Railway route selection and demolition cost evaluation method based on big data model

Abstract

This invention belongs to the field of demolition cost assessment technology, specifically disclosing a method for assessing railway route selection and land acquisition costs based on a big data model. The method includes the following steps: constructing a graphical database and a core land acquisition database; acquiring policy-related land acquisition documents and dynamic data for the target area; creating an external database; identifying key factors affecting land acquisition costs; constructing a big data prediction model for land acquisition costs; training the model; and integrating it into a railway route selection design system. When inputting or modifying route plans in the route selection design system, the system calls the big data prediction model for land acquisition costs in real time. Based on the geometric parameters and spatial location of the route, it dynamically calculates and feeds back the corresponding land acquisition cost prediction results, adjusting the route plan until cost control requirements are met. This technical solution achieves dynamic and accurate assessment of land acquisition costs through the deep integration of multi-source heterogeneous databases and a big data prediction model.

Description

Technical Field

[0001] This invention belongs to the field of demolition cost assessment technology, and relates to a method for assessing railway route selection and demolition costs based on a big data model. Background Technology

[0002] In high-speed railway and conventional railway construction projects, land acquisition and demolition costs are a significant component of the total project investment, typically accounting for a high percentage. However, the composition of these costs is extremely complex, involving various factors such as land type (e.g., arable land, forest land, construction land), building structure (e.g., brick-concrete, brick-wood), ancillary facilities, local economic development level, population density, and constantly changing local policies and regulations.

[0003] In traditional railway alignment design, the calculation of land acquisition and demolition costs mainly relies on manual field surveys to collect data and estimates based on historical experience. This method has the following drawbacks:

[0004] Data collection channels are scattered, relying on obtaining data one by one from railway construction departments, local governments and archives departments, which is inefficient and prone to outdated calculation basis due to information lag;

[0005] Traditional estimation models struggle to handle complex relationships involving nonlinearity and multivariate coupling, resulting in insufficient computational accuracy and often failing to meet the requirements for precise control of project investment. Statistics show that in many railway projects, land acquisition and resettlement costs have generally exceeded the initial budget.

[0006] Existing computer-aided design software is mostly general-purpose tools, lacking dedicated modules for calculating land acquisition and demolition costs. When designers adjust route plans (such as modifying horizontal and vertical profiles), they cannot obtain real-time, dynamic feedback on changes in land acquisition and demolition costs, making it difficult to optimize route selection in a timely manner to avoid high-cost areas. Therefore, there is an urgent need for an intelligent method for calculating land acquisition and demolition costs that can integrate multi-source heterogeneous data, achieve dynamic and accurate prediction, and be deeply integrated with the route selection and design process. Summary of the Invention

[0007] The purpose of this invention is to address the aforementioned problems in existing technologies by proposing a method for evaluating railway route selection and land acquisition costs based on a big data model.

[0008] To achieve the above objectives, the basic solution of this invention is: a method for evaluating railway route selection and land acquisition costs based on a big data model, comprising the following steps:

[0009] Acquire geographical information along the railway line, railway design information, core land acquisition and demolition data, and historical land acquisition and demolition data, and preprocess them to build a graphic database and a core land acquisition and demolition database;

[0010] External databases are created by obtaining policy-related demolition and relocation documents and dynamic data for the target areas through external dynamic links.

[0011] Based on graph databases, core land acquisition and demolition databases, and external databases, key factors affecting land acquisition and demolition costs are identified, a big data prediction model for land acquisition and demolition costs is constructed, the model is trained, and then integrated into the railway alignment design system.

[0012] When inputting or modifying route plans in the route selection and design system, the system calls the big data prediction model for land acquisition and demolition costs in real time. Based on the geometric parameters and spatial location of the route, it dynamically calculates and feeds back the corresponding prediction results of land acquisition and demolition costs.

[0013] The route plan will be adjusted based on the predicted land acquisition and demolition costs until the cost control requirements are met.

[0014] The working principle and beneficial effects of this basic solution are as follows: This technical solution achieves dynamic and accurate assessment of land acquisition and demolition costs by deeply integrating multi-source heterogeneous databases with big data prediction models.

[0015] By integrating the evaluation model with the route selection and design system, the traditional fragmented process of "design-estimation-redesign" is broken down, enabling designers to obtain cost feedback in real time and optimize the plan. This effectively avoids the subjectivity and lag of manual estimation, significantly reduces the risk of exceeding the budget for land acquisition and demolition costs in railway projects, and greatly improves the intelligence level and investment control accuracy of railway route selection and design.

[0016] Furthermore, geographical information along the railway line, railway design information, core land acquisition and demolition data, and historical land acquisition and demolition data are acquired and preprocessed to construct a graphic database and a core land acquisition and demolition database, specifically:

[0017] The collected land area, land ownership, and building area data are cleaned to remove noisy, duplicate, and erroneous data. The data is then coded and normalized to convert different types of data into a format suitable for model analysis.

[0018] The land type and building structure type classification data are coded;

[0019] Normalize the numerical data of land area and building area to eliminate the influence of data units;

[0020] If the database of unit prices for demolition costs is empty, the data in the policy index will be used by default; if it is not empty, the cost information will be created according to the file.

[0021] The data composition of the unit price of demolition costs includes: during the design phase, slope line data is saved, and "land compensation fee information," "forest land and timber fee information," and "social security fee information" are also saved in the database;

[0022] The collected historical land acquisition and demolition data were cleaned to remove noise, duplicates and errors, and the categorical data and numerical data were coded and normalized respectively.

[0023] Establish a graphical database containing geographic information along the railway line and railway design parameters. The graphical database includes multiple objects used to form interactive route selection, parameter variables of the objects, and multiple parameter variables used to construct a numerical model system for predicting land acquisition and demolition costs and to establish triangular network data.

[0024] Establish an independent core database for land acquisition and demolition, which should include at least a land price sub-database, a housing replacement price sub-database, and a demolition compensation standard sub-database.

[0025] By performing operations such as "cleaning, encoding, and normalization," the quality of the input data to the model is ensured, eliminating model bias caused by inconsistent data formats and different units of measurement. Independent sub-databases (such as land prices and building replacement costs) are established and linked to policy documents, making the database highly maintainable and scalable. When local policies are adjusted, only the corresponding sub-database needs to be updated, without needing to reconstruct the entire model.

[0026] Furthermore, the steps for obtaining multiple objects from the graph database used to construct interactive line selection, the parameter variables of these objects, and multiple parameter variables used to construct a numerical model system for land acquisition and demolition cost prediction and to establish triangular network data are as follows:

[0027] Select the layer containing the entity to be collected and collect the entity. After all the entities are collected, optimize the data, including points, arcs, circles, lines, diagonal lines, text, layers, block names, text insertion points, ellipses, curves, ground lines, control points, graphic annotations, tables, elevation ranges, rectangles, and each object contains at least one parameter variable.

[0028] The ground line data format is "imported data mileage and elevation"; specify the ground line information table. If the specified file does not exist, it will be created automatically; if it exists, the data will be loaded and the original data will be sorted.

[0029] The generated files are sorted according to the group settings and marked on the design line;

[0030] The optimization scheme is set up so that duplicate points at the same location within 1 meter are considered duplicates, and the mileage information of the last input is retained.

[0031] The control point data format is set to "prefix import data elevation description", and the content is saved in the graphic database.

[0032] By setting rules such as "removing duplicate points within 1 meter" and "automatic sorting," data redundancy and errors caused by shaky hands or repeated data collection are avoided. Binding control point information with "prefix" and "description" ensures that the graphic data includes not only geometric lines but also semantic information (such as elevation attributes), providing accurate input for subsequent model understanding of geographical features.

[0033] Furthermore, the specific steps for creating an external database are as follows: This involves obtaining policy-related demolition and relocation documents and dynamic data for the target area through external dynamic links.

[0034] An interactive module is set up, which is connected to an external database. Local land acquisition and demolition compensation standards are selected, and the land acquisition area, demolition quantity, control points, and leveling points are saved to the external database.

[0035] Select cost estimation to calculate key land acquisition and demolition data in an external database;

[0036] Land pre-approval fee: On the selected slope section, the fee is calculated at 0.65 × (1 + 20%) RMB per kilometer of main line, starting from the previous slope change point. The input slope length cannot exceed the current slope section length. For independent large-scale station buildings, maintenance bases, EMU depots, and container center stations that are separately listed, the fee is calculated at 110 × (1 ± 10%) RMB per mu, based on the amount of land used in the project. The number of statistical points is increased based on the input demolition quantity and cost. Referring to the cleared projects, the fee is calculated at 247 RMB per mu.

[0037] Selecting the progress verification entity will save key land acquisition and demolition data to a database file. Users can input, modify, or delete design parameters in the external database through an interactive program.

[0038] The external database uses a custom entity with design information to store multiple design parameters input by the user. This custom entity, representing the interactive alignment selection, sets multiple design parameters, including names and values, specifically:

[0039] Store different categories of design parameters in different tables, convert single lines to double lines, and keep the curve elements of the right line the same as those of the left line after conversion.

[0040] To convert a polyline into a design line (single or double line), select the polyline to be converted and create a new design line input template. When generating a double line, the inner and outer curve elements differ by one line spacing.

[0041] Save the design lines as a data file. The data file is a text file. Single lines have the extension ".1xl" and double lines have the extension ".2xl". Load the data file from the planar data file into the drawing and convert it into design lines as needed.

[0042] When Gype=1, it is equivalent to the command "zx3", which converts a polyline to a design line;

[0043] When Gype=2, it is equivalent to the command "zx2", which converts a single line to a double line;

[0044] When Gype=3, a double line is converted into two single lines;

[0045] When Gype=4, the intersection points of multiple aligned edges are combined into a single line.

[0046] By providing built-in formulas for calculating specific expenses such as "land pre-approval fees" and "large station buildings," standardized estimation tools are offered, reducing errors caused by manual review and calculation. Based on the conversion of design lines (single / double lines) and file storage (e.g., .1xl, .2xl), structured storage and cross-platform reuse of design data are achieved, enabling external dynamic data to be accurately mapped to design entities.

[0047] Furthermore, based on graph databases, core land acquisition and demolition databases, and external databases, the key factors affecting land acquisition and demolition costs were identified. The specific steps are as follows:

[0048] Identify specific land acquisition and demolition projects, and clarify their geographical location, scale, and nature.

[0049] Set research objectives: Clearly define the scope of expropriation and demolition costs to be analyzed, including direct costs: land compensation fees and resettlement subsidies, indirect costs: demolition management fees and unforeseen expenses, and research objectives: cost drivers and prediction of expropriation and demolition costs;

[0050] Collect cost data for historical land acquisition and demolition projects, including specific amounts of fees, scope of acquisition and demolition, time periods, land and building information: understand the nature of land within the acquisition and demolition area: state-owned land, collective land, area, use, and the number, structure, area, and age of buildings; policy and regulatory documents: collect relevant policies, regulations, and compensation standards related to land acquisition and demolition, and understand the government's regulations and requirements for land acquisition and demolition compensation; market data: monitor local real estate market price trends, land transfer prices, and building material prices.

[0051] Preliminary screening of potential influencing factors includes:

[0052] Land factors: location, area, use, and ownership of the land;

[0053] Building factors: the building's area, structure, quality, and decoration;

[0054] Policy and regulatory factors: government-formulated policies on land acquisition and demolition compensation and resettlement;

[0055] Market factors: supply and demand in the real estate market, land prices, and building material prices;

[0056] The key factors were identified through correlation analysis using data analysis methods, specifically:

[0057] By using statistical analysis methods, the correlation coefficients between various factors and the cost of land acquisition and demolition are calculated, and the factors with a strong correlation to the cost of land acquisition and demolition are identified.

[0058] Regression analysis: Establish a regression model, take the cost of land acquisition and demolition as the dependent variable, and take the possible influencing factors as independent variables, and determine the degree of influence of each factor on the cost of land acquisition and demolition through regression analysis;

[0059] Sensitivity analysis: By changing the value of a certain factor, observe the changes in land acquisition and demolition costs to determine the sensitivity of that factor to land acquisition and demolition costs.

[0060] It is simple to operate and easy to use.

[0061] Furthermore, the big data prediction model for land acquisition and demolition costs adopts a neural network algorithm model, a random forest algorithm model, or a support vector machine algorithm model.

[0062] As needed, adopt machine learning algorithms that can handle nonlinear, high-dimensional, and complex problems for ease of use.

[0063] Furthermore, the steps for training the big data prediction model for land acquisition and demolition costs are as follows:

[0064] Select a preset graphic template, divide the historical land acquisition and demolition data into training set and test set, associate the parameter variables of the training set objects with the design parameters of the relevant test set, and use the training set data to train the big data prediction model for land acquisition and demolition costs.

[0065] Optimize model performance by adjusting model parameters: number of layers and number of nodes;

[0066] Insert test points and input short codes to improve input efficiency: 1 = "paddy field", 2 = "dry land", 3 = "vegetable garden"; input other text and record it accurately.

[0067] The trained big data prediction model for land acquisition and demolition costs was validated and evaluated using test set data. The model parameters were adjusted based on the evaluation results until the model achieved the preset prediction accuracy.

[0068] By dividing the training and testing sets and tuning hyperparameters, the model's generalization ability and prediction accuracy are ensured. The introduction of a "short code input" mechanism (e.g., 1 = paddy field) greatly improves the efficiency of designers during data annotation and input stages, and reduces the error rate of manual text input.

[0069] Furthermore, based on the positional relationships of objects within the graphic template, objects in the graphic template are loaded sequentially or in parallel from the graphic database. The steps are as follows:

[0070] S71, in the matched graphic template, establish the position and size correspondence between different objects in the order from left to right and from top to bottom, including annotation tangent, annotation feature point, intersection information, intersection number, intersection coordinates, curve deflection angle, curve radius, transition curve length, tangent length, and curve length;

[0071] Based on the relevant design parameters associated with the object's parameter variables, including design line width, mileage text height, font, number of decimal places, mileage display method, hundred-meter marker style, chain break control method, and the value of the starting intersection, update the positional relationship of each object in the graphic template, and obtain the relevant design line attribute information of each object in the graphic template after updating the positional relationship;

[0072] S72, taking the center of the image template as the origin, divides the interactive line selection graphic template into four quadrants—upper left, lower left, lower right, and upper right—using a cross, and displays the five major feature points of the intersection and the center of the circle;

[0073] If the distance between the center position of an object in each quadrant and the origin is less than the distance threshold, the color scheme is entered. The custom configuration is performed according to "pure white" first line color = "0", second line color = "0", transition curve color = "0", circular curve color = "0", tangent color = "0", mileage text color = "0", whole kilometer and broken chain text color = "0". Then, the objects in the graphic template are loaded sequentially from the graphic database to generate a big data prediction model for land acquisition and demolition costs.

[0074] If the distance between the center position of the object in each quadrant and the origin is less than the distance threshold, proceed to step S73; the distance threshold ranges from one-quarter to one-third of the diagonal length of the image template, and the image template is a rectangle or a square.

[0075] S73 uses the chaotic artificial fish swarm algorithm, GA-BP algorithm, GEP algorithm, BP neural network, and optimized RBF valuation model to cluster the surrounding position coordinates of all objects, obtain multiple cluster centers and the position coordinates of each cluster center, and assign each object to the cluster center that is closest to the center position of the object, thereby realizing the division of the graphic template area.

[0076] When creating a new design line, set the interactive prefix; the default prefix is ​​"K".

[0077] Each land acquisition and demolition scenario is represented by "planned progress | on-site data". A "data management method" is introduced, which is divided into the preferred implementation method of "on-site land acquisition" and the preferred implementation method of "house demolition". The algorithm is integrated to calculate the land acquisition and demolition costs.

[0078] For complex route design diagrams, the system can automatically determine the complexity of the graphic layout: if the layout is regular, it loads quickly in sequence; if the layout is complex, it introduces an intelligent clustering algorithm to divide the region, ensuring that the model can accurately capture key feature points under complex terrain.

[0079] Furthermore, when inputting or modifying the route plan in the route selection and design system, the track route is obtained and a new longitudinal profile is created. If the “.GJKOs” graphic template is selected, the route selection and design system will automatically execute the linkage of adding land acquisition and demolition objects.

[0080] When constructing a new land acquisition and demolition plan, the road alignment design system cuts a triangulation network according to the alignment position to obtain the initial ground points, including GU1, GU2, ..., GUN;

[0081] When the alignment changes, a statistical ledger is generated based on the statistical conditions. The statistical ledger is organized according to the administrative divisions at all levels. The statistical items include the total investment amount, the completed amount, and the quantity and price statistics of various land acquisition and demolition features. This indicates that the section has changed. The road alignment design system deletes points that are not on the alignment.

[0082] By selecting the statistical area, land use type, land use zone, demolition and relocation status, and start and end time, the demolition and relocation objects participating in the statistical calculation can be screened to improve the efficiency of the linkage and only cut out the changed locations.

[0083] When designing the plane, an AI model is used to map natural language to the 3D scene. When ground lines intersect, the ground points are recut according to the line positions to achieve intelligent interpretation of complex rules such as compensation standards and ownership relationships.

[0084] When horizontal and vertical design are coordinated, the multi-source policy, regulations and compensation standard data will not change, and the structured land acquisition and demolition knowledge graph will be reconstructed.

[0085] When the circuit is modified, the system no longer recalculates the entire circuit, but only "cuts out the changed parts", which greatly improves the calculation efficiency.

[0086] Furthermore, in the graphical template of the big data prediction model for land acquisition and demolition costs, when users modify the values ​​of some design parameters through the interactive module, the position and size of objects in the two-dimensional design drawing are adjusted through the position and size correspondence between different objects;

[0087] The location of the chain break does not change. When the starting point is moved, the location of the chain break will not change, but the mileage before the chain break will change.

[0088] When users perform operations such as "Calculation of House Demolition Unit Price or Fee", "Calculation of Unit Demolition Unit Price or Fee", and "Calculation of House Demolition Fee", the house demolition unit price may change due to the inability to use the weighted average comprehensive unit price based on line length. When "Modify Design Line Start Point", if the starting mileage is located after a part of the existing broken chain, the broken chain part will be deleted from the system. When moving the intersection point, if a short chain becomes a long chain, it violates the principle that "long chains must change the prefix", and a prompt will be generated for user interaction.

[0089] The line has a short chain, K0+130|K0+150. Change the starting mileage of the designed line from K0+0 to K0+50. When the chain break control method and mileage remain unchanged, the chain break location will be re-determined based on the mileage before the chain break.

[0090] When the chain break control method = position remains unchanged, the chain break becomes K0+180|K0+150, which violates the principle that the prefix of the long chain must be changed. When the prefix of the long chain remains unchanged, add a "#" sign before the prefix.

[0091] The position and size correspondence between different objects is expressed as follows: Chain break control method = position remains unchanged, move the intersection point. When there is a chain break on the straight line, move the intersection point and give a prompt.

[0092] The system intelligently identifies mileage jumps and repetitions, automatically adjusts the mileage attribution of land acquisition and demolition targets, ensures the continuity and accuracy of mileage calculation, and avoids errors in engineering quantity statistics caused by improper handling of chain breaks.

[0093] Furthermore, when a railway crosses a highway, river, reservoir, sea area, or passes through mountainous terrain, there are no breaks in the chain in the straight line, and no land acquisition quantity or cost will be counted.

[0094] If local governments at the county level or above have policy documents requiring compensation for the water surface, riverbank, beachhead, and sea area of ​​rivers, reservoirs, and sea areas, the compensation fee shall be calculated according to the compensation scope and unit price stipulated by the local government, and shall not be included in the land acquisition quantity. It can be moved. If the movement results in a broken chain on the curve, the broken chain will be automatically deleted.

[0095] If there is no stipulated fee for requisitioning construction land, it shall be calculated by referring to the unit price of dry land collective land in the same region. Compensation for ground attachments and greening shall be determined separately based on the survey data.

[0096] For large temporary stations, the land area is calculated based on the area outside the perimeter wall and drainage ditch. In the "Design Line Properties -> Advanced -> Clips -> Mileage Points" section, in addition to the site layout, the land area for railway branch lines is also considered. When the design line is selected, the broken chain will be displayed as a yellow clip.

[0097] The land use for railway branch lines is calculated based on the area outside the slope toe of the drainage ditch on both sides of the branch line. When modifying the mileage before the chain break, the previous mileage is used to determine the location of the chain break, and the input value is within the current mileage system. All chain breaks between the new location and the original location will be deleted. When modifying the mileage after the chain break, the newly input mileage after the chain break is within the current mileage system. If a chain break smaller than this mileage is deleted and is not within the current mileage system, then the route selection and design system will re-order the mileage.

[0098] When the route selection and design system automatically follows the mileage, it always processes the route based on the chain break control method, ensuring that the mileage and position remain unchanged.

[0099] When the demolition unit price adopts the transaction price, it includes the land use fee. The corresponding land acquisition quantity is included in the permanent land acquisition quantity, but the land acquisition compensation fee is no longer listed. The land acquisition unit price is set to zero, and the parameter variables of objects with the same name in the graphic template are linked together.

[0100] For special work sites such as "crossing rivers and roads" and "temporary projects", specific billing boundaries and rules (such as outside drainage ditches) have been clarified to avoid double counting of costs and make cost assessments more in line with actual financial expenditures. Attached Figure Description

[0101] Figure 1 This is a flowchart illustrating the railway route selection and land acquisition cost assessment method based on a big data model, as proposed in this invention. Detailed Implementation

[0102] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0103] This invention discloses a method for evaluating land acquisition and demolition costs for railway route selection based on a big data model. By integrating multi-source data and constructing a dedicated prediction model, it solves the problems of data lag and low calculation accuracy in traditional methods. By integrating the model with the design system and providing real-time feedback, it achieves integrated linkage of "design-cost-optimization," transforming route selection from a purely engineering-driven approach to a dual-driven approach of "technology + economy," fundamentally reducing the risk of project cost overruns.

[0104] like Figure 1 As shown, the railway route selection and land acquisition cost assessment method based on big data models includes the following steps:

[0105] Acquire geographical information along the railway line, railway design information, core land acquisition and demolition data, and historical land acquisition and demolition data, and preprocess them to build a graphic database and a core land acquisition and demolition database;

[0106] Geographic information refers to information related to spatial geographic distribution. It reflects the characteristics and relationships of natural and human elements at a specific location on the Earth's surface, including:

[0107] Natural geographic information includes topography and landforms, such as altitude, slope, and aspect data for mountains, plains, plateaus, and basins. For example, in the construction of railways in mountainous areas, where the terrain is highly undulating, slope and aspect data are crucial for route selection and the design of bridges and tunnels.

[0108] Geological conditions include rock type, geological structure, and seismic intensity. Geological information can help assess the stability of railway construction areas and avoid building railways in areas prone to geological hazards.

[0109] Statistical data on meteorological elements such as temperature, precipitation, wind speed, and wind direction. In railway design, the impact of extreme weather conditions on railway facilities must be considered; for example, high temperatures, heavy rain, and strong winds can lead to problems such as rail deformation and roadbed collapse.

[0110] Information on the distribution, water level, flow rate, and water quality of water bodies such as rivers, lakes, and oceans. When a railway crosses a river, the height, span, and foundation type of the bridge need to be designed based on hydrological information.

[0111] Railway design information refers to various technical data and information involved in the early planning and design processes of railway construction, mainly including:

[0112] Route design information: Route alignment: The specific direction and geographical location of the railway line, usually represented by the coordinates of the centerline. The route alignment needs to comprehensively consider factors such as topography, geology, urban planning, and economic development to ensure the rationality and economy of the route.

[0113] Curve radius: The degree of curvature of a curve on a railway line. The size of the curve radius directly affects the train's operating speed and safety. During the design process, the curve radius needs to be selected appropriately based on the railway's grade and design speed.

[0114] Gradient: The longitudinal gradient of the track. The gradient affects the train's traction power and operating energy consumption. To ensure the normal operation of the train, the gradient needs to be set reasonably based on the train's traction performance and track conditions.

[0115] Track structure information: Track type, such as conventional track, seamless track, etc. Seamless track has advantages such as reducing train vibration and noise and improving running smoothness, and is widely used in high-speed railways.

[0116] Railway sleeper types include wooden sleepers and reinforced concrete sleepers. Different types of sleepers have different characteristics and applicable ranges, and the selection needs to be based on the actual conditions of the railway line.

[0117] Ballast track types: ballasted track and ballastless track. Ballasted track has advantages such as good elasticity and low cost, but requires regular maintenance; ballastless track has advantages such as good stability and less maintenance workload, but is more expensive.

[0118] Bridge design includes the type of bridge (e.g., beam bridge, arch bridge, cable-stayed bridge), span, height, foundation type, etc. Bridge design must consider the topography, geology, hydrology, and other conditions of the bridge's location to ensure its safety and durability.

[0119] Tunnel design: The length, cross-sectional shape, and support methods of the tunnel. Tunnel design needs to take into account factors such as geological conditions, groundwater conditions, ventilation and lighting to ensure the safety of tunnel construction and operation.

[0120] Station layout design information includes the station's location, size, number of platforms, and track arrangement. The station layout needs to consider passenger and freight transport demands, as well as its connection to urban transportation.

[0121] Station building design: This includes the station's floor area, functional zoning, and architectural style. The station design must meet the needs of passengers for waiting, ticketing, and transfers, while also reflecting local cultural characteristics.

[0122] Core data on land acquisition and resettlement refers to the fundamental data that plays a crucial role in the land acquisition and resettlement work of railway construction projects. It mainly includes:

[0123] Land expropriation data, land ownership: clearly defines the ownership and use rights of the expropriated land, including the distinction between state-owned land and collectively owned land, as well as information on land owners and users.

[0124] Land area: Accurately measure the area of ​​the land to be expropriated, including the area of ​​different types of land such as cultivated land, forest land, and construction land.

[0125] Land use: Understand the original use of the land to be expropriated, such as agricultural land, industrial land, commercial land, etc., in order to determine a reasonable compensation standard.

[0126] Housing demolition data: Property rights: Determine the ownership of the demolished houses, including information on relevant documents such as the house ownership certificate and land use certificate.

[0127] House area: The building area and usable area of ​​the demolished house are measured as the basis for calculating demolition compensation.

[0128] Building Structure and Quality: Understand the building structure (such as brick-concrete structure, frame structure, etc.) and quality condition of the building to assess its value.

[0129] Data on above-ground attachments, including types and quantities of above-ground attachments such as trees, seedlings, wells, and utility poles.

[0130] Value of attachments: The value of attachments is assessed based on their type, quantity, and market price, serving as a reference for compensation.

[0131] Compensation and resettlement data, compensation standards: Formulate compensation standards for land acquisition, house demolition and above-ground attachments, and clarify the methods and amounts of compensation.

[0132] Resettlement method: Determine the resettlement method for those whose land is being expropriated, such as monetary compensation, property exchange, or relocation to another location. Resettlement location: The planned construction site of the resettlement housing or the location of the resettlement housing provided.

[0133] Historical land acquisition and demolition data refers to relevant data accumulated during previous railway construction or other project land acquisition and demolition processes, mainly including:

[0134] Land Acquisition and Demolition Project Information: Project Name: Record the specific name of each land acquisition and demolition project for easy querying and management.

[0135] Project Location: Clearly define the geographical location of the land acquisition and demolition project, including administrative divisions and specific addresses.

[0136] Project Timeline: Record the start and end times of the land acquisition and demolition project to understand the cycle of the work.

[0137] Scope and scale of land acquisition and demolition: The total area of ​​land acquired for each project and the area of ​​different types of land are counted.

[0138] Number and area of ​​houses to be demolished: Record the number and total area of ​​houses to be demolished, as well as the area of ​​houses with different structures and uses.

[0139] Compensation and resettlement details: The total amount of compensation for each project, as well as the amounts for different types of compensation (such as land compensation, housing compensation, and compensation for attachments).

[0140] Resettlement methods and effects: Analyze the implementation and effects of different resettlement methods, such as the construction quality of resettlement housing and resident satisfaction.

[0141] Problems and experiences encountered during the land acquisition and demolition process: Record various problems encountered during the land acquisition and demolition process, such as residents' resistance, compensation disputes, legal disputes, etc.

[0142] External databases are created by obtaining policy-related demolition and relocation documents and dynamic data for the target areas through external dynamic links.

[0143] Based on graph databases, core land acquisition and demolition databases, and external databases, key factors affecting land acquisition and demolition costs are identified, a big data prediction model for land acquisition and demolition costs is constructed, the model is trained, and then integrated into the railway alignment design system.

[0144] When inputting or modifying route plans in the route selection and design system, the system calls upon the land acquisition and demolition cost big data prediction model in real time. Based on the route's geometric parameters and spatial location, it dynamically calculates and feeds back the corresponding land acquisition and demolition cost prediction results. Preferably, the land acquisition and demolition cost big data prediction model employs a neural network algorithm model, a random forest algorithm model, or a support vector machine algorithm model. Specifically:

[0145] Regularly or in real-time, update both basic and dynamic data. For example, update market data monthly and promptly update relevant information when policies and regulations change.

[0146] Land compensation costs are calculated based on the nature and area of ​​the land and current land compensation standards. For example, state-owned land may be compensated according to market appraisal prices, while collectively owned land is compensated according to a multiple of the local unified annual output value or the comprehensive land price of the area.

[0147] Building compensation costs: The replacement cost method can be used, which calculates the cost of demolishing the building and rebuilding a building of the same standard based on current building material prices and labor costs. Adjustments should also be made to account for the building's depreciation.

[0148] Resettlement costs: Calculated based on population size and resettlement method (monetary compensation or property exchange). Monetary compensation is typically provided on a per capita basis; property exchange requires consideration of the construction or purchase cost of the resettlement housing. Other costs: Include relocation subsidies, temporary resettlement subsidies, and business interruption losses (for commercial premises), calculated according to relevant policy standards.

[0149] Visual Presentation: The predicted costs of land acquisition and resettlement are displayed in intuitive charts (such as bar charts and line graphs) or reports for easy viewing and understanding by relevant personnel. Real-time Feedback: The predicted results are fed back to decision-makers, implementers, and those affected by the land acquisition and resettlement project in real time through information systems or platforms. For example, those affected can check their predicted costs anytime via a mobile application.

[0150] Dynamic adjustment: Adjust the forecast results in a timely manner according to changes in the actual situation, such as the progress of land acquisition and demolition, market fluctuations, etc., and then provide feedback to relevant personnel.

[0151] The route plan will be adjusted based on the predicted land acquisition and demolition costs until cost control requirements are met, specifically as follows:

[0152] Collect multi-source data including predicted land acquisition and resettlement costs, geographical information related to the route (such as topography, landforms, and geological conditions), traffic demand (such as population distribution and travel volume), and environmental factors (such as ecological protection zones and water sources). Perform data mining on historical route construction cases to analyze land acquisition and resettlement costs, construction costs, and operational benefits under different route options. Utilize data analysis techniques to identify the correlation between land acquisition and resettlement costs and factors such as route alignment, length, and the areas traversed. For example, establish regression models to analyze the relationship between route length and land acquisition and resettlement costs, determining the quantitative impact between the two.

[0153] Intelligent algorithms are introduced to adjust the route. Ant colony optimization abstracts the route planning problem into the problem of an ant colony finding the optimal path. Each ant, during its search, chooses its direction based on pheromone concentrations such as land acquisition costs and terrain difficulty. The pheromone update rule is related to land acquisition costs; paths with lower costs receive increased pheromone levels, guiding more ants to choose those paths. After multiple iterations, the ant colony gradually finds a relatively optimal route. Particle swarm optimization initializes a swarm of particles, each representing a route plan. Particles move in the solution space, adjusting their speed and position based on their own historical best position and the group's historical best position. Using land acquisition costs as the fitness function, particle positions are iteratively updated until a route plan that meets cost control requirements is found.

[0154] By combining geographic information with spatial analysis functions, the feasibility of different route options is assessed. For example, it analyzes whether the route avoids areas with high land acquisition and demolition costs and whether it conforms to topographical requirements. Through buffer zone analysis, the scope of land acquisition and demolition and the potential impact area within a certain range around the route are determined. Visualization displays the route options and land acquisition and demolition cost predictions on a map. The distribution of land acquisition and demolition costs for different route options is presented intuitively, facilitating further analysis and adjustments.

[0155] A dynamic model is established to simulate the construction and operation of the railway line, considering multiple factors such as land acquisition and resettlement costs, construction progress, and operational efficiency. The model simulates cost changes for different route options at different time points. A multi-objective evaluation is conducted, comprehensively considering multiple objectives beyond land acquisition and resettlement costs, including construction costs, operational efficiency, and environmental impact. Methods such as the analytic hierarchy process (AHP) are used to determine the weights of each objective, enabling a comprehensive evaluation of the route options.

[0156] Establish a feedback mechanism to feed back land acquisition and demolition costs and other relevant data from the actual construction process to the adjustment system. Based on the feedback, optimize algorithm parameters and models to improve the accuracy of subsequent route adjustments. Utilize machine learning and deep learning technologies to learn from and analyze large amounts of route construction data. Continuously improve route adjustment strategies to adapt to different project needs and changing environments.

[0157] In a preferred embodiment of the present invention, geographical information along the railway line, railway design information, core land acquisition and demolition data, and historical land acquisition and demolition data are acquired, preprocessed, and used to construct a graphic database and a core land acquisition and demolition database, specifically as follows:

[0158] The collected land area, land ownership, and building area data are cleaned to remove noisy, duplicate, and erroneous data. The data is then coded and normalized to convert different types of data into a format suitable for model analysis.

[0159] The land type and building structure type classification data are coded, and the land area and building area numerical data are normalized to eliminate the influence of data units. If the demolition cost unit price database is empty, the data in the policy index is used by default; if it is not empty, the cost information is created according to the file.

[0160] The data composition of the unit price of demolition costs includes: during the design phase, slope line data is saved, and "land compensation fee information," "forest land and timber fee information," and "social security fee information" are also saved in the database;

[0161] The collected historical land acquisition and demolition data were cleaned to remove noise, duplicates and errors, and the categorical data and numerical data were coded and normalized respectively.

[0162] Establish a graphical database containing geographic information along the railway line and railway design parameters. The graphical database includes multiple objects used to form interactive route selection, parameter variables of the objects, and multiple parameter variables used to construct a numerical model system for predicting land acquisition and demolition costs and to establish triangular network data.

[0163] The objects and parameter variables used to construct interactive line selection:

[0164] Target Route: The core target for route selection, representing the planned transportation routes, such as railways and highways. It consists of a series of points and lines, connecting the starting point, the ending point, and key intermediate nodes.

[0165] Control points are important locations that a route must pass through or avoid, possessing specific geographical coordinates and attributes. For example, important cities, industrial areas, and tourist attractions are control points that need to be connected; while nature reserves, military restricted areas, etc., are control points that need to be avoided.

[0166] Topography and landforms: This includes natural landform features such as mountains, rivers, plains, and valleys, as well as man-made landforms such as buildings, bridges, and tunnels. Topography and landforms can affect the route alignment and construction difficulty.

[0167] Transportation network: Existing transportation lines such as railways, highways, and waterways. The planning of new lines needs to consider the connection and coordination with the existing transportation network in order to achieve interconnectivity.

[0168] Parameters and Variables: Line Parameters: Length: The total length of the line, affecting construction costs and operational efficiency. Curvature: Reflects the degree of curvature of the line; excessive curvature may limit train or vehicle speed. Gradient: The vertical inclination of the line; a steep gradient increases construction difficulty and operational energy consumption. Control Point Parameters: Coordinates: Precisely represent the geographical location of the control points. Importance Weight: Assigns different weights to control points based on their importance, prioritizing important control points during route selection. Topographic Parameters: Elevation: The altitude of the terrain, affecting the design of uphill and downhill sections of the line. Gradient and Aspect: The degree and direction of the terrain's inclination, significantly impacting the alignment and stability of the line.

[0169] Obstacle types: such as mountains, rivers, buildings, etc. Different types of obstacles require different crossing or avoidance methods. Traffic network parameters: Distance between the new route and the existing traffic network, affecting the ease of connection. Connection methods: such as at-grade crossings, grade-separated crossings, etc. Different connection methods have different impacts on construction costs and traffic flow.

[0170] The parameters used to construct the numerical model system for predicting land acquisition and demolition costs are:

[0171] Land-related parameters: Land area: The total area of ​​the expropriated area, which is the basis for calculating land compensation. Land type: Such as agricultural land, construction land, unused land, etc. Different types of land have different compensation standards. Land grade: A grade classified according to factors such as geographical location and quality of the land. The higher the grade, the higher the compensation standard usually is.

[0172] Building-related parameters: Building area: The total area of ​​the buildings to be demolished.

[0173] Building structure: Such as brick-concrete structure, frame structure, etc., different structures have different replacement costs. Construction age: Affects the building's depreciation rate and compensation value. Use: Divided into residential, commercial, industrial, etc., different uses have different compensation standards. Population-related parameters: Registered population size: Involved in the calculation of population resettlement costs.

[0174] Number of households: This is an important reference for information such as housing resettlement.

[0175] Policy and regulatory parameters: Land compensation standards: Compensation unit prices for different types of land stipulated by the government. Building compensation coefficient: Compensation adjustment coefficients determined based on factors such as the structure and age of the building. Resettlement policy parameters: Such as per capita resettlement area, resettlement subsidy standards, etc.

[0176] The parameter plane coordinates of the points used to establish triangular network data: determine the position and elevation of each point in the triangular network on the plane.

[0177] Establish an independent core database for land acquisition and demolition, which should include at least a land price sub-database, a housing replacement price sub-database, and a demolition compensation standard sub-database.

[0178] In a preferred embodiment of the present invention, the steps of obtaining multiple objects in a graphical database used to constitute interactive line selection, the parameter variables of the objects, and multiple parameter variables used to construct a numerical model system for predicting land acquisition and demolition costs and to establish triangular network data are as follows:

[0179] Select the layer containing the entity to be collected. The default is to collect polylines, with one point collected every 10 meters. After all data is collected, optimize the data, including points, arcs, circles, lines, diagonal lines, text, layers, block names, text insertion points, ellipses, curves, ground lines, control points, graphic annotations, tables, elevation ranges, and rectangles. Each object (each object refers to optimized data, including points, arcs, circles, lines, diagonal lines, text, layers, block names, text insertion points, ellipses, curves, ground lines, control points, graphic annotations, tables, elevation ranges, and rectangles) contains at least one parameter variable. For example, during the operation, points can be collected out of order; the system will automatically sort them.

[0180] The ground line data format is "imported data mileage and elevation", and the land acquisition and demolition data is imported into the database;

[0181] Specify the ground line information table. If the specified file does not exist, it will be created automatically; if it exists, the data will be loaded and sorted.

[0182] The generated files are sorted according to the group settings and marked on the design line;

[0183] The optimization scheme is set up so that duplicate points at the same location within 1 meter are considered duplicates, and the mileage information of the last input is retained.

[0184] The control point data format is set to "prefix import data elevation description", and the content is saved in the graphic database.

[0185] When entering descriptions, you can input simplified codes, such as code 1 for paddy fields, code 2 for dry land, and code 3 for vegetable gardens. These simplified codes are saved in the image database, and are only converted into corresponding text when referenced.

[0186] In a preferred embodiment of the present invention, the specific steps for creating an external database by obtaining policy-related demolition and relocation documents and dynamic data of the target area through external dynamic links are as follows:

[0187] An interactive module is set up, which is linked to an external database. The local land acquisition and demolition compensation standard is selected, and the land acquisition area, demolition quantity, control points, and leveling points are saved to the external database. The relocation location of the three utilities (electricity, water, and electricity) is determined by the previous mileage. For example, when the starting point is moved, the relocation location of the three utilities also moves accordingly. The relocation location of the three utilities is determined by the mileage.

[0188] Select cost estimation to accurately calculate key land acquisition and resettlement data in an external database, specifically:

[0189] In land acquisition and demolition projects, cost estimation involves pre-estimating and calculating various costs involved in the process. These costs typically encompass land acquisition compensation, building demolition compensation, resettlement costs, and compensation for above-ground attachments, among others. "Choosing a cost estimation method" means selecting the appropriate approach from different estimation methods, models, or plans to conduct the cost estimation work.

[0190] For example, there may be different methods for estimating land expropriation compensation, such as the market comparison approach and the income capitalization approach. The most appropriate estimation method needs to be selected based on the specific circumstances of the expropriation and demolition project.

[0191] Key data for land acquisition and demolition: This refers to data that plays a crucial role in estimating land acquisition and demolition costs and managing the entire project. This data includes, but is not limited to, the area and nature of the land to be acquired or demolished (e.g., agricultural land, construction land), the building area, structural type, and condition of the buildings, as well as the number of people involved and the resettlement methods.

[0192] Based on relevant policies, regulations, market conditions, and other factors, we accurately calculate key data related to land acquisition and demolition. For example, we accurately calculate land acquisition compensation based on local land acquisition compensation standards and the area of ​​land to be acquired; and we calculate building demolition compensation based on the building's replacement cost and depreciation rate.

[0193] External database: This is a data storage and management system independent of the current land acquisition and demolition project calculation system. Storing the calculation results of key land acquisition and demolition data in an external database can, on the one hand, achieve centralized management and sharing of data, making it convenient for different departments and personnel to query and use it when needed; on the other hand, it also helps with the long-term preservation and analysis of data, providing reference and guidance for subsequent land acquisition and demolition projects.

[0194] For example, in the land acquisition and demolition work of a large-scale infrastructure construction project in a city, the staff first selects an appropriate cost estimation method to estimate the land acquisition and demolition costs, and then accurately calculates key data such as land area, building value, and resettlement costs, and stores them in the external database of the city construction management department for subsequent project audits, statistical analysis, and reference for similar projects.

[0195] Land pre-approval fee: On the selected slope section, the fee is calculated at 0.65 × (1 + 20%) RMB per kilometer of main line, starting from the previous slope change point. The input slope length cannot exceed the current slope section length. For independent large-scale station buildings, maintenance bases, EMU depots, and container center stations that are separately listed, the fee is calculated at 110 × (1 ± 10%) RMB per mu, based on the amount of land used in the project. The number of statistical points is increased based on the input demolition quantity and cost. Referring to the cleared projects, the fee is calculated at 247 RMB per mu.

[0196] Selecting the progress verification entity will save key land acquisition and demolition data to a database file. Users can input, modify, or delete design parameters in the external database through an interactive program.

[0197] The external database uses a custom entity with design information to store multiple design parameters input by the user. This custom entity, representing the interactive alignment selection, sets multiple design parameters, including names and values, specifically:

[0198] Store different categories of design parameters in different tables, convert single lines to double lines, and keep the curve elements of the right line the same as those of the left line after conversion.

[0199] To convert a polyline into a design line (single or double line), select the polyline to be converted and create a new design line input template. When generating a double line, the inner and outer curve elements differ by one line spacing.

[0200] Save the design lines as a data file. The data file is a text file. Single lines have the extension ".1xl" and double lines have the extension ".2xl". Load the data file from the planar data file into the drawing and convert it into design lines as needed.

[0201] When Gype=1, it is equivalent to the command "zx3", which converts a polyline to a design line;

[0202] When Gype=2, it is equivalent to the command "zx2", which converts a single line to a double line;

[0203] When Gype=3, a double line is converted into two single lines;

[0204] When Gype=4, the intersection points of multiple aligned edges are combined into a single line.

[0205] The entity includes tunnel segments. The process of laying tunnel segments in sequence by calling the segments in the graphic database along the railway line includes: starting with the lower tunnel segments, arranging the segments alternately on the left and right sides, then assembling adjacent segments, and finally arranging wedge-shaped segments. Based on the mileage of the connecting passage, the connecting passages for the left and right lines are created, thus completing the creation of the three-dimensional model of the high-speed railway.

[0206] In a preferred embodiment of the present invention, key factors affecting land acquisition and demolition costs are determined based on a graph database, a core land acquisition and demolition database, and an external database. The specific steps are as follows:

[0207] Identify specific land acquisition and demolition projects, and clarify their basic information such as geographical location, scale, and nature; for example, whether it is land acquisition and demolition for commercial development in the city center or land acquisition and demolition for infrastructure construction in the suburbs.

[0208] Set research objectives: Clarify the scope of land acquisition and demolition costs to be analyzed, such as direct costs (land compensation fees, resettlement subsidies, etc.), indirect costs (demolition management fees, unforeseen expenses, etc.), and the research objectives to be achieved, such as identifying the main cost drivers and predicting land acquisition and demolition costs.

[0209] Collect relevant historical demolition and relocation data: Collect cost data from similar past demolition and relocation projects, including the specific amounts of various expenses, the scope of demolition and relocation, and the time period. This data can be obtained from government departments, demolition and relocation agencies, or relevant project archives. Land and building information: Understand the nature (e.g., state-owned land, collective land), area, and use of the land within the demolition and relocation area, as well as the number, structure, area, and age of buildings. This information can be obtained through land management departments, real estate registration departments, etc. Policy and regulatory documents: Collect relevant policies, regulations, compensation standards, and other documents related to demolition and relocation to understand the government's regulations and requirements regarding demolition and relocation compensation. Market data: Monitor local real estate market price trends, land transfer prices, building material prices, and other information, as these market factors can affect demolition and relocation costs.

[0210] Preliminary screening of potential influencing factors: Land factors: The location, area, use, and ownership of land all affect the cost of land acquisition and demolition. For example, land located in the core area of ​​a city usually has higher acquisition and demolition costs; compensation standards for commercial land may be higher than those for industrial or residential land.

[0211] Building factors: The area, structure, quality, and decoration of a building are important factors affecting the cost of land acquisition and demolition. Generally speaking, the larger the building area, the more complex the structure, and the higher the decoration standard, the higher the cost of land acquisition and demolition.

[0212] Policy and regulatory factors: Government-formulated policies on land acquisition and demolition compensation and resettlement directly affect land acquisition and demolition costs. Policies and regulations may differ across regions and time periods, leading to variations in land acquisition and demolition costs.

[0213] Market factors: Market factors such as supply and demand in the real estate market, land prices, and building material prices can affect the cost of land acquisition and demolition. For example, during periods of real estate market prosperity, land and housing prices rise, and the cost of land acquisition and demolition will increase accordingly.

[0214] Social factors: Social circumstances during the land acquisition and demolition process, residents' demands and attitudes, and other social factors may also affect the cost of land acquisition and demolition.

[0215] Data analysis methods are used to determine the correlation between key factors: Statistical analysis is employed to calculate the correlation coefficients between each factor and the cost of land acquisition and demolition, identifying factors with a strong correlation to these costs. For example, the Pearson correlation coefficient is used to measure the linear relationship between factors such as land area and building area and the cost of land acquisition and demolition.

[0216] Choose an appropriate statistical analysis method to calculate the correlation coefficient:

[0217] The Pearson correlation coefficient is used to analyze the linear correlation between two continuous variables. For example, it can be used to analyze the correlation between land acquisition and demolition costs and continuous variables such as land area and building area.

[0218] Spearman's correlation coefficient is used when variables do not follow a normal distribution or are ordinal variables. For example, building structure type can be considered an ordinal variable (brick-concrete structure, frame structure, etc.), and Spearman's correlation coefficient can be used to analyze its correlation with land acquisition and demolition costs.

[0219] The closer the absolute value of the correlation coefficient is to 1, the stronger the correlation; the closer it is to 0, the weaker the correlation. Common criteria are as follows: Strong correlation: The absolute value of the correlation coefficient is greater than or equal to 0.8.

[0220] Strong correlation: The absolute value of the correlation coefficient is between 0.6 and 0.8.

[0221] Moderate correlation: The absolute value of the correlation coefficient is between 0.3 and 0.6.

[0222] Weak correlation: The absolute value of the correlation coefficient is less than 0.3.

[0223] By calculating the correlation coefficients between various factors and demolition costs, factors with absolute correlation coefficients between 0.6 and 0.8 are identified as those with a strong correlation to demolition costs. For example, calculations show that the correlation coefficient between land area and demolition costs is 0.75, and the correlation coefficient between building area and demolition costs is 0.7. Therefore, land area and building area are factors with a strong correlation to demolition costs.

[0224] Regression Analysis: A regression model is established, with land acquisition and demolition costs as the dependent variable and potential influencing factors as independent variables. Regression analysis determines the degree of influence of each factor on land acquisition and demolition costs. For example, a multiple linear regression model can be used to analyze the combined impact of land factors, building factors, and policy and regulatory factors on land acquisition and demolition costs. Specific identification of potential influencing factors is required.

[0225] After establishing a regression model to determine the extent of each factor's influence on land acquisition and demolition costs, the effectiveness of the regression analysis and the impact of each factor are evaluated from multiple perspectives:

[0226] Goodness of fit R 2 R represents the degree to which the regression model explains the changes in the dependent variable, and its value ranges from 0 to 1. 2 The closer the value is to 1, the better the model fits the data, meaning the independent variables explain a higher proportion of the variation in the dependent variable. The number of independent variables in the model is considered to avoid excessive R-values ​​due to too many independent variables. 2 The price is artificially inflated. Adjust R. 2 A value greater than 0.7 indicates a good model fit, but the specific criteria need to be determined based on the actual problem and data characteristics.

[0227] For example, in the analysis of land acquisition and demolition costs, if R 2 =0.8, indicating that the model can explain 80% of the changes in land acquisition and demolition costs. The F-test is used to test the significance of the entire regression model, that is, to determine whether all independent variables as a whole have a significant impact on the dependent variable.

[0228] The null hypothesis is that the regression coefficients of all independent variables are 0. If the null hypothesis is rejected, it means that the model is significant. By checking the p-value of the F-test, if the p-value is less than the set significance level (usually 0.05), the null hypothesis is rejected, and the model is considered to be significant overall, that is, at least one independent variable has a significant effect on the dependent variable.

[0229] The coefficients of independent variables are used to assess the impact of regression on the dependent variable. Regression coefficients represent the average change in the dependent variable when the independent variable changes by one unit, assuming other independent variables remain constant. A positive coefficient indicates a positive correlation between the independent and dependent variables, while a negative coefficient indicates a negative correlation. The magnitude and sign of the coefficient determine the direction and extent of the independent variable's influence on the dependent variable. For example, in a land acquisition and demolition cost regression model, if the regression coefficient for land area is 500, it means that, assuming other factors remain constant, for every 1 square meter increase in land area, the average cost of land acquisition and demolition increases by 500 yuan.

[0230] The t-test is used to test whether the regression coefficient of a single independent variable is significantly different from zero, that is, to determine whether the independent variable has a significant effect on the dependent variable. The null hypothesis is that the regression coefficient of the independent variable is 0. Check the p-value of the t-test; if the p-value is less than the set significance level (usually 0.05), the null hypothesis is rejected, and the independent variable is considered to have a significant effect on the dependent variable. For example, the p-value of the t-test for building area is 0.02, which is less than 0.05, indicating that building area has a significant impact on land acquisition and demolition costs. The residual is the difference between the observed value and the model's predicted value. The residual should follow a normal distribution, which is one of the basic assumptions of regression analysis.

[0231] The approximate normality of the residuals can be determined by plotting histograms or QQ plots. If the residuals do not follow a normal distribution, it may affect the reliability and effectiveness of the model. The residuals should be independent of each other and free from autocorrelation. Autocorrelation can lead to inaccurate estimation of the model's standard error, thus affecting the reliability of the parameter estimates.

[0232] The Durbin-Watson test can be used to determine if autocorrelation exists in the residuals. The Durbin-Watson statistic ranges from 0 to 4, with values ​​close to 2 indicating no autocorrelation. Outliers are observations that deviate significantly from other data points and may have a substantial impact on the regression model. Outliers can be identified by plotting scatter plots, box plots, etc. For outliers, further analysis is needed to determine their cause—whether it's a data entry error or a special case. Impact points are observations that significantly influence the parameter estimation and prediction results of the regression model. Impact points can be identified using statistics such as Cook's distance. The larger the Cook's distance, the greater the impact of the observation on the model. Impact points require careful handling and may need to be considered for removal or special processing.

[0233] Sensitivity analysis: By changing the value of a certain factor, the changes in land acquisition and demolition costs are observed to determine the sensitivity of that factor to the costs. For example, analyzing the impact of land price fluctuations on land acquisition and demolition costs can help identify factors that have a significant impact on these costs.

[0234] The key factors affecting land acquisition and demolition costs are multifaceted, and can be analyzed from several dimensions, including land-related factors, building-related factors, policies and regulations, market environment, and social factors.

[0235] Land Location: Land located in prime urban centers, bustling commercial districts, and transportation hubs typically incurs significantly higher expropriation and demolition costs due to its high economic value and development potential compared to urban fringe or remote areas. For example, the expropriation and demolition costs in the city center of a first-tier city can be several times, or even dozens of times, higher than in the suburbs. Land Area and Shape: Generally, the larger the expropriated land area, the higher the land compensation and resettlement subsidies required. Furthermore, irregularly shaped land may present development challenges, increasing additional planning and construction costs, thus affecting expropriation and demolition costs. Land Use: Expropriation and demolition compensation standards vary considerably depending on the land use. Commercial land is generally more valuable than industrial and residential land, resulting in relatively higher expropriation and demolition costs. For instance, the compensation for converting industrial land into a commercial center will increase due to the change in land use. Land Ownership: Expropriation and demolition policies and compensation standards differ between state-owned and collectively owned land. Expropriation and demolition of collectively owned land involves the interests of rural collective economic organizations and farmers. In addition to land compensation, resettlement of farmers must be considered, making the process more complex and potentially more costly.

[0236] Building area and number of floors: The larger the building area and the more floors, the higher the compensation fees and relocation subsidies will be during demolition and relocation. For example, the demolition and relocation costs of a large commercial complex are much higher than those of an ordinary residential building. Building structure and quality: Different building structures, such as steel structures, reinforced concrete structures, and brick-concrete structures, have different construction costs, and therefore different demolition and relocation compensation. Buildings with better quality and lower maintenance costs will receive relatively higher demolition and relocation compensation.

[0237] Building Age and Depreciation: Newer buildings typically have advantages in functionality, facilities, and appearance, resulting in higher compensation for demolition and relocation. Older buildings may suffer from depreciation and damage, leading to lower compensation. Decoration Condition: The decoration standards and style of a building also affect demolition and relocation costs. Luxury-decorated houses require additional compensation for the decoration in addition to the value of the house itself.

[0238] Compensation Policies for Land Acquisition and Demolition: The compensation standards set by the government are a direct factor affecting land acquisition and demolition costs. Compensation policies may vary across different regions and at different times, including different calculation methods and standards for land compensation fees, resettlement subsidies, and compensation for attachments and crops on the land. Resettlement Policies: Resettlement methods (such as monetary compensation and property exchange) and resettlement conditions have a significant impact on land acquisition and demolition costs. Monetary compensation requires paying a considerable amount of resettlement fees to those whose land is being acquired; property exchange requires providing suitable resettlement housing, involving the construction or purchase costs of the housing. Land Transfer Policies: Policies such as the method and price of land transfer affect developers' land acquisition costs, which in turn indirectly affect land acquisition and demolition costs. If the land transfer price is high, developers may try to control costs in the land acquisition and demolition process to ensure the project's profit margin, but they may also raise compensation standards to accelerate project progress due to intense market competition.

[0239] Real Estate Market Conditions: The supply and demand relationship and price trends in the local real estate market directly affect the cost of land acquisition and demolition. During periods of real estate market prosperity, rising housing prices increase the value of land and buildings, leading to higher compensation standards for land acquisition and demolition; conversely, during market downturns, the cost of land acquisition and demolition may be relatively lower. Building Material and Labor Prices: The land acquisition and demolition process involves the demolition of buildings and the construction of new resettlement housing. Fluctuations in the prices of building materials and labor affect the cost of land acquisition and demolition. For example, rising prices of major building materials such as steel and cement will increase the construction cost of resettlement housing, thereby increasing the total cost of land acquisition and demolition.

[0240] Residents' demands and attitudes: The demands and attitudes of residents facing demolition and relocation have a significant impact on the cost of such projects. Social and media oversight: Social and media oversight can also influence demolition and relocation projects.

[0241] In a preferred embodiment of the present invention, the step of training the big data prediction model for land acquisition and demolition costs is as follows:

[0242] Select a preset graphic template, divide the historical land acquisition and demolition data into training set and test set, associate the parameter variables of the training set objects with the design parameters of the relevant test set, and use the training set data to train the big data prediction model for land acquisition and demolition costs.

[0243] In the scenario of predicting land acquisition and demolition costs, the preset graphical template can be understood as a structured and standardized form of data presentation and organization, which can assist in the processing and analysis of historical land acquisition and demolition data:

[0244] Based on maps, the system combines geographical information of the land acquisition and demolition area with land acquisition and demolition data. The map can mark information such as the acquisition and demolition scope, land type, and building distribution of different plots. When dividing the training and test sets, it can be done according to geographical regions, such as allocating land acquisition and demolition data from different administrative districts and streets to the training and test sets respectively. Simultaneously, map templates can visually display the distribution of land acquisition and demolition data in different areas, helping analysts better understand the spatial characteristics of the data. This provides a geospatial basis for subsequently associating the parameter variables of the training set objects (such as the land area and building density of a certain area) with the design parameters of the test set (such as the geographical factors to be considered when predicting the future land acquisition and demolition costs of the area). 3D terrain template description: For land acquisition and demolition projects involving complex terrain, 3D terrain templates can more accurately present the topography of the acquisition and demolition area. It can display terrain information such as mountains, rivers, and slopes, as well as the relative positional relationship between buildings and the terrain.

[0245] Bar chart templates are used to compare land acquisition and demolition data across different categories or time periods. For example, bar charts can be used to display the area of ​​land acquisition and demolition in different years, the number of different types of buildings acquired and demolished, etc. Functions: When dividing the training and test sets, observing the distribution characteristics of the bar chart allows for segmentation based on data size, trends, and other patterns. When associating parameters, the category information (such as building type) shown in the bar chart can be linked to design parameters such as land acquisition and demolition costs to analyze the impact of different categories on costs.

[0246] Line chart templates are primarily used to display the changing trends of land acquisition and demolition data over time. For example, changes in land acquisition and demolition costs and land prices over years can be clearly presented using line charts. Functions: When partitioning data, the data can be divided into training and test sets based on the fluctuation cycle and trend inflection points of the line chart. When associating parameters, the characteristics of data changes in the time series can be used as parameter variables to correlate with design parameters for predicting future land acquisition and demolition costs, helping the model capture the dynamic patterns of data change.

[0247] The flowchart template for land acquisition and demolition projects displays the entire process from start to finish in flowchart form, including various stages such as project initiation, assessment, compensation, and demolition. Its function is to categorize and organize the data and parameters involved in each stage when dividing data and associating parameters. For example, different stages may have different cost expenditures and influencing factors. The flowchart template clearly identifies these stages and associates the relevant parameter variables (such as assessment fees and standards in the assessment stage) with the design parameters of the test set (such as the stage factors considered when predicting the overall cost of the land acquisition and demolition project), enabling the model to take into account the procedural characteristics of land acquisition and demolition projects.

[0248] By adjusting the model's parameters—number of layers and number of nodes—the performance of the model can be optimized. Specifically, the basic principle for adjustment is to balance model complexity and generalization ability: increasing the number of layers and nodes increases model complexity, enabling it to learn more complex patterns, but may lead to overfitting, meaning the model performs well on training data but poorly on new data; reducing the number of layers and nodes may cause underfitting, preventing the model from learning the key features of the data. Therefore, it is necessary to find a balance between complexity and generalization ability.

[0249] Data feature matching: The complexity and number of features of the data will affect the appropriate number of layers and nodes. If the data features are simple, too many layers and nodes may make the model too complex; if the data features are complex, too few layers and nodes will not be able to fully learn the data features.

[0250] Manually adjust the number of layers gradually: Start with a simple model, such as a neural network with only one hidden layer. Add one layer at a time and observe the performance change on the validation set. For example, first train the model using a single hidden layer and record metrics such as accuracy and mean squared error on the validation set; then add another hidden layer, train again, and evaluate, comparing the performance. If performance improves, consider continuing to increase the number of layers; if performance deteriorates, reduce the number of layers. Adjusting the number of nodes: A similar gradual adjustment method can be used for the number of nodes in each layer. First, set a relatively small number of nodes, such as 10, train the model, and evaluate its performance. Then gradually increase the number of nodes, such as adding 10 nodes at a time, and observe the performance change. When performance no longer improves or even deteriorates, stop increasing the number of nodes.

[0251] Grid search determines the parameter range: Based on experience and problem complexity, determine the possible ranges for the number of layers and nodes. For example, the number of layers can be set to 1-5, and the number of nodes per layer can be set to 10-100. Parameter combinations are generated: Different values ​​for the number of layers and nodes are combined to form a parameter grid. For example, if there are 5 possible values ​​for the number of layers and 10 possible values ​​for the number of nodes per layer, 50 different parameter combinations will be generated. Each combination is evaluated: For each parameter combination, the model is trained using the training set, and its performance is evaluated on the validation set. The parameter combination with the best performance is selected as the final model parameters.

[0252] Random search defines the parameter distribution: This specifies the probability distribution for the number of layers and nodes. For example, the number of layers can be uniformly and randomly selected from 1 to 10, and the number of nodes per layer can be uniformly and randomly selected from 10 to 200. Random sampling: This involves randomly sampling a certain number of parameter combinations from the defined parameter distribution. For example, randomly sampling 20 parameter combinations.

[0253] Evaluation and selection: Train and evaluate each parameter combination, and select the parameter combination with the best performance.

[0254] Use Hyperopt, an automated hyperparameter tuning tool: a Python library for hyperparameter optimization that can automatically search for optimal parameters by defining a parameter space and an objective function.

[0255] Optuna: Also a powerful hyperparameter optimization library, it features efficient search algorithms and an intuitive API, helping to quickly find the optimal combination of layers and nodes. Important considerations during tuning: Use of a validation set: During parameter tuning, use a validation set to evaluate model performance, not a test set. The test set should only be used for the final model evaluation to ensure the objectivity of the results. Multiple trials: Due to the inherent randomness of model training, it is recommended to perform multiple training and evaluations for each parameter combination, taking the average performance as the evaluation result for that parameter combination. Recording and analysis: Record the parameters adjusted each time and the corresponding model performance, analyzing the trend of performance changes to better understand the impact of the number of layers and nodes on model performance.

[0256] Insert test points and input short codes to improve input efficiency: 1 = "paddy field", 2 = "dry land", 3 = "vegetable garden"; input other text and record it accurately.

[0257] The trained big data prediction model for land acquisition and demolition costs was validated and evaluated using test set data. The model parameters were adjusted based on the evaluation results until the model achieved the preset prediction accuracy.

[0258] The big data model comprises multiple objects, such as initial support, firewalls, and ballast lining. Each object has at least one associated design parameter, and each object corresponds to an engineering quantum list. The engineering quantum list includes multiple materials and their quantities. Materials preferably include, but are not limited to, anchor rods, steel mesh, and reinforcing bars. A correspondence is established between the relevant design parameter values ​​of each object and the quantities of materials in the corresponding engineering quantum list. This correspondence is preferably, but not limited to, a linear or matrix mapping, and can be established based on prior data. When new design parameters are changed or re-entered, the correspondence between the relevant design parameter values ​​of each object and the quantities of materials in the corresponding engineering quantum list is used to obtain the engineering quantum lists of all objects and to calculate the total bill of quantities for the project. This allows for a rapid understanding of the composition of the big data model and the quantities of materials used.

[0259] In a preferred embodiment of the present invention, to improve the speed and efficiency of generating a big data model for the calculation method of railway alignment and land acquisition costs, the objects in the graphic template are loaded serially or in parallel from the graphic database according to the positional relationships of each object in the graphic template. The steps are as follows:

[0260] S71, in the matched graphic template, establish the position and size correspondence between different objects in the order from left to right and from top to bottom, including annotation tangent, annotation feature point, intersection information, intersection number, intersection coordinates, curve deflection angle, curve radius, transition curve length, tangent length, and curve length;

[0261] Based on the relevant design parameters associated with the object's parameter variables, including design line width, mileage text height, font, number of decimal places, mileage display method, hundred-meter marker style, chain break control method, and the value of the starting intersection, update the positional relationship of each object in the graphic template, and obtain the relevant design line attribute information of each object in the graphic template after updating the positional relationship;

[0262] S72, taking the center of the image template as the origin, divides the interactive line selection graphic template into four quadrants—upper left, lower left, lower right, and upper right—using a cross, and displays the five major feature points of the intersection and the center of the circle;

[0263] If the distance between the center position of an object in each quadrant and the origin is less than the distance threshold, the color scheme is entered. The custom configuration is performed according to "pure white" first line color = "0", second line color = "0", transition curve color = "0", circular curve color = "0", tangent color = "0", mileage text color = "0", whole kilometer and broken chain text color = "0". Then, the objects in the graphic template are loaded sequentially from the graphic database to generate a big data prediction model for land acquisition and demolition costs.

[0264] If the distance between the center position of the object in each quadrant and the origin is less than the distance threshold, proceed to step S73; the distance threshold ranges from one-quarter to one-third of the diagonal length of the image template, and the image template is a rectangle or a square.

[0265] S73 uses the chaotic artificial fish swarm algorithm, GA-BP algorithm, GEP algorithm, BP neural network, and optimized RBF valuation model to cluster the surrounding position coordinates of all objects, obtain multiple cluster centers and the position coordinates of each cluster center, and assign each object to the cluster center that is closest to the center position of the object, thereby realizing the division of the graphic template area.

[0266] When creating a new design line, set the interactive prefix; the default prefix is ​​"K".

[0267] Each land acquisition and demolition scenario is represented by "Planned Progress | On-site Data," and a "data management method" is introduced, divided into a preferred implementation method for "on-site land acquisition" and a preferred implementation method for "house demolition." An algorithm is used to calculate land acquisition and demolition costs, specifically:

[0268] Information Acquisition for Project Progress: Clearly define the planned arrangements for the land acquisition and demolition project at different stages, such as the timeline for land acquisition, the specific steps for house demolition, and the estimated completion time. This information can be obtained from the project's planning documents and schedules. Data Processing: Organize the planned progress in chronological order, marking each key milestone, such as the start date of land acquisition, the completion date of land surveying, and the commencement date of house demolition.

[0269] On-site data for land acquisition includes the area of ​​the land to be acquired, land type (e.g., arable land, forest land, construction land), location, and surrounding environment. This data can be obtained through on-site measurements and land registration documents. Housing data records the building area, building structure (e.g., brick-concrete structure, frame structure), usage status (owner-occupied, rented), and condition of the demolished houses. This information can be collected through on-site surveys and property registration documents.

[0270] The preferred implementation method under the "Data Management Approach" is "On-site Land Acquisition." This involves determining compensation standards based on local land acquisition policies and market conditions. For example, the compensation price for cultivated land may include land compensation fees, resettlement subsidies, and crop compensation fees, requiring comprehensive consideration of local regulations and actual circumstances. It also considers influencing factors such as land flatness and the presence of underground pipelines, which may increase acquisition costs and must be taken into account when calculating expenses. The preferred implementation method for "House Demolition" is house appraisal. This involves using professional appraisal methods to assess the value of the houses to be demolished. Common appraisal methods include the market comparison method and the cost method. The appraised value of the house is determined based on factors such as its building area, structure, and condition. Resettlement costs are considered, including the resettlement methods for the affected residents, such as monetary compensation or property exchange. If monetary compensation is used, resettlement costs need to be calculated; if property exchange is used, the price difference of the replacement house needs to be considered.

[0271] The algorithm for calculating land acquisition and demolition costs uses a weighted summation method: Weights are assigned to various costs based on different data management methods and influencing factors. For example, land area may have a higher weight for land acquisition costs, while the assessed value of the house may have a higher weight for house demolition costs. Each cost is multiplied by its corresponding weight and then summed to obtain the total land acquisition and demolition cost. Machine learning algorithms can also be used: Regression algorithms (such as linear regression and decision tree regression) can be used to build a predictive model for land acquisition and demolition costs. Using project progress and on-site data as input features and land acquisition and demolition costs as output variables, the model is trained to fit the data and predict costs. Specific calculation steps include: Land acquisition cost calculation: For each piece of land to be acquired, land compensation fees are calculated based on land type and compensation standards.

[0272] For example, the land compensation fee for cultivated land = cultivated land area × cultivated land compensation unit price. Other influencing factors, such as the cost of leveling due to uneven land surface, should be considered and added to the land compensation fee to obtain the total land acquisition cost for that plot. The total land acquisition cost is obtained by adding up the total land acquisition costs for all acquired land.

[0273] The demolition compensation is determined based on the assessed value of the house. For example, demolition compensation = assessed value of the house × building area. Resettlement costs are calculated, such as monetary compensation or the price difference for property exchange. The demolition compensation and resettlement costs are added together to obtain the total demolition cost. Total cost calculation: The total land acquisition cost and the total demolition cost are added together to obtain the total land acquisition and demolition cost for the entire project. Example code (using Python to implement a simple weighted summation method) Python land acquisition data:

[0274] land_areas = [1000, 2000, 1500], land area (square meters).

[0275] land_types = ['arable land', 'forest land', 'construction land'],

[0276] land_compensation_prices = {'cultivated land': 500, 'forest land': 300, 'construction land': 800}, land compensation unit price (yuan / square meter).

[0277] land_other_costs = [10000, 5000, 8000], Other influencing factors costs (yuan) for housing data.

[0278] house_areas = [120, 150, 180], where the building area is in square meters.

[0279] house_evaluation_values ​​= [5000, 6000, 7000], the appraised value of the house (yuan / square meter).

[0280] resettlement_costs = [30000, 40000, 50000], resettlement costs (yuan).

[0281] Calculate land acquisition costs

[0282] total_land_cost = 0

[0283] for i in range(len(land_areas)):

[0284] land_cost = land_areas[i] * land_compensation_prices[land_types[i]] + land_other_costs[i]

[0285] total_land_cost += land_cost

[0286] Calculate house demolition costs

[0287] total_house_cost = 0

[0288] for i in range(len(house_areas)):

[0289] house_cost = house_areas[i] * house_evaluation_values[i] +resettlement_costs[i]

[0290] total_house_cost += house_cost

[0291] Calculate the total cost

[0292] total_cost = total_land_cost + total_house_cost

[0293] print("Total land acquisition cost:", total_land_cost, "yuan")

[0294] print("Total house demolition cost:", total_house_cost, "yuan")

[0295] print("Total cost of land acquisition and demolition:", total_cost, "yuan")

[0296] By following the steps and methods above, land acquisition and demolition costs can be calculated using a fusion algorithm based on "planned progress | on-site data" and different data management methods. In practical applications, the algorithm and parameters need to be adjusted and optimized according to specific circumstances.

[0297] In a preferred embodiment of the present invention, when inputting or modifying a route plan in the route selection and design system, the track route is obtained and a new longitudinal profile is created. If the “.GJKOs” graphic template is selected, the route selection and design system automatically executes the linkage of adding land acquisition and demolition objects.

[0298] When constructing a new land acquisition and demolition plan, the road alignment design system cuts a triangulation network according to the alignment position to obtain the initial ground points, including GU1, GU2, ..., GUN;

[0299] When the alignment changes, a statistical ledger is generated based on the statistical conditions. The statistical ledger is organized according to the administrative divisions at all levels. The statistical items include the total investment amount, the completed amount, and the quantity and price statistics of various land acquisition and demolition features. This indicates that the section has changed. The road selection and design system deletes points that are not on the alignment, such as GU10~GU20.

[0300] By selecting the statistical area, land use type, land use zone, demolition and relocation status, and start and end time, the demolition and relocation objects participating in the statistical calculation can be screened to improve the efficiency of the linkage and only cut out the changed locations.

[0301] When designing the floor plan, an AI model is used to map natural language to the 3D scene. When ground lines intersect, the ground points are re-cut according to the line positions, enabling intelligent interpretation of complex rules regarding compensation standards and ownership relationships. Specifically:

[0302] The process can be divided into several stages: data preparation, AI model building, ground line intersection processing, and intelligent interpretation of complex rules. First, prepare 3D scene data and rules described in natural language. Then, build an AI model that can map natural language to the 3D scene. Next, handle the ground line intersection problem. Finally, intelligently interpret the compensation standards and ownership rules.

[0303] 3D scene data: Collect 3D information such as terrain data, building models, and ground lines of the design area, and store it in the form of point cloud data and 3D mesh models. It can be obtained through methods such as laser scanning and UAV photogrammetry.

[0304] Natural Language Data: This involves organizing natural language texts containing rules regarding compensation standards and ownership relationships, such as relevant laws and regulations, policy documents, and project descriptions. These texts are then annotated to highlight key information, such as the compensation recipients, the method of calculating compensation amounts, and the definition of ownership scope.

[0305] Model selection: Deep learning models, such as those based on the Transformer architecture, can be used. Transformers possess powerful natural language processing and feature extraction capabilities, enabling them to handle long sequences of text and are well-suited for processing complex natural language rules. Convolutional Neural Networks (CNNs) can be combined to process 3D scene data, as CNNs excel at processing images and 3D data.

[0306] Training data: The prepared 3D scene data and natural language data are paired to form a training dataset. For example, natural language text describing the compensation standard of a certain area is associated with the corresponding 3D scene data.

[0307] Model training: The model is trained using a training dataset. By adjusting the model's parameters, it learns the mapping relationship between natural language and 3D scenes. During training, an appropriate loss function, such as cross-entropy loss, is used to measure the difference between the model's predictions and the actual results.

[0308] Ground line intersection handling: Intersection detection: Geometric algorithms are used to analyze ground lines and determine if intersections exist. Ground lines can be represented as line segments, and intersections are detected by calculating the intersection points between these segments. In terms of code implementation, Python's Shapely library can be used, which provides convenient geometric operation functions. Example code is as follows:

[0309] Python

[0310] from shapely.geometry import LineString

[0311] line1 = LineString([(0, 0), (1, 1)])

[0312] line2 = LineString([(0, 1), (1, 0)])

[0313] if line1.intersects(line2):

[0314] print("ground lines intersect")

[0315] ```

[0316] Ground points are re-cut according to line positions: When ground line intersections are detected, a cutting scheme is determined based on the intersection positions. Ground point cloud data in a 3D scene can be segmented according to intersecting lines to obtain different regions. Point cloud processing libraries such as PCL (Point Cloud Library) can be used to implement point cloud cutting operations.

[0317] Intelligent Interpretation of Complex Rules: Rule Extraction: A trained AI model analyzes compensation standards and ownership rules described in natural language to extract key information. For example, it extracts information such as compensation amounts and conditions corresponding to different ownership types from the text. Rule Application: The extracted rules are applied to the re-segmented ground point data. Based on the region and ownership information of each ground point, the compensation standard corresponding to each point is determined. For example, if a certain area belongs to collective land, the compensation amount is calculated according to the compensation standard for collective land. Result Output: The results of the intelligent interpretation are presented in a visual manner, such as marking different ownership areas and corresponding compensation amounts in a 3D scene, providing designers with intuitive decision-making support.

[0318] Model evaluation: The model is evaluated using a test dataset. Metrics such as accuracy and recall are calculated to assess the model's performance in natural language and 3D scene mapping as well as complex rule interpretation.

[0319] Continuous optimization: Based on the evaluation results, optimize the model by adjusting its structure and parameters, or by adding training data to improve its performance and accuracy. Simultaneously, continuously update the compensation criteria and ownership rules to ensure the model can adapt to new rule changes.

[0320] When horizontal and vertical design are coordinated, the multi-source policy, regulation, and compensation standard data will remain unchanged, but the structured land acquisition and demolition knowledge graph will be reconstructed, specifically as follows:

[0321] Existing data should be comprehensively reviewed, including multi-source policies, regulations, and compensation standards, clarifying the scope, format, and content of the data. While the data itself remains unchanged, its completeness and accuracy must be ensured to provide a solid foundation for knowledge graph construction. Other data related to land acquisition and demolition, such as land ownership information, building data, and population information, should also be collected; this data will enrich the content of the knowledge graph.

[0322] Remove noise, duplication, and errors from the data to ensure data quality. For example, standardize the format of policy and regulatory texts and correct typos. Standardize data from different sources to ensure consistent data structure and coding rules, facilitating subsequent knowledge extraction and fusion.

[0323] Entity recognition: Utilizing natural language processing techniques, such as Named Entity Recognition (NER) algorithms, this involves identifying important entities from policy regulations, compensation standards, and other relevant data. These entities include the entities involved in land acquisition and demolition (government departments, developers, etc.), the objects of acquisition and demolition (land, housing, etc.), and compensation items (monetary compensation, resettlement compensation, etc.). Pre-trained language models, such as BERT, can be used in conjunction with labeled training data to train the entity recognition model, thereby improving recognition accuracy.

[0324] Relationship extraction: Determining the relationships between entities, such as "expropriation entity - expropriation target" and "compensation item - compensation standard". This can be achieved using rule-based methods, defining relationships based on explicit statements in policies and regulations; or using machine learning or deep learning methods to automatically extract relationships by learning from large amounts of text data. For example, by analyzing text such as "A government department expropriates a plot of land and provides monetary compensation," the relationship between "expropriation entity (government department) - expropriation target (land plot)" and "compensation item (monetary compensation) - expropriation target (land plot)" can be extracted.

[0325] Entity alignment: Because different data sources may contain different representations of the same entity, entity alignment is necessary. This involves merging the same entity from different data sources by comparing its attributes, characteristics, and contextual information. For example, different documents might use "a certain plot of land" and "the plot number is XXXX" to refer to the same plot of land; these need to be aligned to form the same entity.

[0326] Relationship integration: This involves consolidating the extracted relationships to eliminate conflicts and inconsistencies. For example, different policies and regulations may have different compensation standards for the same compensation item, requiring unification and coordination.

[0327] Knowledge Graph Construction and Storage: Graph Construction: A graph database (such as Neo4j) is used to construct the knowledge graph. Extracted and merged entities and relationships are stored in the form of a graph, with entities as nodes and relationships as edges. Attributes are added to each node and edge, such as the entity's name, type, and description, and the relationship's name and weight, enriching the information in the knowledge graph.

[0328] Establish a knowledge graph storage and management system to achieve efficient storage, retrieval, and updating of knowledge. Corresponding interfaces can be developed to facilitate access to and use of the knowledge graph during the horizontal and vertical design processes.

[0329] Quality Assessment: Develop assessment metrics such as accuracy, recall, and completeness to evaluate the quality of the knowledge graph. Identify problems and shortcomings in the knowledge graph through manual review and comparative testing. Optimization and Improvement: Based on the assessment results, optimize and improve the knowledge graph. For example, adjust the algorithms for entity recognition and relation extraction, increase training data, and refine knowledge fusion strategies to continuously improve the quality and usability of the knowledge graph.

[0330] In a preferred embodiment of the present invention, in the graphic template of the big data prediction model for land acquisition and demolition costs, when the user modifies the values ​​of some design parameters through the interactive module, the position and size of the objects in the two-dimensional design drawing are adjusted through the position and size correspondence between different objects;

[0331] The location of the chain break does not change. When the starting point is moved, the location of the chain break will not change, but the mileage before the chain break will change.

[0332] When users perform operations such as "Calculation of House Demolition Unit Price or Fee", "Calculation of Unit Demolition Unit Price or Fee", and "Calculation of House Demolition Fee", the house demolition unit price may change due to the inability to use the weighted average comprehensive unit price based on line length. When "Modify Design Line Start Point", if the starting mileage is located after a part of the existing broken chain, the broken chain part will be deleted from the system. When moving the intersection point, if a short chain becomes a long chain, it violates the principle that "long chains must change the prefix", and a prompt will be generated for user interaction.

[0333] The line has a short chain, K0+130|K0+150. Change the starting mileage of the designed line from K0+0 to K0+50. When the chain break control method and mileage remain unchanged, the chain break location will be re-determined based on the mileage before the chain break.

[0334] When the chain break control method = position remains unchanged, the chain break becomes K0+180|K0+150, which violates the principle that the prefix of the long chain must be changed. When the prefix of the long chain remains unchanged, add a "#" sign before the prefix.

[0335] The position and size correspondence between different objects is expressed as follows: Chain break control method = position remains unchanged, move the intersection point. When there is a chain break on the straight line, move the intersection point and give a prompt.

[0336] In a preferred embodiment of the present invention, when the railway crosses highways, rivers, reservoirs, sea areas or passes through mountains, there are no breaks in the straight line, and the amount and cost of land acquisition are not considered.

[0337] If local governments at the county level or above have policy documents requiring compensation for the water surface, riverbank, beachhead, and sea area of ​​rivers, reservoirs, and sea areas, the compensation fee shall be calculated according to the compensation scope and unit price stipulated by the local government, and shall not be included in the land acquisition quantity. It can be moved. If the movement results in a broken chain on the curve, the broken chain will be automatically deleted.

[0338] After selecting the insertion point, the system will display the distance as 10 meters rounded up by default.

[0339] Since there are no stipulated fees for requisitioning construction land, the compensation will be calculated by referring to the unit price of dry land collective land in the same region. Compensation for surface attachments and greening will be determined separately based on the survey data, as follows:

[0340] Ground attachment survey: Types and quantity statistics: Conduct a detailed survey of all ground attachments on the requisitioned construction land, identify their types, such as buildings, structures (including simple sheds, warehouses, etc.), farmland water conservancy facilities (irrigation wells, canals, etc.), roads, etc., and count their respective quantities.

[0341] Feature Recording: Record the characteristics of the attachments, such as the building area, structural type (e.g., brick-concrete structure, brick-wood structure), and service life of the building; the species, diameter at breast height, height, and crown width of the trees. These characteristics are crucial for determining the compensation value later.

[0342] Green Space Survey: Green Space Layout and Area: Understand the distribution and size of green areas, and distinguish between different types such as public green spaces and private green spaces. Plant Species and Specifications: Record in detail the types of green plants, such as flowers, lawns, trees, and shrubs, and determine their specifications, such as the number of flowers, the area of ​​lawn coverage, and the diameter at breast height and height of trees.

[0343] Market Price Information Collection: Local Market: Collect recent transaction price information for similar ground attachments and green plants in the local area, including the construction cost of new buildings, the transfer price of second-hand buildings, and the market price of different types of trees and flowers. Industry Standards: Refer to relevant industry cost standards and price indices, such as engineering quotas in the construction industry and seedling price guidelines in the landscaping industry, to provide a reference for determining the compensation value.

[0344] Valuation of surface attachments can be categorized into three approaches: **Cost Approach:** For newly constructed or relatively new buildings and structures, the replacement cost approach is used to calculate their value. This involves calculating the cost of rebuilding the same or similar attachments based on current building material prices and labor costs, then adjusting for depreciation. **Market Comparison Approach:** If there are market transactions of similar attachments, the market comparison approach can be used. By comparing the transaction prices with those of the attachments being expropriated, the transaction prices are adjusted to determine the value of the expropriated attachments. **Income Approach:** For attachments with income-generating functions, such as commercial shops and factories, the income approach can be used to assess their value. The present value is calculated based on the expected income, the income period, and the discount rate.

[0345] Green space compensation value assessment: Seedling value assessment: For greening plants, their value is determined based on factors such as variety, size, and market price. For rare and valuable plants, it may be necessary to invite a professional landscape appraiser for evaluation. Green space project cost assessment: The construction cost of the green space project is considered, including land preparation, seedling procurement, planting, and maintenance expenses. Simultaneously, the landscape effect and ecological value of the green space should be considered, and appropriate compensation should be provided.

[0346] Compensation Plan: Inform the rights holders of the expropriated land of the preliminary assessment results, including the calculation methods and amounts for compensation for surface attachments and green spaces. Listen to their opinions and demands, and conduct thorough communication and negotiation. Adjustment of Compensation Plan: Based on the feedback from the rights holders, make appropriate adjustments to the compensation plan. If the rights holders have objections to the assessment results, they can invite a professional assessment agency to conduct a reassessment to ensure the compensation plan is fair and reasonable.

[0347] When determining the final compensation plan, in addition to considering the value of ground attachments and greenery, local policies and regulations, social stability, and the overall interests of the project must also be taken into account. A final compensation plan for ground attachments and greenery is then formulated based on all these factors and submitted to the relevant departments for approval. For large temporary stations, the land area is calculated based on the area outside the drainage ditch outside the perimeter wall. The process is described in "Design Line Attributes -> Advanced -> Junction Point -> Mileage Point". In addition to the site layout, the track yard also considers the land use for railway branch lines. When the design line is selected, the break point will be displayed as a yellow junction point.

[0348] The land use for railway branch lines is calculated based on the area outside the slope toe of the drainage ditch on both sides of the branch line. When modifying the mileage before the chain break, the previous mileage is used to determine the location of the chain break, and the input value is within the current mileage system. All chain breaks between the new location and the original location will be deleted. When modifying the mileage after the chain break, the newly input mileage after the chain break is within the current mileage system. If a chain break smaller than this mileage is deleted and is not within the current mileage system, then the route selection and design system will re-order the mileage.

[0349] When the route selection and design system automatically follows the mileage, it always processes the route based on the chain break control method, ensuring that the mileage and position remain unchanged.

[0350] When the demolition unit price adopts the transaction price, it includes the land use fee. The corresponding land acquisition quantity is included in the permanent land acquisition quantity, but the land acquisition compensation fee is no longer listed. The land acquisition unit price is set to zero, and the parameter variables of objects with the same name in the graphic template are linked together.

[0351] In a preferred embodiment of the present invention, the objective function and loss function for constructing the big data prediction model for land acquisition and demolition costs are specifically as follows:

[0352] In the prediction of land acquisition and demolition costs, assuming that the dataset has m samples and each sample has n features, the goal is to minimize the mean of the sum of squared errors between the predicted values ​​of all samples and the actual land acquisition and demolition costs.

[0353] x i y is the feature vector of the i-th sample. iLet be the actual demolition cost corresponding to the i-th sample. The loss function for linear regression is the squared error loss, which measures the degree of error between the predicted and actual values ​​of a single sample. The objective function of the decision tree regression model is to partition the feature space so that the target values ​​of samples within each partition are as close as possible. For the regression problem, the decision tree is constructed by minimizing the mean squared error of samples within each leaf node. During tree construction, the goal is to select the optimal splitting features and splitting points at each node, minimizing the sum of the mean squared errors of the child nodes after the partition. Let R_m be the sample set corresponding to the m-th leaf node of the decision tree, and c_m be the mean of the target values ​​of the samples in R_m. The objective function is:

[0354] J=sum{m=1} M \sum_{x i \inR_m}(y i -cm) 2 ,

[0355] Where M is the number of leaf nodes.

[0356] The loss function, the loss function for a single sample in decision tree regression, can also be expressed as mean squared error: L(x i ,y i )=f(x i ) - y i2， Where, f(x) i ) is a decision tree for sample x( i The predicted value of ).

[0357] In another preferred approach, the objective function of the random forest is an ensemble of multiple decision trees. Its goal is to reduce the model's variance and improve the accuracy and stability of predictions by combining the predictions from multiple decision trees. Let the random forest consist of n_{trees} decision trees, and let the j-th decision tree be the most suitable for sample x. i The predicted value is f_j(x) i The predicted value of the random forest is hat{y}. i The objective function is:

[0358] J=frac{1} m \sum_{i=1} m (hat{y} i - y i ) 2 ,

[0359] The loss function for a single sample in a random forest is the mean squared error loss.

[0360] L(\hat{y} i ,y i=(hat{y} i - y i ) 2 ,

[0361] In another preferred approach, a multilayer perceptron (NNP) model is used. The objective function of the NLP is to minimize the error between the predicted and actual values ​​by adjusting the network weights W and biases b. Similar to linear regression, the mean squared error is commonly used as the objective function.

[0362] J(W,b)=frac{1} 2m \sum_{i=1} m (h_{W,b}(x i ) - y i ) 2 ,

[0363] Where h{W,b}(x i ) is a multilayer perceptron for processing sample x i The predicted values ​​are calculated by neurons in the input layer, hidden layer, and output layer.

[0364] The loss function for a single sample is L(h_{W,b}(x)). i ),y i )=(h_{W,b}(x i )-y i ) 2 The selection of these objective functions and loss functions is based on the characteristics of different models and the needs of land acquisition and demolition cost prediction, with the aim of enabling the model to predict land acquisition and demolition costs more accurately.

[0365] The specific embodiments described herein are merely illustrative examples of the present invention. Those skilled in the art can make various modifications or additions to the described embodiments or use similar methods to substitute them, without departing from the technology of the present invention or exceeding the scope defined by the appended claims.

Claims

1. A method for evaluating land acquisition and demolition costs for railway route selection based on a big data model, characterized in that, Includes the following steps: Acquire geographical information along the railway line, railway design information, core land acquisition and demolition data, and historical land acquisition and demolition data, and preprocess them to build a graphic database and a core land acquisition and demolition database; By acquiring policy-related land acquisition and demolition documents and dynamic data of the target area through external dynamic links, an external database is created; based on the graph database, the core land acquisition and demolition database, and the external database, the key factors affecting land acquisition and demolition costs are identified, a big data prediction model for land acquisition and demolition costs is constructed, the model is trained, and then integrated into the railway alignment design system. When inputting or modifying route plans in the route selection and design system, the system calls the big data prediction model for land acquisition and demolition costs in real time. Based on the geometric parameters and spatial location of the route, it dynamically calculates and feeds back the corresponding prediction results of land acquisition and demolition costs. The route plan will be adjusted based on the predicted land acquisition and demolition costs until the cost control requirements are met.

2. The railway route selection and land acquisition cost assessment method based on a big data model according to claim 1, characterized in that, Acquire geographical information along the railway line, railway design information, core land acquisition and demolition data, and historical land acquisition and demolition data, and perform preprocessing to construct a graphic database and a core land acquisition and demolition database, specifically as follows: The collected land area, land ownership, and building area data are cleaned to remove noisy, duplicate, and erroneous data. The data is then coded and normalized to convert different types of data into a format suitable for model analysis. The land type and building structure type classification data are coded; Normalize the numerical data of land area and building area to eliminate the influence of data units; If the database for unit prices of demolition costs for demolished houses is empty, the data in the policy index will be used by default. If not empty, then create the cost information according to the file; The data composition of the unit price of demolition costs includes: during the design phase, slope line data is saved, and "land compensation fee information", "forest land and timber fee information" and "social security fee information" are also saved in the database; The collected historical land acquisition and demolition data were cleaned to remove noise, duplicates and errors, and the categorical data and numerical data were coded and normalized respectively. Establish a graphical database containing geographic information along the railway line and railway design parameters. The graphical database includes multiple objects used to form interactive route selection, parameter variables of the objects, and multiple parameter variables used to construct a numerical model system for predicting land acquisition and demolition costs and to establish triangular network data. Establish an independent core database for land acquisition and demolition, which should include at least a land price sub-database, a housing replacement price sub-database, and a demolition compensation standard sub-database.

3. The railway route selection and land acquisition cost assessment method based on a big data model according to claim 2, characterized in that, The steps to obtain multiple objects from the graph database used to construct interactive line selection, the parameter variables of these objects, and multiple parameter variables used to construct a numerical model system for land acquisition and demolition cost prediction and to establish triangular network data are as follows: Select the layer containing the entity to be collected and collect the entity. After all the entities are collected, optimize the data, including points, arcs, circles, lines, diagonal lines, text, layers, block names, text insertion points, ellipses, curves, ground lines, control points, graphic annotations, tables, elevation ranges, rectangles, and each object contains at least one parameter variable. The ground line data format is "imported data mileage and elevation"; Specify the ground line information table. If the specified file does not exist, it will be created automatically; if it exists, the data will be loaded and sorted. The generated files are sorted according to the group settings and marked on the design line; The optimization scheme is set up so that duplicate points at the same location within 1 meter are considered duplicates, and the mileage information of the last input is retained. The control point data format is set to "prefix import data elevation description", and the content is saved in the graphic database.

4. The railway route selection and land acquisition cost assessment method based on a big data model according to claim 1, characterized in that, The specific steps for creating an external database to obtain policy-related land acquisition and demolition documents and dynamic data for the target area through external dynamic links are as follows: An interactive module is set up, which is connected to an external database. Local land acquisition and demolition compensation standards are selected, and the land acquisition area, demolition quantity, control points, and leveling points are saved to the external database. Select cost estimation to calculate key land acquisition and demolition data in an external database; Land pre-approval fee: On the selected slope section, the fee is calculated at 0.65 × (1 + 20%) ten thousand yuan per kilometer of main line, starting from the previous slope change point. The input slope length cannot be greater than the current slope section length. For independent large-scale station buildings, maintenance bases, EMU depots, and container center stations that are listed separately, the calculation is based on the amount of land used for the project, at 110 × (1 ± 10%) yuan / mu. The statistical points are increased based on the input demolition quantity and cost, and the calculation is based on 247 yuan / mu, referring to the cleared projects. Selecting the progress verification entity will save key land acquisition and demolition data to a database file. Users can input, modify, or delete design parameters in the external database through an interactive program. The external database uses a custom entity with design information to store multiple design parameters input by the user. This custom entity, representing the interactive alignment selection, sets multiple design parameters, including names and values, specifically: Store different categories of design parameters in different tables, convert single lines to double lines, and keep the curve elements of the right line the same as those of the left line after conversion. To convert a polyline into a design line (single or double line), select the polyline to be converted and create a new design line input template. When generating a double line, the inner and outer curve elements differ by one line spacing. Save the design lines as a data file. The data file is a text file. Single lines have the extension ".1xl" and double lines have the extension ".2xl". Load the data from the planar data file into the drawing and convert it into a design line as needed. When Gype=1, it is equivalent to the command "zx3", which converts a polyline to a design line; When Gype=2, it is equivalent to the command "zx2", which converts a single line to a double line; When Gype=3, a double line is converted into two single lines; When Gype=4, the intersection points of multiple aligned edges are combined into a single line.

5. The method for evaluating railway route selection and land acquisition costs based on a big data model according to claim 1, characterized in that, Based on graph databases, core land acquisition and demolition databases, and external databases, the key factors affecting land acquisition and demolition costs were identified. The specific steps are as follows: Identify specific land acquisition and demolition projects, and clarify their geographical location, scale, and nature. Set research objectives: Clearly define the scope of expropriation and demolition costs to be analyzed, including direct costs: land compensation fees and resettlement subsidies, indirect costs: demolition management fees and unforeseen expenses, and research objectives: cost drivers and prediction of expropriation and demolition costs; Collect cost data for historical land acquisition and demolition projects, including specific amounts of fees, scope of acquisition and demolition, time periods, land and building information: understand the nature of land within the acquisition and demolition area: state-owned land, collective land, area, use, and the number, structure, area, and age of buildings; policy and regulatory documents: collect relevant policies, regulations, and compensation standards related to land acquisition and demolition, and understand the government's regulations and requirements for land acquisition and demolition compensation; market data: monitor local real estate market price trends, land transfer prices, and building material prices. Preliminary screening of potential influencing factors includes: Land factors: location, area, use, and ownership of the land; Building factors: the building's area, structure, quality, and decoration; Policy and regulatory factors: government-formulated policies on land acquisition and demolition compensation and resettlement; Market factors: supply and demand in the real estate market, land prices, and building material prices; The key factors were identified through correlation analysis using data analysis methods, specifically: Statistical analysis methods were used to calculate the correlation coefficients between various factors and land acquisition and demolition costs, and to identify the factors related to land acquisition and demolition costs. Regression analysis: Establish a regression model, take the cost of land acquisition and demolition as the dependent variable and the influencing factors as independent variables, and determine the degree of influence of each factor on the cost of land acquisition and demolition through regression analysis; Sensitivity analysis: By changing the value of a certain factor, observe the changes in land acquisition and demolition costs to determine the sensitivity of that factor to land acquisition and demolition costs; The big data prediction model for land acquisition and demolition costs adopts a neural network algorithm model, a random forest algorithm model, or a support vector machine algorithm model.

6. The railway route selection and land acquisition cost assessment method based on a big data model according to claim 5, characterized in that, The steps for training a big data prediction model for land acquisition and demolition costs are as follows: Select a preset graphic template, divide the historical land acquisition and demolition data into training set and test set, associate the parameter variables of the training set objects with the design parameters of the relevant test set, and use the training set data to train the big data prediction model for land acquisition and demolition costs. Optimize model performance by adjusting model parameters: number of layers and number of nodes; Insert test points and add short codes to improve input efficiency: 1="paddy field", 2="dry land", 3="vegetable garden"; input other text and record it accurately. The trained big data prediction model for land acquisition and demolition costs was validated and evaluated using test set data. The model parameters were adjusted based on the evaluation results until the model achieved the preset prediction accuracy.

7. The railway route selection and land acquisition cost assessment method based on a big data model according to claim 6, characterized in that, The steps for loading objects from the graphics template from the graphics database sequentially or in parallel, based on the positional relationships of the objects in the graphics template, are as follows: S71, in the matched graphic template, establish the position and size correspondence between different objects in the order from left to right and from top to bottom, including annotation tangent, annotation feature point, intersection information, intersection number, intersection coordinates, curve deflection angle, curve radius, transition curve length, tangent length, and curve length; Based on the relevant design parameters associated with the object's parameter variables, including design line width, mileage text height, font, number of decimal places, mileage display method, hundred-meter marker style, chain break control method, and the value of the starting intersection, update the positional relationship of each object in the graphic template, and obtain the relevant design line attribute information of each object in the graphic template after updating the positional relationship; S72, taking the center of the image template as the origin, divides the interactive line selection graphic template into four quadrants—upper left, lower left, lower right, and upper right—using a cross, and displays the five major feature points of the intersection and the center of the circle; If the distance between the center position of an object in each quadrant and the origin is less than the distance threshold, the color scheme is entered. The custom configuration is performed according to "pure white", first line color="0", second line color="0", transition curve color="0", circular curve color="0", tangent color="0", mileage text color="0", whole kilometer and broken chain text color="0". Then, each object in the graphic template is loaded serially from the graphic database to generate a big data prediction model for land acquisition and demolition costs. If the distance between the center position of the object in each quadrant and the origin is less than the distance threshold, proceed to step S73; the distance threshold ranges from one-quarter to one-third of the diagonal length of the image template, and the image template is a rectangle or a square. S73 uses the chaotic artificial fish swarm algorithm, GA-BP algorithm, GEP algorithm, BP neural network, and optimized RBF valuation model to cluster the surrounding position coordinates of all objects, obtain multiple cluster centers and the position coordinates of each cluster center, and assign each object to the cluster center that is closest to the center position of the object, thereby realizing the division of the graphic template area. When creating a new design line, set the interactive prefix; the default prefix is ​​"K". Each land acquisition and demolition scenario is represented by "planned progress | on-site data". A "data management method" is introduced, which is divided into "on-site land acquisition" preferred implementation method and "house demolition" preferred implementation method. The algorithm is integrated to calculate the land acquisition and demolition costs.

8. The method for evaluating railway route selection and land acquisition costs based on a big data model according to claim 7, characterized in that, When inputting or modifying a route plan in the route selection and design system, the system obtains the track route and creates a new longitudinal profile. If the ".GJKOs" graphic template is selected, the route selection and design system automatically adds land acquisition and demolition objects in conjunction with the system. When constructing a new land acquisition and demolition plan, the road alignment design system cuts a triangulation network according to the alignment position to obtain the initial ground points, including GU1, GU2, ..., GUN; When the alignment changes, a statistical ledger is generated based on the statistical conditions. The statistical ledger is organized according to the administrative divisions at all levels. The statistical items include the total investment amount, the completed amount, and the quantity and price statistics of various land acquisition and demolition features. This indicates that the section has changed. The road alignment design system deletes points that are not on the alignment. By selecting the statistical area, land use type, land use zone, demolition and relocation status, and start and end time, the demolition and relocation objects participating in the statistical calculation can be screened to improve the efficiency of the linkage and only cut out the changed locations. When designing the plan, an AI model is used to map natural language to the 3D scene. When ground lines intersect, the ground points are recut according to the line positions to interpret complex rules such as compensation standards and ownership relationships. When horizontal and vertical design are coordinated, the multi-source policy, regulations and compensation standard data will not change, and the structured land acquisition and demolition knowledge graph will be reconstructed.

9. The method for evaluating railway route selection and land acquisition costs based on a big data model according to claim 8, characterized in that, In the graphical template of the big data prediction model for land acquisition and demolition costs, when users modify the values ​​of some design parameters through the interactive module, the position and size of objects in the two-dimensional design drawing are adjusted through the position and size correspondence between different objects. The location of the chain break does not change. When the starting point is moved, the location of the chain break will not change, but the mileage before the chain break will change. When users perform operations such as "Calculation of House Demolition Unit Price or Cost", "Calculation of Unit Demolition Unit Price or Cost", and "Calculation of House Demolition Cost", the house demolition unit price cannot use the weighted average comprehensive unit price based on line length, resulting in a change. When "Modify Design Line Start Point", if the starting mileage is located after a part of the existing broken chain, the broken chain part will be deleted from the system. When moving the intersection point, if a short chain becomes a long chain, it violates the principle that "long chains must change the prefix", and a prompt will be generated for user interaction. The line has a short chain, K0+130|K0+150. Change the starting mileage of the designed line from K0+0 to K0+50. When the chain break control method and mileage remain unchanged, the chain break location will be re-determined based on the mileage before the chain break. When the chain break control method = position remains unchanged, the chain break becomes K0+180|K0+150, which violates the principle that the prefix of the long chain must be changed. When the prefix of the long chain remains unchanged, add "#" before the prefix. The position and size correspondence between different objects is expressed as follows: Chain break control method = position remains unchanged, move the intersection point. When there is a chain break on the straight line, move the intersection point and give a prompt.

10. The method for evaluating railway route selection and land acquisition costs based on a big data model according to claim 9, characterized in that, When a railway crosses a highway, river, reservoir, sea, or mountainous area, and there are no breaks in the chain in the straight line, no land acquisition quantity or cost will be counted. If local governments at the county level or above have policy documents requiring compensation for the water surface, riverbank, beachhead, and sea area of ​​rivers, reservoirs, and sea areas, the compensation fee shall be calculated according to the compensation scope and unit price stipulated by the local government, and shall not be included in the land acquisition quantity. It can be moved. If the movement results in a broken chain on the curve, the broken chain will be automatically deleted. If there is no stipulated fee for requisitioning construction land, it shall be calculated by referring to the unit price of dry land collective land in the same region. Compensation for ground attachments and greening shall be determined separately based on the survey data. For large temporary stations, the land area is calculated based on the area outside the perimeter wall and drainage ditch. In "Design Line Properties -> Advanced -> Clips -> Mileage Points", the track yard is not only based on the site layout, but also considers the land use of railway branch lines. When the design line is selected, the broken chain will be displayed as a yellow clip. The land use for railway branch lines is calculated based on the area outside the slope toe of the drainage ditch on both sides of the branch line. When modifying the mileage before the chain break, the previous mileage is used to determine the location of the chain break, and the input value is within the current mileage system. All chain breaks between the new location and the original location will be deleted. When modifying the mileage after the chain break, the newly input mileage after the chain break is within the current mileage system. If a chain break smaller than this mileage is deleted and is not within the current mileage system, then the route selection and design system will re-order the mileage. When the route selection and design system automatically follows the mileage, it always processes the route based on the chain break control method, ensuring that the mileage and position remain unchanged. When the demolition unit price adopts the transaction price, it includes the land use fee. The corresponding land acquisition quantity is included in the permanent land acquisition quantity, but the land acquisition compensation fee is no longer listed. The land acquisition unit price is set to zero, and the parameter variables of objects with the same name in the graphic template are linked together.

Images