A method and device for generating the dynamic carbon footprint of port cargo circulation

Abstract

The present application discloses a method and device for generating a dynamic carbon footprint of port cargo circulation. The method includes: collecting image data, positioning data, and environmental variables generated during the circulation of port cargo, and performing time and space alignment to form a multi-modal fusion dynamic data set; using artificial intelligence to detect the activity level information of logistics equipment from the dynamic data set; calculating the carbon emissions of port cargo in different logistics links through a multi-factor dynamic emission model, in combination with a dynamic adjustment coefficient and the activity level information, where the dynamic adjustment coefficient is used to correct power emissions and fuel emissions; constructing a dynamic association model of time, space, and logistics equipment in combination with the image data and positioning data, and then generating a full-process dynamic carbon footprint of port cargo during the circulation process by using the carbon emissions of port cargo in different logistics links, so as to at least solve the technical problem of relatively low accuracy of carbon emission calculation in the process of generating a carbon footprint in the prior art.

Description

Technical Field

[0001] The present application relates to the technical fields of logistics management and carbon emission calculation. Specifically, it relates to a method and device for generating dynamic carbon footprints of port cargo circulation. Background Art

[0002] This section aims to provide background or context for the content stated in the claims or the specification. The content described here is not admitted to be prior art merely because it is included in this section.

[0003] The concept of "carbon footprint" originates from "ecological footprint", which mainly represents the total emissions of greenhouse gases emitted during the production and consumption activities of human beings in terms of carbon dioxide emission equivalent (CO2equivalent, abbreviated as: CO2eq). Compared with single carbon dioxide emissions, the carbon footprint evaluates the greenhouse gas emissions directly or indirectly generated by the research object during its life cycle by using the life cycle assessment method. For the same object, the calculation difficulty and scope of the carbon footprint are greater than those of carbon emissions, and its calculation result contains information on carbon emissions.

[0004] As an important node in the global logistics chain, the carbon emission management of the cargo circulation process in ports has become the key to achieving the goal of green development. However, the existing technologies have the following prominent problems in the calculation of carbon emissions in separate links and the generation of full-process carbon footprints:

[0005] 1) Insufficient correlation between activity levels and emission factors: The current carbon emission calculation methods usually adopt fixed emission factors and fail to dynamically reflect the impacts of equipment operation status, load changes, and environmental variables, resulting in large calculation errors;

[0006] 2) Lack of dynamics and accuracy in the carbon emission calculation method for separate links: Most of the existing technologies calculate using a single fuel emission factor or power emission factor and fail to comprehensively consider the dynamic impacts of multiple factors such as power output, fuel consumption, load changes, and environmental variables;

[0007] 3) Lack of dynamic association in the generation of full-process carbon footprints: The current methods mostly rely on the simple accumulation of carbon emissions in separate links and fail to generate continuous carbon footprints from the perspectives of time, space, and dynamic association of equipment, making it difficult to accurately describe the complexity of port logistics;

[0008] 4) Low utilization rate of multi-modal data: A large amount of video data, GPS trajectory data, and environmental variable data are generated during port operation. The existing technologies fail to fully utilize the complementarity of these data, resulting in insufficient accuracy in calculating equipment activity levels and carbon emissions.

[0009] In response to the above problems, no effective solutions have been proposed yet. Summary of the Invention

[0010] The embodiments of the present application provide a method and device for generating a dynamic carbon footprint of port cargo circulation, so as to at least solve the technical problem of low accuracy in calculating carbon emissions during the carbon footprint generation process in the prior art. Through the complementary analysis of video images and positioning trajectories (such as GPS trajectories), this solution accurately obtains the activity levels of equipment during the port cargo circulation process. At the same time, combined with an innovative multi-factor fusion dynamic emission calculation method, dynamic adjustment coefficients, and full-process dynamic modeling technology, it generates a dynamic carbon footprint of port cargo circulation, providing scientific decision-making support for the construction of green ports.

[0011] According to one aspect of the embodiments of the present application, a method for generating a dynamic carbon footprint of port cargo circulation is provided, including: collecting image data, positioning data, and environmental variables generated during the circulation of port cargo, and performing time and space alignment on the image data, positioning data, and environmental variables to form a multi-modal fusion dynamic data set; using artificial intelligence to detect the activity level information of logistics equipment from the dynamic data set, where the activity level information includes the equipment type, operation status, operation duration, and load change of the logistics equipment; calculating the carbon emissions of port cargo in different logistics links through a multi-factor dynamic emission model, in combination with dynamic adjustment coefficients and activity level information, where the dynamic adjustment coefficients are used to correct power emissions and fuel emissions; constructing a dynamic association model of time, space, and logistics equipment in combination with the image data and positioning data, and then generating a full-process dynamic carbon footprint of port cargo during the circulation process by using the carbon emissions of port cargo in different logistics links.

[0012] Optionally, collecting image data, positioning data, and environmental variables generated during the circulation of port cargo, and performing time and space alignment on the image data, positioning data, and environmental variables to form a multi-modal fusion dynamic data set includes: collecting image data of logistics equipment through fixed cameras and drones, obtaining the running trajectory, running speed, and position information of the logistics equipment through the positioning module on the logistics equipment, collecting the temperature, humidity, and wind speed of the port operation environment through temperature and humidity sensors and anemometers, where the positioning data includes the running trajectory, running speed, and position information, and the environmental variables include the temperature, humidity, and wind speed; based on timestamp synchronization technology and space alignment technology, performing time and space alignment on the image data, positioning data, and environmental variables to form a multi-modal fusion dynamic data set.

[0013] Optionally, using artificial intelligence to detect the activity level information of logistics equipment from the dynamic data set includes: using a deep learning model to perform object detection on the image data to identify the equipment category, operation status, and load change; using the time series model LSTM to track and predict the dynamic changes of the operation status; calculating the operation duration of the logistics equipment in the operation status through the image data, and estimating the dynamic load of the logistics equipment in combination with the load change. Lcurrent =L base +ΔL , L base is the load before change, ΔL is the load change amount; verify the operation status and operation duration by combining the positioning data.

[0014] Optionally, the operation status includes the loading / unloading status, the idle status, and the transportation status. Verifying the operation status by combining the positioning data includes: when the running speed determined according to the positioning data is greater than 0, it is verified that the logistics equipment is in the transportation status; when the running speed determined according to the positioning data is equal to 0, it is verified that the logistics equipment is in the loading / unloading status or the idle status.

[0015] Optionally, through the multi-factor dynamic emission model, calculate the carbon emissions of port goods in different logistics links by combining the dynamic adjustment coefficient and the activity level information, including: calculating the carbon emissions of the logistics equipment in different operation statuses based on the multi-factor dynamic emission model CF :

[0016]

[0017] CF represents the carbon emissions per link, T statei represents the i operation duration of the logistics equipment in the P i represents the output power of the logistics equipment, EF poweri represents the power emission factor, Fi represents the fuel consumption, EF fueli represents the fuel emission factor, α(Ei,Li) represents the dynamic adjustment coefficient, and the formula for the dynamic adjustment coefficient α(Ei,Li) is:

[0018]

[0019] ΔT = T−T ref , represents the current temperature T and the reference temperature T ref the difference between them, ΔH = H−H ref , represents the current humidity H and the reference humidity H ref the difference between them, W represents the current wind speed, ΔL = L−L ref , represents the current loadL The difference from the reference load k1, k2, k3, k4 represents the weight of the dynamic adjustment coefficient.

[0020] Optionally, a dynamic association model of time, space, and logistics equipment is constructed by combining image data and positioning data, and then the full-process dynamic carbon footprint of port goods during the circulation process is generated by using the carbon emissions of port goods in different logistics links, including: modeling based on time to capture the dynamic conversion time points between the loading and unloading state, idle state, and transportation state of the logistics equipment, and the operation duration of each operation state; modeling based on space, combining positioning data and regional division algorithms to capture the dynamic path of the logistics equipment at different spatial positions; modeling based on the operation state, using a deep learning model and positioning data to analyze the state switching and load changes of the logistics equipment; generating the full-process carbon emission footprint using the following formula:

[0021]

[0022] CF total represents the full-process carbon emissions, CF i (t) represents the i th logistics equipment's instantaneous carbon emissions at time t and T represents the total time of the logistics process.

[0023] Optionally, in the process of constructing a dynamic association model of time, space, and logistics equipment by combining image data and positioning data, and then generating the full-process dynamic carbon footprint of port goods during the circulation process by using the carbon emissions of port goods in different logistics links, the method further includes: verifying the dynamic path, operation state, and load changes of the quasi-logistics equipment by combining image data and positioning data, and outputting a complete carbon footprint report.

[0024] According to another aspect of the embodiments of the present application, there is also provided a device for generating a dynamic carbon footprint of port cargo circulation, including: a collection unit, configured to collect image data, positioning data, and environmental variables generated during the circulation of port cargo, and perform time and space alignment on the image data, positioning data, and environmental variables to form a multi-modal fusion dynamic data set; a detection unit, configured to use artificial intelligence to detect the activity level information of logistics equipment from the dynamic data set, where the activity level information includes the equipment type, operation status, operation duration, and load change of the logistics equipment; a calculation unit, configured to calculate the carbon emissions of port cargo in different logistics links through a multi-factor dynamic emission model, in combination with a dynamic adjustment coefficient and the activity level information, where the dynamic adjustment coefficient is used to correct power emissions and fuel emissions; a generation unit, configured to construct a dynamic association model of time, space, and logistics equipment in combination with the image data and positioning data, and then generate the full-process dynamic carbon footprint of port cargo during the circulation process by using the carbon emissions of port cargo in different logistics links.

[0025] According to another aspect of the embodiments of the present application, there is also provided a computer-readable storage medium, which includes a stored program that, when running, executes the above method.

[0026] According to another aspect of the embodiments of the present application, there is also provided an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the above method through the computer program.

[0027] According to one aspect of the present application, there is provided a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps of any one of the above methods.

[0028] In the embodiments of the present application, image data, positioning data, and environmental variables generated during the circulation of port goods are collected, and the image data, positioning data, and environmental variables are aligned in time and space to form a dynamic dataset with multi-modal fusion. Artificial intelligence is used to detect the activity level information of logistics equipment from the dynamic dataset. The activity level information includes the equipment type, operation status, operation duration, and load change of the logistics equipment. Through a multi-factor dynamic emission model, the carbon emissions of port goods in different logistics links are calculated by combining the dynamic adjustment coefficient and the activity level information. The dynamic adjustment coefficient is used to correct the power emission and fuel emission. A dynamic association model of time, space, and logistics equipment is constructed by combining the image data and the positioning data, and then the full-process dynamic carbon footprint of port goods during the circulation process is generated by using the carbon emissions of port goods in different logistics links. It can solve the technical problem of low accuracy in calculating carbon emissions during the carbon footprint generation process in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The illustrative embodiments and descriptions thereof of the present application are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0030] Figure 1 is a flowchart of an optional method for generating the dynamic carbon footprint of port goods circulation according to an embodiment of the present application;

[0031] Figure 2 is a schematic diagram of an optional solution for generating the dynamic carbon footprint of port goods circulation according to an embodiment of the present application;

[0032] Figure 3 is a schematic diagram of an optional device for generating the dynamic carbon footprint of port goods circulation according to an embodiment of the present application;

[0033] Figure 4 is a structural block diagram of a terminal according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0035] It should be noted that the terms "first", "second", etc. in the description, claims and the above drawings of this application are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or devices.

[0036] According to an aspect of the embodiments of the present application, a method embodiment of a method for generating a dynamic carbon footprint of port cargo circulation is provided. A solution for generating a dynamic carbon footprint of port cargo circulation based on multi-modal data fusion and intelligent optimization algorithms is provided. Through complementary analysis of video images and GPS trajectories, combined with a multi-factor fusion dynamic emission model and dynamic adjustment coefficients, and supported by full-process dynamic modeling technology, the emission factors are optimized in real time to generate the dynamic carbon footprint of port cargo circulation.

[0037] Based on multi-modal data fusion and intelligent optimization algorithms, the present invention constructs a system for generating a dynamic carbon footprint of port cargo circulation, which mainly includes the following core modules: 1) Multi-modal data acquisition and fusion module: Collect video image data and GPS data, and combine environmental variable data to obtain the equipment activity level through complementary analysis; 2) Activity level calculation module: Dynamically calculate the activity time and load changes of the equipment in different states based on multi-modal data; 3) Innovative carbon emission calculation module for each link: Accurately calculate the carbon emissions of different logistics links through a multi-factor dynamic emission model combined with dynamic adjustment coefficients; 4) Full-process carbon footprint dynamic modeling module: Based on the combination of video and GPS data, construct a dynamic association model of time, space and equipment to generate the full-process dynamic carbon footprint of the port cargo circulation process.

[0038] Figure 1 is a flowchart of an optional method for generating a dynamic carbon footprint of port cargo circulation according to an embodiment of the present application, as Figure 1 shown, the method may include the following steps:

[0039] Step S102, collect image data, positioning data and environmental variables generated during the circulation of port cargo, and perform time and space alignment on the image data, positioning data and environmental variables to form a dynamic data set of multi-modal fusion.

[0040] Step S104: Use artificial intelligence to detect the activity level information of logistics equipment from the dynamic dataset. The activity level information includes the equipment type of the logistics equipment, the operation status (or equipment status), the operation duration (or running duration, running time, etc.), and the load change.

[0041] Step S106: Through a multi-factor dynamic emission model, combine the dynamic adjustment coefficient and the activity level information to calculate the carbon emissions of port goods in different logistics links. The dynamic adjustment coefficient is used to correct the power emission and fuel emission.

[0042] Step S108: Combine the image data and the positioning data to construct a dynamic association model of time, space, and logistics equipment, and then use the carbon emissions of port goods in different logistics links to generate the full-process dynamic carbon footprint of port goods during the circulation process.

[0043] In the above solution, it has the following advantages and can solve the technical problem of low accuracy in calculating carbon emissions during the carbon footprint generation process mentioned:

[0044] Accurate calculation of activity level: Use the multi-modal fusion method of video images and GPS data to dynamically identify the equipment status and obtain the activity time and load change in different states.

[0045] Innovative carbon emission calculation method for each link: Propose a dynamic emission model based on multi-factor fusion, combine the dynamic adjustment coefficient, and comprehensively consider power, fuel emission factors, load, environmental variables, and equipment dynamic characteristics to accurately calculate the carbon emissions in different logistics links.

[0046] Full-process dynamic modeling: Based on the combination of video and GPS data, establish a dynamic association model among time, space, and equipment, and generate a complete dynamic carbon footprint of port goods during the circulation process.

[0047] Multi-modal data fusion technology: Capture the equipment operation status and load change through video images, and calibrate the activity

[0048] As an alternative embodiment, the technical solution of this application will be described in detail below in combination with specific implementation manners. Figure 2 Describe the technical solution of this application in detail:

[0049] Step 1: Multi-modal data collection and fusion (collect the image data, positioning data, and environmental variables generated during the circulation process of port goods, and perform time and space alignment on the image data, the positioning data, and the environmental variables to form a multi-modal fusion dynamic dataset)

[0050] 1) Data collection sources:

[0051] Video image data, which runs videos through fixed cameras and drone acquisition devices to identify device categories, operating status, and load changes; GPS data, which collects the operating trajectory and location information of the device through the GPS module on the device for calibrating video analysis results; environmental variable data: which collects real-time environmental conditions, including variables such as temperature, humidity, and wind speed, through temperature and humidity sensors and anemometers.

[0052] 2) Data fusion method:

[0053] Complementary analysis of video and GPS, which uses video image recognition to identify device status (such as loading, idling, transporting), and calibrates the device's path and activity area through GPS data to achieve complementary analysis of status and trajectory; time and space alignment, which synchronizes timestamps and aligns spatial coordinates for video, GPS, and environmental variable data to form a dynamic dataset for multi-modal fusion.

[0054] Step 2: Activity level calculation (using artificial intelligence to detect the activity level information of the logistics device from the dynamic dataset, where the activity level information can be abbreviated as the activity level, including the device type, operating status, operating duration, and load change of the logistics device)

[0055] 1) Activity level definition

[0056] Running time: The activity time of the device in different states (such as loading, idling, transporting).

[0057] Load change: Capturing the change in the device's load status (such as loading weight) through video image data, and analyzing the transportation distance of the device in combination with the GPS trajectory.

[0058] 2) Activity level calculation method

[0059] 2.1) Multi-modal data input for status recognition

[0060] Status recognition depends on video data and the device's auxiliary sensor data (such as GPS, IMU, etc.), and high-precision recognition of the device status is achieved through multi-modal input.

[0061] The input data includes: video data: which runs the device operation video through a fixed camera or drone for detecting the device's operating status (such as transporting, loading, idling, etc.); GPS data: which provides the device's location information and speed for assisting in verifying the status recognition result; load change data: which captures the change in the device's loading volume through video or sensors, reflecting the device's operation intensity.

[0062] 2.2) Design and optimization of the deep learning model

[0063] In the present invention, deep learning models (YOLOv5 and LSTM) are used to process video data to complete state recognition and time calculation. The following are the specific design and optimization details of the models:

[0064] (1) YOLOv5: Real-time object detection

[0065] Detection of equipment and loads in videos of complex port scenarios:

[0066] Detection targets: Equipment types (such as cranes, transport vehicles), operating states of equipment (such as loading / unloading, transporting, idling), and cargo loads (such as container volume, bulk cargo volume);

[0067] Output results: Through the bounding boxes output by YOLOv5, the spatial positions of equipment or goods are located in real time, and their categories and states are labeled.

[0068] Model optimization:

[0069] Data augmentation: Data augmentation (such as rotation, scaling, noise addition, etc.) is performed on the video data of port scenarios to improve the model's adaptability to complex scenarios;

[0070] Custom dataset: Based on specific port tasks (such as detection of loading / unloading and transporting states), the model is trained to improve detection accuracy;

[0071] Multi-scale detection: Detect targets of different sizes (such as large equipment and small goods) to ensure the robustness of the model in complex scenarios.

[0072] (2) LSTM: Time series analysis and state tracking

[0073] Applicable to the dynamic change analysis of port equipment states:

[0074] Input sequence: The equipment state sequence (such as loading / unloading, idling, transporting) output by YOLOv5 is input into the LSTM model to analyze the change trend of equipment states in the time dimension;

[0075] Output state: Predict the current state of the equipment and its duration, for example:

[0076] State current : Current state (such as transporting, loading / unloading), T state =t stop −t start : State duration.

[0077] Model optimization:

[0078] State smoothing processing: Filter short-term fluctuations through LSTM to avoid misjudgment of states (such as misidentifying a short-term pause as idling).

[0079] State transition analysis: Based on the historical state sequence, predict the next possible state (such as transitioning to transportation after loading and unloading are completed).

[0080] (3)Multi-modal fusion: Combining video and sensor data

[0081] To improve the accuracy of state recognition, the deep learning model fuses video data with GPS or other sensor data:

[0082] Time synchronization: Align the timestamps of video frames with GPS data to ensure the consistency of data in the time dimension;

[0083] State verification: Verify the device state recognized by the video through the speed and trajectory features of GPS. For example, when the video detects a "transportation state", verify it by combining with GPS speed Vgps>0; when the video detects an "idling state", verify it by combining with the zero speed of GPS Vgps = 0.

[0084] 2.3)State recognition and time calculation process

[0085] The entire process of state recognition and time calculation is as follows:

[0086] (1)Object detection and state classification

[0087] Video input: Preprocess the port operation video (such as denoising, resolution adjustment);

[0088] Object detection: Use YOLOv5 to detect the equipment, goods, and their operation states;

[0089] State classification: Conduct a preliminary classification of the equipment state and output the following results: State i : The current state of the equipment (such as transportation, loading and unloading, idling); Box i : The detection location and bounding box of the equipment.

[0090] (2)Time series analysis and state tracking

[0091] Sequence input: Use the state sequence output by YOLOv5 as the input of LSTM.

[0092] State tracking: Analyze the state changes of the equipment through LSTM and predict its current state and duration.

[0093] Time calculation: Combine the video frame timestamps to calculate the duration of each state:

[0094] T state =t stop −t start

[0095] Among them, t start and t stop are the start and end times of the state respectively.

[0096] (3)Dynamic calculation of load changes

[0097] Load detection: Detect the presence and quantity changes of goods through YOLOv5. For example, detect the change in the number of containers (such as the number of containers decreases from 1 to 0 during loading and unloading); detect the change in the volume of bulk goods (such as the volume increases during goods loading);

[0098] Dynamic load adjustment: Dynamically update the load L of the device according to the detection results:

[0099] L current =L base +ΔL,

[0100] Among them, L base is the initial load, ΔL is the load change amount.

[0101] (4)State and time verification

[0102] Verify the state recognition and time calculation through GPS data:

[0103] State verification: Combine the GPS speed V gps Verify the state detected by the video. For example: If V gps >0 , verify that the device is in the "transportation" state. If V gps =0 , verify that the device is in the "loading and unloading" or "idle" state;

[0104] Path verification: Combine the GPS trajectory to determine whether the device is in the correct operation area.

[0105] 2.4) Output

[0106] Through the above calculation process, generate the activity level data of port equipment, including:

[0107] Device status sequence: {State1, State2, …, Staten};

[0108] Time corresponding to each status T state and load L;

[0109] Status duration: Time interval of each status T state ;

[0110] Load change sequence: Dynamic change of device load L current .

[0111] Step 3: Calculate carbon emissions in innovative sub - links (Calculate the carbon emissions of port goods in different logistics links through a multi - factor dynamic emission model, combined with a dynamic adjustment coefficient and the activity level information. The dynamic adjustment coefficient is used to correct power emissions and fuel emissions)

[0112] 1) Dynamic adjustment coefficient α(Ei,Li)

[0113] Dynamic adjustment coefficient α(Ei,Li) is a dynamic parameter used to correct the power emission factor and the fuel emission factor, considering the following two types of influencing factors:

[0114] Environmental variable Ei: including temperature (T), humidity (H) and wind speed (W), which affect the combustion efficiency and emission characteristics of the device.

[0115] Load status Li: Dynamically capture the load and operation status (such as no - load, full - load, idle, etc.) of the device through video images and GPS data, reflecting the impact of the actual operation load of the device on emissions.

[0116] The calculation formula of the dynamic adjustment coefficient is:

[0117]

[0118] ΔT = T−T ref : The difference between the current temperature and the reference temperature;

[0119] ΔH = H−H ref : The difference between the current humidity and the reference humidity;

[0120] W: The current wind speed;

[0121] ΔL = L−L ref : The difference between the current load and the reference load;

[0122] k1, k2, k3, k4: Weights of environmental and load variables, determined by historical data and model training.

[0123] The correction formula for the emission factor by dynamically adjusting the coefficient is as follows:

[0124] Correction of power emission factor:

[0125]

[0126] Correction of fuel emission factor:

[0127]

[0128] 2) Dynamic adjustment method combining video and GPS

[0129] To ensure the calculation accuracy of the dynamic adjustment coefficient α(Ei,Li) of the present invention, through the multi-modal analysis technology combining video and GPS data, the operation state and load changes of the device are dynamically captured, and fused with the real-time data of environmental variables to form a complete dynamic adjustment framework.

[0130] 2.1) Role of video data

[0131] Video data realizes the following functions through deep learning models (such as YOLOv5, LSTM):

[0132] Recognition of device operation state: Using video images to recognize the current operation state of the device (such as loading and unloading, idling, transportation), providing state input for activity level calculation, and judging whether it is in the loading and unloading state by monitoring the boom movement frequency of the crane through video;

[0133] Estimation of load quantity: By analyzing information such as the volume of goods and the loading actions of the device through video, estimate the current load quantity L. Dynamically update the load value by detecting the quantity or volume of goods on the transport vehicle through video;

[0134] Capture of operation time: According to the start and stop times of the device in the video, accurately calculate the activity time in different states.

[0135] 2.2) Role of GPS data

[0136] GPS data realizes the following functions through trajectory analysis and state calibration:

[0137] Path calibration and activity area analysis: Through the displacement and speed changes between GPS trajectory points, calibrate the device state and activity area recognized by video. When the device moves at a low speed and is located in the yard area, it can assist in verifying that it is in the loading and unloading state;

[0138] Capture of dynamic state changes: Combining speed and acceleration (calculated from GPS trajectories), assist in judging the dynamic state of the device (such as transportation, idling). Judge whether the device is in the transportation or empty driving state by the speed change of the trajectory points;

[0139] Time and space alignment: Align the timestamps of GPS data with the video data to ensure the synchronization of multimodal data in time and space.

[0140] 2.3) Complementary analysis of video and GPS

[0141] The combination of video and GPS data enables high-precision dynamic capture of the device operating state and load changes:

[0142] Status verification: The device status identified by the video (such as loading / unloading, idling) is calibrated through GPS trajectory data to eliminate misjudgments that may be caused by a single data source. When the video identifies the device as in the loading / unloading state, combining with GPS to confirm that the device is indeed located in the yard area can avoid misjudgments caused by video perspective occlusion;

[0143] Dynamic capture of load changes: The load estimated by the video is further verified through the distance and speed characteristics of the GPS trajectory to form a multi-dimensional dynamic input of the load state;

[0144] Improve the calculation accuracy of activity level: Through the complementary analysis of video and GPS data, accurately calculate the activity time and load changes of the device in different states, providing high-quality input data for dynamically adjusting coefficients.

[0145] Multi-factor fusion dynamic emission model

[0146] The dynamic emission model proposed by the present invention comprehensively considers the following factors: device operating state, output power, fuel consumption, load, and environmental variables, and accurately calculates carbon emissions in different links.

[0147] Emission calculation formula:

[0148]

[0149] CF: Carbon emissions in different links;

[0150] T statei ： The i Activity time of the

[0151] Pi: Device output power;

[0152] EF poweri ： Power emission factor;

[0153] Fi: Fuel consumption;

[0154] EFfueli ： Fuel emission factor;

[0155] α(Ei,Li): Dynamic adjustment coefficient.

[0156] Step 4: Dynamic modeling of the full-process carbon footprint (construct a dynamic association model of time, space, and logistics equipment by combining image data and positioning data, and then generate the full-process dynamic carbon footprint of port goods during the circulation process using the carbon emissions of port goods in different logistics links)

[0157] 1) Dynamic association model of time, space, and equipment

[0158] The dynamic association model of time, space, and equipment designed by the present invention takes multi-modal data as the core input, and through the dynamic analysis of the combination of video and GPS, establishes the correlation of the equipment operation state in time, space, and between equipment, and constructs a dynamic carbon emission path.

[0159] 1.1) Model input

[0160] Video data: Real-time capture of the operation state and load changes of the equipment, used for state recognition and activity level calculation;

[0161] GPS trajectory data: Provide the operation path, location information, and speed changes of the equipment, used for calibration of the state, path analysis, and activity area division;

[0162] Environmental variable data: Include parameters such as temperature, humidity, and wind speed, used to correct the emission factor and dynamically adjust the carbon emission calculation result.

[0163] 1.2) Model structure

[0164] The model is divided into three core parts: time modeling, space modeling, and equipment state modeling.

[0165] (1) Time modeling

[0166] Time modeling captures the operation rules and carbon emission changes of the equipment at different times through the time series analysis of the equipment state changes;

[0167] Time series analysis method: Use a recurrent neural network (RNN) or a long short-term memory network (LSTM) to model the time series data of the equipment, and identify the dynamic conversion between the states of the equipment such as running, idling, loading and unloading, and transportation;

[0168] Association of equipment state with time: Extract the key state change points Tstart and Tstop of the equipment through the time stamps of the video data, and combine with the GPS speed change points to establish the activity time of each state:

[0169] Tstate =T stop −T start

[0170] (2)Spatial Modeling

[0171] Spatial modeling captures the dynamic path of the equipment at different spatial positions by combining the GPS trajectory information of the equipment with the job area division.

[0172] Trajectory analysis and activity area division: Based on GPS data, key job areas of the equipment (such as yards, loading and unloading areas, transportation channels) are identified through trajectory clustering algorithms (such as DBSCAN), and the trajectories are divided into different job scenarios;

[0173] Path optimization and calibration: The movement path of the equipment is calibrated through GPS trajectories to ensure the continuity and accuracy of the path, and video data is combined to verify whether the equipment is truly in a certain job area;

[0174] Spatial association modeling formula: The spatial path modeling formula of the equipment is:

[0175]

[0176] where S path is the total path length, xi, yi are the coordinates of the trajectory points.

[0177] (3)Equipment Status Modeling

[0178] Equipment status modeling uses video data combined with deep learning algorithms (such as YOLOv5) to detect the running status (loading and unloading, idling, transportation) and load changes of the equipment in real time, and combines GPS data to calibrate the status switch.

[0179] Multi - equipment interaction modeling: A graph network (Graph Neural Network, GNN) is used to construct the dynamic interaction relationship between the states of equipment, such as the impact of the collaborative operation between loading and unloading equipment and transportation vehicles on carbon emissions;

[0180] Status recognition and update formula: Status recognition dynamically updates the equipment status by analyzing the action features Fvideo of the equipment through video and combining the GPS speed change Vgps:

[0181]

[0182] 1.3) Generation of Carbon Emission Path

[0183] Combining the dynamic associations of time, space, and equipment status, the full - process carbon emission path of the equipment is generated, and the formula is:

[0184]

[0185] CFtotal: Full-process carbon footprint;

[0186] CFi(t): The i instantaneous carbon emissions of the device at time t;

[0187] T： Total time of the logistics process.

[0188] 2) Generation of dynamic footprint by combining video and GPS

[0189] 2.1) Role of video data

[0190] Device status recognition: Use a deep learning model (such as YOLOv5) to analyze video images in real time to identify the operating status (loading / unloading, idle, transportation) of the device and load changes;

[0191] Activity time capture: Through the recognition of device actions in the video, capture the key activity time points of the device, and generate time series data in combination with timestamps;

[0192] Load change estimation: Analyze the operating actions of the device (such as loading and unloading goods) through video to dynamically estimate the load of the device.

[0193] 2.2) Role of GPS data

[0194] Path calibration and activity area division: Calibrate the path of the device through GPS trajectory data to identify key operating areas (such as yards, loading and unloading areas);

[0195] Speed and status verification: Combine the speed and acceleration changes of GPS data to verify the device status identified by the video (such as transportation or idle);

[0196] Time and space alignment: Align the timestamps of GPS data with video data to ensure the synchronization of multimodal data.

[0197] 2.3) Complementary analysis of combining video and GPS

[0198] Status verification: The device status identified by the video is calibrated through the GPS trajectory to ensure the accuracy of status recognition. For example, when the video identifies that the device is in the loading / unloading state, combine the GPS trajectory to confirm whether the device is located within the loading / unloading area;

[0199] Load change capture: The load estimated by the video (such as the weight of goods) is further verified through the transportation path of the GPS trajectory, and the load status is dynamically adjusted;

[0200] Optimization of Activity Level Calculation: By combining video and GPS data, accurately calculate the activity time and path of the device in different states, providing high-quality input for carbon emission calculation.

[0201] This application also provides a preferred embodiment, which further details the technical solution of this application in combination with specific scenarios:

[0202] Background: A certain port processes 1 million twenty-foot equivalent units (TEU) of container cargo annually. The port management needs to evaluate the dynamic carbon emissions of a single batch of goods (1 TEU, 20-foot container) during loading, unloading, transportation, and yard operations. Since port operations involve multiple devices and complex scenarios, traditional static carbon emission calculation methods cannot capture dynamic changes (such as equipment state switching, load changes, environmental impacts, etc.). Therefore, the solution of this invention is adopted for full-process dynamic carbon footprint calculation.

[0203] Case data, the logistics process of handling a single batch of goods involves the links and data shown in Table 1 below:

[0204] Table 1

[0205]

[0206] Environmental variables:

[0207] Temperature T = 35°C, reference temperature T ref = 25°C;

[0208] Humidity H = 80%, reference humidity H ref = 50%;

[0209] Wind speed W = 5 m / s;

[0210] Carbon emission factors:

[0211] Fuel emission factor EF fuel = 2.68 kgCO2;

[0212] Power emission factor EF power = 0.85 kgCO2 / kWh.

[0213] Implementation steps

[0214] Step 1: Multi-modal data collection and fusion

[0215] Data collection:

[0216] Video data: Collect video of equipment operation through fixed cameras and drones to identify equipment states (loading, unloading, idling).

[0217] GPS data: Record the running paths, speeds, and positions of transport trucks and stackers.

[0218] Environmental variable data: Collect the real-time temperature, humidity, and wind speed of the port.

[0219] Data fusion:

[0220] Synchronize multi-modal data using the time and space alignment algorithm of the present invention:

[0221] Align the video frame timestamps with the GPS trajectory data;

[0222] Map the device status to specific scenarios (such as yards, transport channels, docks, etc.) according to the device spatial coordinates.

[0223] Multi-modal data fusion results:

[0224] Crane status: Loading and unloading goods, with a status switching time of 0.5 seconds;

[0225] Transport truck status: The transport path is 5 kilometers, and the speed varies in the range of 0 - 30 km / h;

[0226] Stacker status: Handling goods in the yard, with a load change of +2 m³.

[0227] Step two: Activity level calculation

[0228] Dynamically calculate the device activity level using the YOLOv5+LSTM model of the present invention.

[0229] Device status recognition:

[0230] The YOLOv5 model recognizes the device status through video images:

[0231] Crane status: Loading and unloading;

[0232] Transport truck status: Transporting, idling;

[0233] Stacker status: Handling goods.

[0234] Model accuracy: The status recognition accuracy rate reaches 96%.

[0235] Time series analysis:

[0236] Use the LSTM model to perform time series prediction on the device status changes and calculate the activity time of the device:

[0237] Crane loading and unloading time: 15 minutes;

[0238] Transport truck transportation time: 30 minutes;

[0239] Operating time of the reach stacker: 20 minutes;

[0240] Idle time of the crane and the truck: 25 minutes.

[0241] Load change calculation:

[0242] During the loading and unloading process of the crane, the load change is ΔL = +1 (loading 1 container).

[0243] During the handling process of the reach stacker, the load change is ΔL = +2 m³ (handling bulk cargo).

[0244] Verify that the load of the transport truck changes from full load to empty load (ΔL = −1) according to the GPS data.

[0245] Step 3: Multi-factor dynamic carbon emission calculation

[0246] Dynamic adjustment coefficient calculation:

[0247] According to the dynamic adjustment coefficient formula:

[0248]

[0249] Weight setting: k1 = 0.01, k2 = 0.005, k3 = 0.02, k4 = 0.05.

[0250] Crane (loading and unloading):

[0251] α = 1 + (0.01×10) + (0.005×30) + (0.02×5) + (0.05×1) = 1.45.

[0252] Transport truck (transportation):

[0253] α = 1 + (0.01×10) + (0.005×30) + (0.02×5) + (0.05×(−1)) = 1.35.

[0254] Reach stacker (handling): α = 1 + (0.01×10) + (0.005×30) + (0.02⋅5) + (0.05×2) = 1.55.

[0255] Carbon emission calculation:

[0256] According to the formula:

[0257]

[0258] Crane (loading and unloading):

[0259] CF = 15 / 60×(0 + 5×2.68)×1.45 = 2.91 kgCO 2

[0260] Transport truck (transportation):

[0261] CF = 30 / 60×(150×0.85 + 0)×1.35 = 51.53 kgCO 2

[0262] Stacker (for handling):

[0263] CF = 20 / 60 × (0 + 3 × 2.68) × 1.55 = 4.16 kgCO 2

[0264] Idle speed (crane / truck):

[0265] CF = 25 / 60 × (0 + 2 × 2.68) × 1.35 = 3.02 kgCO 2

[0266] Step 4: Full-process dynamic carbon footprint modeling

[0267] Full-process carbon emission path generation:

[0268] Summarize carbon emissions of each link:

[0269] CF total = 2.91 + 51.53 + 4.16 + 3.02 = 61.62 kgCO 2

[0270] Spatial distribution:

[0271] Yard area (loading / unloading + yard operation): Total 7.07 kgCO 2;

[0272] Transportation path (truck transportation): Total 51.53 kgCO 2.

[0273] Analysis of high-emission links:

[0274] The carbon emissions in the transportation link of transport trucks account for the highest proportion: 51.53÷61.62≈83.6%.

[0275] Results and optimization suggestions

[0276] Total carbon emissions: The dynamic carbon footprint of a single batch of goods is 61.62 kgCO2.

[0277] Optimization suggestions:

[0278] Path optimization: Adjust the driving route of transport trucks to reduce the path length.

[0279] Equipment replacement: Gradually introduce electric or hybrid transport trucks.

[0280] Improve operation efficiency: Reduce the idle time of equipment and optimize the operation process through intelligent scheduling.

[0281] The multi-modal fusion algorithm, dynamic carbon emission model, and full-process modeling method adopted in this solution not only quantify the carbon emissions of the entire process, but also identify high-emission links and propose optimization directions, providing a scientific basis for the construction of green ports.

[0282] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0283] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of this application.

[0284] According to another aspect of the embodiments of this application, there is also provided a port cargo circulation dynamic carbon footprint generation device for implementing the above port cargo circulation dynamic carbon footprint generation method. Figure 3 is a schematic diagram of an optional port cargo circulation dynamic carbon footprint generation device according to the embodiments of this application, as Figure 3 shown, the device may include:

[0285] An acquisition unit 31, configured to acquire image data, positioning data, and environmental variables generated during the circulation of port cargo, and perform time and space alignment on the image data, the positioning data, and the environmental variables to form a multi-modal fusion dynamic data set.

[0286] A detection unit 33, configured to use artificial intelligence to detect the activity level information of logistics equipment from the dynamic data set, where the activity level information includes the equipment type, operation status, operation duration, and load change of the logistics equipment.

[0287] A calculation unit 35 for calculating the carbon emissions of port goods in different logistics links through a multi-factor dynamic emission model in combination with a dynamic adjustment coefficient and the activity level information, wherein the dynamic adjustment coefficient is used to correct power emissions and fuel emissions.

[0288] A generation unit 37 for constructing a dynamic association model of time, space, and logistics equipment by combining image data and positioning data, and then generating a full-process dynamic carbon footprint of port goods during the circulation process by using the carbon emissions of port goods in different logistics links.

[0289] It should be noted here that the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run in the hardware environment as shown in Figure 1 and can be implemented by software or by hardware.

[0290] Optionally, the acquisition unit can also be used to: collect image data of logistics equipment through fixed cameras and drones, obtain the running trajectory, running speed, and position information of the logistics equipment through the positioning module on the logistics equipment, collect the temperature, humidity, and wind speed of the port operation environment through temperature and humidity sensors and anemometers, the positioning data includes the running trajectory, the running speed, and the position information, and the environmental variables include temperature, humidity, and wind speed; based on timestamp synchronization technology and spatial alignment technology, perform time and space alignment on the image data, the positioning data, and the environmental variables to form a multi-modal fusion dynamic data set.

[0291] Optionally, the detection unit can also be used to: perform object detection on the image data by using a deep learning model to identify the equipment category, operation status, and load change; use the time series model LSTM to track and predict the dynamic changes of the operation status; calculate the operation duration of the logistics equipment in the operation state through the image data, and estimate the dynamic load of the logistics equipment in combination with the load change L current =L base +ΔL , where L base is the load before change, ΔL is the load change amount; verify the operation status and the operation duration in combination with the positioning data.

[0292] Optionally, the detection unit can also be used to: when the running speed determined according to the positioning data is greater than 0, verify that the logistics equipment is in the transportation state; when the running speed determined according to the positioning data is equal to 0, verify that the logistics equipment is in the loading and unloading state or the idle state.

[0293] Optionally, the calculation unit can also be used to calculate the carbon emissions of the logistics equipment in different operating states based on the multi-factor dynamic emission model. CF :

[0294]

[0295] Wherein, CF represents the carbon emissions per link, T statei represents the operating duration of the logistics equipment in the i th operating state, P i represents the output power of the logistics equipment, EF poweri represents the power emission factor, Fi represents the fuel consumption, EF fueli represents the fuel emission factor, α(Ei,Li) represents the dynamic adjustment coefficient, and the calculation formula of the dynamic adjustment coefficient α(Ei,Li) is:

[0296]

[0297] ΔT = T−T ref , represents the current temperature T and the reference temperature T ref The difference between them, ΔH = H−H ref , represents the current humidity H and the reference humidity H ref The difference between them, W represents the current wind speed, ΔL = L−L ref , represents the current load L The difference between and the reference load, k1, k2, k3, k4 represents the weight of the dynamic adjustment coefficient.

[0298] Optionally, the generation unit can also be used to capture the dynamic transition time points between the loading and unloading state, idle state, and transportation state of the logistics equipment, and the operating duration of each operating state based on time modeling; capture the dynamic path of the logistics equipment at different spatial positions based on spatial modeling, combined with positioning data and regional division algorithms; analyze the state switching and load changes of the logistics equipment based on operating state modeling by combining a deep learning model and positioning data; generate the full-process carbon emission footprint using the following formula:

[0299]

[0300] Among them, CF total represents the total carbon emissions of the whole process, CF i (t) represents the i th logistics device's instantaneous carbon emissions at time t and T represents the total time of the logistics process.

[0301] Optionally, the generating unit can also be used to: combine the image data and the positioning data to verify the dynamic path, operation status, and load changes of the quasi-logistics device, and output a complete carbon footprint report.

[0302] According to another aspect of the embodiments of the present application, a server or a terminal for implementing the above-mentioned method for generating the dynamic carbon footprint of port cargo circulation is also provided.

[0303] Figure 4 is a structural block diagram of a terminal according to an embodiment of the present application. As Figure 4 shown, the terminal may include: one or more (only one is shown in the figure) processors 401, a memory 403, and a transmission device 405. As Figure 4 shown, the terminal may further include an input / output device 407.

[0304] Among them, the memory 403 can be used to store software programs and modules, such as program instructions / modules corresponding to the method and device for generating the dynamic carbon footprint of port cargo circulation in the embodiments of the present application. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 403, that is, implements the above-mentioned method for generating the dynamic carbon footprint of port cargo circulation. The memory 403 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 403 may further include a memory remotely disposed relative to the processor 401, and these remote memories can be connected to the terminal through a network. Examples of the above network include but are not limited to the Internet, enterprise intranets, local area networks, mobile communication networks, and their combinations.

[0305] ​The above-mentioned transmission device 405 is used to receive or send data via a network, and can also be used for data transmission between the processor and the memory. Specific examples of the above-mentioned network can include wired networks and wireless networks. In one example, the transmission device 405 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers through network cables, so as to communicate with the Internet or local area network. In one example, the transmission device 405 is a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0306] Specifically, the memory 403 is used to store application programs.

[0307] The processor 401 can call the application programs stored in the memory 403 through the transmission device 405 to execute the following steps:

[0308] Collect image data, positioning data, and environmental variables generated during the circulation of port goods, and perform time and space alignment on the image data, the positioning data, and the environmental variables to form a multi-modal fusion dynamic dataset;

[0309] Use artificial intelligence to detect the activity level information of logistics equipment from the dynamic dataset, where the activity level information includes the equipment type, operation status, operation duration, and load change of the logistics equipment;

[0310] Through a multi-factor dynamic emission model, combine the dynamic adjustment coefficient and the activity level information to calculate the carbon emissions of port goods in different logistics links, where the dynamic adjustment coefficient is used to correct power emissions and fuel emissions;

[0311] Combine the image data and the positioning data to construct a dynamic association model of time, space, and logistics equipment, and then use the carbon emissions of port goods in different logistics links to generate the full-process dynamic carbon footprint of port goods during the circulation process.

[0312] Optionally, the specific examples in this embodiment can refer to the examples described in the above embodiments, and this embodiment will not be elaborated here.

[0313] Those of ordinary skill in the art can understand that Figure 4 The structure shown is only schematic. The terminal can be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, and a Mobile Internet Devices (MID), a PAD and other terminal devices. Figure 4 It does not limit the structure of the above-mentioned electronic device. For example, the terminal may further include moreFigure 4 more or fewer components shown therein (such as network interfaces, display devices, etc.), or having a configuration different from that Figure 4 shown.

[0314] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium. The storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, an optical disc, etc.

[0315] The embodiments of the present application also provide a storage medium. Optionally, in this embodiment, the above storage medium can be used to execute the program code of the method for generating the dynamic carbon footprint of port cargo circulation.

[0316] Optionally, in this embodiment, the above storage medium may be located on at least one of the multiple network devices in the network shown in the above embodiments.

[0317] Optionally, in this embodiment, the storage medium is set to store the program code for performing the following steps:

[0318] Collect image data, positioning data, and environmental variables generated during the circulation of port cargo, and perform time and space alignment on the image data, the positioning data, and the environmental variables to form a multi-modal fusion dynamic data set;

[0319] Use artificial intelligence to detect the activity level information of logistics equipment from the dynamic data set, where the activity level information includes the equipment type, operation status, operation duration, and load change of the logistics equipment;

[0320] Through a multi-factor dynamic emission model, combine the dynamic adjustment coefficient and the activity level information to calculate the carbon emissions of port cargo in different logistics links, where the dynamic adjustment coefficient is used to correct power emissions and fuel emissions;

[0321] Combine the image data and the positioning data to construct a dynamic association model of time, space, and logistics equipment, and then use the carbon emissions of port cargo in different logistics links to generate the full-process dynamic carbon footprint of port cargo during the circulation process.

[0322] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and this embodiment will not be elaborated here.

[0323] Optionally, in this embodiment, the above storage medium may include, but is not limited to: various media that can store program codes, such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs.

[0324] The serial numbers of the embodiments of the present application are only for description and do not represent the superiority or inferiority of the embodiments.

[0325] If the integrated unit in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in the above computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in the storage medium and includes several instructions for causing one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.

[0326] In the above embodiments of the present application, the descriptions of the various embodiments have their own focuses. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0327] In the several embodiments provided by the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. 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 displayed or discussed mutual coupling or direct coupling or communication connection can be through some interfaces. The indirect coupling or communication connection of the units or modules can be in an electrical or other form.

[0328] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0329] In addition, the functional units in the various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0330] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A method for generating dynamic carbon footprint of port cargo circulation, characterized in that: include: Collect image data, positioning data and environmental variables generated by port cargo during circulation, and perform time and space alignment on the image data, positioning data and environmental variables to form a multi-modal fused dynamic data set; Using artificial intelligence to detect activity level information of logistics equipment from the dynamic data set, wherein the activity level information includes equipment type, operation status, operation duration, and load change of the logistics equipment; The carbon emissions of port cargo in different logistics links are calculated by combining the dynamic adjustment coefficient and the activity level information through a multi-factor dynamic emission model, including: based on the multi-factor dynamic emission model, the carbon emissions of logistics equipment in different operating states are calculated CF : ,in, CF Indicates the carbon emissions at each stage, T statei Indicates that logistics equipment is i The duration of each job in each job state, P i Represents the output power of logistics equipment, EF poweri represents the power emission factor, F i Indicates fuel consumption, EF fueli represents the fuel emission factor, α(E i , L i ) Indicates the dynamic adjustment coefficient. α(E i ,L i ) The calculation formula is: , ΔT=T current −T ref , indicating the current temperature T current With reference temperature T ref The difference between ΔH=H−H ref , indicating the current humidity H With reference humidity H ref The difference between W Indicates the current wind speed. ΔL2=L current −L ref , indicating the current load L current The difference between the reference load and k1, k2, k3, k4 represents a weight of a dynamic adjustment coefficient, wherein the dynamic adjustment coefficient is used to correct power emission and fuel emission; Combine image data and positioning data to build a dynamic correlation model of time, space and logistics equipment, including: based on time modeling, capture the dynamic conversion time points of logistics equipment between loading and unloading state, idling state and transportation state, and the operation duration of each operation state; based on space modeling, combine positioning data and area division algorithm to capture the dynamic path of logistics equipment in different spatial locations; based on operation state modeling, use deep learning model and positioning data to analyze the state switching and load changes of logistics equipment; The carbon emissions of port cargo in different logistics links are used to generate a dynamic carbon footprint of the entire process of port cargo circulation.

2. The method according to claim 1, characterized in that The image data, positioning data and environmental variables generated by the port cargo during the circulation process are collected, and the image data, positioning data and environmental variables are aligned in time and space to form a multi-modal fused dynamic data set, including: The image data of logistics equipment is collected by fixed cameras and drones, the running track, running speed and location information of the logistics equipment are obtained by the positioning module on the logistics equipment, and the temperature, humidity and wind speed of the port operating environment are collected by temperature and humidity sensors and anemometers. The positioning data includes the running track, the running speed and the location information, and the environmental variables include temperature, humidity and wind speed; Based on the timestamp synchronization technology and the space alignment technology, the image data, the positioning data and the environmental variables are aligned in time and space to form a multi-modal fused dynamic data set.

3. The method according to claim 1, characterized in that Artificial intelligence is used to detect the activity level information of logistics equipment from the dynamic data set, including: Using a deep learning model to perform target detection on the image data to identify equipment categories, operating states, and load changes; Use the time series model LSTM to track and predict the dynamic changes of job status; The operating time of the logistics equipment in the operating state is calculated through the image data, and the dynamic load of the logistics equipment is estimated in combination with the load change. L current =L base +ΔL1 ,in, L base is the load before the change, ΔL1 is the load variation; The operation status and the operation duration are verified in combination with the positioning data.

4. The method according to claim 3, characterized in that: The operation status includes loading and unloading status, idling status and transportation status. The operation status is verified in combination with the positioning data, including: When the running speed determined according to the positioning data is greater than 0, it is verified that the logistics equipment is in a transport state; When the running speed determined according to the positioning data is equal to 0, it is verified that the logistics equipment is in a loading and unloading state or an idling state.

5. The method according to claim 1, characterized in that The carbon emissions of port cargo in different logistics links are used to generate the dynamic carbon footprint of the entire process of port cargo circulation, including: The full process carbon footprint is generated using the following formula: , in, CF total represents the carbon emissions of the entire process, Indicates j Logistics equipment at a time t The instantaneous carbon emissions, Represents the total time of the logistics process.

6. The method according to claim 5, characterized in that In the process of combining the image data and the positioning data to construct a dynamic correlation model of time, space and logistics equipment, and then using the carbon emissions of port cargo in different logistics links to generate a full-process dynamic carbon footprint of the port cargo in the circulation process, the method also includes: The image data and the positioning data are combined to verify the dynamic path, operation status and load change of the logistics equipment, and a complete carbon footprint report is output.

7. A device for generating dynamic carbon footprint of port cargo circulation, characterized in that: include: A collection unit, used to collect image data, positioning data and environmental variables generated by port cargo during circulation, and to perform time and space alignment on the image data, the positioning data and the environmental variables to form a multi-modal fused dynamic data set; a detection unit, configured to detect activity level information of logistics equipment from the dynamic data set using artificial intelligence, wherein the activity level information includes equipment type, operation status, operation duration, and load change of the logistics equipment; A calculation unit is used to calculate the carbon emissions of port cargo in different logistics links through a multi-factor dynamic emission model, combined with a dynamic adjustment coefficient and the activity level information: Based on the multi-factor dynamic emission model, the carbon emissions of logistics equipment in different operating states are calculated CF : ,in, CF Indicates the carbon emissions at each stage, T statei Indicates that logistics equipment is i The duration of each job in each job state, P i Represents the output power of logistics equipment, EF poweri represents the power emission factor, F i Indicates fuel consumption, EF fueli represents the fuel emission factor, α(E i ,L i ) Indicates the dynamic adjustment coefficient. α(E i ,L i ) The calculation formula is: , ΔT=T current −T ref , indicating the current temperature T current With reference temperature T ref The difference between ΔH=H−H ref , indicating the current humidity H With reference humidity H ref The difference between W Indicates the current wind speed. ΔL2=L current −L ref , indicating the current load L current The difference between the reference load and k1, k2, k3, k4 represents a weight of a dynamic adjustment coefficient, wherein the dynamic adjustment coefficient is used to correct power emission and fuel emission; The generation unit is used to combine image data and positioning data to build a dynamic correlation model of time, space and logistics equipment: based on time modeling, it captures the dynamic conversion time points of logistics equipment between loading and unloading state, idling state and transportation state, and the operation duration of each operation state; based on space modeling, it combines positioning data and area division algorithm to capture the dynamic path of logistics equipment in different spatial locations; based on operation state modeling, it uses deep learning model and positioning data to analyze the state switching and load changes of logistics equipment; The generation unit is also used to generate a full-process dynamic carbon footprint of the port cargo in the circulation process by using the carbon emissions of the port cargo in different logistics links.

8. A computer-readable storage medium, characterized in that: The storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 6 when executed.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor executes the method described in any one of claims 1 to 6 through the computer program.

Images