Intelligent decision-making method and system for distribution of city-level garbage cans

Abstract

The invention relates to the technical field of urban planning, in particular to an intelligent decision-making method and system for urban-level garbage can distribution, and the method comprises the steps: obtaining the multi-dimensional historical data, affecting the garbage generation amount, of each region of a city, and taking the multi-dimensional historical data as factor data; combining each piece of factor data of each region with the corresponding garbage output to serve as a training data set; constructing a garbage output prediction model based on a neural network; training the junk output prediction model by using the training data set, and learning the relationship between the factor data and the junk output; inputting the current factor data of any region of the city into the trained garbage output prediction model, and predicting the garbage output of the region; and determining the distribution condition of the intelligent garbage cans in the region according to the garbage output of the region. The distribution condition of the urban intelligent garbage cans can be more scientifically set.

Description

Technical Field

[0001] The present invention relates to the field of urban planning technology, and more specifically to a method and system for intelligent decision-making on urban garbage bin distribution. Background Art

[0002] In urban life, the trash cans set up at intervals on both sides of the road have brought great convenience to people's lives. However, due to the large number of trash cans and their wide distribution, the management of the trash cans is also difficult. At present, there are mainly the following problems in the management of trash cans:

[0003] 1) When cleaning the trash cans, the cleaning staff usually cleans them in the order in which they are arranged. The trash cans with less trash are cleaned at the same frequency as the trash cans with more trash or even overflowing. This results in high frequency but low efficiency of cleaning work.

[0004] 2) The managers of garbage areas cannot know the garbage distribution information of the entire city or region. Therefore, the garbage cans on both sides of the city roads are generally placed at equal intervals. In fact, the number of garbage cans required for different sections of the road is different depending on the frequency of use by passers-by, which leads to unreasonable planning of urban garbage cans.

[0005] Therefore, how to scientifically arrange the distribution of urban garbage bins is a problem that those skilled in the art need to solve urgently. Summary of the invention

[0006] In view of this, the present invention provides a city-level trash can distribution intelligent decision-making method and system, which can set the distribution of urban smart trash cans more scientifically.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides a city-level garbage bin distribution intelligent decision-making method, comprising:

[0009] Obtain multi-dimensional historical data on the impact of garbage generation in various areas of the city as factor data;

[0010] The data of each factor in each area is combined with the corresponding amount of garbage generated as a training data set;

[0011] Construct a neural network-based waste generation prediction model;

[0012] Use the training data set to train the garbage generation prediction model and learn the relationship between each factor data and garbage generation;

[0013] Input the current factor data of any area in the city into the trained garbage generation prediction model to predict the garbage generation in that area;

[0014] The distribution of smart trash cans in the area is determined based on the amount of garbage generated in the area.

[0015] Furthermore, each factor data includes: one or more of urban geographic information data, demographic data, industrial and commercial statistical data, traffic flow data and environmental weather data.

[0016] Furthermore, after feature extraction of each factor data, it is used as a training data set;

[0017] The urban geographic information features include: one or more of the topological structure and attribute information of the urban road network, the distribution and attribute information of urban buildings, the distribution and area information of urban green spaces, and the distribution and attribute information of urban peripheral facilities;

[0018] Demographic characteristics include: one or more of population density, population flow, population growth rate, population age structure, and population quantity characteristics;

[0019] Industrial and commercial statistical characteristics include: one or more of the following: number of enterprises, type of enterprise, and number of employees;

[0020] Traffic flow characteristics include: one or more of road capacity, traffic congestion, and public transportation line coverage;

[0021] Ambient weather characteristics include one or more of temperature, humidity, precipitation and wind speed.

[0022] Furthermore, for a certain area, the calculation formula for the number of trash cans required is:

[0023] N=G / (C×F 清运 )

[0024] Where N is the number of smart trash cans in the area, G is the amount of garbage generated in the area every day or every week predicted by the garbage generation prediction model, C is the capacity of each smart trash can, and F is the capacity of each smart trash can. 清运 Indicates the frequency of garbage collection.

[0025] Furthermore, after the smart trash cans in a certain area are distributed, it also includes:

[0026] Collect real-time data from smart trash can sensors and calculate the usage frequency of each smart trash can:

[0027] F 使用 =T 满溢 / T 周期

[0028] Among them, F 使用 represents the usage frequency of the smart trash can, T满溢 It represents the time from empty to full for the smart trash can, T 周期 Indicates the cycle of cleaning the trash can.

[0029] Furthermore, the method also includes: using a correlation analysis method or a regression analysis method to analyze the relationship between the frequency of use of the smart trash can and a certain factor data.

[0030] Furthermore, after the smart trash cans in a certain area are distributed, it also includes:

[0031] Connect the sensors of each smart trash can to the Internet of Things to remotely monitor the filling status characteristics and garbage classification characteristics of each smart trash can in each area in real time, and issue a reminder message when the capacity of the smart trash can exceeds the preset value.

[0032] Furthermore, when the smart trash cans in a certain area are distributed and run for a period of time, data analysis methods or machine learning methods are used to perform correlation analysis on the filling trend of the smart trash cans in the area during this period of time and one or more factor data to obtain the relationship between the filling trend of the smart trash cans and one or more factor data.

[0033] Furthermore, the garbage generation prediction model includes an input layer, a convolutional pooling layer, a fully connected layer and an output layer; wherein the number of nodes in the input layer is related to the number of input factor data, and the output layer sets one node or multiple nodes according to the decision-making tasks of one area or multiple areas.

[0034] In a second aspect, the present invention provides a city-level garbage bin distribution intelligent decision-making system, comprising:

[0035] A garbage generation prediction module, used to predict the garbage generation in a certain urban area using the trained garbage generation prediction model as described above;

[0036] The decision module is used to determine the distribution of smart trash cans in a certain area according to the city-level trash can distribution intelligent decision method as described above.

[0037] It can be seen from the above technical solutions that, compared with the prior art, the present invention has the following beneficial effects:

[0038] The present invention adopts a deep learning model to analyze the relationship between various environmental factors and garbage generation in a certain urban area, and predicts the garbage generation in the area based on the current environmental factor data, determines the number and distribution positions of garbage bins based on the garbage generation, and transforms the original equal-spacing setting of garbage bins into an on-demand setting, thereby improving the utilization efficiency of garbage bins and reducing the occurrence of overflow.

[0039] The present invention makes the trash cans intelligent and can remotely monitor the filling status, usage frequency, overflow and garbage classification of each intelligent trash can in real time, which is convenient for timely dispatching and route planning of garbage trucks and reducing garbage accumulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0041] Figure 1 A flow chart of the city-level garbage bin distribution intelligent decision-making method provided by the present invention;

[0042] Figure 2 This is a schematic diagram of the structure of the city-level trash can distribution intelligent decision-making system provided by the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] like Figure 1 As shown, the embodiment of the present invention discloses a city-level garbage bin distribution intelligent decision-making method, comprising the following steps:

[0045] S1. Obtain multi-dimensional historical data of the impact of garbage generation in various areas of the city as factor data;

[0046] S2, combining the data of each factor in each area with the corresponding amount of garbage generated as a training data set;

[0047] S3, building a garbage generation prediction model based on neural network;

[0048] S4, using the training data set to train the garbage generation prediction model to learn the relationship between each factor data and the garbage generation;

[0049] S5, inputting the current factor data of any area of ​​the city into the trained garbage generation prediction model to predict the garbage generation of the area;

[0050] S6. Determine the distribution of smart trash cans in the area based on the amount of garbage generated in the area.

[0051] The above steps are further described below.

[0052] S1. Obtain multi-dimensional historical data on the impact of garbage generation in various areas of the city as factor data; each factor data includes: one or more of urban geographic information data, demographic data, industrial and commercial statistics, traffic flow data and environmental weather data.

[0053] Among them, urban geographic information data includes geographic information data such as urban roads, buildings, and green spaces. The data may include information such as road network diagrams, building distribution, and green space distribution to help determine the best layout of trash cans. Building information includes information such as the location, height, type, and purpose of the building, as well as detailed data such as the structure, material, and age of the building; infrastructure information includes information such as the location, type, and capacity of infrastructure such as roads, bridges, tunnels, parks, squares, schools, hospitals, fire stations, and police stations. The data source is the urban planning department.

[0054] Demographic data, including information on population size, population density, population mobility, etc. in each area. It contains census data, annual demographic data, etc. to help determine the need for trash can layout. The data source is the government statistics bureau.

[0055] Industrial and commercial statistics include various data such as the number of enterprises, enterprise types, registered capital, operating income, profits, employment, etc. This type of statistics is collected and released by government industrial and commercial administration departments or relevant statistical agencies. It is used to monitor the development trend of industry and commerce, evaluate the business conditions of enterprises, and formulate policy plans. The data source is the industrial and commercial administration department.

[0056] Traffic flow data, including road traffic flow, traffic congestion, public transportation routes, etc. It includes road traffic monitoring data, GPS trajectory data, etc. to help identify busy traffic areas and traffic bottlenecks. The data source is the traffic management department.

[0057] Environmental weather data, including meteorological information such as temperature, humidity, wind speed, precipitation, and environmental indicators such as air quality (such as PM2.5) and noise pollution. These data are used to monitor and evaluate changes in environmental conditions such as the atmosphere, soil, and water quality. Such data help optimize decisions in waste management, urban planning, and environmental protection. The data comes from the meteorological department.

[0058] S2. After obtaining the data of each factor, it is necessary to extract its features to facilitate model training. After that, the extracted data features of various factors in each area are combined with the garbage generation amount corresponding to the area as a training data set, where the garbage generation amount can be obtained from the historical garbage collection data of the urban sanitation department.

[0059] Among them, urban geographic information features include: the topological structure and attribute information of the urban road network, including road length, road type (such as expressways, main roads, branch roads), road width, etc.; the distribution and attribute information of urban buildings, including building types (such as residential, commercial, office buildings), building density, building height, etc.; the distribution and area information of urban green spaces, including the distribution density and coverage area of ​​green spaces such as parks, gardens, and woodlands; the distribution and attribute information of urban peripheral facilities, including the location and type of facilities such as schools, hospitals, shopping centers, and transportation hubs. GIS tools or map data can be used for processing, such as extracting road length, calculating the distance to the nearest park or commercial area, etc. One or more of the above data can be used;

[0060] Demographic characteristics include: population density, flow of people, population growth rate, population age structure and one or more of the population quantity characteristics; it is possible to consider standardizing or normalizing the population quantity of each area in the city to better reflect the population distribution. Extracting population quantity characteristics from demographic data includes the total population of each area, population density, etc.

[0061] Industrial and commercial statistical features include: one or more of the number of enterprises, enterprise type and number of employees; the number of enterprises, enterprise type and number of employees are extracted from industrial and commercial data to determine whether it is a business district or commercial street. For example, an economically active area may generate more garbage and require more resources to manage garbage bins.

[0062] Traffic flow characteristics include: road capacity, such as average traffic volume on each road, traffic congestion index and other indicators; traffic congestion conditions, including average traffic volume on each road, traffic volume during peak hours, traffic congestion degree, etc.; public transportation line coverage, including bus line coverage, station distribution, bus line passenger flow, etc. One or more of the above data can be extracted.

[0063] Environmental weather characteristics include one or more of temperature, humidity, precipitation, and wind speed. High temperature and humidity will accelerate the decay of garbage, produce odors and bacteria, increase the frequency of garbage bin cleaning, and increase the need for sterilization and disinfection. Weather factors such as precipitation and wind speed will affect the placement and structural design of garbage bins to ensure that they do not tip over or overflow in bad weather. Air quality and pollution levels will affect garbage management strategies, especially in areas with severe air pollution, where more protection and monitoring equipment may be needed to ensure public health. Optimize the distribution, cleaning frequency, and maintenance of garbage bins through environmental and weather data.

[0064] S3. Build a garbage generation prediction model based on neural network.

[0065] The Keras deep learning framework can be used to build a garbage generation prediction model. The number of nodes in the input layer should match the dimension of the features. The complexity and learning ability of the model can be increased by adding hidden layers.

[0066] The model specifically includes:

[0067] 1. Input layer: Design an appropriate input layer to process the input data. For image data, it is usually a three-dimensional shape (width, height, number of channels), and for time series data, it may be a two-dimensional or three-dimensional shape (time step, feature dimension, number of channels). The input layer accepts feature data related to the amount of urban garbage generated, which may include urban geographic information data, surrounding environment, weather conditions, etc. These feature data may be two-dimensional or three-dimensional.

[0068] 2. Convolution layer and pooling layer: To improve the representation ability of the model, a Flatten layer is added before the fully connected layer to flatten the feature map. The convolution layer is used to extract feature information. The input data is convolved by a sliding convolution kernel to generate a feature map. The pooling layer is used to reduce the dimension of the feature map, reduce the amount of calculation and retain important information. Usually, the maximum pooling or average pooling operation is used.

[0069] 3. Fully connected layer for classification or regression tasks. The output of the pooling layer is flattened into a one-dimensional vector as the input of the fully connected layer. The fully connected layer is used to learn the relationship between features. It can contain multiple hidden layers and neurons, and introduce nonlinearity through activation functions.

[0070] 4. Output layer: The output layer is determined according to the task requirements. It can be one node (for predicting the amount of garbage generated in a single area) or multiple nodes (for predicting the amount of garbage generated in multiple areas).

[0071] S4. Use the training data set to train the garbage generation prediction model to learn the relationship between each factor data and the garbage generation.

[0072] For a single node, you can use the binary_crossentropy loss function; for multiple nodes, you can use the categorical_crossentropy loss function; for regression tasks, you can use the mse loss function.

[0073] Among them, an optimizer suitable for the model and data is selected to adjust the weights of the model to minimize the loss function. The Adam optimizer is a common choice, which can usually effectively adjust the learning rate and adapt to different data sets and model structures.

[0074] Select the appropriate loss function according to the task type of the model. For regression tasks, the mean_squared_error loss function can be used because the goal is to predict the specific amount of garbage generated in each urban area, which is a regression task. You can use the mean square error (MSE) as the loss function and select a suitable optimizer.

[0075] Select one or more metrics to evaluate the model performance. For classification problems, common evaluation metrics include accuracy, precision, recall, and F1 score.

[0076] S5. Input the current factor data of any area of ​​the city into the trained garbage generation prediction model to predict the garbage generation of the area; after the model is trained, the garbage generation of one or more areas of the city can be predicted.

[0077] S6. Determine the distribution of smart trash cans in the area based on the amount of garbage generated in the area, including the required number of trash cans and their installation locations.

[0078] The formula for calculating the number of trash cans required for a certain area is:

[0079] N=G / (C×F 清运 )

[0080] Where N is the number of smart trash cans in the area, G is the amount of garbage generated in the area every day or every week predicted by the garbage generation prediction model, C is the capacity of each smart trash can, and F is the capacity of each smart trash can. 清运 Indicates the frequency of garbage collection.

[0081] As for the installation location of the trash can, we can combine the road network map, building distribution, green space distribution and other information to help determine the best layout location for the trash can.

[0082] In one embodiment, after the smart trash cans in a certain area are distributed, the method further includes:

[0083] Collect real-time data from smart trash can sensors and calculate the usage frequency of each smart trash can:

[0084] F 使用 =T 满溢 / T 周期

[0085] Among them, F 使用 represents the usage frequency of the smart trash can, T 满溢 It represents the time from empty to full for the smart trash can, T 周期 Indicates the cycle of cleaning the trash can.

[0086] In one embodiment, a correlation analysis method or a regression analysis method is used to analyze the relationship between the usage frequency of the smart trash can and certain factor data (such as population density, environment, etc.).

[0087] Suppose you want to analyze the frequency of use of the trash can F 使用 With population density D 人口 The commonly used analysis method is Pearson correlation coefficient or linear regression:

[0088] Correlation = Covariance (F 使用 ,D 人口 ) / σ F使用 ·σ D人口

[0089] Through calculation, we can determine whether the frequency of trash can use is significantly correlated with population density, and then optimize the trash can configuration strategy.

[0090] You can also use a logistic regression model to determine whether a certain area needs to be equipped with trash cans. Assuming that two features need to be collected, namely population density and surrounding environment score, the problem can be modeled as a binary classification problem, where 1 indicates that it is suitable to set up trash cans and 0 indicates that it is not suitable to set up trash cans.

[0091] The logistic formula of the logistic regression model can be expressed as:

[0092] P(y=1|x)=1 / (1+e -(b0+b1x1+b2x2 ))

[0093] Among them, P(y=1|x) represents the probability of y=1 given the feature x, b0, b1, b2 are the coefficients of the logistic regression model, and x1, x2 are the feature values.

[0094] In one embodiment, after the smart trash cans in a certain area are distributed, the method further includes:

[0095] Connect the sensors of each smart trash can (such as pressure sensor, weight sensor, infrared sensor, etc.) to the Internet of Things to remotely monitor the filling status and garbage classification characteristics of each smart trash can in each area in real time, and perform timely processing and scheduling. At the same time, when the capacity of the smart trash can exceeds the preset value, a reminder message will be issued.

[0096] The collection routes of garbage trucks can also be scheduled and optimized according to the overflow status of garbage bins in various areas of the city, so as to achieve optimal garbage collection route planning, reduce travel and time costs, improve efficiency and save resources.

[0097] When the smart trash cans in a certain area are distributed and run for a period of time, data analysis methods or machine learning methods are used to perform correlation analysis on the filling trend of the smart trash cans in the area during this period of time and one or more factor data, so as to obtain the relationship between the filling trend of the smart trash cans and one or more factor data, so as to predict the future filling status of the trash cans in a certain area.

[0098] At the same time, reminders for waste sorting and recycling can be sent to citizens to promote environmental awareness and behavioral changes.

[0099] In general, the layout principles of smart trash cans must meet the requirements of coverage area, service scope, spacing, etc., to ensure that the trash cans can meet the needs of surrounding residents and businesses. Density control: According to the needs and characteristics of different areas, control the density of trash cans to avoid waste of resources or visual pollution caused by dense layout. Intelligent management: Use intelligent technology to achieve real-time monitoring and management of trash cans, including filling status monitoring, positioning services, data analysis and other functions to optimize the layout and scheduling of trash cans. Classification setting: According to the garbage classification policy and actual situation, set up different types of trash cans, including recyclables, hazardous waste, wet garbage, etc., to promote garbage classification and resource recycling. Environmentally friendly: Consider the material, design and color of the trash can to make it consistent with the surrounding environment and enhance the overall aesthetics and environmental quality of the city. Regular evaluation: Regularly evaluate and adjust the layout and management of trash cans, optimize and improve according to actual conditions, and ensure the long-term effectiveness and adaptability of trash can planning and management.

[0100] like Figure 2 As shown, the embodiment of the present invention also provides a city-level garbage bin distribution intelligent decision-making system, including:

[0101] A garbage generation prediction module, used to predict the garbage generation in a certain urban area using the trained garbage generation prediction model as described above;

[0102] The decision module is used to determine the distribution of smart trash cans in a certain area according to the city-level trash can distribution intelligent decision method as described above.

[0103] Specifically, the garbage generation prediction module receives one or more factor data (urban geographic information data, demographic data, industrial and commercial statistics, traffic flow data and environmental weather data) of one or more regions, and outputs the garbage generation of the corresponding region.

[0104] The decision-making module mainly implements the overflow management and planning management of garbage bins. Specifically, it calculates the number of smart garbage bins required according to the amount of garbage generated in each area, and determines the best installation location of the garbage bins based on the city's geographic information data. At the same time, based on the sensor data of the smart garbage bins, its filling status and garbage classification characteristics are monitored and warned in real time, and the routes of garbage trucks are dispatched and optimized according to the overflow status of garbage bins in each area.

[0105] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0106] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A city-level garbage bin distribution intelligent decision-making method, characterized in that: include: Obtain multi-dimensional historical data on the impact of garbage generation in various areas of the city as factor data; The data of each factor in each area is combined with the corresponding amount of garbage generated as a training data set; Construct a neural network-based waste generation prediction model; Use the training data set to train the garbage generation prediction model and learn the relationship between each factor data and garbage generation; Input the current factor data of any area in the city into the trained garbage generation prediction model to predict the garbage generation in that area; The distribution of smart trash cans in the area is determined based on the amount of garbage generated in the area.

2. The city-level garbage bin distribution intelligent decision-making method according to claim 1 is characterized in that: The data of each factor includes one or more of: urban geographic information data, demographic data, industrial and commercial statistical data, traffic flow data and environmental weather data.

3. The city-level garbage bin distribution intelligent decision-making method according to claim 1 is characterized in that: After feature extraction of each factor data, it is used as training data set; The urban geographic information features include: one or more of the topological structure and attribute information of the urban road network, the distribution and attribute information of urban buildings, the distribution and area information of urban green spaces, and the distribution and attribute information of urban peripheral facilities; Demographic characteristics include: one or more of population density, population flow, population growth rate, population age structure, and population quantity characteristics; Industrial and commercial statistical characteristics include: one or more of the following: number of enterprises, type of enterprise, and number of employees; Traffic flow characteristics include: one or more of road capacity, traffic congestion, and public transportation line coverage; Ambient weather characteristics include one or more of temperature, humidity, precipitation and wind speed.

4. The city-level garbage bin distribution intelligent decision-making method according to claim 1 is characterized in that: The formula for calculating the number of trash cans required for a certain area is: N=G / (C×F 清运 ) Where N is the number of smart trash cans in the area, G is the amount of garbage generated in the area every day or every week predicted by the garbage generation prediction model, C is the capacity of each smart trash can, and F is the capacity of each smart trash can. 清运 Indicates the frequency of garbage collection.

5. The city-level garbage bin distribution intelligent decision-making method according to claim 4 is characterized in that: When the smart trash cans in a certain area are distributed, it also includes: Collect real-time data from smart trash can sensors and calculate the usage frequency of each smart trash can: F 使用 =T 满溢 / T 周期 Among them, F 使用 represents the usage frequency of the smart trash can, T 满溢 It represents the time from empty to full for the smart trash can, T 周期 Indicates the cycle of cleaning the trash can.

6. The city-level garbage bin distribution intelligent decision-making method according to claim 5 is characterized in that: It also includes: using correlation analysis method or regression analysis method to analyze the relationship between the frequency of use of the smart trash can and a certain factor data.

7. The city-level garbage bin distribution intelligent decision-making method according to claim 1 is characterized in that: When the smart trash cans in a certain area are distributed, it also includes: Connect the sensors of each smart trash can to the Internet of Things to remotely monitor the filling status characteristics and garbage classification characteristics of each smart trash can in each area in real time, and issue a reminder message when the capacity of the smart trash can exceeds the preset value.

8. The city-level garbage bin distribution intelligent decision-making method according to claim 1 is characterized in that: When the smart trash cans in a certain area are distributed and run for a period of time, data analysis methods or machine learning methods are used to perform correlation analysis on the filling trend of the smart trash cans in the area during this period of time and one or more factor data to obtain the relationship between the filling trend of the smart trash cans and one or more factor data.

9. The city-level garbage bin distribution intelligent decision-making method according to claim 1 is characterized in that: The garbage generation prediction model includes an input layer, a convolutional pooling layer, a fully connected layer and an output layer; the number of nodes in the input layer is related to the number of input factor data, and the output layer sets one node or multiple nodes according to the decision-making tasks of one area or multiple areas.

10. A city-level garbage bin distribution intelligent decision-making system, characterized in that: include: A garbage generation prediction module, used to predict the garbage generation in a certain urban area using the trained garbage generation prediction model according to any one of claims 1 to 9; A decision module is used to determine the distribution of smart trash cans in a certain area according to the city-level trash can distribution intelligent decision method as described in any one of claims 1-9.

Images