CN-121998499-A - Underground space value evaluation method integrating multi-source space big data

CN121998499ACN 121998499 ACN121998499 ACN 121998499ACN-121998499-A

Abstract

The invention relates to the technical field of asset management, in particular to an underground space value evaluation method for fusing multi-source space big data, which comprises the steps of acquiring the multi-source space big data in real time, performing space-time fusion processing on the acquired multi-source space big data, and generating a space-time data set of a target underground space; the method comprises the steps of calculating a plurality of evaluation indexes of a target underground space based on a space-time data set, predicting the value of the target underground space by utilizing the plurality of evaluation indexes, comprehensively evaluating the plurality of evaluation indexes through a machine learning model, and correcting the predicted value based on a comprehensive evaluation result to obtain a final evaluation value.

Inventors

SUN PING
DONG JIE
WANG ZHONGSHENG
YU PENG
HUANG YUXIAO
XU MEIJUN

Assignees

青岛地质工程勘察院(青岛地质勘查开发局)

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. A method for evaluating the value of a subsurface space fused with big data of a multi-source space, the method comprising: acquiring multi-source space big data in real time, and performing space-time fusion processing on the acquired multi-source space big data to generate a space-time data set of a target underground space; Calculating a plurality of evaluation indexes of the target underground space based on the space-time data set, and predicting the value of the target underground space by using the plurality of evaluation indexes; And comprehensively evaluating the multiple evaluation indexes through a machine learning model, and correcting the predicted value based on the comprehensive evaluation result to obtain a final evaluation value.
2. The method for evaluating the value of an underground space fused with multi-source spatial big data according to claim 1, wherein the acquiring the multi-source spatial big data in real time, performing space-time fusion processing on the acquired multi-source spatial big data, and generating a space-time data set of a target underground space, comprises: acquiring space data, wherein the space data comprises urban GIS road network data and BIM drawing data of a target underground space; acquiring dynamic people stream data, including Wi-Fi or Bluetooth probe data and mobile phone signaling data; acquiring environment and operation data, including business interest points, business state distribution data, sales/lease data and weather data; The method for generating the space-time data set of the target underground space by carrying out space-time fusion processing on the obtained multi-source space big data comprises the following steps: unifying time references, namely unifying acquired space data, dynamic people stream data, environment and operation data into the same city coordinate system; The unified space standard aligns the acquired space data, dynamic people stream data, environment and operation data to UTC+8 (Beijing time) on a time axis, wherein the granularity is 15 minutes; and preprocessing the multisource space big data after unifying the time reference and the space reference to form a space-time data set for space value evaluation.
3. The method for evaluating the value of a subsurface space in combination with big data of a multi-source space according to claim 2, wherein the calculating a plurality of evaluation indexes of the target subsurface space based on the spatio-temporal data set, and predicting the value of the target subsurface space using the plurality of evaluation indexes, comprises: calculating the reachability index of the target underground space in the regional network where the target underground space is located; calculating attractive contribution indexes of structural features of the target underground space to the people stream; predicting the flow of people entering a target underground space, and quantifying consumption conversion indexes of the flow; The value of the target underground space is predicted by using the reachability index, the consumption conversion index and the attractive contribution index.
4. A method of evaluating the value of a subsurface space in combination with multi-source spatial big data according to claim 3, wherein said calculating the reachability index of the target subsurface space in the area network thereof comprises: constructing a three-dimensional space network of an area where a target underground space is located, comprising: with the position of departure of the stream of people as the source point Taking a transfer point or a path junction point in a traffic network as a connection point Taking the underground space to be evaluated as an entrance set and a target point Forming a key node set of the three-dimensional space network by utilizing the source point, the connecting point and the target point; the connecting paths of key nodes on the same horizontal plane are used as horizontal edges, and each horizontal edge As a weight for the horizontal edge ; The connecting paths of key points of planes with different heights are taken as vertical edges, and each vertical edge Is taken as the weight of the vertical side ; Forming a key edge set of the three-dimensional space network by utilizing the horizontal edges and the vertical edges; Graph with three-dimensional space network formed by key node set and key edge set , wherein, A set of keypoints is represented and, Representing a set of key edges, wherein, ; In the drawings A minimum cumulative transit cost, comprising: Searching a path sequence with minimum passing cost from each source point to a target point by using a shortest path Dijkstra algorithm Wherein, the method comprises the steps of, Represent the first Source point to the first A set of paths for the individual target points; , , 、 Respectively representing the number of source points and the number of target points; Then the minimum accumulated traffic cost is Wherein, the method comprises the steps of, Representing a path One edge of (a); three-dimensional reachability index for target point Wherein, the method comprises the steps of, Represent the first The weights of the individual source points are chosen, Represent the first A distance attenuation coefficient of the traffic mode; the weight of the traffic pattern is represented, Representing the number of traffic patterns; Integrating multiple source points to obtain multiple integral target underground space stereoscopic accessibility indexes 。
5. The method for evaluating the value of a subsurface space in combination with multi-source spatial big data according to claim 4, wherein said calculating the reachability index of the target subsurface space in the area network thereof further comprises: considering traffic conditions for different time periods, the distance decay index for each traffic pattern is time-varying, i.e.: for a target point Three-dimensional reachability index of time ; Representation of Time of day (time) A distance attenuation coefficient of the traffic mode; obtaining a stereoscopic reachability index time series Wherein, the method comprises the steps of, Representing an evaluation period; based on the stereoscopic reachability index time sequence, recalculating stereoscopic reachability indexes of a plurality of integral target underground spaces to obtain dynamic reachability indexes 。
6. The method for evaluating the value of a subsurface space with integrated multi-source spatial big data according to claim 5, wherein the calculating the attractive contribution index of the structural feature of the target subsurface space to the stream of people comprises: Obtaining a scale index of the target subterranean space for reflecting a capacity of the target subterranean space, comprising: calculating scale index of target subsurface space , wherein, The function of the normalization is represented by a function of the normalization, Representing the total rentable area of the target underground space, A common area headroom representing a target subsurface space; the weight of the area index is represented, Representing height index weights; obtaining a target underground space internal communication efficiency index comprising: Dividing a channel part in a target underground space plane into a plurality of grids based on BIM drawing data of the target underground space, wherein the center of each grid is a node, and the nodes are connected through straight line segments Wherein Representing nodes To the node The shortest topology step number of (a); integration level of each node Wherein, the method comprises the steps of, Representing the number of nodes; the average integration level of the target underground space is ; Obtaining a functional mix index for a target subsurface space, comprising: Acquiring shop data in a target underground space, and classifying shops; counting the proportion of the area of each type of shops to the total rentable area Wherein, the method comprises the steps of, Representation of Area of the class shop; Based on Calculating entropy of store type distribution ; Obtaining attractive contribution index Wherein, the method comprises the steps of, 、 And Representing the normalized scale index, the average integration and the entropy; 、、 And representing the index fusion weight coefficient.
7. The method for evaluating the value of a subsurface space with integrated multisource spatial big data according to claim 6, wherein predicting the flow of people into the target subsurface space and quantifying the consumption conversion index thereof comprises: predicting a flow of people into a target subterranean space, comprising: dynamic people stream data based acquisition Time of day passes through the target point Nearby people flow ; Cut-off rate of target point ; People stream from target point Utility function in accessing a target subsurface space ; The term of the constant is represented by a term, 、、 The weight coefficient is represented by a number of weight coefficients, The characteristics of the time period are represented, ; The weather-state index is indicated as such, Or (b) Utility function without entering target subsurface space ; Predicting slave target points The flow of people entering the target underground space is ; The incoming traffic of the entire target underground space is ; Quantifying consumption conversion metrics of a stream of people entering a target underground space, comprising: consumption conversion index Wherein, the method comprises the steps of, The expression of the consumption is given, Indicating access to the target subsurface space; representing a probability of consumption of the access target subsurface space; A consumer trend index; ; The index constant term is consumed and the index value is calculated, 、、 The weight coefficient representing the consumption index, Representing the density of the people stream in real time, Representing a target subsurface space area; the predicted number of consumers entering the target subterranean space is ; The predicting the value of the target underground space by using the reachability index, the consumption conversion index and the attraction contribution index comprises the following steps: Obtaining Predicted consumption of class shops ; Wherein, the Time of day per capita consumption , Represents the characteristic adjustment coefficient of the time period, Representation of Average sales of class shops, overflow coefficient ; Representing the premium weight; Predicting a value of the target subsurface space based on the consumption, comprising: calculating annual revenue for a target subsurface space Wherein, the method comprises the steps of, Representing the annual consumption of all shops, Indicating the rate of rental/sales, Representing annual operating expenses of all shops; ; Representation of Annual average number of days of operation of the shop; The commercial value of the target underground space is , wherein, The cost-effectiveness is indicated by the index of the fundamentals, , Indicating that there is no risk of interest, Representing the risk factor of the underground space, Representing market risk premium; the long-term growth rate is expressed as the sum of the expansion rate of the currency and the urban development premium.
8. The method for estimating the value of the underground space in combination with the multi-source spatial big data according to claim 7, wherein the comprehensive estimating the multiple evaluation indexes by the machine learning model, correcting the predicted value based on the comprehensive estimating result, and obtaining the final estimated value comprises: 、、 Constructing an index vector ; Input index vector to pre-trained deep neural network output comprehensive evaluation index Comprising: The deep neural network comprises an input layer, a feature extraction layer, an attention layer and an output layer; Attention weighting of an attention layer Wherein, the method comprises the steps of, Representing the first of the index vectors An embedded representation of the individual indicators; 、、 Representing a learnable parameter; outputting the comprehensive evaluation index Wherein, the method comprises the steps of, Representing the first of the index vectors A number of indicators; final evaluation value of , wherein, The correction coefficient is represented by a number of coefficients, Representing the reference index.
9. The method for evaluating the value of a subsurface space fused with multi-source spatial big data as recited in claim 8, further comprising optimizing the stereoscopic reachability based on the structural features of the target subsurface space, comprising: Will be Cost of extending from source point to target point and to target location inside target subsurface space , wherein, Representing the target point from To a target location within a target subterranean space Navigation costs of (2); To target position Nodes of the grid As an alternative to the target position, i.e. ; Wherein, the Indicating the effective road-finding speed and, ; The basic walking speed is indicated as well, Gain coefficient, three-dimensional accessibility after optimization ; And predicting the value of the target underground space again by using the optimized stereoscopic accessibility, consumption conversion index and attractive contribution index.
10. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program, which is executed by a processor to implement a method of estimating a value of a subsurface space fused with multi-source spatial big data according to any of the above claims 1-9.

Description

Underground space value evaluation method integrating multi-source space big data Technical Field The invention relates to the technical field of asset management, in particular to an underground space value assessment method integrating multi-source space big data. Background Assessing the value of a subsurface space is a comprehensive topic involving multiple disciplines and factors. The evaluation dimension includes economic value (most common), social environmental benefit, strategic location value, and the like. The multisource spatial big data brings innovation for the underground spatial value assessment, and the analysis of static, local and supposition in the traditional assessment is changed into accurate decision support of dynamic, panoramic and predictable. The multi-source big data mainly comprises urban operation big data, underground space structure big data, environment big data, socioeconomic and interest point data and the like. The people stream heating power and activity modes can be identified through the urban operation big data, for example, people stream scale and stay time of different time periods of the ground functional area can be described through Wi-Fi data, and matching requirements of the underground space and the ground functional area can be analyzed by combining the POI and the people stream data. The prior art generally realizes the evaluation of underground space value by utilizing multi-source big data by constructing a multi-layer system, specifically, the multi-source big data is collected and fused through a data layer, the multi-source big data is analyzed through a model in an analysis layer, and an evaluation result is visually output through an application layer so as to support decision. However, in the evaluation process, the problems of single use model, isolated and static evaluation indexes, incapability of capturing the relation among different evaluation indexes and the like exist, so that the value evaluation is inaccurate. Disclosure of Invention According to the invention, the value of the target underground space is predicted by calculating a plurality of evaluation indexes and utilizing the fusion analysis of the plurality of evaluation indexes, so that the reliability and the accuracy of the value estimation are improved. The technical scheme provided by the invention is that the underground space value assessment method for fusing multi-source space big data comprises the following steps: acquiring multi-source space big data in real time, and performing space-time fusion processing on the acquired multi-source space big data to generate a space-time data set of a target underground space; Calculating a plurality of evaluation indexes of the target underground space based on the space-time data set, and predicting the value of the target underground space by using the plurality of evaluation indexes; And comprehensively evaluating the multiple evaluation indexes through a machine learning model, and correcting the predicted value based on the comprehensive evaluation result to obtain a final evaluation value. Preferably, the acquiring the multi-source spatial big data in real time performs space-time fusion processing on the acquired multi-source spatial big data to generate a space-time data set of the target underground space, including: acquiring space data, wherein the space data comprises urban GIS road network data and BIM drawing data of a target underground space; acquiring dynamic people stream data, including Wi-Fi or Bluetooth probe data and mobile phone signaling data; acquiring environment and operation data, including business interest points, business state distribution data, sales/lease data and weather data; The method for generating the space-time data set of the target underground space by carrying out space-time fusion processing on the obtained multi-source space big data comprises the following steps: unifying time references, namely unifying acquired space data, dynamic people stream data, environment and operation data into the same city coordinate system; The unified space standard aligns the acquired space data, dynamic people stream data, environment and operation data to UTC+8 (Beijing time) on a time axis, wherein the granularity is 15 minutes; and preprocessing the multisource space big data after unifying the time reference and the space reference to form a space-time data set for space value evaluation. Preferably, the calculating a plurality of evaluation indexes of the target underground space based on the space-time data set, and predicting the value of the target underground space by using the plurality of evaluation indexes includes: calculating the reachability index of the target underground space in the regional network where the target underground space is located; calculating attractive contribution indexes of structural features of the target underground space to the people stream; predicting the flow of