15results about How to "Reduce redundant data" patented technology

The present application relates to coke oven operation control technical field, especially to a kind of applied to electric heating coke oven expansion pressure detection method, comprising: based on the coal material state change in the historical coking cycle of electric heating coke oven division coking stage, and set reference collection frequency for each coking stage;For each coking stage, based on reference collection frequency, real-time data acquisition is carried out, and expansion pressure detection data are obtained;Expansion pressure detection data are calculated for pressure anomaly, and expansion pressure change rate and expansion pressure distribution characteristics are obtained;And according to this, reference collection frequency is revised, and the dynamic collection frequency corresponding to the coking stage is obtained;In response to at least one of expansion pressure detection data, change rate, distribution characteristics and dynamic collection frequency exceeds corresponding preset warning threshold, expansion pressure detection alarm information is generated.It can make collection frequency dynamic adjustment with coal material expansion characteristics, ensure the timeliness of pressure peak capture, reduce redundant data in stable stage.

The invention provides a document synchronization method and device, a storage medium and electronic equipment. The method comprises the steps of obtaining first document content of a to-be-synchronized target local document, and obtaining a document content version of a target cloud document corresponding to the target local document in current cloud storage; second document content is determined based on the document content version, and a first document change operation set is determined according to the difference between the first document content and the second document content; generating a first document synchronization instruction according to the first document change operation set; and sending the first document synchronization instruction to the target server, so that the target server updates the content of the target cloud document based on the first document synchronization instruction. Only the change data is transmitted to the server, redundant data and meaningless time consumption generated by full-amount synchronization are reduced, the change operation is re-determined based on the document content indicated by the cloud document version, data loss or coverage errors of the cloud document are avoided, and data consistency and integrity in the data synchronization process are improved.

The invention relates to the technical field of artificial intelligence image recognition and data mining, and discloses an image data mining processing system and method combining a visual large model and a visual small model. The method aims at solving the problems that in the prior art, the image recognition recall rate and accuracy are difficult to consider, the calculation cost and the processing speed are unbalanced, the data management efficiency is low, and the model generalization ability is insufficient. The system comprises a data input module, a frame extraction deduplication module, a picture library construction module, a visual large model judgment module, a visual small model judgment module, a label preprocessing module, a manual inspection module, a training data generation module, a model training module and a data security module. According to the method, the generalization ability is guaranteed through preliminary screening of an LPM (Deic) large model, missing detection is complemented through secondary verification in cooperation with a YOLO small model, false alarms are eliminated in combination with manual verification, meanwhile, double models are optimized through targeted sample training, the pain point of missing detection or misjudgment of a single model is effectively solved, and accurate and comprehensive consideration of target recognition is achieved.

The invention relates to the technical field of data processing, in particular to a multi-dimensional automobile accessory big data processing method and system based on a vertical class model, and the method comprises the steps: obtaining multi-dimensional data information of automobile accessories at regular time; according to key text information in the multi-dimensional data information, constructing a text feature vector of the automobile accessory; determining a first similarity between each attribute of each automobile accessory and the automobile accessory with the same name; obtaining the attribute importance degree of each attribute of each automobile accessory; based on a vertical class model, evaluating the similarity degree of the text feature vectors between any two automobile accessories obtained in a single time, and classifying the automobile accessories; and determining a first similarity between each attribute of any automobile accessory in the same kind of automobile accessories and the remaining automobile accessories, and calculating the automobile accessory similarity between the any automobile accessory and the remaining automobile accessories by combining the attribute importance of each attribute of the any automobile accessory so as to perform duplicate removal processing on the same kind of automobile accessories. Therefore, the duplicate removal effect of the automobile accessory data is improved.

The invention relates to a distributed optical fiber bridge foundation deformation monitoring system and a construction and monitoring method, the distributed optical fiber bridge foundation deformation monitoring system comprises a top optical fiber tube unit, a middle optical fiber tube unit and a bottom optical fiber tube unit, and a connecting optical fiber and a free optical fiber penetrate out of the top of the top optical fiber unit and are respectively connected with a BOTDA (Brillouin Optical Time Domain Analysis) demodulator. A bridge foundation deformation monitoring system is pre-buried by drilling holes in the ground of a bridge foundation area, millimeter-level-precision pile foundation deformation continuous monitoring is achieved through parameter changes, geometric transformation and mechanical transformation of optical fiber signals, and data are stable and reliable; the monitoring system in an optical fiber form is buried in a soil body, so that the pile foundation structure can be prevented from being damaged; meanwhile, the method is suitable for a newly-built bridge and an existing bridge, supports the adjustment of the monitoring range as required in an optical fiber laying area, reduces redundant data, and effectively improves the monitoring efficiency.

The application discloses a bridge structure self-adaptive sampling monitoring method and device suitable for an extreme road environment, and belongs to the technical field of bridge structure health monitoring.The core target of the application is to solve the problems of repeated arrangement of existing monitoring equipment, road-bridge data fragmentation, and low self-adaptive sampling regulation precision.Through innovative design of a road-bridge multiplexing integrated sensing node, single-node multi-parameter and full-scene monitoring are realized, and the arrangement cost and construction difficulty are reduced; through construction of a self-adaptive sampling strategy based on fuzzy logic control, a feasible algorithm model is given, continuous dynamic regulation of sampling parameters is realized, monitoring precision and low power consumption are taken into account, deep collaborative linkage of road environment monitoring and bridge structure monitoring is realized, and the safe operation of the bridge structure under the extreme road environment is ensured.

A lead screw detection device disclosed by the present invention comprises a detection cabinet body, the detection cabinet body is provided with a control host, the control host is internally provided with a driving assembly, the control host is connected with a clamping mechanism driven by the driving assembly to rotate, and the clamping mechanism is used for clamping one end of a lead screw piece and driving the lead screw piece to rotate; a supporting column is fixed in the detection cabinet body, a horizontal adjusting mechanism and a pneumatic control assembly are arranged on the supporting column, the horizontal adjusting mechanism is connected with a supporting detection block, and electric sliding rails with grating rulers are symmetrically fixed on the supporting detection block; a first sliding plate and a second sliding plate are arranged on the electric sliding rail in a sliding mode, a detection mechanism is arranged on the first sliding plate, and a fixing mechanism is arranged on the second sliding plate. The invention belongs to the technical field of lead screw detection, and particularly relates to a lead screw detection device.

This application discloses a memory management method, related devices, and program products for a multi-agent interaction system, relating to the field of artificial intelligence technology. The method includes: responding to a user-initiated task command, calling a multi-agent ensemble to infer the task result, and feeding the task result back to the user; acquiring memory data generated by each agent in the multi-agent ensemble during the inference process; and storing the memory data of each agent hierarchically according to task memory level, dialogue memory level, and state memory level. A single task memory includes one or more rounds of dialogue memory related to a single task processed by the agent, and a single dialogue memory includes one or more state memories related to one round of dialogue by the agent. The state memory is used to record the state data of the model inference process called by the agent in one round of dialogue. This application provides more flexible memory data management, offering more flexible, accurate, and efficient context management and organization capabilities in human-machine and multi-agent collaboration processes.

The invention discloses a mining lithium battery locomotive data monitoring device and method, and relates to the technical field of data monitoring, and the device comprises a gateway collection module which collects multi-source initial data of a mining lithium battery locomotive, and generates a standardized multi-source locomotive signal after analysis, data verification and preprocessing through an analysis protocol algorithm; the data transmission module combines and packages the multi-source locomotive signals; the monitoring display module analyzes the packaged multi-source locomotive signals, and the decoded locomotive signals are displayed through the monitoring display module; and the host computer module reversely transmits the decoded locomotive signal to the data transmission module so as to control the acquisition strategy of the gateway acquisition module and monitor the mining lithium battery locomotive data in real time. According to the invention, two-way communication between the underground and the ground is realized, and the acquisition strategy is dynamically adjusted based on the operation condition of the locomotive, so that the real-time performance and accuracy of monitoring data are improved, redundant data are reduced, and the communication adaptability of the underground environment is enhanced.

The invention discloses a point cloud alignment method and device based on adaptive voxels and electronic equipment. According to the embodiment of the invention, two point cloud frames which need to be aligned and are based on different coordinate systems are acquired, and corresponding voxels are determined according to the density of the first point cloud frames and the density of the second point cloud frames, so that the two point cloud frames are divided into a plurality of first point cloud frame voxel units and a plurality of second point cloud frame voxel units; using a KD tree algorithm to identify each point cloud voxel unit and extract plane features of the voxel unit, and fusing the extracted plane features to generate voxel features of the first point cloud frame and the second point cloud frame, and calculating a rotation matrix for converting the first point cloud frame into a coordinate system where the second point cloud frame is located based on the generated voxel features. Therefore, under the condition that key information is reserved, redundant data and the calculation amount can be reduced, the calculation efficiency can be improved, and the rotation matrix which more accurately reflects the geometrical relationship between the planes can be obtained, so that the alignment accuracy can be improved.

The invention provides an image compression method, an image decompression method, a model training method, a robot and a server, and relates to the technical field of image processing. The method comprises the following steps: acquiring a first image and a second image; wherein the first image and the second image are images of different modes in the same view field; performing feature extraction on the first image and the second image to obtain a first original feature and a second original feature corresponding to different positions in the field of view; performing joint coding on the first original feature and the second original feature corresponding to the same position in the view field to obtain a first fusion feature and a second fusion feature corresponding to the position; and respectively compressing the first fusion feature and the second fusion feature corresponding to each position in the field of view to obtain a first compressed image and a second compressed image. According to the invention, through joint coding of the first image and the second image, redundant data between modes can be reduced, and the image compression effect is improved.

The invention relates to a vehicle positioning method and device, a controller, a medium, a program product and a vehicle, the vehicle is equipped with positioning equipment, and the positioning method comprises the following steps: performing feature matching on a target image and a field image acquired by the vehicle in real time; a position of the vehicle is determined based on the coordinates of the positioning device. On the basis, the problem that the vehicle depends on GNSS positioning in a GNSS signal missing scene can be solved, and the blank of vehicle position judgment when no positioning signal exists is filled.

This invention provides a method and system for joint formation control of unmanned aerial vehicles (UAVs) and unmanned vehicles (UAVs), relating to the field of vehicle-machine collaboration technology. This invention uses a virtual UAV as the leader, treating both the UAV and the UAV as followers. A fixed formation offset is set for each follower. Communication topology is defined based on graph theory. Real-time acquisition of measured state vectors from sensors is performed. The current estimated state is processed by a constructed radial basis function neural network observer. The estimated disturbance value is output by combining the weight matrix, and the estimated state is updated according to the core equation. Simultaneously, the weight matrix is ​​updated based on the observation error. The desired state of the followers is calculated based on the leader's reference trajectory and the fixed formation offset, and the formation error is calculated. Event-triggered communication conditions are constructed, and it is determined whether the conditions are met. Only when the conditions are met do the followers interact with the state and error according to the communication topology; otherwise, the previously triggered data is used. Finally, the current control input of the followers is calculated through a distributed control law.

This invention relates to the fields of data processing and industrial internet technology, and discloses information management methods, systems, electronic devices, and storage media. The methods include receiving heterogeneous data from multiple sources such as carbon emissions and energy consumption, mapping them into holographic data packets containing a unified spatiotemporal reference; dynamically constructing a spatiotemporal causal correlation spectrum based on the correlation weights between data nodes; calculating the topological potential energy value of a physical region, and generating hardware control commands when the potential energy value exceeds a warning threshold to adaptively adjust the sampling frequency of the front-end device; monitoring the correlation entropy of data nodes, and when a phase transition condition is met, serializing and migrating a stable subgraph to a second storage layer and generating a hash index; retrieving data based on the index and performing integrity verification. This invention achieves dynamic adjustment of the sampling frequency through topological potential energy feedback and uses correlation entropy to drive storage evolution, optimizing hardware resource configuration while ensuring the spatiotemporal correlation and reliable traceability of data.

The application provides a sea-land segmentation method based on prior information. Firstly, a geometric imaging model on a satellite is constructed, and image coordinate system, satellite body coordinate system, satellite orbit coordinate system, J2000 coordinate system and WGS84 coordinate system are sequentially established. Through conversion between the coordinate systems, the image coordinates and the ground coordinates are normalized to the same reference system, and the longitude and latitude information corresponding to the image points of the remote sensing image is obtained. Then, a GIS database is constructed to store a global sea-land information binary graph, and the sea-land marks of the image points are obtained through longitude and latitude query based on the prior information, so that the land area is removed. The application further reduces the algorithm complexity while ensuring the high accuracy segmentation performance, and can be applied to on-board implementation.