Plant utility optimizing method and an optimizing system

• US 5,886,895 A
• Filed: 07/01/1996
• Issued: 03/23/1999
• Est. Priority Date: 09/26/1994
• Status: Expired due to Term

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Kita, Yoichi, Kamibayashi, Tohru
• Primary Examiner(s): Cosimano, Edward R.
• Application Number:
  US08/640,949
• Time in Patent Office: 995 Days
• Field of Search: 364/138, 364/148.01, 364/148.08, 364/149, 364/151, 364/152, 364/528.23
• US Class Current: 700/28
• CPC Class Codes: F01K 13/02 Controlling, e.g. stopping or starting
• CPC Class Codes: G05B 13/00 Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion G05B19/00 takes precedence
• CPC Class Codes: G05B 13/02 electric
• CPC Class Codes: G05B 13/042 in which a parameter or coefficient is automatically adjusted to optimise the performance

Abstract

For calculating optimum operation parameters of a BTG plant including a plurality of boilers having a linear input/output characteristic between an incoming fuel and steam output and a plurality of turbines having a non-convex input/output characteristic between the incoming steam which is an output of the boiler and power output, a dynamic programming process is performed twice and a linear programming process is performed once with respect to the steam and power to find, as an optimum solution, the power output of the turbines. Finally, an optimum solution is found in which a total fuel cost is minimized.

Claims

1. A utility optimizing method for calculating optimum operation parameters of a plant including a plurality of first plant elements having a linear input/output characteristic with first physical amounts as an input and second physical amounts as an output and a plurality of second plant elements having a non-convex input/output characteristic with the second physical amounts output from the first plant elements as an input and third physical amounts as an output, the method comprising:

a first step of calculating a division of the second physical amounts to the second plant elements by using a dynamic programming to maximize the total of the third physical amounts output from the second plant elements associated with the total of the second physical amounts in a predetermined range satisfying a first demand for the second physical amounts;

a second step of calculating a division of the second physical amounts to the second plant elements by using a dynamic programming based on a set of the total of the second physical amounts, total of the third physical amounts and division of the second physical amounts to the second plant elements obtained by the first step at each of the second physical amounts to maximize the total of the third physical amounts output from said all the second plant elements and corresponding to the total of the second physical amounts to all the second plant elements;

a third step of calculating, by using one of a linear programming or a nonlinear programming, the second physical amounts input to the respective second plant elements and a first cost required to supply the first physical amounts with respect to each set of the total of the second physical amounts, total of the third physical amounts and division of the second physical amounts to the second plant elements obtained by the second step; and

a fourth step of, based on each set of the total of the third physical amounts and first cost obtained at the third step, calculating a second cost required to receive a power which corresponds to a shortage in a second power demand for the total of the third physical amounts from an external source to calculate a total cost for each set and obtaining, as an optimum solution, each value belonging to a set in which the total cost is minimized.

2. A utility optimizing method for calculating optimum operation parameters of a plant including a plurality of first plant elements having a linear input/output characteristic with first physical amounts as an input and second physical amounts as an output and a plurality of second plant elements having a non-convex input/output characteristic with the second physical amounts output from the first plant elements as an input and third physical amounts as an output, the method comprising:

a first step of calculating a division of the second physical amounts to the second plant elements by using a dynamic programming to maximize the total of the third physical amounts output from the second plant elements associated with the group corresponding to the total of the second physical amounts in a predetermined range satisfying a first demand for a group of second physical amounts;

a second step of calculating a division of the second physical amounts to the second plant elements by using a dynamic programming based on a set of the total of the second physical amounts, total of the third physical amounts and division of the second physical amounts to the second plant elements obtained by the first step at each of the second physical amounts in a group to maximize the total of the third physical amounts output from said all the second plant elements and corresponding to the total of the second physical amounts to all the second plant elements;

a third step of calculating, by using one of a linear programming or a nonlinear programming, the second physical amounts input to the respective second plant elements and a first cost required to supply the first physical amounts with respect to each set of the total of the second physical amounts, total of the third physical amounts and division of the second physical amounts to the second plant elements obtained by the second step; and

a fourth step of, based on each set of the total of the third physical amounts and first cost obtained at the third step, calculating a second cost required to receive a power which corresponds to a shortage in a second power demand for the total of the third physical amounts from an external source to calculate a total cost for each set and obtaining, as an optimum solution, each value belonging to a set in which the total cost is minimized.

3. The method according to claim 2, wherein when, in the fourth step, the total of the third physical amounts exceeds the second power demand, the second cost is made zero or the second cost is made to a negative value when the second cost is obtained by selling an excess of the total of the third physical amounts over the second power demand.

4. A utility optimizing method for calculating optimum operation parameters of a power generation plant including a plurality of boilers, a plurality of turbines operated by steams generated from the boilers and generators driven by the respective turbines to generate electric power in which the electric power is supplied to electric power loads while the steams generated from the respective turbines are supplied to steam loads via a plurality of turbine discharge systems, the method comprising:

a first step of calculating a steam division to the respective turbine by using a dynamic programming to maximize the total of the generation powers of the turbines connected to the same turbine discharge system corresponding to the steam amounts in a predetermined range satisfying a steam demand to the respective turbine discharge system;

a second step of calculating a steam division to the respective turbines by using the dynamic programming based on a set of the steam amount, generation output power and steam division obtained by the first step at each turbine discharge system to maximize a total generation power output from the turbines corresponding to a total input steam amount to all the turbines; and

a third step of calculating, by using one of a linear programming and a nonlinear programming, a supply fuel division to the respective boilers and total fuel cost with respect to each set of the total input steam amount, total generation output power and steam division to the respective turbines.

5. A utility optimizing method according to claim 4, further comprising:

a fourth step of, based on each set of a total generation output power and total fuel cost obtained by the third step, calculating an external cost required to receive a power which corresponds to a shortage in a power demand for the total generation output power from an outside source to calculate a total cost for each set and obtaining, as an optimum solution, each value belonging to a set in which the total cost is minimized.

6. The method according to claim 5 wherein when, in the fourth step, the total of the third physical amounts exceeds the second power demand, the second cost is made zero or the second cost is made to a negative value when the second cost is obtained by selling an excess of the total of the third physical amounts over the second power demand.

7. A plant utility optimizing system adapted to be connected to a power generation plant including boilers for generating steams, turbines driven by the steams generated from the boilers, a steam header for collecting the steams generated from the boilers, distributing steams by a given ratio and supplying distributed steams to the turbines, and electric power generators driven by the turbines, the plant utility optimizing system comprising:

turbine output maximum point calculating means for calculating a division ratio of the steam header to maximize a total output amount of the turbines with respect to a given total amount of steam, the calculating means using a dynamic programming scheme;

fuel cost minimizing point calculating means for calculating respective amounts of steams to be generated from each of the boilers at a minimum fuel cost based on the given total amount of steam and the division ratio of the steam header calculated by said turbine output maximum point calculating means, the calculating means using a linear programming scheme; and

total energy cost minimizing point calculating means for calculating a total energy cost minimum operation point based on an electric power demand, the given total amount of steam, the division ratio of the steam header calculated by said turbine output maximum point calculating means, and the amounts of steams to be generated from the boilers calculated by said fuel cost minimizing point calculating means.

8. The plant utility optimizing system according to claim 7, further comprising:

plant model creating means for creating/editing system graphics for indicating a model of the power generation plant; and

data generating means, based on the model, for generating data necessary for calculations by said turbine output maximum point calculating means, said fuel cost minimizing point calculating means, and said total energy cost minimizing point calculating means.