Container ship operation carbon intensity estimation method and system based on historical data

Abstract

This invention belongs to the field of carbon intensity prediction technology, specifically relating to a method and system for predicting the carbon intensity of container ship operations based on historical data. The method includes: collecting and parsing fixed-format messages sent from the ship to accumulate historical data; filtering the collected messages; decomposing a complete voyage into multiple different states, and establishing fuel consumption analysis models for each state and equipment; calculating the total fuel consumption of the complete voyage, and predicting the CII (Carbon Intensity Index) of the complete voyage based on the total fuel consumption. The system includes: a data accumulation module, a data filtering module, a data modeling module, and a CII prediction model. This invention solves the problems of existing methods for predicting ship CII requiring a large amount of data and experimental data, being complex and costly, and the inability to achieve CII prediction due to the large amount of input data required for these models, much of which is unavailable.

Description

Technical Field

[0001] This invention belongs to the field of carbon intensity prediction technology, specifically relating to a method and system for predicting the carbon intensity of container ship operations based on historical data. Background Technology

[0002] The Carbon Intensity Indicator (CII) is a mandatory rule introduced by the International Maritime Organization (IMO) to reduce greenhouse gas emissions from the shipping industry. It aims to promote decarbonization in the industry by quantifying ship operating efficiency. According to the rules, a ship's annual actual CII is categorized into five levels, from A (best) to E (worst). If a ship is rated D for three consecutive years or E in a single year, it must submit a corrective action plan and may be subject to close scrutiny during Port State Control (PSC) inspections, potentially even being detained. This places higher demands on shipping companies' ship operation management. Shipping companies need to predict their ships' CII in advance and take timely measures (such as optimizing routes and reducing speed) if performance is poor to avoid regulatory penalties and operational restrictions.

[0003] Current methods primarily utilize theoretical formulas or machine learning models, leveraging ship data, model test data, and data collected by shipborne sensors to predict ship fuel consumption, thereby forecasting ship CII (Common Intake). However, this approach requires substantial data and experimental data, or necessitates the installation of navigation data monitoring and acquisition systems on the ship (such as energy efficiency management systems), making it complex and costly. Furthermore, these methods require significant input data for predicting ship CII, such as ship cargo load (or draft) and navigation weather conditions, often failing to provide specific numerical values ​​and thus hindering CII prediction. Summary of the Invention

[0004] This invention addresses the problems of existing methods for predicting the carbon intensity (CI) of ships, which require a large amount of data and experimental data, are complex and costly, and whose models require a large amount of input data but many specific data are unavailable, thus making CII prediction impossible. The invention provides a method and system for predicting the carbon intensity of container ship operations based on historical data.

[0005] The technical solution claimed by this invention is as follows:

[0006] A method for predicting the carbon intensity of container ship operations based on historical data includes the following steps:

[0007] S1: Data accumulation: Collect and parse the fixed-format messages sent back from the ship to accumulate historical data; the messages include: navigation afternoon reports, ship departure reports and ship berthing reports;

[0008] S2: Data Filtering: Filter the message emails in the historical data obtained in S1. Specifically, the navigation midday report filters the data of navigation throughout the day and the berthing data, and removes data with severe winds and waves and abnormal data. Abnormal data is removed from the ship departure report and the ship berthing report. The data of navigation throughout the day is called the navigation midday report, and the berthing data is called the berthing midday report.

[0009] S3: Data Modeling: Decompose a complete voyage of a ship into multiple different states. Based on the data obtained from S2, establish fuel consumption analysis models for each state and each piece of equipment. Wherein: the complete voyage refers to the voyage from the departure of a ship from one port to the departure of a ship from another port.

[0010] S4: CII Prediction: Based on the fuel consumption predicted by each fuel consumption analysis model in S3, the total fuel consumption of the complete flight segment is calculated, and the CII of the complete flight segment is predicted based on the total fuel consumption.

[0011] Preferably, the voyage report includes the ship's voyage time, mileage, equipment fuel consumption, and wind and waves encountered that day; the ship departure report includes information on the cargo on board; the ship berthing report includes information on the fuel consumption of the ship during the berthing process; and the equipment fuel consumption includes: main engine fuel consumption, auxiliary engine fuel consumption, and boiler fuel consumption.

[0012] Preferably, the selection criteria for the navigation afternoon report, berthing afternoon report, ship departure report, and ship berthing report in S2 are as follows:

[0013] Midday Navigation Report: Navigation duration greater than or equal to 23 hours, navigation distance greater than or equal to 200 nautical miles, wave force less than or equal to 5, wind force less than or equal to 6, and fuel consumption of both main engine and auxiliary engine greater than zero;

[0014] Midday report from anchorage: sailing time is zero, main engine fuel consumption is zero, auxiliary engine fuel consumption is greater than zero, and boiler fuel consumption is greater than or equal to zero;

[0015] Ship departure report: Refrigerant container capacity information is not empty;

[0016] Ship berthing report: The time of entry and exit from the port and the fuel consumption are not empty.

[0017] Preferably, the various states in S3 include: normal navigation, port entry / exit navigation, and berthing / anchoring; for normal navigation, two fuel consumption analysis models are established: daily fuel consumption of the main engine and daily fuel consumption of the auxiliary engine; for port entry / exit navigation, one fuel consumption analysis model is established, which is the total fuel consumption for port entry / exit; for berthing / anchoring, two fuel consumption analysis models are established: daily fuel consumption of the berthed auxiliary engine and daily fuel consumption of the berthed boiler.

[0018] Preferably, under normal navigation conditions, the main engine fuel consumption is modeled using the midday navigation report, while the auxiliary engine fuel consumption is modeled using a combination of the midday navigation report and the ship's departure report. Since the main engine fuel consumption is primarily related to the ship's speed, the daily main engine fuel consumption under normal navigation conditions is expressed as a function of the ship's speed Vg, i.e.:

[0019] Normal flight engine daily fuel consumption = fme(Vg)

[0020] Among them: the function fme adopts the form of a power function or a cubic function; the undetermined coefficient method is used to fit the air speed and main engine fuel consumption data from the midday report.

[0021] Under normal navigation conditions, auxiliary engine fuel consumption is mainly related to the number of cold storage containers on board. The daily fuel consumption of auxiliary engines under normal navigation conditions can be expressed as a function of the number of cold storage containers on board, nref, i.e.:

[0022] Daily fuel consumption of auxiliary aircraft during normal navigation = fae1(nref)

[0023] Among them, the function fae1 adopts the form of a linear function, and uses the auxiliary engine fuel consumption data of the midday navigation report and the cold box data of the departure report to fit the data using the method of undetermined coefficients.

[0024] Preferably, when a ship is entering or leaving port, the fuel consumption of the main engine and auxiliary engines is modeled as a whole, and the total fuel consumption for entering and leaving port is expressed as a function of the time t between entering and leaving port, that is:

[0025] Total fuel consumption for inbound and outbound traffic = fall(t)

[0026] Among them, fall adopts the form of an exponential function, and uses the arrival and departure times and fuel consumption data of ships berthing reports to fit the data using the method of undetermined coefficients.

[0027] Preferably, when the ship is berthed / anchored, fuel consumption includes auxiliary engine fuel consumption and boiler fuel consumption. Auxiliary engine fuel consumption is related to the number of cold containers. The auxiliary engine fuel consumption during anchoring uses the auxiliary engine fuel consumption function established during normal navigation. The auxiliary engine fuel consumption during berthing uses the same function form, but since the number of cold containers changes during loading and unloading while berthing, the input becomes the average number of cold containers between two departures, i.e.:

[0028] Daily fuel consumption of auxiliary machinery at rest = fae2[(nref1 + nref2) / 2]

[0029] Among them: the function fae2 adopts the form of a linear function, using auxiliary engine fuel consumption data from the midday berthing report, the cold container data from the previous ship departure report and the ship departure report after this berthing, and uses the method of undetermined coefficients for fitting; the daily boiler fuel consumption during berthing or anchoring can be represented by a constant value, namely:

[0030] Daily fuel consumption of the boiler during shutdown = Cboiler

[0031] Cboiler used the average boiler oil consumption from the midday berthing report to obtain the data.

[0032] Preferably, the specific process of S4 is as follows: refer to the common route schedules of container shipping companies to obtain the ship route schedule information, and based on the ship route schedule information and the fuel consumption predicted by various fuel consumption analysis models, estimate the total fuel consumption of a ship on the route for one voyage; the ship route schedule information includes: arrival and departure times of each port, port entry and exit time, mileage between each port, and the number of departing refrigerated containers at each port, wherein if the number of departing refrigerated containers cannot be provided, historical average data can be used.

[0033] Preferably, the formula for calculating the total fuel consumption is:

[0034]

[0035] Where: d i Indicates the distance from port i to Shanghai port; d i+1 This indicates the distance from Hong Kong i+1 to Shanghai Port; Ta i Indicates the arrival time at port i; Ta i+1 Indicates the arrival time at port i+1; Td i Indicates the departure time from port i; Td i+1 Indicates the departure time from Hong Kong (i+1 time); to i Indicates the departure time (i); to i+1 Indicates the departure time i+1; ti i Indicates the duration of arrival at port; ti i+1 Indicates the inbound i+1 duration; nref i Indicates the departing refrigerated container volume; nref i+1 Indicates the volume of the refrigerated container departing from port i+1;

[0036] The operational carbon intensity (CII) of a voyage is calculated using the following formula:

[0037]

[0038] Where: C F The carbon emission factor represents the type of fuel consumed, with the default coefficient of 3.114 for heavy fuel oil; DWT represents the deadweight tonnage of the vessel.

[0039] The present invention also provides a container ship operation carbon intensity prediction system based on historical data, comprising: a data accumulation module, a data filtering module, a data modeling module and a CII prediction model connected in sequence;

[0040] The data accumulation module is used to collect and parse fixed-format messages sent back from the ship, and accumulate them to form historical data; the messages include: navigation afternoon reports, ship departure reports and ship berthing reports;

[0041] The data filtering module is used to filter the message emails collected by the data accumulation module;

[0042] The data modeling module decomposes a complete voyage of a ship into multiple different states. Based on the data filtered by the data filtering module, fuel consumption analysis models are established for each state and each piece of equipment. The complete voyage refers to the voyage from when the ship departs from one port to when it departs from another port.

[0043] The CII prediction module calculates the total fuel consumption for the entire flight segment based on the fuel consumption predicted by each fuel consumption analysis model in the data modeling module, and predicts the CII for the entire flight segment based on the total fuel consumption.

[0044] The beneficial effects of this invention are as follows:

[0045] This invention provides a method and system for predicting the carbon intensity of container ship operations based on historical data. It collects and parses fixed-format messages sent back from the ship to accumulate historical data. It can predict CII without the need for ship data or the installation of a data monitoring and acquisition system. The message data is readily available to shipping companies. This invention solves the problem that existing methods for predicting ship CII require a lot of data and experimental data, and are complex and costly. This method decomposes a complete ship voyage into multiple different states, and establishes fuel consumption analysis models for each state and equipment. Targeted function forms are adopted for different stages and equipment, ultimately enabling the prediction of ship fuel consumption for a given route, thus achieving CII (Carbon Intensity Index). This method simplifies the input of CII prediction, taking into account the relatively fixed routes of container ships. It does not require specific cargo load (or draft) and route weather conditions data. It uses historical message data to model ship fuel consumption, incorporating the influence of ship cargo load and weather conditions into the model. This ensures good accuracy in predicting operational carbon intensity throughout the year, solving the problem that existing methods for predicting ship CII require a large amount of input data, but many specific data are unavailable, leading to the inability to achieve CII prediction. Attached Figure Description

[0046] Figure 1 This is a flowchart of the container ship operation carbon intensity prediction method based on historical data in an embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram of a complete flight segment in an embodiment of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions will be further described clearly and completely below with reference to the accompanying drawings.

[0049] This invention provides a method for predicting the carbon intensity of container ship operations based on historical data, such as... Figure 1 As shown, it includes the following steps:

[0050] S1: Data Accumulation: Collect and parse the fixed-format messages sent back from the ship to accumulate historical data. The messages are divided into three types: noon navigation report, departure report, and berthing report. The noon navigation report should include information such as the ship's sailing time, mileage, equipment fuel consumption (main engine / auxiliary engine / boiler fuel consumption), and wind and waves encountered that day. The departure report will include information on the cargo on board, such as cargo volume, container volume, and refrigerated container volume. The berthing report will include information on the fuel consumption of the ship during the berthing process, such as the time taken to enter the port and the total fuel consumption.

[0051] S2: Data Filtering: Filter the message emails in the historical data obtained in S1. Specifically, for the navigation midday report, filter the data for the entire day's navigation and berthing, removing data related to severe winds and waves or abnormal data. For the ship departure and berthing reports, remove abnormal data. The data for the entire day's navigation is referred to as the navigation midday report, and the data for berthing is referred to as the berthing midday report. The filtering criteria for the navigation midday report, berthing midday report, ship departure report, and ship berthing report are as follows:

[0052] Midday Navigation Report: Navigation duration greater than or equal to 23 hours, navigation distance greater than or equal to 200 nautical miles, wave force less than or equal to 5, wind force less than or equal to 6, and fuel consumption of both main engine and auxiliary engine greater than zero (not empty);

[0053] Midday report on berthing: sailing time is zero, main engine fuel consumption is zero, auxiliary engine fuel consumption is greater than zero (not empty), boiler fuel consumption is greater than or equal to zero (not empty);

[0054] Ship departure report: Refrigerant container capacity information is not empty;

[0055] Ship berthing report: The time of entry and exit from the port and the fuel consumption are not empty.

[0056] S3: Data Modeling: Decompose a complete voyage of a ship into multiple different states. Based on the data obtained from S2, establish fuel consumption analysis models for each state and each piece of equipment. Wherein: the complete voyage refers to the voyage from the departure of a ship from one port to the departure of a ship from another port.

[0057] In specific embodiments of the present invention, such as Figure 2As shown, a complete voyage (departure from port A to port B) can be broken down as follows: After departing from port A and leaving the port boundary, the ship begins normal navigation. Upon arrival at port B, if waiting for berthing is required, the ship will anchor for a period of time before weighing anchor and entering the port to berth. Alternatively, if waiting for berthing is not required, the ship will directly enter the port to berth after completing normal navigation and remain berthed at port B until departure. This complete voyage can be divided into three states: normal navigation, port entry / exit navigation, and berthing / anchoring. Fuel consumption analysis and modeling are performed for each of these three states.

[0058] Normal navigation: Fuel consumption during normal navigation is divided into main engine and auxiliary engine fuel consumption (during normal navigation, boiler operation mainly utilizes main engine exhaust gas, and boiler fuel consumption is very small and can be ignored). Main engine fuel consumption is modeled using the midday navigation report, while auxiliary engine fuel consumption is modeled using a combination of the midday navigation report and the ship's departure report. Main engine fuel consumption is mainly related to ship speed; therefore, the daily main engine fuel consumption is expressed as a function of ship speed Vg, i.e.:

[0059] Normal flight engine daily fuel consumption = fme(Vg)

[0060] The function fme is implemented using a power function or a cubic function, and then fitted using the midday report's speed and main engine fuel consumption data using the method of undetermined coefficients. Auxiliary engine fuel consumption is mainly related to the number of refrigerated containers on board the ship; therefore, the daily auxiliary engine fuel consumption is expressed as a function of the number of refrigerated containers on board, nref, i.e.:

[0061] Daily fuel consumption of auxiliary aircraft during normal navigation = fae1(nref)

[0062] Among them, the function fae1 adopts the form of a linear function, and uses the auxiliary engine fuel consumption data of the noon navigation report and the cold container data of the departure report (the cold container data remains unchanged from departure to arrival), and uses the undetermined coefficient method for fitting.

[0063] Arrival and Departure: Since ships are typically accelerating and decelerating during arrival and departure, fuel consumption is complex. Therefore, the fuel consumption of the main engine and auxiliary engines is modeled as a whole, and the total fuel consumption during arrival and departure is expressed as a function of the arrival / departure time t, i.e.:

[0064] Total fuel consumption for inbound and outbound traffic = fall(t)

[0065] Among them, fall adopts the form of an exponential function, and uses the arrival and departure times and fuel consumption data of ships berthing reports to fit the data using the method of undetermined coefficients.

[0066] c) Berthing / Anchoring: Fuel consumption during berthing / anchoring includes auxiliary engine fuel consumption and boiler fuel consumption. Auxiliary engine fuel consumption is also related to the number of cold containers. Auxiliary engine fuel consumption during anchoring uses the auxiliary engine fuel consumption function established during normal navigation. Auxiliary engine fuel consumption during berthing uses the same function form, but since the number of cold containers on the vessel changes during berthing, the input becomes the average number of cold containers between two departures, i.e.:

[0067] Daily fuel consumption of auxiliary machinery at rest = fae2[(nref1 + nref2) / 2]

[0068] The function fae2 also adopts a linear function form, using auxiliary engine fuel consumption data from the berthing midday report, the cold container data from the previous departure report, and the departure report after this berthing, and employs the method of undetermined coefficients for fitting. The daily boiler fuel consumption during berthing or anchoring can be represented by a constant value, namely...

[0069] Daily fuel consumption of the boiler during shutdown = Cboiler

[0070] Cboiler used the average boiler oil consumption from the midday berthing report to obtain the data.

[0071] S4: CII Prediction: Based on the fuel consumption predicted by each fuel consumption analysis model in S3, the total fuel consumption of the complete flight segment is calculated, and the CII of the complete flight segment is predicted based on the total fuel consumption.

[0072] In a specific embodiment of the present invention, referring to the common route schedules of container shipping companies, the route schedule information required for ship CII prediction is designed as shown in Table 1, which includes the arrival and departure times of each port, the duration of entry and exit from the port, the mileage between each port, and the departing refrigerated container volume of each port. If the departing refrigerated container volume cannot be provided, historical average data can be used.

[0073] Table 1. Route and schedule information required for ship CII prediction

[0074]

[0075] Based on fuel consumption data modeling and schedule information, the total fuel consumption of a vessel on this route for one voyage can be estimated using the following formula:

[0076]

[0077] Then, the operational carbon intensity (CII) of a voyage can be calculated using the following formula:

[0078]

[0079] Where: C FThe carbon emission factor is the type of fuel consumed, with the default coefficient of 3.114 for heavy oil (in the modeling process, all fuels used are converted to heavy oil based on their calorific value), and DWT is the deadweight tonnage of the ship.

[0080] The present invention also provides a container ship operation carbon intensity prediction system based on historical data, comprising: a data accumulation module, a data filtering module, a data modeling module and a CII prediction model connected in sequence;

[0081] The data accumulation module is used to collect and parse fixed-format messages sent back from the ship, and accumulate them to form historical data; the messages include: navigation afternoon reports, ship departure reports and ship berthing reports;

[0082] The data filtering module is used to filter the message emails collected by the data accumulation module;

[0083] The data modeling module decomposes a complete voyage of a ship into multiple different states. Based on the data filtered by the data filtering module, fuel consumption analysis models are established separately for each state and each piece of equipment. Here, a complete voyage refers to the segment from when the ship departs from one port to when it departs from another. In a specific embodiment of the invention, such as... Figure 2 As shown, a complete voyage (departure from port A to port B) can be broken down as follows: After departing from port A and leaving the port boundary, the ship begins normal navigation. Upon arrival at port B, if waiting for berthing is required, the ship will anchor for a period of time before weighing anchor and entering the port to berth. Alternatively, if waiting for berthing is not required, the ship will directly enter the port to berth after completing normal navigation and remain berthed at port B until departure. This complete voyage can be divided into three states: normal navigation, port entry / exit navigation, and berthing / anchoring. Fuel consumption analysis and modeling are performed for each of these three states.

[0084] The CII prediction module calculates the total fuel consumption for the entire flight segment based on the fuel consumption predicted by each fuel consumption analysis model in the data modeling module, and predicts the CII for the entire flight segment based on the total fuel consumption.

[0085] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail with reference to the accompanying drawings and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention. In short, all technical solutions and improvements that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention patent.

Claims

1. A method for predicting the carbon intensity of container ship operations based on historical data, characterized in that, The steps include the following: S1: Data accumulation: Collect and parse the fixed-format messages sent back from the ship to accumulate historical data; the messages include: navigation afternoon reports, ship departure reports and ship berthing reports; S2: Data Filtering: Filter the message emails in the historical data obtained in S1. Specifically, the navigation midday report filters the data of navigation throughout the day and the berthing data, and removes data with severe winds and waves and abnormal data. Abnormal data is removed from the ship departure report and the ship berthing report. The data of navigation throughout the day is called the navigation midday report, and the berthing data is called the berthing midday report. S3: Data Modeling: Decompose a complete voyage of a ship into multiple different states. Based on the data obtained from S2, establish fuel consumption analysis models for each state and each piece of equipment. Wherein: the complete voyage refers to the voyage from the departure of a ship from one port to the departure of a ship from another port. S4: CII Prediction: Based on the fuel consumption predicted by each fuel consumption analysis model in S3, the total fuel consumption of the complete flight segment is calculated, and the CII of the complete flight segment is predicted based on the total fuel consumption.

2. The method for predicting the carbon intensity of container ship operations based on historical data as described in claim 1, characterized in that, The voyage report includes the ship's sailing time, mileage, equipment fuel consumption, and wind and waves encountered that day; the departure report includes information on the cargo on board; the berthing report includes information on the fuel consumption of the ship during the berthing process; and the equipment fuel consumption includes: main engine fuel consumption, auxiliary engine fuel consumption, and boiler fuel consumption.

3. The method for predicting the carbon intensity of container ship operations based on historical data as described in claim 1 or 2, characterized in that, The selection criteria for the S2 navigation afternoon report, berthing afternoon report, vessel departure report, and vessel berthing report are as follows: Midday Navigation Report: Navigation duration greater than or equal to 23 hours, navigation distance greater than or equal to 200 nautical miles, wave force less than or equal to 5, wind force less than or equal to 6, and fuel consumption of both main engine and auxiliary engine greater than zero; Midday report from anchorage: sailing time is zero, main engine fuel consumption is zero, auxiliary engine fuel consumption is greater than zero, and boiler fuel consumption is greater than or equal to zero; Ship departure report: Refrigerant container capacity information is not empty; Ship berthing report: The time of entry and exit from the port and the fuel consumption are not empty.

4. The method for predicting the carbon intensity of container ship operations based on historical data as described in claim 3, characterized in that, S3 includes several different states: normal navigation, port entry / exit navigation, and berthing / anchorage. Two fuel consumption analysis models are established for the normal navigation state: daily fuel consumption of the main engine and daily fuel consumption of the auxiliary machinery. One fuel consumption analysis model is established for the port entry / exit navigation state, which is the total fuel consumption for port entry / exit. Two fuel consumption analysis models are established for the berthing / anchorage state: daily fuel consumption of the auxiliary machinery and daily fuel consumption of the boiler.

5. The method for predicting the carbon intensity of container ship operations based on historical data as described in claim 4, characterized in that, Under normal navigation conditions, main engine fuel consumption is modeled using the midday navigation report, while auxiliary engine fuel consumption is modeled using a combination of the midday navigation report and the ship's departure report. Main engine fuel consumption is primarily related to ship speed; therefore, the daily main engine fuel consumption under normal navigation conditions is expressed as a function of the ship's speed Vg, i.e.: Normal flight engine daily fuel consumption = fme(Vg) Among them: the function fme adopts the form of a power function or a cubic function; the undetermined coefficient method is used to fit the air speed and main engine fuel consumption data from the midday report. Under normal navigation conditions, auxiliary engine fuel consumption is mainly related to the number of cold storage containers on board. The daily fuel consumption of auxiliary engines under normal navigation conditions can be expressed as a function of the number of cold storage containers on board, nref, i.e.: Daily fuel consumption of auxiliary aircraft during normal navigation = fae1(nref) Among them, the function fae1 adopts the form of a linear function, and uses the auxiliary engine fuel consumption data of the midday navigation report and the cold box data of the departure report to fit the data using the method of undetermined coefficients.

6. The method for predicting the carbon intensity of container ship operations based on historical data as described in claim 5, characterized in that, When a ship is entering or leaving port, the fuel consumption of the main engine and auxiliary engines is modeled as a whole, and the total fuel consumption for entering and leaving port is expressed as a function of the time t between entering and leaving port, i.e.: Total fuel consumption for inbound and outbound traffic = fall(t) Among them, fall adopts the form of an exponential function, and uses the arrival and departure times and fuel consumption data of ships berthing reports to fit the data using the method of undetermined coefficients.

7. The method for predicting the carbon intensity of container ship operations based on historical data as described in claim 6, characterized in that, When a ship is moored / anchored, fuel consumption includes auxiliary engine fuel consumption and boiler fuel consumption. Auxiliary engine fuel consumption is related to the number of cold containers. The auxiliary engine fuel consumption during anchoring uses the auxiliary engine fuel consumption function established during normal navigation. The auxiliary engine fuel consumption during mooring uses the same function form, but since the number of cold containers on a ship loading and unloading cargo changes during mooring, the input becomes the average number of cold containers between two departures, i.e.: Daily fuel consumption of auxiliary machinery at rest = fae2[(nref1 + nref2) / 2] Among them: the function fae2 adopts the form of a linear function, using auxiliary engine fuel consumption data from the midday berthing report, the cold container data from the previous ship departure report and the ship departure report after this berthing, and uses the method of undetermined coefficients for fitting; the daily boiler fuel consumption during berthing or anchoring can be represented by a constant value, namely: Daily fuel consumption of the boiler during shutdown = Cboiler Cboiler used the average boiler oil consumption from the midday berthing report to obtain the data.

8. The method for predicting the carbon intensity of container ship operations based on historical data as described in claim 7, characterized in that, The specific process of S4 is as follows: refer to the common route schedules of container shipping companies to obtain the ship route schedule information, and based on the ship route schedule information and the fuel consumption predicted by various fuel consumption analysis models, estimate the total fuel consumption of a ship on the route for one voyage; the ship route schedule information includes: arrival and departure times of each port, port entry and exit time, mileage between each port, and the number of departing refrigerated containers at each port. If the number of departing refrigerated containers cannot be provided, historical average data can be used.

9. The method for predicting the carbon intensity of container ship operations based on historical data as described in claim 7, characterized in that, The formula for calculating the total fuel consumption is: Where: d i Indicates the distance from port i to Shanghai port; d i+1 This indicates the distance from Hong Kong i+1 to Shanghai Port; Ta i Indicates the arrival time at port i; Ta i+1 Indicates the arrival time at port i+1; Td i Indicates the departure time from port i; Td i+1 Indicates the departure time from Hong Kong (i+1 time); to i Indicates the departure time (i); to i+1 Indicates the departure time i+1; ti i Indicates the duration of arrival at port; ti i+1 Indicates the inbound i+1 duration; nref i Indicates the departing refrigerated container volume; nref i+1 Indicates the volume of the refrigerated container departing from port i+1; The operational carbon intensity (CII) of a voyage is calculated using the following formula: Where: C F The carbon emission factor represents the type of fuel consumed, with the default coefficient of 3.114 for heavy fuel oil; DWT represents the deadweight tonnage of the vessel.

10. A container ship operation carbon intensity prediction system based on historical data, characterized in that, It includes, in sequence: a data accumulation module, a data filtering module, a data modeling module, and a CII prediction model; The data accumulation module is used to collect and parse fixed-format messages sent back from the ship, and accumulate them to form historical data; the messages include: navigation afternoon reports, ship departure reports and ship berthing reports; The data filtering module is used to filter the message emails collected by the data accumulation module; The data modeling module decomposes a complete voyage of a ship into multiple different states. Based on the data filtered by the data filtering module, fuel consumption analysis models are established for each state and each piece of equipment. The complete voyage refers to the voyage from when the ship departs from one port to when it departs from another port. The CII prediction module calculates the total fuel consumption for the entire flight segment based on the fuel consumption predicted by each fuel consumption analysis model in the data modeling module, and predicts the CII for the entire flight segment based on the total fuel consumption.

Images