JPH08100607A - Optimization method for plant utility - Google Patents

Abstract

PURPOSE: To reduce operation cost by obtaining a candidate for the most suitable solution after having satisfied the first demand quantity using dynamic programming and linear programming which reflect nonconvex characteristic, and obtaining total cost in the case where the second demand quantity is satisfied for each candidate, and minimizing the total cost. CONSTITUTION: The steam exhaust quantity, in each turbine, to maximize total output electric power is obtained using dynamic programming in a fluid demand system optimizing section 21, and maintained by a list holding section 23. These data and steam-electric power demands are inputted into a fluid input system optimizing section 24, and the steam exhaust quantity, in each turbine, which maximizes the total output electric power to satisfy the steam-electric power demands is obtained by the dynamic programming, and a distributing ratio to minimize fuel cost is obtained using linear programming in a fluid generating system optimizing section 27. Next, operation parameter to maximize total electric output to each fuel cost after having satisfied each restricting condition and steam demand is obtained in a list retrieval section 28, and total cost on each line is computed for finding the line having the minimum total cost, and given to the most suitable solution output section 29.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a plant utility optimizing system for optimizing operating conditions of a plant system while satisfying various requirements for the plant system.

[0002]

2. Description of the Related Art A conventional plant utility optimization system takes the operating cost of a plant as a cost function, and sets the operating conditions of the plant to some equality or inequality constraints, and then minimizes the cost function. Nonlinear programming or linear programming was applied to. According to such a method, a sufficiently satisfactory solution can be obtained depending on the characteristics of the target plant.

By the way, the target plant may include an element having a non-convex characteristic. For example, the input / output characteristics of the turbine have non-convex characteristics called valve point characteristics as shown in FIG. This is due to the reduction in shaft output due to the pressure reduction that occurs when the steam input valve is in the half open state. Therefore, when the lower limit and the upper limit of the total output of the turbine group are given, the search range becomes a non-convex region.

However, the algorithm of the non-linear programming method is based on the assumption that the search area is convex, and there is a problem in the search of the non-convex area that not only the optimality of the solution but also the convergence of the algorithm is not guaranteed. .

Therefore, when using a non-linear programming method for such a plant, the starting point of the search (for example, the current operating state) is determined, and the domain of each parameter is set to an appropriate neighborhood of the searching start point. The local optimal solution can only be searched in the vicinity of the search starting point by the successive quadratic approximation method, etc., and it is difficult to obtain the optimal solution in the entire search range.

On the other hand, when applying the linear programming method to a plant including elements having non-convex characteristics, after linearly approximating the characteristics of the elements to be bought, for example, the input / output characteristics of the boiler or turbine, some linear constraint conditions are applied. It is possible to apply the simplex method or the like below. However, even if the non-convex characteristic is linearly approximated, it is extremely difficult to accurately reflect the original characteristic and solve the problem, and an optimal solution cannot be obtained.

[0007]

As described above, in the conventional plant utility optimization method based on the linear programming method, the optimization that accurately reflects the valve point characteristic of the turbine cannot be performed. Further, in the conventional plant utility optimization method based on the nonlinear programming method, only the local optimum solution can be searched, and it is difficult to obtain the optimum solution in the entire search range.

The present invention has been made in consideration of the above circumstances, and is a plant that satisfies various requirements for plant systems and that can find an optimum operating state that reflects the non-convex characteristics of plant elements. -The purpose is to provide a utility optimization method.

[0009]

SUMMARY OF THE INVENTION According to the present invention, a plurality of first plant elements each having a first physical quantity as an input and a second physical quantity as an output and having linear input / output characteristics, and these first plant elements are provided. Input the second physical quantity that is the output of
In the utility optimization method for searching for optimum operating parameters of a plant including a plurality of second plant elements each having an input / output characteristic including a non-convex characteristic portion as an output, a group of the second physical quantities The third physical quantity output by the second plant element associated with the group corresponding to each sum of the second physical quantities within a predetermined range that satisfy the first demand quantity for the group. A second for each second plant element that maximizes the sum of
A first processing step for determining the distribution of physical quantities of the second physical quantity by using a dynamic programming method, and a total sum of the second physical quantities obtained for each second physical quantity of the group in the first processing step,
Based on the set of the sum total of the third physical quantities and the allocation of the second physical quantity to each of the second plant elements, all corresponding to the sum totals of the second physical quantities to all the second plant elements, A second processing step for obtaining a distribution of the second physical quantity to each of the second plant elements that maximizes the sum of the third physical quantities output by the second plant element of For each set of the total sum of the second physical quantity, the total sum of the third physical quantity, and the distribution of the second physical quantity to each of the second plant elements obtained in the second processing step, A second processing step for determining the first physical quantity to be input to each of the two plant elements and the first cost required to supply the first physical quantity by using a linear programming method or a non-linear programming method; Three The third obtained in process step
It is necessary to externally supply a shortage to the second demand amount given in advance of the total sum of the third physical quantities based on the set of the total sum of the physical quantities and the first cost required for the first physical quantity. A fourth processing step of calculating a second cost to obtain a total cost for each set, and determining each value belonging to the set that minimizes the total cost as an optimum solution.

Preferably, in the fourth processing step, when the sum of the third physical quantities exceeds the second demand quantity, the second cost is set to 0 or the third physical quantity The second cost, which is obtained by supplying to the outside the sum of which exceeds the second demand amount, may be calculated as a negative value.

Further, the present invention is a utility optimization method for searching optimum operating parameters of a plant for driving a plurality of turbines by steam generated by a plurality of boilers to generate electric power, in each turbine exhaust system. The steam flow rate for each turbine that maximizes the total output power of the turbines connected to the same turbine exhaust system, corresponding to each steam flow rate within a predetermined range that satisfies the steam demand amount for each system. Based on the first processing step for obtaining the distribution using the dynamic programming method and the set of steam flow rate, generated output power and steam flow rate distribution obtained for each turbine exhaust system in this first processing step, Corresponding to the total input steam flow to all turbines, the steam flow distribution to each turbine is maximized to maximize the total output power generated by all turbines. For each of the boilers for the second processing step obtained by using the dynamic programming method and for each set of the total input steam flow rate, the total generated output power, and the steam flow rate distribution to each turbine obtained in the second processing step. To the third processing step for obtaining the supply fuel distribution and the total fuel cost to the vehicle using the linear programming method or the non-linear programming method, and the total generated output power and the total fuel cost obtained in the third processing step, respectively. Based on this, the external cost required to externally supply the power shortage with respect to the given power demand amount of the total generated output power is calculated to obtain the total cost for each set, and the total cost is set to the set that minimizes this total cost. A fourth processing step of determining each belonging value as an optimum solution.

Preferably, in the fourth processing step, when the total generated output power exceeds the power demand amount, the external cost is set to 0 or the total generated output power exceeds the power demand amount. It is also possible to calculate the external cost obtained by supplying the power corresponding to the external power as a negative value.

[0013]

In the present invention, a plurality of first plant elements (for example, energy fluid generation) having a first physical quantity (for example, fuel) as an input and a second physical quantity (for example, steam quantity) as an output and each having linear input / output characteristics are provided. Device) and a plurality of second physical quantities each having an input / output characteristic including a second physical quantity that is an output of these first plant elements as an input and a third physical quantity (for example, output power) as an output. When searching for optimum operating parameters of a plant (for example, a turbine power generation system) including plant elements (for example, a turbine),
First, the dynamic programming that can reflect the non-convex characteristic is performed twice to satisfy the first demand amount (for example, vapor demand amount) that is the requirement for the second physical quantity, and then various second If the total amount of physical quantities is assumed, the total amount of the second physical amount, the distribution amount of this total amount to all the second plant elements, and the second amount output by all the second plant elements. The set of the total amount of the physical quantities of 3 is obtained. Then, based on the values of this determined pair,
For each set, the first physical quantity input to each second plant element and the first cost required to supply these first physical quantities are obtained using linear programming or nonlinear programming. Finally, based on the value of the set of the total sum of the third physical quantities thus obtained and the first cost (for example, the total fuel cost) required for the first physical quantity, the third value is set for each group.
If the sum of the physical quantities of is smaller than the second demand amount (for example, power demand amount), the second cost (for example, external cost) required to supply the shortfall from the outside is calculated, and the total cost for each group is calculated. And each value belonging to the set that minimizes this total cost (for example, the sum of the second physical quantities, the distribution of the second physical quantity to each of the second plant elements, the sum of the third physical quantities, the second plant). The first physical quantity input to each element, the first cost required to supply the first physical quantity, the total cost, etc.) can be determined as the optimum solution.

As described above, according to the plant utility optimizing system of the present invention, it is possible to find various optimum conditions in the plant system and to find an optimum operating state that reflects the non-convex characteristics of plant elements such as turbines. You can Therefore, the operation cost can be significantly reduced as compared with the conventional case.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. First, the present embodiment will be schematically described. In this embodiment, an in-house power generation system will be described as an example of a plant system targeted by the plant utility optimization system. This private power generation system reuses the discharged steam, and there is a demand for the discharged steam amount and the output power amount.

The plant utility optimization system of the present embodiment utilizes dynamic programming to satisfy the steam demand, the power demand, and various constraint conditions while minimizing the total cost required for the plant system. It is to find out the state.

By the way, the above-mentioned dynamic programming is a problem in which the decision is generally made in multiple stages, and the result is prescribed at each stage, and the result is the premise of the decision problem in the subsequent stages. It is one of the solutions to the decision problem that is applied to such problems. In dynamic programming, a sequence of feasible decision choices is called a "policy", and a certain function (criteri) of the final state variable is called.
A policy that maximizes on function is called an "optimal policy." And dynamic programming consistently states that "optimal policy is such that, regardless of the initial state and the initial decision, the subsequent decision sequence gives the optimal policy for the state resulting from this initial decision. It has the property that it must be. ”The optimality principle (principle o
f optimality) is used. Regarding dynamic programming, for example, “Dynamic programming” written by R. Bellman.
g, Princeton University P
Press, 1957 ".

In an actual plant as dealt with in the present embodiment, an optimum solution must be found under a complicated constraint condition because of steam reuse, the presence of a boiler group, and the existence of various steam systems. Therefore, it is impossible to derive the optimum solution by the conventional dynamic programming method.

In the plant utility optimization system of this embodiment, in addition to the optimization means of the turbine group output, various optimization means are sequentially and organically combined according to the configuration of the plant, thereby making it possible to obtain complicated requirements. To find the optimum operating condition of the plant that satisfies the above conditions. In general, first, a process that uses dynamic programming is performed twice and a process that uses linear programming is performed once to optimize the output of the turbine group for each operating state (that is, a candidate for the optimal solution). ) Is calculated, and then the operating condition that minimizes the total cost is determined as the optimum solution.

The plant utility optimization system of this embodiment will be described in more detail below. First, FIG.
The outline of the plant / energy system according to the present embodiment is shown in FIG. As shown in the figure, a private power generator 11 and an external power input device 1 using a BTG (boiler, turbine, generator)
2 and a steam system 13, a plant system 6, a plant control device 4 for controlling these, a plant control device 4 connected to the plant control device 4, and obtaining optimum operating parameters of each component of the plant system 6 and transferring them to the plant control device 4. It is composed of the plant utility optimization system 2. The plant utility optimizing system 2 may input / output data to / from the plant utility optimizing system 2 using an information storage medium such as a magnetic disk, without connecting to the plant system 6.

FIG. 2 shows an example of a schematic processing flow relating to the plant utility optimization system 2.
The detailed system configuration information of the plant system 6 is input to the plant utility optimization system 2 (step S11).

The power demand / steam demand is determined, and the values are input to the plant controller 4 (step S12). The plant control device 4 sends power demand / steam demand to the plant utility optimization system 2 and requests execution of optimization processing (step S13).

The plant utility optimization system 2 searches for an optimum solution (step S14). The plant utility optimization system 2 transmits the searched optimum solution to the plant control device 4 (step S
15).

The plant controller 4 controls the plant system 6 based on the given optimum solution (step S16). FIG. 3 shows the plant system 6 in FIG.
It is a block diagram showing in more detail.

In this embodiment, four steam generating boilers (hereinafter referred to as boilers) B1 to B4, five steam turbines (hereinafter referred to as turbines) T1 to T5, and five generators G are provided. I have it.

Each of the boilers B1 to B4 generates steam by using the supplied fuel. The amounts of fuel supplied to the boilers B1 to B4 are f1 to f4, respectively, and the amounts of steam generated by the boilers B1 to B4 are S1 to S4, respectively.

L1 and L2 are steam systems having different pressures (L1 is on the high pressure side). Also, S5
And S6 are the input steam amount and the output steam amount of the pressure reducing device 20, respectively.

Each of the turbines T1 to T5 has an input steam amount-output power characteristic function p1, p2, p3, p4, p5.
And generate electric power according to the amount of input steam. The amounts of steam input to the turbines T1 to T5 are s1 to s5, respectively.

The exhaust steam from each of the turbines T1 to T5 is connected to steam systems L3 and L4 having different pressures. Then, the steam amount S1 ′ is output from the steam system L3, and the steam amount S2 ′ is output from the steam system L4.

On the other hand, the electric power generated in each of the turbines T1 to T5 is combined and output. Where each boiler B
Upper limit values and lower limit values are set for the fuel amounts f1 to f4 injected into 1 to B4, respectively. In addition, restrictions on the upper limit value and the lower limit value of the fuel input amount based on the amount of fuel currently burning in each boiler B1 to B4 may be added.

Input steam quantity s1 to turbines T1 to T5
An upper limit value and a lower limit value are determined for each of s5. Regarding such a plant system 6, steam demand SD for each steam system L3, L4
1, SD2 (minimum value that S1 ′ should satisfy and minimum value that S2 ′ should satisfy), and the power demand ED are determined. When the power demand cannot be satisfied with the total amount of power generation by all the boilers B1 to B4, power is purchased from the outside to compensate. If the total amount of power generated by all the boilers B1 to B4 exceeds the power demand, measures such as discarding the excess or selling the power are taken.

In the plant system 6 of this embodiment, the turbines T1 to T5 are assumed to have no bleed air. However, regarding the plant system including the bleed turbine, the bleed turbine is not a plurality of bleed turbines. Can be reduced to a plant / system configuration that does not include the extraction turbine as in FIG.

FIG. 4 shows a schematic configuration of a main part of a plant utility optimization system 2 according to an embodiment of the present invention. As shown in the figure, the plant utility optimization system of the present embodiment includes a fluid demand system optimization unit 21, a list conversion unit 22, a list holding unit 23, a fluid input system optimization unit 24, and a power / fluid demand input unit 25. ,plant·
The data holding unit 26, the fluid generation optimization unit 27, the list search unit 28, and the optimum solution output unit 29 are provided.

The plant data holding unit 26 stores the upper and lower limit values of the fuel amounts f1 to f4 input to the boilers B1 to B4, and the input steam amount (steam discharge amount) s1 to s5 of the turbines T1 to T5. Upper and lower limits, each turbine T
Input steam amount-output power characteristic functions p1, p2 of 1 to T5
p3, p4, p5, and other necessary constants and functions such as input / output characteristic functions of the boilers B1 to B4 are stored. As will be described later, the plant / data storage unit 26 provides the numerical values required for the fluid demand system optimization unit 21, the list conversion unit 22, the fluid input system optimization unit 24, and the fluid generation system optimization unit 27. Give a function.

First, the fluid demand system optimizing unit 21
In the case where the steam flow rate of the steam system L3 is assumed to be each value, each turbine T1, so as to maximize the total output power of the turbines T1, T2, T4 succeeding to the steam system L3 by using the dynamic programming method. T2 and T4 steam discharges s1 and s
2, s4 is determined.

The fluid demand system optimizing unit 21 in this case
An example of the operation of is shown in FIGS. 5 and 6. 5 and 6,
p1, p2, ..., p5 are characteristic functions of each turbine,
M1, M2 and M4 are upper limit values of the exhaust steam amount of each turbine T1, T2 and T4. To simplify the explanation, M
1, M2 and M4 are integers, and each turbine T
It is assumed that the lower limit values of the exhaust steam amounts of 1, T2 and T4 are all 0. Further, T and T'represent an array, and T and T '
The first [] after the is a row, the next [] is a column, and the next [0] [1] [2] is the first to third elements in the row / column.

Here, the steam flow rate is changed from 0 to M1 + M2 +.
It is calculated for each of the cases of assuming integer values up to M4. By the operation shown in FIGS. 5 and 6, the table a as shown in FIG. 7 is obtained.

Similarly, the fluid demand system optimizing unit 2
In the case where the steam flow rate of the steam system L4 is assumed to be each value, 1 uses the dynamic programming method to maximize the total output power of the turbines T3 and T5 succeeding the steam system L4.
The steam discharge amounts s3, s5 of the turbines T3, T5 are determined.

The fluid demand system optimizing section 21 in this case
The operation of, for example, M1 and M2 of FIG.
5, and if "no" in step S27, "to A" is corrected to end.

Here again, the calculation will be made for the case where the vapor flow rate is assumed to be an integer value between 0 and M3 + M5. By this operation, for each steam flow rate of the steam system L4, a table b as shown in FIG. 8 in which the optimum steam distribution and output power of the turbines T3 and T5 are described is obtained.

These tables a and b are sent to the next list conversion unit 22. Here, the boilers B1 to B4 to L
The amount of steam supplied to the 1 system and the L2 system is σ1,
Let σ2. That is, σ1 = S1 + S2 σ2 = S3 + S4.

Therefore, the following relational expression holds. s1 + s2 + s3 = σ1−S5 s4 + s5 = σ2 + S6 Here, a linear relationship between S5 and S6, S5 = λ · S6
Therefore, the equation u + v = σ holds.

Here, u = s1 + s2 + λ · s4 v = s3 + λ · s5 σ = σ1 + λ · σ2. σ is the total amount of generated steam converted by high-pressure steam.

The list conversion unit 22 adds the value of u to each column of the table a, adds the value of v to the table b, and adds the table c as shown in FIG. 9 and the table d as shown in FIG. To create.
The value of u _{M in} Table c is the maximum time value that does not exceed M1 + M2 + λ · M4, and the value of v _{M in} Table d is M3 + λ · M.
It is the maximum tick value that does not exceed 5. The list conversion unit 22
The table c and the table d are sent to the list holding unit 23.

By the way, the values of u and v calculated from the respective amounts of steam of the steam system L3 and the respective amounts of steam of the steam system L4 are simply added to the tables a and b to prepare the tables c and d. Therefore, the values of u and v are not necessarily arranged at equal intervals.

Therefore, in order to make the pitch of the value of u and the value of v constant, the list conversion unit 22 determines the pitch of the value of u and the value of v, and for each value of u and v. It is preferable to appropriately interpolate the amount of steam, the optimum steam distribution, and the output voltage based on the values in Table a and Table b. As the interpolation method, for example, the following method can be considered. One is that when the total vapor amount corresponding to the value of u or the value of v is present in the rows of the vapor amounts of L3 and L4 in Tables a and b, the column is adopted as it is. This is a method of adopting the minimum value sequence that is equal to or exceeds the vapor amount. In other methods, if the total steam amount corresponding to the value of u or the value of v is present in Table a or Table b, that column is adopted as it is. This is a method of linear interpolation using a sequence of values and a sequence of maximum values less than the total amount of steam.

The list holding unit 23 stores the given tables c and d. When the plant / system is determined, the processing up to this point can be performed.

Next, the power demand and steam demand given from the plant controller 4 are sent from the power / fluid demand input unit 25 to the fluid input system optimization unit 24. When the fluid input system optimizing unit 24 receives from the list holding unit 23 the tables c and d already created as described above, from the table c, the steam amount of the steam system L3 is L.
Cut out the rows that satisfy the vapor demand SD1 for the three systems. Similarly, a row satisfying the vapor demand SD2 for the L4 system is cut out from Table d.

By this operation, the table e as shown in FIG. 11 is obtained.
Then, a table f as shown in FIG. 12 is obtained. However, in Table e, L3l and L3u are the lower limit value and the upper limit value of the L3 steam amount in Table c that satisfy the steam demand for the L3 system. In addition, Um, (sm1, sm2, sm
4) and Em3 are u, (s1, s) of the L3l column, respectively.
2, s4) and the output power value. UM, (sM
1, sM2, sM4), and EM3.
The meaning of each symbol in Table f is the same as that in Table e.

Next, the fluid input system optimizing section 24 uses the dynamic programming method for the case where the total input steam amount in high pressure conversion is assumed to be each value based on Tables e and f. The steam discharge amounts s1 to s5 of the turbines T1 to T5 are determined so as to maximize the total output power of the turbines T1 to T5.

The tables e and f are regarded as 4 × l1 and 4 × l2 two-dimensional arrays, and the arrays are referred to as A and B. However, l1 and l2 are the lengths of the table e and the table f, respectively. For example, A [0] [l1-1] is UM.

The operation of the fluid input system optimizing section 24 will be described with reference to FIG.
Shown in In FIG. 13, T represents an array, the first [] after T is a row, the next [] is a column, the next [0] [1] is the first element in the row / column, and Represents two elements.

In FIG. 13, A [2] [j] + B
The expression [2] [k] indicates an operation of merging the list B [2] [k] with the list A [2] [j]. For example, when the elements of the list A [2] [j] are (a1, a2, a4) and the elements of the list B [2] [k] are (a3, a5), the result of the merge operation is A [2 ] [J] + B [2]
The elements of [k] are (a1, a2, a3, a4, a5).

By the operation of FIG. 13, a table g as shown in FIG. 14 is obtained as the array T. This table g is sent to the fluid generation optimization unit 27. Here, in Table g, since the steam distribution (s1, ..., S5) that maximizes the total power output is obtained for each column (that is, each total input steam amount), L1 for each total input steam amount is obtained. , L2, the amounts of steam σ1, σ2 flowing through each system are uniquely determined.

The fluid generating system optimizing unit 27 satisfies the lower limit value and the upper limit value of the boiler combustion fuel amounts f1 to f4 for each column of Table g, that is, each total input steam amount, and each of L1 and L2. Under the condition that steam of σ1 and σ2 is supplied to the system,
Find fuel allocations that minimize fuel costs. The boiler characteristics of the boilers B1 to B4 are all linear functions,
Since the fuel cost is also a linear function of fuel, the above fuel allocation and fuel cost can be obtained by using a linear programming method or a non-linear programming method.

Thus, for each column of table g, S1
Each value of S6 and f1 to f4 is obtained. The fluid generation system optimizing unit 27 adds these values to the table g and
A table h as shown in FIG.

This table h is sent to the list search unit 28. At this point, it is possible to obtain an operating parameter that maximizes the total electric power output for each fuel cost, while satisfying each constraint condition and steam demand. That is, it can be said that each column is a candidate for the optimum solution to be obtained.

The list retrieving unit 28 makes the table h as follows.
The optimal solution is determined from among the optimal solution candidates of. First, the list search unit 28 calculates the total cost for each column, that is, each optimal solution candidate, based on the values in the table h.

The total cost is determined, for example, by the following formula. (Total cost) = (Total fuel cost) + e (ED- (Total output power)) Here, e is the unit price of power purchase and ED is the power demand amount.

In this case, it is premised that the total output power is less than the power demand amount. However, if the total output power exceeds the power demand amount, for example, (ED- (total output power)) is set to 0. good. Alternatively, if it is possible to sell the amount exceeding the power demand amount, let e ′ be the selling price, and if ED ≧ total output power, (total cost) = (total fuel cost) + e (ED− (total Output power)), and when ED <total output power, it may be calculated as (total cost) = (total fuel cost) + e ′ (ED− (total output power)).

The list search unit 28 adds the calculated total cost to each column of the table h to create a table i as shown in FIG. Then, from Table i, a column with the minimum total cost is found, and a set of various values forming this column is given to the optimal solution output unit 29 as an optimal solution for the entire system.

The optimum solution output unit 29 holds the optimum solution sent from the list search unit 28. Then, appropriately, necessary information among the values included in the optimum solution is transmitted to the plant control device 4 as the optimum operation parameter.

As described above, according to the plant utility optimizing system of the present embodiment, it is possible to find various optimum conditions in the plant system and to find the optimum operating state reflecting the non-convex characteristics of the turbine. Therefore, the operation cost can be significantly reduced as compared with the conventional case.

In the present embodiment, the operation of the plant utility optimization system according to the plant system of FIG. 3 has been described, but the present invention can be applied even if the number of boilers and turbines and the number of types of steam systems are changed. It is possible to apply. Furthermore, even if other elements are added to the plant system of FIG. 3, an optimum solution can be obtained by appropriately modifying the plant utility optimization system. Further, the present invention can be generally applied to not only the private power generation system as in the present embodiment but also a plant system including an element having a non-convex characteristic. Further, the present invention is not limited to the above-described embodiments, and various modifications can be carried out without departing from the scope of the invention.

[0065]

According to the plant utility optimization system of the present invention, the first demand amount is satisfied by the process of using the dynamic programming method capable of reflecting the non-convex characteristic twice and the linear programming method once. In the above, after obtaining the optimal solution candidate, the total cost when the second demand amount is satisfied is obtained for each optimal solution candidate, and the optimal solution candidate that minimizes this is determined as the optimal solution. Therefore, it is possible to find an optimum operating condition that satisfies various requirements in the plant system and that reflects the non-convex characteristics of the plant element such as the turbine. Therefore, compared to the conventional system, the operating cost can be significantly reduced.

[Brief description of drawings]

FIG. 1 is a diagram showing an outline of a plant / energy system according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a schematic process flow of a plant utility optimization system according to the embodiment.

FIG. 3 is a diagram showing a configuration of main parts of a plant system according to the embodiment.

FIG. 4 is a diagram showing a schematic configuration of main parts of a plant utility optimization system according to the embodiment.

FIG. 5 is a flowchart showing an example of the operation of a fluid demand system optimizing unit.

FIG. 6 is a flowchart showing an example of the operation of a fluid demand system optimizing unit.

FIG. 7 is a diagram showing a relationship between optimal steam distribution and output power for each steam amount.

FIG. 8 is a diagram showing a relationship between optimal steam distribution and output power for each steam amount.

FIG. 9 is a diagram showing a relationship between optimal steam distribution and output power for each steam amount.

FIG. 10 is a diagram showing a relationship between optimal steam distribution and output power for each steam amount.

FIG. 11 is a diagram showing a relationship between optimal steam distribution and output power for each steam amount.

FIG. 12 is a diagram showing a relationship between optimal steam distribution and output power for each steam amount.

FIG. 13 is a flowchart showing an example of the operation of the fluid input system optimizing unit.

FIG. 14 is a diagram showing the relationship between the optimum steam distribution and the total output power with respect to the total input steam amount.

FIG. 15 is a diagram showing a relationship among optimum steam distribution, total output power, fuel distribution, and total fuel cost with respect to total input steam amount.

FIG. 16 is a diagram showing each optimal solution candidate.

FIG. 17 is a diagram showing input / output characteristics of the turbine.

[Explanation of symbols]

2 ... Plant utility optimization system, 4 ... Plant control device, 6 ... Plant system, 11 ... Private power generation device, 12 ... External power input device, 13 ... Steam system,
21 ... Fluid demand system optimization unit, 22 ... List conversion unit, 23 ... List holding unit, 24 ... Fluid input system optimization unit,
25 ... Power / fluid demand input section, 26 ... Plant
Data holding unit, 27 ... Fluid generation optimization unit, 28 ... List search unit, 29 ... Optimal solution output unit, B1 to B4 ... Steam generation boiler, T1 to T5 ... Steam turbine, G ... Generator, ... Steam system L1, L2, 20 ... Pressure reducing device

Claims (4)

[Claims] 1. A plurality of first plant elements each having a first physical quantity as an input, a second physical quantity as an output, and linear input / output characteristics, and a second physical quantity which is an output of these first plant elements. , A third physical quantity is output, and a plurality of second plant elements each having an input / output characteristic including a non-convex characteristic portion are included in the utility optimization method for searching an optimum operating parameter of the plant. For each of the second physical quantities, corresponding to each sum of the second physical quantities within a predetermined range that satisfies the first demand quantity for the group, output of the second plant element related to the group. A first processing step of obtaining a distribution of the second physical quantity to each of the second plant elements that maximizes the total sum of the third physical quantity by using a dynamic programming method; Based on the set of the sum of the second physical quantities, the sum of the third physical quantities, and the distribution of the second physical quantities for each of the second plant elements, which are obtained for each group of the second physical quantities. Corresponding to each sum of the second physical quantities to the second plant element,
A second processing step of obtaining a distribution of the second physical quantity to each of the second plant elements that maximizes the sum of the third physical quantities output by all the second plant elements, using dynamic programming; For each set of the total sum of the second physical quantity, the total sum of the third physical quantity, and the distribution of the second physical quantity to each of the second plant elements, obtained in the second processing step, A third processing step of obtaining the first physical quantity input to each of the second plant elements and the first cost required to supply the first physical quantity by using a linear programming method or a non-linear programming method; Based on the set of the total sum of the third physical quantity and the first cost required for the first physical quantity obtained in the third processing step, the second predetermined sum of the total sum of the third physical quantity is given. Demand A fourth processing step of calculating a second cost required to supply the shortfall from the outside to obtain a total cost for each set, and determining each value belonging to the set that minimizes the total cost as an optimum solution. A method for optimizing a plant utility, comprising: 2. In the fourth processing step, when the sum of the third physical quantities exceeds the second demand quantity, the second cost is set to 0 or the sum of the third physical quantities. 2. The plant utility optimization method according to claim 1, wherein the second utility cost is calculated as a negative value by calculating a second cost obtained by supplying an amount exceeding the second demand amount to the outside. 3. A utility optimization method for searching optimum operating parameters of a plant that drives a plurality of turbines with steam generated by a plurality of boilers to generate electric power, the method comprising: The steam flow distribution to each turbine is maximized to maximize the total output power of the turbines connected to the same turbine exhaust system, corresponding to each steam flow within the specified range that satisfies the steam demand to the system. Based on the set of the steam flow rate, the generated output power, and the steam flow rate distribution obtained for each turbine exhaust system in the first process step obtained by using the planning method, Dynamic programming is used to allocate steam flow to each turbine to maximize total output power generated by all turbines, corresponding to each total input steam flow. The second processing step obtained by the above, and the distribution of the supply fuel to each boiler and the set of the total input steam flow rate, the total generated output power, and the steam flow rate distribution to each turbine obtained in this second processing step. Based on the third processing step of obtaining the total fuel cost by using the linear programming method or the non-linear programming method, and the total output power and the total fuel cost obtained in the third processing step, respectively, the total generated output is calculated. Calculate the external cost required to supply the power shortage for the given power demand amount from the outside, find the total cost for each set, and optimize each value that belongs to the set that minimizes this total cost A fourth utility step of determining a solution, and a method for optimizing a plant utility. 4. In the fourth processing step, if the total generated output power exceeds the power demand amount, the external cost is set to 0, or the total generated output power exceeds the power demand amount. The plant-utility optimizing method according to claim 3, wherein the external cost obtained by supplying the electric power for the outside to the outside is calculated as a negative value.