8 results about "Rail freight transport" patented technology

The application discloses a kind of railway freight vehicle key component real-time detection method and system based on Transformer, the method includes: S1, the original image of railway freight vehicle key component to be detected is obtained;S2, the original image of railway freight vehicle key component after pre-processing is carried out feature extraction, and multiple scale feature maps are obtained;S3, the edge texture information of small scale component in deep multi-scale feature map is extracted and strengthened, and the feature map after strengthening is obtained;S4, the feature map after strengthening is input as first branch, shallow multi-scale feature and middle multi-scale feature are input as second branch, and splicing and fusion are carried out, and the fusion feature map is obtained;S5, the positioning loss of the boundary box of fusion feature map is calculated, to realize the accurate detection of railway freight vehicle key component;The application realizes the real-time accurate detection of railway freight vehicle key component by the multi-scale feature fusion of lightweight dynamic hybrid convolution and polarized self-attention mechanism and Inner-Shape-IoU loss optimization.

The application discloses a railway freight safety evaluation method and system based on data analysis, and relates to the field of railway freight, and comprises the following steps: an uploading module is used for uploading and storing attribute data of transported goods corresponding to a current railway freight task; a calibration module is used for traversing the attribute data of goods stored in the uploading module, and evaluating and calibrating an upper limit value of a transport speed corresponding to the freight task based on the attribute data of goods; and the application can accurately evaluate the upper limit of the freight speed and the optimal driving speed of each section according to the characteristics of goods, the load distribution of the carriage and the real-time road conditions of the line, smoothly transition when the speed is changed, effectively reduce the risk of goods overturning and deformation, and effectively improve the safety of the whole railway freight.

The invention relates to the technical field of tank car welding equipment, and provides railway freight tank car welding equipment which can adapt to the matched welding working condition of a railway freight tank car, achieves supporting and welding integrated welding operation, effectively improves the overall welding construction efficiency, can replenish and fill sintered flux in real time in the operation process, and is stable and reliable in welding operation. The welding quality is stable and controllable, position fine adjustment can be conducted at any time, adaptability is high, practical performance is better, the welding trolley comprises a welding trolley and a supporting auxiliary structure, a penetrating shaft and two auxiliary wheels are rotationally installed on the welding trolley, follower wheels are fixedly connected to the two ends of the penetrating shaft, and the supporting auxiliary structure comprises a fixed supporting frame and a movable supporting frame; the fixed supporting frame is fixedly connected with a supporting rail, the movable supporting frame is connected with the supporting rail in a sliding mode, the fixed supporting frame and the movable supporting frame are each provided with a driving supporting wheel set and an auxiliary supporting wheel set, the welding trolley is fixedly connected with a supporting main frame, and the supporting main frame is fixedly connected with an assembling arm.

This application provides a method, device, storage medium, and program product for predicting railway freight volume. The method includes: acquiring multiple feature data sequences, each feature data sequence including the same feature data corresponding to multiple consecutive time intervals, the feature data including historical railway freight volume data and railway freight volume influencing factor data; inputting the multiple feature data sequences into a prediction model, the prediction model using the feature data sequences as data processing units to process data, and outputting a prediction result for railway freight volume data at a future preset time. In this embodiment, by forming feature data sequences from the same feature data corresponding to multiple consecutive time intervals, and by inputting the feature data sequences as data processing units into the prediction model, the model can learn the relationship between different feature sequences, understand the complex changing patterns of railway freight volume influenced by multiple factors, and thus improve the prediction accuracy of the model.

This invention discloses an edge computing-based autonomous decision-making and real-time communication system and method for loading depots, relating to the field of railway freight automation. The system includes: deploying edge computing nodes to construct an end-to-end distributed processing architecture, collecting real-time data and performing localized processing; utilizing an autonomous decision-making module to construct a closed-loop mechanism of "state awareness - strategy reasoning - instruction output" based on uploaded multi-source data, autonomously generating material unloading control instructions and multi-device collaborative strategies; and employing a real-time communication protocol integrating Time-Sensitive Networking (TSN) and 5G URLLC technologies to ensure real-time communication between edge computing nodes, between edge computing nodes and the autonomous decision-making module, and when equipment executes instructions, achieving synchronization between locomotive arrival and material unloading initiation, and seamless connection between leveling machine operation and material unloading completion. This invention can improve the unmanned, highly collaborative, and low-latency capabilities of loading operations, thereby better adapting to the needs of the "heavy-load" transportation mode in railway freight transport.

The present application relates to a kind of railway freight loading state image recognition method and system.Current freight car loading state anomaly detection method has the problem of low detection accuracy.This method is based on railway freight vehicle abnormal image to establish abnormal data set, and is enhanced and expanded;Introduce bidirectional weighted BiFPN structure in YOLOv5 network model;Coordinate attention mechanism module is embedded in bidirectional weighted BiFPN structure;Abnormal data set is input into model for training;Target detection data set is input into the model after training for railway freight vehicle anomaly recognition.The present application uses weighted bidirectional feature pyramid to capture information on context, improves the fusion degree of different weight multiscale features, uses embedded improved coordinate attention module, makes receptive field more focused on interesting area, maintains the integrity of features, significantly improves detection accuracy, achieves the purpose of reducing operator labor intensity and improving freight inspection work quality.

This invention relates to the field of railway freight positioning technology, specifically a railway freight positioning method and system based on BeiDou and multi-sensor fusion. The method includes: constructing a dynamic positioning observation model with BeiDou coordinates, inertial measurement unit output, wheel axle speedometer output, and environmental monitoring data as initial inputs; performing spatiotemporal consistency registration on multi-source heterogeneous positioning data to form a spatiotemporally synchronized observation sequence; establishing a positioning information flow and constructing a positioning observation confidence distribution map, separating high-confidence and low-confidence observation information; iteratively calibrating the dynamic positioning observation model using high-confidence observation information; reconstructing low-confidence observation information using the calibrated model, fusing the two to generate a globally consistent fused positioning result; performing a continuity check on the fused positioning result, and outputting the final positioning information. This method can improve the consistency and effectiveness of railway freight positioning and is adaptable to complex operating scenarios.

The application discloses a railway freight multimodal intelligent security check method and system, relates to the technical field of railway freight, and comprises the following steps: before a train arrives at a station, based on in-transit monitoring data and freight document information, a pre-check risk assessment result for each carriage is generated; when the train arrives at the station, according to the guidance of the pre-check risk assessment result, a plurality of non-destructive detection devices are synchronously triggered to scan the train, and spatiotemporally aligned multimodal detection data are generated; and for the multimodal detection data, a differentiated analysis strategy is executed according to the pre-check risk assessment result, multimodal feature fusion is performed on high-risk targets, and a deep analysis result list containing suspicious targets and their features is generated. The application forms a complete business closed loop from intelligent early warning to accurate disposal, and realizes the leap of overall operation efficiency while improving the safety margin.