WO2023095389A1

WO2023095389A1 - Tank operation plan derivation system and tank operation plan derivation method

Info

Publication number: WO2023095389A1
Application number: PCT/JP2022/029773
Authority: WO
Inventors: 孔一郎手塚; 雅之口井; 康太渡邊
Original assignee: 大阪瓦斯株式会社
Priority date: 2021-11-24
Filing date: 2022-08-03
Publication date: 2023-06-01
Also published as: JP2023077351A; TW202326580A; JP7121180B1

Abstract

The present invention provides a tank operation plan derivation system and a tank operation plan derivation method in which burden of computing processes is reduced. A tank operation plan derivation system S is provided with a storage unit 1 and a computing processing unit 2. The computing processing unit 2 executes, for each of a plurality of tanks, a first process that derivates an operation plan by using the calorific value or density of a stored liquefied fuel as a constant or a linearly changing variable and thus solving a mixed integer linear programming problem and a second process that calculates a change in the calorific value or density of a liquefied fuel stored in each of the plurality of tanks, in the operation plan derived in the first process. When re-executing the first process after the second process, the computing processing unit 2 uses, as a constant or a linearly changing variable, the change in the calorific value or density of a liquefied fuel stored in each of the plurality of tanks calculated in the second process and thus executes the first process.

Description

Tank operation plan derivation system and tank operation plan derivation method

The present invention provides a tank operation plan derivation system and tank operation for deriving an operation plan for a tank storing liquefied fuel (for example, fuel stored in a liquid state such as liquefied natural gas, propane, hydrogen, and ammonia; the same shall apply hereinafter). It relates to a plan derivation method.

Mathematical programming is known as a method of formulating plans for efficient delivery of goods and operation of equipment, in which a problem is reduced to a mathematical expression to find a solution. Examples of problems solved by mathematical programming include mixed integer programming problems in which decision variables include both continuous and discrete variables, linear programming problems in which constraints and objective functions are linear, and linear programming problems in which constraints and objective functions are linear. There are nonlinear programming problems expressed by arbitrary continuous functions that are not

However, when using mathematical programming to solve the operation plan for tanks that store liquefied fuel, it is necessary to solve sophisticated and complex problems, so there is the problem of an enormous amount of computation. Specifically, the calorific value of liquefied fuel not only varies depending on the place of production and the environment during transportation, but also varies depending on the reception and transfer, etc. The calorific value is expressed by a complex nonlinear formula. be done. Furthermore, the increase or decrease of the liquefied fuel stored in the tank is a continuous variable, but the variable of whether or not the tank receives or expels the liquefied fuel is a binary discrete variable, and both are mixed. In this way, in order to solve the operation plan of the tank that stores the liquefied fuel by the mathematical programming method, it is necessary to solve a highly complex mixed-integer non-linear programming problem, resulting in an enormous amount of computation.

In this regard, in Patent Document 1, when solving an operation plan for a liquefied fuel tank by mathematical programming, a mixed integer programming problem in which nonlinearity is linearly approximated and a nonlinear programming problem in which discrete variables are relaxed to continuous variables are A system has been proposed in which the computational load is reduced by alternately obtaining the solution two or more times.

JP 2013-092162 A

However, in the system proposed in Patent Document 1, although the load of arithmetic processing is alleviated by treating discrete variables as continuous variables, it is necessary to solve the nonlinear programming problem two or more times, so the arithmetic processing is slowed down. The effect of reducing the burden is limited.

The present invention has been made to solve the above problems, and provides a tank operation plan derivation system and a tank operation plan derivation method that reduce the burden of arithmetic processing.

In order to achieve the above object, the tank operation plan derivation system disclosed below provides information on the inflow and outflow of the liquefied fuel with respect to all of the plurality of tanks storing the liquefied fuel and information on the state of each of the plurality of tanks. a storage unit for storing input information and constraint information including constraint conditions for inflow/outflow and storage of the liquefied fuel in each of the plurality of tanks; and based on the input information and the constraint information, the plurality of a calculation processing unit for deriving an operation plan, which is a plan for the inflow and outflow of the liquefied fuel in each of the plurality of tanks, wherein the calculation processing unit calculates the heat quantity or density of the stored liquefied fuel for each of the plurality of tanks A first process for deriving the operation plan by solving a mixed-integer linear programming problem assuming that is a constant or a linearly changing variable, and each of the plurality of tanks in the operation plan derived in the first process and a second process of calculating the transition of the calorie or density of the stored liquefied fuel, and when the first process is re-executed after the second process, the plurality of values calculated in the second process The transition of the calorific value or density of the liquefied fuel stored in each tank is regarded as a constant, and the first process is re-executed.

According to the tank operation plan derivation system described above, the heat quantity or density of the liquefied fuel stored in the nonlinearly fluctuating tank is not treated as a variable in the mathematical programming problem and is calculated separately, thereby solving the mixed integer linear programming problem. to derive the operation plan. Therefore, according to the above-described tank operation plan derivation system, it is possible to reduce the load of arithmetic processing.

FIG. 1 is a schematic diagram showing a simplified example of the configuration of a storage facility including tanks. FIG. 2 is a block diagram showing a schematic configuration of this system. FIG. 3 is a table showing the principal variables and constants used in solving the mixed integer linear programming problem for deriving the operational plan for tank 10. As shown in FIG. FIG. 4 is a flow chart showing an outline of arithmetic processing by the arithmetic processing unit 2 of the system S. As shown in FIG. FIG. 5 is a flow chart showing details of the arithmetic processing of step #1 by the arithmetic processing section 2 of the system S. As shown in FIG. FIG. 6 is a table showing an example of a ship allocation plan. FIG. 7 is a table showing an example of tank specifications. FIG. 8 is a table showing an example of line demand. FIG. 9 is a table showing an example of acceptance patterns. FIG. 10 is a flow chart showing details of the arithmetic processing of step #2 by the arithmetic processing section 2 of the system S according to the first embodiment. FIG. 11 is a table showing an example of a transfer table. FIG. 12 is a flow chart showing details of the arithmetic processing of step #3 by the arithmetic processing section 2 of the system S according to the first embodiment. FIG. 13 is a flow chart showing details of the arithmetic processing of step #2 by the arithmetic processing unit 2 of the system S according to the second embodiment. FIG. 14 is a flow chart showing details of the arithmetic processing of step #3 by the arithmetic processing unit 2 of the system S according to the second embodiment. FIG. 15 is a flow chart showing details of the arithmetic processing of step #2 by the arithmetic processing unit 2 of the system S according to the third embodiment. FIG. 16 is a flow chart showing details of the arithmetic processing of step #3 by the arithmetic processing unit 2 of the system S according to the third embodiment. FIG. 17 is a flow chart showing details of the arithmetic processing of step #2 by the arithmetic processing unit 2 of the system S according to the fourth embodiment. FIG. 18 is a flow chart showing details of the arithmetic processing of step #3 by the arithmetic processing unit 2 of the system S according to the fourth embodiment.

An embodiment of the present invention will be described below based on the drawings. The present invention is not limited to the following embodiments, and design changes can be made as appropriate within the scope of satisfying the configuration of the present invention. Further, in the following description, the same reference numerals are used in common for the same parts or parts having similar functions in different drawings, and the repeated description thereof will be omitted. Also, each configuration described in the embodiment and modification may be appropriately combined or changed. Also, in order to make the explanation easier to understand, in the drawings referred to below, the configuration is shown in a simplified or schematic form, or some constituent members are omitted.

<<Overall composition, etc.>>
First, the tank operation plan derivation system according to the embodiment of the present invention (hereinafter referred to as "this system") will describe the configuration of the tank, etc. for which the operation plan is derived, as well as the details of various operations performed on the tank. , variables and constraints used in calculations are explained. In the following description, a tank storing liquefied natural gas (LNG) will be described as an example, but the system can also derive an operation plan for tanks storing liquefied fuels other than LNG.

Fig. 1 is a schematic diagram showing a simplified example of the configuration of a storage facility including tanks. As shown in FIG. 1, LNG transported from a production site or other storage facility or the like is supplied to an arbitrary tank 10 by a transportation means 11 such as an LNG tanker. This operation is called "accept". Also, as shown in FIG. 1, a portion of the LNG stored in any tank 10 is supplied to another tank 10 via transfer line 12 . This operation is called "transfer". Also, as shown in FIG. 1, LNG stored in an arbitrary tank 10 is supplied for demand such as city gas and power generation via a payout line 14 associated with the tank 10. This operation is called "payout".

Further, as shown in FIG. 1, the tank 10 includes a transfer pump 13 for transferring and a dispensing pump 15 for dispensing. A plurality of these pumps (for example, two each) may be provided for one tank 10 so that transfer or dispensing can be performed even if one of them becomes unusable due to failure or the like. In addition, for the purpose of maintaining the inside of the pipes through which LNG flows in a cryogenic state, "cooling" can also be performed in which LNG is sent to the transfer line 12 or the discharge line 14 and collected in the tank 10. Incidentally, when deriving an operation plan in this system, cooling may be treated as an operation separate from transfer or payout, or may be treated as a part of the transfer or payout operation. Note that FIG. 1 shows a simplified configuration of the storage facility, and the actual storage facility may be more complicated than shown in FIG. For example, in FIG. 1, one tank 10 and one payout pump 15 are shown to be able to pay out only one payout line 14, but they may be able to pay out to a plurality of payout lines. Also, for example, FIG. 1 shows lines and equipment for cooling, and lines and equipment for compressing and liquefying BOG (Boil Off Gas) generated by vaporizing LNG in the tank 10. omitted.

FIG. 2 is a block diagram showing the schematic configuration of this system. As shown in FIG. 2, the system S includes a storage unit 1 that stores input information and constraint information, and an arithmetic processing unit 2 that derives an operation plan for the tank 10 based on the input information and constraint information. The storage unit 1 includes a non-volatile storage device such as a HDD (Hard Disk Drive) or an SSD (Solid State Drive), and a volatile storage device such as a RAM (Random Access memory). The arithmetic processing unit 2 is composed of an arithmetic processing device such as a CPU (Central Processing Unit). Note that the storage unit 1 may be configured by a plurality of storage devices, and may be configured by combining a nonvolatile storage device and a volatile storage device, for example. The storage unit 1 may be configured as part of a server or the like located away from the arithmetic processing unit 2, and may exchange information with the arithmetic processing unit 2 via a network such as the Internet. Moreover, the storage unit 1 and the arithmetic processing unit 2 may be configured as a part of one personal computer.

Based on the input information and the constraint information stored in the storage unit 1, the arithmetic processing unit 2 plans the input and output of LNG in each of the plurality of tanks 10 for a predetermined planning target period (for example, 30 days). Derive the operation plan. Therefore, at least the input information includes information on the input/output of LNG with respect to the entire tank 10 from which the operation plan is to be derived, and information on the status of each tank 10 . In addition, the restriction information includes information regarding restrictions on LNG inflow/outflow and storage in each of the plurality of tanks 10 . The arithmetic processing unit 2 outputs the output information including the operation plan of the tank 10 by displaying it on a display, transmitting it as data to other devices, or printing it on a printer.

Although the details will be described later, the arithmetic processing unit 2 solves the mixed integer linear programming problem based on the input information and constraint information acquired from the storage unit 1 to derive the operation plan for the planning target period. The main continuous variables, discrete variables, constants, and constraints for solving this problem will be described below. In the following, it is assumed that the planning target period T is 30 days, the unit period is 1 day, and the time t (t=1 to 30) within the planning target period T is expressed in units of days.

　Figure 3 is a table showing the main variables and constants used in solving the mixed integer linear programming problem for deriving the tank operation plan. The symbols of continuous variables, discrete variables, and constants are described in the left column of each table in FIGS. 3A to 3C, and the contents are described in the right column. The dimensions of the received amount, transferred amount, discharged amount, BOG generation amount, cooling return amount, BOG amount, and planned delivery amount shown in FIGS. 3A and 3C are volume (liquid state).

The amount of heat of LNG (amount of received heat, amount of heat stored in tank 10, amount of heat in payout line 14, etc.) can be converted to density (mass per unit volume) in a liquid state. Therefore, all continuous variables and constants relating to the calorific value of LNG can be replaced with continuous variables and constants relating to density. During this replacement, for example, the heat quantity of LNG is converted to the heat quantity per unit volume in the standard state of the vaporized gas [MJ/Nm ³ ] (standard state vaporization heat quantity), and the standard state vaporization heat quantity of LNG and the density of LNG (liquid state) to a linear expression of the other. In addition, since the density (liquid state) of LNG can be converted into the standard state vaporization heat quantity of LNG, the calorie constraint for city gas obtained by vaporizing the delivered LNG is replaced by the constraint on the density (liquid state) of the delivered LNG. Can be replaced with conditions.

Next, I will explain the constraints. Constraints are mainly defined as physical quantity constraints and heat quantity constraints for various facilities at each stage of receiving, storing, transferring, discharging, and cooling. Quantity constraints include constraints on the volume of LNG and constraints on possible combinations of tanks 10, transfer lines 12, and payout lines 14 subject to each operation. The calorific value constraint is a constraint on the calorific value of LNG at each operation stage, but as described above, the calorific value and the density can be replaced with each other, so it is also a constraint on the density of the LNG. Normally, the heat quantity constraint is a non-linear constraint, but as will be described later, the heat quantity is treated as a constant or a variable that varies linearly in this embodiment, so it is not a non-linear constraint in this embodiment.

Below, we will specifically explain the main constraints. First,

equations

1 and 2 below show the constraint condition expressions regarding the volume change of each tank 10 on a daily basis. Both

Equations

1 and 2 are constraints on continuous variables. Each of the constraint condition expressions of Equation 1 and Equation 2 exists in the same number as the product of the number of days in the planning target period T and the number of tanks 10 .

Equation 1 expresses that in tank j, the initial storage volume at time t becomes the initial storage volume at time t+1 after undergoing volume changes due to various operations occurring at time t and the occurrence of BOG. Specifically, to the initial storage volume at time t, the received amount, the inflow transfer amount, and the inflow cooling return amount at time t are added, and the outflow transfer amount, the discharge amount, and the BOG generation amount are subtracted. is the initial storage volume at the next time t+1. Therefore, the constraint condition shown in Equation 1 indicates a physical quantity constraint including all operations of receiving, storing, transferring, discharging, and cooling. Here, the amount of BOG generated in the seventh term on the right side of Equation 1 is the total amount of BOG generated by each operation at time t. , it is obtained by using values that are tabulated in advance for each operation.

Equation 2 is a constraint condition formula that defines physical quantity constraints at the storage stage regarding the upper and lower limits of the storage amount in each tank 10. Specifically, the storage amount in each tank 10 at time t is It stipulates that the volume should be equal to or less than the upper limit of volume determined by the storage capacity and equal to or more than the lower limit of volume.

Next, the constraint condition formulas related to physical quantity restrictions in acceptance are shown in

Equations

3 and 4 below. Equation 3 is a constraint on continuous variables, and Equation 4 is a constraint on continuous and discrete variables (mixed integer constraint).

Equation 3 expresses that the acceptance amount of the acceptance plan whose acceptance date is time t is the sum of the acceptance amounts accepted by each tank 10 at time t. Of the tanks 10 on the left-hand side of Equation 3, the tanks 10 that are not used as receiving tanks have a receiving amount of zero. The number 3 is the same number as the number of days in the planning period T. If there is no acceptance plan at time t, the value of the constant on the right side of Equation 3 is zero.

"·" in Equation 4 is a multiplication symbol (the same shall apply hereinafter). Equation 4 indicates that if the amount of acceptance in tank j at time t+1 is acceptance at time t, it is equal to or less than the upper volume limit of tank j. It represents 0 if there is no acceptance at time t. In addition, the number of formulas 4 is the same as the product of the number of days in the planning target period T and the number of tanks 10 .

Next, the constraint condition formulas related to physical quantity restrictions in payout are shown in Equations 5 to 7 below. Equations 5 to 7 are all constraints on continuous variables. There are as many constraint conditional expressions as the number of days in the planning target period T multiplied by the number of payout lines 14 . There are as many constraint condition expressions as number 6 as the product of the number of days in the planning target period T and the number of tanks 10 . There are as many constraint conditional expressions as the product of the number of days in the planning period T, the number of tanks 10 and the number of delivery lines 14 .

Equation 5 defines the volume balance between the total payout amount to payout line l at time t and the planned delivery amount of payout line l. Specifically, the sum of the amounts dispensed to the dispensation line 1 at time t is the sum of the cooling return amounts from the dispensation line 1 to each tank 10, and the destination m of the dissemination plan for the dispensation line 1 at time t. ' (m' indicates the destination corresponding to payout line 1) by adding the total planned amount of each delivery and subtracting the total amount of each BOG mixed in the delivery destination m' at time t; be equal. For example, the amount of BOG that is mixed in with dispensing can be obtained by using values tabulated in advance for each operation using the density of BOG at the time of the operation as a parameter.

Equation 6 defines that the total amount dispensed from tank j to each dispensing line l at time t is the total amount dispensed from the dispensing pumps connected to tank j.

Equation 7 expresses that the amount dispensed from tank j to dispensing line l at time t is limited to an upper limit determined by the performance of dispensing pump 15 provided between tank j and dispensing line l.

Next, the constraint condition formula related to the calorie constraint in payout is shown in Equation 8 below. Equation 8 is a constraint on continuous variables. There are as many constraint conditional expressions as number 8 as the product of the number of days in the planning target period T and the number of payout lines 14 .

Equation 8 prescribes that the density of each payout line l at time t falls within the range below its upper limit and above its lower limit.

Next, Expressions 9 to 11 below show the constraint condition expressions related to quantity restrictions in transportation. Equations 9 to 11 are all constraints on continuous variables. There are as many constraint condition expressions as number 9 as the product of the number of days in the planning target period T, the number of tanks 10, and the number of tanks 10 minus 1. 10 and 11, the same number as the product of the number of days in the planning target period T and the number of tanks 10 exists.

Equation 9 expresses that the transfer amount from tank i to tank j at time t is limited to an upper limit determined by the performance of transfer pump 13 provided in tank i.

Equation 10 expresses that the total amount of transfer from one tank to another tank is equal to or less than the upper limit value at time t. Equation 11 expresses that at time t, the total transfer amount to a certain tank from other tanks is equal to or less than the upper limit.

Next, the constraint condition formula related to physical quantity restrictions in cooling is shown in Equation 12 below. Equation 12 is a constraint on continuous variables. There are as many constraint conditional expressions as the number of days in the planning target period T multiplied by the number of payout lines 14 .

Equation 12 defines the volume balance for cooling LNG provided via payout line l at time t. Specifically, the total cooling return amount from the payout line 1 to each tank j at time t is equal to the cooling return amount from the payout line 1 to the predetermined transfer line 12 at time t.

Next, as in the third and fourth embodiments to be described later, when the density or heat quantity of the LNG stored in the tank is treated as a variable that varies linearly, the constraint condition expression related to the heat quantity constraint is given by Equation 13 below. 15. 13 to 15 are present in the same number as the product of the number of days in the planning target period T and the number of tanks 10 .

Equations

13 and 14 are formulas that express the constraints on changes in tank density on the day of acceptance.

Equations

13 and 14 update the density of the tank with the receiving density (the density of LNG received from the transportation means 11) in the receiving plan for the tank with receiving, but change to the density for the tank without receiving shall not occur. Here, M is a constant value sufficiently larger than the incoming density and the tank density. Equation 15 is an expression that expresses a constraint on fluctuations in tank density on days when there is no reception.

Above, we explained in detail the main constraints by classifying them into physical quantity constraints and heat quantity constraints at each stage of receiving, storing, transferring, discharging, and cooling. In addition, since the constraints greatly depend on the configuration of the LNG storage facility, the number of individual components, the attributes of individual components (size, performance, etc.), etc., the main constraints illustrated in Equations 1 to 12 In addition to , another constraint may be set. Furthermore, some of the main constraints may be changed to other constraints.

As an example, as a constraint when LNG flows into the tank 10 by receiving, transferring, or cooling, the LNG inlet is determined based on the relationship between the storage amount of the tank 10 and the density difference between the inflowing LNG and the stored LNG. There are constraints that determine either the top of the tank or the bottom of the tank. This is a constraint condition (stratification determination condition) related to the heat amount constraint for preventing stratification due to density distribution occurring in the LNG in the tank 10 due to the above relationship. For example, when there are multiple receiving tanks and the LNG inlet is unified at the bottom of the tank, the storage amount of the receiving tank whose LNG inlet is at the top of the tank is determined in advance. It can be changed to a constraint that is the lower storage amount. In either case, the variable that determines whether the LNG inlet to tank 10 is at the top or bottom of the tank is a discrete variable. For example, in the case where there are a plurality of tanks 10 to receive LNG, the constraint when it is necessary to unify the inlet of LNG to the tank upper part or the tank lower part becomes the constraint using the discrete variable.

As another example, in the case of a plurality of receiving tanks, the difference in tank liquid level between receiving tanks determined by each storage amount before receiving is within a predetermined range. If there are multiple receiving tanks, the constraint that stipulates that the difference in the tank liquid level determined by each storage amount after receiving is equal between the receiving tanks, If there are multiple receiving tanks, the Constraints that stipulate that the ratio of incoming volumes between tanks is a predetermined ratio, and that stipulate that the number of incoming tanks is determined by the incoming volume of the acceptance plan, etc. may be added.

As another example, as a constraint condition related to quantity restrictions in payout, for example, a constraint condition that stipulates that the correspondence between the payout pump 15 and the payout line 14 is fixed without being changed for a certain period of time, and that each payout line 14 is operated A constraint condition or the like may be added that specifies that the number of payout pumps 15 is set to be one more than the required number determined by the planned delivery amount for each payout line 14 . Among the former constraints, the correspondence between the dispensing pump 15 and the dispensing line 14 and the fixed period for fixing the correspondence may be configured to be appropriately changeable as part of the input information.

As another example, as a constraint condition related to the heat quantity constraint in dispensing, for example, the constraint conditions shown in

Equations

11 and 12 are satisfied even if one of the dispensing pumps 15 fails, etc. Constraint conditions stipulating that the density in the tank 10, which is the delivery source before starting the pump 15, is within a predetermined range, and that the amount of BOG mixed in the LNG delivered from the delivery line 14 is not more than a predetermined upper limit value. Constraint conditions or the like that define that there is a certain condition may be added.

As another example, as a constraint condition related to quantity restrictions in transfer, for example, it is specified that the tank 10 that is the transfer source and the tank 10 that is the transfer destination are fixed to a specific tank 10 for each transfer line 12. Constraints, Constraints stipulating that transfer is not performed within the same area when receiving within the same area, Storage amount or liquid level in tank 10 to be transferred before starting transfer pump 13 Constraint conditions or the like that define being within a predetermined range may be added.

As another example, as a constraint condition related to quantity restrictions in cooling, for example, for each transfer line 12, a possible combination of the discharge line 14 from which the LNG for cooling is sent out and the cooling return tank is fixed in advance to a predetermined combination. Constraints and the like may be added.

<<Details of arithmetic processing>>
<First embodiment>
Hereinafter, the content of arithmetic processing by the arithmetic processing unit 2 of the system S according to the first embodiment will be described. First, an overview of the processing of the arithmetic processing unit 2 will be described. FIG. 4 is a flow chart showing an outline of arithmetic processing by the arithmetic processing unit 2 of the system S. As shown in FIG. Note that FIG. 4 shows an outline of arithmetic processing by the arithmetic processing unit 2 of the present system S according to each of the first embodiment described below and the second embodiment described later. This is common to the two embodiments.

As shown in FIG. 4, the arithmetic processing unit 2 first sets a plurality of tank groups 100 (see FIG. 1) formed by dividing the plurality of tanks 10 into groups, each of which receives LNG from the transportation means 11. Create an acceptance pattern (step #1). Next, the arithmetic processing unit 2 solves a mixed integer linear programming problem for the acceptance pattern created in step #1 to derive an operation plan (step #2). Further, based on the operation plan derived in step #2, the arithmetic processing unit 2 solves the mixed integer linear programming problem under the condition that at least one of the number of transfers and the transfer amount is reduced, and derives the operation plan (step # 3). Finally, the arithmetic processing unit 2 outputs the operation plan derived in step #3 (step #4). When multiple acceptance patterns are created in step #1, the operation plans derived for all acceptance patterns in step #4 may be output as output information. The operation plan may be selectively output as output information.

Note that the arithmetic processing unit 2 may be configured to execute only steps #1, #2 and #4, and even in this case it is possible to derive the operation plan. In this case, the arithmetic processing unit 2 outputs the operation plan derived in step #2. However, although the details will be described below, the execution of step #3 by the arithmetic processing unit 2 reduces at least one of the number of transfers and the transfer amount, making it possible to output an operation plan that is easy to execute. Alternatively, the arithmetic processing unit 2 may create only one acceptance pattern in step #1. However, when the arithmetic processing unit 2 creates a plurality of receiving patterns in step #1, it also optimizes the receiving from the transportation means 11 to the tank 10 by comparing the operation plan obtained for each receiving pattern. becomes possible.

The details of step #1 shown in FIG. 4 will be described. FIG. 5 is a flow chart showing details of the arithmetic processing of step #1 by the arithmetic processing section 2 of the system S. As shown in FIG. It should be noted that FIG. 5 is also common to the first embodiment and the second embodiment, as is the case with FIG.

As shown in FIG. 5, the arithmetic processing unit 2 first acquires input information (step #10). The input information includes, for example, a ship allocation plan illustrated in FIG. 6, a tank specification illustrated in FIG. 7, and a line demand illustrated in FIG.

The ship allocation plan exemplified in FIG. 6 includes, for each means of transport 11, a "date of acceptance" that is the time to visit the storage facility, a "heat amount" of the LNG to be supplied, and a "amount of acceptance" that is the amount of LNG to be supplied. included. The unit of "calorie" is [MJ/Nm ³ ], and the unit of "acceptance amount" is [Nm ³ ]. The tank specifications exemplified in FIG. "Liquid level upper limit" that is the upper limit of the level, "Liquid level lower limit" that is the lower limit of the liquid level that can be stored, "Initial stock heat amount" that is the heat amount of LNG stored at the start of the planning target period T, Tank 10 "cross-sectional area", and a "tie line" which is a payout line 14 from which the tank 10 can pay out. The unit of “initial stock liquid level”, “upper limit of liquid level” and “lower limit of liquid level” is [m], the unit of “initial heat quantity” is [MJ/Nm ³ ], and the unit of “cross-sectional area” is [ m ² ]. The line demand illustrated in FIG. 8 includes the demand for each unit period t in the planned period T for each payout line 14, and the unit of the demand is [Nm ³ ] (gas).

Next, the arithmetic processing unit 2 creates a plurality of tank groups 100 by grouping the plurality of tanks 10 (step #11). A tank group 100 is created based on a predetermined rule. For example, acceptable tanks are included in a group of tanks, the "tie line" is the same, and so on. The tank group 100 may be created in advance and the information thereof may be stored in the storage unit 1 .

Next, the arithmetic processing unit 2 creates a plurality of acceptance patterns (step #12). As illustrated in FIG. 9 , the receiving pattern is a combination of vehicles 11 and tanks 100 . Acceptance patterns are also created based on predetermined rules. For example, on the premise that one transport means 11 supplies LNG only to one tank group 100, the heat quantity of the ship allocation plan illustrated in FIG. For example, the transport means 11 to which the LNG belongs is supplied to the tank group 100 in which the heat quantity of the stored LNG is high. All patterns that can be combined with the transportation means 11 and the group of tanks 100 may be created as receiving patterns.

Next, the arithmetic processing unit 2 calculates changes in the calorific value of LNG, etc. for each of the acceptance patterns for each tank group 100 (step #13). At this time, for each tank group 100, the arithmetic processing unit 2 adds the amount of LNG supplied from the transportation means 11 according to the receiving pattern, calculates the transition of the heat amount, and calculates the amount of LNG paid out as the line demand. Subtract

Next, the arithmetic processing unit 2 narrows down the acceptance patterns based on the calculation result of step #13 (step #14). For example, for the LNG stored in the tank group 100, the acceptance patterns are narrowed down by excluding acceptance patterns with large violation amounts such as the storage amount and the average heat quantity. Then, the arithmetic processing unit 2 stores the narrowed-down acceptance pattern in the storage unit 1 (step #15), and ends step #1.

The details of step #2 shown in FIG. 4 will be described. FIG. 10 is a flow chart showing details of the arithmetic processing of step #2 by the arithmetic processing section 2 of the system S according to the first embodiment. Note that the arithmetic processing shown in FIG. 10 is performed for each acceptance pattern created in step #1.

As shown in FIG. 10, the arithmetic processing unit 2 first acquires input information (step #20). The input information includes, for example, a ship allocation plan illustrated in FIG. 6, a tank specification illustrated in FIG. 7, a line demand illustrated in FIG. 8, and a transfer table illustrated in FIG. In the transfer table illustrated in FIG. 11, a weighting factor W _i in an objective function F, which will be described later, is defined for each combination of the tanks 10 of the transfer source and the transfer destination. Further, the arithmetic processing unit 2 acquires constraint information (step #21). The constraint information is information representing the constraint conditions described above.

Next, the arithmetic processing unit 2 solves a mixed integer linear programming problem for each of the plurality of tanks 10 based on the acquired input information and constraint information, regarding the heat quantity of the LNG to be stored as a constant. is derived (step #22). At this time, an objective function F shown in Equation 16 below is used as an objective function. In Equation 16, P _i is a penalty indicated by the deviation from the value of a predetermined monitoring target item or a predetermined reference value, and W _i is a weighting factor when weighting and adding the penalty P _i . For example, items to be monitored include the amount of heat dispensed, the number of transfers, and the amount transferred. In addition, the decision variables are, for example, the amount received and discharged for each tank 10, the presence or absence of operation of each discharge pump 15, the presence or absence of transfer between tanks 10, and the transfer amount for each transfer, and these are included in the operation plan. be

The calorific value of the LNG stored in the tank 10 varies non-linearly depending on the supply of LNG from the outside. Therefore, if the heat quantity of the LNG stored in the tank 10 is used as a variable, the mixed-integer nonlinear programming problem must be solved. Therefore, in step #22, the calorific value of the LNG stored in the tank 10 is regarded as a constant, and the operation plan can be derived by solving the mixed integer linear programming problem. Considering the heat quantity of the LNG stored in the tank 10 as a constant means treating the heat quantity as a constant, and is not limited to fixing the heat quantity to a constant value. For example, in step #22 executed for the first time, even if LNG is supplied to the tank 10 from the outside, the heat amount of the LNG stored in the tank 10 does not change before and after that, or the amount of LNG in the tank 10 and the amount of LNG from the outside It is updated to a predetermined value independent of the amount of LNG supplied. Specifically, when LNG is supplied to the tank 10 by transfer or cooling return, the heat quantity of the LNG stored in the tank 10 does not change, but when the tank 10 receives LNG, the amount of LNG stored in the tank 10 The heat value is updated with the heat value of the incoming LNG.

Next, the arithmetic processing unit 2 calculates changes in the heat quantity of the LNG stored in the tank 10 in the operation plan derived in step #22 (step #23). Specifically, the arithmetic processing unit 2 also calculates changes in the volume of LNG stored in the tank 10 in addition to changes in the amount of heat of the LNG stored in the tank 10 . This is because, as described above, heat quantity and density are interchangeable, and if the heat quantity of LNG fluctuates, the density also fluctuates, and if the density fluctuates, the volume also fluctuates.

Next, the arithmetic processing unit 2 calculates the amount of additional fuel required for dispensing in each of the plurality of tanks 10 based on the operation plan derived in step #22 (step #24). Additional fuel is, for example, LPG. In addition, the required amount of additional fuel can be obtained by the calculation of (target heat amount - payout heat amount) / (additional fuel heat amount - payout heat amount) x payout amount x payout production amount / additional fuel production amount. . The production volume is the volume of gas when a predetermined amount of liquefied fuel is vaporized, and the target heat quantity is, for example, 45 MJ/Nm ³ .

Next, the arithmetic processing unit 2 updates the line demand as a new demand with a value obtained by subtracting the amount of additional fuel calculated in step #24 from the line demand included in the input information (step #25). Then, the arithmetic processing unit 2 determines whether or not the result of calculating the transition of the amount of heat, etc. in step #23 violates the restrictions on the LNG stored in the tank 10 (step #26). Note that the constraint for confirming whether or not there is a violation in step #26 may be a constraint regarding the LNG discharged from the tank 10 in addition to (or instead of) the constraint regarding the LNG stored in the tank 10, or in step #22. The constraints may be the same as or different from the constraints used in solving the mixed integer linear programming problem.

In step #22, the arithmetic processing unit 2 finds a solution so as to satisfy the constraint conditions. As a result of correctly calculating the transition of , the constraint may be violated (step #26, YES). Specifically, the heat amount of the LNG stored in the tank 10 or the heat amount of the LNG paid out to the payout line 14 is out of the limited range, or the volume of the LNG stored in the tank 10 (for example, the liquid level) is out of the limited range. It is possible that In this case, if the solution-finding process of step #22 has not reached the N-th time (N is a natural number of 2 or more) (step #27, NO), the arithmetic processing unit 2 changes the amount of heat calculated in step #23, The new line demand updated in step #25 is acquired (step #28), and the mixed integer linear programming problem is solved again by reflecting these as new constant changes (step #22). As a result, compared with step #22 executed immediately before, the mixed integer linear programming problem can be solved and the operation plan can be derived in a state in which the change in the amount of heat is accurately reflected.

Then, if the calculation result of the transition of the amount of heat and the like in step #23 does not violate the restrictions on the LNG stored in the tank 10 (step #26, NO), the arithmetic processing unit 2 finally determines in step #22 A calculation result based on the derived operation plan is stored in the storage unit 1 (step #29), and the process of step #2 is terminated. At this time, the calculation results include, for example, the amount of fuel received for each tank 10, changes in demand allocation and calorific value, a receiving plan, and the amount of additional fuel required.

Also, if there is a constraint violation (step #26, YES), but the processing of step #22 has reached the N-th time (step #27, YES), the operation plan finally derived in step #22 is stored in the storage unit 1. (step #29), and the process of step #2 is terminated. In this case, since the constraint violation has not been resolved, it is preferable to store this fact together as a calculation result and include it in the output information to alert the operator of the storage facility.

The details of step #3 shown in FIG. 4 will be described. FIG. 12 is a flow chart showing details of the arithmetic processing of step #3 by the arithmetic processing section 2 of the system S according to the first embodiment. Note that, as shown in FIG. 12, the arithmetic processing unit 2 also performs the same processing as in step #2 shown in FIG. 10 in step #3. In the following, regarding the processing of step #3, detailed descriptions of the same portions as in step #2 will be omitted, and different portions will be described.

As shown in FIG. 12, the arithmetic processing unit 2 acquires not only the input information but also the calculation result of step #2 (step #30). Next, the arithmetic processing unit 2 acquires constraint information (step #31), solves a mixed integer linear programming problem, and derives an operation plan (step #32). At this time, for example, the arithmetic processing unit 2 may derive the operation plan under the same kind of conditions as in step #2. Specifically, for example, in step #22 of step #2, the arithmetic processing unit 2 derives an operation plan under conditions that minimize the number of transfers, and in step #32, transfers more than the operation plan derived in step #2. The operation plan may be derived under the condition that the number of times is reduced, or the operation plan is derived under the condition that the transfer amount is minimized in step #22 of step #2, and in step #32, the operation plan is derived in step #2. An operation plan may be derived on the condition that the transfer amount is further reduced than the operation plan. Further, for example, the arithmetic processing unit 2 may derive the operation plan under conditions different from those in step #2. Specifically, for example, the arithmetic processing unit 2 may derive the operation plan under the condition that the number of transfers is minimized in step #22, and derive the operation plan under the condition that the transfer amount is further reduced in step #32. Alternatively, in step #22, the operation plan may be derived under the condition of minimizing the transfer amount, and in step #32, the operation plan may be derived under the condition of further reducing the number of transfers. As a condition for further reducing the number of times of transfer and the transfer amount in step #32, the transfer amount based on the transfer amount in the calculation result of step #2 is not transferred for the combination of tanks that are not transferred in the calculation result of step #2. and allow transfer within the upper and lower limits of

Also, when solving the mixed integer linear programming problem, whether or not each payout pump 15 is in operation is fixed based on the calculation result of step #2, so it is excluded from the decision variables. Subsequent processing (steps #33 to #39) is the same as the processing of step #2 (steps #23 to #29) shown in FIG. Then, the operation plan derived in step #3 becomes the finally derived operation plan.

As described above, in the system S according to the first embodiment, the heat quantity of the liquefied fuel stored in the tank 10 that varies nonlinearly is not treated as a variable in the mathematical programming problem, but is calculated separately. Solve as an integer linear programming problem to derive an operational plan. Specifically, the system S calculates the calorific value of the LNG stored in the tank 10 separately from the mixed integer linear programming problem and treats it as a constant in the mixed integer linear programming problem, thereby reflecting the change in the calorific value. to solve. Thereby, this system S becomes possible [ reducing the load of arithmetic processing ].

In addition, in this system S, in steps #2 and #3, while repeatedly solving the mixed integer linear programming problem, the required amount of additional fuel is calculated and the line demand is updated (steps #24, #25, #34 and #35). Therefore, it is possible to derive an operation plan including the amount of fuel to be delivered, which is determined by taking into account the amount of additional fuel to be supplied.

<Second embodiment>
Next, the content of arithmetic processing by the arithmetic processing unit 2 of the system S according to the second embodiment will be described. However, in the arithmetic processing of the arithmetic processing unit 2 of the system S according to the second embodiment, the detailed explanation of the parts that are the same as in the first embodiment will be omitted, and the parts that are different from the first embodiment will be explained. .

FIG. 13 is a flow chart showing details of the arithmetic processing of step #2 by the arithmetic processing unit 2 of the system S according to the second embodiment. As shown in FIG. 13, in the second embodiment, after there is no constraint violation (step #26, NO) or the solution-finding process reaches the N-th time (step #27, NO), that is, in step #2, mixed integer linear After solving the planning problem, the amount of additional fuel required is calculated (step #24). In addition, the arithmetic processing unit 2 does not update the line demand (step #25 in FIG. 10), and when resolving the mixed integer linear programming problem in step #22, the heat amount of the LNG stored in the tank 10 Only the transition is acquired (step #28A).

FIG. 14 is a flow chart showing details of the arithmetic processing of step #3 by the arithmetic processing unit 2 of the system S according to the second embodiment. Step #3 is similar to step #2, and after solving the mixed integer linear programming problem in step #3, the amount of additional fuel required is calculated (step #34). In addition, the arithmetic processing unit 2 does not update the line demand (step #35 in FIG. 12), and when resolving the mixed integer linear programming problem in step #32, the heat amount of the LNG stored in the tank 10 Only transitions are acquired (step #38A).

As described above, in the system S according to the second embodiment as well, the heat quantity of the liquefied fuel stored in the nonlinearly varying tank 10 is calculated separately without being treated as a variable in the mathematical programming problem. Since the operation plan is derived by solving the problem as an integer linear programming problem, it is possible to reduce the computational load. Furthermore, in the present system S according to the second embodiment, it is possible to minimize the computational processing related to the amount of additional fuel, thereby reducing the computational processing load.

<Third Embodiment>
Next, the content of arithmetic processing by the arithmetic processing unit 2 of the system S according to the third embodiment will be described. However, in the arithmetic processing of the arithmetic processing unit 2 of the system S according to the second embodiment, the detailed explanation of the parts that are the same as in the first embodiment will be omitted, and the parts that are different from the first embodiment will be explained. .

FIG. 15 is a flow chart showing details of the arithmetic processing of step #2 by the arithmetic processing unit 2 of the system S according to the third embodiment. As shown in FIG. 15, in the third embodiment, similarly to the first embodiment, acquisition information and constraint information are acquired (steps #20 and #21). Next, in the third embodiment, the arithmetic processing unit 2 regards the heat quantity of the LNG to be stored for each of the plurality of tanks 10 as a variable that varies linearly rather than as a constant, and solves the mixed integer linear programming problem. An operation plan is derived (step #22B). At this time, the arithmetic processing unit 2 uses the objective function F shown in Equation 16 as in the first embodiment as the objective function for deriving the operation plan. Based on the formula, the operation plan is derived by regarding the heat quantity of the LNG stored in the tank as a variable that varies linearly.

After that, the arithmetic processing unit 2 performs the same processing as in the first embodiment (steps #23 to #29). However, when performing step #22B again after step #28, the arithmetic processing unit 2 recalculates the mixed integer linear programming problem by reflecting the change in the amount of heat as a constant change in the first embodiment. , In the third embodiment, the amount of change in the amount of heat regarded as a variable that varies linearly so as to correspond to the transition (difference) of the amount of heat (for example, when the storage tank j in the above-mentioned equation 13 receives the time t density change α, inventory density change β at time t in storage tank j in Equation 15, density change γ when transferred from storage tank i at time t in storage tank j in Equation 15, etc.) to solve the mixed-integer linear programming problem again.

FIG. 16 is a flow chart showing details of the arithmetic processing of step #3 by the arithmetic processing unit 2 of the system S according to the third embodiment. As shown in FIG. 16, the arithmetic processing unit 2 also performs the same processing as step #2 shown in FIG. 15 in step #3. In addition, the arithmetic processing unit 3 performs the processing of step #3 in the same manner as in the first embodiment, but as described in step #2, when solving the mixed integer linear programming problem and deriving the operation plan, For each of the plurality of tanks 10, the mixed-integer linear programming problem is solved by considering the heat quantity of the LNG to be stored not as a constant but as a variable that varies linearly (step #32B).

As described above, in the system S according to the third embodiment, the heat quantity of the liquefied fuel stored in the nonlinearly varying tank 10 is calculated as a variable that varies linearly, so that the mixed integer linear programming problem Since the operation plan is derived by finding the solution, it is possible to reduce the load of arithmetic processing. Also, as in the first embodiment, while solving the mixed integer linear programming problem repeatedly in steps #2 and #3, the required amount of additional fuel is calculated and the line demand is updated (steps #24, #25, #34, and #35), it is possible to derive an operation plan including a payout amount determined by taking into consideration the amount of additional fuel to be supplied.

<Fourth Embodiment>
Next, the content of arithmetic processing by the arithmetic processing unit 2 of the system S according to the fourth embodiment will be described. However, among the arithmetic processing of the arithmetic processing unit 2 of the present system S according to the fourth embodiment, detailed explanations of the same parts as those in the first to third embodiments are omitted, and the first to third embodiments are omitted. A different part will be explained.

FIG. 17 is a flow chart showing details of the arithmetic processing of step #2 by the arithmetic processing unit 2 of the system S according to the third embodiment. As shown in FIG. 17, the arithmetic processing unit 2 calculates the required amount of additional fuel (step #24) after solving the mixed integer linear programming problem in step #2, as in the second embodiment. ), updating the line demand (step #25 in FIGS. 10 and 15) is not executed, and when resolving the mixed integer linear programming problem in step #22B, only the change in the heat amount of LNG stored in the tank 10 is acquired. (step #28A). Further, similarly to the third embodiment, when the arithmetic processing unit 2 solves the mixed integer linear programming problem and derives the operation plan, the heat quantity of the LNG to be stored for each of the plurality of tanks 10 is set to a constant A mixed-integer linear programming problem is solved by regarding variables that change linearly instead of (step #22B).

FIG. 18 is a flow chart showing details of the arithmetic processing of step #3 by the arithmetic processing unit 2 of the system S according to the fourth embodiment. Step #3 is similar to step #2, and after solving the mixed integer linear programming problem in step #3, the amount of additional fuel required is calculated (step #34). In addition, the arithmetic processing unit 2 does not update the line demand (step #35 in FIGS. 12 and 16), and when resolving the mixed integer linear programming problem in step #32B, the LNG stored in the tank 10 is acquired (step #38A). Further, similarly to the third embodiment, when the arithmetic processing unit 2 solves the mixed integer linear programming problem and derives the operation plan, the heat quantity of the LNG to be stored for each of the plurality of tanks 10 is set to a constant A mixed-integer linear programming problem is solved by regarding variables that change linearly instead of (step #32B).

As described above, in the present system S according to the fourth embodiment as well, the heat quantity of the liquefied fuel stored in the nonlinearly varying tank 10 is calculated as a variable that varies linearly, so that the mixed integer linear programming problem Since the operation plan is derived by finding the solution, it is possible to reduce the load of arithmetic processing. Furthermore, in the present system S according to the fourth embodiment, similarly to the second embodiment, it is possible to minimize the computational processing related to the amount of additional fuel, thereby reducing the computational processing load.

<<Modification>>
The above-described embodiments are merely examples for carrying out the present invention. Therefore, the present invention is not limited to the above-described embodiment, and it is possible to modify the above-described embodiment appropriately without departing from the scope of the invention.

For example, in the above first to fourth embodiments, if there is a constraint violation (steps #26, #36, YES), update the heat amount of LNG stored in the tank and solve the mixed integer linear programming problem again. (Steps #22, #32, #22B, #32B). However, even if there are no constraint violations, the mixed integer linear programming problem may be solved again by updating the calorific value of the LNG stored in the tanks, for example to improve the accuracy of the derived operational plan. Specifically, after confirming that there is no constraint violation (or without confirming whether there is a constraint violation), the calorific value of the LNG stored in the tank is updated a predetermined number of times, and the mixed integer linear programming problem can be solved again.

Also, for example, in the first to fourth embodiments, the input information includes the transfer table, but the input information may not include the transfer table. For example, by setting a condition that the weight is smaller when transferring from a tank with a large calorific value of LNG to be stored to a tank with a smaller calorific value, and the weight is larger when transferring from a tank with a small calorific value of LNG to be stored to a tank with a large calorific value. , it is possible to derive an operational plan without using a transfer table.

Further, for example, in the first to fourth embodiments, when solving the mixed integer linear programming problem and deriving the operation plan (steps #22, #32, #22B, #32B), the solution is solved according to some rule. may For example, the arithmetic processing unit 2 of the system S may perform this solution based on a predetermined operating rule that suppresses nonlinear fluctuations in the heat quantity of LNG. As such an operation rule, for example, a tank with a large calorific value of LNG to be stored and a tank with a small calorific value are distinguished in advance, and a tank that receives the LNG loaded by the transport means 11 is determined according to the calorific value of the LNG loaded by the transport means 11 ( If the calorific value of the LNG to be stored is large, it is received in a tank with a large calorific value of the LNG to be stored, and if the calorific value of the LNG loaded by the transport means 11 is small, the LNG to be stored is received in a tank with a small calorific value), and the calorific value of the LNG to be stored is Between tanks where the difference is greater than a certain level, a large weight is set for transfer to avoid transfer as much as possible. to prevent it from happening, and so on. By solving the mixed-integer linear programming problem using such operation rules, the arithmetic processing unit 2 can easily derive the optimum operation plan, so that the burden of arithmetic processing can be reduced more appropriately. Become.

Further, for example, in the first to fourth embodiments, the arithmetic processing unit 2 solves the mixed integer linear programming problem by regarding the heat amount of LNG stored in the tank as a constant or a linearly changing variable, and further calculates the heat amount It was explained that the transition of is calculated as a change in a constant or linearly changing variable when solving the mixed integer linear programming problem, but the density instead of the heat quantity can be regarded as a constant or linearly changing variable, Both heat content and density may be considered as constant or linearly varying variables.

Further, for example, in the above first to fourth embodiments, the case where the amount of LNG such as the amount of LNG received, transferred, and discharged was calculated by volume was described, but it is calculated by mass (for example, the unit is tons). may

Also, for example, in the above-described first to fourth embodiments, step #1 in FIG. 4 may be omitted by separately determining the acceptance pattern in advance. Further, for example, in the above-described first to fourth embodiments, step #1 of FIG. 4 may be omitted by solving the acceptance pattern as a mixed integer linear programming problem in step #2 of FIG.

The tank operation plan derivation system and tank operation plan derivation method described above can be explained as follows.

The tank operation plan derivation system includes input information including information on the inflow and outflow of the liquefied fuel with respect to the entire plurality of tanks storing the liquefied fuel and information on the state of each of the plurality of tanks, and input information in each of the plurality of tanks a storage unit for storing constraint information including constraint conditions for the input/output and storage of the liquefied fuel; and a plan for the input/output of the liquefied fuel in each of the plurality of tanks based on the input information and the constraint information an arithmetic processing unit for deriving a certain operation plan, wherein the arithmetic processing unit considers the calorific value or density of the liquefied fuel stored in each of the plurality of tanks as a constant or a variable that varies linearly, and calculates a mixed integer linear A first process of deriving the operation plan by solving a planning problem, and calculating changes in the calorific value or density of the liquefied fuel stored in each of the plurality of tanks in the operation plan derived by the first process. and when re-executing the first process after the second process, the heat quantity or density of the liquefied fuel stored in each of the plurality of tanks calculated in the second process is regarded as a change in a constant or linearly changing variable, and the first process is re-executed (first configuration). According to this configuration, the calorific value or density of the liquefied fuel stored in the nonlinearly fluctuating tank is not treated as a variable in the mathematical programming problem but is calculated separately, so that it is solved as a mixed integer linear programming problem and operated. Derive the plan. Therefore, according to this configuration, it is possible to reduce the load of arithmetic processing.

In the first configuration, the arithmetic processing unit may execute the first processing based on an operating rule that suppresses nonlinear fluctuations in the calorific value or density of the liquefied fuel (second configuration). According to this configuration, it becomes easier to derive the optimum operation plan, so that it is possible to more preferably reduce the load of arithmetic processing.

If the result of the second process violates the constraint of at least one of the liquefied fuel stored in the plurality of tanks and the liquefied fuel discharged from the plurality of tanks, the arithmetic processing unit performs The first process may be re-executed by regarding the transition of the calculated calorific value or density of the liquefied fuel stored in each of the plurality of tanks as a constant or linearly changing variable. According to this configuration, it is possible to derive an operational plan that does not violate the constraints.

In the third configuration, the arithmetic processing unit calculates changes in the volume of the liquefied fuel stored in each of the plurality of tanks in the second processing, and the result is the volume of the liquefied fuel stored in each of the plurality of tanks. If the restriction on the volume of the liquefied fuel is violated, the change in the calorific value or density of the liquefied fuel stored in each of the plurality of tanks calculated in the second process is regarded as a constant or linearly changing variable, and the second process is performed. 1 processing may be re-executed (fourth configuration). According to this configuration, it is possible to derive an operation plan in which the volume of liquefied fuel stored in the tank does not violate the constraints.

In the third or fourth configuration, if the result of the second process violates constraints on the calorific value or density of the liquefied fuel dispensed from the plurality of tanks, the arithmetic processing unit calculates in the second process The first process may be re-executed by regarding the change in the calorific value or density of the liquefied fuel stored in each of the plurality of tanks as a constant or linearly changing variable (fifth configuration). According to this configuration, it is possible to derive an operation plan in which the heat quantity of the liquefied fuel discharged from the tank does not violate the constraints.

In any one of the third to fifth configurations, the arithmetic processing unit determines that the result of the second processing is the liquefied fuel stored in the plurality of tanks and the liquefied fuel discharged from the plurality of tanks. Even if at least one of the constraints is violated, if the number of times the first process has already been executed has reached N times (N is a natural number equal to or greater than 2), the first process must be re-executed. (sixth configuration). According to this configuration, when it is difficult to expect that the constraint violation will be resolved, it is possible to prevent occurrence of an excessive computational load by terminating further solution-finding processing.

In any one of the first to sixth configurations, if the first process is not re-executed after the second process, the arithmetic processing unit performs Under the condition that at least one of the number of times of transfer and the amount of transfer is reduced from the operation plan, the calorific value or density of the liquefied fuel stored in each of the plurality of tanks is regarded as a constant or a variable that varies linearly, and is a mixed integer A third process of deriving the operation plan may be executed by solving a linear programming problem (seventh configuration). According to this configuration, it is possible to derive an operation plan that reduces at least one of the number of times of transfer and the amount of transfer and that is easy to execute.

In any one of the first to seventh configurations, the arithmetic processing unit determines, based on the operation plan derived by the first processing, the amount of additional fuel required to be dispensed from each of the plurality of tanks. is calculated, and a new demand is derived by subtracting the amount corresponding to the amount of the additional fuel from the demand corresponding to the payout, and when the first process is executed again after the second process, The first process may be re-executed based on the new demand (eighth configuration). According to this configuration, it is possible to derive an operation plan including a payout amount determined by also considering the amount of additional fuel to be supplied.

In any one of the first to seventh configurations, if the first process is not re-executed after the second process, the arithmetic processing unit performs A required amount of additional fuel may be calculated based on the operational plan (ninth configuration). According to this configuration, it is possible to minimize the computational processing related to the amount of additional fuel and reduce the computational processing load.

In any one of the first to ninth configurations, the arithmetic processing unit is a combination in which each of a plurality of tank groups formed by grouping the plurality of tanks receives the supply of the liquefied fuel from the transportation means. A plurality of acceptance patterns may be created, and the first process and the second process may be executed for each acceptance pattern to derive a plurality of the operation plans (tenth configuration). According to this configuration, by comparing the operation plan obtained for each receiving pattern, it becomes possible to optimize the receiving from the transportation means to the tank.

In another embodiment of the present invention, input information including information on the inflow and outflow of the liquefied fuel with respect to all of the plurality of tanks storing the liquefied fuel and information on the state of each of the plurality of tanks; Based on constraint information including constraints on the inflow and outflow and storage of the liquefied fuel in each of the plurality of tanks, the heat quantity or density of the liquefied fuel stored in each of the plurality of tanks is regarded as a constant or a variable that varies linearly. Mixed integer A first step of deriving an operation plan that is a plan for the liquefied fuel in and out of each of the plurality of tanks by solving a linear programming problem; and in the operation plan derived in the first step, the plurality of a second step of calculating transitions in the calorific value or density of the liquefied fuel stored in each of the tanks; A method for deriving a tank operation plan, wherein the change in the calorific value or density of the liquefied fuel stored in each of the plurality of tanks is regarded as a change in a constant or linearly changing variable, and the first step is re-executed (11th composition).

S... Tank operation plan derivation system 1... Storage unit 2... Arithmetic processing unit 10... Tank 11... Transportation means 12... Transfer line 13... Transfer pump 14... Payout line 15... Payout pump 100... group of tanks

Claims

input information including information on the input/output of the liquefied fuel to/from the entire plurality of tanks storing the liquefied fuel and information on the state of each of the plurality of tanks; input/output and storage of the liquefied fuel in each of the plurality of tanks; a storage unit that stores constraint information including constraint conditions of
an arithmetic processing unit that derives an operation plan, which is a plan for entering and exiting the liquefied fuel in each of the plurality of tanks, based on the input information and the constraint information;
The arithmetic processing unit is
a first process of deriving the operation plan by solving a mixed-integer linear programming problem regarding each of the plurality of tanks, regarding the heat quantity or density of the stored liquefied fuel as a constant or a linearly changing variable;
a second process of calculating changes in the calorific value or density of the liquefied fuel stored in each of the plurality of tanks in the operation plan derived in the first process,
When re-executing the first process after the second process, a constant or linearly changing variable of the transition of the calorie or density of the liquefied fuel stored in each of the plurality of tanks calculated in the second process A tank operation plan derivation system that regards a change and re-executes the first process.
The tank operation plan derivation system according to claim 1, wherein the arithmetic processing unit executes the first process based on an operation rule that suppresses nonlinear fluctuations in the calorific value or density of the liquefied fuel.
If the result of the second process violates the constraint of at least one of the liquefied fuel stored in the plurality of tanks and the liquefied fuel discharged from the plurality of tanks, the arithmetic processing unit performs 3. The tank according to claim 1 or 2, wherein the calculated change in the calorific value or density of the liquefied fuel stored in each of the plurality of tanks is regarded as a change in a constant or linearly changing variable, and the first process is re-executed. Operation plan derivation system.
The arithmetic processing unit calculates a transition of the volume of the liquefied fuel stored in each of the plurality of tanks in the second processing, and the result is a constraint on the volume of the liquefied fuel stored in each of the plurality of tanks. is violated, the transition of the calorific value or density of the liquefied fuel stored in each of the plurality of tanks calculated in the second process is regarded as a constant or linear variable change, and the first process is re-executed. 4. The tank operation plan derivation system according to claim 3.
When the result of the second processing violates the constraint of the calorific value or density of the liquefied fuel discharged from the plurality of tanks, the arithmetic processing unit determines that each of the plurality of tanks calculated in the second processing is stored. 5. The tank operation plan deriving system according to claim 3 or 4, wherein the change in the calorific value or density of said liquefied fuel is regarded as a change in a constant or linearly changing variable, and said first process is re-executed.
The arithmetic processing unit already performs The tank according to any one of claims 3 to 5, wherein the first process is not re-executed when the number of times the first process has been executed has reached N times (N is a natural number of 2 or more). Operation plan derivation system.
When the first process is not re-executed after the second process, the arithmetic processing unit determines that at least one of the number of transfers and the transfer amount is higher than the operation plan derived by the last executed first process. is reduced, the operation plan is derived by solving a mixed-integer linear programming problem with the calorific value or density of the liquefied fuel stored in each of the plurality of tanks as a constant or linearly changing variable. The tank operation plan derivation system according to any one of claims 1 to 6, which executes a third process.
The arithmetic processing unit is
Based on the operation plan derived by the first process, the amount of additional fuel required for dispensing in each of the plurality of tanks is calculated, and the amount of additional fuel is calculated from the demand corresponding to the dispensing. Deriving new demand by subtracting
The tank operation plan according to any one of claims 1 to 7, wherein when the first process is re-executed after the second process, the first process is re-executed based on the new demand. derivation system.
When the first process is not re-executed after the second process, the arithmetic processing unit calculates the required amount of additional fuel based on the operation plan derived by the last executed first process. The tank operation plan derivation system according to any one of claims 1 to 7, which calculates.
The arithmetic processing unit creates a plurality of receiving patterns that are combinations in which each of the plurality of tank groups formed by grouping the plurality of tanks receives the supply of the liquefied fuel from the transportation means, and the first receiving pattern for each of the receiving patterns. The tank operation plan derivation system according to any one of claims 1 to 9, wherein the first process and the second process are executed to derive a plurality of the operation plans.
input information including information on the input/output of the liquefied fuel to/from the entire plurality of tanks storing the liquefied fuel and information on the state of each of the plurality of tanks; input/output and storage of the liquefied fuel in each of the plurality of tanks; Based on the constraint information including the constraint of, for each of the plurality of tanks, the heat quantity or density of the liquefied fuel to be stored is regarded as a constant or linearly changing variable to solve a mixed integer linear programming problem, a first step of deriving an operation plan, which is a plan for entering and exiting the liquefied fuel in each of the plurality of tanks;
a second step of calculating changes in the calorific value or density of the liquefied fuel stored in each of the plurality of tanks in the operation plan derived in the first step;
When re-executing the first process after the second process, the change in the calorific value or density of the liquefied fuel stored in each of the plurality of tanks calculated in the second step is a constant or a variable that changes linearly. A method for deriving a tank operation plan, which considers a change and re-executes the first step.