5results about How to "Improve digging effect" patented technology

The invention belongs to the technical field of dredging engineering, and discloses a method for calculating a standardized drag head excavation model of a drag suction dredger, which comprises the following steps: determining a standardized drag head corresponding to the drag suction dredger based on water area position information of a current water area, and determining corresponding rake tooth information based on the standardized drag head; the rake tooth information and the water area position information calculate the mechanical ground breaking amount in unit time and a corresponding rake tooth ground breaking parameter table based on a mechanical ground breaking amount calculation model; on the basis of high-pressure water outlet characteristic information corresponding to the trailing suction dredger, the hydraulic ground breaking amount in unit time and a corresponding high-pressure water outlet characteristic table are calculated on the basis of a hydraulic ground breaking amount calculation model according to the high-pressure water outlet characteristic information; the hydrological environment information is divided into dredging influence grades according to the dredging requirements of the to-be-dragged suction area through a fuzzy function; and a multi-factor coupling excavation yield calculation model is established, and the excavation yield of the drag suction dredger under the action of the standardized drag head is obtained by considering the environmental constraint of the dredging influence level on the mechanical ground breaking amount and the hydraulic ground breaking amount in unit time.

The invention discloses a remote sensing image weak supervision target detection method based on weight redistribution and task alignment. The method mainly solves the problems that neighborhood targets are misclassified, only the most significant region of the targets is concerned, and classification branches are inconsistent with regression branches in an existing detection scheme. According to the scheme, the method comprises the steps of introducing high-quality positive instance mining to obtain seed instances, and determining neighborhood instances and background instances according to space and feature similarity between the instances; by calculating a spatial relation matrix between positive instances, screening instances covering a complete target, and completing weight redistribution of the positive instances; meanwhile, the initial instability of a weak supervision target detection model is considered, task alignment frame regression loss is introduced, the weight is dynamically adjusted by designing a task alignment modulation factor, and then the modulation factor and the difficulty degree of an instance are combined to serve as the weight of a regression branch. According to the method, misclassification of neighborhood instances can be avoided, the whole target is concerned, the coupling of classification and regression branches is improved, and the detection performance is effectively improved.

The present application belongs to the technical field of dredging engineering, and discloses a method and system for calculating the multi-factor coupled excavation yield of a drag suction dredger. Specifically, the present application determines the corresponding tooth information of the drag suction dredger based on the position information of the current water area; substitutes the tooth information and the position information of the water area into a mechanical soil breaking amount calculation model to calculate the mechanical soil breaking amount per unit time and generate a corresponding tooth soil breaking parameter table; and substitutes the high-pressure water discharge characteristic information of the drag suction dredger into a hydraulic soil breaking amount calculation model to calculate the hydraulic soil breaking amount per unit time and generate a corresponding high-pressure water discharge characteristic table. The hydrological environment information is processed by a fuzzy function, and the dredging influence level is divided according to the dredging demand of the drag suction area. Under uncertain external environment, the multi-factor coupled excavation yield calculation model is established through parameter identification learning, which takes into account the environmental constraints of the dredging influence level on the mechanical soil breaking amount and the hydraulic soil breaking amount per unit time, and finally calculates the excavation yield of the drag suction dredger.

The utility model relates to the technical field of civil engineering, in particular to a civil engineering foundation pit construction excavation device which comprises a bucket body, and first fixing blocks are fixed to the two sides of the bucket body. The reinforcing plate replacing assembly is arranged at the bottom end of the bucket body; and the digging tooth replacing assembly is arranged at one end of the bottom of the bucket body. According to the excavator bucket, the bucket body and the multiple sets of excavating tooth replacing assemblies are installed and used in cooperation, and the specific excavating tooth body can be adjusted and replaced according to the actual situation so that the optimal excavating effect can be kept. The bucket body, the first fixing block and the reinforcing plate replacement assembly are matched for use, so that each group of reinforcing plate bodies can be replaced and fixed, the abraded and damaged reinforcing plate bodies can be easily replaced, and the working continuity and efficiency are improved.

The invention discloses a map POI knowledge graph task intelligent mining method based on a large language model. The method comprises the steps of obtaining map POI data; based on the map POI data, a finetune data set is constructed according to the task type, and data amplification is carried out on the finetune data set; carrying out large language model pre-training through the amplified data set, then carrying out SFT retraining and carrying out reward learning optimization; reinforcement learning is carried out on the large language model based on a PPO algorithm; and processing map POI data based on the trained large language model to carry out knowledge graph task mining. According to the method, the unstructured data of the map POI can be processed, and various tasks of the map POI data can be processed.