Port operation optimization method based on big data and air rail transportation technology fusion

Abstract

The invention provides a port operation optimization method based on big data and air rail transportation technology fusion, and belongs to the technical field of intelligent ports, and the method comprises the steps: obtaining port geographic data, setting a four-layer intelligent system, and collecting basic data based on the four-layer intelligent system; performing multi-modal data fusion processing on the basic data to obtain multi-modal fusion data, performing big data analysis based on the multi-modal fusion data, and constructing an intelligent decision model according to a big data analysis result; the dynamic scheduling of the air rail transport system is guided based on the output of the intelligent decision model, the dynamic scheduling is executed, the execution result is evaluated according to the real-time basic information obtained by the four-layer intelligent system, and then the port operation link is optimized according to the evaluation result; and designing a continuous learning mechanism based on an optimization result, and carrying out comprehensive optimization on the intelligent decision-making system and a port transportation link. And the port competitiveness is enhanced, and the data security and the autonomous controllable level are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent port technology, and in particular to a port operation optimization method based on the integration of big data and sky-rail transportation technology. Background Art

[0002] With the rapid growth of global trade, the importance of ports as logistics hubs has become increasingly prominent. Traditional port operation models face bottlenecks in efficiency, safety and energy efficiency, and are in urgent need of improving operation levels through intelligent technologies. In addition, in the current construction of intelligent ports, data security awareness and autonomous controllability of core technologies need to be improved.

[0003] Therefore, the present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology. Summary of the invention

[0004] The present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology. The method senses the port operation status through a four-layer intelligent system, collects multimodal data (such as video, sensor data, etc.), performs fusion analysis, builds an intelligent decision-making model, dynamically dispatches the air-rail transportation system, and uses the evaluation results to drive the system to continuously learn, optimize port operations, improve efficiency, safety and energy efficiency, and ultimately realize intelligent port operations, enhance the autonomous controllability of core technologies, ensure data security, improve port operation efficiency, safety and energy efficiency, enhance port competitiveness, and improve data security and autonomous controllability.

[0005] The present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology, comprising: Step 1: Obtain port geographic data, set up a four-layer intelligent system, and collect basic data based on the four-layer intelligent system; Step 2: Perform multimodal data fusion processing on the basic data to obtain multimodal fusion data, perform big data analysis based on the multimodal fusion data, and build an intelligent decision model according to the big data analysis results; Step 3: guiding the dynamic scheduling of the air-rail transportation system based on the output of the intelligent decision-making model, and executing the dynamic scheduling, evaluating the execution results according to the real-time basic information obtained by the four-layer intelligent system, and then optimizing the port operation links according to the evaluation results; Step 4: Design a continuous learning mechanism based on the optimization results to comprehensively optimize the intelligent decision-making system and port transportation links.

[0006] The present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology, which obtains port geographic data, sets up a four-layer intelligent system, and collects basic data based on the four-layer intelligent system, including: Obtain port geographic data, deploy an air-to-rail transportation system based on the port geographic data, capture port operation scenes, install ground IoT sensors and ship Beidou positioning terminals, collect port operation data, and integrate the port operation scenes and port operation data to derive a perception layer; Build a dedicated network, deploy edge computing nodes according to the port geographic data, establish a data transmission channel, and derive the transport layer; Deploy logistics optimization engine, safety warning engine and energy efficiency management engine to obtain the decision-making layer; Provide visual command platform, mobile terminal app and API open interface to check port operation status and obtain application layer; Combining the perception layer, transmission layer, decision-making layer and application layer, a four-layer intelligent system is set up to collect basic data.

[0007] The present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology, which performs multimodal data fusion processing on the basic data to obtain multimodal fusion data, performs big data analysis based on the multimodal fusion data, and constructs an intelligent decision model according to the big data analysis results, including: Attach a unified timestamp to all basic data and perform spatial registration based on the coordinate system of the port digital twin; and performing standardization processing on the basic data based on the spatial registration, and extracting visual features, sensor features, and operation features based on the standardization processing result; Performing multimodal data fusion processing on the visual features, sensor features, and operation features to obtain multimodal fusion data; Based on multimodal fusion data, big data analysis technology is used to conduct logistics optimization analysis, safety warning analysis and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model.

[0008] The present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology. Based on multimodal fusion data, big data analysis technology is used to perform logistics optimization analysis, safety warning analysis and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model, including: Construct the logistics optimization engine objective function based on the logistics optimization analysis results: , where W represents the objective function of the logistics optimization engine; Indicates the weight coefficient of the ship waiting time; represents the waiting time of the ship at time tt; represents the utilization rate of port resources at time t; The weight coefficient representing resource utilization; represents the maximum available capacity of port resources; T represents the overall time range of objective function calculation; , where A represents the objective function of the security early warning engine; The weight coefficient representing the accuracy of anomaly detection; Represents the number of anomalies correctly detected by the model; Indicates the number of normal targets correctly detected by the target detection model in the security early warning engine; Indicates the number of anomalies that are incorrectly detected by the target detection model in the security early warning engine; Indicates the number of anomalies that the target detection model in the security early warning engine failed to detect; The weight coefficient representing the false alarm rate.

[0009] The present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology. Based on multimodal fusion data, big data analysis technology is used to perform logistics optimization analysis, safety warning analysis and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model, which also includes: , where N represents the energy efficiency management engine objective function; Represents the weight coefficient of energy consumption cost; represents the energy consumption cost of the port at time t; represents the energy efficiency of the port at time t; Represents the weight coefficient of energy utilization efficiency; It represents the theoretical maximum value of port energy utilization; An intelligent decision-making model is constructed by integrating the objective function of the logistics optimization engine, the objective function of the safety warning engine and the objective function of the energy efficiency management engine.

[0010] The present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology, guiding the dynamic scheduling of the air-rail transportation system based on the output of the intelligent decision-making model, and executing the dynamic scheduling, evaluating the execution results according to the real-time basic information obtained by the four-layer intelligent system, and then optimizing the port operation links according to the evaluation results, including: Determine the corresponding scheduling situation according to the output of each engine in the intelligent decision-making model, and obtain the dynamic scheduling of the sky rail transportation system by combining all the scheduling situations; Execute the dynamic scheduling, collect real-time basic information, evaluate the execution results, identify execution problems in the evaluation results, and optimize the port operation links based on the execution problems.

[0011] The present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology, which determines the corresponding scheduling situation according to the output of each engine in the intelligent decision-making model, and obtains the dynamic scheduling of the air-rail transportation system by integrating all the scheduling situations, including: The logistics optimization engine outputs the coordinated scheduling strategy of berths, quay cranes and container trucks; the safety warning engine outputs abnormal behavior detection results and warning information; and the energy efficiency management engine outputs energy consumption hotspot predictions and energy optimization suggestions; The aerial rail transport path and transport strategy are dynamically adjusted according to the collaborative scheduling strategy, the abnormal area is determined and the transport priority is adjusted according to the abnormal behavior detection results and early warning information, and the energy consumption distribution of the transport task is optimized according to the energy consumption hotspot prediction and energy optimization suggestions, thereby obtaining the dynamic scheduling of the aerial rail transport system.

[0012] The present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology, designs a continuous learning mechanism based on the optimization results, and comprehensively optimizes the intelligent decision-making system and port transportation links, including: Based on the application layer in the four-layer intelligent system, the operation and maintenance personnel's evaluation of the optimization results is collected, and a continuous learning mechanism is designed based on the evaluation; The intelligent decision-making system is updated according to the continuous learning mechanism, and the port transportation links are comprehensively optimized based on the updated intelligent decision-making model.

[0013] Compared with the existing technology, the beneficial effects of this application are as follows: through a four-layer intelligent system, the port operation status is perceived, multimodal data (such as video, sensor data, etc.) is collected, fusion analysis is performed, an intelligent decision-making model is constructed, the air-rail transportation system is dynamically dispatched, and the evaluation results drive the system to continuously learn, optimize port operations, improve efficiency, safety and energy efficiency, and ultimately realize intelligent port operations, enhance the autonomous controllability of core technologies, ensure data security, improve port operation efficiency, safety and energy efficiency, enhance port competitiveness, and improve data security and autonomous controllability.

[0014] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0015] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 It is a flow chart of a port operation optimization method based on the integration of big data and sky-rail transportation technology provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0018] Embodiment 1: The embodiment of the present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology, such as Figure 1 As shown, including: Step 1: Obtain port geographic data, set up a four-layer intelligent system, and collect basic data based on the four-layer intelligent system; Step 2: Perform multimodal data fusion processing on the basic data to obtain multimodal fusion data, perform big data analysis based on the multimodal fusion data, and build an intelligent decision model according to the big data analysis results; Step 3: guiding the dynamic scheduling of the air-rail transportation system based on the output of the intelligent decision-making model, and executing the dynamic scheduling, evaluating the execution results according to the real-time basic information obtained by the four-layer intelligent system, and then optimizing the port operation links according to the evaluation results; Step 4: Design a continuous learning mechanism based on the optimization results to comprehensively optimize the intelligent decision-making system and port transportation links.

[0019] In this embodiment, the port geographic data is data describing the physical spatial characteristics of the port, including port area maps, shoreline coordinates, water depth data, berth locations and sizes, road network maps, building distribution maps, etc.

[0020] In this embodiment, the four-layer intelligent system includes a perception layer, a transmission layer, a decision layer, and an application layer.

[0021] In this embodiment, the basic data includes visual data, sensor data and operational data. Visual data: deploy low-altitude drone clusters and fixed cameras, collect 4K resolution video streams (30fps), and mark the bounding boxes and behavior labels of key targets (containers, vehicles, and personnel); sensor data: install IoT devices such as temperature and humidity, vibration, and GPS, collect device status and environmental parameters at a frequency of 1Hz, and mark abnormal events (such as equipment overload and leakage); operational data: connect to the port management system, obtain structured data (ship ETA, cargo list, operation plan), and build a spatiotemporal index table.

[0022] In this embodiment, multimodal data fusion processing is the process of integrating multiple types of data together, for example, using a deep learning model to fuse visual features, sensor features, and operational features.

[0023] In this embodiment, the specific process of big data analysis is to use the big data processing platform to perform real-time analysis on the pre-processed data. Data mining, machine learning and other algorithms are used to mine the potential value in the data and provide decision support for port operations. For example, the coordinated scheduling of berths, air rails and container trucks is realized through the logistics optimization engine to improve logistics efficiency; abnormal behavior detection is realized through the safety warning engine to warn potential safety hazards in advance; energy consumption hotspots are predicted through the energy efficiency management engine to optimize energy use.

[0024] In this embodiment, the input of the intelligent decision-making model is multimodal fusion data, and the output is logistics optimization strategy, safety warning information, energy efficiency management suggestions, and the final air-rail transportation scheduling plan. These outputs will be fed back to the four-layer intelligent system to guide port operations.

[0025] In this embodiment, dynamic scheduling is a scheduling plan adjusted according to real-time conditions rather than a pre-set static plan, including the scheduling of the logistics optimization engine, the scheduling of the safety warning engine, and the scheduling of the energy efficiency management engine. For example, due to the early arrival of the "Ocean" ship, the system automatically adjusted the operation plans of the container truck and the quay crane, and started unloading in advance to avoid delays in the berthing of subsequent ships. This is a real-time response to emergencies.

[0026] In this embodiment, the real-time basic information is data collected in real time during the operation of the system, including the operating efficiency of the quay crane, the position and speed of the container truck, the number of containers in the yard, weather conditions, energy consumption data of each device, etc.

[0027] In this embodiment, the execution result evaluation is an evaluation of the dynamic scheduling execution effect, and the measurement indicators include: ship turnaround time, container storage time, average mileage of container trucks, energy consumption, safety accident rate, etc. The system will compare the actual data with the expected target.

[0028] In this embodiment, the optimization is to improve the port operation process based on the execution problem, such as optimizing algorithm parameters, adjusting the scheduling strategy (for example, increasing the number of container trucks or adjusting the route), improving the equipment maintenance process, strengthening safety training, or adjusting the port layout to improve efficiency and reduce congestion.

[0029] In this embodiment, based on the output results of the three engines, the scheduling of the air-rail transportation system is coordinated. The logistics optimization engine determines the collaborative scheduling strategy of berths, quay cranes and container trucks; the safety warning engine identifies anomalies and adjusts transportation priorities; and the energy efficiency management engine optimizes energy consumption distribution. The outputs of the three are integrated to form a dynamic schedule.

[0030] In this embodiment, the setting process of the continuous learning mechanism is to design a feedback loop: the warning results confirmed by the operation and maintenance personnel are transmitted back as new labeled data, and an elastic weight consolidation method is used to retain important parameters when updating the model to prevent catastrophic forgetting.

[0031] In this embodiment, comprehensive optimization is to use the updated intelligent decision-making model to comprehensively optimize the port transportation links, redeploy, deploy the updated intelligent decision-making model into the system, continuously monitor, continuously monitor the system operation, and collect new data for further optimization of the model, cyclically improve, repeat the process of evaluation, update and optimization, and continuously improve the performance of the system. After deploying the updated model, the system will re-plan the AGV route and dynamically adjust it according to the real-time situation, striving to achieve the best balance of efficiency, safety and energy saving.

[0032] The working principle and beneficial effects of the above technical solution are: through a four-layer intelligent system to perceive the port operation status, collect multimodal data (such as video, sensor data, etc.), conduct fusion analysis, build an intelligent decision-making model, dynamically dispatch the air-rail transportation system, and evaluate the results to drive the system to continuously learn, optimize port operations, improve efficiency, safety and energy efficiency, and ultimately realize intelligent port operations, enhance the autonomous controllability of core technologies, ensure data security, improve port operating efficiency, safety and energy efficiency, enhance port competitiveness, and improve data security and autonomous controllability.

[0033] Embodiment 2: The embodiment of the present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology, which obtains port geographic data, sets up a four-layer intelligent system, and collects basic data based on the four-layer intelligent system, including: Obtain port geographic data, deploy an air-to-rail transportation system based on the port geographic data, capture port operation scenes, install ground IoT sensors and ship Beidou positioning terminals, collect port operation data, and integrate the port operation scenes and port operation data to derive a perception layer; Build a dedicated network, deploy edge computing nodes according to the port geographic data, establish a data transmission channel, and derive the transport layer; Deploy logistics optimization engine, safety warning engine and energy efficiency management engine to obtain the decision-making layer; Provide visual command platform, mobile terminal app and API open interface to check port operation status and obtain application layer; Combining the perception layer, transmission layer, decision-making layer and application layer, a four-layer intelligent system is set up to collect basic data.

[0034] In this embodiment, the perception layer is composed of an air-rail transport system (equipped with high-definition cameras / lidar), ground-based IoT sensors (pressure / temperature and humidity / vibration), and a ship Beidou positioning terminal; the transmission layer uses 5G private network + edge computing nodes to establish three data transmission channels: high-bandwidth channel: transmission of 4K video stream (50Mbps / channel), low-latency channel: transmission of control instructions (end-to-end latency <10ms), and high-reliability channel: transmission of key status data (99.999% availability); the decision-making layer deploys three core engines, the logistics optimization engine: berth-quay crane-container truck collaborative scheduling based on an improved genetic algorithm, the safety warning engine: application of YOLOv7+Transformer model to realize abnormal behavior detection, and the energy efficiency management engine: prediction of energy consumption hotspots through digital twin simulation; the application layer provides a visual command platform, mobile terminal APP, and API open interface.

[0035] In this embodiment, the sky rail transportation system is an automated, efficient port internal transportation system, which usually uses rails to transport goods, including AGV (automatic guided vehicle) systems, rail crane systems, light rail transportation systems, etc.

[0036] In this embodiment, the port operation scene is a video or image record of various operation activities occurring in the port, including video surveillance images of container loading and unloading processes, vehicle driving conditions, personnel activities, and equipment operating status.

[0037] In this embodiment, the ground IoT sensor is a sensor deployed on the ground to collect various environmental and operational data, including temperature sensors, humidity sensors, pressure sensors, weight sensors, video surveillance cameras, etc.

[0038] In this embodiment, the ship Beidou positioning terminal is installed on the ship, and the equipment used for real-time positioning and data transmission includes a Beidou navigation receiver with a data transmission function.

[0039] In this embodiment, the port operation data is various data recording the port operation status including container throughput, ship arrival and departure time, equipment utilization rate, energy consumption data, personnel work efficiency, etc.

[0040] In this embodiment, the dedicated network is a network specially built for the port intelligent system, which is used for data transmission and communication, including 5G-based dedicated networks, private cloud networks, etc. The edge computing node is a computing device deployed close to the data source, which is used for real-time data processing and analysis, including edge servers, edge gateways, etc. The data transmission channel is the path for data transmission between various devices and nodes, including network lines, wireless communication links, etc.

[0041] In this embodiment, the logistics optimization engine, the safety warning engine, and the energy efficiency management engine are software modules used to perform specific tasks. The logistics optimization engine may use algorithms to optimize the loading and unloading and transportation sequence of containers; the safety warning engine may use machine learning algorithms to detect abnormal behavior; and the energy efficiency management engine may optimize port energy consumption through prediction and control.

[0042] In this embodiment, the visual command platform, mobile terminal app, and API open interface are tools for human-computer interaction, including a visual command platform that may display the real-time operating status of the port; a mobile terminal app that may provide mobile monitoring and management functions; and an API open interface that may allow third-party systems to access port data.

[0043] In this embodiment, the port operation status is a comprehensive reflection of the overall operation status of the port including indicators such as throughput, efficiency, safety, and energy consumption.

[0044] The working principle and beneficial effects of the above technical solution are: by constructing a perception layer (sensors, video surveillance), a transmission layer (dedicated network, edge computing), a decision layer (optimization engine) and an application layer (visualization platform), a four-layer intelligent system is formed, and the port geographic data is used to deploy the air-rail system and collect multi-source data for real-time analysis to achieve logistics optimization, safety warning and energy efficiency management. Finally, the application layer provides visualization and convenient interfaces to improve port operation efficiency and safety.

[0045] Embodiment 3: The embodiment of the present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology, performs multimodal data fusion processing on the basic data to obtain multimodal fusion data, performs big data analysis based on the multimodal fusion data, and constructs an intelligent decision model according to the big data analysis results, including: Attach a unified timestamp to all basic data and perform spatial registration based on the coordinate system of the port digital twin; and performing standardization processing on the basic data based on the spatial registration, and extracting visual features, sensor features, and operation features based on the standardization processing result; Performing multimodal data fusion processing on the visual features, sensor features, and operation features to obtain multimodal fusion data; Based on multimodal fusion data, big data analysis technology is used to conduct logistics optimization analysis, safety warning analysis and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model.

[0046] In this embodiment, the port digital twin is a virtual, digital port model that corresponds to the real port environment and can reflect the operating status of the port in real time. For example, a three-dimensional model includes digital representations of all buildings, equipment, roads, etc. in the port area, and can display information such as ship location, container status, and equipment operation status in real time.

[0047] In this embodiment, spatial registration is to align data from different data sources according to a unified coordinate system so that they have a corresponding relationship in space. For example, the target position in the video surveillance screen is matched with the sensor data (for example, GPS coordinates), or the data collected by different sensors (for example, temperature sensor data from different locations) are unified into the coordinate system of the port digital twin.

[0048] In this embodiment, standardization processing is to pre-process the data to make it conform to a unified standard and format to facilitate subsequent analysis and processing, including data cleaning, data normalization, data conversion, etc., including visual data: downsampling to 1024×576 resolution, histogram equalization enhancement, sensor data: continuous parameters (such as temperature) are normalized using Z-score, discrete parameters (such as equipment ID) are built into an embedded lookup table, operational data: categorical fields (such as cargo type) are One-hot encoded, and numerical fields (such as tonnage) are normalized to [0,1].

[0049] In this embodiment, visual features are features extracted from video surveillance images, including the shape, color, motion trajectory, etc. of the target object; sensor features are features extracted from sensor data, including temperature, humidity, pressure, weight, etc.; operational features are features extracted from operational data, including container throughput, ship waiting time, equipment utilization, etc.

[0050] In this embodiment, multimodal fusion data fuses data from different modalities (e.g., vision, sensor, and operation) to obtain a more comprehensive and richer description. For example, the movement trajectory of the container in the video, the container weight collected by the sensor, and the loading and unloading time in the operation data are combined together to obtain a comprehensive description of the container loading and unloading process.

[0051] In this embodiment, the logistics optimization analysis result is the optimal container loading and unloading and transportation plan, the safety warning analysis result is the warning information of potential safety hazards and abnormal events, and the energy efficiency management analysis result is the prediction and optimization suggestions of port energy consumption.

[0052] The working principle and beneficial effects of the above technical solution are: by adding timestamps and spatial coordinates to the basic data, data standardization processing is achieved, and visual, sensor and operational features are extracted. Multimodal data fusion technology integrates these features, and then big data analysis technology is used to perform logistics optimization, safety warning and energy efficiency management analysis. Finally, a four-layer intelligent system is combined to build an intelligent decision-making model to achieve intelligent management of port operations, optimize port logistics, and improve safety and energy efficiency.

[0053] Embodiment 4: The embodiment of the present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology. Based on multimodal fusion data, big data analysis technology is used to perform logistics optimization analysis, safety warning analysis, and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis, and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model, including: Construct the logistics optimization engine objective function based on the logistics optimization analysis results: , where W represents the objective function of the logistics optimization engine; Indicates the weight coefficient of the ship waiting time; represents the waiting time of the ship at time tt; represents the utilization rate of port resources at time t; The weight coefficient representing resource utilization; represents the maximum available capacity of port resources; T represents the overall time range of objective function calculation; , where A represents the objective function of the security early warning engine; The weight coefficient representing the accuracy of anomaly detection; Represents the number of anomalies correctly detected by the model; Indicates the number of normal targets correctly detected by the target detection model in the security early warning engine; Indicates the number of anomalies that are incorrectly detected by the target detection model in the security early warning engine; Indicates the number of anomalies that the target detection model in the security early warning engine failed to detect; The weight coefficient representing the false alarm rate.

[0054] The working principle and beneficial effects of the above technical solution are: by constructing the objective functions of logistics optimization and safety warning engines, the optimization objectives are quantified. The logistics optimization objective function minimizes the ship waiting time and maximizes the resource utilization rate. The safety warning objective function maximizes the anomaly detection accuracy and minimizes the false alarm rate. By optimizing these two objective functions, the system can automatically adjust the parameters to improve the efficiency and safety of port operations, increase port throughput, reduce ship waiting time, and improve resource utilization.

[0055] Embodiment 5: The embodiment of the present invention provides a port operation optimization method based on the integration of big data and sky-rail transportation technology. Based on multimodal fusion data, big data analysis technology is used to perform logistics optimization analysis, safety warning analysis, and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis, and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model, which also includes: , where N represents the energy efficiency management engine objective function; Represents the weight coefficient of energy consumption cost; represents the energy consumption cost of the port at time t; represents the energy efficiency of the port at time t; Represents the weight coefficient of energy utilization efficiency; It represents the theoretical maximum value of port energy utilization; An intelligent decision-making model is constructed by integrating the objective function of the logistics optimization engine, the objective function of the safety warning engine and the objective function of the energy efficiency management engine.

[0056] The working principle and beneficial effects of the above technical solution are: by constructing the objective functions of the three engines of logistics optimization, safety warning and energy efficiency management, the efficiency, safety and energy consumption of port operations are quantitatively evaluated respectively. The three objective functions are combined to build an intelligent decision-making model to achieve multi-objective optimization and reduce energy consumption costs while ensuring safety and efficiency.

[0057] Embodiment 6: The embodiment of the present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology, guiding the dynamic scheduling of the air-rail transportation system based on the output of the intelligent decision-making model, and executing the dynamic scheduling, evaluating the execution results according to the real-time basic information obtained by the four-layer intelligent system, and then optimizing the port operation links according to the evaluation results, including: Determine the corresponding scheduling situation according to the output of each engine in the intelligent decision-making model, and obtain the dynamic scheduling of the sky rail transportation system by combining all the scheduling situations; Execute the dynamic scheduling, collect real-time basic information, evaluate the execution results, identify execution problems in the evaluation results, and optimize the port operation links based on the execution problems.

[0058] In this embodiment, the scheduling status is the specific scheduling instructions or suggestions output by each engine, including the scheduling status of the logistics optimization engine, the scheduling status of the safety warning engine, and the scheduling status of the energy efficiency management engine. For example, the logistics optimization engine prioritizes 100 of the 200 containers of the "Ocean" ship to berth No. 3, 50 to berth No. 4, and the remaining 50 to berth No. 5; at the same time, 10 container trucks are arranged to prioritize the handling of containers at berth No. 3, which includes specific allocation, priority and sequence. The safety warning engine finds that there are vehicles operating illegally near yard No. 8, and recommends temporarily stopping the handling of containers in the area and dispatching security personnel to deal with it. The energy efficiency management engine predicts that the peak electricity consumption will be from 2 to 4 pm, and recommends reducing the operation of empty container trucks and giving priority to the operation of container trucks with sufficient battery power, which includes suggestions on energy management.

[0059] In this embodiment, the execution problem is a problem existing in the scheduling execution found during the evaluation process, such as too long waiting time for a container truck, too many containers piled up in a certain yard, energy consumption exceeding expectations, a minor collision accident, etc.

[0060] The working principle and beneficial effects of the above technical solution are: the output of the intelligent decision-making model (including logistics optimization, safety warning and energy efficiency management engine) formulates a dynamic scheduling plan for the sky-rail transportation system. After the system executes the scheduling, it collects real-time data for evaluation, identifies execution problems and optimizes the port operation links, forms a closed-loop control, realizes the refined management of the sky-rail transportation system, and improves scheduling efficiency.

[0061] Embodiment 7: The embodiment of the present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology, which determines the corresponding scheduling situation according to the output of each engine in the intelligent decision-making model, and obtains the dynamic scheduling of the air-rail transportation system by integrating all the scheduling situations, including: The logistics optimization engine outputs the coordinated scheduling strategy of berths, quay cranes and container trucks; the safety warning engine outputs abnormal behavior detection results and warning information; and the energy efficiency management engine outputs energy consumption hotspot predictions and energy optimization suggestions; The aerial rail transport path and transport strategy are dynamically adjusted according to the collaborative scheduling strategy, the abnormal area is determined and the transport priority is adjusted according to the abnormal behavior detection results and early warning information, and the energy consumption distribution of the transport task is optimized according to the energy consumption hotspot prediction and energy optimization suggestions, thereby obtaining the dynamic scheduling of the aerial rail transport system.

[0062] In this embodiment, the berth-quay crane-container truck collaborative scheduling strategy is an overall strategy that aims to optimize the collaborative work of the three key links of berth, quay crane (container crane) and container truck. For example, assume that a large cargo ship is docked at berth 1. The collaborative scheduling strategy will consider the following factors: berth allocation, allocating berth 1 to the ship, quay crane allocation, allocating a suitable quay crane (for example, a quay crane that can handle containers of different sizes) to berth 1 according to the ship's loading and unloading plan, and container truck scheduling: planning the route of container trucks to ensure that container trucks can deliver containers to the yard or other destinations in a timely manner and avoid conflicts or congestion between container trucks, including optimizing the route of container trucks and arranging the waiting order of container trucks.

[0063] In this embodiment, the abnormal behavior detection results and warning information are abnormal situations detected by the safety warning engine and their corresponding warning information, such as abnormal behavior, speeding of container trucks, driving in prohibited areas, abnormal operation of quay cranes, dumping of containers in the yard, etc. The system will issue an alarm to indicate the location, severity and possible risks of the abnormal behavior, for example: the truck with the license plate number ABC-1234 is speeding in the yard, please pay attention to safety! Or: abnormal vibration occurs in quay crane No. 5, it is recommended to suspend the operation and conduct an inspection.

[0064] In this embodiment, the energy consumption hotspot prediction and energy optimization suggestions are predictions and optimization suggestions of the energy efficiency management engine on energy consumption. For example, the energy consumption hotspot prediction predicts that the power load will reach a peak from 3 to 5 pm due to a large number of container loading and unloading operations. The energy optimization suggestion recommends that during this period, electric-driven container truck operations should be given priority to reduce the use of diesel container trucks; and the operation plan of the quay crane should be adjusted to stagger the power consumption.

[0065] In this embodiment, adjusting the aerial rail transport path and transport strategy is to adjust the path and strategy of the aerial rail transport system (e.g., AGV) according to the above information. For example, if the safety warning engine detects that there is a safety hazard in a certain area, the system will automatically adjust the path of the AGV to avoid the area.

[0066] In this embodiment, abnormal areas and adjustment of transportation priorities are based on the abnormal behavior detection results to determine the areas and tasks that need to be processed first. For example, if congestion is found in a certain yard, the system will give priority to arranging container trucks to transport the containers in the yard to alleviate the congestion.

[0067] In this embodiment, the energy consumption distribution of the transportation task is optimized by adjusting the energy distribution according to the energy consumption hotspot prediction and energy optimization suggestions. For example, during the peak power consumption period, the system will give priority to the use of electric-driven AGVs to reduce energy consumption; or choose a shorter transportation route to reduce energy consumption.

[0068] The working principle and beneficial effects of the above technical solution are: based on the output results of the three engines, the air-rail transportation system scheduling is coordinated, the logistics optimization engine determines the collaborative scheduling strategy of berths, quay cranes and container trucks; the safety warning engine identifies anomalies and adjusts transportation priorities; the energy efficiency management engine optimizes energy consumption distribution, integrates the outputs of the three, forms a dynamic air-rail transportation scheduling plan, improves air-rail transportation efficiency, reduces energy consumption, and improves safety.

[0069] Embodiment 8: The embodiment of the present invention provides a port operation optimization method based on the integration of big data and air-rail transportation technology, designs a continuous learning mechanism based on the optimization results, and comprehensively optimizes the intelligent decision-making system and the port transportation link, including: Based on the application layer in the four-layer intelligent system, the operation and maintenance personnel's evaluation of the optimization results is collected, and a continuous learning mechanism is designed based on the evaluation; The intelligent decision-making system is updated according to the continuous learning mechanism, and the port transportation links are comprehensively optimized based on the updated intelligent decision-making model.

[0070] In this embodiment, the evaluation of the optimization results is a key step to obtain feedback from the operation and maintenance personnel from the application layer. This process is not a simple question of good or bad, but requires a structured feedback mechanism, such as: data collection method, multiple methods can be used to collect evaluations, such as questionnaire survey, design standardized questionnaires, evaluate the improvement of optimization results in terms of efficiency, safety, energy consumption, etc., and allow operation and maintenance personnel to provide specific improvement suggestions, scoring system, set up a scoring system, let the operation and maintenance personnel score the optimization results according to different indicators, for example, 1-5 star evaluation, feedback system, establish an online feedback system, allow operation and maintenance personnel to submit feedback and suggestions at any time, face-to-face interview, conduct face-to-face interviews, and deeply understand the views and suggestions of operation and maintenance personnel on the optimization results, evaluation indicators, evaluation indicators should cover all aspects of the optimization goals, such as: efficiency: container turnover time, ship berthing time, AGV utilization rate, etc. The operation and maintenance personnel are satisfied with the average container turnover time after optimization, which is shortened by 15 minutes, but the AGV path optimization effect in some specific areas is not ideal, and propose modification suggestions.

[0071] In this embodiment, the update is based on the evaluation of the operation and maintenance personnel, the intelligent decision-making system is updated, the collected evaluation data is analyzed, and the aspects that need to be improved and the specific improvement directions are identified. For example, if most operation and maintenance personnel are not satisfied with the AGV path optimization, it is necessary to analyze the reasons, such as unreasonable algorithm parameter settings, inaccurate data, etc. According to the data analysis results, the parameters of the intelligent decision-making model are adjusted, or the algorithm of the model is improved. This may require the participation of data scientists to improve or replace the algorithm used in the intelligent decision-making model, for example, using more advanced machine learning algorithms to improve the prediction accuracy and decision-making ability of the model, and updating or supplementing training data to improve the generalization ability of the model. If deviations are found in the data, these deviations need to be corrected. For example, according to the feedback from the operation and maintenance personnel, the parameters of the AGV path planning algorithm are adjusted, and predictions of traffic congestion are added to improve the effect of path optimization.

[0072] The working principle and beneficial effects of the above technical solution are as follows: the output results of the three engines coordinate the scheduling of the air-rail transport system. The logistics optimization engine determines the coordinated scheduling strategy of berths, quay cranes and container trucks; the safety warning engine identifies anomalies and adjusts transportation priorities; the energy efficiency management engine optimizes energy consumption distribution, integrates the outputs of the three, forms a dynamic air-rail transport scheduling plan, and realizes intelligent and sustainable port operations.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A port operation optimization method based on the integration of big data and air-rail transportation technology, characterized in that: include: Step 1: Obtain port geographic data, set up a four-layer intelligent system, and collect basic data based on the four-layer intelligent system; Step 2: Perform multimodal data fusion processing on the basic data to obtain multimodal fusion data, perform big data analysis based on the multimodal fusion data, and build an intelligent decision model according to the big data analysis results; Step 3: guiding the dynamic scheduling of the air-rail transportation system based on the output of the intelligent decision-making model, and executing the dynamic scheduling, evaluating the execution results according to the real-time basic information obtained by the four-layer intelligent system, and then optimizing the port operation links according to the evaluation results; Step 4: Design a continuous learning mechanism based on the optimization results to comprehensively optimize the intelligent decision-making system and port transportation links.

2. The port operation optimization method based on the integration of big data and sky-rail transportation technology according to claim 1 is characterized in that: Obtain port geographic data, set up a four-layer intelligent system, and collect basic data based on the four-layer intelligent system, including: Obtain port geographic data, deploy an air-to-rail transportation system based on the port geographic data, capture port operation scenes, install ground IoT sensors and ship Beidou positioning terminals, collect port operation data, and integrate the port operation scenes and port operation data to derive a perception layer; Build a dedicated network, deploy edge computing nodes according to the port geographic data, establish a data transmission channel, and derive the transport layer; Deploy logistics optimization engine, safety warning engine and energy efficiency management engine to obtain the decision-making layer; Provide visual command platform, mobile terminal app and API open interface to check port operation status and obtain application layer; Combining the perception layer, transmission layer, decision-making layer and application layer, a four-layer intelligent system is set up to collect basic data.

3. The port operation optimization method based on the integration of big data and sky-rail transportation technology according to claim 1 is characterized in that: Performing multimodal data fusion processing on the basic data to obtain multimodal fusion data, performing big data analysis based on the multimodal fusion data, and building an intelligent decision model according to the big data analysis results, including: Attach a unified timestamp to all basic data and perform spatial registration based on the coordinate system of the port digital twin; and performing standardization processing on the basic data based on the spatial registration, and extracting visual features, sensor features, and operation features based on the standardization processing result; Performing multimodal data fusion processing on the visual features, sensor features, and operation features to obtain multimodal fusion data; Based on multimodal fusion data, big data analysis technology is used to conduct logistics optimization analysis, safety warning analysis and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model.

4. The port operation optimization method based on the integration of big data and sky-rail transportation technology according to claim 3 is characterized in that: Based on multimodal fusion data, big data analysis technology is used to conduct logistics optimization analysis, safety warning analysis, and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis, and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model, including: Construct the logistics optimization engine objective function based on the logistics optimization analysis results: , where W represents the objective function of the logistics optimization engine; The weight coefficient representing the ship waiting time; represents the waiting time of the ship at time tt; represents the utilization rate of port resources at time t; The weight coefficient representing resource utilization; represents the maximum available capacity of port resources; T represents the overall time range of objective function calculation; , where A represents the objective function of the security early warning engine; The weight coefficient representing the accuracy of anomaly detection; Represents the number of anomalies correctly detected by the model; Indicates the number of normal targets correctly detected by the target detection model in the security early warning engine; Indicates the number of anomalies that are incorrectly detected by the target detection model in the security early warning engine; Indicates the number of anomalies that the target detection model in the security early warning engine failed to detect; The weight coefficient representing the false alarm rate.

5. The port operation optimization method based on the integration of big data and sky-rail transportation technology according to claim 4 is characterized in that: Based on multimodal fusion data, big data analysis technology is used to conduct logistics optimization analysis, safety warning analysis, and energy efficiency management analysis. The results of logistics optimization analysis, safety warning analysis, and energy efficiency management analysis are integrated with the four-layer intelligent system to build an intelligent decision-making model, which also includes: , where N represents the energy efficiency management engine objective function; Represents the weight coefficient of energy consumption cost; represents the energy consumption cost of the port at time t; represents the energy efficiency of the port at time t; Represents the weight coefficient of energy utilization efficiency; It represents the theoretical maximum value of port energy utilization; The objective function of the logistics optimization engine, the objective function of the safety warning engine and the objective function of the energy efficiency management engine are integrated with the four-layer intelligent system to build an intelligent decision-making model.

6. The port operation optimization method based on the integration of big data and sky-rail transportation technology according to claim 1 is characterized in that: Based on the output of the intelligent decision-making model, the dynamic scheduling of the air-rail transportation system is guided and executed. According to the real-time basic information obtained by the four-layer intelligent system, the execution results are evaluated, and then the port operation links are optimized according to the evaluation results, including: Determine the corresponding scheduling situation according to the output of each engine in the intelligent decision-making model, and obtain the dynamic scheduling of the sky rail transportation system by combining all the scheduling situations; Execute the dynamic scheduling, collect real-time basic information, evaluate the execution results, identify execution problems in the evaluation results, and optimize the port operation links based on the execution problems.

7. The port operation optimization method based on the integration of big data and sky-rail transportation technology according to claim 6 is characterized in that: The corresponding scheduling situation is determined according to the output of each engine in the intelligent decision-making model, and the dynamic scheduling of the sky rail transportation system is obtained by integrating all the scheduling situations, including: The logistics optimization engine outputs the coordinated scheduling strategy of berths, quay cranes and container trucks; the safety warning engine outputs abnormal behavior detection results and warning information; and the energy efficiency management engine outputs energy consumption hotspot predictions and energy optimization suggestions; The aerial rail transport path and transport strategy are dynamically adjusted according to the collaborative scheduling strategy, the abnormal area is determined and the transport priority is adjusted according to the abnormal behavior detection results and early warning information, and the energy consumption distribution of the transport task is optimized according to the energy consumption hotspot prediction and energy optimization suggestions, thereby obtaining the dynamic scheduling of the aerial rail transport system.

8. The port operation optimization method based on the integration of big data and sky-rail transportation technology according to claim 2 is characterized in that: Based on the optimization results, a continuous learning mechanism is designed to comprehensively optimize the intelligent decision-making system and port transportation links, including: Based on the application layer in the four-layer intelligent system, the operation and maintenance personnel's evaluation of the optimization results is collected, and a continuous learning mechanism is designed based on the evaluation; The intelligent decision-making system is updated according to the continuous learning mechanism, and the port transportation links are comprehensively optimized based on the updated intelligent decision-making model.

Images