A real-time calculation method and system for global airport taxes and fuel charges based on multi-level cache
Through the calculation method based on multi-level caching, the performance bottleneck problem of real-time computing systems for airport taxes and fuel fees is solved, efficient real-time computing is achieved, and the service efficiency and service quality of airlines are improved.
Patent Information
- Application Number
- CN202411238119.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-09-05
AI Technical Summary
The global airport tax and fuel fee real-time computing system has low computing performance due to its complex computing logic and performance bottlenecks of traditional cache technology, which limits the airline's customized services for the sales environment and the improvement of passenger service quality.
Using a multi-level cache-based calculation method, the real-time calculation efficiency of airport tax and fuel expenses is improved by forming K-V structure cache data, combining high-frequency route prediction and real-time calculation rules.
On the premise of ensuring computing accuracy, it significantly improves real-time computing performance, improves the efficiency of airlines' external air ticket inquiry, booking and refund services, meets the needs of high throughput and fast response time, and provides passengers with efficient and accurate Passenger ticket service.
Smart Images

Figure CN118780844B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of real-time calculation of global airport taxes and fuel charges, and in particular relates to a real-time calculation method and system of global airport taxes and fuel charges based on multi-level cache. Background Art
[0002] With the continuous increase in global tourism and business travel demand, the access needs of various users to airline e-commerce ticketing systems have increased significantly; and the real-time calculation of global airport taxes and fuel charges often faces computing performance challenges due to its extremely complex calculation logic. On the other hand, because traditional calculation methods have performance bottlenecks, they often limit airlines from further combining their own advantages to customize the service transformation of the sales environment, which restricts the quality of service to passengers.
[0003] In addition, as the core sales system of each airline, the global airport tax and fuel surcharge real-time calculation system is also facing large-scale access requirements from different sales channels, and traditional cache technology often restricts the performance of the system. Therefore, it is necessary to design a new and effective multi-level cache-based calculation method to improve the real-time calculation efficiency of global civil aviation ticket airport taxes and fuel surcharges. Summary of the invention
[0004] In view of the above problems, the purpose of the present invention is to provide a real-time calculation method and system for global airport taxes and fuel charges based on multi-level cache, which can be applied to the civil aviation e-commerce ticketing scenario, effectively deal with the performance bottleneck caused by large-scale concurrent requests, and greatly improve the performance of real-time calculation while ensuring the accuracy of the calculation of airport taxes and fuel charges for passenger tickets. After the ticketing system obtains the original data of airport taxes and fuel charges and parses them into the physical database, the system can intelligently adapt the high-frequency route prediction cache, real-time calculation rule cache, original calculation rule cache, and intelligently select the most suitable cache data, so as to more effectively support the real-time calculation needs of civil aviation passenger tickets airport taxes and fuel charges, and provide an important technical solution for the real-time calculation of the scale of civil aviation passenger ticket sales.
[0005] The specific technical solution for achieving the purpose of the present invention is:
[0006] A real-time calculation method for global airport taxes and fuel charges based on multi-level cache includes the following steps:
[0007] Step 1: Based on the original ticket business rule data of airport tax and fuel fee, form KV structure cache data based on ticket business rules;
[0008] Step 2: Based on the KV structure cache data based on the original ticket business rules generated in step 1, the real-time calculation rule structure is extracted in combination with the attribute dependency library based on real-time calculation, and the KV structure cache data oriented to the real-time calculation rules is generated;
[0009] Step 3: Receive and record the real-time calculation requests for airport taxes and fuel charges from various ticket sales channels, and extract route information;
[0010] Step 4: Based on the real-time calculation request information of airport tax and fuel fee extracted in step 3, adapt it with the KV structure cache data of the high-frequency route prediction results in the high-frequency route prediction cache data; if the adaptation is successful, the calculation result is directly fed back; if the adaptation is unsuccessful, it is distributed in turn to the KV structure cache data based on the real-time calculation rules and the KV structure cache data based on the original ticket business rules for further adaptation, and the corresponding calculation results are fed back.
[0011] The present invention also provides a global airport tax and fuel fee real-time calculation system based on multi-level cache, comprising the following modules:
[0012] Original calculation rule cache module: used for original ticket business rule data based on airport tax and fuel fee, forming KV structure cache data based on original ticket business rules;
[0013] Real-time calculation rule cache module: used to generate KV structure cache data based on the generated original ticket business rules, extract the real-time calculation rule structure based on the attribute dependency library based on real-time calculation, and generate KV structure cache data for real-time calculation rules;
[0014] Route analyzer module: used to receive and record the real-time calculation requests for airport taxes and fuel charges from various channels, extract route information, and obtain the real-time calculation results of airport taxes and fuel charges based on the route information;
[0015] High-frequency route prediction module: used to build a high-frequency route prediction model based on a deep time series structure, and to train based on a high-frequency route model training data set. The data set obtains the real-time calculation request log data of airport taxes and fuel charges from each channel according to a set period, and extracts high-frequency route features in the log data, including flight frequency, sales popularity, and flight time, thereby forming a high-frequency route model training data set, and iteratively trains the high-frequency route prediction model based on the continuously updated high-frequency route model training data set;
[0016] High-frequency route prediction cache module: It is used to perform real-time calculation based on the high-frequency route data predicted by the high-frequency route prediction module according to the KV structure cache data for real-time calculation rules, so as to determine the airport tax and fuel fee calculation results of each high-frequency route, and form the airport tax and fuel fee calculation results corresponding to the high-frequency route into KV structure cache data, where K represents the high-frequency route input parameter combination data, and V represents the airport tax and fuel fee calculation results corresponding to the high-frequency route input parameters.
[0017] Compared with the prior art, the present invention has the following beneficial effects:
[0018] The solution of the present invention can effectively handle the performance bottleneck problem caused by large-scale concurrent requests while ensuring the accuracy of the original system calculation, thereby improving the efficiency of airlines' external ticket query, booking, refund and change services, meeting the performance index requirements of civil aviation ticket sales for real-time calculation throughput and request response time, and providing important support for providing passengers with efficient and accurate ticket services;
[0019] The solution of the present invention designs a route analysis and multi-level cache architecture. The component design concept of high cohesion and low coupling is adopted between caches at each level. While ensuring computing efficiency, rapid horizontal expansion or component iteration can be performed. At the same time, based on the form of multi-level cache, multiple levels are used to gradually eliminate the access traffic generated by scale concurrency. The log data generated during the real-time computing process can more effectively record the real-time computing process, providing a basis for subsequent fault location for system operation and maintenance technicians.
[0020] The solution of the present invention generates and stores multi-level cache structure data based on a multi-level cache mode. When a large number of concurrent requests occur, the calculation results of airport taxes and fuel charges can be quickly fed back based on the multi-level cache structure data; at the same time, in the operation of the overall system and process, the high-frequency route prediction model can be iterated more accurately based on the accumulation of request data, thereby further improving the real-time calculation efficiency of global airport taxes and fuel charges.
[0021] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 The present invention is a flowchart of a method for real-time calculation of global airport taxes and fuel charges based on multi-level cache.
[0023] Figure 2 The present invention is a schematic diagram of the architecture of a real-time calculation system for global airport taxes and fuel charges based on multi-level cache. DETAILED DESCRIPTION
[0024] Example
[0025] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0026] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an kind" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.
[0027] Combination Figure 1 , a real-time calculation method for global airport taxes and fuel charges based on multi-level cache, comprising the following steps:
[0028] Step 1: Based on the original ticket business rule data of airport tax and fuel fee, form KV structure cache data based on the original ticket business rules:
[0029] The original ticket business rule data of the airport tax and fuel surcharge includes an exchange rate data table, a flight distance data table, an airport tax and fuel surcharge core rule data table, and an airport tax and fuel surcharge auxiliary rule data table;
[0030] The KV structure cache data based on the original ticket business rules includes a set corresponding to the above data, that is, the above rule data is extracted from the original business physical database to form KV structure cache data, where K is the attribute or attribute combination of the corresponding table in the relational database corresponding to the original ticket business rule data of airport tax and fuel fee, and V is the value corresponding to the attribute or attribute combination of the table to which it belongs;
[0031] For example, some data in the KV structure cache data of the original ticket business rule data in this embodiment is:
[0032] Key: USDCNY, Value: 7.13325485; the Key consists of the three-letter codes of the two currencies that need to be converted, and the Value is the exchange rate corresponding to the two currencies.
[0033] The above data is the table structure data required for the calculation of airport taxes and fuel charges extracted from the original exchange rate data table. The KV structure cache data based on the original ticket business rule data is stored in the memory to prepare for the subsequent real-time calculation of airport taxes and fuel charges.
[0034] Step 2: Based on the KV structure cache data based on the original ticket business rules generated in step 1, extract the real-time calculation rule structure in combination with the attribute dependency library based on real-time calculation, and generate KV structure cache data for real-time calculation rules:
[0035] Step 2-1: For the KV structure cache data based on the original ticket business rules, build an attribute dependency library based on real-time calculation, and convert the original ticket business rules into attribute dependencies applicable to the real-time calculation of airport taxes and fuel charges;
[0036] Step 2-2, using the attribute dependency library based on real-time calculation, extract the real-time calculation rule structure from the KV structure cache data based on the original ticket business rules, and generate KV structure cache data for real-time calculation rules, where K represents the key attribute set for real-time calculation rules transformed from the original ticket business rules based on the attribute dependency relationship, and V represents the value corresponding to the key attribute set.
[0037] For example, when processing the distance structure in the original ticket business rules, the distance structure of the original ticket business rules is the flight distance between regions, while the input parameters obtained in the actual calculation process are airports and airports. Through the designed attribute dependency relationship for real-time calculation, the distance between regions can be converted into the distance between each airport. During calculation, it can be directly adapted to the distance structure for real-time calculation. The following is the example data extracted based on the KV structure for real-time calculation rules:
[0038] Key: SHAPEK, Value: 1178; the Key is composed of the three-letter codes of the airports extracted from the original ticket business rule structure based on the attribute dependency, and the Value is the corresponding flight distance between the two airports.
[0039] The KV structure cache data for real-time calculation rules generated in this step is the basis for real-time calculation of all airport taxes and fuel charges in this solution. When real-time calculation of airport taxes and fuel charges is required, the input parameters corresponding to the route need to be adapted in the KV structure cache data for real-time calculation rules, and the corresponding V value is determined based on the value of K, so as to obtain the calculation parameters of airport taxes and fuel charges and complete the calculation of airport taxes and fuel charges.
[0040] Step 3: Receive and record the real-time calculation requests for airport taxes and fuel charges from various ticket sales channels, and extract route information, specifically:
[0041] Calculate request data in real time based on airport taxes and fuel charges, extract route information contained in the request data, and split and adapt the request input parameters;
[0042] During the splitting process, if the route information includes multiple flight segments, the multiple flight segments are iteratively split and adapted according to the verification rule library. For example, if the route has 3 flight segments, the corresponding travel time should have 3 segments.
[0043] During the splitting process, the data format of the input parameters is also verified based on the verification rule library. For example, the passenger type parameter (adult, child) must be filled in the three-letter code format. If the input parameter format is abnormal, an error message is returned and the airport tax and fuel fee cannot be calculated.
[0044] After the split is completed, the data in the split calculation request is stored in the real-time calculation log library of fuel charges and airport taxes. The log library is used to provide a training database for the subsequent high-frequency route prediction model. It can be said that the method and system in this solution will continuously accumulate data during operation, thereby continuously iterating the training of the high-frequency route prediction model. The longer it runs, the better the implementation effect of this solution.
[0045] This example provides a query parameter sample that complies with the verification rules:
[0046] Route: SHA-PEK, i.e. Shanghai Hongqiao International Airport to Beijing Capital International Airport (SHA-PEK); Ticketing date: 2024-08-21; Departure date: 2024-08-25 08:00 Landing date: 2024-08-25 10:15; Currency type: CNY; Passenger type: ADT; Ticketing party code: MU; Market party code: MU; Carrier code: MU; Aircraft type: ARJ; Cabin: S
[0047] The route type includes one of the following types: international one-way, international round-trip, domestic one-way, domestic round-trip, domestic circle, domestic open-jaw, international circle, and international open-jaw.
[0048] In the above example, the route is SHA-PEK, and the route type is a purely domestic one-way flight.
[0049] The received airport tax and fuel fee real-time calculation request input is split, and the single-segment flight itinerary, cabin, ticketing time, take-off and landing time, ticketing airline, marketing airline, carrier airline, passenger type, aircraft model, and fare information in the input are recorded in the request past log fuel fee and airport tax real-time calculation log library.
[0050] Step 4: Based on the airport tax and fuel fee real-time calculation request information extracted in step 3, the KV structure cache data of the high-frequency route prediction result in the high-frequency route prediction cache data is adapted. If the adaptation is successful, the calculation result is fed back; if the adaptation is unsuccessful, further adaptation is performed based on the KV structure cache data for real-time calculation rules and the KV structure cache data based on the original ticket business rule data, and the corresponding calculation results are fed back;
[0051] Among them, the high-frequency route prediction cache data uses a high-frequency route prediction model to predict the high-frequency routes, and the prediction result data is calculated in real time according to the KV structure cache data for real-time calculation rules, so as to determine the airport tax and fuel fee calculation results of each high-frequency route, and the airport tax and fuel fee calculation results corresponding to the high-frequency route are formed into KV structure cache data, where K represents the high-frequency route input parameter combination data, and V represents the airport tax and fuel fee calculation results corresponding to the high-frequency route input parameters.
[0052] This embodiment provides example data of extracting KV structure cache data of high-frequency route prediction results:
[0053] Key:SHA-PVG;|MU;|MU;|S;|CNY|Mon Aug 21 00:00:00 CST 2024|zh|;
[0054] Value:{code='CN',currency=CNY, amount='50', description='Airport construction fee'},{code='YQ', currency=CNY, amount='70', description='Fuel fee'};
[0055] In addition, the high-frequency route prediction model is constructed based on a deep time series structure and trained based on a high-frequency route model training data set;
[0056] Obtain the real-time calculation request log data of airport taxes and fuel charges from each channel according to the set period, and extract the high-frequency route features in the log data, including flight frequency, sales popularity, and flight time, to form a high-frequency route model training data set;
[0057] The high-frequency route prediction model is iteratively trained based on the continuously updated high-frequency route model training data set, and the high-frequency route prediction model is updated while meeting the precision and recall evaluation indicators. Finally, the updated high-frequency route prediction model is used in the inference stage, and the model is connected and applied to the high-frequency route prediction cache.
[0058] The high-frequency route model dataset provides flight frequencies, for example: PVG-PEK: 1000 times / day, PVGSHA-NKGPEK: 500 times / day, combined with the above-mentioned deep training of the high-frequency route prediction model based on the deep time series structure.
[0059] High-frequency route prediction model inference route input example: route: PVG-PEK and assemble various inputs, such as: cabin, ticketing time, take-off and landing time, ticketing airline, market airline, carrier airline, passenger type, aircraft type, fare information, to form high-frequency route inputs to participate in the calculation of high-frequency route airport tax and fuel charges;
[0060] The specific process of feedback calculation results is:
[0061] Step 4-1: Based on the extracted airport tax and fuel fee real-time calculation request information, the route input parameter information is adapted to the KV structure cache data of the high-frequency route prediction result in the high-frequency route prediction cache data. If the adaptation is successful, the calculation result data in the KV structure cache data is directly fed back, otherwise, it goes to step 4-2;
[0062] Generally speaking, based on the aforementioned high-frequency route prediction model and the corresponding KV structure cache data of the high-frequency route prediction results, in actual operation, most requests can obtain the corresponding calculation result data in this step. Due to the multi-level cache design, the airport tax and fuel fee calculation results with considerable timeliness can be obtained quickly after the client's fee calculation request is issued;
[0063] Step 4-2: Adapt the extracted route input information of the airport tax and fuel fee real-time calculation request information to the KV structure cache data for real-time calculation rules. If the adaptation is successful, perform the airport tax and fuel fee real-time calculation based on the KV structure cache data for real-time calculation rules, and feedback the calculation result; otherwise, proceed to step 4-3;
[0064] Step 4-3: Adapt the unadapted information to the KV structure cache data based on the original ticket business rule data, and convert the adapted KV structure cache data based on the original ticket business rule data into the KV structure cache data for real-time calculation rules, and perform real-time calculation of airport taxes and fuel charges based on this, and feedback the calculation results.
[0065] The above two steps are mainly to provide a bottom line for real-time calculation requests for airport taxes and fuel charges. When the request does not belong to a high-frequency route, the KV structure cache data for real-time calculation rules is used for fast real-time calculation. If the corresponding rule data does not exist in the KV structure cache data for real-time calculation rules due to some data update cycle, memory limitations and other issues, the KV structure cache data based on the original ticket business rules, or even the original airport tax and fuel charge rule relational database, is used to extract the corresponding data, and then returns to the real-time calculation part for calculation. However, this situation rarely occurs, especially when large-scale concurrent requests occur, and most of the requests may be cached in the KV structure cache of high-frequency route prediction results. The calculation results can be directly obtained from the stored data for feedback. A small number of requests can be calculated in real time based on the KV structure cache data for real-time calculation rules. In very rare cases, data adaptation is required in the KV structure cache data based on the original ticket business rules. This ensures that the accuracy of the original system calculations can be guaranteed while effectively handling the performance bottleneck caused by large-scale concurrent requests. In addition, the log data generated during the real-time calculation process can be used to more effectively record the real-time calculation process, thereby improving the service efficiency of airlines in external ticket inquiries, bookings, refunds and changes, meeting the performance index requirements of civil aviation ticket sales for real-time calculation throughput and request response time, and providing important support for providing passengers with efficient and accurate ticket services.
[0066] In addition, combined Figure 2 The present invention also provides a real-time calculation system for global airport taxes and fuel charges based on multi-level cache, comprising the following modules:
[0067] Original calculation rule cache module: used for original ticket business rule data based on airport tax and fuel fee, forming KV structure cache data based on original ticket business rule data;
[0068] Real-time calculation rule cache module: used to generate KV structure cache data based on the generated original ticket business rule data, extract the real-time calculation rule structure based on the attribute dependency library based on real-time calculation, and generate KV structure cache data for real-time calculation rules;
[0069] Route analyzer module: used to receive and record the real-time calculation requests for airport taxes and fuel charges from various channels, extract route information, and obtain the real-time calculation results of airport taxes and fuel charges based on the route information;
[0070] High-frequency route prediction module: used to build a high-frequency route prediction model based on a deep time series structure, and to train based on a high-frequency route model training data set. The data set obtains the real-time calculation request log data of airport taxes and fuel charges from each channel according to a set period, and extracts high-frequency route features in the log data, including flight frequency, sales popularity, and flight time, thereby forming a high-frequency route model training data set, and iteratively trains the high-frequency route prediction model based on the continuously updated high-frequency route model training data set;
[0071] High-frequency route prediction cache module: It is used to perform real-time calculation based on the high-frequency route data predicted by the high-frequency route prediction module according to the KV structure cache data for real-time calculation rules, so as to determine the airport tax and fuel fee calculation results of each high-frequency route, and form the airport tax and fuel fee calculation results corresponding to the high-frequency route into KV structure cache data, where K represents the high-frequency route input parameter combination data, and V represents the airport tax and fuel fee calculation results corresponding to the high-frequency route input parameters.
[0072] The process of the route analyzer module obtaining the real-time calculation results of airport tax and fuel charges based on route information is as follows:
[0073] Based on the extracted airport tax and fuel fee real-time calculation request information, the route input parameter information is adapted to the KV structure cache data of the high-frequency route prediction result in the high-frequency route prediction cache data of the high-frequency route prediction cache module. If the adaptation is successful, the calculation result data in the KV structure cache data is directly fed back. Otherwise:
[0074] The extracted route input parameter information of the airport tax and fuel fee real-time calculation request information is adapted to the KV structure cache data for real-time calculation rules. If the adaptation is successful, the airport tax and fuel fee are calculated in real time based on the KV structure cache data for real-time calculation rules, and the calculation result is fed back. Otherwise:
[0075] The unadapted information is adapted to the KV structure cache data based on the original ticket business rules, and the adapted KV structure cache data based on the original ticket business rules is converted into KV structure cache data oriented to real-time calculation rules, and based on this, real-time calculation of airport taxes and fuel charges is performed, and the calculation results are fed back.
[0076] The above embodiments show and describe the basic principles and main features of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.
Claims
1. A real-time calculation method for global airport taxes and fuel charges based on multi-level cache, characterized in that: The following steps are involved: Step 1: Based on the original ticket business rule data of airport tax and fuel fee, form KV structure cache data based on ticket business rules: The original ticket business rule data of the airport tax and fuel surcharge includes global exchange rate data, flight distance data, airport tax and fuel surcharge core rule data, and airport tax and fuel surcharge auxiliary rule data; The KV structure cache data based on the original ticket business rules includes a set corresponding to the above data, where K is the primary attribute set of the corresponding table of the original ticket business rule data of airport tax and fuel fee in the relational database, and V is the non-primary attribute set of the corresponding table; Step 2: Based on the KV structure cache data based on the original ticket business rules generated in step 1, extract the real-time calculation rule structure in combination with the attribute dependency library based on real-time calculation, and generate KV structure cache data for real-time calculation rules: Step 2-1: For the KV structure cache data based on the original ticket business rules, build an attribute dependency library based on real-time calculation, and convert the original ticket business rules into attribute dependencies applicable to the real-time calculation of airport taxes and fuel charges; Step 2-2: Using the attribute dependency library based on real-time calculation, extract the real-time calculation rule structure from the KV structure cache data based on the original ticket business rules, and generate KV structure cache data for real-time calculation rules, where K represents the key attribute set for real-time calculation rules corresponding to the transformation of the original ticket business rules based on the attribute dependency relationship, and V represents the value corresponding to the key attribute set; Step 3: Receive and record the real-time calculation requests for airport taxes and fuel charges from various ticket sales channels, and extract route information; Step 4: Based on the real-time calculation request information of airport tax and fuel fee extracted in step 3, the KV structure cache data of the high-frequency route prediction results in the high-frequency route prediction cache data is adapted. If the adaptation is successful, the calculation result is directly fed back. If the adaptation is unsuccessful, it is distributed to the KV structure cache data based on the real-time calculation rules and the KV structure cache data based on the original ticket business rules in turn for further adaptation, and the corresponding calculation results are fed back.
2. The real-time calculation method for global airport taxes and fuel charges based on multi-level cache according to claim 1 is characterized in that: The route information extraction in step 3 is specifically as follows: Calculate request data in real time based on airport taxes and fuel charges, extract route information contained in the request data, and split and adapt the request input parameters; During the splitting process, if the route information includes multiple flight segments, the multiple flight segments are iteratively split according to the verification rule base; During the splitting process, the input data format is verified based on the verification rule library; After the splitting is completed, the data in the split calculation request is stored in the fuel fee and airport tax real-time calculation log library.
3. The real-time calculation method for global airport taxes and fuel charges based on multi-level cache according to claim 2 is characterized in that: The high-frequency route prediction cache data in step 4 is specifically: The high-frequency route prediction model is used to predict the high-frequency routes. The prediction result data is calculated in real time according to the KV structure cache data for real-time calculation rules, so as to determine the airport tax and fuel fee calculation results of each high-frequency route, and the airport tax and fuel fee calculation results corresponding to the high-frequency route are formed into KV structure cache data, where K represents the input parameter combination data of the high-frequency route, and V represents the airport tax and fuel fee calculation results corresponding to the input parameters of the high-frequency route.
4. The real-time calculation method for global airport taxes and fuel charges based on multi-level cache according to claim 3 is characterized in that: The high-frequency route prediction model is constructed based on a deep time series structure and trained based on a high-frequency route model training data set; Obtain the real-time calculation request log data of airport taxes and fuel charges from each channel according to the set period, and extract the high-frequency route features in the log data, including flight frequency, sales popularity, and flight time, to form a high-frequency route model training data set; The high-frequency route prediction model is iteratively trained based on the continuously updated high-frequency route model training data set.
5. The real-time calculation method for global airport taxes and fuel charges based on multi-level cache according to claim 3 is characterized in that: The process of adapting and feeding back the calculation result based on the real-time calculation request information of airport tax and fuel fee extracted in step 3 in step 4 is as follows: Step 4-1: Based on the extracted airport tax and fuel fee real-time calculation request information, adapt the route input information and the KV structure cache data of the high-frequency route prediction result in the high-frequency route prediction cache data; if the adaptation is successful, directly feedback the calculation result data in the KV structure cache data, otherwise proceed to step 4-2; Step 4-2, adapt the extracted route input parameter information of the airport tax and fuel fee real-time calculation request information to the KV structure cache data oriented to the real-time calculation rule; if the adaptation is successful, perform the airport tax and fuel fee real-time calculation based on the KV structure cache data oriented to the real-time calculation rule, and feedback the calculation result, otherwise proceed to step 4-3; Step 4-3: adapt the unadapted information to the KV structure cache data based on the original ticket business rules, and convert the successfully adapted KV structure cache data based on the original ticket business rules into KV structure cache data for real-time calculation rules, and perform real-time calculation of airport taxes and fuel charges based on this, and feedback the calculation results.
6. A real-time calculation system for global airport taxes and fuel charges based on multi-level cache, characterized in that: Includes the following modules: Original calculation rule cache module: used for original ticket business rule data based on airport tax and fuel fee, forming KV structure cache data based on original ticket business rules; The original ticket business rule data of the airport tax and fuel surcharge includes global exchange rate data, flight distance data, airport tax and fuel surcharge core rule data, and airport tax and fuel surcharge auxiliary rule data; The KV structure cache data based on the original ticket business rules includes a set corresponding to the above data, where K is the primary attribute set of the corresponding table of the original ticket business rule data of airport tax and fuel fee in the relational database, and V is the non-primary attribute set of the corresponding table; Real-time calculation rule cache module: used to generate KV structure cache data based on the generated original ticket business rules, extract the real-time calculation rule structure based on the attribute dependency library based on real-time calculation, and generate KV structure cache data for real-time calculation rules, specifically: For the KV structure cache data based on the original ticket business rules, build an attribute dependency library based on real-time calculation to convert the original ticket business rules into attribute dependencies applicable to the real-time calculation of airport taxes and fuel charges; By using the attribute dependency library based on real-time calculation, the real-time calculation rule structure is extracted from the KV structure cache data based on the original ticket business rules, and the KV structure cache data for real-time calculation rules is generated, where K represents the key attribute set for real-time calculation rules transformed from the original ticket business rules based on the attribute dependency relationship, and V represents the value corresponding to the key attribute set; Route analyzer module: used to receive and record the real-time calculation requests for airport taxes and fuel charges from various channels, extract route information, and obtain the real-time calculation results of airport taxes and fuel charges based on the route information; High-frequency route prediction module: used to build a high-frequency route prediction model based on a deep time series structure, and to train based on a high-frequency route model training data set. The data set obtains the real-time calculation request log data of airport taxes and fuel charges from each channel according to a set period, and extracts high-frequency route features in the log data, including flight frequency, sales popularity, and flight time, thereby forming a high-frequency route model training data set, and iteratively trains the high-frequency route prediction model based on the continuously updated high-frequency route model training data set; High-frequency route prediction cache module: It is used to perform real-time calculation based on the high-frequency route data predicted by the high-frequency route prediction module according to the KV structure cache data for real-time calculation rules, so as to determine the airport tax and fuel fee calculation results of each high-frequency route, and form the airport tax and fuel fee calculation results corresponding to the high-frequency route into KV structure cache data, where K represents the high-frequency route input parameter combination data, and V represents the airport tax and fuel fee calculation results corresponding to the high-frequency route input parameters.
7. The real-time calculation system for global airport taxes and fuel charges based on multi-level cache according to claim 6 is characterized in that: The process of the route analyzer module obtaining the real-time calculation results of airport tax and fuel charges based on route information is as follows: Based on the extracted airport tax and fuel fee real-time calculation request information, the route input parameter information is adapted to the KV structure cache data of the high-frequency route prediction result in the high-frequency route prediction cache data of the high-frequency route prediction cache module; if the adaptation is successful, the calculation result data in the KV structure cache data is directly fed back, otherwise: The extracted route input parameter information of the airport tax and fuel fee real-time calculation request information is adapted to the KV structure cache data for real-time calculation rules; if the adaptation is successful, the airport tax and fuel fee are calculated in real time based on the KV structure cache data for real-time calculation rules, and the calculation result is fed back; otherwise: The unadapted information is adapted to the KV structure cache data based on the original ticket business rules, and the adapted KV structure cache data based on the original ticket business rules is converted into KV structure cache data oriented to real-time calculation rules, and based on this, real-time calculation of airport taxes and fuel charges is performed, and the calculation results are fed back.
