figure shows an example of a data selection screen. [Figure 6] FIG. 1 is a diagram showing an example of a trend display screen (maximum Q value trend). [Figure 7] FIG. 2 is a diagram showing an example of a trend display screen (Q value trend within one batch). [Figure 8] FIG. 3 is a diagram showing an example of a trend display screen (Q value contribution degree within one batch). [Figure 9] FIG. 4 is a diagram showing an example of a trend display screen (measured value trend within one batch). DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, an embodiment of the present invention will be described. In this embodiment, a data visualization system 1 that can assist in identifying the cause of an abnormality by visualizing various data obtained from an abnormality diagnosis system that performs abnormality diagnosis by MSPC for the process of a batch plant (that is, a batch process) will be described. Note that targeting a batch plant is an example, and it is possible to target various devices, apparatuses, facilities, systems, robots, etc. whose processes are batch processes or processes similar thereto.

[0010] <Overall Configuration Example of Data Visualization System 1> An overall configuration example of the data visualization system 1 according to this embodiment is shown in FIG. 1. As shown in FIG. 1, the data visualization system 1 according to this embodiment includes a data visualization device 10, an abnormality diagnosis system 20, a control device 30, and a batch plant 40. The data visualization device 10 and the abnormality diagnosis system 20 are communicably connected via an arbitrary communication network. Similarly, the abnormality diagnosis system 20 and the control device 30 are connected via an arbitrary communication network, and the control device 30 and the batch plant 40 are connected via an arbitrary communication network.

[0011] The data visualization device 10 acquires various types of data (e.g., abnormal diagnosis results, Q values, measured values of process variables of the batch process, etc.) from an abnormal diagnosis system 20 that performs abnormal diagnosis by multivariate statistical process control (MSPC), and visualizes these various types of data. Hereinafter, the process variables of the batch process are also referred to as "diagnosis parameters" or simply "parameters". Note that what diagnosis parameters exist can vary depending on the process, and examples include temperature, pressure, flow rate, gas concentration, current, voltage, frequency, rotation speed, output, various control information, etc.

[0012] The abnormal diagnosis system 20 performs abnormal diagnosis of the process (batch process) of the batch plant 40 by multivariate statistical process control. More specifically, the abnormal diagnosis system 20 acquires measured values of various diagnosis parameters from the control device 30 for each measurement period, and calculates a statistic called the Q value from these measured values and a model created in advance (since this model is created from normal data of the batch process, it is also called a normal model). Then, the abnormal diagnosis system 20 diagnoses that an abnormality (or a sign of an abnormality) has occurred in the batch process when the maximum value of the Q values in the batch process (hereinafter, also referred to as the maximum Q value) exceeds a certain predetermined threshold value called the management limit. Abnormal diagnosis by MSPC is a known method, and for details thereof, for example, refer to Patent Documents 1 to 4 described above. Note that in this embodiment, instead of the Q value, the present invention is similarly applicable when using a statistic called the T 2 value. Hereinafter, as an example, the case where the statistic is the Q value will be described.

[0013] Note that the abnormal diagnosis system 20 may have a plurality of abnormal diagnosis units (or may be called an abnormal diagnosis engine) in which models and the like are different according to, for example, the type or kind of the batch process.

[0014] The control device 30 acquires measured values ​​of various process variables from the batch plant 40 and controls the batch plant 40 based on these measured values. Examples of the control device 30 include a PLC (Programmable Logic Controller).

[0015] A batch plant 40 is a collection of various machines or equipment used to carry out batch processes (hereinafter also simply referred to as "batch"). Specific examples of batch plants 40 include petrochemical plants, steel plants, food processing plants, and so on.

[0016] Note that the overall configuration of the data visualization system 1 shown in Figure 1 is just one example, and other configurations are also possible. For example, the anomaly diagnosis system 20 and the control device 30 may be integrated into a single unit.

[0017] <Screen transition example> Here, the data visualization device 10 displays two screens: a screen called the data selection screen and a screen called the trend display screen. The data selection screen is where the user selects the diagnostic result data to be visualized from among the diagnostic result data representing the diagnostic results of each batch. On the other hand, the trend display screen is where the maximum Q value, each Q value, and the measured values ​​of each diagnostic parameter for each batch corresponding to the diagnostic result data selected on the data selection screen are displayed as trends (time-series display). On the trend display screen, the user can select a desired batch from the trend of the maximum Q value and sequentially check the Q value trend, the contribution of the Q value, and the trend of the measured values ​​of each diagnostic parameter within that batch. The contribution of the Q value (hereinafter referred to as the Q value contribution) is expressed as the sum (accumulation) of the Q values ​​of each diagnostic parameter. It is also possible to express the Q value contribution within a batch as the ratio when the maximum value of the total value of each diagnostic parameter within a batch is set to 100.

[0018] Figure 2 shows an example of screen transitions in the data visualization device 10. As shown in Figure 2, the data visualization device 10 first displays a data selection screen 1000. When the user selects one or more diagnostic result data (in particular, a large number of diagnostic result data within a certain period) on this data selection screen 1000, the data visualization device 10 displays a trend display screen 2100 that visualizes the trend of the maximum Q value for each batch corresponding to those diagnostic result data. Next, when the user selects a batch on this trend display screen 2100, the data visualization device 10 displays a trend display screen 2200 that visualizes the trend of the Q value within that batch. Next, when the user performs an operation on this trend display screen 2200 to display the Q value contribution, the data visualization device 10 displays a trend display screen 2300 that visualizes the Q value contribution within that batch (more precisely, the sum of the Q value contributions of the same diagnostic parameter within that batch) in descending order. Then, when a user selects a Q-value contribution on this trend display screen 2400, the data visualization device 10 displays a trend display screen 2400 that visualizes the trend of the measured values ​​of the diagnostic parameters corresponding to that Q-value contribution in that batch.

[0019] In this way, users can select a desired batch from the maximum Q-value trend of each batch corresponding to a large amount of diagnostic result data within a certain period, and then select a desired Q-value contribution from the Q-value contributions of that batch, thereby displaying the measured value trend of the diagnostic parameter corresponding to that Q-value contribution. This makes it possible, for example, to easily identify the diagnostic parameter causing the abnormality or signs of abnormality and its measured value trend from the batch in which the abnormality (or signs of abnormality) occurred, enabling rapid identification of the cause of the abnormality.

[0020] <Example hardware configuration of data visualization device 10> Figure 3 shows an example of the hardware configuration of the data visualization device 10 according to this embodiment. As shown in Figure 3, the data visualization device 10 according to this embodiment includes an input device 101, a display device 102, an external I / F 103, a communication I / F 104, a RAM (Random Access Memory) 105, a ROM (Read Only Memory) 106, an auxiliary storage device 107, and a processor 108. Each of these hardware components is connected to each other via a bus 109 so as to be able to communicate.

[0021] The input device 101 is, for example, a keyboard, mouse, touch panel, or physical buttons. The display device 102 is, for example, a display or display panel.

[0022] External I / F 103 is an interface with external devices such as recording media 103a. Examples of recording media 103a include CD (Compact Disc), DVD (Digital Versatile Disk), SD memory card (Secure Digital memory card), and USB (Universal Serial Bus) memory card.

[0023] The communication interface 104 is an interface for connecting the data visualization device 10 to a communication network. The RAM 105 is a volatile semiconductor memory (storage device) that temporarily holds programs and data. The ROM 106 is a non-volatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off. The auxiliary storage device 107 is a non-volatile storage device such as an HDD (Hard Disk Drive) or SSD (Solid State Drive), where programs and data are stored. The processor 108 is a type of arithmetic unit such as a CPU (Central Processing Unit).

[0024] Note that the hardware configuration shown in Figure 3 is just one example, and the data visualization device 10 may have other hardware configurations. For example, the data visualization device 10 may have multiple auxiliary storage devices 107 and multiple processors 108, or it may have various hardware other than the hardware shown in the figure.

[0025] <Example of functional configuration of data visualization device 10> Figure 4 shows an example of the functional configuration of the data visualization device 10 according to this embodiment. As shown in Figure 4, the data visualization device 10 according to this embodiment has a user interface unit 201. The user interface unit 201 is realized, for example, by processing that one or more programs installed on the data visualization device 10 cause the processor 108 to execute. The data visualization device 10 according to this embodiment also has a diagnostic results DB 202, a measurement value DB 203, and an analysis results DB 204. Each of these DBs (databases) is realized, for example, by an auxiliary storage device 107. At least one of these DBs may be realized, for example, by a storage device (database server, etc.) connected to the data visualization device 10 via a communication network.

[0026] The user interface unit 201 displays various screens (such as the data selection screen and trend display screen mentioned above) on the display device 102 and accepts various user operations (various input operations by the input device 101) on these screens.

[0027] The diagnostic results DB202 stores diagnostic result data. Each diagnostic result data represents the diagnostic result of each batch and includes items such as "Batch ID," "Measurement Start Date and Time," "Diagnostic Start Date and Time," "Equipment," "Diagnostic System," "Diagnostic Unit," "Quality," "Comments," and "Metadata." The "Batch ID" is set to an ID that identifies the diagnostic result data or the batch corresponding to that diagnostic result data (e.g., the batch index). The "Measurement Start Date and Time" is set to the date and time when the measurement of the diagnostic parameters of that batch started. The "Diagnostic Start Date and Time" is set to the date and time when the abnormal diagnosis of that batch started. The "Equipment" is set to an ID or name that identifies the equipment corresponding to that batch (i.e., the batch plant 40). The "Diagnostic System" is set to an ID or name that identifies the abnormal diagnosis system 20 that performed the abnormal diagnosis of that batch. The "Diagnostic Unit" is set to an ID or name that identifies the diagnostic unit of the abnormal diagnosis system 20 that performed the abnormal diagnosis of that batch. The "Quality" is set to information that represents the diagnostic result of that batch (e.g., "Normal" or "Abnormal" indicating the quality of that batch). The "Comment" field contains an arbitrary comment (string) set by the person responsible for monitoring the batch. The "Metadata" field contains the diagnostic parameters for the batch.

[0028] The measurement data DB203 stores measurement data for each batch, representing the measured values ​​of the diagnostic parameters for that batch. For example, the measurement data might use i as the index of the diagnostic parameter, j as the index of the batch, and x as the measured value of diagnostic parameter i in batch j. ij If t is the relative time within the batch, then {x ij This can be expressed as (t)|t=1,···,T} (where t=1 is the batch start time and t=T is the batch end time), etc.

[0029] The analysis result DB204 stores analysis result data representing the results of analyzing measurement data (such as Q value, maximum Q value, Q value contribution degree, total within batch of Q value contribution degree, etc.) when the abnormality diagnosis system 20 performs an abnormality diagnosis. For the Q value, for example, if the batch index is j and the relative time within the batch is t, it is expressed as {Q j (t)|t = 1, ···, T} (where t = 1 is the batch start time and t = T is the batch end time), etc. Also, the maximum Q value is expressed as Q j = max t {Q j (t)}. Also, if the number of diagnosis parameters is I, Q j (t) is the sum of the squared prediction errors of each diagnosis parameter Q j (t)=Q 1j (t)+···+Q Ij (t). Therefore, the Q value contribution degree of diagnosis parameter i is expressed as Q ij (t). Furthermore, the total within batch of the Q value contribution degree of diagnosis parameter i is expressed as Q ij (1)+···+Q ij (T). Hereinafter, the total within batch of the Q value contribution degree is also referred to as the "total within batch of Q value contribution degree".

[0030] The above diagnosis result data, measurement data, and analysis result data are obtained from the abnormality diagnosis system 20 by, for example, FTP (File Transfer Protocol) and stored in the diagnosis result DB202, measurement value DB203, and analysis result DB204, respectively. However, using FTP is just an example, and the acquisition method of each data is not limited to this. For example, at least some of the diagnosis result data, measurement data, and analysis result data may be stored in a recording medium 103a or the like, and then the data may be obtained from this recording medium 103a.

[0031] <Data Selection Screen 1000> An example of the data selection screen 1000 is shown in Figure 5. The data selection screen 1000 shown in Figure 5 is a screen for the user to select the diagnostic result data to be visualized from the diagnostic result data representing the diagnostic results of each batch. This data selection screen 1000 is displayed by the user interface unit 201, and various user operations on this data selection screen 1000 are accepted by the user interface unit 201.

[0032] As shown in Figure 5, the data selection screen 1000 includes a diagnostic result display area 1001 that displays a list of diagnostic result data stored in the diagnostic result DB 202. The user can select the desired diagnostic result data from this diagnostic result display area 1001. At this time, the user may select the desired diagnostic result data using the checkbox 1002, or select all diagnostic result data using the select all button 1003. Note that one diagnostic result data (in the example shown in Figure 5, one row in the diagnostic result display area 1001 represents one diagnostic result data) corresponds to one batch.

[0033] Furthermore, by pressing the search button 1004, the user can set desired search conditions (for example, measurement start date and time, or period relative to the diagnosis start date and time) for the diagnostic result data displayed in the diagnostic result display field 1001 and perform a search. Therefore, the user may search the diagnostic result data displayed in the diagnostic result display field 1001 using the desired search conditions and then select the diagnostic result data from the search results using the checkbox 1002 or the select all button 1003.

[0034] Furthermore, the user can deselect all currently selected diagnostic result data by pressing the deselect button 1005.

[0035] When one or more diagnostic result data (in particular, a large number of diagnostic result data within a certain period) are selected and the user presses the trend button 1006, the user interface unit 201 of the data visualization device 10 displays the trend display screen 2100. Hereinafter, assuming that a certain batch plant 40 diagnostic result data for a certain period is selected on the data selection screen 1000, these selected diagnostic result data will be referred to as "selected diagnostic result data".

[0036] <Trend display screen 2100> An example of the trend display screen 2100 is shown in Figure 6. The trend display screen 2100 shown in Figure 6 visualizes the trend of the maximum Q value for each batch corresponding to each selected diagnostic result data. This trend display screen 2100 is displayed by the user interface unit 201, and various user operations on this trend display screen 2100 are accepted by the user interface unit 201.

[0037] As shown in Figure 6, the trend display screen 2100 includes a maximum Q value trend display area 2101 that displays a graph representing the trend of the maximum Q value of the batch corresponding to the selected diagnostic result data. In this maximum Q value trend display area 2101, the maximum Q value of each batch is displayed as a trend (time series display) with time on the horizontal axis and the maximum Q value on the vertical axis. The maximum Q value of each batch is stored in the analysis results DB 204.

[0038] In the example shown in Figure 6, the maximum Q value begins to increase around April 18, and by April 28, the maximum Q value exceeds 6000. This indicates that abnormal signs begin to appear in the batch process around April 18, and abnormalities are occurring around April 28.

[0039] Therefore, in order to identify the cause of the above-mentioned abnormalities and abnormal signs, the user selects, for example, a batch from around the time when the maximum Q value began to increase. For example, suppose the user presses the button indicated by symbol 2102 (specifically Apr18). As a result, the batch corresponding to the location pressed by the user (i.e., the Apr18 batch) is selected, and the data visualization device 10 displays a trend display screen 2200 in which the Q value trend within that batch is visualized by the user interface unit 201. Hereafter, assuming that the Apr18 batch is selected in the maximum Q value trend display field 2101, this selected batch will be referred to as the "selected batch".

[0040] <Trend display screen 2200> An example of the trend display screen 2200 is shown in Figure 7. The trend display screen 2200 shown in Figure 7 visualizes the trend of Q values ​​within the selected batch. This trend display screen 2200 is displayed by the user interface unit 201, and various user operations on this trend display screen 2200 are accepted by the user interface unit 201.

[0041] As shown in Figure 7, the trend display screen 2200 includes a batch-specific Q-value trend display area 2201, which displays a graph representing the trend of Q-values ​​within the selected batch. In this batch-specific Q-value trend display area 2201, the Q-values ​​within the selected batch are displayed as a trend (time-series display) with time on the horizontal axis and Q-values ​​on the vertical axis. Note that each Q-value in each batch is stored in the analysis results DB 204.

[0042] Furthermore, the trend display screen 2200 shown in Figure 7 includes a diagnostic parameter display area 2202 where a legend of colors or patterns representing the Q-value contribution of each diagnostic parameter is displayed. In the Q-value trend display area 2201 within a batch, each Q-value of the Q-value trend is displayed with its constituent Q-value contributions distinguished by color or pattern.

[0043] This allows the user to see the trend between each Q value within the selected batch and the contribution of each Q value that makes up those Q values. The maximum Q value trend display area 2101 also displays a line 2103 representing the selected batch.

[0044] On the other hand, the trend display of the Q value and the contribution of each Q value that constitutes it makes it difficult to determine which diagnostic parameter is contributing to the abnormality or abnormality. Therefore, it is necessary to check the magnitude of each Q value contribution within the selected batch. To this end, the user presses the Q value contribution batch total display button 2203 to display the total Q value contribution within the selected batch in descending order. As a result, the data visualization device 10 displays a trend display screen 2300 that visualizes the total Q value contribution within the selected batch in descending order.

[0045] <Trend display screen 2300> An example of the trend display screen 2300 is shown in Figure 8. The trend display screen 2300 shown in Figure 8 visualizes the total Q value contribution within a selected batch in descending order. This trend display screen 2300 is displayed by the user interface unit 201, and various user operations on this trend display screen 2300 are accepted by the user interface unit 201.

[0046] As shown in Figure 8, the trend display screen 2300 displays a pop-up window 2301 showing the Q-value contribution within a batch in descending order. However, displaying in descending order is just one example and is not the only option.

[0047] In the batch-specific Q-value contribution display window 2301 shown in Figure 8, the total Q-value contributions within each batch are displayed in the following order: total Q-value contribution of diagnostic parameter "Parameter E", total Q-value contribution of diagnostic parameter "Parameter G", total Q-value contribution of diagnostic parameter "Parameter B", total Q-value contribution of diagnostic parameter "Parameter A", and so on. This allows the user to easily confirm whether a diagnostic parameter contributes to an abnormality or an abnormality sign within the selected batch. The total Q-value contribution for each batch is stored in the analysis results DB 204.

[0048] In general, diagnostic parameters with a large total Q-value contribution within a batch are likely to be the cause of the abnormality or abnormality in that batch. Therefore, the user selects the diagnostic parameters that are displayed relatively high in the Q-value contribution display window 2301 within a batch as the diagnostic parameters that are considered to be the cause of the abnormality or abnormality.

[0049] For example, if the user presses the area indicated by symbol 2302 (specifically, the total Q-value contribution of the diagnostic parameter "parameter G" within a batch), the diagnostic parameter corresponding to the area indicated by symbol 2302 (i.e., "parameter G") is selected. As a result, the data visualization device 10 displays a trend display screen 2400 in which the measured value trend of that diagnostic parameter is visualized by the user interface unit 201. Hereafter, assuming that the diagnostic parameter "parameter G" has been selected in the batch Q-value contribution display window 2301, this diagnostic parameter will be referred to as the "selected diagnostic parameter".

[0050] To close the batch Q-value contribution display window 2301, the user simply needs to press the close button 2303.

[0051] <Trend display screen 2400> An example of the trend display screen 2400 is shown in Figure 9. The trend display screen 2400 shown in Figure 9 is a screen that visualizes the measured value trend of the selected diagnostic parameter. This trend display screen 2400 is displayed by the user interface unit 201, and various user operations on this trend display screen 2400 are accepted by the user interface unit 201.

[0052] As shown in Figure 9, the trend display screen 2400 includes a batch-specific measurement value trend display area 2401, which displays a graph representing the trend of the measured values ​​of the selected diagnostic parameters in the selected batch. In this batch-specific measurement value trend display area 2401, the measured values ​​of the selected diagnostic parameters are displayed as a solid line trend (time series display) with the horizontal axis representing time and the vertical axis representing the measured value, and the upper and lower limits of the normal range are displayed as dashed lines. Here, the upper and lower limits of the normal range can be set in various ways, but for example, it is conceivable to set the normal range to [-σ,σ], [-2σ,2σ], [-3σ,3σ], etc., with σ being the standard deviation of the measured value under normal conditions for that diagnostic parameter. The measured values ​​of each diagnostic parameter are stored in the measurement value DB 203.

[0053] This allows users to see the trend of the measured values ​​of selected diagnostic parameters (for example, diagnostic parameters with a relatively large total Q-value contribution within a batch) and the upper and lower limits of their normal range. Therefore, users can find out, for example, whether the measured values ​​of selected diagnostic parameters are within the normal range, and whether the trend is upward or downward. However, displaying the upper and lower limits of the normal range is just one example and is not limited to this; for example, only the upper or lower limit of the normal range may be displayed.

[0054] If the user wishes to display the measurement trend of a different diagnostic parameter, they can simply select that parameter in the batch-specific Q-value contribution display window 2301.

[0055] <Summary> As described above, the data visualization device 10 according to this embodiment can visualize various data (diagnosis result data, measured value data, analysis result data) acquired from the abnormality diagnosis system 20 that performs abnormality diagnosis of batch processes by multivariate statistical process management. At this time, the data visualization device 10 according to this embodiment can sequentially display the following using the diagnosis result data selected on the data selection screen 1000: a trend display screen 2100 that visualizes the maximum Q value trend of each batch, a trend display screen 2200 that visualizes the Q value trend within the selected batch, a trend display screen 2300 that visualizes the total Q value contribution within the selected batch, and a trend display screen 2400 that visualizes the measured value trend of the selected diagnostic parameters in the selected batch. Furthermore, by performing three operations in sequence—selecting a batch, displaying the total Q-value contribution within the batch, and selecting diagnostic parameters—the user can transition sequentially from the trend display screen 2100 to the trend display screen 2200, from the trend display screen 2200 to the trend display screen 2300, and from the trend display screen 2300 to the trend display screen 2400. In addition, when selecting a batch, the user can view the data not as points (i.e., individual batches) but as a line (i.e., the trend of the maximum Q-value of each batch) to select the desired batch. As a result, it becomes possible to select an appropriate batch considering the data characteristics, and to exclude batches from selection that contain negligible anomalies.

[0056] Therefore, by performing just three operations in sequence, users can determine which diagnostic parameter is causing the anomaly or abnormality in a batch, and identify the true cause of the anomaly from the measured value trend of that diagnostic parameter.

[0057] For example, suppose that batch plant 40 consists of gas piping and a motor that adjusts the flow rate of that gas. In this case, suppose that the total Q-value contribution within the batch is relatively high for the diagnostic parameters "piping pressure" and "motor rotation speed," and that the measured value of "piping pressure" is below the range of values ​​that can be taken under normal conditions, while the measured value of "motor rotation speed" is above the range of values ​​that can be taken under normal conditions. In such a case, the user can infer that the cause of the abnormality is a gas leak from the piping, given that the pressure has not risen compared to normal conditions, but the motor rotation speed has risen compared to normal conditions. Therefore, the user can identify the gas leak in the piping as the cause of the abnormality and implement preventive maintenance measures such as inspection, repair, or replacement of the piping.

[0058] The present invention is not limited to the embodiments specifically disclosed above, and various modifications, changes, and combinations with known technologies are possible without departing from the scope of the claims. [Explanation of symbols]

[0059] 1. Data Visualization System 10 Data Visualization Devices 20 Anomaly Diagnosis System 30 Control device 40 batch plant 101 Input Device 102 Display device 103 External I / F 103a Recording medium 104 Communication I / F 105 RAM 106 ROM 107 Auxiliary storage 108 processors 109 Bus 201 User Interface Section 202 Diagnostic Results Database 203 Measurement Value Database 204 Analysis result DB

Claims

1. A storage unit configured to store measured values ​​relating to process variables of a batch process, statistics calculated from the measured values ​​using a multivariate statistical process control method, the contribution of the process variables to the statistics, and information indicating whether the batch process is normal or abnormal. A first display unit is configured to display a first screen that visualizes, as a trend graph, the maximum value of the statistical quantity for each of the multiple batch processes selected by the user within a predetermined period of time, and A second display unit is configured to display a second screen that visualizes the total contribution of the batch process selected by the user on the first screen in a selectable order, A third display unit is configured to display a third screen that visualizes the measured values ​​of the process variable corresponding to the sum of the contributions selected by the user on the second screen as a trend graph, A data visualization device having the following features.

2. The system further includes a data selection screen display unit configured to display a data selection screen that visualizes the plurality of batch processes in a list for the user to select from, The first display unit is, The data visualization device according to claim 1, configured to display the first screen with respect to the plurality of batch processes selected by the user on the data selection screen.

3. The second screen described above is, The data visualization device according to claim 1 or 2, wherein the first screen visualizes the total contributions of the batch processes selected by the user in descending order for selection.

4. The third screen is, The data visualization device according to any one of claims 1 to 3, wherein the second screen visualizes the measured values ​​for the process variable corresponding to the sum of the contributions selected by the user, and at least one of the upper and lower limits of the normal range for the process variable, as a trend graph.

5. A storage procedure for storing in a storage unit measured values ​​relating to process variables of a batch process, statistics calculated from the measured values ​​using a multivariate statistical process control method, the contribution of the process variables to the statistics, and information indicating whether the batch process is normal or abnormal. A first display procedure that displays a first screen that visualizes, as a trend graph, the maximum value of the statistical quantity for each of the multiple batch processes selected by the user for multiple abnormal batch processes within a predetermined period, and A second display procedure, which involves displaying a second screen that visualizes the total contribution of the batch process selected by the user on the first screen in a predetermined order, and A third display procedure, which involves displaying a third screen that visualizes the measured values ​​for the process variable corresponding to the sum of the contributions selected by the user on the second screen as a trend graph, A data visualization method performed by a computer.

6. A storage procedure for storing in a storage unit measured values ​​relating to process variables of a batch process, statistics calculated from the measured values ​​using a multivariate statistical process control method, the contribution of the process variables to the statistics, and information indicating whether the batch process is normal or abnormal. A first display procedure that displays a first screen that visualizes, as a trend graph, the maximum value of the statistical quantity for each of the multiple batch processes selected by the user for multiple abnormal batch processes within a predetermined period, and A second display procedure, which involves displaying a second screen that visualizes the total contribution of the batch process selected by the user on the first screen in a predetermined order, and A third display procedure, which involves displaying a third screen that visualizes the measured values ​​for the process variable corresponding to the sum of the contributions selected by the user on the second screen as a trend graph, A program that causes a computer to execute something.

Images