Environmental Benefit Evaluation and Visualization Understanding Method for Shared Bicycles

Abstract

The present invention belongs to the field of information technology and discloses a method for environmental benefit assessment and visual understanding of shared bicycles, which includes the following steps: Step 1: Obtain a shared bicycle riding dataset and preprocess the data; Step 2: Based on the starting and ending positions of shared bicycle rides, obtain detailed riding trajectories and driving trajectories; Step 3: Define the environmental benefits obtained by replacing driving with cycling based on the route trajectories; Step 4: Divide the city into grids, calculate the environmental benefits obtained by each grid according to the definition in Step 3, and draw a grid heat map; Step 5: Obtain all POI records in the city and visualize the semantic association between the regional function and the environmental benefits of shared bicycles based on the POI embedding in different grids. The present invention enables relevant personnel to intuitively grasp the impact of land use attributes in different regions on energy conservation and emission reduction, so as to better deploy shared bicycles in the city and contribute to the realization of the carbon neutrality goal.

Description

Technical Field

[0001] The present invention belongs to the field of information technology, and particularly relates to an environmental benefit evaluation and visualization understanding method for shared bicycles. Background Art

[0002] Facing the increasingly serious urban traffic and environmental problems, shared bicycles provide residents with a flexible, convenient, low-carbon and environmentally friendly travel mode, solving the "last mile" problem of travel. If shared bicycles are used to replace driving, it can reduce waste gas emissions, relieve traffic pressure, and build a sustainable transportation system.

[0003] Although it is recognized that shared bicycles are environmentally friendly, few studies have quantitatively estimated the environmental benefits of using shared bicycles to replace motor vehicle travel. With the help of GPS devices, we can obtain the trip data of users using the shared bicycle system, including the starting and ending points of all users' rides within a period of time. With the help of these trip big data, existing studies have quantitatively evaluated the environmental benefits of using shared bicycles to replace driving, mainly calculating the reduced energy consumption and air pollutant emissions, but there are two major defects. 1) Existing studies either simply use the Euclidean distance between the starting and ending points or require specific trajectory information of the ride to estimate the riding distance; 2) Although existing methods can calculate the environmental benefits of shared bicycles in different regions, they do not deeply analyze the semantic association between the obtained environmental benefits and geographical regions. There are challenges in how to better estimate the real riding distance and how to analyze the association between the environmental benefits of different regions and the regional usage functions using POI information. Summary of the Invention

[0004] The purpose of the present invention is to provide an environmental benefit evaluation and visualization understanding method for shared bicycles to solve the above technical problems.

[0005] To solve the above technical problems, the specific technical solution of the environmental benefit evaluation and visualization understanding method for shared bicycles of the present invention is as follows:

[0006] An environmental benefit evaluation and visualization understanding method for shared bicycles includes the following steps:

[0007] Step 1: Obtain a shared bicycle ride data set and preprocess the data;

[0008] Step 2: Obtain detailed riding trajectories and driving trajectories based on the starting and ending positions of the shared bicycle rides;

[0009] Step 3: Define the environmental benefits obtained by replacing driving with cycling based on the route trajectories;

[0010] Step 4: Divide the city into grids, calculate the environmental benefits obtained by each grid according to the definition in Step 3, and draw a grid heat map;

[0011] Step 5: Obtain all POI records in the city, and visualize and analyze the semantic association between the regional function and the environmental benefits of shared bicycles based on the POI embeddings of different grids.

[0012] Furthermore, Step 1 includes the following specific steps:

[0013] Step 1.1: Obtain the riding data set of shared bicycles over a period of time and store them in a database:

[0014] A riding record R DBS is represented as follows:

[0015] R DBS =(bID, startT, startLoc, endT, endLoc)

[0016] where bID represents the shared bicycle number, startT and startLoc are the borrowing time and location, and endT and endLoc are the returning time and location;

[0017] Step 1.2: Delete invalid data:

[0018] For all riding records, delete the records with a riding distance less than 0.2 km or greater than 15 km, delete the records with a riding time less than 1 minute or greater than 3 hours, and delete the records with a riding speed greater than 400 m / minute, and finally obtain meaningful riding records.

[0019] Furthermore, Step 2 includes the following specific steps:

[0020] Estimate the actual distances of cycling and driving between the starting and ending points of a riding record: Use the route planning API of Amap to obtain the travel trajectories considering the urban road network. For cycling, obtain the riding distance d bike , riding time and riding trajectory. For driving, obtain the driving distance d drive , driving time and driving trajectory, where a trajectory consists of multiple line segments, reflecting the detailed cycling or driving path.

[0021] Furthermore, Step 3 includes the following specific steps:

[0022] Environmental benefits include the energy consumption saved by cycling instead of driving and the reduction in exhaust emissions. The life cycle analysis method of fuel is used to calculate the total energy consumption and environmental impact in the fuel production and operation stages, and the energy consumption saved by cycling instead of driving is calculated using formula (1):

[0023]

[0024] Where N represents the saved energy consumption, d bike and d drive represent the cycling distance and driving distance obtained from the route planning API of Gaode Map respectively, T d is a distance threshold. For a single trip, when the distance d bike is less than the threshold T d , there is no energy consumption for the trip. When the distance d bike is greater than the threshold T d , replacing driving with cycling will produce environmental benefits. In actual calculation, T d is set to 1 km, indicating that trips with a cycling distance exceeding 1 km have the potential for energy conservation and emission reduction. p = 0.088 is the gasoline consumption per unit driving distance, ρ = 0.72 is the gasoline density, λ e = 0.87 and λ t = 0.95 represent the crude oil utilization rate and transportation efficiency respectively;

[0025] The reduced exhaust gas emissions from replacing driving with cycling are calculated using formula (2):

[0026]

[0027] Where E g represents the emission reduction of a certain exhaust gas g. g represents CO 2 or NO X , f g represents the emission coefficient of exhaust gas g generated by driving a car. For CO 2 emission, f g = 2.93. For NO X emission, f g = 0.006. Further, step 4 includes the following specific steps:

[0028] First, obtain the cycling trajectories of shared bikes in the whole city, remove trips with a distance less than the threshold T d , and obtain eligible cycling trajectories; then, divide the city into grids of 500m * 500m, and map the corresponding driving trajectories of the cycling trajectories into the grids; then, calculate the sum of the environmental benefits of all sub-trajectories falling into each grid according to formulas (1) and (2), including the energy consumption savings N, CO 2 emission reduction and NO X emission reduction By drawing a grid heat map, present the environmental benefits of each grid, including the grid heat map of energy consumption savings, CO 2Grid heat map of emission reduction, NO X Grid heat map of emission reduction.

[0029] Furthermore, step 5 includes the following specific steps:

[0030] Step 5.1: Obtain all POI records in the city using the Gaode Map API;

[0031] Step 5.2: Use the Word2Vec word embedding model to obtain the embedding vectors of POI classes based on the geographical adjacency relationship between different POI types;

[0032] Step 5.3: For each grid, draw the POI embedding map of the grid, visualize the functional semantics of each grid, and analyze the semantic association between regional functions and environmental benefits.

[0033] Furthermore, step 5.1 includes the following specific steps:

[0034] A POI record can be represented as POI i =[loc i , mainType i , midType i , where loc i represents the location of the i-th POI, and mainType i and midType i represent the major category and middle category of the POI to which the i-th POI belongs, respectively. Only 13 major categories and 117 middle categories that can reflect human activities and travel purposes are obtained. The 13 major categories include: "Food and Beverage Services, Shopping Services, Life Services, Sports and Leisure Services, Healthcare Services, Accommodation Services, Scenic Spots, Commercial Residences, Government Agencies and Social Organizations, Science, Education and Culture Services, Transportation Facilities Services, Financial Insurance Services, Companies and Enterprises".

[0035] Furthermore, step 5.2 includes the following specific steps:

[0036] Learning the embedding of POI types in POI classification is analogous to learning word embeddings in a corpus. Regarding each POI middle category as a word, for each POI middle category, find the POI middle categories within a radius of 500 meters with it as the center. Take the POI middle category corresponding to the center as the first word of the sentence, and arrange the other POI middle categories within the circular range in sequence as the following words to construct a sentence. The set composed of all POI middle categories is regarded as a corpus, and the Word2Vec word embedding model is used to calculate the embedding vectors corresponding to each POI middle category, reflecting the semantic relevance of POI types in the geographical space.

[0037] Further, step 5.3 includes the following specific steps:

[0038] Since the obtained embeddings of POI types are usually high-dimensional vectors, the t-SNE dimensionality reduction algorithm is used to map them onto a two-dimensional semantic space, and a scatter plot of POI embeddings is drawn. One dot represents one type of POI, and the position of the dot is determined by the coordinates after dimensionality reduction, and the color of the dot is determined by the major category to which the type of POI belongs, that is, dots of the same color belong to the same major POI category. The size of the dot is proportional to the number of this type of POI contained in the grid. The more the number, the larger the dot. The POI embedding map can visually show the relationships between different types of POIs. Among them, similar or related types of POIs are closer to each other in the figure. When a dot is clicked, the name of the corresponding type of POI, the name of the major category it belongs to, and the number of types of POIs will be displayed. According to the POI distribution within the grid, the functional attributes of the area can be analyzed. Combining with the grid heat map, the reasons for the environmental benefits generated by replacing driving with shared bicycles in this grid area can be further analyzed.

[0039] The method for evaluating and visually understanding the environmental benefits of shared bicycles according to the present invention has the following advantages: The method for evaluating and visually understanding the environmental benefits of shared bicycles proposed by the present invention first uses the Gaode Map API to obtain the cycling trajectories and driving trajectories of residents' trips; then divides the city into grids and maps the trajectories into the grids, and uses the energy conservation and exhaust gas emission reduction formulas to calculate the environmental benefits obtained by replacing driving with shared bicycles in each grid; finally, based on the urban POI dataset, a word embedding model is used to extract and measure the semantic relationships between POI types, and a POI embedding map of the grid is designed to present the functional attributes of the area, thereby assisting in analyzing the relationship between the regional function and the environmental benefits. The analysis results can enable relevant personnel to intuitively grasp the impact of land use attributes in different regions on energy conservation and emission reduction, so as to better deploy shared bicycles in the city and contribute to achieving the carbon neutral goal. Description of the Drawings

[0040] Figure 1 is a flowchart of the method for evaluating and visually understanding the environmental benefits of shared bicycles according to the present invention;

[0041] Figure 2 is the grid heat map of energy consumption savings according to the present invention;

[0042] Figure 3 is the POI embedding map of grid 59_20 according to the present invention. Detailed Embodiments

[0043] To better understand the purpose, structure, and function of the present invention, the following further describes in detail a method for evaluating and visually understanding the environmental benefits of shared bicycles according to the present invention in conjunction with the accompanying drawings.

[0044] As Figure 1 shown, the method for evaluating and visually understanding the environmental benefits of shared bicycles according to the present invention includes the following steps:

[0045] Step 1: Obtain a shared bicycle riding data set and preprocess the data.

[0046] Step 1.1: Obtain a shared bicycle riding data set within a period of time and store them in a database.

[0047] A riding record R DBS is represented as follows:

[0048] R DBS =(bID, startT, startLoc, endT, endLoc)

[0049] where bID represents the shared bicycle number, startT and startLoc are the borrowing time and location, and endT and endLoc are the returning time and location.

[0050] Step 1.2: Delete invalid data.

[0051] For all riding records, delete records with a riding distance less than 0.2 km or greater than 15 km, delete records with a time consumption less than 1 minute or greater than 3 hours, and delete records with a riding speed greater than 400 meters per minute. These records may be caused by vehicle failures, GPS positioning errors, vehicle maintenance, etc. Finally, obtain riding records with practical significance.

[0052] Step 2: Based on the starting and ending point locations of shared bicycle rides, obtain detailed riding trajectories and driving trajectories.

[0053] To accurately evaluate the environmental benefits obtained by replacing driving with cycling, it is necessary to estimate the actual distances of cycling and driving between the starting and ending points of a riding record. Use the route planning API of Amap to obtain travel trajectories considering the urban road network. For cycling, obtain the riding distance d bike , riding time, and riding trajectory. For driving, obtain the driving distance d drive , driving time, and driving trajectory. One trajectory consists of multiple line segments, reflecting the detailed cycling or driving path.

[0054] Step 3: Define the environmental benefits obtained by replacing driving with cycling based on the route trajectory.

[0055] Environmental benefits include the energy consumption saved by cycling instead of driving and the reduction in emissions of waste gases (CO 2 and NO X ). The life cycle analysis method of fuel is adopted to calculate the total energy consumption and environmental impact in the fuel production and operation stages.

[0056] The energy consumption saved by cycling instead of driving is calculated using formula (1):

[0057]

[0058] where N (unit: kg) represents the saved energy consumption. d bike (unit: km) and d drive (unit: km) represent the cycling distance and driving distance obtained from the route planning API of Gaode Map respectively. T d (unit: km) is a distance threshold. For a single trip, when the distance d bike is less than the threshold T d , there is no energy consumption for the trip. When the distance d bike is greater than the threshold T d , cycling instead of driving will produce environmental benefits. In actual calculation, T d is set to 1 km, indicating that trips with a cycling distance exceeding 1 km have the potential for energy conservation and emission reduction. p = 0.088 (unit: L / km) is the gasoline consumption per unit driving distance, and ρ = 0.72 (unit: kg / L) is the gasoline density. λ e = 0.87 and λ t = 0.95 represent the crude oil utilization rate and transportation efficiency respectively.

[0059] The reduction in waste gas emissions by cycling instead of driving is calculated using formula (2):

[0060]

[0061] where E g represents the reduction in emissions of a certain waste gas g (unit: kg). g represents CO 2 or NO X . f g (unit: kg / kg) represents the emission factor of waste gas g generated by driving a car. For CO 2 emissions, f g = 2.93, and for NO X emissions, f g = 0.006.

[0062] Step 4: Divide the city into grids, calculate the environmental benefits obtained by each grid according to the definition in Step 3, and draw a grid heat map.

[0063] First, obtain the riding trajectories of shared bikes in the entire city and remove trips with a distance less than the threshold T d to obtain eligible riding trajectories. Then, divide the city into grids of 500m * 500m and map the driving trajectories corresponding to the riding trajectories into the grids. Then, calculate the sum of the environmental benefits of all sub-trajectories falling into each grid according to formulas (1) and (2), including the energy consumption savings N, CO 2 emission reduction and NO X emission reduction By drawing a grid heat map, present the environmental benefits of each grid, including the grid heat map of energy consumption savings, the grid heat map of CO 2 emission reduction, and the grid heat map of NO X emission reduction. As Figure 2 shown, use a gradually changing color from light to dark to encode the environmental benefits of the grids. The darker the grid color, the more energy consumption is saved, or the greater the reduction in waste gas emissions, that is, the greater the environmental benefits obtained. The lighter the color, the opposite is true.

[0064] Step 5: Obtain all POI records in the city and visualize the semantic association between the functions of the analysis area and the environmental benefits of shared bikes based on the POI embeddings of different grids.

[0065] Step 5.1: Use the Gaode Map API to obtain all POI records in the city.

[0066] A POI record can be expressed as POI i = [loc i , mainType i , midType i . Where loc i represents the location of the i-th POI, and mainType i and midType i represent the major category and middle category of the i-th POI respectively. Since the analysis purpose is related to human travel, only 13 major categories and 117 middle categories that can reflect human activities and travel purposes are obtained. The 13 major categories include: "Food and Beverage Services, Shopping Services, Life Services, Sports and Leisure Services, Medical and Health Services, Accommodation Services, Scenic Spots, Commercial Residences, Government Agencies and Social Organizations, Science, Education and Culture Services, Transportation Facilities Services, Financial Insurance Services, Companies and Enterprises".

[0067] Step 5.2: Use the Word2Vec word embedding model to obtain the embedding vectors of the middle categories of POIs based on the geographical adjacency relationship between different POI types.

[0068] Learning the embedding of POI types in POI classification is analogous to learning word embeddings in a corpus. Consider each class in a POI as a word. For each class in a POI, taking it as the center, find the classes in POIs within a radius of 500 meters. Take the class in the POI corresponding to the center as the first word of the sentence, and arrange the other classes in the POIs within the circle in sequence as the following words to construct a sentence. The set composed of all classes in POIs can be regarded as a corpus. Use the Word2Vec word embedding model to calculate the embedding vectors corresponding to each class in a POI, which reflects the semantic relevance of POI types in the geographical space.

[0069] Step 5.3: For each grid, draw the POI embedding map of the grid, visualize the functional semantics of each grid, and analyze the semantic association between the regional function and the environmental benefits.

[0070] Since the obtained embedding of POI types is usually a high-dimensional vector, use the t-SNE dimensionality reduction algorithm to map it onto a two-dimensional semantic space and draw the POI embedding map in the form of a scatter plot. As Figure 3 shown, one dot represents one class in a POI. The position of the dot is determined by the coordinates after dimensionality reduction, and the color of the dot is determined by the major category to which the class in the POI belongs, that is, dots of the same color belong to the same major POI category. The size of the dot is proportional to the number of such classes in the grid. The more the number, the larger the dot. The POI embedding map can intuitively show the relationships between different classes in POIs. Among them, the points of similar or related classes in POIs are closer in the figure. When clicking on a dot, the name of the corresponding class in the POI, the name of the major category it belongs to, and the number of classes in the POI will be displayed. According to the POI distribution within the grid, the functional attributes of the area, such as residential areas, shopping places, tourist attractions, etc., can be analyzed. Combining with the grid heat map, the reasons for the environmental benefits generated by replacing car trips with shared bikes in this grid area can be further analyzed.

[0071] It can be understood that the present invention is described through some embodiments. Those skilled in the art know that without departing from the spirit and scope of the present invention, various changes or equivalent replacements can be made to these features and embodiments. In addition, under the teaching of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application belong to the scope protected by the present invention.

Claims

1. An environmental benefit evaluation and visualization understanding method for shared bicycles, characterized in that, it includes the following steps: Step 1: Obtain the shared bicycle riding data set and preprocess the data; Step 2: Based on the start and end positions of the shared bicycle rides, obtain detailed riding trajectories and driving trajectories; Step 3: Based on the route trajectories, define the environmental benefits obtained by replacing driving with cycling; The environmental benefits include the energy consumption saved by replacing driving with cycling and the reduction in exhaust gas emissions. The life cycle analysis method of fuel is used to calculate the total energy consumption and environmental impact in the fuel production and operation stages. The energy consumption saved by replacing driving with cycling is calculated using formula (1): where N represents the saved energy consumption, d bike and d drive represent the cycling distance and driving distance obtained from the route planning API of Amap respectively, T d is a distance threshold. For a trip, when the distance d bike is less than the threshold T d , there is no energy consumption for the trip. When the distance d bike is greater than the threshold T d , using cycling instead of driving will produce environmental benefits. In actual calculation, T d is set to 1 km, indicating that trips with a cycling distance exceeding 1 km have the potential for energy conservation and emission reduction. p = 0.088 is the gasoline consumption per unit driving distance, ρ = 0.72 is the gasoline density, λ e = 0.87 and λ t = 0.95 represent the crude oil utilization rate and transportation efficiency respectively; The reduction in exhaust gas emissions by replacing driving with cycling is calculated using formula (2): Among which E g represents the reduced emission amount of a certain exhaust gas g, where g represents CO 2 or NO X , f g represents the emission coefficient of exhaust gas g generated by driving a vehicle. For CO 2 emission, f g = 2.93, and for NO X emission, f g = 0.006; Step 4: Divide the city into grids, calculate the environmental benefits obtained by each grid according to the definition in Step 3, and draw a grid heat map; First, obtain the riding trajectories of shared bicycles in the whole city and remove the trips with a distance less than the threshold T d to get the riding trajectories that meet the conditions; then, divide the city into grids of 500m * 500m and map the driving trajectories corresponding to the riding trajectories into the grids; then, calculate the sum of the environmental benefits of all sub-trajectories falling into each grid according to formulas (1) and (2), including the energy consumption savings N, CO 2 emission reduction and NO X emission reduction By drawing a grid heat map, present the environmental benefits of each grid, including the grid heat map of energy consumption savings, the grid heat map of CO 2 emission reduction, the grid heat map of NO X emission reduction; Step 5: Obtain all the POI records in the city, and based on the POIs in different grids, embed and visualize the semantic association between the regional functions and the environmental benefits of shared bicycles; Step 5.1: Use the Gaode Map API to obtain all the POI records in the city; Step 5.2: Use the Word2Vec word embedding model to obtain the embedding vectors of the classes in the POIs based on the geographical adjacency relationship between different POI types; Step 5.3: For each grid, draw the POI embedding map of the grid, visualize the functional semantics of each grid, and analyze the semantic association between the regional functions and the environmental benefits.

2. The environmental benefit evaluation and visualization understanding method for shared bicycles according to claim 1, characterized in that, the specific steps of Step 1 include: Step 1.1: Obtain the shared bicycle riding data set within a period of time and store them in a database: A cycling record R DBS is expressed as follows: R DBS = (bID, startT, startLoc, endT, endLoc) where bID represents the shared bicycle number, startT and startLoc are the borrowing time and location, and endT and endLoc are the returning time and location; Step 1.2: Delete invalid data: For all the riding records, delete the records with a riding distance less than 0.2 km or greater than 15 km, delete the records with a riding time less than 1 minute or greater than 3 hours, and delete the records with a riding speed greater than 400 meters per minute. Finally, obtain the riding records with practical significance.

3. The environmental benefit evaluation and visualization understanding method for shared bicycles according to claim 1, characterized in that, the specific steps of Step 2 include: Estimate the actual distances of cycling and driving between the start and end points of a cycling record: Use the route planning API of Amap to obtain the travel trajectories considering the urban road network. For cycling, obtain the cycling distance d bike , the cycling time, and the cycling trajectory. For driving, obtain the driving distance d drive , the driving time, and the driving trajectory. One trajectory consists of multiple line segments, reflecting the detailed cycling or driving path.

4. The environmental benefit evaluation and visualization understanding method for shared bicycles according to claim 1, characterized in that, the specific steps of Step 5.1 include: A POI record can be expressed as POI i =[loc i , mainType i , midType i , where loc i represents the location of the i-th POI, mainType i and midType i respectively represent the major category and middle category to which the i-th POI belongs. Only 13 major categories and 117 middle categories that can reflect human activities and travel purposes are obtained. The 13 major categories include: "Food and Beverage Services, Shopping Services, Life Services, Sports and Leisure Services, Healthcare Services, Accommodation Services, Scenic Spots, Commercial Residences, Government Agencies and Social Organizations, Science, Education and Culture Services, Transportation Facilities Services, Financial and Insurance Services, Companies and Enterprises".

5. The environmental benefit evaluation and visualization understanding method for shared bicycles according to claim 1, characterized in that, the specific steps of Step 5.2 include: Learning the embedding of POI types in POI classification can be analogized to learning word embeddings in a corpus. Each class in a POI is regarded as a word. For each class in a POI, taking it as the center, find the classes in POIs within a radius of 500 meters. The class in the POI corresponding to the center is used as the first word of the sentence, and the other classes in POIs within the circle are arranged in sequence as the following words to construct a sentence. The set composed of all classes in POIs is regarded as the corpus, and the Word2Vec word embedding model is used to calculate the embedding vector corresponding to each class in the POI, reflecting the semantic relevance of POI types in the geographical space.

6. The method for environmental benefit evaluation and visual understanding for shared bicycles according to claim 1, characterized in that the step 5.3 includes the following specific steps: Since the obtained embedding of the POI type is usually a high-dimensional vector, the t-SNE dimensionality reduction algorithm is used to map it to a two-dimensional semantic space, and a scatter plot of POI embeddings is drawn. One dot represents one class in the POI. The position of the dot is determined by the coordinates after dimensionality reduction, and the color of the dot is determined by the major category to which the class in the POI belongs, that is, dots of the same color belong to the same major POI category. The size of the dot is proportional to the number of such classes in the grid. The more the number, the larger the dot. The POI embedding map can intuitively display the relationships between different classes in the POI. Among them, the dots of similar or related classes in the POI are closer in the figure. When a dot is clicked, the name of the corresponding class in the POI, the name of the major category to which it belongs, and the number of classes in the POI will be displayed. According to the distribution of POIs in the grid, the functional attributes of the area can be analyzed. Combining with the grid heat map, the reasons for the environmental benefits generated by replacing driving with shared bicycles in this grid area can be further analyzed.

Images