JPH0259801A - Optimization controller - Google Patents

Abstract

PURPOSE:To ensure the effective operation and to reduce the load of an operator by discontinuing the operation of optimization together with execution of the second operation in case the information on the initialization point is changed or fluctuated and therefore eliminating the useless time of operation as well as the calculation of the improper set value. CONSTITUTION:An initialization point information monitor part 6 monitors the optimization point information C even in an executing period of operation of the optimization when an optimization calculation part 4 calculates the set value based on the information C issued from an initial point information calculation part 3. When the information C is fluctuated, an optimization operation control part 7 issues a discontinuing instruction and then a re-executing instruction of the optimization operation to the part 4. Thus the part 4 discontinues the optimization operation under execution and restarts another optimization operation to calculate the set value based on the fluctuated information C. As a result, the useless operation time is eliminated and the calculation of the improper set value is avoided. Thus it is possible to improve the operating efficiency of a plant 1 and to automatically apply set value to the plant 1. Then the operator's load is reduced.

Description

[Detailed Description of the Invention] [Objective of the Invention] (Industrial Application Field) The present invention calculates set values of various manipulated variables of a plant by optimization calculation using plant data indicating the current state of the plant as initial point information. The present invention relates to an optimization control device that applies energy to a plant, and particularly relates to an optimization control device that immediately redoes the optimization calculation when initial point information changes during the execution of the optimization calculation.

(Prior Art) Generally, in an optimization control device, set values of various manipulated variables in a plant are calculated using optimization calculations. When performing this optimization calculation, initial point information is first calculated using plant data indicating the current state of the plant.

This initial point information is, for example, a function value of a target value for plant operation found from plant data, a function value of a constraint condition, a current value of an operating state coefficient of the plant, etc.

Once the initial point information is calculated, the setting values for each of the manipulated variables are calculated by optimization calculation using this initial point information as a starting point. The calculated set value is output to the plant to be controlled via the set value output section.

The plant is then operated at the set values and transitions to the target operating state.

In such an optimization control device, for example, when an operator manually changes the output value of a plant, or when the amount of fuel supplied to a boiler in a boiler plant fluctuates due to changes in the conditions of the fuel supply process, etc. Due to changes in the status of other plants related to this plant, etc.
Plant data indicating the current state of the plant is often changed or fluctuated. When the plant data changes or fluctuates, the initial point information calculated from this plant data also fluctuates.

On the other hand, in an actual plant, the number and types of plant data are large, and the number of manipulated variables is also large. As a result, the number of initial point information and the number of necessary setting values increase, and the processing time of the optimization calculation that calculates the setting value using the initial point information becomes longer. For example, the time required for one calculation is 20~ Sometimes it takes 30 minutes. Then, such optimization calculation processing is repeated every 3 to 4 hours.

Therefore, the initial point information may fluctuate or change during execution of the optimization calculation. In this way, if the initial point information fluctuates or changes during the execution of a calculation, the setting value obtained by the optimization calculation using the initial point information before the fluctuation will not correspond to the initial point information after the fluctuation or change. Therefore, if this set value is applied to the plant, there will be a problem that the plant will be operated in an unintended direction.

In order to avoid such inconveniences, once the set value is calculated by the optimization calculation process, the operator should check the set value and make sure it is a reasonable value instead of immediately applying the set value to the plant. The voltage is applied to the plant only when it is determined that the voltage is present. If an abnormal value is found, the operator cancels that value, runs the optimization calculation again, and when a reasonable set value is obtained, manually applies the set value to the plant. I have to.

However, if the operator confirms the setting values obtained through the optimization calculation process and then applies the voltage to the plant, even if the initial point information changes or fluctuates during the execution of the optimization calculation, the initial point information Even if the optimization calculation does not change until the end of the optimization calculation, the optimization calculation is executed to the end using the initial point information at the start of the calculation. As a result, if the initial point information changes or fluctuates during the process, the calculation result will not be used, resulting in wasted calculation time.

Therefore, the waiting time until a new set value that is required due to a change or variation in the initial point information is calculated becomes long, and as a result, a problem arises in which the plant cannot be operated efficiently.

Furthermore, the operator must constantly monitor the set values calculated by the optimization calculation, which increases the burden on the operator.

Furthermore, since the operator determines whether or not the calculated set value is appropriate, there is a risk that a judgment error may occur. If an incorrect set value is applied to the plant, the set value becomes a disturbance factor for the plant.

(Problem to be solved by the invention) In this way, in the conventional optimization control device, even if the initial point information changes or fluctuates during the process, the optimization calculation continues as is using the initial point information before the change. Therefore, there was a problem that unnecessary calculation time was consumed and the plant could not be operated efficiently.

In addition, the operator has to judge the calculation results each time, which increases the burden on the operator and raises the possibility of errors in judgment.

The present invention monitors the initial point information calculated from the current plant data, and when the initial point information changes or fluctuates, cancels the optimization calculation and performs the recalculation, thereby eliminating wasted calculation time. We have created an optimization control device that can prevent inappropriate set values from being calculated, thereby allowing the plant to operate efficiently, and further reducing the burden on the operator by eliminating the operator's need to judge the set values. The purpose is to provide.

[Structure of the Invention] (Means for Solving the Problems) In order to solve the above problems, the present invention includes a plant data detection unit that detects plant data indicating the current state of the plant, and a system for detecting data in the plant data detection unit. An initial point information calculation unit that calculates initial point information such as target values and constraint conditions for plant operation using the plant data that has been calculated, and a In an optimization control device that has an optimization calculation unit that calculates set values of various manipulated variables by optimization calculation, and a set value output unit that outputs the set values of various manipulated variables calculated by this optimization calculation unit to a plant. , an initial point information monitoring unit that monitors whether or not a change has occurred in the initial point information calculated by the initial point information calculation unit during the execution period of the optimization calculation in the optimization calculation unit; and this initial point information monitoring unit. When a change in the initial point information is detected, a command to stop the execution of the optimization calculation is sent to the optimization calculation unit, followed by a command to re-execute the optimization calculation using the initial point information after the change. and an optimization calculation control section.

(Function) By taking the above-mentioned means, the present invention allows the initial point information monitoring section to perform the initial point information even during the execution period of the optimization calculation in which the optimization calculation section calculates the setting value based on the initial point information. Point information is monitored.

When the initial point information changes, the optimization calculation control unit sends an optimization calculation stop command to the optimization calculation unit, followed by an optimization calculation re-execution command. As a result, the optimization calculation section stops the optimization calculation currently being executed and re-executes the optimization calculation for calculating the set value based on the initial point information after the change.

Therefore, wasted calculation time is greatly reduced. Furthermore, the optimization calculation unit no longer outputs the set value calculated using the initial point information before the change. Therefore, the set value can be automatically sent to the plant without the operator having to monitor the set value.

(Example) An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing an optimization control device according to an embodiment. In other words, as shown in the figure, the optimization control device is broadly divided into a plant 1 to be controlled, a plant data detection unit 2 that detects plant data a, and an initial point that calculates initial point information using the detected plant data b. The basic system consists of an information calculation section 3, an optimization calculation section 4 that calculates a setting value by optimization calculation based on the calculated initial point information C, and a setting value output section 5 that outputs the calculated setting value d to the plant 1. It is composed of Furthermore, in the optimization control device of this embodiment, the initial point information e calculated by the initial point information calculation section 3
An initial point information monitoring section 6 that monitors fluctuations in , and an optimization calculation control section 7 that controls the calculation operation of the optimization calculation section 4 in response to the monitoring results are added.

Next, the operation of each part will be explained in order.

Plant data a indicating the current state is always inputted to the plant data detection unit 2 from the plant 1 as a controlled object.

A timer 8 and an input terminal 9 consisting of a keyboard and a CR7 display are connected to the optimization calculation unit 4, and the optimization calculation unit 4 receives output from the timer 8 at a cycle of, for example, 3 to 4 hours. When receiving a periodic calculation request f or a manual calculation request g sent from the input terminal 9 operated by an operator, it sends an initial point information request to the initial point information calculation section 3.

Upon receiving the initial point information request, the initial point information calculating section 3 takes in the plant data b necessary for calculating the initial point information from the plant data detecting section 2.

Then, from the inputted plant data b, the initial point information at the current point in time, which is indicated by the function value of the target value for the plant operation, the constraint function value for the target value, the current value of the operating state coefficient of the plant, etc., is calculated. The calculated initial point information C is sent to the optimization calculation section 4.

The optimization calculation unit 4 that has received the initial point information C starts an optimization calculation based on the initial point information C. At the same time, a monitoring start command i and initial point information j used in the optimization calculation are sent to the initial point information monitoring section 6.

The initial point information monitoring unit 6 that has received the monitoring start command i and the initial point information j stores the input initial point information j in its own storage unit until the optimization calculation in the optimization calculation unit 4 is completed. Then, the current initial point information request k is sent to the initial point information calculation unit 3 at regular intervals ΔT. The initial point information calculation unit 3 calculates the current initial point information e at regular intervals ΔT according to the initial point information request k using the plant data b from the plant data detection unit 2, and outputs the current initial point information e to the initial point information monitoring unit 6. Send to.

The initial point information monitoring section 6 compares the current initial point information e, which is sequentially input at regular intervals ΔT, with the initial point information j, which is input from the optimization calculation section 4 and which is stored in its own storage section. Specifically, each element of the input current initial point information e is compared with each element of the initial point information j used for the optimization calculation, and the current initial point information e is It is determined whether the initial point information j has varied by a preset tolerance value or more in each element or all elements.

If the fluctuation is less than the allowable value, nothing is done and the input of initial point information e for the next cycle is waited.

If the variation is equal to or greater than the allowable value, a variation detection signal is sent to the optimization calculation control section 7.

Upon receiving the fluctuation detection, the optimization calculation control unit 7 sends an optimization calculation execution stop command m to the optimization calculation unit 4. Subsequently, after a predetermined period of time has elapsed, a command n to re-execute the optimization calculation using the changed initial point information is similarly sent to the optimization calculation unit 4.

The optimization calculation unit 4 continues to execute the optimization calculation unless an execution stop command m is input from the optimization calculation control unit 7. When the optimization calculation is completed and the set value d is determined, the set value d is sent to the set value output section 5. Set value output section 5
sends the set value p corresponding to the input set value d to the plant 1. Thus, the plant 1 is operated at the set value p and shifts to the target operating state.

On the other hand, when the optimization calculation unit 4 receives an execution stop command m from the optimization calculation control unit 7 during execution of the optimization calculation, it stops the optimization calculation currently being executed.

Then, when a re-execution command n is input, an initial point information request is sent to the initial point information calculation section 3, and the current initial point information C after the change is fetched. Then, the optimization calculation is restarted based on the initial point information C after this change.

At the same time, a monitoring start command i and initial point information j used in the optimization calculation are sent to the initial point information monitoring section 6.

If the execution stop command m is not inputted from the optimization calculation control unit 7 during the execution of the optimization calculation, the set value d corresponding to the initial point information after the fluctuation obtained by the optimization calculation
is sent to the set value output section 5. Thus, the plant 1 is operated toward the target of the new operating state after the change.

In this way, in the optimization control device shown in FIG.
If the plant data changes during execution of an optimization calculation and the initial point information fluctuates or is changed, the optimization calculation in progress is automatically stopped and a new one is created using the initial point information after the change. The optimization operation is re-executed. Therefore, the time during which the optimization calculation unit 4 executes wasteful optimization calculations can be minimized, so that after the initial point information changes, a new set value can be obtained based on the initial point information after the change. The time it takes is shortened. As a result, the response of the plant becomes quick and the plant can be operated efficiently.

In addition, since the optimization calculation unit 4 is automatically prevented from outputting the set value based on the initial point information before variation, the operator can monitor and judge the set value obtained by the optimization calculation unit 4. There's no need. Therefore, since the application of set values to the plant can be automated, the burden on the operator can be significantly reduced.

Further, since human judgment is not involved in setting values, disturbances are not applied to the plant due to judgment errors as in conventional devices.

[Effects of the Invention] As explained above, according to the optimization control device of the present invention,
The initial point information calculated from the current plant data is monitored, and if the initial point information changes or fluctuates, the optimization calculation is stopped, and then recalculation is performed using the initial point information after the change. There is.

Therefore, it is possible to eliminate unnecessary calculation time and prevent inappropriate set values from being calculated, thereby allowing the plant to operate efficiently. Further, since the operator can automatically apply the set value to the plant without having to judge the set value, the burden on the operator can be significantly reduced.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an optimization control device according to an embodiment of the present invention. 1...Plant, 2...Plant data detection unit, 3
... initial point information calculation section, 4 ... optimization calculation section, 5.
...Set value output section, 6...Initial point information monitoring section, 7...
・Optimization calculation control unit. Applicant's agent Patent attorney Takehiko Suzue

Claims (1)

[Claims] A plant data detection unit that detects plant data indicating the current state of the plant, and an initial point of target values, constraint conditions, etc. regarding plant operation using the plant data detected by the plant data detection unit. an initial point information calculation unit that calculates information; an optimization calculation unit that calculates set values of various manipulated variables in the plant by optimization calculation based on the initial point information calculated by the initial point information calculation unit; In an optimization control device having a set value output unit that outputs set values of various manipulated variables calculated by the optimization calculation unit to the plant, the initial an initial point information monitoring section that monitors whether or not a change has occurred in the initial point information calculated by the point information calculation section; and when the initial point information monitoring section detects a change in the initial point information, the optimization is performed. and an optimization calculation control unit that sends a command to the calculation unit to stop execution of the optimization calculation, and then sends a command to re-execute the optimization calculation using the initial point information after the fluctuation. Optimization control device.