4results about How to "Improve travel efficiency" patented technology

The invention provides an urban circle traffic demand prediction method based on a super network. The method comprises the following steps: constructing a multi-level, multi-attribute and three-dimensional interconnected super network covering highways, subways and railways in an urban circle; constructing a multimode generalized cost function based on the super network; dividing a comprehensive traffic zone, generating a passenger transport demand starting point and terminal point OD matrix, and constructing a travel mode division model considering multimode generalized cost and environmental influence; and based on the travel mode division model, traffic flow distribution is carried out by using a user balanced traffic flow distribution algorithm based on a quasi-Newton method, and a traffic flow prediction result of the urban circle is obtained. According to the method, a solution is provided for efficiently constructing a digital base of a traffic transportation industry vertical domain large model, improving the traffic mode division precision and the like, the bottleneck restricting the improvement of urban circle travel efficiency can be deeply analyzed based on the traffic demand prediction result, and policy suggestions are provided for constructing an efficient urban circle traffic transportation system.

The application relates to the technical field of urban bus line network and timetable integration optimization, and particularly discloses a bus line network and timetable integration optimization method and device based on graph reinforcement learning. The method is used for generating a bus line network structure and a timetable in a multi-mode public transport network containing BRT, subway and shared bicycles. Firstly, a multi-mode public transport road network is constructed, and passenger flow demand data is obtained; a bus line network design and a synchronous timetable compilation process are modeled as a Markov decision process; according to a current bus line network state, the road network is reconstructed into a heterogeneous graph containing time sequence evolution arcs and spatial topology arcs, and a double attention graph neural network is used to extract candidate node features; according to the node features, an action is selected through an action network, and the bus line network state is updated; the node feature extraction, the action selection are repeated until a termination condition is met, and the bus line network structure and the line timetable are output. The application can generate a bus line network and timetable which are more coordinated with other public transport modes, and reduce the operation cost of bus enterprises and the travel and transfer time of passengers.

The application discloses a compact micro-bubble air floatation device for removing oil from oil and gas field production water, and belongs to the technical field of micro-bubble air floatation machines, which comprises a main tank body, wherein an oscillating assembly and a cleaning assembly are arranged in the main tank body, a main rotating shaft is rotationally connected to the main tank body through a bearing, the main rotating shaft is in transmission connection with the oscillating assembly and the cleaning assembly, and a solid-removing assembly is fixedly connected to the bottom of the main tank body. In the application, the oscillating assembly is arranged, the main rotating shaft drives a plurality of extrusion blocks to rotate through a connecting ring, the extrusion blocks drive a sliding rod to make a reset movement in cooperation with a reset spring, thereby realizing the reciprocating movement of the sliding rod, the sliding rod drives the baffle to reciprocate, the baffle drives the high-pressure spray pipe to reciprocate through a connecting rod, thereby expanding the range of the high-pressure spray pipe for spraying bubbles, enabling the bubbles to more fully fill the main tank body, and enabling the bubbles to more comprehensively adhere to oil and grease impurities, thereby improving the oil removal efficiency.

This invention relates to the field of robotics, and more particularly to a gas-driven variable-cell robot for disaster relief and its detection method. The robot comprises: a torso module, which includes several torso links rotatably connected end-to-end and a walking unit connected to the torso links. The walking unit includes a hip joint rotatably connected to the torso links, a thigh rotatably connected to the hip joint, a lower leg rotatably connected to the thigh, and a foot connected to the lower leg; a gas-driven module, which includes a hip joint rotary cylinder disposed on the torso links, and the hip joint is driven and connected to the hip joint rotary cylinder; a detection module disposed on the torso links; and an explosion-proof energy module installed on the torso links. This invention solves the technical problem of poor robot morphology and environmental adaptability in the prior art, and improves the robot's adaptability to complex environments.