Optimal operation controller of plant

• US 7,269,466 B2
• Filed: 07/08/2003
• Issued: 09/11/2007
• Est. Priority Date: 07/24/2002
• Status: Expired due to Fees

• Current Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Inventors: Hirai, Norio
• Primary Examiner(s): Knight; Anthony
• Assistant Examiner(s): Chang; Sunray
• Application Number:
  US10/614,129
• Publication Number:
  US 20040019392A1
• Time in Patent Office: 1,526 Days
• Field of Search: 700 28- 29, 340/511, 606/60, 250/339.12, 738/631.2, 123/568, 123/674, 123/568.22
• US Class Current: 700/28
• CPC Class Codes: G05B 13/024 in which a parameter or coefficient is automatically adjusted to optimise the performance

Abstract

The present invention aims to control elements to improve operation efficiency of a plant based on a correlation value of the past accumulated data. According to the invention, an optimal operation controller obtains a correlation between a state of a predetermined process and each element based on an operation status of the plant to be controlled, storing the correlation in a correlation table, and compute an operation efficiency for each element based on the operation status of the plant. Also, the operation controller has a categorization efficiency table for storing the computed operation efficiency of the predetermined process. The operation controller refers to the categorization efficiency table based on a data input from the plant, and outputs an instruction to control the each element accordingly.

Claims

1. An optimal operation controller of a plant comprising:

a correlation analyzing unit for performing regression analysis to obtain a correlation between an operative parameter of a predetermined process and each of a plurality of operation elements based on an operation status of the plant to be controlled, storing the correlation in a correlation table, and computing operation efficiency for each operation element based on the operation status of the plant, wherein said correlation analyzing unit uses data collected from the plant during a prior implementation of the predetermined process to perform the regression analysis and compute the operation efficiencies of the operation elements;

a categorization efficiency table for storing the operation efficiency of each operation element computed by the correlation analyzing unit in association with a highly correlative operative parameter of the predetermined process as determined by the regression analysis, wherein the categorization efficiency table further stores a relationship between the operation efficiency of each operation element and the overall efficiency of the predetermined process determined based on the data collected during the prior implementation; and

an optimal pattern searching unit for performing a look-up of the categorization efficiency table based on data collected from the plant during a subsequent implementation of the predetermined process in order to output an instruction to control each of the operation elements during the subsequent implementation of the predetermined process without simulating or computing an operation efficiency.

2. The optimal operation controller of the plant of claim 1, wherein:

the categorization efficiency table stores the operation efficiency for operation element and the operation efficiency of an entire plant, and

the optimal pattern searching unit controls each operation element in consideration of the operation efficiency of the entire plant.

3. The optimal operation controller of the plant of claim 1, wherein the correlation analyzing unit categorizes the correlation between the state of the predetermined process and each operation element into specific steps based on the data input from the plant to be controlled, and writes the correlation in the correlation table.

4. The optimal operation controller of the plant of claim 3, wherein

the categorization efficiency table stores data of an approximated curve generated by the categorized correlation, and

the optimal pattern searching unit outputs the instruction by referring to the data of the approximated curve.

5. An method of performing optimal operation control of a plant, comprising:

performing regression analysis to obtain a correlation between an operative parameter of a predetermined process and each a plurality of operation elements based on an operation status of the plant to be controlled, wherein the regression analysis is performed using data collected from the plant during a prior implementation of the predetermined process,

storing the correlation obtained by the correlation analyzing step into a correlation table,

computing efficiency for each of the one or more operation elements based on the operation status of the plant, wherein the computing step uses the data collected from the plant during the prior implementation of the predetermined process,

storing, in a categorization efficiency table, the operation efficiency of each operation element obtained by the efficiency computing step and a highly correlative operative parameter of the predetermined process as determined by the regression analysis,

storing, in the categorization efficiency table, a relationship between the operation efficiency of each operation element and the overall efficiency of the predetermined process determined based on the data collected during the prior implementation, and

performing a look-up of the categorization efficiency table based on data collected from the plant during a subsequent implementation of the predetermined process in order to output an instruction to control each of the operation elements during the subsequent implementation of the predetermined process without simulating or computing operation efficiency.

6. The optimal operation controller of claim 5, wherein the plurality of operation elements are devices for configuring the plant during the predetermined process.

7. The method of claim 5, further comprising:

storing the operation efficiency for each operation element and of the entire plant in the categorization efficiency table, and

controlling each operation element in consideration of the operation efficiency of the entire plant.

8. The method of claim 5, further comprising:

categorizing the correlation between the state of the predetermined process and each operation element into specific steps based on the data input from the plant to be controlled, and

writing the correlation in the correlation table.

9. The method of claim 8, further comprising:

storing data of an approximated curve generated by the categorized correlation in the categorization correlation table, and

referring to the data of the approximated curve when performing the look-up of the categorization correlation table in order to output the control instruction.

10. The method of claim 5, wherein the predetermined process involves the plurality of operation elements.

11. The method of claim 10, wherein the plurality of operation elements are devices for configuring the plant during the predetermined process.

12. A method for controlling devices in a plant in order to optimize the operation of the plant during a process, comprising:

collecting operational data from the plant during a first implementation of the process, the collected operational data relating to operational parameters of the plant and outputs of the devices;

calculating an efficiency value for each of the devices based on the collected operational data;

performing regression analysis to correlate changes in the one or more operational parameters to the calculated efficiency values of the respective devices based on the collected operational data to generate correlation values;

storing the generated correlation values in a correlation table;

storing each of the calculated efficiency values of the devices in a look-up table in association with a highly correlative operating parameter of the process as determined by the regression analysis;

storing a relationship between each of the calculated efficiency values of the devices and an overall efficiency of the process in the look-up table, the relationship being determined based on the collected operational data; and

referring to the look-up table in order to output a control instruction to each of the devices during a second implementation of the process without simulating or computing efficiency, wherein the second implantation of the process occurs subsequent to the first implementation.

13. The method of claim 12, wherein

the generated correlation values are stored in the look-up table in response to determining that an operational parameter has a strong effect on efficiency of at least one of the devices based on the regression analysis being performed on the generated correlation values, and

the generated correlation values are stored in the look-up table in response to the determination of the strong effect.

14. The method of claim 13, wherein

the collected operational data includes a plurality of variables related to the operation of the plant during the process, and

the correlating step includes calculating a correlation coefficient for each of the variables according to the performed regression analysis, the one or more operational parameters being selected from the variables on the basis of the correlation coefficients.

15. The method of claim 12, wherein the collected operational data is continuously sampled as an output quantity of the process changes, the method further comprising:

categorizing the sampling points of the collected operational data into a number of specific steps corresponding to a number of instructions for specifying an output quantity of the process.

16. The method of claim 12,

wherein the correlating step determines a relationship between efficiency of at least one of the devices and efficiency of the entire plant, the control instruction being output to the at least one of the devices based on the determined relationship.

17. The method of claim 16, wherein

the control instructions are output to the plurality of devices in order to maximize the efficiency of the entire plant.

18. The method of claim 12, wherein the method is performed for a water conveyance system to optimize conveyance of water to a plurality of pump stations.