Sequential liquid conveying pipeline scheduling method based on multi-time scale nesting

Abstract

The invention relates to a sequential liquid conveying pipeline scheduling method based on multi-time scale nesting, and the method comprises the steps: obtaining input information which comprises pipeline and station yard information, a scheduling period and preset batch plan information; a rough scheduling plan, namely a daily scheduling plan, of the operation amount of each station yard is solved by taking day as a scheduling time unit and using a first objective function and a first constraint condition; dividing a fine scheduling time window from the current coarse scheduling time; for the fine scheduling time window, solving a fine scheduling plan of the operation amount of each station yard in the fine time window and a coarse scheduling plan of the operation amount of each station yard in the remaining coarse time window by using a preset time unit less than one day and using the day as the time unit for the remaining coarse scheduling time according to a second objective function and a second constraint condition; updating the daily scheduling plan; repeatedly iterating until the current coarse scheduling time cannot be divided into the fine scheduling time window; and superposing and outputting the fine scheduling plans of the stations according to a time sequence.

Description

Technical Field

[0001] The invention relates to the technical field of refined oil transportation, and in particular to a method for scheduling sequential liquid transportation pipelines based on multiple time scale nesting. Background Art

[0002] The refined oil pipeline is an important means of transportation for the spatial transfer of oil resources on land. When the refined oil pipeline is put into market operation, it faces requests for changes in the demand plans of various shippers. Therefore, the core link of the operation and management of the refined oil pipeline network is to quickly formulate a scheduling plan that meets market demand and ensures safe and efficient transportation of the pipeline network. The optimized scheduling plan can achieve accurate matching between the oil supply side and the market demand side. Due to the strong uncertainty of upstream production plans and downstream demand plans, during the operation process, the dispatchers also need to adjust the scheduling plan in real time according to changes in production, demand and pipeline status.

[0003] The optimization model for refined oil pipeline scheduling is usually constructed as a mixed integer linear programming (MILP) model, which requires a large number of binary variables to characterize the operational decisions of different stations at different stages. The number of binary variables grows exponentially with the length of the planning cycle, the scale of the station, and the number of batches. When facing long-cycle, large-scale pipeline systems, a large number of binary variables can easily cause a "combinatorial explosion" in the search space. The branch and bound method that traditional scheduling optimization relies on takes more than half an hour when facing pipelines with 7 stations. The timeliness cannot meet the needs of dynamic adjustment. Failure to adjust in time will cause production accidents such as emergency shutdown of pipelines, a sharp increase in mixed oil volume, and the inability to deliver subsequent oil products on time. Summary of the invention

[0004] In view of the above problems, the purpose of the present invention is to provide a sequential liquid pipeline scheduling method based on multi-time scale nesting, which can be suitable for the rapid preparation of large-scale, long-period sequential transportation plans for refined oil pipelines, and greatly reduce the scale of binary variables in large-scale scheduling optimization models, thereby achieving the effect of accelerating the solution and realizing the rapid response of pipeline plans to the dynamic market.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present application provides a method for scheduling a sequential liquid transport pipeline based on multiple time scale nesting, the method comprising:

[0007] (1) obtaining input information, including: pipeline and station information of the sequential liquid transport pipeline, scheduling cycle and preset batch plan information;

[0008] (2) Based on the input information, taking day as the basic time unit of rough scheduling, and using the preset first objective function and first constraint condition, solve the daily scheduling plan of each station within the rough scheduling time window. The initial value of the rough scheduling time window is the scheduling period;

[0009] (3) Perform the following iterative processing:

[0010] Based on the current daily scheduling plan, the coarse scheduling time window is updated by subtracting a fine scheduling time window of a set length from the current coarse scheduling time window;

[0011] Among them, for the fine scheduling time window, a preset time unit of less than one day is used, and for the updated rough scheduling time window, the time unit is still one day. The second objective function and the second constraint condition are used to solve the fine scheduling plan of the operation quantity of each station in the fine time window and the rough scheduling plan in the rough scheduling time window;

[0012] Update the daily scheduling plan according to the rough scheduling plan within the rough scheduling time window;

[0013] Repeat this step until the current coarse scheduling time window can no longer be divided into a fine scheduling time window, and use the remaining time of the coarse scheduling time window as a preset time unit less than one day, and use the second objective function and the second constraint condition to solve the fine scheduling plan of the operation volume of each station;

[0014] (4) The detailed scheduling plans of each station obtained by the above iterative processing are superimposed in chronological order to output the final scheduling plan covering the complete scheduling cycle.

[0015] In one implementation, the first objective function is: the deviation between the actual injection / distribution volume accumulated at the station in the rough scheduling time window and the planned injection / distribution volume is minimized.

[0016] In one implementation, the first constraint condition includes: batch volume constraint, daily maximum injection / distribution volume constraint, daily maximum pipeline transport flow constraint, and operation logic constraint.

[0017] In one implementation, the second objective function is: the deviation between the actual injection / distribution volume accumulated at the station in the fine scheduling time window and the coarse scheduling time window and the planned injection / distribution volume is minimized.

[0018] In one implementation, the second constraint condition includes: batch volume constraint, maximum injection / distribution flow constraint, minimum injection / distribution flow constraint, maximum pipeline transport flow constraint, minimum pipeline transport flow constraint, and operation logic constraint.

[0019] In one implementation, the preset time unit smaller than one day is evenly divisible by the fine scheduling time window.

[0020] In a second aspect, a method system for sequential liquid pipeline scheduling based on multi-time scale nesting is provided, the system comprising:

[0021] An input module is used to obtain input information, including: pipeline and station information of the sequential liquid delivery pipeline, scheduling cycle and preset batch plan information;

[0022] An initial calculation module is used to solve the daily scheduling plan of each station within the rough scheduling time window according to the input information, taking the day as the basic time unit of the rough scheduling, and using the preset first objective function and the first constraint condition. The initial value of the rough scheduling time window is the scheduling period;

[0023] The iterative processing module is used to perform the following iterative processing: based on the current daily scheduling plan, a fine scheduling time window of a set length is subtracted from the current coarse scheduling time window to update the coarse scheduling time window; wherein, for the fine scheduling time window, a preset time unit of less than one day is used, while for the updated coarse scheduling time window, the time unit is still day, and the second objective function and the second constraint are used to solve the fine scheduling plan of the operating quantity of each station in the fine time window and the coarse scheduling plan in the coarse scheduling time window; the daily scheduling plan is updated according to the coarse scheduling plan in the coarse scheduling time window; the processing of this step is repeated until the current coarse scheduling time window can no longer be divided into fine scheduling time windows, and the remaining time of the coarse scheduling time window is used in a preset time unit of less than one day, and the second objective function and the second constraint are used to solve the fine scheduling plan of the operating quantity of each station;

[0024] The output module is used to superimpose the detailed scheduling plans of each station obtained by the aforementioned iterative processing in chronological order and output the final scheduling plan covering the complete scheduling cycle.

[0025] The present invention uses a "fine + coarse" mixed time expression to construct a method and system for accelerating the solution of sequential liquid pipeline scheduling decisions based on nested multi-time scales, aiming to effectively solve the problem of "combinatorial explosion" in the search space caused by the presence of a large number of binary variables when compiling a large-scale refined oil pipeline system scheduling plan, thereby laying a technical foundation for the dynamic adjustment of the refined oil pipeline scheduling plan. The proposed method uses the solution information of a small-scale coarse scheduling cycle to quickly construct a dynamic set of time, station field, and batch, and embeds it in the decision variable and logical constraint construction process of the next round of fine scheduling cycle, guiding the removal of a large number of redundant and invalid search spaces in the fine scheduling cycle, accelerating the convergence of the fine scheduling model without losing optimality, and greatly improving the solution speed of large-scale, long-cycle, and multi-batch scheduling plans. According to case verification, the solution speed can be increased by 99.40%. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1It is a schematic diagram of a mixed time scale nested rolling solution framework provided in an embodiment of the present application;

[0027] Figure 2 is a detailed scheduling plan diagram obtained in a calculation example of this application;

[0028] Figures 3(a) to (g) are schematic diagrams of flow rate changes at various stations obtained in a calculation example of the present application;

[0029] 4(a) to (f) are schematic diagrams of flow rate changes in various pipe sections obtained in a calculation example of the present application. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0031] In view of the problems of the prior art, an embodiment of the present invention provides a method for scheduling a sequential liquid transport pipeline based on multi-time scale nesting, the method comprising:

[0032] (1) obtaining input information, including: pipeline and station information of the sequential liquid transport pipeline, scheduling cycle and preset batch plan information;

[0033] (2) Based on the input information, taking day as the basic time unit of rough scheduling, and using the preset first objective function and first constraint condition, solve the daily scheduling plan of each station within the rough scheduling time window. The initial value of the rough scheduling time window is the scheduling period;

[0034] (3) Perform the following iterative processing:

[0035] Based on the current daily scheduling plan, the coarse scheduling time window is updated by subtracting a fine scheduling time window of a set length from the current coarse scheduling time window;

[0036] Among them, for the fine scheduling time window, a preset time unit of less than one day is used, and for the updated rough scheduling time window, the time unit is still one day. The second objective function and the second constraint condition are used to solve the fine scheduling plan of the operation quantity of each station in the fine time window and the rough scheduling plan in the rough scheduling time window;

[0037] Update the daily scheduling plan according to the rough scheduling plan within the rough scheduling time window;

[0038] Repeat this step until the current coarse scheduling time window can no longer be divided into a fine scheduling time window, and use the remaining time of the coarse scheduling time window as a preset time unit less than one day, and use the second objective function and the second constraint condition to solve the fine scheduling plan of the operation volume of each station;

[0039] (4) The detailed scheduling plans of each station obtained by the above iterative processing are superimposed in chronological order to output the final scheduling plan covering the complete scheduling cycle.

[0040] The following is based on Figure 1 , the above method is described in a more detailed embodiment.

[0041] like Figure 1 , a rolling solution method nested with coarse and fine mixed time scales.

[0042] Specifically, the solution method steps include:

[0043] (1) For batch plans with a duration of TD, the scheduling cycle is split into daily (24 h) time intervals. The daily injection / distribution volume of each station is used as the decision variable. The batch volume constraint, the daily maximum injection / distribution volume constraint, the daily maximum pipeline transportation flow constraint, and the operation logic constraint are comprehensively considered. The objective function is to minimize the deviation between the actual injection / distribution volume and the planned injection / distribution volume of the station. A rough scheduling optimization model for the refined oil pipeline is constructed to obtain the daily scheduling plan, that is, the injection / distribution volume of each batch at each station along the pipeline every day.

[0044] (2) A "fine + coarse" hybrid time scale nested rolling solution framework is constructed based on the daily scheduling plan. The daily scheduling cycle TD is divided into "fine + coarse" cycles, where the total number of days for fine scheduling is Δd, and the time interval is Δdt hours (Δdt<24h and can be divided by 24), and the total number of days for coarse scheduling is TD-Δd, and the time interval is day (24h). For the fine scheduling cycle, firstly, based on the batch movement and the station download volume of the previous Δd days of the daily scheduling plan, a dynamic set of time windows, stations, and batches is defined. The injection / distribution flow of each station in each fine time window is used as the decision variable to construct the corresponding batch volume constraint, maximum injection / distribution flow constraint, minimum injection / distribution flow constraint, maximum pipeline transportation flow constraint, minimum pipeline transportation flow constraint, and operation logic constraint. Then, for the coarse scheduling cycle (TD-Δd), the daytime injection / distribution volume of each station in the cycle is used as the decision variable to construct the coarse scheduling related constraints in step (1). The objective function is to minimize the station injection / distribution deviation. Finally, a scheduling optimization model with a "fine + coarse" mixed time scale is formed to obtain a mixed scheduling plan, in which the first Δd days are the fine scheduling plan and the next (TD-Δd) days are the daily scheduling plan.

[0045] (3) If the coarse scheduling period TD-Δd>0, take the daily scheduling plan solved in step (2) as the known condition, update the daily scheduling period TD=TD-Δd, the total number of days of fine scheduling time Δd=min(Δd,TD), and the planned injection / sub-transmission volume of each station, and repeat step (2); otherwise, end the calculation, splice the fine scheduling plans of each iteration in time, and combine them into a scheduling plan for a complete cycle.

[0046] Furthermore, the above solution process is represented by a mathematical model:

[0047] (1) Dynamic Collection

[0048] The model uses a mixed discrete time representation, where T represents a set of time points, T D Represents a set of detailed time nodes; T A represents a set of coarse time nodes, T = T D ∪T A . T D The last time node in the set is T A The first node of the time window, and the number of time windows is one less than the number of time nodes.

[0049] N represents the station set, N R represents the set of injection stations, N D represents the set of substations, N = N R ∪N D .

[0050] I represents a batch set. n represents the dynamic batch set that may pass through station n in the whole cycle, I t,n The dynamic set of batches that may pass through station n within time window t. Dynamic set I n and I t,n It needs to be updated based on the daily scheduling results of the previous iteration. and Indicates the batch oil head and tail position at each time node in the last iteration of the daily scheduling plan. If in the last round of daily scheduling plan, the oil tail of batch i does not exceed station n at the beginning of the fine scheduling cycle, and its oil head exceeds station n at the end, then the batch i is in the dynamic batch set of station n, see formula (1). If the oil tail of batch i does not exceed station n at the beginning of the fine scheduling time window t, and its oil head exceeds station n at the end, then the batch i is in the dynamic set I t,n The purpose of introducing this batch dynamic set is to reduce the binary variables associated with the subscripts t, n, and i in the subsequent modeling, reduce the complexity of model solution, and improve the solution efficiency.

[0051]

[0052] (2) Objective function

[0053] The objective function is to minimize the sum of injection and download deviations of each batch at the station, where They represent the volume of injection and distribution batch i at station n within time window t respectively; They respectively represent the planned injection and distribution volumes of batch i at station n within time window t.

[0054]

[0055] (3) Batch volume constraints

[0056] The variable RC in formula (4) t,i and LC t,i Used to represent the oil head coordinate and oil tail coordinate of batch i at time point t. The oil head coordinate of batch i is calculated by summing the volumes of all batches with i′≥i in the same pipeline. Formula (5) obtains the left coordinate by subtracting the batch volume from the right coordinate of the batch. Formula (6) is used to characterize the conservation of batch volume. Specifically, for non-initial time nodes t>0, the volume of batch i at time point t is equal to the volume of batch i at the previous time point plus the total volume of batch i injected within the time window t-1, minus the total volume of the distribution. The volume of batch i at the initial time node t=0 is equal to the known Parameters. Due to the incompressibility of oil, Equation (7) limits the total volume of all batches to be equal to the pipeline capacity v L .

[0057]

[0058]

[0059] (4) Fine Scheduling Cycle Operation Logic Constraints

[0060] Binary variables Indicates whether station n is injecting or distributing batch i within time window t. For fine scheduling time window t<|T D |, only when the oil head of batch i at the starting time node t exceeds that of station n and the oil tail at the ending node t+1 does not exceed that of station n, can station n inject or distribute the batch. Therefore, each station can only inject or distribute a single batch in the same time window, and the operation volume cannot exceed the product of the upper and lower limits of the operation flow and the length of the fine time window Δdt. Formulas (8)-(10) are the injection operation logic constraints in the fine scheduling time window, and formulas (11)-(13) are the distribution operation logic constraints, where The upper and lower limits of the injection flow and the lower and upper limits of the distribution flow of station n are respectively represented. Formula (14) represents the logic constraint of starting distribution. If station n does not have distribution batch i in time window t-1, but is distributing batch i in time window t, then station n is regarded as starting distribution batch i at time node t. The binary variable =1. In order to reduce the start-stop and distribution operations of the pipeline section and improve the stability of pipeline operation, the number of times each station starts and distributes a single batch needs to be controlled within m times, as shown in formula (15).

[0061]

[0062] (5) Coarse Scheduling Cycle Operation Logic Constraints

[0063] Since the results of the coarse scheduling cycle will be brought into the fine scheduling model for further update in the next round of iteration, the coarse scheduling cycle only considers the rough injection and distribution process constraints, and relaxes or ignores the logical constraints of the fine scheduling part, so as to achieve the purpose of reducing the complexity of model solution. For example, for the coarse scheduling time window t∈T A For \|T|, only when the oil tail of batch i at the start time node t does not exceed that of station n and the oil head at the end node t+1 exceeds that of station n, station n can inject or distribute the batch. Therefore, each station can inject or distribute multiple batches in the same time window, and the total operation volume in the time window cannot exceed the upper limit of the operation flow multiplied by the length of the coarse time window Δt. Formulas (16)-(19) are the injection operation logic constraints in the fine scheduling time window, and formulas (20)-(23) are the distribution operation logic constraints. Formula (24) limits the maximum volume of batch i distributed by station n in the coarse scheduling time window t, which cannot exceed the total volume of the batch before the station plus the total injection volume of the upstream station of station n minus the total distribution volume.

[0064]

[0065]

[0066] (4) Pipeline flow constraints

[0067] In any time window t<|T|, the first pipe section delivers volume FS t,0 is equal to the total injection volume of the first station, as shown in formula (25); while for the remaining pipe sections 1≤n<|N|, the transport volume FS t,n Equal to the upstream pipe section conveying volume FS t,n-1 Add the total injection volume of station n minus the total distribution volume of station n, as shown in formula (26). Formula (27) limits the fine scheduling time window t<|T D |The flow rate of the pipe section within the pipe section is within the upper and lower limits of the flow rate Formula (28) limits the pipeline flow rate within the rough scheduling time window to be less than the upper limit of the pipeline flow rate, and the lower limit is no longer constrained. Therefore, the rough scheduling period does not need to introduce binary variables Reduce the size of discrete variables.

[0068]

[0069] In the following, a specific example is solved by the above method and compared with the prior art to illustrate the technical effect of the method.

[0070] A case study is conducted on a refined oil pipeline system in the western region. There are 7 stations along the pipeline, of which S0 is an injection station and the rest are distribution stations. The basic data of the stations are shown in Table 1, and the pipeline operation flow limit range is shown in Table 2.

[0071] Table 1 Basic data of the station

[0072] Station No. Station Type <![CDATA[Volume coordinates (m 3 )]]> <![CDATA[Flow upper limit (m 3 / h)]]> <![CDATA[Lower flow limit (m 3 / h)]]> S0 Injection station 0 750 750 S1 Substation 20001 200 200 S2 Substation 43484 200 100 S3 Substation 56842 300 100 S4 Substation 84808 120 120 S5 Substation 152652 1100 200 S6 Substation 221124 1471 525

[0073] Table 2 Pipe segment operating flow limit range

[0074] Pipe segment number <![CDATA[Flow rate upper limit (m 3 / h)]]> <![CDATA[Lower flow limit (m 3 / h)]]> Pipe0 1471 525 Pipe1 1471 525 Pipe2 1471 350 Pipe3 1471 120 Pipe4 1471 525 Pipe5 1471 525

[0075] The dispatching cycle is about 25 days. At the initial moment, there are 4 old batches in the pipeline. The order of batch oil products is 0# diesel (B0: 16960m 3 )、92# gasoline (B1:1040m 3 )、0# diesel(B2:8956m 3 )、92# gasoline (B3:1015m 3 )、0# diesel(B4:89658m 3 )、92# gasoline (B5:36607m 3 )、0# diesel(B6:66888m 3 ), the planned injection batch sequence in this cycle is 0# diesel (B6) - 92# gasoline (B7) - 92# component gasoline (B8) - 0# diesel (B9) - 92# component gasoline (B10). The planned injection and distribution volume of each batch of oil products at each station are shown in Table 3.

[0076] Table 3 Planned oil injection volume and distribution volume Unit: m 3

[0077]

[0078] The fine time window period is set to 3 days, the fine time window length is 3 hours, and a total of 6 iterations are performed. Python and Gurobi11.0.3 solver are used for solving, and the total solving time is 13 seconds. Compared with the prior art (Table 4), the present invention greatly shortens the scheduling plan preparation time, and can achieve zero deviation, which greatly improves the scheduling decision efficiency. The scale comparison between the present invention and the complete model is shown in Table 5. It can be seen that the number of binary variables in the single iteration model of the present invention is much lower than that of the complete model. This is because in each round of iteration, the proposed method will make full use of the solution information of the previous round of rough scheduling cycle to construct a dynamic set of time, station field, and batch, guide the removal of a large number of redundant and invalid decision variables and constraints in the fine scheduling cycle, and accelerate the convergence of the fine scheduling model without losing optimality, greatly improving the solution speed of large-scale, long-cycle, and multi-batch scheduling plans.

[0079] The complete and detailed scheduling plan obtained is shown in Figure 2 The horizontal axis in the figure represents time, the left vertical axis represents the batch distribution status along the pipeline at the initial moment, the right vertical axis represents the distance between each station and the first station, and the black oblique line represents the migration process of the batch interface. The rectangular bars represent the batch injection / distribution operations of the stations along the line, the colors represent the oil products injected / distributed, and the corresponding time span on the horizontal axis represents the start and end time nodes of the operation. Figure 2 It can be seen that the intermediate distribution station basically completes the distribution tasks of each batch through one operation, which is conducive to the smooth operation of the pipeline. The operating flow changes of each station and each pipe section along the pipeline are shown in Figure 3-4. The flow of each station and pipe section always changes within the allowable limit range, which meets the process requirements.

[0080] Table 4 Comparison of solution results

[0081] The present invention (decomposition iteration) Complete model solution Manual compilation Calculation time 13 seconds 2161 seconds 1-2 days Plan Deviation 0 6525 0

[0082] Table 5 Model scale comparison

[0083]

[0084] The present invention uses a "fine + coarse" mixed time expression to construct a method and system for accelerating the solution of sequential liquid pipeline scheduling decisions based on nested multi-time scales, aiming to effectively solve the problem of "combinatorial explosion" in the search space caused by the presence of a large number of binary variables when compiling a large-scale refined oil pipeline system scheduling plan, thereby laying a technical foundation for the dynamic adjustment of the refined oil pipeline scheduling plan. The proposed method uses the solution information of a small-scale coarse scheduling cycle to quickly construct a dynamic set of time, station field, and batch, and embeds it in the decision variable and logical constraint construction process of the next round of fine scheduling cycle, guiding the removal of a large number of redundant and invalid search spaces in the fine scheduling cycle, accelerating the convergence of the fine scheduling model without losing optimality, and greatly improving the solution speed of large-scale, long-cycle, and multi-batch scheduling plans. According to case verification, the solution speed can be increased by 99.40%.

[0085] The system provided in the above embodiment obtains the basic pipeline information and preset transportation batch information of the finished oil pipeline with multiple injection points, and forms batch scheduling information based on the basic pipeline information and the preset transportation batch information; then the batch scheduling information is input into the preset scheduling model for time series simulation, and the solution is performed in combination with the set objective function and constraint conditions to calculate the remaining transportation capacity information that meets the preset target. Compared with the prior art, the complexity of the multi-injection point pipeline system and the constraints such as the consignor's transportation volume, arrival time, and adjacent batch requirements can be fully considered to quickly and accurately calculate the remaining transportation capacity information, so as to facilitate timely announcement to the public and improve social and economic benefits.

[0086] The above system can be implemented in a computer device in the form of hardware or software, so that the computer device can implement the remaining transportation capacity assessment method of a multi-injection point refined oil pipeline in the embodiment of the present application. The details of the method can be referred to the description of the aforementioned embodiment, and will not be repeated here.

[0087] In an embodiment of the present application, a computer-readable storage medium is also provided, in which a computer program is stored. When a computer device executes the computer program, the remaining transportation capacity assessment method of a multi-injection point refined oil pipeline in an embodiment of the present application is implemented.

[0088] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-mentioned system (device) and module unit can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0089] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system device embodiments described above are only schematic. For example, the division of the above-mentioned module units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0090] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (Processor) to perform some steps of the above-mentioned method of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0091] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for scheduling sequential liquid pipelines based on multiple time scale nesting, characterized in that: The method comprises: (1) obtaining input information, including: pipeline and station information of the sequential liquid transport pipeline, scheduling cycle and preset batch plan information; (2) Based on the input information, taking day as the basic time unit of rough scheduling, and using the preset first objective function and first constraint condition, solve the daily scheduling plan of each station within the rough scheduling time window. The initial value of the rough scheduling time window is the scheduling period; (3) Perform the following iterative processing: Based on the current daily scheduling plan, the coarse scheduling time window is updated by subtracting a fine scheduling time window of a set length from the current coarse scheduling time window; Among them, for the fine scheduling time window, a preset time unit of less than one day is used, and for the updated rough scheduling time window, the time unit is still one day. The second objective function and the second constraint condition are used to solve the fine scheduling plan of the operation quantity of each station in the fine time window and the rough scheduling plan in the rough scheduling time window; Update the daily scheduling plan according to the rough scheduling plan within the rough scheduling time window; Repeat this step until the current coarse scheduling time window can no longer be divided into a fine scheduling time window, and use the remaining time of the coarse scheduling time window as a preset time unit less than one day, and use the second objective function and the second constraint condition to solve the fine scheduling plan of the operation volume of each station; (4) The detailed scheduling plans of each station obtained by the above iterative processing are superimposed in chronological order to output the final scheduling plan covering the complete scheduling cycle.

2. The method for sequential liquid pipeline scheduling based on multi-time scale nesting according to claim 1 is characterized in that: The first objective function is: the deviation between the actual injection / distribution volume accumulated at the station in the rough scheduling time window and the planned injection / distribution volume is minimized.

3. The method for sequential liquid pipeline scheduling based on multi-time scale nesting according to claim 2 is characterized in that: The first constraint conditions include: batch volume constraint, daily maximum injection / distribution volume constraint, daily maximum pipeline transport flow constraint, and operation logic constraint.

4. The method for sequential liquid pipeline scheduling based on multi-time scale nesting according to claim 1 is characterized in that: The second objective function is: the deviation between the actual injection / distribution volume of the station accumulated in the fine scheduling time window and the coarse scheduling time window and the planned injection / distribution volume is minimized.

5. The method for sequential liquid pipeline scheduling based on multi-time scale nesting according to claim 4 is characterized in that: The second constraint conditions include: batch volume constraint, maximum injection / distribution flow constraint, minimum injection / distribution flow constraint, maximum pipeline transportation flow constraint, minimum pipeline transportation flow constraint, and operation logic constraint.

6. A method system for sequential liquid pipeline scheduling based on multi-time scale nesting, characterized in that: The system comprises: An input module is used to obtain input information, including: pipeline and station information of the sequential liquid delivery pipeline, scheduling cycle and preset batch plan information; An initial calculation module is used to solve the daily scheduling plan of each station within the rough scheduling time window according to the input information, taking the day as the basic time unit of the rough scheduling, and using the preset first objective function and the first constraint condition. The initial value of the rough scheduling time window is the scheduling period; The iterative processing module is used to perform the following iterative processing: based on the current daily scheduling plan, a fine scheduling time window of a set length is subtracted from the current coarse scheduling time window to update the coarse scheduling time window; wherein, for the fine scheduling time window, a preset time unit of less than one day is used, while for the updated coarse scheduling time window, the time unit is still day, and the second objective function and the second constraint are used to solve the fine scheduling plan of the operating quantity of each station in the fine time window and the coarse scheduling plan in the coarse scheduling time window; the daily scheduling plan is updated according to the coarse scheduling plan in the coarse scheduling time window; the processing of this step is repeated until the current coarse scheduling time window can no longer be divided into fine scheduling time windows, and the remaining time of the coarse scheduling time window is used in a preset time unit of less than one day, and the second objective function and the second constraint are used to solve the fine scheduling plan of the operating quantity of each station; The output module is used to superimpose the detailed scheduling plans of each station obtained by the aforementioned iterative processing in chronological order and output the final scheduling plan covering the complete scheduling cycle.

Images