A Gantt chart display method for business jet flights
By optimizing flight takeoff and landing time period data through the LSTM model and combining HTML5 and Canvas technologies, a Gantt chart framework for business jet flights was constructed. This solved the problem that existing technology could not meet the complex needs of modern aviation management, and achieved real-time display and efficient management of flight progress.
Patent Information
- Application Number
- CN202410407622.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-07
- Publication Date
- 2025-10-03
- Estimated Expiration
- 2044-04-07
AI Technical Summary
The existing flight Gantt chart is unable to display detailed flight information and conduct detailed management, and cannot meet the complex needs of modern aviation management.
The LSTM model is used to optimize flight takeoff and landing time period data. Combined with HTML5 and Canvas technologies, a Gantt chart framework for business jet flights is constructed to achieve real-time display and dynamic updates of flight progress.
It achieves a comprehensive, intuitive and real-time display of flight progress, improves the efficiency and experience of users in obtaining information, increases the operational efficiency of business jets by 20%-50%, and reduces the loss costs caused by mission delays by 10%-15%.
Smart Images

Figure CN118245153B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of computer graphics, and in particular to a method for displaying a Gantt chart of a business jet flight. Background Art
[0002] In the aviation industry, precise control over flight scheduling and progress is at the heart of management. Gantt charts have long been the primary visual tool for effectively monitoring flight progress. However, while Gantt charts provide a high-level overview of progress, they fall short in presenting detailed flight information and enabling detailed management. Furthermore, technological advancements are driving increasing demands for data presentation and user interfaces, making the traditional Gantt chart approach inadequate for the complex demands of modern aviation management. Summary of the Invention
[0003] In order to overcome the deficiencies of the above technologies, the present invention provides a Gantt chart display method for business jet flights, which provides a comprehensive, intuitive and real-time flight progress display platform for flight management personnel.
[0004] The technical solution adopted by the present invention to overcome the technical problems is:
[0005] A method for displaying a Gantt chart of a business jet flight comprises the following steps:
[0006] a) Establish a Gantt chart framework for business jet flights to display flight progress information;
[0007] b) Obtain the take-off and landing time period data set T2 based on the take-off and landing time data of the business jet, where T2 = {H1, H2, ..., H k ,...,H K}, H k is the kth takeoff and landing time period array, k∈{1,2,...,K}, K is the total number of takeoff and landing time period arrays;
[0008] c) Through the kth takeoff and landing time period array H k , the k+1th takeoff and landing time period array H k+1 , the k+2th takeoff and landing time period array H k+2 Train the LSTM model to obtain the optimized LSTM model;
[0009] d) Array H of missing time periods for business jet takeoff and landing q The first three takeoff and landing time period array H q-1 、H q-2 、H q-3 Input them into the optimized LSTM model respectively to obtain the missing time period array H of business jet takeoff and landing q Take-off time tup and the missing time period array H for business jet takeoff and landing q Landing time t down , H q =[t up ,t down ], q-1∈{1,2,...,K}, q-2∈{1,2,...,K}, q-3∈{1,2,...,K};
[0010] e) Based on the missing time period array H of business jet takeoff and landing q Draw a rectangle for the flight Gantt chart in the business jet flight Gantt chart frame;
[0011] f) Dynamically display the rectangles of the drawn flight Gantt chart.
[0012] Furthermore, in step a), a Gantt chart framework of business jet flights is constructed using HTML5.
[0013] Furthermore, step b) comprises the following steps:
[0014] b-1) Obtain the take-off and landing time data of M business jets, and obtain the data set T1, T1 = {t1, t2, ..., t m ,...,t M}, where t m is the take-off and landing time data of the mth business jet, m∈{1,2,…,M}, m is an odd number, and t m is the take-off time of the business jet, and when m is an even number, t m The landing time of the business jet;
[0015] b-2) The take-off and landing time data of the mth business jet t m Convert to ISO8601 format;
[0016] b-3) Through formula c m =t m -s Calculate and convert the mth business jet takeoff and landing time data t into ISO8601 format m The time difference c from the epoch time s m ;
[0017] b-4) Through the formula Calculate the number of minutes h m , where y is the time difference c m The number of years, m1 is the time difference c m The number of months, d is the time difference c m The number of days, h is the time difference c m hours, m2 is the time difference c mThe number of minutes is obtained, and the transformed data set T1′ is obtained, T1′={h1,h2,...,h m ,...,h M};
[0018] b-5) By formula H k =[h m-1 ,h m ] k Create the kth takeoff and landing time period array H k ,
[0019] Furthermore, step c) includes the following steps:
[0020] c-1) Set the kth takeoff and landing time period array H k Input into the LSTM model and output the hidden state vector i k ;
[0021] c-2) Use the torch.cat function in Pytorch to convert the hidden state vector i k and the k+1th takeoff and landing time period array H k+1 Perform splicing operation to obtain vector i k ′, use the Xavier function in Pytorch to initialize a matrix w, and use the mul function in Pytorch to convert the vector i k ′ is multiplied by the initialized matrix w to obtain the vector i k ″, vector i k ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i k+1 ;
[0022] c-3) Use the torch.cat function in Pytorch to convert the hidden state vector i k+1 and the k+2th takeoff and landing time period array H k+2 Perform splicing operation to obtain vector i k+1 ′, use the Xavier function in Pytorch to initialize a matrix w′, and use the mul function in Pytorch to convert the vector i k+1 ' and the initialized matrix w' dot multiplication operation to obtain the vector i k+1 ″, vector i k+1 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i k+2 ;
[0023] c-4) The hidden state vector i k+2 Input into the LSTM model and output the predicted output
[0024] c-5) Calculate the predicted output and the k+3th takeoff and landing time period array H k+3 The mean square error loss function between them is used, and the Adam optimizer is used to train the LSTM model using the mean square error loss function to obtain the optimized LSTM model.
[0025] Furthermore, step d) comprises the following steps:
[0026] d-1) Set the takeoff and landing time period array H q-3 Input into the optimized LSTM model and output the hidden state vector i -3 ;
[0027] d-2) Use the torch.cat function in Pytorch to convert the hidden state vector i -3 and the takeoff and landing time period array H q-2 Perform splicing operation to obtain vector i -3 ′, use the Xavier function in Pytorch to initialize a matrix w″, and use the mul function in Pytorch to convert the vector i -3 ′ and the initialized matrix w″ dot multiplication operation to obtain vector i -3 ″, vector i -3 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i -2 ;
[0028] d-3) Use the torch.cat function in Pytorch to convert the hidden state vector i -2 and the takeoff and landing time period array H q-1 Perform splicing operation to obtain vector i -2 ′, use the Xavier function in Pytorch to initialize a matrix w″′, and use the mul function in Pytorch to convert the vector i -2 ' and the initialized matrix w'' dot multiplication operation to obtain the vector i -2 ″, vector i -2 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i -1 ;
[0029] d-4) The hidden state vector i -1 Input into the optimized LSTM model, and output the missing time period array H of business jet takeoff and landing q .
[0030] Furthermore, step e) comprises the following steps:
[0031] e-1) In the Gantt chart framework for business jet flights, a Gantt chart display area is established with index as the vertical axis and time as the horizontal axis, where index = {0, 1, 2, ..., N}, where N is the number of business jets;
[0032] e-2) The coordinates of the four vertices of the rectangle of the Gantt chart of the nth business jet are (t up ,n)、(t down ,n)、(t up ,n+g)、(t down ,n+g), n={1,2,...,N}, g is the height of the rectangle of the Gantt chart;
[0033] e-3) the vertex (t up ,n) and vertex (t down ,n) as the two endpoints of the line segment, the decision parameter p1 is calculated by the formula p1=Δy1-Δx1, where Δx1 is the difference in the horizontal coordinates, Δx1=|t up -t down |, Δy1 is the difference of the horizontal coordinates, Δy1=|nn|, if the decision parameter p1 is less than or equal to 0, then the vertex (t up ,n) and point (t up +Δx1,n) connects the first line segment of the rectangle of the output Gantt chart. If the decision parameter p1 is greater than 0, the vertex (t up ,n) and point (t up ,n+Δy1) connects the first line segment of the rectangle of the output Gantt chart;
[0034] e-4) the vertex (t up ,n+g) and vertex (t down ,n+g) as the two endpoints of the line segment, the decision parameter p2 is calculated by the formula p2=Δy2-Δx2, where Δx2 is the difference in the horizontal coordinates, Δx2=|t up -t down |, Δy2 is the difference of the horizontal coordinates, Δy2=|(n+g)-(n+g)|, if the decision parameter p2 is less than or equal to 0, then the vertex (t up ,n+g) and point (t up +Δx2,n+g) connects the second line segment of the rectangle of the output Gantt chart. If the decision parameter p2 is greater than 0, the vertex (t up ,n+g) and point (t up ,n+g+Δy2) connects the second line segment of the rectangle of the output Gantt chart;
[0035] e-5) the vertex (t up ,n) and vertex (t up,n+g) as the two endpoints of the line segment, the decision parameter p3 is calculated by the formula p3=Δy3-Δx3, where Δx3 is the difference in the horizontal coordinates, Δx3=|t up -t up |, Δy3 is the difference of the horizontal coordinates, Δy3=|n-(n+g)|, if the decision parameter p3 is less than or equal to 0, then the vertex (t up ,n) and point (t up +Δx3,n) connects the third line segment of the rectangle of the output Gantt chart. If the decision parameter p3 is greater than 0, the vertex (t up ,n) and point (t up ,n+Δy3) connects the third line segment of the rectangle of the output Gantt chart;
[0036] e-6) The vertex (t down ,n) and vertex (t down ,n+g) as the two endpoints of the line segment, the decision parameter p4 is calculated by the formula p4=Δy4-Δx4, where Δx4 is the difference in the horizontal coordinates, Δx4=|t down -t down |, Δy4 is the difference of the horizontal coordinates, Δy4=|n-(n+g)|, if the decision parameter p4 is less than or equal to 0, then the vertex (t down ,n) and point (t down +Δx4,n) connects the fourth line segment of the rectangle of the output Gantt chart. If the decision parameter p4 is greater than 0, the vertex (t down ,n) and point (t down ,n+Δy4) connects the fourth line segment of the rectangle of the output Gantt chart.
[0037] Further, step f) comprises the following steps:
[0038] f-1) Obtain the take-off and landing times of the business jet from the database, and calculate the total flight duration by subtracting the take-off time from the landing time of the business jet;
[0039] f-2) Use the setInterval function in JavaScript to set a timer to trigger the function execution every 1 second to update the display content of the Gantt chart;
[0040] f-3) Construct the updateGanttChart function and use it to calculate the difference between the current time and the business jet's takeoff time. Calculate the elapsed time percentage based on this difference and the total flight duration. Update the length of the progress bar in the Gantt chart based on the elapsed time percentage.
[0041] The beneficial effects of the present invention are: through H5 technology, real-time updating and display of flight Gantt charts are achieved, ensuring that users can obtain the latest flight information at any time. The Canvas drawing algorithm is used to achieve rapid drawing and rendering of flight Gantt charts, improving the efficiency and experience of users in obtaining information. Through the visual flight Gantt chart, the flight plan is intuitively displayed to the user, making it convenient for users to quickly understand flight information. This patented technology can be applied to various terminal devices, including PCs, mobile phones, tablets, etc., to meet the user's usage needs in different scenarios. It can not only be used for the visual display of flight Gantt charts, but can also be extended to progress management in other fields, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION
[0043] The following is combined with Figure 1 The present invention is further described.
[0044] A method for displaying a Gantt chart of a business jet flight comprises the following steps:
[0045] a) Create a Gantt chart framework for business jet flights to display flight progress information.
[0046] b) Obtain the take-off and landing time period data set T2 based on the take-off and landing time data of the business jet, where T2 = {H1, H2, ..., H k ,...,H K}, H k is the kth takeoff and landing time period array, k∈{1,2,…,K}, and K is the total number of takeoff and landing time period arrays.
[0047] c) Through the kth takeoff and landing time period array H k , the k+1th takeoff and landing time period array H k+1 , the k+2th takeoff and landing time period array H k+2 Train the LSTM model to obtain the optimized LSTM model.
[0048] d) Array H of missing time periods for business jet takeoff and landing q The first three takeoff and landing time period array H q-1 、H q-2 、H q-3 Input them into the optimized LSTM model respectively to obtain the missing time period array H of business jet takeoff and landing q Take-off time t up and the missing time period array H for business jet takeoff and landing q Landing time t down , Hq =[t up ,t down ], q-1∈{1,2,...,K}, q-2∈{1,2,...,K}, q-3∈{1,2,...,K}.
[0049] e) Based on the missing time period array H of business jet takeoff and landing q Draw the rectangle for the flight Gantt chart in the Business Jet Flight Gantt Chart frame.
[0050] f) Dynamically display the rectangles of the drawn flight Gantt chart.
[0051] Leveraging the new features of HTML5 technology, the data analysis capabilities of the LSTM algorithm, the styling capabilities of CSS3, and the interactivity and dynamism of JavaScript, this platform provides flight managers with a comprehensive, intuitive, and real-time flight progress display platform, better supporting flight scheduling and progress control. This platform improves overall business jet operational efficiency by approximately 20%-50%, reduces costs associated with mission delays by approximately 10%-15%, and reduces staff turnover.
[0052] In one embodiment of the present invention, in step a), a business jet flight Gantt chart framework is constructed using HTML5. HTML5 technology was used to develop the basic framework of the flight Gantt chart. First, the design of the flight Gantt chart should include a view that displays the overall progress of the flights, which can help users understand at a glance the current status of all flights and their position relative to the entire timeline. Next, to allow users to obtain more details about a specific flight, each flight item should expand to display detailed information such as take-off and landing times, flight number, aircraft type, and current status when selected. In addition, the action button area is an essential part, which allows users to perform various operations, such as adding a new flight plan, modifying the time of an existing flight, or deleting a flight plan. The design of this area should be intuitive and easy to use to enhance the user experience. In order to adapt the layout on both PC and mobile devices, a responsive layout is adopted to ensure good performance and compatibility on both PC and mobile devices.
[0053] In one embodiment of the present invention, step b) comprises the following steps:
[0054] b-1) Obtain the take-off and landing time data of M business jets, and obtain the data set T1, T1 = {t1, t2, ..., t m ,...,t M}, where t m is the take-off and landing time data of the mth business jet, m∈{1,2,...,M}, m is an odd number, t m is the take-off time of the business jet, and when m is an even number, tm The landing time of the business jet.
[0055] b-2) The take-off and landing time data of the mth business jet t m Convert to ISO86011 (YYYY-MM-DDHH:MM) format.
[0056] b-3) Through formula c m =t m -s Calculate and convert the mth business jet takeoff and landing time data t into ISO8601 format m The time difference c from the epoch time s m .
[0057] b-4) Through the formula Calculate the number of minutes h m , where y is the time difference c m The number of years, m1 is the time difference c m The number of months, d is the time difference c m The number of days, h is the time difference c m hours, m2 is the time difference c m The number of minutes is obtained, and the transformed data set T1′ is obtained, T1′={h1,h2,...,h m ,...,h M}.
[0058] b-5) By formula H k =[h m-1 ,h m ] k Create the kth takeoff and landing time period array H k ,
[0059] In one embodiment of the present invention, step c) comprises the following steps:
[0060] c-1) Set the kth takeoff and landing time period array H k Input into the LSTM model and output the hidden state vector i k .
[0061] c-2) Use the torch.cat function in Pytorch to convert the hidden state vector i k and the k+1th takeoff and landing time period array H k+1 Perform splicing operation to obtain vector i k ′, use the Xavier function in Pytorch to initialize a matrix w, and use the mul function in Pytorch to convert the vector i k ′ is multiplied by the initialized matrix w to obtain the vector i k″, vector i k ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i k+1 .
[0062] c-3) Use the torch.cat function in Pytorch to convert the hidden state vector i k+1 and the k+2th takeoff and landing time period array H k+2 Perform splicing operation to obtain vector i k+1 ′, use the Xavier function in Pytorch to initialize a matrix w′, and use the mul function in Pytorch to convert the vector i k+1 ' and the initialized matrix w' dot multiplication operation to obtain the vector i k+1 ″, vector i k+1 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i k+2 .
[0063] c-4) The hidden state vector i k+2 Input into the LSTM model and output the predicted output
[0064] c-5) Calculate the predicted output and the k+3th takeoff and landing time period array H k+3 The mean square error loss function between them is used, and the Adam optimizer is used to train the LSTM model using the mean square error loss function to obtain the optimized LSTM model.
[0065] In one embodiment of the present invention, step d) comprises the following steps:
[0066] d-1) Set the takeoff and landing time period array H q-3 Input into the optimized LSTM model and output the hidden state vector i -3 .
[0067] d-2) Use the torch.cat function in Pytorch to convert the hidden state vector i -3 and the takeoff and landing time period array H q-2 Perform splicing operation to obtain vector i -3 ′, use the Xavier function in Pytorch to initialize a matrix w″, and use the mul function in Pytorch to convert the vector i -3 ′ and the initialized matrix w″ dot multiplication operation to obtain vector i -3 ″, vector i -3 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i -2.
[0068] d-3) Use the torch.cat function in Pytorch to convert the hidden state vector i -2 and the takeoff and landing time period array H q-1 Perform splicing operation to obtain vector i -2 ′, use the Xavier function in Pytorch to initialize a matrix w″′, and use the mul function in Pytorch to convert the vector i -2 ' and the initialized matrix w'' dot multiplication operation to obtain the vector i -2 ″, vector i -2 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i -1 .
[0069] d-4) The hidden state vector i -1 Input into the optimized LSTM model, and output the missing time period array H of business jet takeoff and landing q .
[0070] In one embodiment of the present invention, step e) comprises the following steps:
[0071] e-1) In the Gantt chart framework for business jet flights, a Gantt chart display area is established with index as the vertical axis and time as the horizontal axis, where index = {0, 1, 2, ..., N}, where N is the number of business jets.
[0072] e-2) The coordinates of the four vertices of the rectangle of the Gantt chart of the nth business jet are (t up ,n)、(t down ,n)、(t up ,n+g)、(t down ,n+g), n={1,2,...,N}, g is the height of the rectangle of the Gantt chart.
[0073] e-3) the vertex (t up ,n) and vertex (t down ,n) as the two endpoints of the line segment, the decision parameter p1 is calculated by the formula p1=Δy1-Δx1, where Δx1 is the difference in the horizontal coordinates, Δx1=|t up -t down |, Δy1 is the difference of the horizontal coordinates, Δy1=|nn|, if the decision parameter p1 is less than or equal to 0, then the vertex (t up ,n) and point (t up +Δx1,n) connects the first line segment of the rectangle of the output Gantt chart. If the decision parameter p1 is greater than 0, the vertex (t up ,n) and point (t up,n+Δy1) connects the first line segment of the rectangle of the output Gantt chart.
[0074] e-4) the vertex (t up ,n+g) and vertex (t down ,n+g) as the two endpoints of the line segment, the decision parameter p2 is calculated by the formula p2=Δy2-Δx2, where Δx2 is the difference in the horizontal coordinates, Δx2=|t up -t down |, Δy2 is the difference of the horizontal coordinates, Δy2=|(n+g)-(n+g)|, if the decision parameter p2 is less than or equal to 0, then the vertex (t up ,n+g) and point (t up +Δx2,n+g) connects the second line segment of the rectangle of the output Gantt chart. If the decision parameter p2 is greater than 0, the vertex (t up ,n+g) and point (t up ,n+g+Δy2) connects the second line segment of the rectangle of the output Gantt chart.
[0075] e-5) the vertex (t up ,n) and vertex (t up ,n+g) as the two endpoints of the line segment, the decision parameter p3 is calculated by the formula p3=Δy3-Δx3, where Δx3 is the difference in the horizontal coordinates, Δx3=|t up -t up |, Δy3 is the difference of the horizontal coordinates, Δy3=|n-(n+g)|, if the decision parameter p3 is less than or equal to 0, then the vertex (t up ,n) and point (t up +Δx3,n) connects the third line segment of the rectangle of the output Gantt chart. If the decision parameter p3 is greater than 0, the vertex (t up ,n) and point (t up ,n+Δy3) connects the third line segment of the rectangle of the output Gantt chart.
[0076] e-6) The vertex (t down ,n) and vertex (t down ,n+g) as the two endpoints of the line segment, the decision parameter p4 is calculated by the formula p4=Δy4-Δx4, where Δx4 is the difference in the horizontal coordinates, Δx4=|t down -t down |, Δy4 is the difference of the horizontal coordinates, Δy4=|n-(n+g)|, if the decision parameter p4 is less than or equal to 0, then the vertex (t down ,n) and point (t down +Δx4,n) connects the fourth line segment of the rectangle of the output Gantt chart. If the decision parameter p4 is greater than 0, the vertex (t down ,n) and point (tdown ,n+Δy4) connects the fourth line segment of the rectangle of the output Gantt chart.
[0077] By using the timer function in JavaScript, the display content of the Gantt chart can be updated regularly to ensure that the progress of the flight can be reflected in the Gantt chart in real time. Specifically, in one embodiment of the present invention, step f) includes the following steps:
[0078] f-1) Obtain the take-off and landing times of the business jet from the database, and calculate the total flight duration by subtracting the take-off time from the landing time of the business jet.
[0079] f-2) Use the setInterval function in JavaScript to set a timer to trigger the function execution every 1 second to update the display content of the Gantt chart.
[0080] f-3) Construct an updateGanttChart function, use it to calculate the difference between the current time and the business jet's takeoff time, calculate the elapsed time ratio based on the difference and the total flight duration, and update the length of the progress bar in the Gantt chart based on the elapsed time ratio. This allows for a dynamic flight Gantt chart that reflects the flight's progress in real time, providing users with a more intuitive and timely information display. Finally, it should be noted that the foregoing is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will appreciate that they can modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.
Claims
1. A method for displaying a Gantt chart of a business jet flight, characterized in that: The steps include: a) Establish a Gantt chart framework for business jet flights to display flight progress information; b) Obtain the take-off and landing time period data set T2 based on the take-off and landing time data of the business jet, where T2 = {H1, H2, ..., H k ,...,H K }, H k is the kth takeoff and landing time period array, k∈{1,2,...,K}, K is the total number of takeoff and landing time period arrays; c) Through the kth takeoff and landing time period array H k , the k+1th takeoff and landing time period array H k+1 , the k+2th takeoff and landing time period array H k+2 Train the LSTM model to obtain the optimized LSTM model; d) Array H of missing time periods for business jet takeoff and landing q The first three takeoff and landing time period array H q-1 、H q-2 、H q-3 Input them into the optimized LSTM model respectively to obtain the missing time period array H of business jet takeoff and landing q Take-off time t up and the missing time period array H for business jet takeoff and landing q Landing time t down , H q =[t up ,t down ], q-1∈{1,2,...,K}, q-2∈{1,2,...,K}, q-3∈{1,2,...,K}; e) Based on the missing time period array H of business jet takeoff and landing q Draw a rectangle for the flight Gantt chart in the business jet flight Gantt chart frame; f) Dynamically display the rectangles of the drawn flight Gantt chart; Wherein, step d) comprises the following steps: d-1) Set the takeoff and landing time period array H q-3 Input into the optimized LSTM model and output the hidden state vector i -3 ; d-2) Use the torch.cat function in Pytorch to convert the hidden state vector i -3 and the takeoff and landing time period array H q-2 Perform splicing operation to obtain vector i -3 ′, use the Xavier function in Pytorch to initialize a matrix w″, and use the mul function in Pytorch to convert the vector i -3 ′ and the initialized matrix w″ dot multiplication operation to obtain vector i -3 ″, vector i -3 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i -2 ; d-3) Use the torch.cat function in Pytorch to convert the hidden state vector i -2 and the takeoff and landing time period array H q-1 Perform splicing operation to obtain vector i -2 ′, use the Xavier function in Pytorch to initialize a matrix w″′, and use the mul function in Pytorch to convert the vector i -2 ' and the initialized matrix w'' dot multiplication operation to obtain the vector i -2 ″, vector i -2 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i -1 ; d-4) The hidden state vector i -1 Input into the optimized LSTM model, and output the missing time period array H of business jet takeoff and landing q .
2. The business jet flight Gantt chart display method according to claim 1, characterized in that: In step a), a Gantt chart framework of business jet flights is constructed using HTML5.
3. The business jet flight Gantt chart display method according to claim 1, characterized in that: Step b) comprises the following steps: b-1) Obtain the take-off and landing time data of M business jets, and obtain the data set T1, T1 = {t1, t2, ..., t m ,...,t M }, where t m is the take-off and landing time data of the mth business jet, m∈{1,2,...,M}, m is an odd number, t m is the take-off time of the business jet, and when m is an even number, t m The landing time of the business jet; b-2) The take-off and landing time data of the mth business jet t m Convert to ISO8601 format; b-3) Through formula c m =t m -s Calculate and convert the mth business jet takeoff and landing time data t into ISO8601 format m The time difference c from the epoch time s m ; b- 4) Through the formula Calculate the number of minutes h m , where y is the time difference c m The number of years, m1 is the time difference c m The number of months, d is the time difference c m The number of days, h is the time difference c m hours, m2 is the time difference c m The number of minutes is obtained, and the transformed data set T1′ is obtained, T1′={h1,h2,...,h m ,...,h M }; b-5) by formula H k =[h m-1 ,h m ] k Create the kth takeoff and landing time period array H k , 4. The business jet flight Gantt chart display method according to claim 1, characterized in that: Step c) comprises the following steps: c-1) Set the kth takeoff and landing time period array H k Input into the LSTM model and output the hidden state vector i k ; c-2) Use the torch.cat function in Pytorch to convert the hidden state vector i k and the k+1th takeoff and landing time period array H k+1 Perform splicing operation to obtain vector i k ′, use the Xavier function in Pytorch to initialize a matrix w, and use the mul function in Pytorch to convert the vector i k ′ is multiplied by the initialized matrix w to obtain the vector i k ″, vector i k ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i k+1 ; c-3) Use the torch.cat function in Pytorch to convert the hidden state vector i k+1 and the k+2th takeoff and landing time period array H k+2 Perform splicing operation to obtain vector i k+1 ′, use the Xavier function in Pytorch to initialize a matrix w′, and use the mul function in Pytorch to convert the vector i k+1 ' and the initialized matrix w' dot multiplication operation to obtain the vector i k+1 ″, vector i k+1 ″ is input into the Sigmoid function in Pytorch, and the output is the hidden state vector i k+2 ; c-4) The hidden state vector i k+2 Input into the LSTM model and output the predicted output c-5) Calculate the predicted output and the k+3th takeoff and landing time period array H k+3 The mean square error loss function between them is used, and the Adam optimizer is used to train the LSTM model using the mean square error loss function to obtain the optimized LSTM model.
5. The business jet flight Gantt chart display method according to claim 1, characterized in that: Step e) comprises the following steps: e-1) In the Gantt chart framework for business jet flights, a Gantt chart display area is established with index as the vertical axis and time as the horizontal axis, where index = {0, 1, 2, ..., N}, where N is the number of business jets; e-2) The coordinates of the four vertices of the rectangle of the Gantt chart of the nth business jet are (t up ,n)、(t down ,n)、(t up ,n+g)、(t down ,n+g), n={1,2,...,N}, g is the height of the rectangle of the Gantt chart; e-3) the vertex (t up ,n) and vertex (t down ,n) as the two endpoints of the line segment, the decision parameter p1 is calculated by the formula p1=Δy1-Δx1, where Δx1 is the difference in the horizontal coordinates, Δx1=|t up -t down |, Δy1 is the difference of the horizontal coordinates, Δy1=|nn|, if the decision parameter p1 is less than or equal to 0, then the vertex (t up ,n) and point (t up +Δx1,n) connects the first line segment of the rectangle of the output Gantt chart. If the decision parameter p1 is greater than 0, the vertex (t up ,n) and point (t up ,n+Δy1) connects the first line segment of the rectangle of the output Gantt chart; e-4) the vertex (t up ,n+g) and vertex (t down ,n+g) as the two endpoints of the line segment, the decision parameter p2 is calculated by the formula p2=Δy2-Δx2, where Δx2 is the difference in the horizontal coordinates, Δx2=|t up -t down |, Δy2 is the difference of the horizontal coordinates, Δy2=|(n+g)-(n+g)|, if the decision parameter p2 is less than or equal to 0, then the vertex (t up ,n+g) and point (t up +Δx2,n+g) connects the second line segment of the rectangle of the output Gantt chart. If the decision parameter p2 is greater than 0, the vertex (t up ,n+g) and point (t up ,n+g+Δy2) connects the second line segment of the rectangle of the output Gantt chart; e-5) the vertex (t up ,n) and vertex (t up ,n+g) as the two endpoints of the line segment, the decision parameter p3 is calculated by the formula p3=Δy3-Δx3, where Δx3 is the difference in the horizontal coordinates, Δx3=|t up -t up |, Δy3 is the difference of the horizontal coordinates, Δy3=|n-(n+g)|, if the decision parameter p3 is less than or equal to 0, then the vertex (t up ,n) and point (t up +Δx3,n) connects the third line segment of the rectangle of the output Gantt chart. If the decision parameter p3 is greater than 0, the vertex (t up ,n) and point (t up ,n+Δy3) connects the third line segment of the rectangle of the output Gantt chart; e-6) The vertex (t down ,n) and vertex (t down ,n+g) as the two endpoints of the line segment, the decision parameter p4 is calculated by the formula p4=Δy4-Δx4, where Δx4 is the difference in the horizontal coordinates, Δx4=|t down -t down |, Δy4 is the difference of the horizontal coordinates, Δy4=|n-(n+g)|, if the decision parameter p4 is less than or equal to 0, then the vertex (t down ,n) and point (t down +Δx4,n) connects the fourth line segment of the rectangle of the output Gantt chart. If the decision parameter p4 is greater than 0, the vertex (t down ,n) and point (t down ,n+Δy4) connects the fourth line segment of the rectangle of the output Gantt chart.
6. The business jet flight Gantt chart display method according to claim 1, characterized in that: Step f) comprises the following steps: f-1) Obtain the take-off and landing times of the business jet from the database, and calculate the total flight duration by subtracting the take-off time from the landing time of the business jet; f-2) Use the setInterval function in JavaScript to set a timer to trigger the function execution every 1 second to update the display content of the Gantt chart; f-3) Construct the updateGanttChart function and use it to calculate the difference between the current time and the business jet's takeoff time. Calculate the elapsed time percentage based on this difference and the total flight duration. Update the length of the progress bar in the Gantt chart based on the elapsed time percentage.
Citation Information
Patent Citations
Method and system for simultaneously displaying flight news and plans on same Gantt chart interface
CN103544672A
Intelligent flight delay prediction method based on deep learning
CN113128769A