JP2004129322A - Predictive control system for power demand - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a predictive control system for power demand which can predictively control the total power demand of a group of factories equipped with many manufacturing lines accurately so that it may not go wide of contracted power qualifications. <P>SOLUTION: Many manufacture lines are classified into a plurality of patterns by the quantity of used power and the fluctuation of the quantity of used power. To be concrete, the manufacture lines are classified into the first pattern where the quantity of used power and the fluctuation of the quantity of used power are both large, the second pattern of the small fluctuation in the quantity of used power though the quantity of used power is large, and the third pattern where the quantity of used power is small. The power demand is predicted according to the property in every pattern, and the predicted value of their total power demand is obtained. When there is a fear of the predicted value going wide of the contacted power qualifications, an energy concentrated control center 20 issues a power use limiting command. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention is a predictive control system for the total power demand in a group of factories equipped with a large number of manufacturing lines, and when power is purchased from a power company, and / or by sending back excess power to the power company. This is a power demand prediction control system that accurately performs predictive control so as not to deviate from a contract power condition with a power company when selling power. The contract power condition when the power is purchased is generally an upper limit of the power amount, and the contract power condition when the power is sold is a lower limit of the power amount. The contract power referred to here is the contract power when a contract is made to purchase part or all of the power from a power company.
[0002]
[Prior art]
[Patent Document 1] Japanese Patent Publication No. Hei 5-22157 (page 2, column 3, lines 26 to 29)
For example, in a group of factories having a large number of production lines, such as a steel mill, an energy central management center for managing the total power demand is set up. The centralized energy management center constantly monitors the total power demand so that it does not deviate from the contract power condition with the power company (see Patent Document 1). If the contracted power condition deviates from the contracted power condition even for a short time (for example, the accumulated power value for 30 minutes), a large penalty must be paid. However, if the contracted power condition is set to include an allowance for safety, for example, If the power is purchased from a power company, setting the upper limit of the contracted power to a higher value will increase the contract fee. In the case where power is sold to a power company, setting the lower limit of the contracted power to a lower value lowers the contract fee for the power sale. For this reason, companies perform daily power control so that the maximum value, average value, and standard deviation of the total power demand are close to the restricted regulation value within the contract power condition and do not deviate from the contract power condition.
[0003]
For this reason, the power demand of each production line has been monitored in real time, and if the total power demand is likely to fall outside the contracted power conditions with the power company, the centralized energy management center will limit the power usage to the production line. Is giving instructions. In addition, the long-term and / or short-term power demand based on the production plan of each manufacturing line is calculated by the computer system for process management, and the results are aggregated by the computer system for energy management at the centralized energy management center, and then required. By adjusting the production plan accordingly, the total power demand is adjusted so as not to deviate from the contract power condition.
[0004]
However, in the conventional power demand prediction control system, the predicted value and the actual value may deviate by about 10% at the maximum. The reason is that the production plan does not always match the actual operation, and the frequently changed operation / stop schedule cannot be immediately reflected in the power demand prediction control system. In addition, when the total power demand is likely to deviate from the contracted power conditions with the power company, the decision on which production line to issue the power use restriction instruction at that time depends on the experience of the operator of the centralized energy management center. Depends on the experience of the production line. In addition, since the invalid time from when the power use restriction instruction is issued to the production line to when it is executed varies depending on the line, the execution of the power use restriction may be delayed.
[0005]
[Problems to be solved by the invention]
The present invention solves the conventional problems described above, improves the prediction accuracy of power demand, and sets a situation in which power demand deviates from contract power conditions (exceeds contract power conditions in the case of power purchase, in the case of power sales). It is a main object of the present invention to provide a power demand prediction control system capable of preventing occurrence of (below the contract power condition) in advance. Further, another object of the present invention is to provide an instruction of power use control to which production line at which timing when total power demand is likely to deviate from the contracted power condition or when private power generation equipment suddenly stops. The purpose of the present invention is to provide a power demand forecasting demand system that can provide guidance to an operator of the centralized energy management center on whether to issue power.
[0006]
[Means to solve the problem]
The power demand prediction control system of the present invention made to solve the above-mentioned problem is a system for predictively controlling the total power demand of a group of factories having a large number of production lines so as not to deviate from the contract power. A large number of manufacturing lines are classified into a plurality of patterns according to power consumption and fluctuations in power consumption, and the power demand is predicted for each pattern, and the total power demand is predicted from the total thereof. It is characterized in that the power usage is restricted when there is a possibility that the condition may be violated. Further, in the present invention, the contract power condition is an upper limit of the power amount, the contract power condition is a lower limit of the power amount, or the contract power condition is both the upper limit and the lower limit of the power amount. There may be. In the present invention, the manufacturing line is divided into a first pattern in which both the power consumption and the fluctuation in the power consumption are large, a second pattern in which the power consumption is large and the fluctuation in the power consumption is small, and a third pattern in which the power consumption is small. It is preferable to classify and predict the power demand for each pattern.
[0007]
In addition, a production line capable of restricting power use and an invalid time from when a power use restriction instruction is issued to the production line until execution are set in advance. By providing a function of simulating whether to limit the power usage by selecting a line, it is possible to provide an operator of the centralized energy management center with appropriate judgment criteria. Further, it is preferable to add a function of simulating which production line should be selected to restrict the electric power usage when the private power generation equipment is stopped (especially at a sudden stop). For example, as a group of factories having a large number of production lines, it is effective to apply to a business having a plastic working process and a plating process, such as a steel mill, having the above-mentioned three patterns.
[0008]
According to the present invention described above, it is possible to greatly improve the accuracy of the prediction of the power demand as compared with the conventional case, and when the total power demand is likely to deviate from the contract power condition or when the private power generation equipment is stopped (particularly, In the case of a sudden stop), it is possible to inform the operator of the centralized energy management center of which production line the power use control instruction should be issued.
[0009]
Hereinafter, embodiments of the present invention will be described. There are three cases where the contract power condition is the upper limit of the power amount, the contract power condition is the lower limit, and the contract power condition is both the upper limit and the lower limit of the power amount. The power condition will be described with an example of the upper limit of the power amount. From this description, those skilled in the art can understand the power demand prediction system of the present invention when the contract power condition is the lower limit of the power amount, and when the contract power condition is both the upper limit and the lower limit of the power amount. It is possible to do it.
[0010]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 2 is a conceptual diagram of a group of factories having a large number of production lines in FIG. 1, representative examples of which are a hot rolling factory 1, a cold rolling factory 2, a plating factory 3, another factory group 4, and an office 5 in an iron mill. Is illustrated. The hot rolling factory 1, the cold rolling factory 2, the plating factory 3 and the like include many production lines. The production line of the hot rolling factory 1 is naturally a hot rolling line, the production line of the cold rolling factory 2 is a cold rolling line, and the production line of the plating factory 3 is an electroplating line 8. Although FIG. 1 shows a plurality of production lines in each factory, even if a factory consisting of a single production line exists in a group of factories, the number of production lines may be any as a group of factories.
[0011]
Each factory is provided with a computer system 10 for operation control, and the actual operation is performed under the control of these computer systems 10. The daily operation schedule is input to each computer system 10, and all the operation schedules of the computer system 10 that are frequently changed and the schedules of operation and suspension are all input to the computer system 10, so that each computer system 10 The operation schedule has a good agreement with the actual operation. Therefore, the computer system 10 of each factory and the power management computer system 21 of the centralized energy management center 20 are connected by a communication line, and the power demand is predicted as follows.
[0012]
In the present invention, it is preferable to classify a large number of production lines into a plurality of patterns of power consumption and power consumption fluctuation, and to predict power demand for each pattern. Specifically, it is classified into a first pattern in which both the power consumption and the power consumption fluctuation are large, a second pattern in which the power consumption is large and the power consumption fluctuation is small, and a third pattern in which the power consumption is small. . Of course, it can be classified into more patterns.
[0013]
The hot rolling line 6 and the cold rolling line 7 are classified into the first pattern, and the power demand is as shown in the graph of FIG. The operating conditions of the hot rolling line 6 and the cold rolling line 7 greatly affect the total power demand. Therefore, the computer systems 10 of the hot rolling mill 1 and the cold rolling mill 2 predict the power demand by using a regression equation obtained from the rolling reduction, the sheet pressure, and the like in the input rolling schedule. This regression equation is obtained by performing multiple regression analysis on past data, and is, for example, as shown in FIG.
[0014]
It is the electroplating line 8 that is classified into the second pattern in which the power consumption is large but the fluctuation of the power consumption during operation is small. The power demand is as shown in the graph of FIG. 3, and although the power consumption is large, there is almost no change in the power consumption during operation. For this reason, if only the operation schedule can be grasped, it is possible to accurately predict the power demand.
[0015]
Classified into the third pattern that uses a small amount of power is a small pattern such as the power of the office 5 and a manufacturing line that is installed in another factory group 4. The power demand is as shown in the graph of FIG. Therefore, it can be assumed that if the time is relatively short, it proceeds as it is, and the normal air power consumption within a certain period in the past is used as the predicted value. However, it is necessary to review the predicted value at regular intervals. For example, a moving average of the past 15 minutes is calculated every minute and used as the predicted value of the power demand one hour ahead.
[0016]
The power demands classified and predicted for each pattern as described above are totaled, and the total power demand of the entire factory group is predicted. The estimated value and the actual value are displayed on the power demand management screen 22 of the centralized energy management center 20, as shown in FIG. In this example, the latest actual value is indicated by a solid line, and the predicted value is indicated by a broken line. For example, the predicted value of the power demand up to one hour ahead is displayed on the power demand management screen 22, and when the predicted value may exceed the upper limit regulation of the contract power condition, the operator of the energy central management center 20 Restrict power usage.
[0017]
The power demand prediction control system according to the present invention described above has better prediction accuracy than the conventional prediction control system. According to the results of the applicant company, as described above, the conventional method had a prediction error of about 10% at maximum, but according to the present invention, the prediction error could be significantly reduced to 1%. .
[0018]
Restriction of power use by the operator is performed by a method of instructing each factory by telephone to restrict power use. However, there are few things that can be stopped immediately at the production line of each factory, and in many cases, the work on the line cannot be stopped unless the work on the line is completely eliminated. . Therefore, the operator looks at the ever-changing movement of the predicted value on the power demand management screen 22 and, if it is likely to exceed the upper limit of the contract power condition as it is, considers the invalid time and sets the power use restriction in advance. I have to give directions.
[0019]
However, it is not easy for an operator to accurately determine which manufacturing line should be selected to limit the power usage in a short time. In addition, there are manufacturing lines for which electric power use cannot be restricted for safety reasons. Therefore, in the invention of claim 3, the power management computer system 21 determines the production lines on which power use can be restricted and the priority order, and the invalid time from when the power use restriction instruction is issued to the production line to when the power use restriction is executed. Is set in advance. Then, a simulation is performed as to whether the power use should be limited, and the result can be displayed on the screen.
[0020]
Using this simulation function, the operator can know in a very short time which production line to select and limit the power usage to avoid exceeding the upper limit of the contracted power condition, and even in an emergency It is possible to issue an instruction for power use restriction without any need.
[0021]
In a group of factories in which private power generation equipment bears part of the power demand, when the private power generation equipment stops due to an accident (especially when it suddenly stops), the power demand purchased from the power company Since power consumption rises instantaneously, it is necessary to limit the power usage of each production line in the shortest possible time. In this case as well, if the power management computer system 21 has a function of simulating which production line should be selected, the operator can instruct the power use restriction accurately and promptly without having to be in an emergency. Can be issued.
[0022]
【The invention's effect】
As described above, according to the power demand prediction control system of the first to fifth aspects of the present invention, the prediction accuracy of the total power demand of a group of factories equipped with a large number of production lines can be greatly improved compared to the related art. Thus, it is possible to prevent in advance that the total power demand is out of the contract power condition. According to the sixth and seventh aspects of the present invention, when the total power demand is likely to deviate from the contract power condition or when the private power generation equipment stops (especially when it suddenly stops), Can inform the operator of the centralized energy management center which production line should be instructed to use the power usage restriction, and the operator can issue the power usage restriction command accurately and quickly without having to panic in an emergency. it can.
[0023]
The applicable range of the present invention is not limited to a steel company, but can be widely applied to a group of factories having a large number of production lines having different power consumptions and fluctuations in the power consumption.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram of a group of factories having a large number of production lines. FIG. 2 is a graph showing power demand of production lines classified into a first pattern.
FIG. 3 is a graph showing a power demand of a manufacturing line classified into a second pattern.
FIG. 4 is a graph showing a power demand of a manufacturing line classified into a third pattern.
FIG. 5 is a regression equation of electric power obtained from a rolling reduction, a plate thickness, and the like.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Hot rolling factory 2 Cold rolling factory 3 Plating factory 4 Other factories 5 Office 6 Hot rolling line 7 Cold rolling line 8 Electroplating line 10 Computer system for operation control 20 Energy central management center 21 Power management Computer system 22 for power demand management screen

Claims (7)

This system predicts and controls the total power demand of a group of factories equipped with a large number of production lines so that it does not deviate from the contracted power. In addition, the power demand is predicted for each pattern, the total power demand is predicted from the total thereof, and the power usage is limited when the predicted value may deviate from the contract power condition. Predictive control system. 2. The power demand forecasting system according to claim 1, wherein the contract power condition is an upper limit of the power amount. The power demand forecasting system according to claim 1, wherein the contract power condition is a lower limit of power amount. 2. The power demand forecasting system according to claim 1, wherein the contract power condition is both upper limit and lower limit of the amount of power. The production line is classified into a first pattern in which both the power consumption and the power consumption fluctuation are large, a second pattern in which the power consumption is large and the power consumption fluctuation is small, and a third pattern in which the power consumption is small. The power demand forecasting system according to any one of claims 1 to 4. In advance, set a production line capable of restricting power use and an invalid time from when a power use restriction instruction is issued to the production line until execution is performed. The power demand forecasting system according to any one of claims 1 to 5, further comprising a function of simulating whether to select and limit the power use. The power demand forecasting system according to any one of claims 1 to 6, further comprising a function of simulating which production line should be selected to limit the power usage when the private power generation facility is stopped.

Images