Methods for construction planning of charging piles in the smart cities and internet of things systems thereof

An Internet of Things system for construction planning of a charging pile in a smart city, including a user platform, a service platform, a management platform, a sensor network platform, and an object platform;

• the user platform being configured as a terminal device, the terminal device including a mobile device and a tablet computer;
• the management platform including a processing device, the processing device including a server or a server group;
• the sensor network platform including a plurality of sensor network sub-platforms;
• different regions to be expanded correspond to different the sensor network sub-platforms;
• the sensor network platform being configured as a communication network and a gateway, the plurality of sensor network sub-platforms being configured as independent gateways;
• the object platform being configured to include at least one charging pile, the at least one charging pile including a positioning device and a camera device;
• the user platform being configured to issues a query instruction of urban charging pile construction planning to the service platform;
• the service platform issues the query instruction of urban charging pile construction planning to the management platform, and the service platform receives an urban charging pile construction planning scheme uploaded by the management platform;
• the management platform including a general database and the plurality of management sub-platforms;
• the sensor network platform being configured to, in response to a query instruction issued by a management sub-platform of the management platform, obtain charging pile related data from a corresponding region in the object platform through a corresponding sensor network sub-platform and upload the charging pile related data to a corresponding management sub-platform of the management platform;
• the management platform includes a general database of the management platform and a plurality of management sub-platforms;
• the object platform is configured to obtain region feature of the region to be expanded;
• the region feature of the region to be expanded being obtained based on the object platform and uploaded to the corresponding management sub-platform based on a sensor network sub-platform corresponding to the region to be expanded;
• the sensor network sub-platform is configured to obtain the region feature of the corresponding region to be expanded based on the object platform, and upload the region feature to the corresponding management sub-platform;
• the management sub-platform is configured to perform operations including:
• determining at least one candidate construction site based on the region feature, wherein the region feature at least includes distribution of people flow in the region to be expanded and distribution of existing charging piles in the region to be expanded;
• determining at least one target construction site based on the at least one candidate construction site, comprising:
• constructing a first candidate feature map based on the at least one candidate construction site, wherein the first candidate feature map includes nodes and edges, the nodes correspond to preset facilities in the region to be expanded, the edges represent roads between the preset facilities corresponding to the nodes, a feature of each of the edges includes a length and convenience of a road corresponding to the edge, and the convenience includes an average spending time;
• retaining at least a part of first nodes that satisfy a preset node condition in the first candidate feature map, deleting remaining first nodes and corresponding edges, and obtaining a second candidate feature map based on a modified first candidate feature map, the preset node condition including:
• an expected average queuing time not exceeding a time threshold, and an expected electric vehicle flow not being lower than a preset flow threshold;
• wherein the expected electric vehicle flow and the expected average queuing time of each of the at least a part of first nodes are predicted through a second prediction model, the second prediction model is a graph neural network (GNN) model;
• an input of the second prediction model is the first candidate feature map, and an output of the second prediction model includes expected average queuing times and expected electric vehicle flows of candidate construction sites corresponding to all first nodes in the first candidate feature map;
• the output of the second prediction model is obtained based on an output of the nodes, and the expected electric vehicle flow is characterized by an estimated monthly average flow or an annual average flow;
• the second prediction model is trained by a plurality of training samples with labels, comprising:
• inputting the plurality of training samples with labels into an initial second prediction model, constructing a loss function by using the labels and results of the initial second prediction model, and iteratively updating parameters of the initial second prediction model based on the loss function;
• when the loss function of the initial second prediction model satisfies a preset condition, completing model training to obtain a trained second prediction model, the preset condition including that the loss function converges or a number of iterations reaches a threshold;
• and
• determining the at least one target construction site based on at least one second candidate feature map;
• and
• transmitting at least one target construction site to the service platform through the general database of the management platform;
• the service platform is configured to upload the at least one target construction site to the user platform;
• and
• the user platform receiving the urban charging pile construction planning scheme uploaded by the service platform and feeding back a construction planning result of urban charging piles to a user through the terminal device.