CN-121998375-A - Ship-field-vehicle-machine cooperative scheduling optimization method for dry bulk port

CN121998375ACN 121998375 ACN121998375 ACN 121998375ACN-121998375-A

Abstract

The invention discloses a ship-field-vehicle-machine collaborative scheduling optimization method for a dry bulk cargo port, which relates to the technical field of port logistics scheduling or the field of intelligent transportation systems and comprises the following steps of S1, acquiring ship date, business bill and operation plan data of a ship, preprocessing the data, S2, predicting the shipment mode, shipment period and stack average shipment period of the ship through a combined prediction model based on the preprocessed data, and generating ship loading and unloading operation data. The core of the invention is to construct a full-flow optimization system which takes data prediction as a drive and takes multi-module intelligent cooperation as a context, including but not limited to a prospective shipping behavior prediction method based on time sequence prediction, a cooperation allocation mechanism of berths and storage yard resources, a mechanical scheduling and shipping dispatching strategy based on multi-objective optimization and a shipping port scheduling method.

Inventors

WANG TAO
TIAN ZHENDONG
ZHANG TAO
Mao Ruiwen
YANG YUGANG
WANG JINGYING
LI YUJUAN
DING XIAOQIN
Liang Yiyao
WANG YINGQI
Lv Zhengxing
JIANG NAN

Assignees

山东港口日照港集团有限公司
山东港口科技集团日照有限公司
海博泰科技(青岛)有限公司

Dates

Publication Date: 20260508
Application Date: 20260309

Claims (10)

1. The ship-field-vehicle-machine cooperative scheduling optimization method for the dry bulk port is characterized by comprising the following steps of: (1) Data acquisition and preprocessing: Docking the port production management system, acquiring date, business bill, operation plan, equipment state, storage yard capacity and traffic flow data of the ship, and cleaning, normalizing and supplementing the data with missing values; (2) Prediction of shipping behavior: constructing a combined prediction model based on a gray prediction model, a BP neural network and a multiple linear regression model, carrying out multi-layer prediction on a cargo shipping mode, a shipping period and a stack average shipping period, and outputting a prediction result; (3) And (3) collaborative decision of berth and storage yard: The ship scheduling module synthesizes the stockyard stockpile information and the berth resource information based on a berth-yard cooperative scheduling algorithm to generate an optimal berth allocation scheme and a ship unloading stack position scheme; (4) Scheduling and planning mechanical resources: According to the berth allocation and stacking scheme, constructing a mixed integer planning model aiming at minimizing the operation cost and maximizing the equipment utilization rate, and generating a fixed machine and flow machine proportioning scheme for a gantry crane, a stacker-reclaimer and a loader; (5) And (3) shipping and dispatching and executing: Based on the transportation scheme and the mechanical scheduling scheme, combining the real-time vehicle state, realizing one-key intelligent assignment of transportation machinery and tasks by utilizing a multi-objective optimization algorithm, and dynamically adjusting in real time according to the site; (6) Dispatching and monitoring of the air traffic harbor: according to the harbor plan and real-time traffic information in the harbor, constructing a scheduling model combining queuing theory and mixed integer planning, dynamically generating a vehicle release sequence and a loader scheduling instruction, and outputting operation efficiency, cost and performance indexes when the vehicle stops in the harbor; And (3) to (6) perform real-time data interaction and feedback through a standardized data interface to form a full-flow closed-loop scheduling system, so that collaborative optimization of port ships, yards, vehicles and machinery is realized.
2. The ship-field-vehicle-machine collaborative scheduling optimization method for the dry bulk cargo port, which is disclosed in claim 1, is characterized in that the combined prediction model adopts a two-layer structure, wherein the first layer performs parallel prediction by using gray prediction, BP neural network and multiple linear regression model respectively, and the second layer calculates dynamic weight according to the prediction error of the first layer and outputs a weighted prediction result.
3. The ship-field-vehicle-machine cooperative scheduling optimization method for the dry bulk cargo port, which is disclosed in claim 1, is characterized in that the berth-yard cooperative scheduling algorithm is used for realizing bidirectional matching of berths and stacks by taking ship dimensions, water depths, cargo species and yard distance constraints into consideration, wherein the ship waiting time is minimized, the berth utilization rate is maximized as an optimization target.
4. The ship-field-vehicle-machine collaborative scheduling optimization method for the dry bulk cargo port, which is characterized in that the stack position planning algorithm comprehensively considers the available area of a stack field, the stack height limit, the shipping distance, the stacking time and the ship unloading sequence constraint, and adopts simulated annealing or genetic algorithm to solve.
5. The ship-field-vehicle-machine collaborative scheduling optimization method for a dry bulk port according to claim 1, wherein the mechanical resource scheduling planning adopts a mixed integer planning model, constraint conditions comprise mechanical capacity, operation time period, regional accessibility and operation proportion, multiple sets of alternatives are output, and optimal selection is performed based on cost-efficiency weights.
6. The ship-yard-vehicle-machine collaborative scheduling optimization method for the dry bulk port, which is disclosed in claim 1, is characterized in that the shipping dispatching module performs resource allocation through a double-layer optimization model, wherein the first layer completes matching binding of vehicles and tasks, and the second layer optimizes a working route and a time window based on a nearby allocation principle.
7. The ship-field-vehicle-machine collaborative scheduling optimization method for the dry bulk cargo port, which is disclosed in claim 1, is characterized in that the vehicle entering port rhythm is adjusted by utilizing a dynamic release strategy by combining real-time data of vehicle queuing length, road congestion degree and working state of a loader through the vehicle transporting port scheduling model, so that the port traffic efficiency is optimized.
8. The ship-field-vehicle-machine collaborative scheduling optimization method for a dry bulk port according to claim 1, wherein the whole collaborative model is solved by a Benders decomposition algorithm, a main problem is used for processing integer variables of berth allocation and stack allocation, and a sub problem is used for solving continuous variables of operation flow and resource proportion.
9. The ship-field-vehicle-machine collaborative scheduling optimization method for the dry bulk cargo port, according to claim 1, is characterized in that the system dynamically generates cutting constraint according to real-time operation data after each Benders iteration, updates the main problem and realizes real-time convergence and self-adaptive adjustment of a model.
10. The ship-field-vehicle-machine collaborative scheduling optimization method for the dry bulk cargo port according to claim 1, wherein the system is provided with a multi-scheme decision mechanism, and can output a scheduling scheme with optimal cost, optimal efficiency or balance type according to weight parameters for a scheduler to select and execute.

Description

Ship-field-vehicle-machine cooperative scheduling optimization method for dry bulk port Technical Field The invention relates to the technical field of port logistics scheduling or the field of intelligent traffic systems, in particular to a ship-field-vehicle-machine cooperative scheduling optimization method for a dry bulk cargo port. Background In the field of dry bulk port operation, the conventional dispatching management mode generally regards links such as ship berthing, yard planning, horizontal transportation, mechanical control and the like as independent modules for management. Each module often adopts an independent dispatching system or relies on the experience of a dispatcher to make a decision, and the sectional management mode simplifies the operation flow of each link to a certain extent, but causes the cooperative breakage of the whole operation chain. The lack of effective information interaction mechanism among each operation unit forms a data barrier, so that the port operation efficiency is difficult to realize the integral breakthrough. In the actual operation process, the traditional mode is often focused on the improvement of the equipment utilization rate of a single link, and the overall efficiency of the multi-equipment collaborative operation is ignored. Due to the lack of efficient data sharing and coordination mechanisms between systems, yard planning often fails to adequately account for the operating efficiency of ship unloaders or the transport capacity of trucks, which can easily lead to the occurrence of ship port congestion. Meanwhile, lack of linkage between the port dredging plan and stock positions of a storage yard and states of a loader often causes long-time waiting or idle running of a truck, and resource utilization efficiency is seriously reduced. The existing system is mostly based on static planning schemes, and is difficult to cope with inherent uncertainty in the port operation environment. The arrival time and the cargo quantity of the ship often deviate from the expected arrival time, and the yard space allocation scheme cannot be dynamically adjusted accordingly, so that the pre-allocation scheme is not feasible or serious waste of yard space resources is caused. The contradiction between the local optimization and the overall efficiency directly leads to the reduction of the utilization rate of the operation resources and the increase of the operation cost. Although some ports have been attempting to introduce informationized management systems in recent years, most systems still stay in independent scheduling management within each module, and lack an overall collaborative scheduling method. The existing solution realizes digital monitoring of single links, such as ship berthing management, storage yard partition monitoring, vehicle transportation scheduling or mechanical resource allocation, but fails to break the data barriers among modules, and cannot form a unified and optimized decision system. These systems are often limited to statistical analysis of historical operating data, lack modeling capabilities for multi-link association constraints, and are more difficult to cope with real-time dynamically changing harbour operating environments. When complex working conditions such as ship arrival time fluctuation, cargo quantity deviation, stock yard sudden congestion and the like are faced, the rapid reconfiguration of cross-module resources cannot be realized, a globally optimal emergency scheduling scheme cannot be provided, and finally the overall operation efficiency of the port is difficult to further improve. Therefore, developing an intelligent dispatching method capable of realizing deep fusion and collaborative optimization of ships, sites, vehicles and machines has become an urgent need for pushing the dry bulk port to intelligent and green transformation and upgrading. In the field of intelligent port scheduling, the prior art has proposed various technical schemes for improving the operation efficiency. For example, a patent "a car-harbor-ship cooperative scheduling method and system" (CN 111785028 a) published by the university of maritime at Shanghai "establishes five working links of berth, quay bridge, automatic guided car, yard bridge and yard through a chain structure, and adjusts a scheduling plan based on the predicted arrival time of the ship and the vehicle. Although the method realizes the linkage of multiple links to a certain extent, the chain model is relatively fixed, the links are tightly coupled, and when the method faces the common uncertainty (such as ship arrival time delay and variable cargo types) in the operation of a dry bulk cargo port, the method lacks sufficient dynamic adjustment capability and global optimization view angle. Another "an intelligent port work vehicle dispatch system and dispatch robot" (CN 120746080 a) disclosed by Sun Zhijun et al, uses two-part map matching and multi-objective optimization me