Knowledge graph-driven intelligent railway line selection method, medium and equipment in karst areas

Abstract

The present invention relates to the technical field of railway intelligent line selection, and specifically to a knowledge graph-driven method for railway intelligent line selection in karst areas. A knowledge graph for line selection in karst areas is constructed according to the requirements of relevant design specifications for railway lines and relevant specifications for the disposal of karst areas; a risk-cost dual-objective line optimization model is constructed based on the knowledge graph for line selection in karst areas; a railway intelligent line selection algorithm is used to generate multiple alternative lines in a certain study area based on the knowledge graph query function; the dual-objective solution results of each alternative line are obtained respectively based on the risk-cost dual-objective line optimization model; the dual-objective solution results of each alternative line are ranked according to their preference based on multi-criteria tournament decision-making, and the alternative line corresponding to the optimal dual-objective solution result is the final recommended line. The present invention can take into account the influence of multi-dimensional environmental factors, meet multiple complex coupling constraints, and balance multi-evaluation objectives that are difficult to quantify and conflict with each other to obtain high-quality line plans.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent railway route selection, and in particular to a knowledge graph-driven intelligent railway route selection method in a karst area. Background Art

[0002] Railway route selection and design is a core task that takes into account the overall situation. It usually needs to consider the impact of multi-dimensional environmental factors, meet multiple complex coupling constraints, and balance multiple evaluation objectives that are difficult to quantify and conflict with each other. Especially under complex conditions, it is usually faced with the superposition effect of the above three factors, which is a recognized NP problem. In order to ensure the economy and safety of the entire life cycle of the railway, traditional manual route selection usually requires time-consuming and laborious iterative design work. However, due to limited manpower, time and resources, designers may miss some high-quality solutions.

[0003] At present, in order to make up for the shortcomings of manual design and ensure the safety of the entire life cycle of railway lines, some intelligent line selection methods have made outstanding contributions. Aiming at engineering economic goals, bionic evolutionary algorithms represented by genetics and particle swarms suitable for intelligent line selection in flat terrain, distance change algorithms suitable for intelligent line selection in complex and dangerous mountainous areas, and Monte Carlo tree search algorithms with global perception have been proposed by scholars for efficient and intelligent generation of line plans. Aiming at geological safety goals, researchers have established a hazard assessment model that combines typical geological disaster susceptibility represented by landslides and structural vulnerability, as well as a long-term low-frequency geological disaster risk quantitative assessment model represented by earthquakes.

[0004] However, as a common geological disaster in areas with developed water systems, karst development areas also pose a great threat to the construction and operation of railway lines, but they have rarely received attention from researchers in the field of intelligent line selection. Although line selection designers have accumulated rich experience in dealing with karst disasters, the relevant knowledge has not been systematically refined. In addition, railways crossing karst areas are also affected by a variety of other environmental factors, such as other geological disasters, environmental sensitive points, rivers, lakes, and existing structures. Therefore, the line design knowledge in karst areas has the characteristics of highly coupled environmental conditions and complex and multidimensional semantic relationships, making it difficult to directly apply line selection knowledge to intelligent line selection.

[0005] In summary, there is an urgent need for a knowledge graph-driven intelligent railway line selection method in karst areas to solve the problems existing in the existing technology. Summary of the invention

[0006] The present invention aims to provide a knowledge graph-driven intelligent railway line selection method in karst areas that can consider the influence of multidimensional environmental factors, meet multiple complex coupling constraints, and balance multivariate evaluation objectives that are difficult to quantify and conflict with each other. The specific technical scheme is as follows:

[0007] The present invention provides a knowledge graph-driven intelligent route selection method for karst area railways, comprising the following steps:

[0008] S1: Construct a knowledge graph for karst area line selection based on the relevant railway line design specifications and karst area treatment specifications;

[0009] S2: Construct a risk-cost dual-objective route optimization model based on the knowledge graph of route selection in karst areas; use the railway intelligent route selection algorithm based on the knowledge graph query function to generate multiple alternative routes in a certain study area;

[0010] S3: Obtain the dual-objective solution results of each alternative route based on the risk-cost dual-objective route optimization model;

[0011] S4: Based on the multi-criteria tournament decision, the dual-objective solution results of each alternative route are ranked in priority, and the alternative route corresponding to the optimal dual-objective solution result is the final recommended route.

[0012] Optionally, the S1 includes:

[0013] S1.1. Principles for selecting railway lines in karst areas are formulated based on the relevant design specifications for railway lines and the relevant specifications for karst area treatment;

[0014] S1.2. Based on the basic semantic logic of knowledge graph triples, a customized semantic relationship of the knowledge graph for railway line selection in karst areas is constructed. The basic semantic logic is: head entity → relationship → tail entity, and the customized semantic relationship is: environmental conditions → intersection with the line → treatment measures;

[0015] S1.3. Carry out the attribute design of environmental conditions and treatment measures entities respectively, and construct the knowledge ontology model of railway line selection in karst areas, where: the environmental condition attributes include risk value, and the treatment measures attributes include measure type, parameter index, threshold size, structure and priority level;

[0016] S1.4. Based on the matching relationship between environmental conditions, a knowledge graph for line selection in karst areas is established.

[0017] Optionally, constructing a risk-cost dual-objective line optimization model based on the karst area line selection knowledge graph in S2 specifically includes:

[0018] S2.1.1, based on the karst area railway line selection principle in step S1.1, extract the factors that pose the greatest threat to the railway engineering in the karst area, where the influencing factors include karst structure, surface depressions, underground rivers and other geological disasters;

[0019] S2.1.2. Based on the risk value attributes of environmental conditions in S1.3 and the influencing factors in S2.1.1, a risk assessment model for line selection in karst areas is constructed, wherein: the risk value attributes include syncline karst risk, anticline karst risk, surface karst depression risk and other disaster risk areas;

[0020] S2.1.3. Integrate the karst area line selection risk assessment model with the railway construction project economic assessment model to obtain a risk-cost dual-objective line optimization model.

[0021] Optionally, the karst area line selection risk assessment model is as follows:

[0022]

[0023] Among them: Min represents the minimum value of the calculated objective function; F hazard is the karst risk objective function; HPIs is the line plane, VPIs is the longitudinal section design vector; H SK is the syncline karst risk, I represents the total number of piles in the syncline karst area where the line passes; H AK is the anticline karst risk, J is the total number of piles in the anticline karst area where the line passes; H KD is the risk of surface karst depression, V is the total number of piles passing through the karst depression; H GH is the surface disaster risk, C is the total number of piles where the line passes through other disaster risk areas, and the total number of piles is the total number of mileage piles that the line passes;

[0024] The dual-objective optimization model is as follows:

[0025] Min:F(HPIs,VPIs)=[F cost (HPIs, VPIs), F hazard (HPIs, VPIs)] T ;

[0026] Among them, F cost is the engineering economic objective function; T is the transpose.

[0027] Optionally, the step of using the railway intelligent line selection algorithm based on the knowledge graph query function to generate multiple candidate lines for a certain research area in S2 specifically includes:

[0028] S2.2.1. Construct a comprehensive geographic information model of a study area based on the grid GIS model, and discretize the study area into a series of regular cell grids;

[0029] S2.2.2. Obtain the basic geographic information of each unit grid in the study area based on the comprehensive geographic information model;

[0030] S2.2.3. Use the railway intelligent route selection algorithm to generate the route plan for the study area;

[0031] S2.2.4. Input the basic geographic information of each unit grid corresponding to the area where the route plan passes into the karst area route selection knowledge map for query, obtain the environmental condition attributes and treatment measure attributes of the corresponding unit grid respectively, and store them back in the corresponding unit grid;

[0032] S2.2.5. Based on the treatment measure attributes of the unit grid, the line plan is constrained and checked. If the constraints are met, the line plan is an alternative line. Otherwise, the line plan is abandoned and the process returns to S2.2.3 until no new line plan is generated by the railway intelligent line selection algorithm and multiple alternative lines are obtained.

[0033] Optionally, the environmental condition attributes also include: level, length, width, height and azimuth; the risk value and parameter indicators are quantitative indicators; the constraint detection includes bridge and tunnel clearance, limit detection, plane avoidance distance and intersection angle detection.

[0034] Optionally, the dual-objective solution result includes a karst risk objective function value and an engineering economic objective function value;

[0035] The S4 includes:

[0036] S4.1. Select a non-dominated solution set based on the multiple alternative routes obtained in S2.2.5, and then compare the dual-objective solution results of each alternative route to obtain the karst risk success number and engineering economic success number of each alternative route. The specific formula is as follows:

[0037]

[0038]

[0039] Where: L is the total number of candidate routes in the non-dominated solution set; P A is the lth candidate route, l is a natural number from 1 to L; P X P is the non-dominated solution set A Another alternative route other than A ) is P A The objective function value, F(P X ) is P X The objective function value of t l To represent P A Whether it is a victory or not, victory is recorded as 1, and failure is recorded as 0; T l P A Relative to each P in the non-dominated solution set X The number of victories, that is, the average number of victories;

[0040] S4.2. Calculate the comprehensive number of wins for each candidate route based on the set dual-objective weights. The specific formula is as follows:

[0041]

[0042] Where: R T (P A ) is P A The total number of victories; W C and W H are the dual objective weights set, W C is the economic weight, W H is the risk weight; T lC (P A ) indicates P A The average number of victories of economic goals; T lH (P A ) indicates P A Average number of victories for karst risk objectives.

[0043] S4.3. Sort the alternative routes according to their comprehensive victory times from large to small, and the alternative route with the largest comprehensive victory times is the final recommended route.

[0044] The present invention also provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implements the knowledge graph-driven intelligent line selection method for karst area railways as described above.

[0045] The present invention also provides an electronic device, comprising: at least one processor, at least one memory and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, the knowledge graph-driven intelligent route selection method for karst area railways as described above is implemented.

[0046] The present invention provides a knowledge graph-driven intelligent line selection method for karst area railways. Firstly, the line selection experience and relevant normative provisions in the karst area are summarized, and the line selection principles in the karst area are sorted out. Based on this, a knowledge graph for line selection in the karst area is constructed to guide the construction of a line selection optimization model in the karst area and intelligent line search. Then, based on the knowledge graph for line selection in the karst area, a karst risk quantitative evaluation model is constructed, and a risk-cost dual-objective optimization model is jointly constructed with a mature engineering construction economic model for the comprehensive evaluation of line plans in the karst area. Finally, a comprehensive geographic information model is constructed, and the information flow relationship between the model and the knowledge graph is structured. The knowledge guidance of the intelligent search process is realized through knowledge graph retrieval. In addition, the contradictions and conflicts between the dual objectives are resolved through multi-criteria tournament decision-making based on the Pareto optimal principle, so that the influence of multi-dimensional environmental factors can be considered, multiple complex coupling constraints can be satisfied, and multi-evaluation objectives that are difficult to quantify and conflict with each other can be balanced to obtain high-quality line plans.

[0047] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0049] Figure 1 It is a schematic diagram of the process of the knowledge graph-driven intelligent route selection method for karst area railways in an embodiment of the present invention;

[0050] Figure 2 is a schematic diagram of the knowledge ontology model for railway line selection in karst areas in S1-3 in an embodiment of the present invention;

[0051] Figure 3 It is a schematic diagram of the application of the karst area line selection knowledge graph in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The embodiments of the present invention are described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0053] In one embodiment, if Figure 1 As shown in FIG. 1 , a knowledge graph-driven intelligent railway line selection method for karst areas includes the following steps:

[0054] S1: Construct a knowledge graph for karst area line selection based on the relevant railway line design specifications and karst area treatment specifications;

[0055] Specifically, according to the railway line design specifications (TB 10098-2017) Railway Line Design Specifications and karst area treatment requirements (Theory and Technology of Railway Disaster Reduction Route Selection in Complex and Dangerous Mountainous Areas), summarize the experience of route selection in artificial karst areas, construct a knowledge ontology, and establish a knowledge map for route selection in karst areas.

[0056] S1 includes:

[0057] S1.1. According to the relevant design specifications of railway lines and the relevant specifications for karst area treatment, the principles for railway line selection in karst areas are formed; specifically, the principles for railway line selection in karst areas are formed by summarizing, sorting out and refining the experience of manual line selection;

[0058] S1.2. Based on the basic semantic logic of knowledge graph triples, a customized semantic relationship of the knowledge graph for railway line selection in karst areas is constructed. The basic semantic logic is: head entity → relationship → tail entity, and the customized semantic relationship is: environmental conditions → intersection with the line → treatment measures;

[0059] S1.3. Carry out the attribute design of environmental conditions and treatment measures entities respectively, and construct the knowledge ontology model of railway line selection in karst areas, such as Figure 2 As shown, the environmental condition attributes include risk value, and the treatment measure attributes include measure type, parameter index, threshold value, structure and priority level;

[0060] Environmental condition attributes also include: level, length, width, height and azimuth; risk values ​​and parameter indicators are both quantitative indicators; risk values ​​and parameter indicators can also be dimensionless parameters.

[0061] In this embodiment, the environmental condition attributes include code, name, risk value, level, length, width, height and azimuth, where the code and name are used to build a one-to-one association relationship in the retrieval process, and the risk value is a customized attribute of the karst area line selection knowledge graph, which is a dimensionless value in the range of 0-10. The larger the value, the higher the risk. The remaining attributes are inherent attributes and objectively exist in the environmental conditions.

[0062] The treatment condition attributes include measure type, parameter index, threshold value, structure, and priority level. The measure type represents the type of treatment to be taken, such as detour, crossing, intersection, underpass, etc. The parameter index is the distance of detour, the clearance requirements for crossing and underpass, the intersection angle, etc. In this embodiment, the line needs to cross a karst depression with a minimum bridge height of 30m, so the parameter index is 30m; the threshold value indicates whether the parameter index is the maximum or minimum value requirement; the structure indicates which forms of roadbed, bridge, and tunnel can be used for the line; the priority level indicates the applicable scenario of the treatment measure, for example, the priority of the treatment measure applicable to general conditions should be higher than that of the measures applicable to difficult conditions.

[0063] S1.4. Based on the matching relationship between environmental conditions, a knowledge graph for line selection in karst areas is established.

[0064] Based on the above steps, a knowledge graph for karst area line selection containing 230 nodes and 480 relationships was constructed.

[0065] S2: Based on the knowledge graph of route selection in karst areas, a risk-cost dual-objective route optimization model is constructed; based on the knowledge graph query function, a railway intelligent route selection algorithm is used to generate multiple alternative routes in a certain study area, as follows: Figure 3 As shown;

[0066] S2.1. Constructing a risk-cost dual-objective route optimization model based on the knowledge graph of route selection in karst areas specifically includes:

[0067] S2.1.1, based on the karst area railway line selection principle in step S1.1, extract the factors that pose the greatest threat to the railway engineering in the karst area, where the influencing factors include karst structure, surface depression, underground river and other geological disasters (i.e., geological disasters other than karst structure, surface depression and underground river);

[0068] S2.1.2. Based on the risk value attributes of environmental conditions in S1.3 and the influencing factors in S2.1.1, a risk assessment model for line selection in karst areas is constructed, wherein: the risk value attributes include syncline karst risk, anticline karst risk, surface karst depression risk and other disaster risk areas;

[0069] The risk assessment model for line selection in karst areas is as follows:

[0070]

[0071] Among them: Min represents the minimum value of the calculated objective function; F hazard is the karst risk objective function; HPIs is the line plane, VPIs is the longitudinal section design vector; H SK is the syncline karst risk, I represents the total number of piles in the syncline karst area where the line passes; H AK is the anticline karst risk, J is the total number of piles in the anticline karst area where the line passes; H KD is the risk of surface karst depression, V is the total number of piles passing through the karst depression; H GH is the surface disaster risk, C is the total pile number of the line passing through other disaster risk areas, and the total pile number is the total number of mileage piles passed by the line; the pile number represents the mileage pile number of the line. The mileage piles are set at equal distances from the starting point along the center line of the line. The spacing can be set according to actual needs, usually 100m, 20m, etc.

[0072] S2.1.3. Integrate the karst area line selection risk assessment model with the railway construction project economic assessment model to obtain a risk-cost dual-objective line optimization model.

[0073] The dual-objective optimization model is as follows:

[0074] Min:F(HPIs,VPIs)=[F cost (HPIs, VPIs), F hazard (HPIs, VPIs)] T ;

[0075] Among them, F cost is the engineering economic objective function; T is the transposition;

[0076] The engineering economic objective function is as follows:

[0077] Min:F cost (HPIs, VPIs) = C B +C T +C E +C L +C R ;

[0078] Where: C B is the total cost of bridge construction along the entire line; C T is the total cost of tunnel construction along the entire line; C E is the total cost of earthwork construction; C L C is the engineering construction cost related to the line length; R The total cost of land acquisition for the entire line.

[0079] In this embodiment, the engineering economic evaluation objective function includes the economic costs of bridges, tunnels, length-related structures (such as tracks), roadbed projects, and land acquisition areas.

[0080] S2.2. Using the railway intelligent line selection algorithm based on the knowledge graph query function to generate multiple alternative routes in a certain study area specifically includes:

[0081] S2.2.1. Construct a comprehensive geographic information model of a study area based on the grid GIS model, and discretize the study area into a series of regular cell grids;

[0082] S2.2.2. Obtain basic geographic information of each unit grid in the study area based on the comprehensive geographic information model; the basic geographic information includes topography, landforms, geology, and land price information;

[0083] S2.2.3. Use the railway intelligent route selection algorithm to generate the route plan for the study area;

[0084] S2.2.4. Input the basic geographic information of each unit grid corresponding to the area where the line passes into the karst area line selection knowledge map for query, obtain the environmental condition attributes and treatment measure attributes of the corresponding unit grid respectively, and store them back in the corresponding unit grid;

[0085] The cell grid of the comprehensive geographic information model (CGIM) contains information on terrain, landforms, geology, and land prices. Based on the cell grid information in CGIM, the data of the cell grid through which the route plan passes is taken as input information, matched with the environmental condition attributes in the knowledge graph, and the risk value and treatment measures related attributes of the cell grid through which the route plan passes are retrieved and stored back in the current cell grid to facilitate the call of the route search process, so as to guide the subsequent objective function calculation and constraint detection process.

[0086] S2.2.5. Based on the treatment measures attributes of the unit grid, the line plan is constrained and checked. If the constraints are met, the line plan is an alternative line. Otherwise, the line plan is abandoned and the process returns to S2.2.3 until no new line plan is generated by the railway intelligent line selection algorithm. All the obtained plans are recorded as alternative line plans.

[0087] Constraint detection includes bridge and tunnel clearance, limit detection, plane avoidance distance and intersection angle detection.

[0088] S3: Obtain the dual-objective solution results of each alternative route based on the risk-cost dual-objective route optimization model;

[0089] S4: Based on the multi-criteria tournament decision, the dual-objective solution results of each alternative route are ranked in priority, and the alternative route corresponding to the optimal dual-objective solution result is the final recommended route.

[0090] The dual-objective solution result includes the fitness value of the karst risk objective function and the fitness value of the engineering economic objective function;

[0091] The S4 includes:

[0092] S4.1. Select a non-dominated solution set based on the multiple alternative routes obtained in S2.2.5, and then compare the dual-objective solution results of each alternative route to obtain the karst risk success number and engineering economic success number of each alternative route. The specific formula is as follows:

[0093]

[0094] Where: L is the total number of candidate routes in the non-dominated solution set; P A is the lth candidate route, l is a natural number from 1 to L; P X P is the non-dominated solution set A Another alternative route other than A ) is PA The objective function value, F(P X ) is P X The objective function value of t l To represent P A Whether it is a victory or not, victory is recorded as 1, and failure is recorded as 0; T l P A Relative to each P in the non-dominated solution set X The number of victories, that is, the average number of victories;

[0095] Specifically, for each P A The sum of the number of victories of the solution is divided by the total number of solutions other than itself. In this embodiment, L = 10, that is, there are 10 alternative routes in the non-dominated solution set, P A There are 10 options, and the 10 options will be used as P A Participate in the calculation, among which, the number of victories of the first alternative route compared with the other 9 options is 7, so T1 is 7 / 9. The number of victories of the 10 alternative routes is calculated separately: T1, T2, ..., T 10 .

[0096] S4.2. Calculate the comprehensive number of wins for each candidate route based on the set dual-objective weights. The specific formula is as follows:

[0097]

[0098] Where: R T (P A ) is P A The total number of victories; W C and W H are the dual objective weights set, W C is the economic weight, W H is the risk weight; T lC (P A ) indicates P A The average number of victories of economic goals; T lH (P A ) indicates P A Average number of victories for karst risk objectives.

[0099] S4.3. Sort the alternative routes according to their comprehensive victory times from large to small, and the alternative route with the largest comprehensive victory times is the final recommended route.

[0100] According to the application of this embodiment in a high-speed railway design section in a karst mountainous area in China, the comparison parameters of the calculated intelligent line selection scheme (machine selection scheme) and the manual line selection scheme designed by experienced engineers of the design institute are shown in Table 1.

[0101] Table 1 Comparison results of manual and machine selected solutions

[0102]

[0103] It can be seen that this method can generate an intelligent line selection scheme with better economic cost and karst risk than the manual design scheme, which verifies the effectiveness of the method of the present invention.

[0104] This embodiment also includes a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the above-mentioned knowledge graph-driven intelligent route selection method for karst railways is implemented:

[0105] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0106] This embodiment also includes an electronic device, including: at least one processor, at least one memory, and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, the knowledge graph-driven intelligent route selection method for karst area railways is implemented as described above.

[0107] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0108] The electronic device may be a computing device such as a mobile phone, a desktop computer, a notebook, a PDA, a cloud server, etc. The electronic device may include, but is not limited to, a processor and a memory. For example, the electronic device may also include an input / output device, a network access device, a bus, etc.

[0109] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0110] The memory can be used to store the computer program and / or module, and the processor implements the computer program by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0111] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A knowledge graph-driven intelligent railway line selection method in karst areas. It is characterized in that The steps include: S1: Construct a knowledge graph for karst area line selection based on the relevant railway line design specifications and karst area treatment specifications; S1 includes: S1.

1. Principles for selecting railway lines in karst areas are formulated based on the relevant design specifications for railway lines and the relevant specifications for karst area treatment; S1.

2. Based on the basic semantic logic of knowledge graph triples, a customized semantic relationship of the knowledge graph for railway line selection in karst areas is constructed. The basic semantic logic is: head entity → relationship → tail entity, and the customized semantic relationship is: environmental conditions → intersection with the line → treatment measures; S1.

3. Carry out the attribute design of environmental conditions and treatment measures entities respectively, and construct the knowledge ontology model of railway line selection in karst areas, where: the environmental condition attributes include risk value, and the treatment measures attributes include measure type, parameter index, threshold size, structure and priority level; S1.

4. Based on the matching relationship between environmental conditions, a knowledge graph for line selection in karst areas is established; S2: Constructing a risk-cost dual-objective route optimization model based on the karst area route selection knowledge graph; Using the railway intelligent route selection algorithm based on the knowledge graph query function to generate multiple candidate routes in a certain study area; Constructing a risk-cost dual-objective route optimization model based on the karst area route selection knowledge graph in S2 specifically includes: S2.1.1, based on the karst area railway line selection principle in step S1.1, extract the factors that pose the greatest threat to the railway engineering in the karst area, where the influencing factors include karst structure, surface depressions, underground rivers and other geological disasters; S2.1.

2. Based on the risk value attributes of environmental conditions in S1.3 and the influencing factors in S2.1.1, a risk assessment model for line selection in karst areas is constructed, wherein: the risk value attributes include syncline karst risk, anticline karst risk, surface karst depression risk and other disaster risk areas; S2.1.

3. Integrate the karst area route selection risk assessment model with the railway construction project economic assessment model to obtain a risk-cost dual-objective route optimization model; The risk assessment model for line selection in the karst area is as follows: Among them: Min represents the minimum value of the calculated objective function; F hazard is the karst risk objective function; HPIs is the line plane, VPIs is the longitudinal section design vector; H SK is the syncline karst risk, I represents the total number of piles in the syncline karst area where the line passes; H AK is the anticline karst risk, J is the total number of piles in the anticline karst area where the line passes; H KD is the risk of surface karst depression, V is the total number of piles passing through the karst depression; H GH is the surface disaster risk, C is the total number of piles where the line passes through other disaster risk areas, and the total number of piles is the total number of mileage piles that the line passes; The risk-cost dual-objective line optimization model is as follows: Min:F(HPIs,VPIs)=[F cost (HPIs,VPIs),F hazard (HPIs, VPIs)] T ; Among them, F cost is the engineering economic objective function; T is the transposition; The engineering economic objective function is as follows: Min:F cost (HPIs,VPIs)=C B +C T +C E +C L +C R ; Where: C B is the total cost of bridge construction along the entire line; C T is the total cost of tunnel construction along the entire line; C E is the total cost of earthwork construction; C L C is the engineering construction cost related to the line length; R The total cost of land acquisition for the entire line; S3: Obtain the dual-objective solution results of each alternative route based on the risk-cost dual-objective route optimization model; S4: Based on the multi-criteria tournament decision, the dual-objective solution results of each alternative route are ranked in priority, and the alternative route corresponding to the optimal dual-objective solution result is the final recommended route.

2. According to the knowledge graph-driven intelligent route selection method for karst railways in claim 1, It is characterized in that The S2 uses the railway intelligent line selection algorithm based on the knowledge graph query function to generate multiple candidate lines in a certain research area, specifically including: S2.2.

1. Construct a comprehensive geographic information model of a study area based on the grid GIS model, and discretize the study area into a series of regular cell grids; S2.2.

2. Obtain the basic geographic information of each unit grid in the study area based on the comprehensive geographic information model; S2.2.

3. Use the railway intelligent route selection algorithm to generate the route plan for the study area; S2.2.

4. Input the basic geographic information of each unit grid corresponding to the area where the route plan passes into the karst area route selection knowledge map for query, obtain the environmental condition attributes and treatment measure attributes of the corresponding unit grid respectively, and store them back in the corresponding unit grid; S2.2.

5. Based on the treatment measure attributes of the unit grid, the line plan is constrained and checked. If the constraints are met, the line plan is an alternative line. Otherwise, the line plan is abandoned and the process returns to S2.2.3 until no new line plan is generated by the railway intelligent line selection algorithm and multiple alternative lines are obtained.

3. According to the knowledge graph-driven intelligent route selection method for karst railways in claim 2, It is characterized in that The environmental condition attributes also include: level, length, width, height and azimuth; the risk value and parameter indicators are quantitative indicators; the constraint detection includes bridge and tunnel clearance, limit detection, plane avoidance distance and intersection angle detection.

4. According to claim 2, the knowledge graph-driven intelligent route selection method for karst railways, It is characterized in that The dual-objective solution results include a karst risk objective function value and an engineering economic objective function value; The S4 includes: S4.

1. Select a non-dominated solution set based on the multiple alternative routes obtained in S2.2.5, and then compare the dual-objective solution results of each alternative route to obtain the karst risk success number and engineering economic success number of each alternative route. The specific formula is as follows: Where: L is the total number of candidate routes in the non-dominated solution set; P A is the lth candidate route, l is a natural number from 1 to L; P X P is the non-dominated solution set A Another alternative route other than A ) is P A The objective function value, F(P X ) is P X The objective function value of t l To represent P A Whether it is a victory or not, victory is recorded as 1, and failure is recorded as 0; T l P A Relative to each P in the non-dominated solution set X The number of victories, that is, the average number of victories; S4.

2. Calculate the comprehensive number of wins for each candidate route based on the set dual-objective weights. The specific formula is as follows: Where: R T (P A ) is P A The total number of victories; W C and W H are the dual objective weights set, W C is the economic weight, W H is the risk weight; T lC (P A ) indicates P A The average number of victories of economic goals; T lH (P A ) indicates P A Average number of victories for karst risk objectives; S4.

3. Sort the alternative routes according to their comprehensive victory times from large to small, and the alternative route with the largest comprehensive victory times is the final recommended route.

5. A computer-readable storage medium, It is characterized in that Computer program instructions are stored thereon, and when the computer program instructions are executed by a processor, a knowledge graph-driven intelligent line selection method for karst area railways as described in any one of claims 1 to 4 is implemented.

6. An electronic device, It is characterized in that include: At least one processor, at least one memory, and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, the knowledge graph-driven intelligent route selection method for karst area railways as described in any one of claims 1 to 4 is provided.

Images