119results about How to "Avoid falling into" patented technology

The invention provides an unmanned ship autonomous navigation control method and system based on artificial intelligence, and relates to the technical field of control, and the method comprises the steps: collecting the real-time navigation data of an unmanned vehicle through a multi-source sensor, inputting the navigation environment feature data into a pre-trained reinforcement learning model, and enabling the reinforcement learning model to adopt a dual Q network structure, performing iterative optimization on the navigation strategy based on the environment reward function to generate a navigation strategy optimization parameter; constructing a multi-target route optimization model according to the navigation strategy optimization parameters, performing global optimization on a route by adopting an improved ant colony algorithm, and outputting optimal navigation path data; a hierarchical navigation control model is established according to the optimal navigation path data, the hierarchical navigation control model comprises a strategic layer, a tactical layer and an execution layer, a navigation control instruction is output through the hierarchical navigation control model, and autonomous navigation control of the unmanned vehicle is achieved.

The invention provides a wave height prediction method for multi-strategy snow ablation optimization, and belongs to the field of ocean engineering. The wave height prediction method comprises the following five steps: preprocessing sea condition data, performing feature extraction on the sea condition data by using a CNN model, training the model, optimizing parameters and evaluating the model. According to the method, the multi-strategy snow ablation optimization algorithm, the convolutional neural network, the bidirectional gating circulation unit and the multi-head attention mechanism are comprehensively utilized, and efficient processing and accurate prediction of the sea condition data are achieved; the introduction of the multi-strategy snow ablation optimization algorithm increases the diversity of the population, prevents the model from falling into local optimum, and improves the robustness of the algorithm. The CNN and the bidirectional GRU are combined, so that the model has excellent performance in the aspect of processing high-dimensional, nonlinear and time sequence data; the application of the attention mechanism can highlight important information, perform weighting processing on features, and further improve the prediction performance of the model.

The invention relates to the technical field of radio communication, and discloses an XL-RIS-assisted near-field multi-beam design method and system, and the method comprises the steps: constructing an XL-RIS-assisted near-field communication multi-user receiving signal model; constructing a near-field multi-beam optimization model; the near-field multi-beam optimization model is solved, so that an XL-RIS beam reflection matrix can be obtained; and iteratively solving the near-field multi-beam optimization model to obtain a reflection matrix of a base station transmitting signal to the XL-RIS, and completing XL-RIS assisted system multi-beam optimization. Compared with the existing calculation, the beam optimization method provided by the invention does not need a complicated calculation problem, overcomes the problems of low convergence speed and high calculation complexity in large-scale array processing of the traditional method, and improves the optimization efficiency. The operation speed can be increased under the condition that the system performance is guaranteed.

The utility model provides a medicine bottle cap easy to open, which belongs to the technical field of medicine packaging and comprises a bottle, a bottle neck, a threaded surface, a sealing ring, a fixing frame, aluminum foil paper, a bottle cap, a connecting block and a connecting piece, the bottle neck is arranged at the upper end of the bottle, and when the bottle cap on the bottle is screwed open, the connecting piece in the bottle cap can rotate along with the bottle cap, so that the bottle cap is convenient to open. The connecting piece can pull the pull ring of the fixing frame to move in the unscrewing direction, so that the fixing frame surrounding the outer end of the bottle neck is continuously pulled open in the rotating direction, aluminum-foil paper packaged on the fixing frame is continuously torn open, and the aluminum-foil paper for sealing medicine can be torn open upwards when the bottle cap is opened. And the lower surface of the outer edge of the aluminum-foil paper is packaged on the upper surface of the fixing frame, when the fixing frame is pulled down, the aluminum-foil paper on the fixing frame can be torn open together and taken away from the bottle opening of the bottle mouth, the aluminum foil is prevented from falling into the bottle body, and the safety of the medicine is improved.

The invention belongs to the technical field of hydroelectric generating set vibration prediction, and particularly relates to a hydroelectric generating set mechanical vibration trend prediction method and device, equipment and a medium, and the method comprises the steps: processing a vibration signal of a hydroelectric generating set through employing a variational mode decomposition algorithm; training the deep belief network by using the training set to obtain a trained deep belief network model; using a sparrow search algorithm to optimize parameters of the deep belief network model to obtain a vibration trend prediction model; and collecting a vibration signal of the hydroelectric generating set, processing the vibration signal by using a variational mode decomposition algorithm, inputting the processed vibration signal into the vibration trend prediction model to obtain a vibration prediction value, and taking the obtained vibration prediction value as an input of next prediction to realize multi-step prediction of the vibration trend of the hydroelectric generating set. By means of multi-step prediction, operation and maintenance personnel can know the vibration trend of the hydroelectric generating set in a period of time in the future in advance, potential fault hidden dangers can be found in time, and the operation safety and stability of the hydroelectric generating set are improved.

The present invention relates to the technical field of electronic candles, and discloses an electronic candle power supply device, comprising a power supply seat, wherein a card seat is arranged in the power supply seat, an annular seat is arranged in the card seat, the annular seat is elastically connected to the card seat through a first spring, a first guide rod and a second guide rod are respectively fixed to the annular seat, a power supply plug is arranged in the power supply seat, the power supply plug is elastically connected to the card seat through a second spring, a seesaw assembly is arranged on the power supply seat, two sides of the seesaw assembly are respectively in conflict with the first guide rod and the power plug, and a round hole for the power plug to slide is opened on the card seat; the power plug can be conveniently hidden inside the power supply seat to avoid the power plug being exposed to humid air for a long time, the adhesion of moisture and dust can be reduced, and moisture can be prevented from penetrating into the inside of the electronic candle body to cause corrosion and short circuit of electronic components, thereby improving the protection effect.

The invention discloses a turbonator rotor winding turn-to-turn short circuit fault early warning method and device, and relates to the field of data processing, and the method comprises the steps: obtaining historical monitoring data and an actual value of excitation current, carrying out the preprocessing of the historical monitoring data, obtaining the preprocessed historical monitoring data, and forming a monitoring sequence; constructing an excitation current prediction model based on Bi-LSTM, and optimizing the training parameters to be optimized by adopting an improved pelican optimization algorithm to obtain optimized training parameters; training an exciting current prediction model based on the optimized training parameters to obtain a trained exciting current prediction model; and inputting the monitoring sequence into a trained exciting current prediction model to obtain a predicted value of the exciting current, and judging whether the turbonator has a rotor winding turn-to-turn short circuit fault or not according to the predicted value of the exciting current and the actual value of the exciting current. The problems that an existing model is poor in optimization capacity, low in convergence speed and low in prediction accuracy are solved.

This invention relates to the field of graphite waste treatment technology, and more particularly to a method for preparing special graphite materials from isostatic pressing graphite waste, comprising the following steps: waste collection: firstly, the waste generated from isostatic pressing graphite production is collected and processed centrally; waste treatment: the collected waste is centrally fed into a purification module, where impurities are screened and removed; waste crushing: after the isostatic pressing graphite waste has been screened, it is pushed into a crushing device for crushing; material proportioning, pressing, primary calcination, and graphitization treatment; this invention, by processing the waste from isostatic pressing graphite production through the screening and purification processes described above, followed by crushing, can then be proportioned, fed into a mixing pot, and then pressed, calcined, and graphitized to produce special graphite materials, thus avoiding environmental pollution from waste and achieving full utilization of resources.

This utility model discloses a multi-functional fire-fighting intercom, including a main body and an intercom module, a temperature detection module, a control module, a gas detection module, a stationary detection module, an alarm module, and a battery integrated on the main body. The intercom module has a microphone speaker on its left or right side. The temperature display panel of the temperature detection module is located on the top of the main body. The gas detection module has a gas detection probe and an information display window on its front, displaying the concentration of the detected gas. The stationary detection module includes an accelerometer sensor to sense the speed of the main body. The alarm module includes an alarm speaker and several alarm indicator lights. The control module is electrically connected to the gas detection module, the stationary detection module, and the alarm module. The battery is electrically connected to each module. This significantly expands the uses of the intercom and effectively improves the safety of firefighters during fire rescue operations.

The invention discloses an MPPT (Maximum Power Point Tracking) control method based on a plant cell population-incremental conductance method composite algorithm, which relates to the field of maximum power point tracking of a photovoltaic power generation system under partial shading, and comprises the following steps: S1, setting parameters such as cell number N, maximum growth times M, auxin concentration threshold and the like, and initializing cell positions; s2, calculating the sunshine amount in contact with each cell, and updating the maximum value of the strongest illumination position and the sunshine amount in contact with the cells; s3, distributing auxin for the cells, and calculating the growth speed of the cells; s4, completing primary growth of the cells, and updating the plant cell population; s5, judging whether a termination condition is met or not, and if yes, entering a conductance increment method; otherwise, returning to the step S2; and S6, further adjusting the duty ratio D by adopting an incremental conductance method on the basis of the plant cell group algorithm. The maximum power point tracking method combines the advantages of high global search capability of the plant cell group algorithm and high search speed of the conductance increment method, and can rapidly and accurately track the maximum power point under complex conditions.

The invention discloses a method and a system for testing an automatic neutral section passing function of a train, relates to the technical field of testing, and mainly aims to test the automatic neutral section passing function of the train and prevent the train from falling into a neutral section area. According to the main technical scheme, a rail line for simulating train running comprises a first rail and a phase separation area, the first rail is a train passing forbidding rail, and the first rail is located in front of the phase separation area in the train running direction; determining a preset neutral section passing result of normal processing of the neutral section passing function based on the pre-estimated driving permission terminal point on the neutral section area and the first track; testing the automatic neutral section passing function of the train by using the neutral section area and the first track to obtain a neutral section passing test result; and comparing the preset neutral section passing result with the test neutral section passing result, and determining a test conclusion of the automatic neutral section passing function of the train.

The invention discloses an optimal control method and system for reactive power compensation of a power distribution network, and the method comprises the steps: building a multi-target reactive power optimization model of the power distribution network based on the system parameters of the power distribution network, and carrying out the optimal control of the reactive power compensation of the power distribution network under the constraint condition of the multi-target reactive power optimization model of the power distribution network, solving the target function of the multi-target power distribution network reactive power optimization model by adopting a multi-target whale optimization algorithm to obtain an iterative optimization solution set output by the multi-target power distribution network reactive power optimization model, and processing the multi-target power distribution network reactive power optimization model based on the iterative optimization solution set and an improved multi-target grey wolf optimization algorithm to obtain a multi-target power distribution network reactive power optimization model. And taking a result output by the multi-target power distribution network reactive power optimization model as a node configuration optimal solution set of the power distribution network, and controlling the power distribution network by using the node configuration optimal solution set. According to the optimal control method and system for the reactive compensation of the power distribution network, the optimal control effect of the reactive compensation of the power distribution network is improved.

The utility model discloses a safety protection device for chemical engineering, which comprises a feed hopper and a safety filter screen, the top end of the feed hopper is detachably provided with a protective cover, the top end of the protective cover is communicated with a feed pipe, the top end of the feed pipe is in threaded connection with a sealing cover, and a fixed cylinder is inserted into the feed hopper. The safety filter screen is installed on the bottom side of the interior of the fixing cylinder, a top ring is arranged at the position, below the fixing cylinder, in the feeding hopper, the bottom end of the top ring is fixedly connected with a plurality of connecting rods, the connecting rods are all in sealed sliding fit with the feeding hopper, and the bottom ends of the connecting rods are fixedly connected with a lifting ring; and the fixed cylinder and the safety filter screen can be conveniently ejected out from the interior of the feeding hopper, and people can conveniently take out the safety filter screen from the interior of the feeding hopper.

This invention discloses a container ship trajectory prediction method based on EMD-PSO-GRU-RBFNN, belonging to the field of intelligent shipping technology. The method takes ship AIS data as input, including time, longitude, latitude, speed, and heading angle information; preprocesses the AIS data to divide it into training and test sets; constructs a trajectory prediction model, calculates the difference between the predicted values ​​of the training set and the actual ship position coordinates to obtain the residual signal; adaptively decomposes the residual signal using empirical mode decomposition to obtain intrinsic mode components and residual components; constructs prediction models for high-frequency and mid-to-low-frequency components, reconstructs the prediction results of each component to obtain the residual prediction value, and superimposes it with the baseline prediction results of the test set to obtain the corrected test set output; the corrected test set output is then evaluated. This invention can model and compensate for trajectory errors at different frequency scales, improving the stability and reliability of trajectory prediction.

The invention discloses a task scheduling method and device, electronic equipment, a storage medium and a program product, and relates to the field of big data, and the method comprises the steps: obtaining a task environment state constructed by a to-be-scheduled task and a current data resource environment; calling a preset prompt generator, determining a target scheduling task from the task environment state, and generating a prompt word corresponding to the target scheduling task and the data resource environment; and inputting the cue word into a preset large model to obtain a scheduling scheme of the target scheduling task, and scheduling the target scheduling task based on the scheduling scheme. The technical problem that task scheduling is not efficient enough is solved.

The invention relates to the technical field of unmanned aerial vehicle reconnaissance systems, in particular to an unmanned aerial vehicle multi-target cognitive electronic reconnaissance system which comprises an unmanned aerial vehicle platform, an unmanned aerial vehicle, a ScurterN task simulation module used for intercepting radar signals and a simulator used for modeling unmanned aerial vehicle multi-target cognitive electronic reconnaissance task scenes. The radar is arranged in the reconnaissance task area, a plurality of pseudo target modules are arranged in the reconnaissance task area, a plurality of pseudo targets are obtained from the position of the radar according to a nearest neighbor method, the arc centers of the plurality of pseudo targets are solved and are not updated and iterated continuously, and finally a pseudo target is formed; and the unmanned aerial vehicle multi-target cognitive electronic reconnaissance strategy learning module is used for convergence of a radar reconnaissance strategy and building an unmanned aerial vehicle multi-target cognitive electronic reconnaissance strategy network. According to the method, convergence of more than four radar reconnaissance strategies can be simplified, and the convergence speed of the more than four radar reconnaissance strategies is improved.

A device for preventing a traction steel cable from falling comprises two bottom plates hinged to each other, the upper portion of each bottom plate is provided with a group of bearing seats, each group of bearing seats is rotatably connected with a roll shaft, the steel cable crosses a coal pile under the guidance of the bearing seats and the roll shafts, a lead screw for adjusting the included angle of the two bottom plates is arranged between the two bottom plates, and a fixing rod is arranged on the outer side of each bottom plate. The two hinged bottom plates are placed at the top of a coal pile, so that the steel cable can leave the coal pile on the two sides of the coal pile and is prevented from falling into the coal pile, the outer contour diameter of each bearing seat is larger than the diameter of the roller shaft, and the steel cable can slide between the bearing seats on the two sides when bypassing the device, so that the steel cable can be prevented from being separated from the device; the rotating roll shaft can isolate the steel cable from the coal pile and prevent the steel cable from falling into the coal pile, sliding friction of the steel cable and the roll shaft is converted into rolling friction, friction force is greatly reduced, abrasion of the steel cable is reduced, the included angle of the two bottom plate supports can be adjusted through the lead screw, and the device can adapt to inclination angles of the tops of different coal piles. Meanwhile, the distance between the steel cable and the top of the coal pile can be adjusted, the steel cable is further prevented from making contact with the coal pile, the fixing rods are inserted into the coal pile, the device and the coal pile can be better fixed, and the device is prevented from shaking when pressed.

The utility model relates to a screw jack device for large -scale work piece support leveling, including support frame, screw rod, hand wheel, bracket and bolt, support frame whole presents the trapezoidal structure of little up big, is equipped with the mounting hole on the support frame, is equipped with the stop hole on the upper portion of mounting hole, hand wheel is installed in the stop hole, and can rotate relative to support frame, screw rod lower part passes in hand wheel and mounting hole in proper order, and is fixedly connected through trapezoidal thread and hand wheel screw thread cooperation, upper portion passes through bolt and bracket fixed connection, the utility model discloses through support frame adopts the trapezoidal structure of little up big, greatly enhanced the overall rigidity, effectively reduced the risk of falling when supporting large -scale work piece, provides the stable support foundation for large -scale work piece, guarantees the security of operation process, simultaneously through the trapezoidal thread cooperation of hand wheel and screw rod, single person can easily realize the lifting operation of screw rod, need not many people cooperation, significantly improves the operation efficiency, can satisfy the demand of large -scale work piece support leveling fast.

The application discloses a transformer integrated cooler group and a transfer device thereof, and relates to the technical field of transformer manufacturing equipment.The transformer integrated cooler group comprises cooler group bodies, a frame and mounting supports, the cooler group bodies are provided in plurality, the plurality of cooler group bodies are fixedly arranged on the mounting supports, the mounting supports are fixedly arranged on the frame, and the cooler group bodies are located on the side of the frame which is far away from the transformer body when the frame is located at the mounting position.The plurality of cooler group bodies are pre-installed on the mounting supports of the frame, and the frame is integrally carried to be further installed on the transformer body, so that the installation site of the transformer is saved, the installation process of the transformer cooler group is simplified, the input of the workers is reduced, and the installation process performance is improved.

The application provides a power grid fault traveling wave positioning method based on a fusion optimization algorithm, comprising: establishing a dynamic virtual fault framework model of a power grid, randomly generating a plurality of virtual fault particles containing position coordinates, and constructing an initial virtual fault particle cluster; using a shortest path algorithm to calculate the topological path length of each virtual fault particle in the virtual fault particle cluster to all measurement points in the power grid, and then calculating a virtual fault space-time feature vector; setting the traveling wave arrival time when a real fault occurs as a target clustering centroid in a feature space, and calculating a feature matching fitness; constructing a non-convex optimization problem with the optimal feature matching fitness as a target, solving the problem using a fusion optimization algorithm, and thus obtaining a fault positioning result and a traveling wave velocity, so as to realize traveling wave positioning. The application can solve the problems of local optimum, slow convergence speed and high calculation cost in the prior art.

The application discloses a point pair registration method and system based on feature guidance and double constraints, and belongs to the technical field of computer vision and robot grabbing. The method aims to solve the problems of recognition failure and insufficient estimation accuracy of traditional methods when parts are blocked and stacked, and point cloud features are not obvious. The scheme adopts a two-stage optimization strategy. Firstly, in the coarse registration stage, an adaptive sampling PPF algorithm guided by geometric features is proposed. By preferentially sampling points in the edge, curvature and other feature obvious areas, the robustness and efficiency of matching are improved. Secondly, in the fine registration stage, an ICP algorithm with multi-constraint optimization is proposed. The pose result obtained by coarse registration is used as the initial value. Through rotation and angle constraint, a multi-objective optimization function is constructed to guide the fine registration process to converge in the correct direction and avoid falling into local optimum, thereby ensuring the accuracy and physical feasibility of the final pose estimation. The application significantly improves the accuracy of industrial part pose recognition.

The embodiment of the application discloses a cluster artificial potential field improvement method and device based on a region threat degree; the method comprises the following steps: acquiring a static obstacle position; establishing a multi-agent cluster structure, and adopting a second-order consistency algorithm to form following; when a moving queue is a desired queue, obstacle avoidance control is performed through an artificial potential field method to obtain an overall following path of the cluster; according to the obstacle avoidance path, the second-order consistency algorithm is adopted to perform path following to control the cluster to avoid obstacles and reach a target point; by introducing second obstacle position information and threat weight information, the size and direction of repulsion force suffered by the cluster are adjusted; by introducing position information and threat weight information of obstacles in a farther range, the size and direction of repulsion force suffered by the cluster are adjusted, so that the moving direction of the cluster is changed, and the cluster is prevented from falling into a local minimum value trap region.

This invention relates to the field of electronic information technology and discloses a zero-carbon airport BIPV photovoltaic direct-drive low-voltage lighting system. It addresses the problem in traditional methods where the MPPT algorithm easily gets trapped in local optima under dynamic shading and irradiance variation conditions, failing to achieve efficient global power point tracking. The method includes: a data acquisition module that collects photovoltaic array voltage, current, and power parameters; an environmental monitoring module that detects environmental changes and generates variation signals using irradiance sensors and cameras; an algorithm initialization module that sets the search range and divides the power curve region based on data and signals; a global search module that scans multiple peaks and traverses the global space to identify the maximum power point; a local optimization module that finely tracks and corrects the step size for peak regions; and a control output module that generates commands to drive the inverter to adjust its operating point and provides status feedback. Through module collaboration, efficient operation of the photovoltaic direct-drive low-voltage lighting is achieved, making it suitable for zero-carbon airport environments.

The utility model discloses an OCA optical cement defoaming device which comprises a vacuum box, a cover plate is hinged to the top of the vacuum box, a connecting pipe is arranged on the left side of the vacuum box, a vacuum pump is fixedly connected to the end of the connecting pipe, a connecting plate is fixedly connected to the upper portion of the side face of the vacuum box, and a dust removal mechanism is arranged on the surface of the connecting plate. The dust removal mechanism comprises a supporting plate, a first motor is fixedly installed on the outer side wall of the supporting plate, and a threaded lead screw is arranged on the inner side wall of the supporting plate. By means of the dust removal mechanism, solid dust particles left on the surface of an OCA raw material can be effectively removed before vacuum bubble removal is conducted on the OCA raw material, the raw material is in a clean state to be subjected to vacuum machining treatment, the treatment effect of follow-up vacuum machining is guaranteed, the dust removal mechanism can be pushed to ascend in cooperation with an air cylinder, and the dust removal efficiency is improved. And therefore, OCA adhesive tape raw materials with different thicknesses can be conveniently cleaned, and the adaptability of the dust removal mechanism to the raw materials is improved when the dust removal mechanism is used.

The invention discloses underground air dust removal equipment for phosphorite mining, and relates to the technical field of air dust removal. The fixed end of an elastic telescopic column is fixedly mounted at the bottom of a bottom plate, a fixed plate is fixedly mounted on the front face of the fixed end of the elastic telescopic column, and one end of a long-strip-shaped rod is fixedly mounted at the bottom of the fixed plate; a round hole is formed in the front face of the square box, the square box is used for controlling the descending speed of a bottom plate at the top of the elastic telescopic column, the situation that the height of the spraying machine deviates from the preset requirement due to the fact that the elastic telescopic column descends too fast is avoided, and it is guaranteed that the spraying machine is always at the optimal dust removal height; the moving range of the spraying machine along with the carrier plate is expanded, multi-direction and wider-area spraying dust falling can be achieved, the problem that dust removal is not thorough in a single fixed range when the underground space is complex is solved, and the comprehensiveness and flexibility of dust removal are improved.

The application belongs to the technical field of power equipment fault diagnosis, and specifically discloses a power transmission line fault identification method and a direct current test instrument. The power transmission line fault identification method comprises the following steps: S1, obtaining fault signals and environmental temperature and humidity data of a to-be-tested line and performing pretreatment to obtain fault characteristic data; S2, inputting the fault characteristic data as test samples into a WOA-KFCM fault detection model, wherein the WOA-KFCM fault detection model comprises a WOA algorithm unit and a KFCM algorithm unit; and S3, calculating the initial clustering center of the KFCM algorithm unit through the WOA algorithm unit, and outputting the line fault category through the KFCM algorithm unit according to the initial clustering center and the test samples. Through the direct current test instrument and the above-mentioned fault identification program, the content of the power transmission line fault can be quickly identified, and the troubleshooting efficiency and the identification accuracy of the complex fault type of the power transmission line are effectively improved.

The invention relates to the field of stropharia rugoso-annulata cultivation, in particular to field furrow bed type soil covering equipment for stropharia rugoso-annulata cultivation, which comprises an operation mounting plate and further comprises adjusting moving mechanisms symmetrically arranged on one side of the operation mounting plate, and each adjusting moving mechanism comprises an arc-shaped mounting plate arranged on one side of the operation mounting plate; the swing soil covering and carding mechanism is arranged on the other side of the operation mounting disc and comprises a swing driving module and a soil covering and carding module, the swing driving module comprises a transmission mounting cover arranged on the other side of the operation mounting disc, and the soil covering and carding module comprises a transfer material guide cover arranged at the outer end of an operation rotating shaft. The overstimulation on hyphae can be reduced; continuity and uniformity of soil covering operation can be guaranteed, and soil covering efficiency and quality are improved; the device can comb covering soil of the device, comb and loosen raw soil on the top, and properly stimulate hyphae to promote the hyphae to split, so that the fruiting rate and the fruiting quality are improved.

This invention relates to the field of neural network architecture search and industrial intelligent processing technology, and discloses a structure search method for multi-output dendritic neuron models for industrial classification tasks. The method includes: initializing the synaptic connection weights and dendritic threshold parameters of the multi-input multi-output dendritic neuron model to generate an initial population; uniformly dividing the initial population into several subpopulations of equal size; calculating the temporal and spatial criteria for each subpopulation after task allocation and optimization, and dynamically allocating evolutionary computational resources for each subpopulation based on the fitness change rate represented by the temporal criterion and the population distribution state represented by the spatial criterion; implementing an accelerated sharing penalty mechanism to adjust fitness values ​​according to the crowding degree among individuals to maintain solution set diversity; and updating each subpopulation. This invention solves the problem of low convergence efficiency caused by uniform resource allocation in traditional large-scale multi-objective evolutionary algorithms during neural network architecture search.

The application provides a runoff sediment content prediction method, system, device and storage medium, belongs to the river sediment prediction technical field, and includes the following steps: introducing the position updating mechanism of the spotted hyena optimization algorithm and the position updating mechanism of the gradient optimization algorithm in the position updating mechanism of the gorilla troop optimization algorithm to obtain an improved gorilla troop optimization algorithm; the model parameters of a BiLSTM model are optimized by using the improved gorilla troop optimization algorithm, and a runoff sediment prediction model is obtained by training the BiLSTM model after parameter optimization. The runoff sediment content prediction result and the prediction accuracy are obtained based on the runoff sediment prediction model. Strong support is provided for ecological protection and engineering planning.

The invention provides a blasting vibration peak velocity prediction method fusing parameter uncertainty and data-driven optimization, which comprises the following steps of: firstly, acquiring data such as blasting parameters, lithologic indexes, geological conditions and actually measured vibration peak velocity (PPV), and establishing a basic database; a prior probability model is constructed, a joint uncertainty model is established in combination with probability disturbance and a fuzzy triangular number, a multi-dimensional disturbance sample is generated through a joint central value and a joint standard deviation and is fused with original data, and an extended database is formed. Feature analysis is carried out on the fused data, and input variables which have obvious influence on the vibration peak velocity (PPV) are screened; and constructing a BP neural network model based on the screened features, carrying out global optimization on a network weight and a threshold by adopting a PSO algorithm, and then carrying out local fine tuning by utilizing Adam. Finally, through training and verification, prediction of the vibration peak velocity (PPV) is realized, and model precision is evaluated through RMSE, MAE, MAPE, Rand other indexes. The influence of rock and soil parameter uncertainty on prediction precision can be effectively processed, and the reliability and applicability of blasting vibration prediction are improved.