JP2002358120A - Batch plant operation managing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate the validity or invalidity of a plant state in a real time when each event to be evaluated of a batch operation is generated. SOLUTION: When each operation event to be evaluated of a batch operation is generated, a set value, an instructed value, and associated manipulated variables are extracted for each measuring unit to be monitored, and a deviation from the mean value of the past batch operation is calculated, and the normalized deviations of all the measuring units are averaged, and the validity or invalidity of a process state when the operation event is generated is judged based on the deviation. Thus, it is possible to quickly facilitate countermeasures when each event is generated based on the arithmetic operations in a real time.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a batch plant operation management device, and more particularly to a plant operation state monitoring system for a batch plant, which controls the operation of the batch plant while evaluating the reproducibility of the operation. The present invention relates to a batch plant operation management device.

[0002]

2. Description of the Related Art In recent years, there is an example in which a database for a batch plant is provided in an operation record management function included in an operation control system of a plant. In such a system, operation performance data for each product brand can be systematically organized for each batch, and based on this organized data, the reproducibility of batch operation is examined, and subsequent operation conditions can be set similarly. Available. However, such a method is a follow-up analysis to the last after the batch operation is completed, and there is no example in which the reproducibility of the batch operation during the operation is examined.

[0003]

As described above, in the performance management function for the conventional batch plant, the function is specialized to the performance data accumulation and extraction function. Therefore, it is necessary to extract past results for the same brand and compare it with the current batch in order to examine the inter-batch fluctuations related to the process state quantities such as temperature and pressure and liquid level during the process operation such as charging and reaction. There was a problem that it was difficult to quickly respond to batch operation troubles this time.

[0004] In view of the above, the present invention is to perform a batch operation reproducibility check in real time instead of following up, and to catch a change in the plant state at an early stage.
It is an object of the present invention to provide a batch plant operation management device capable of promptly responding to the state change.

[0005]

SUMMARY OF THE INVENTION The present inventors have found that abnormalities or fluctuations in the quality or yield in a batch plant are caused not only by the manufacturing recipe, but also by fluctuations between batches in other process conditions. I found that there were many,
The present invention has been reached.

[0006] Based on the above knowledge, the batch plant operation management device of the present invention is a batch plant operation management device that manages the operation of a batch plant while monitoring the operation of the batch plant. A storage device for storing process data including at least one of a set value, an instruction value, and an associated operation amount of a plurality of instruments for each instrument and for each event occurrence time;
Calculate the deviation of each process data of this batch from the average value of the corresponding process data obtained in the past predetermined number of batch operations, normalize the deviation according to the corresponding instrument range, and A calculating device for calculating the average value of the deviation values of the respective instruments for each occurrence of the event, and determining whether the overall plant condition is good or bad at each occurrence time of the event based on the average value of the deviation values. Means.

In the batch plant operation management device of the present invention, the quality of the plant state of the current batch can be determined based on the average value of the deviation values of the instruments at each time of occurrence of each event. It can be modified every time an event occurs.

[0008] In a preferred batch plant operation management device of the present invention, one or more monitored instruments whose normalized deviation values are relatively large are displayed. This display may be performed when the determination by the determination means is negative. In this case, it is easy to grasp the measures to be taken for improving the plant condition.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION In a batch plant operation management apparatus according to a preferred embodiment of the present invention, process data at the start of each process operation, such as the start of charging, the start of heating, and the start of a reaction, are stored in the past multiple batches of the same brand. The degree of coincidence compared to the average value of the operation results for each minute is indicated by the index of "reproducibility", and the monitored instruments are displayed in a list in descending order of deviation, and the cause of the decrease in reproducibility is judged in real time It can be so. As a result, if the reproducibility is low, it is possible to easily grasp the necessary measures to increase the low reproducibility.

Hereinafter, the present invention will be described in more detail based on an embodiment of the present invention. FIG. 1 shows a configuration of a batch plant operation management device of the present invention. Main operation management device 10
Is a data collection unit 11, an event management unit 12, a statistical processing unit 13, a performance comparison unit 14, an information display unit 15, an evaluation target event and a monitoring target instrument list storage device 16, which are respectively constructed on a system computer. Results database 17
Consists of The operation management device 10 cooperates with a batch plant control device (not shown) in cooperation with a batch plant 18.
Operation, operation management, collection of data necessary for them, and evaluation of plant condition.

The evaluation target event and monitored instrument list storage device 16 stores a list of evaluation target events required for analysis and a list of monitoring target instruments related thereto when the batch plant 18 is operated. Data collection unit 11
Collects, in real time, process state quantities such as temperature, pressure, and liquid level during the operation of the batch plant from the instrumentation system of the batch plant control device, and provides this to the event management unit 12.

The event management unit 12 refers to the list of events to be evaluated and the list of instruments to be monitored 16 based on the data collected in real time by the data collection unit 11, and sets the event occurrence notification flag for the predefined operation to be evaluated. By examining the on / off state, an operation event to be evaluated that has occurred in the plant is identified. At that time, the data collection unit 1
The process data related to each event acquired from No. 1 is accumulated in the performance database 17 as data at the time of occurrence of the target operation event.

FIG. 2 is a flow chart illustrating events during the operation of a certain batch plant. This batch plant operation consists of nitrogen replacement, charging, heating, reaction, cooling, transfer, receiving,
And, having an operation event consisting of decompression, the preparation to be evaluated,
The heating, reaction, cooling, and transfer operations are registered in the evaluation target event and monitoring target instrument list 16. Here, for example, at the start of charging, a thermometer or the like that measures the charging start temperature is an instrument to be monitored, and at the time of starting temperature rising, the temperature rising start temperature, jacket temperature, steam pressure, etc. The instrument to be measured is the instrument to be monitored. For each operation event, the set temperature of the instrument, the indicated temperature, and the operation amount related to the indication of the instrument are extracted by the event management unit 12 and accumulated in the performance database 17.

The statistical processing unit 13 classifies the instruments in advance according to the device, and calculates one or more instruments to be monitored.
The set value, the instruction value, and the operation amount are input from the event management unit 12 at the time of occurrence of each event to be evaluated. Next, the set value, the indicated value,
Then, for the operation amount, the average value and the standard deviation value for the past n batches of the brand which is the same as the brand manufactured in the current batch are calculated. Here, the default of n is, for example, 50.
Also, when starting up a system for producing a new product, the same value is written as the value of n batches when the result of the first batch is written. If the deviation of the input actual value deviates from three times the standard deviation at that time, the data is excluded and the previous actual value is written, thereby eliminating the exceptional data and real-time. Calculate the mean and standard deviation. At this time, the fact that the actual value has deviated from the standard deviation is recorded in the history. This deviation history can be used to investigate the deviation factors later, consider countermeasures, deviation frequency,
Used when performing trend management.

Here, x _new is the actual value of the current batch, X
n batches of the average value of the _old until the last batch, when the oldest actual value at the time of calculating an average value of the x ₁ up to the last batch, the average value X _{new new} calculated in this batch, X _{new new} = X _old + (x _new -x ₁ ) / n. Also, if ρ _old is the standard deviation up to the previous time, the standard deviation ρ _new calculated this time is

(Equation 1) Is shown.

The performance comparison unit 14 calculates the deviation of the collected current performance value from the previous average value, converts it into a percentage of the full range that can be actually indicated by each instrument (normalized), and accumulates it in the database. Then, the percentage conversion deviation of each instrument monitored by each process operation is 100%.
Is the reproducibility of the instrument.
Further, a value obtained by subtracting the average value of the percentage deviations of all the instruments to be monitored in each process operation from 100% is defined as the reproducibility in the operation event. The reproducibility is stored in the performance database 17. By defining the reproducibility in this way, when the percentage conversion deviation is 0, the reproducibility is the highest 100%, and the percentage conversion deviation is 100%.
In the case of, the reproducibility is the lowest 0.

That is, the reproducibility R for each operation event is represented by the following formula: R = 100.0− (Σα _i * deviation (%)) / number of target instruments. Here, α _i is a weight coefficient for the reproducibility calculation of the i-th instrument, and the default value is 1.0.
The total reproducibility Rt for evaluating the reproducibility of one batch plant operation as a whole is represented by Rt = (＝ R) / m where m is the number of operations to be evaluated.

The information display section 15 displays a bar graph of the reproducibility of the current batch at the start of each process operation, and displays a list of instruments to be monitored in descending order of deviation. In addition, a trend graph of reproducibility for the past 50 batches is also displayed.

FIG. 2 shows an example of a display on the monitoring screen displayed by the information display unit 15, wherein the name of the device used is R10.
1, a batch number is xxx, and a brand name is AAAAA in one batch operation. On the left end, the names of the operation to be evaluated are shown as start of charging, start of temperature rise, start of reaction, start of recovery, and start of transfer. For each operation to be evaluated, a graph showing the transition of the reproducibility for the past 50 batches, the reproducibility for each operation of the current batch, and the worst monitored instrument of the worst 10 which was determined to have the lowest reproducibility in the current batch Is shown, along with its recall. Furthermore, the total reproducibility Rt
Is shown in a graph showing the progress of 50 batches. These displays are sequentially displayed as the process proceeds.

During the operation of the present batch, for example, at the start of the reaction, more precisely, when it is determined that the reproducibility has decreased to a predetermined value or less prior to the start of the reaction, or from the average value of the reproducibility. If it is determined that the reproducibility has decreased over a predetermined width, the reproducibility at that time is displayed in a list with the worst 10 instrument names together with the reproducibility. By referring to the target instrument and the reproducibility displayed in the list, it is possible to easily determine a necessary measure for improving the reproducibility of the process state.
After taking necessary measures, the quality of reproducibility is determined again, and when it is determined to be good, the reaction is actually started. Also,
By checking the data including the reproducibility obtained at the end of the batch operation this time, it is possible to find out what measures to take in the next batch operation to improve product quality and yield.

In the above embodiment, the reproducibility of the process state is evaluated at the time of operating the process to be evaluated, which is a key point of the batch operation, and this reproducibility can be provided to the operator in real time as information. For this reason, it is easy to quickly recognize the difference from the normal state during the batch operation and to take quick action, and it is also possible to perform efficient analysis when investigating the cause of the fluctuation in quality or yield.

When a new system is constructed in order to manufacture a new product, a default value of 1.0 is used as the weight coefficient for calculating the average value of the deviation calculated for each instrument. Then, when manufacturing the same product continuously, weighting is performed for each monitoring target instrument based on the data obtained earlier, and this is sequentially revised during batch operation for the same brand that is repeatedly performed.

As described above, the present invention has been described based on the preferred embodiment. However, the batch plant operation management device of the present invention is not limited to the configuration of the above-described embodiment. Various modifications and changes from the configuration described above are also included in the scope of the present invention.

[0024]

As described above, in the batch plant operation management apparatus of the present invention, the quality of the process state can be determined in real time at the time of occurrence of each event to be evaluated. There is an advantage that the response to improve can be made quickly.

[Brief description of the drawings]

FIG. 1 is a block diagram of a batch plant operation management device according to an embodiment of the present invention.

FIG. 2 is a plant operation flowchart showing operations to be evaluated by the batch plant operation management device of FIG. 1;

FIG. 3 is a display example on a screen of an information display device in the batch plant operation management device of FIG. 1;

[Explanation of symbols]

 10: Batch plant operation management device 11: Data collection unit 12: Event management unit 13: Statistical processing unit 14: Performance comparison unit 15: Information display unit 16: Evaluation target event and monitoring target instrument list storage device 17: Performance database 18: Batch plant

Claims (2)

[Claims] 1. A batch plant operation management device for managing the operation of a batch plant while monitoring the operation of the batch plant, wherein a set value, an indicated value, and a related value of a plurality of instruments are provided at each of a plurality of events to be evaluated. A storage device for storing process data including at least one of the manipulated variables to be performed for each instrument and for each event occurrence time; and a corresponding process obtained by a predetermined number of past batch operations of each process data of the current batch. An arithmetic device that calculates a deviation from an average value of data, normalizes the deviation according to the range of the corresponding instrument, and calculates an average value of the normalized deviation values of the respective instruments at each time of the event occurrence. And determining means for determining whether the overall plant condition is good or bad based on the average value of the deviation values at the time of occurrence of each of the events. Runt operation management device. 2. The method according to claim 1, wherein the normalized deviation value is 1 which is relatively large.
The batch plant operation management device according to claim 1, wherein one or more monitored instruments are displayed.