3962results about How to "Avoid interference" patented technology

The invention discloses a coal mine heat energy monitoring and cooling regulation and control method based on the Internet of Things, and relates to the technical field of coal mine heat energy monitoring and cooling regulation and control. The collected temperature data are transmitted to a central monitoring system in real time through an Internet of Things gateway; redundant temperature sensors are arranged at key positions, and data of the temperature sensors in the same area are regularly compared and verified. According to the invention, the monitoring robustness is improved through a redundant temperature sensor and a data fusion algorithm, a monitoring blind area caused by a single-point fault is avoided, and the accident risk is reduced; a self-adaptive learning algorithm is adopted to dynamically optimize a detection threshold value and a compensation strategy, so that the anomaly detection accuracy is improved, and false alarms and missing alarms are reduced; the temperature trend is pre-judged through a predictive control algorithm, cooling is conducted in advance, balance between safety and energy efficiency is achieved, overheating accidents are prevented, and energy consumption is reduced.

The invention discloses a method for correcting dialogue intention information of old people based on a deep neural network, and particularly relates to the technical field of dialogue optimization, and the method comprises the steps: collecting and cleaning multi-topic multi-round dialogue data, marking multi-round dialogue corpora, and unifying the format; performing feature extraction on the labeled corpus by using the multi-intention recognition sub-model to generate a multi-intention representation vector capable of representing each intention; screening context content related to the target intention in combination with a hierarchical attention mechanism of user context dynamic weight adjustment; processing context content related to the target intention through gating operation based on multiple rounds of interaction history, and reserving semantic fragments with reference value; performing parallel analysis on the reference semantic fragments based on a multi-task learning structure of a deep neural network, performing classification processing on different intentions, and calculating scores; and executing multi-intention fusion correction operation according to classification and scoring results, and outputting intention information capable of distinguishing each intention.

The invention discloses a multi-underwater-robot cooperative control system and method based on predictive control, and relates to the technical field of cooperative control. During operation of the system, underwater environment data are collected through sensor equipment, an underwater three-dimensional dynamic environment model is generated by using a data fusion technology, and the underwater three-dimensional dynamic environment model is established based on task requirements; tasks are decomposed and distributed to all the robots, through an optimization algorithm, optimal paths of the robots and the mechanical arms are planned, the motion trails of the underwater robot mechanical arms are dynamically optimized based on model predictive control MPC, a real-time control strategy is provided for task division among the multiple mechanical arms, and the real-time control strategy of the underwater robot mechanical arms is provided based on environment modeling and the real-time state. Future motion states of the robot and the mechanical arm of the robot are predicted, an optimization objective function including task completion time, energy consumption and mechanical arm operation precision is constructed, a rolling optimization method is utilized, a control instruction is updated in real time, and real-time data sharing among multiple underwater robots is supported based on an acoustic communication technology.

The invention discloses a communication transmission control cabinet based on big data, and relates to the technical field of communication transmission, and the control cabinet comprises the steps: introducing a real-time network state perception and prediction module, collecting the data of a link through an embedded sensor, carrying out the prediction through combining with an LSTM time sequence model, carrying out the estimation of a future link state, introducing a multipath transmission mechanism, and carrying out the real-time network state perception and prediction. When tasks are distributed, bandwidth, delay and load conditions of paths are synthesized, batch uploading is performed through caching and compression strategies of edge computing nodes, occupation of links is effectively reduced, when link performance is reduced, a path switching mechanism can select alternative paths with optimal performance in real time, influences of link interruption on communication tasks are avoided, and meanwhile, the communication efficiency is improved. According to the high-real-time task priority strategy, an optimal path is adopted, delay of key tasks is reduced, non-real-time tasks are compressed and processed in batches at edge nodes, resources are not competed with real-time tasks, interference is avoided, and the requirements for ship navigation and real-time monitoring tasks are met.

The invention provides a hybrid representation learning-based omni-directional motion control method for a quadruped robot in a three-dimensional environment, aiming at solving the problem that an existing motion control method for the quadruped robot is limited by a narrow forward view angle and external perception noise of the robot, so that the omni-directional motion capability in the three-dimensional environment is difficult to obtain. Comprising the steps of designing various input and output contents of a first-stage end-to-end neural network learning framework, designing a training process including mixed representation learning based on a reinforcement learning framework, constructing a specific neural network model of each module, and constructing a quadruped robot simulation motion control training environment. The method has the advantages that network input and output content beneficial to efficiently learning the relation between actual observation of the robot and the environment truth value is provided, and a mixed representation learning method including supervised learning and comparative learning is introduced, so that the robot can effectively deduce environment information beyond the visual field range of the robot in the training process.

The invention discloses a steel structure three-dimensional workpiece welding seam identification and path planning method and system, and relates to the technical field of robot intelligent welding, and the method comprises the steps: building a multi-view fusion point cloud corresponding to a target workpiece; performing welding seam segmentation on the multi-view fusion point cloud by using a welding seam neighborhood point cloud segmentation model to obtain a multi-view fusion welding seam neighborhood point cloud; performing point cloud plane fitting on the multi-view fusion welding seam neighborhood point cloud to obtain a plurality of target welding seam neighborhood planes; and determining a corresponding welding seam straight line based on each pair of intersected target welding seam neighborhood planes, and identifying a welding path and a welding posture according to the welding seam straight line. The method provided by the invention shows good applicability under a complex background, has the advantages of automation, high efficiency, high recognition precision, strong adaptability and the like, and meets the intelligent and flexible operation requirements of the welding robot.

The invention discloses a complex scene multi-source heterogeneous data acquisition and monitoring diagnosis method and device, and the method comprises the steps: collecting multi-source heterogeneous data of an object to be monitored and diagnosed, and carrying out the data cleaning, data preprocessing, feature extraction and fusion of the multi-source heterogeneous data, so as to obtain a feature data set; and constructing and training a causal decoupling representation learning model based on input and output driving by using the feature data set and the data after data cleaning, so as to extract decoupling representation of a to-be-monitored diagnosis object, deduce a causal relationship between variables, and obtain a to-be-monitored diagnosis object. Wherein the causal decoupling characterization learning model comprises an external source control conversion network and a double-view evolution modeling and causal inference network, and the trained causal decoupling characterization learning model is used as a heterogeneous signal monitoring and diagnosis model and is used for monitoring the state of an object to be monitored and diagnosed and diagnosing the root cause of a fault of the object to be monitored and diagnosed. According to the invention, data acquisition, state monitoring and fault reason diagnosis can be integrally realized, the cost is saved, and the application prospect is wide.

The invention discloses an unmanned aerial vehicle control method and system, relates to the technical field of flight path control, and solves the problems of insufficient image acquisition precision, inaccurate routing inspection path planning and limited real-time data analysis capability. Through an image acquisition and enhancement module, a real-time image enhancement algorithm and a multi-image comparative analysis module, the focal length and exposure parameters of a camera are dynamically adjusted, the image definition in a complex environment is improved in combination with evaluation of an image enhancement effect value Enh and an effect threshold E, and a dimensionless size value Fdz, a position value Fdw and a relative distance value Fjl of a fault point are extracted; the image acquisition quality is effectively controlled; the environment modeling module is used for generating a three-dimensional model containing cable and obstacle information, ensuring that the path coverage value Qcd reaches the standard and avoiding an inspection blind area; and an airborne data processing mechanism is combined with an electromagnetic interference prediction and collision prediction model, so that real-time fault analysis and alarm are realized, the detection efficiency is improved, and the data real-time performance is ensured.

The invention provides a context understanding-based software operation prompt automatic optimization method and system, and relates to the technical field of artificial intelligence, and the method comprises the steps: collecting the multi-dimensional context feature data of user software operation, carrying out the feature coding through a multi-modal feature fusion network, modeling the time sequence dependence through a bidirectional long-short-term memory network, and carrying out the feature coding through the bidirectional long-short-term memory network. And a context semantic representation vector is generated based on a multi-head attention mechanism, and a prompt generation strategy is continuously optimized in combination with reinforcement learning. According to the method, personalized customization of software operation prompts can be realized, the accuracy and timeliness of prompt contents are improved, and the software use experience of a user is enhanced.

The invention discloses rapid die casting equipment for aluminum alloy die castings, and relates to the technical field of metal die casting. A first channel is formed in a die plate and an ejector rod, and a spray hole communicating with the first channel is formed in the ejector rod; according to the device, the mold release agent can be sprayed to the relatively-closed die-casting cavity in a mold closing state, the comprehensive uniformity of the mold release agent during adhesion is improved, the demolding effect after a die casting is formed is improved, the mold release agent is sprayed in the relatively-closed die-casting cavity, and the mold release effect is improved. According to the die-casting mold core, the release agent can be prevented from escaping outwards, the surrounding environment can be protected easily, waste of the release agent can be avoided, energy-saving, environment-friendly, green and efficient die-casting production is achieved, meanwhile, the spraying holes are formed in the cylindrical face of the end of the ejector rod, the spraying holes cannot be exposed when the ejector rod retracts into the die core, and the die-casting mold core is more environment-friendly. And interference to die-casting production due to the fact that the spraying holes are exposed in the die-casting cavity can be avoided.

The invention relates to the technical field of medicine information management, in particular to a traceability method and system based on a medicine consumable traceability code. The method comprises the following steps: acquiring medicine purchasing information data; generating a unique code for the number of medicine purchasing information so as to generate a medicine warehousing traceability code; performing drug information association on the drug purchase information data according to the drug warehousing traceability code to generate a drug warehousing associated traceability code; storing the medicine warehousing associated traceability code in an information base to generate a medicine warehousing traceability code information base; performing scanning access mode division on the medicine warehousing traceability code, and generating a medicine dispensing scanning request mode and a medicine returning scanning request mode; and performing drug splitting tracing on the drug warehousing tracing code in a drug dispensing scanning request mode to generate drug splitting tracing data. According to the invention, through full-process closed-loop management, data association enhancement, intelligent scheduling optimization and drug return tracing improvement, the efficiency and precision of the drug consumable traceability code tracing method are improved.

The invention provides a building material surface defect detection method based on infrared polarization imaging, which belongs to the field of image processing, and constructs a building material surface defect detection model through input, image preprocessing, a feature extraction module, a feature fusion module, a building material polarization enhancement image module, a building material surface defect detection module and output. Information of a polarization image, an infrared image and a depth image is fused, more comprehensive feature representation is provided, the model has higher recognition capability on fine defects, a self-attention mechanism further optimizes the focusing effect of the features, interference of background noise is avoided, it is ensured that a defect area is fully recognized, and the recognition efficiency is improved. The method has better adaptability and performance in the field of defect detection, can cope with more complex and variable building material detection tasks, and particularly shows stronger advantages when detecting tiny defects and complex surfaces.

The invention relates to the technical field of lithium batteries, and discloses a lithium battery life decline quantitative prediction and health state intelligent evaluation method, which comprises the following steps: S1, multi-source data acquisition: arranging strain gauge sensors at battery electrode plates and a shell to acquire stress-strain data, integrating a gas sensor in a battery module to acquire gas data, and calculating the data of the stress-strain data; battery data and environment data are acquired by a high-precision synchronous acquisition system through nanosecond clock signals and are processed and stored, and the data quality is monitored at the same time; s2, data preprocessing: generating an extended data set through data transformation and combination and GAN-based data, and processing abnormal data by using a method based on statistics and machine learning; according to the invention, through multi-source data fusion acquisition, a stress-strain sensor and a plurality of gas sensors are added to monitor the internal mechanical change and gas component concentration of the battery, so that the problem that the accuracy and practicability of the model are greatly reduced due to a traditional method based on an electrochemical model is improved.

The invention relates to the technical field of battery module assembly, in particular to a battery cell surface laser cleaning line and a laser cleaning method.The battery cell surface laser cleaning line comprises a main conveying line, a conveying trolley, a polar column surface texturing mechanism, a turnover mechanism, a narrow surface texturing mechanism, a bottom surface texturing mechanism, a wide surface texturing mechanism, a first guide plate and a second guide plate; the conveying trolley comprises a bottom frame movably installed on the main conveying line, a carrier arranged over the bottom frame, a rotating assembly installed in the middle of the top end of the bottom frame, side protection plates arranged on the front side and the rear side of the carrier in a mirror image mode, and a lifting assembly installed at the bottom of the carrier. The first guide plate is mounted on the rear side of the main conveying line and transversely extends between the polar cylindrical surface texturing mechanism and the bottom surface texturing mechanism so as to drive the rotating assembly to rotate, and the second guide plate is mounted on the front side of the main conveying line and transversely extends in the wide surface texturing mechanism so as to drive the lifting assembly to ascend and descend; the device is simple in structural design and process steps, low in input cost and automatic in the whole process, and the cleaning quality can be guaranteed and stable.

The embodiment of the invention discloses a three-dimensional reconstruction method based on structured three-dimensional Gaussian representation. The method comprises the following steps: acquiring a target image containing a target object at multiple visual angles; generating an unstructured Gaussian primitive representation corresponding to the target object according to the target image; establishing a transmission matrix describing the unstructured Gaussian primitive distribution and the structured Gaussian primitive distribution, solving the transmission matrix by using an optimal transmission algorithm, and generating a structured Gaussian primitive representation corresponding to the target object according to the optimal transmission matrix; obtaining a text description for the target object, and complementing the structured Gaussian primitives according to the text description by using a diffusion model obtained by pre-training; and performing three-dimensional reconstruction by using the complete structured Gaussian primitive. According to the method, the optimal transmission algorithm is used for carrying out structured representation on the three-dimensional Gaussian primitives, and text description is used for complementing the Gaussian primitives through the diffusion model, so that the quality of three-dimensional reconstruction is improved.

The invention discloses a data analysis processing method and system based on a water conservancy project, and relates to the technical field of data processing, and the method comprises the steps: carrying out the analysis processing of to-be-processed water conservancy project data in a database system based on a data management terminal, and obtaining a water conservancy project abnormal data position; and based on the data management terminal, analyzing and processing the water conservancy project abnormal data position, and determining a water conservancy project abnormal data type. According to the method, missing data positioning is performed on the to-be-processed water conservancy project data in the database system, then the type of the missing data is analyzed, whether the missing data can be supplemented or not is determined, and if not, the missing data is deleted, so that interference on the water conservancy project power generation equipment during subsequent analysis on the operation state of the water conservancy project power generation equipment is avoided; the data analysis time is shortened, and finally, the supplemented water conservancy project data is sorted through the time continuity, so that the analysis of the water conservancy project data is more convenient and faster when the water conservancy project data is called subsequently.

The invention provides a self-discovery ad-hoc network control method based on a swan-gap operating system, and relates to the technical field of data processing, and the method comprises the steps: based on the swan-gap operating system, extracting the identification information and communication parameters of equipment from all to-be-networked equipment; carrying out compatibility judgment on the communication protocol version and the maximum support bandwidth between the devices; determining a control relationship between the master device and the slave device; dynamically allocating communication resources to each device by adopting a distributed channel allocation algorithm in combination with task requirements of the devices and network bandwidth information; periodically synchronizing the resource state of each device; when the slave device sends a networking quitting request, the master device updates the networking candidate data and re-determines the networking control relationship after releasing the corresponding communication resources; when the master device exits, the device with the highest priority is selected from the slave devices to automatically take over the control role and complete the reconstruction of the networking control relation; according to the invention, the autonomy and accuracy of networking are improved.

The invention relates to the technical field of numerical control machine tool machining, and discloses a numerical control machine tool machining precision optimization method and system. The system comprises a machining data acquisition module, a machining environment data acquisition module, a precision data analysis module and a precision influence early warning module. According to the system, tool bit abrasion data of each tool of the current numerical control machine tool, the flatness of a clamp of the numerical control machine tool, collected cooling liquid spraying uniformity and a workpiece surface scrap residual value are comprehensively calculated to obtain a first precision risk coefficient, and whether a first adjustment strategy is executed to clean the numerical control machine tool or not is judged; based on the electric signal interference frequency and the temperatures of the transmission parts, a second precision risk coefficient is obtained through comprehensive calculation, whether a second adjustment strategy is executed or not is judged, whether other parts need to be overhauled or not is judged, and therefore interference of external factors on the numerical control machine tool is avoided by considering multiple factors.

The invention relates to the technical field of data processing, in particular to a karst area mine groundwater pollution channel tracking method, which comprises the following steps of: acquiring groundwater flow data and pollutant concentration data of a plurality of monitoring points of a karst area mine; the water flow fluctuation degree of the monitoring point and the concentration trend change degree of various pollutants are obtained according to the underground water flow data and the pollutant concentration data, the overall pollutant concentration equilibrium proportion degree of the monitoring point is obtained according to the concentration difference between various pollutants and another matched pollutant, and the overall pollutant concentration equilibrium proportion degree of the monitoring point is calculated according to the water flow fluctuation degree. The characteristic distance between any two monitoring points can be obtained according to the concentration trend change degree of various pollutants in the current monitoring period and the previous monitoring period and the overall pollutant concentration equilibrium ratio degree, so that the monitoring points are clustered, the accuracy of the clustering result is improved, the tracking of the mine groundwater pollution channel is realized, and the monitoring efficiency is improved. And interference of actual mine operation activities on local monitoring points is avoided, so that the tracking accuracy of the pollution channel is enhanced.

The invention relates to the technical field of electric wheelchair control, in particular to a multi-mode navigation control method and system of an electric wheelchair controller. The method comprises the following steps: acquiring a real-time environment monitoring video based on an electric wheelchair high-definition camera; video jitter distortion compensation and dynamic object frame-by-frame continuous optical flow tracking are carried out on the real-time environment monitoring video, and a moving track of each dynamic object is generated; performing environment static obstacle accurate identification on the real-time environment monitoring video, and performing dynamic environment perception modeling based on a moving track of each dynamic object so as to generate a dynamic environment map; a navigation mode switching decision is made based on a real-time wheelchair operation instruction of the user to obtain a current navigation mode of the user, and the current navigation mode of the user comprises a rocker control mode and an intelligent navigation mode. According to the invention, efficient and intelligent multi-mode control of the electric wheelchair is realized, so that the electric wheelchair can more flexibly and safely cope with a complex environment.

The invention relates to the technical field of equipment control, in particular to a self-adaptive switching method and system of a current transformer. The method comprises the following steps: collecting primary side parameters and secondary side parameters of a current transformer to obtain primary side collection parameters of the current transformer and secondary side collection parameters of the current transformer; performing current transformer state segmentation initialization according to the current transformer primary side acquisition parameters and the current transformer secondary side acquisition parameters so as to generate a current transformer operation state classification result; switching result judgment is conducted on the current transformers according to the current transformer operation state classification result, and when the current transformer operation state classification result is a current transformer critical state section, parallel buffering is conducted on the current transformers, and critical current transformer switching data are generated. According to the invention, through accurate state classification, a dynamic switching mechanism, time sequence optimization and visual display, the switching sensitivity and stability of the current transformer are improved.

The invention relates to a composite sensor for acceleration and pressure monitoring and a working method thereof, and belongs to the field of sensors, the composite sensor comprises an upper shell, a middle shell and a lower shell, the upper end of the upper shell is provided with an M6 mounting thread, the lower end of the upper shell is a hexagonal base, and a groove space is reserved in the inner side of the lower end of the hexagonal base; the appearance of the middle shell is a hexagonal base, and M5 mouthpiece installation holes are formed in the left side and the right side of the middle shell. A double-end beam sheet is arranged in the middle of the inner side of the middle shell and is of a sheet-shaped structure, and the two ends of the double-end beam sheet are fixedly connected with the front inner wall and the rear inner wall of the middle shell or integrally formed. The center of the double-end beam sheet is provided with an integrally connected cylindrical screw rod, and the cylindrical screw rod is provided with an acceleration sensor core body structure; the appearance of the lower shell is a hexagonal base, and a mounting groove is formed in the bottom of the lower shell and used for mounting a pressure sensor core structure; the upper shell, the middle shell and the lower shell are connected in a welded mode. The device is simple in structure, can realize synchronous monitoring of acceleration and pressure without mutual interference, and can realize multi-scene application at the same time.

The invention, which relates to the technical field of digital information transmission, discloses an AI technology-based digital information transmission sharing operation and maintenance platform comprising a data acquisition module, an AI intelligent analysis module, a frequency spectrum dynamic allocation module, a data classification and transmission optimization module, and a storage adaptation module. According to the method, the spectrum resource utilization efficiency and the transmission stability are optimized, the platform can accurately master the spectrum use condition in real time through the AI real-time spectrum sensing and dynamic frequency allocation technology, the platform can quickly identify idle spectrums and dynamically allocate optimal frequencies for different services in the face of a complex and changeable wireless environment, spectrum congestion and interference are avoided, and the service quality of the system is improved. The fluency of bandwidth and time delay sensitive services is guaranteed, and the transmission interruption risk is reduced.

The invention relates to the technical field of multi-modal sentiment analysis, in particular to a super-multi-modal sentiment analysis method and system based on bidirectional translation interaction. The method comprises the steps that historical data samples and current data samples of super-multi-mode sentiment analysis are obtained, and the historical data samples and the current data samples comprise five modes including images, audios, texts, electroencephalogram and eye movement; carrying out weighted fusion on similar samples in the historical data and the current data sample to obtain enhanced super-multi-modal features under the condition of uncertain modal deficiency; carrying out self-attention interaction on the super-multi-modal features under the condition of uncertain modal deficiency so as to improve the quality of each modal; performing splicing fusion and self-attention interaction on the four non-electroencephalogram modals to form initial fusion four modals, and translating the electroencephalogram into new fusion four modals through a multi-head attention mechanism; according to the method, the current sample and the similar sample are subjected to weighted average fusion, so that the quality of each mode is enhanced.

The invention relates to the technical field of transistor detection, in particular to a quick test method and system for an MOSFET device. The method comprises the following steps: acquiring and integrating multi-source parallel data of the MOSFET device to obtain real-time I-V characteristic data of the device; performing MOSFET device performance prediction based on the real-time device I-V characteristic data to obtain MOSFET device performance prediction data; screening according to the performance prediction data of the MOSFET device to obtain performance data of a primarily screened qualified device; performing adjustment based on the performance data of the preliminarily screened qualified device to obtain an MOSFET device test scheme; acquiring real-time test data of the MOSFET device, and screening qualified devices to obtain performance data of re-screened qualified devices; and performing device test summarization based on the performance data of the primarily screened qualified device and the performance data of the secondarily screened qualified device to obtain an MOSFET device test report. According to the invention, the test period can be shortened.

The invention discloses a fumigation treatment multi-parameter comprehensive monitoring system which comprises a multi-parameter sensor monitoring module, a multi-modal data processing and analyzing module, a system intelligent monitoring and adjusting module, a fumigation environment control module, a remote maintenance and management module and a system interaction and feedback module. The multi-parameter sensor monitoring module is used for collecting parameter data. The multi-modal data processing and analyzing module is used for fusing and analyzing data. The system intelligent monitoring and adjusting module is used for intelligently monitoring and adjusting fumigation strategies. The fumigation environment control module is used for adjusting environment parameters. The remote maintenance and management module is used for system maintenance and management. And the system interaction and feedback module is used for display interaction of data and a control state. According to the fumigation treatment multi-parameter comprehensive monitoring system, a multi-modal data fusion algorithm based on information channel decomposition is provided for multi-parameter data fusion and analysis; and a multi-parameter intelligent monitoring algorithm based on a multi-layer feedforward neural network is provided to intelligently monitor the fumigation environment.

The utility model discloses an intellectualized safety control take-up straining device, including handle subassembly, base subassembly and long braid, handle subassembly includes handle and ratchet wheel, base subassembly includes base, handle and base are pivoted and rotatingly connected through take-up shaft, ratchet wheel is fixed on take-up shaft, and the winding drum is fixed between the ratchet wheel, and the long braid is fixed on the base. The ratchet wheel, the winding drum and the tape winding shaft rotate synchronously, the winding drum is sleeved outside the tape winding shaft and is coaxially arranged, an opening for the long braid to pass through is arranged on one side of the winding drum, one end of the long braid is sleeved on the tape winding shaft and is led out from one end of the base through the opening on the winding drum, one end of the tape winding shaft is connected with the coil spring assembly, and the coil spring assembly comprises a coil spring, a coil spring box and a coil spring box cover. The coil spring box is covered with the coil spring box, a brake assembly is further arranged in the coil spring box, and one end of the tape winding shaft can be provided with a C-shaped opening, an eccentric shaft or a concentric screw hole according to the structure of the brake assembly. According to the utility model, the long braid can be prevented from swinging to hurt an operator, and the personal safety of the operator is guaranteed.

The embodiment of the invention discloses a voice multitask-based model training method and a voice multitask processing method. Firstly, a target voice signal is acquired, and a first voice feature in the target voice signal is extracted through a pre-trained feature extraction model. Then, processing the first voice feature through a forgery detection model to obtain a first voice forgery probability; thirdly, training a forgery detection model based on the first voice forgery probability, and after training is completed, freezing parameters of the forgery detection model; further, the first voice feature is processed through a voiceprint recognition model to obtain first voiceprint information, and the voiceprint recognition model is trained based on the first voiceprint information. And finally, performing forgery detection and speaker verification on the to-be-processed voice signal by using the trained forgery detection model and voiceprint recognition model.

The invention discloses soil sampling equipment for geological exploration, relates to the field of soil sampling, and solves the problem that the existing soil sampling equipment is difficult to flexibly switch an integral soil sampling state and a local depth soil sampling state. The soil sampling equipment comprises a machine body, a sampling barrel, a driving mechanism, a sampling mechanism and a control mechanism, the driving mechanism comprises a driving motor and a lifting part, the sampling mechanism comprises a control cylinder and a sampling block, and the control mechanism comprises a roller and a control rod. According to the soil sampling equipment for geological exploration, when the driving motor drives the gear column to rotate, the sampling cylinder rotates, and meanwhile, the lifting part is linked to lift the sampling cylinder; soil sampling operation is achieved, the control cylinder is controlled to move upwards by a set distance after a set depth is reached, then drilling continues to enable soil to enter an area between the sampling block and the control cylinder to be collected, and when the sampling cylinder is in a lifting state, the control rod is in linkage to abut against the side face of the rolling wheel, and automatic limiting is achieved.

The invention relates to the technical field of ecological management, in particular to a mine management soil remediation device which comprises an upper shell, a lower shell is fixedly mounted at the bottom of the upper shell, a water tank is fixedly mounted at the top of the upper shell, a separation mechanism is mounted in the upper shell, a smashing mechanism is mounted in the lower shell, and a soil remediation mechanism is mounted in the smashing mechanism. A spraying mechanism is installed between the lower shell and the water tank, an output mechanism is installed in the lower shell and located at the bottom of the smashing mechanism, and a driving mechanism is installed among the separating mechanism, the smashing mechanism and the spraying mechanism. The separating mechanism comprises a first separating roller, a second separating roller and a first bevel gear. According to the mine soil remediation device, stones in mine soil are effectively separated and crushed, the function of agent spraying is combined, the mine soil remediation efficiency and quality are improved, the equipment maintenance cost is reduced, and effective remediation of the mine soil and improvement of the ecological environment are achieved.